British Journal of Surgery

Hardy, R; Lennard, T W J

doi: 10.1002/bjs.6202pmid: 18690616

Adrenalectomy is the recognized treatment for a wide range of hormonally active and inactive adrenal tumours. Recently, subtotal adrenalectomy has re-emerged as a technique, allowing either bilateral adrenal medulla resection or partial adrenal cortex resection in patients with a single remaining adrenal gland. Although not commonly practised in the UK, the rationale is to preserve endogenous steroid production. The authors have assessed the current role of subtotal adrenalectomy by an electronic search of the Medline database for relevant English-language articles. The literature suggests that subtotal adrenalectomy is feasible, and is indicated most strongly in patients with familial predisposition to phaeochromocytoma. Those with aldosterone- or cortisol-producing tumours who have had a contralateral adrenalectomy may also be suitable candidates. Independence from exogenous steroid therapy can be expected in most patients, making this operation one that should be considered in specific situations. The operation of subtotal adrenalectomy was originally described by DeCourcey in 19341. He documented patients with severe hypertension who experienced a reduction in blood pressure after bilateral subtotal adrenalectomy. Over the next seven decades, the popularity of the procedure waxed and waned, with small series of patients being reported by proponents of the technique. Patients who have undergone bilateral adrenalectomy require lifelong exogenous corticosteroid replacement therapy, with its associated risks of undertreatment at times of stress leading to an addisonian crisis2, or overtreatment leading to diabetes, hypertension and osteoporosis. This background, combined with recent improvements in adrenal imaging and operative technique, such as use of the harmonic scalpel, suggest that a review of subtotal adrenalectomy is timely. Is the operation indicated in certain patients, particularly those facing bilateral resection or in whom there is only a single adrenal gland? For subtotal adrenalectomy to be feasible, certain anatomical and pathological criteria must be met. The operation is especially suited to eccentrically placed lesions, because central lesions, particularly those abutting the adrenal vein, make the procedure more demanding or, indeed, impossible. Furthermore, the lesion should be solitary and easily demarcated from normal cortical tissue, allowing planning of resection margins. Finally, there should be a low suspicion of malignancy or of further lesions developing in the remnant. Whether the presence of an intact adrenal vein at the end of the procedure is relevant to remnant function is a matter of debate, but, on balance, the evidence available suggests that the vein may be ligated without adverse consequence3–5. Several series have shown that subtotal adrenalectomy can preserve endogenous steroid production4–7. The largest group of patients in whom subtotal adrenalectomy has been performed are those with bilateral phaeochromocytoma in the context of a familial predisposition, for example multiple endocrine neoplasia type 2 (MEN2) and von Hippel–Landau disease. The tumours generally fulfil some of the criteria for subtotal resection, except that the lesion is often found centrally in the gland and there is a possible risk of further tumour development. Importantly, recent work has demonstrated that the rate of new tumour formation or recurrence in the adrenal remnant is not significantly increased over that noted in the contralateral virgin adrenal8. This reduces concern that tumour ‘spillage’ at the time of subtotal adrenalectomy might lead to an increased local recurrence rate, or that an affected side is more likely to produce a second tumour. Given that such recurrences are readily detected by urinary and/or plasma catecholamine determinations and adrenal imaging, and that distant metastases are unusual, a subtotal procedure may allow patients a prolonged period with adequate endogenous steroid production. Furthermore, bilateral total adrenalectomy does not prevent the development of extra-adrenal paragangliomas, which constituted about 25 per cent of all phaeochromocytomas in an unselected case series from a major endocrine surgical centre9. It is clear that the risk of further tumour development cannot be reduced to zero, even with total adrenal ablative surgery. Subtotal adrenalectomy has also been performed in patients with aldosterone-producing adenomas and, less often, in those with cortisol-producing adenomas and non- functioning adenomas. Reported series document biochemical cure and lack of recurrence in 93–100 per cent of those with aldosterone- or cortisol-producing lesions5–7,10. Conn's adenomas are only rarely malignant; there is less certainty with Cushing's adenomas. Care must therefore be exercised in patient selection. The size of the remnants necessary to ensure adequate corticosteroid production after bilateral excision of adrenal medulla has not been determined precisely. Almost certainly it will vary from patient to patient. In an early study, Brauckhoff and colleagues5 found that remnants of about one-third of the total original gland size predicted adequate function on the basis of synacthen stimulation. Further work by the same group suggested that a remnant size as small as 15 per cent of that of the normal gland would allow exogenous steroid independence, although subclinical adrenal hypofunction, on the basis of an inadequate response to synacthen, was noted in all ten patients studied6. This finding has obvious implications in terms of exogenous steroid use at times of stress. However, both of the above studies involved small numbers of patients and so any conclusions drawn from them should be made with caution. Estimation of adrenal remnant volume has usually been made by computed tomography (CT). Interestingly, recent work suggests that the correlation between the surgeon's estimation of residual volume at operation and the CT estimation is low, with surgeons underestimating adrenal volume. It seems that assessing adequate residual volume is an unpredictable science at best. From the information available at present, it may be concluded that subtotal adrenalectomy is an attractive option in certain circumstances. It is indicated in patients with a familial predisposition to phaeochromocytoma. It may also be indicated in patients with aldosterone-producing adenomas and, possibly, in those with cortisol-producing adenomas who have already undergone contralateral adrenalectomy. Preoperative imaging and intraoperative surgeon assessment may be helpful in deciding suitability for subtotal resection. Many patients undergoing the procedure can be expected to remain independent of exogenous steroids, although this should be confirmed by synacthen stimulation testing. References 1 DeCourcey JL . Subtotal bilateral adrenalectomy for hyperadrenalism (essential hypertension) . Ann Surg 1934 ; 100 : 310 – 318 . Google Scholar Crossref Search ADS PubMed WorldCat 2 Oelkers W . Adrenal insufficiency . N Engl J Med 1996 ; 335 : 1206 – 1212 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Roukounakis N , Dimas S, Kafetzis I, Bethanis S, Gatsulis N, Kostas H et al. Is preservation of the adrenal vein mandatory in laparoscopic adrenal-sparing surgery? JSLS 2007 ; 11 : 215 – 218 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 4 Yip L , Lee JE, Shapiro SE, Waguespack SG, Sherman SI, Hoff AO et al. Surgical management of hereditary pheochromocytoma . J Am Coll Surg 2004 ; 198 : 525 – 534 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Brauckhoff M , Thanh PN, Gimm O, Bar A, Brauckhoff K, Dralle H. Functional results after endoscopic subtotal cortical-sparing adrenalectomy . Surg Today 2003 ; 33 : 342 – 348 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Brauckhoff M , Gimm O, Thanh PN, Bar A, Ukkat J, Brauckhoff K et al. Critical size of residual adrenal tissue and recovery from impaired early postoperative adrenocortical function after subtotal bilateral adrenalectomy . Surgery 2003 ; 134 : 1020 – 1027 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Walz MK , Peitgen K, Diesing D, Petersenn S, Janssen O, Philipp T et al. Partial versus total adrenalectomy by the posterior retroperitoneoscopic approach: early and long-term results of 325 consecutive procedures in primary adrenal neoplasias . World J Surg 2004 ; 28 : 1323 – 1329 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Machens A , Brauckhoff M, Gimm O, Dralle H. Risk-oriented approach to hereditary adrenal pheochromocytoma . Ann N Y Acad Sci 2006 ; 1073 : 417 – 428 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Madani R , Al-Hashmi M, Bliss R, Lennard TW. Ectopic pheochromocytoma: does the rule of tens apply? World J Surg 2007 ; 31 : 849 – 854 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Jeschke K , Janetschek G, Peschel R, Schellander L, Bartsch G, Henning K. Laparoscopic partial adrenalectomy in patients with aldosterone-producing adenomas: indications, technique, and results . Urology 2003 ; 61 : 69 – 72 . Google Scholar Crossref Search ADS PubMed WorldCat Copyright © 2008 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) Copyright © 2008 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd.

Sear, J W; Foex, P

doi: 10.1002/bjs.6323pmid: 18690617

Beta-blockers have for many years been regarded as a means of reducing the risk of perioperative ischaemia in patients with coronary artery disease, or risk factors for that disease, who present for non-cardiac surgery1. An improvement in cardiac outcome at 2 years after non-cardiac surgery was reported by Mangano and colleagues in a randomized controlled trial (RCT) in 19962. This trial, however, had limitations in that it did not include events occurring during hospitalization in its analysis, and an intention-to-treat analysis does not show long-term mortality benefit. Despite these limitations, the American College of Medicine in 1997 advocated the preoperative administration of the beta-blocker atenolol for all patients with coronary artery disease or risk factors for that condition. In 1999, a second RCT involving patients with reversible ischaemia who underwent major elective vascular surgery showed a 100 per cent reduction in myocardial infarction and an 80 per cent reduction in cardiac death in those treated with bisoprolol before and during operation3. Again, the study had limitations. It was unblinded and was stopped at the first interim analysis despite randomizing only 112 patients and recording only 20 events. Nevertheless, these data strengthened the case for beta-blockade, and an American College of Cardiology/American Heart Association (ACC/AHA) guideline in 2002 stated that ‘current studies suggest that appropriately administered beta-blockers reduce perioperative ischaemia and may reduce the risk of myocardial infarction and death in high risk patients’. This acknowledgement of some degree of uncertainty about the efficacy of beta-blockade in surgical patients was appropriate, because other studies failed to show early or short-term outcome benefits4,5. More recent meta-analyses of the use of beta-blockers in cardiac and non-cardiac surgery indicate only a trend toward protection6,7. These findings are reflected in the 2006 and 2007 ACC/AHA guidelines that state ‘current studies suggest that beta-blockers reduce perioperative ischaemia and may reduce the risk of myocardial infarction and death in patients with known coronary artery disease’. To define better the role of beta-blockade in non-cardiac surgery, a large multinational RCT was carried out of beta-blockade versus placebo in patients at risk of perioperative cardiac events, the PeriOperative Ischemia Study Evaluation (POISE trial). This included 8351 patients from 23 countries and the results were published in the Lancet on 31 May 20088. The study drug, metoprolol succinate extended-release, was given to 4174 patients with the goal of achieving 200 mg daily but down-titrated for hypotension or bradycardia to 100 mg daily. Administration of the drug or placebo started between 2 and 4 h before operation and was continued for 30 days. The primary outcomes were cardiovascular death, non-fatal myocardial infarction and non-fatal cardiac arrest. Secondary outcomes included total mortality, stroke, myocardial infarction, coronary revascularization, atrial fibrillation, congestive heart failure, hypotension and bradycardia. Thirty-day results showed a significant reduction in all myocardial infarctions from 239 (5·7 per cent) in the placebo group to 176 (4·2 per cent) in the metoprolol group; metoprolol was also associated with a reduced need for coronary revascularization and a reduction in the number of patients developing atrial fibrillation. In contrast, there was a significant increase in total mortality from 97 (2·3 per cent) in the placebo group to 129 (3·1 per cent) in the metoprolol group. There was also a significant increase in stroke from 19 (0·5 per cent) in controls to 41 (1·0 per cent) with metoprolol. Metoprolol was also associated with excess clinically significant hypotension and bradycardia. The POISE results suggest, for every 1000 patients with a similar risk profile undergoing non-cardiac surgery, that metoprolol will prevent 25 patients from suffering a significant cardiac event. On the other hand, metoprolol could be responsible for an excess of deaths and five strokes. Clearly this study does not support the widespread introduction of perioperative beta-blockers in all patients at risk for coronary disease. It is reasonable to ask whether benefits might be obtained from perioperative beta-blockade if the adverse effects can be avoided. POISE produced evidence of cardiac protection, as shown by reductions in myocardial infarction, need for emergency revascularization and non-fatal cardiac arrest. However, the incidence of stroke with its resulting disability was increased in the treatment group, as was all-cause mortality. Could these adverse effects be avoided? Did the timing of initiation of treatment and the dose of beta-blocker play a role? In the study showing the most marked protective effect, beta-blockade was initiated a week before surgery3; in the POISE study, metoprolol extended-release was given as 100 mg twice daily for the first time on the day of surgery. Was the dose of metoprolol too high, explaining the high incidence of hypotension, or are the findings a drug-specific, rather than a class, effect? Neither is likely given that previous meta-analyses of beta-blockade encompass a variety of drugs and dosages, and demonstrate that clinically significant hypotension can occur. Was control of heart rate insufficient? Although POISE had defined limits of heart rate (fewer than 50 beats/min) for temporarily withholding the study drug, there were no upper limits of heart rate to increase the dosage. The value of tight postoperative heart rate control on morbidity and mortality was shown in the meta-analysis of Beattie and co-workers9. Yet, the role of strict control of heart rate has been questioned10, and may only be achieved at the cost of other adverse effects, such as hypotension. Multivariable analysis of strokes within the POISE study8 confirms a significant association of beta-blockade with clinically significant hypotension. Although routine prescribing of preoperative beta-blockers for all patients with assumed or proven coronary artery disease cannot be recommended, if these drugs are thought to be indicated for an intercurrent medical condition, they should be started well before, rather than at the time of, surgery. For all patients the strict perioperative monitoring of blood pressure and heart rate is necessary to avoid both hypotension and bradycardia. There should be a careful titration of dose to effect rather than the application of a standard dose strategy. Finally, clinicians need to develop and adhere to strict protocols. Although based solely on observational studies, patients receiving chronic beta-blockade should maintain this medication throughout the perioperative period. This recommendation is appropriate as POISE only studied the effects of acute perioperative beta-blockade. For the rest, until future RCTs explore the methods of beta-blocker administration further (time of initiation, dose), these agents should be given on an individual basis only after due consideration has been given to their possible hazards. References 1 Stone JG , Foex P, Sear JW, Johnson LL, Khambatta HJ, Triner L. Risk of myocardial ischaemia during anaesthesia in treated and untreated hypertensive patients . Br J Anaesth 1988 ; 61 : 675 – 679 . Google Scholar Crossref Search ADS PubMed WorldCat 2 Mangano DT , Layug EL, Wallace A, Tateo I. Effect of atenolol on mortality and cardiovascular morbidity after noncardiac surgery. Multicenter Study of Perioperative Ischemia Research Group . N Engl J Med 1996 ; 335 : 1713 – 1720 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Poldermans D , Boersma E, Bax JJ, Thomson IR, van de Ven LL, Blankensteijn JD et al. The effect of bisoprolol on perioperative mortality and myocardial infarction in high-risk patients undergoing vascular surgery. Dutch Echocardiographic Cardiac Risk Evaluation Applying Stress Echocardiography Study Group . N Engl J Med 1999 ; 341 : 1789 – 1794 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Zaugg M , Tagliente T, Lucchinetti E, Jacobs E, Krol M, Bodian C et al. Beneficial effects from beta-adrenergic blockade in elderly patients undergoing noncardiac surgery . Anesthesiology 1999 ; 91 : 1674 – 1686 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Urban MK , Markowitz SM, Gordon MA, Urquhart BL, Kligfield P. Postoperative prophylactic administration of beta-adrenergic blockers in patients at risk for myocardial ischemia . Anesth Analg 2000 ; 90 : 1257 – 1261 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Devereaux PJ , Beattie WS, Choi PT, Badner NH, Guyatt GH, Villar JC et al. How strong is the evidence for the use of perioperative beta blockers in non-cardiac surgery? Systematic review and meta-analysis of randomised controlled trials . BMJ 2005 ; 331 : 313 – 321 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Wiesbauer F , Schlager O, Domanovits H, Wildner B, Maurer G, Muellner M et al. Perioperative beta-blockers for preventing surgery-related mortality and morbidity: a systematic review and meta-analysis . Anesth Analg 2007 ; 104 : 27 – 41 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Poise Study Group . Effects of extended-release metoprolol succinate in patients undergoing non-cardiac surgery (POISE trial): a randomised controlled trial . Lancet 2008 ; 371 : 1839 – 1847 . Crossref Search ADS PubMed WorldCat 9 Beattie WS , Wijeysundera DN, Karkouti K, McCluskey S, Tait G. Does tight heart rate control improve beta-blocker efficacy? An updated analysis of the noncardiac surgical randomized trials . Anesth Analg 2008 ; 106 : 1039 – 1048 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Biccard BM , Sear JW, Foex P. Meta-analysis of the effect of heart rate achieved by perioperative beta-adrenergic blockade on cardiovascular outcomes . Br J Anaesth 2008 ; 100 : 23 – 28 . Google Scholar Crossref Search ADS PubMed WorldCat Copyright © 2008 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) Copyright © 2008 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd.

Koh, C E; Young, C J; Young, J M; Solomon, M J

doi: 10.1002/bjs.6303pmid: 18655219

Abstract Background Pelvic floor dysfunction (PFD) is a type of functional constipation. The effectiveness of biofeedback as a treatment remains unclear. Methods A systematic review of all randomized controlled trials evaluating the effectiveness of biofeedback in adults with PFD was carried out. All online databases from 1950 to 2007 were searched. This was supplemented by hand searching references of retrieved articles. Results Seven trials fulfilled the inclusion criteria. Three compared biofeedback with non-biofeedback treatments and four compared different biofeedback modalities. Electromyography feedback was most widely utilized. The trials were heterogeneous with varied inclusion criteria, treatment protocols and definitions of success. Most had methodological limitations. Quality of life and psychological morbidity were assessed rarely. Meta-analysis of the studies involving any form of biofeedback compared with any other treatment suggested that biofeedback conferred a sixfold increase in the odds of treatment success (odds ratio 5·861 (95 per cent confidence interval 2·175 to 15·794); random-effects model). Conclusion Although biofeedback is the recommended treatment for PFD, high-quality evidence of effectiveness is lacking. Meta-analysis of the available evidence suggests that biofeedback is the best option, but well designed trials that take into account quality of life and psychological morbidity are needed. Introduction Constipation affects 2–30 per cent of the population of Westernized countries1. It is the most common gastrointestinal complaint leading to a medical consultation2–4. Although most constipated individuals will respond to simple remedies such as lifestyle alteration, diet modification or the judicious use of laxatives, some are left with a debilitating problem that adversely affects quality of life (QOL)5. Among those who do not respond to simple remedies, pelvic floor dysfunction (PFD) is particularly common1,6,7 and it has been estimated that up to 50 per cent of constipated patients who warrant referral to a tertiary centre have PFD1,6–8. PFD is a type of functional constipation caused by paradoxical contraction (or failure of adequate relaxation) of pelvic floor muscles during defaecation or the inability to generate adequate propulsive force1. The role of the pelvic floor in constipation was first described in 1978 by Martelli and colleagues9, who treated the condition with anorectal myectomy—a procedure commonly used in patients with ultrashort-segment Hirschsprung's disease. In 1985, Preston and Lennard-Jones10 demonstrated that a much greater force was needed to withdraw an inflated intrarectal balloon from the anal canal of affected patients. This was thought to be the result of pelvic floor muscle spasm and the term anismus was coined, in analogy to vaginismus10. This muscle dysfunction was considered to be an acquired maladaptive behaviour that might be amenable to muscle retraining1,8,10. Clinical characteristics and the diagnostic criteria for PFD as defined by the Rome Foundation11 are summarized in Table 1. It is important to distinguish PFD from structural causes of pelvic outlet obstruction, such as rectal prolapse, a poorly emptying rectocele or low rectal malignancy1,12. These conditions mimic PFD and may present in a similar fashion but, unlike PFD, they have an anatomical basis13. It is also known now that PFD is associated with depression and anxiety, and a diminished QOL13–15. However, whether it is the cause or the result of constipation is unclear15. There is some evidence that psychological status and QOL of affected individuals improve with treatment, although this has been relatively poorly studied15. Table 1 Diagnostic criteria for pelvic floor dysfunction11 Straining during bowel movements Feeling of incomplete evacuation Sensation of blocked stools Manual manoeuvres to facilitate defaecation Difficulty relaxing to allow defaecation Fulfils criteria for functional constipation During repeated attempts to defaecate, must have at least two of the following:  Evidence of impaired evacuation, based on balloon expulsion test   or imaging  Inappropriate contraction of the pelvic floor muscle (sphincter or   puborectalis) or < 20% relaxation of basal resting sphincter   pressure by manometry, imaging or EMG  Inadequate propulsive forces assessed by manometry or imaging Diagnostic criteria fulfilled for the last 3 months with symptom onset at least 6 months before diagnosis Straining during bowel movements Feeling of incomplete evacuation Sensation of blocked stools Manual manoeuvres to facilitate defaecation Difficulty relaxing to allow defaecation Fulfils criteria for functional constipation During repeated attempts to defaecate, must have at least two of the following:  Evidence of impaired evacuation, based on balloon expulsion test   or imaging  Inappropriate contraction of the pelvic floor muscle (sphincter or   puborectalis) or < 20% relaxation of basal resting sphincter   pressure by manometry, imaging or EMG  Inadequate propulsive forces assessed by manometry or imaging Diagnostic criteria fulfilled for the last 3 months with symptom onset at least 6 months before diagnosis EMG, electromyography. Open in new tab Table 1 Diagnostic criteria for pelvic floor dysfunction11 Straining during bowel movements Feeling of incomplete evacuation Sensation of blocked stools Manual manoeuvres to facilitate defaecation Difficulty relaxing to allow defaecation Fulfils criteria for functional constipation During repeated attempts to defaecate, must have at least two of the following:  Evidence of impaired evacuation, based on balloon expulsion test   or imaging  Inappropriate contraction of the pelvic floor muscle (sphincter or   puborectalis) or < 20% relaxation of basal resting sphincter   pressure by manometry, imaging or EMG  Inadequate propulsive forces assessed by manometry or imaging Diagnostic criteria fulfilled for the last 3 months with symptom onset at least 6 months before diagnosis Straining during bowel movements Feeling of incomplete evacuation Sensation of blocked stools Manual manoeuvres to facilitate defaecation Difficulty relaxing to allow defaecation Fulfils criteria for functional constipation During repeated attempts to defaecate, must have at least two of the following:  Evidence of impaired evacuation, based on balloon expulsion test   or imaging  Inappropriate contraction of the pelvic floor muscle (sphincter or   puborectalis) or < 20% relaxation of basal resting sphincter   pressure by manometry, imaging or EMG  Inadequate propulsive forces assessed by manometry or imaging Diagnostic criteria fulfilled for the last 3 months with symptom onset at least 6 months before diagnosis EMG, electromyography. Open in new tab Treatment of pelvic floor dysfunction Historically, PFD has been managed surgically with anorectal myectomy9,16, surgical division of puborectalis17–19, botulinum toxin infiltration into puborectalis20–22 and, as a last resort, the formation of a stoma in patients with intractable symptoms23,24. Anorectal myectomy and surgical division of puborectalis provide only short-term symptom relief and are associated with an unacceptable risk of incontinence9,16–19. Botulinum toxin infiltration remains experimental despite promising initial results and the need for repeated infiltration also makes it less than ideal20–22. Conservative management has also been disappointing8,25, leaving clinicians with the option of biofeedback. Biofeedback is a form of operant conditioning that teaches the patient how to control a physiological function that is not usually under conscious control1,7. By using an instrument that provides visual, auditory or verbal feedback of an action, the action can then be fine tuned and reinforced until the desired response is achieved. Various instruments, such as electromyography (EMG) and manometry, have been used to provide the necessary feedback about individual performance in biofeedback6. The first series of biofeedback in adults with PFD was published in 1987 by Bleijenberg and Kuijpers26, who reported a success rate of 70 per cent. To date, most trials evaluating biofeedback have been small uncontrolled studies with limited follow-up27–32. Previous narrative reviews, meta-analysis and systematic review have attempted to address the issue of effectiveness of biofeedback, and concluded that there was inadequate evidence to support such treatment4,7,25,33–35. However, biofeedback remained popular because it is non-invasive and does not preclude future treatments1,6,7. In recent years, several randomized controlled trials (RCTs) have been completed and an updated systematic review of this topic seems timely. Method The following questions were addressed. First, does biofeedback confer improved outcomes for adults with PFD compared with conservative treatment that does not include biofeedback? Second, what is the relative effectiveness of different types of biofeedback therapy? All RCTs evaluating biofeedback in adults with PFD were considered for review. Paediatric trials were excluded. RCTs had to include biofeedback in at least one treatment arm. For the purposes of this paper, biofeedback was defined as a form of intervention involving visual, auditory or verbal feedback of an activity using an instrument (EMG, manometry or an intrarectal balloon). Search strategy All online databases were searched and included CINAHL (January 1982 to April 2007), EMBASE (January 1980 to April 2007), Medline (January 1950 to April 2007), PsycInfo (January 1967 to April 2007), EBM review and the Cochrane Central Trials Registry. The following search terms were applied: constipation + biofeedback/psychother- apy; anismus + biofeedback/psychotherapy; obstructive defaecation + biofeedback/psychotherapy; dyschezia + biofeedback/psychotherapy; dyssynergia/dyssynergy + biofeedback/psychotherapy; rectocele + biofeedback/psychotherapy; rectal intussusception + biofeedback/psychotherapy; rectal prolapse + biofeedback/psycho-therapy. This search was supplemented by hand searching references of retrieved articles including previous reviews. No language restriction was applied. Potentially eligible studies were identified by one author (C.E.K) by screening titles and abstracts retrieved by applying the search terms above. All trials were then assessed independently by two reviewers (C.E.K., C.J.Y.) for eligibility. These two reviewers then independently extracted the data, assessed the quality of the trial and checked for discrepancies. Any disagreement was resolved with a third independent reviewer (M.J.S). Authors of individual trials were contacted if necessary. Outcomes The primary outcome used was symptomatic improvement (however defined by authors). The secondary outcomes were QOL (however defined by authors) and levels of depression or anxiety (however defined by authors). Methodological quality of included studies The methodological quality of studies was assessed based on methods of randomization, allocation concealment, blinding and loss to follow-up. For methods of randomization, trials were rated as: appropriate randomization procedure (A), unclear (B) or inappropriate randomization procedure (C). Allocation concealment was rated as: concealed (A), unclear (B) or not concealed (C). Blinding was assessed as: double blind (A), blinding of participants or investigators only (B), blinding of analysis only (C), unclear (D) or not blinded (E). Completeness of follow-up was assessed as: 5·0 per cent or less lost to follow-up (A), 5·1–10·0 per cent lost to follow-up (B), 10·1–15·0 per cent lost to follow-up (C), more than 15 per cent lost to follow-up (D) or unclear (E). Statistical analysis For each trial, the proportion of patients who experienced a symptomatic improvement in defaecation was calculated as the primary outcome measure. Unfortunately, QOL and psychological outcomes of treatment were rarely reported, precluding further analysis of secondary outcomes. Meta-analysis was carried out using Comprehensive Meta-Analysis (Biostat, Englewood, New Jersey, USA) using random-effects models, first, for studies comparing any form of biofeedback versus any other treatment and, second, for EMG biofeedback versus non-EMG biofeedback. Odds ratios (OR) and 95 per cent confidence intervals (c.i.) were calculated. Results Of the 5028 articles identified in the literature search, 47 were considered potentially eligible on the basis of the abstract. Of these, only seven were RCTs and eligible for inclusion36–42. The remaining 40 studies were retrospective, uncontrolled or non-randomized trials (Fig. 1). Details of excluded studies are available from the authors. Details of included studies are summarized in Table 2. These seven trials had a total of 413 participants. Fig. 1 Open in new tabDownload slide Selection and screening process for studies of biofeedback Table 2 Details of included studies Reference . Country of origin . n . Sex ratio (F : M) . Mean age (years) . Duration of symptoms (years) . Intervention . No. of sessions . Follow-up (months) . Inclusion criteria . Exclusion criteria . 36 The Netherlands 21 16 : 5 37 8 EMG BF (11) Balloon BF (10) 8 9 Inadequate relaxation on EMG and defaecography — History of severe constipation, incomplete evacuation, obstructed defaecation and digital assistance 37 England 60 53 : 7 41 14 EMG + balloon BF (30) Balloon BF (30) 3 (1–7) Unclear History of < 3 bowel actions/week, excessive straining — 2 (1–4) Failed conservative management 38 Sweden 26 23 : 3 55 11 Manometry BF (13) EMG BF (13) ≤ 10 6 Rome criteria for functional constipation Organic causes PPC on EMG 39 USA 36 26 : 10 61 — EMG BF (9) EMG + balloon BF (9) EMG BF + EMG HT (8) EMG BF + balloon BF + EMG HT (10) Unclear Unclear PPC (not further defined) Failed conservative management Organic causes 40 Italy 109 104 : 5 34 — EMG BF (54) Laxatives (55) 5 12 Rome criteria for functional constipation Slow colonic transit PPC on EMG and manometry Failed BET (> 5 min) Failed conservative management Idiopathic megarectum or megacolon Organic cause 41 USA 77 69 : 8 43 17 Manometry BF (28) Standard treatment (24) Sham feedback (25) 5 (4–6) 3 Rome criteria for functional constipation Organic causes Dyssynergistic defaecation (not further defined) or prolonged BET (> 1 min) or prolonged colonic transit 42 USA 84 71 : 13 50 — EMG BF (30) Diazepam (30) Diazepam placebo (24) 6 3 Rome criteria for functional constipation Previous BF treatment PPC on EMG, manometry or defaecography Diazepam use Failed conservative management Evidence of inadequate propulsive force or incomplete evacuation Reference . Country of origin . n . Sex ratio (F : M) . Mean age (years) . Duration of symptoms (years) . Intervention . No. of sessions . Follow-up (months) . Inclusion criteria . Exclusion criteria . 36 The Netherlands 21 16 : 5 37 8 EMG BF (11) Balloon BF (10) 8 9 Inadequate relaxation on EMG and defaecography — History of severe constipation, incomplete evacuation, obstructed defaecation and digital assistance 37 England 60 53 : 7 41 14 EMG + balloon BF (30) Balloon BF (30) 3 (1–7) Unclear History of < 3 bowel actions/week, excessive straining — 2 (1–4) Failed conservative management 38 Sweden 26 23 : 3 55 11 Manometry BF (13) EMG BF (13) ≤ 10 6 Rome criteria for functional constipation Organic causes PPC on EMG 39 USA 36 26 : 10 61 — EMG BF (9) EMG + balloon BF (9) EMG BF + EMG HT (8) EMG BF + balloon BF + EMG HT (10) Unclear Unclear PPC (not further defined) Failed conservative management Organic causes 40 Italy 109 104 : 5 34 — EMG BF (54) Laxatives (55) 5 12 Rome criteria for functional constipation Slow colonic transit PPC on EMG and manometry Failed BET (> 5 min) Failed conservative management Idiopathic megarectum or megacolon Organic cause 41 USA 77 69 : 8 43 17 Manometry BF (28) Standard treatment (24) Sham feedback (25) 5 (4–6) 3 Rome criteria for functional constipation Organic causes Dyssynergistic defaecation (not further defined) or prolonged BET (> 1 min) or prolonged colonic transit 42 USA 84 71 : 13 50 — EMG BF (30) Diazepam (30) Diazepam placebo (24) 6 3 Rome criteria for functional constipation Previous BF treatment PPC on EMG, manometry or defaecography Diazepam use Failed conservative management Evidence of inadequate propulsive force or incomplete evacuation EMG, electromyography; BF, biofeedback; PPC, paradoxical pelvic contraction; HT, home trainer; BET, balloon expulsion test. Open in new tab Table 2 Details of included studies Reference . Country of origin . n . Sex ratio (F : M) . Mean age (years) . Duration of symptoms (years) . Intervention . No. of sessions . Follow-up (months) . Inclusion criteria . Exclusion criteria . 36 The Netherlands 21 16 : 5 37 8 EMG BF (11) Balloon BF (10) 8 9 Inadequate relaxation on EMG and defaecography — History of severe constipation, incomplete evacuation, obstructed defaecation and digital assistance 37 England 60 53 : 7 41 14 EMG + balloon BF (30) Balloon BF (30) 3 (1–7) Unclear History of < 3 bowel actions/week, excessive straining — 2 (1–4) Failed conservative management 38 Sweden 26 23 : 3 55 11 Manometry BF (13) EMG BF (13) ≤ 10 6 Rome criteria for functional constipation Organic causes PPC on EMG 39 USA 36 26 : 10 61 — EMG BF (9) EMG + balloon BF (9) EMG BF + EMG HT (8) EMG BF + balloon BF + EMG HT (10) Unclear Unclear PPC (not further defined) Failed conservative management Organic causes 40 Italy 109 104 : 5 34 — EMG BF (54) Laxatives (55) 5 12 Rome criteria for functional constipation Slow colonic transit PPC on EMG and manometry Failed BET (> 5 min) Failed conservative management Idiopathic megarectum or megacolon Organic cause 41 USA 77 69 : 8 43 17 Manometry BF (28) Standard treatment (24) Sham feedback (25) 5 (4–6) 3 Rome criteria for functional constipation Organic causes Dyssynergistic defaecation (not further defined) or prolonged BET (> 1 min) or prolonged colonic transit 42 USA 84 71 : 13 50 — EMG BF (30) Diazepam (30) Diazepam placebo (24) 6 3 Rome criteria for functional constipation Previous BF treatment PPC on EMG, manometry or defaecography Diazepam use Failed conservative management Evidence of inadequate propulsive force or incomplete evacuation Reference . Country of origin . n . Sex ratio (F : M) . Mean age (years) . Duration of symptoms (years) . Intervention . No. of sessions . Follow-up (months) . Inclusion criteria . Exclusion criteria . 36 The Netherlands 21 16 : 5 37 8 EMG BF (11) Balloon BF (10) 8 9 Inadequate relaxation on EMG and defaecography — History of severe constipation, incomplete evacuation, obstructed defaecation and digital assistance 37 England 60 53 : 7 41 14 EMG + balloon BF (30) Balloon BF (30) 3 (1–7) Unclear History of < 3 bowel actions/week, excessive straining — 2 (1–4) Failed conservative management 38 Sweden 26 23 : 3 55 11 Manometry BF (13) EMG BF (13) ≤ 10 6 Rome criteria for functional constipation Organic causes PPC on EMG 39 USA 36 26 : 10 61 — EMG BF (9) EMG + balloon BF (9) EMG BF + EMG HT (8) EMG BF + balloon BF + EMG HT (10) Unclear Unclear PPC (not further defined) Failed conservative management Organic causes 40 Italy 109 104 : 5 34 — EMG BF (54) Laxatives (55) 5 12 Rome criteria for functional constipation Slow colonic transit PPC on EMG and manometry Failed BET (> 5 min) Failed conservative management Idiopathic megarectum or megacolon Organic cause 41 USA 77 69 : 8 43 17 Manometry BF (28) Standard treatment (24) Sham feedback (25) 5 (4–6) 3 Rome criteria for functional constipation Organic causes Dyssynergistic defaecation (not further defined) or prolonged BET (> 1 min) or prolonged colonic transit 42 USA 84 71 : 13 50 — EMG BF (30) Diazepam (30) Diazepam placebo (24) 6 3 Rome criteria for functional constipation Previous BF treatment PPC on EMG, manometry or defaecography Diazepam use Failed conservative management Evidence of inadequate propulsive force or incomplete evacuation EMG, electromyography; BF, biofeedback; PPC, paradoxical pelvic contraction; HT, home trainer; BET, balloon expulsion test. Open in new tab Primary and secondary outcomes Table 3 summarizes results from all seven studies. Only three trials40–42 had clearly defined primary and secondary outcomes of interest. In the remaining four trials, a large number of outcome measures were reported including anismus index, EMG quality, bowel frequency, straining and laxative use, but it was unclear which measure was the primary outcome. For the purposes of the present review, primary outcome was symptomatic improvement, and secondary outcomes were changes in depression, anxiety and QOL. A global bowel satisfaction or overall subjective improvement was available in six studies (the exception was Heymen et al. in 199939) whereby only cathartic use and unassisted bowel motions were reported (Table 3). Table 3 Results of included studies Reference . Year . Intervention . n . Symptomatic improvement . Depression/anxiety . Quality of life . Bleijenberg et al.36 1994 EMG BF 11 8 (73) Symptom Checklist 90 Not assessed Balloon BF 10 2 (20) Improved depression  and inadequacy  subscales in EMG B  group Koutsomanis et al.37 1995 EMG + balloon BF 30 16 (53) Not assessed Not assessed Balloon BF 30 18 (60) Glia et al.38 1997 EMG BF 13 8 (62) Not assessed Not assessed Manometry BF 13 6 (46) Heymen et al.39* 1999 EMG BF 9 8 (89) Not assessed Not assessed EMG + balloon BF 9 8 (89) EMG BF + EMG HT 8 5 (63) EMG BF + balloon BF + EMG HT 10 6 (60) Chiarioni et al.40 2006 EMG BF 54 43 (80) Not assessed Not assessed Laxatives 55 12 (22) Rao et al.41 2007 Manometry BF 28 21 (75) Not assessed Not assessed Standard treatment 24 15 (63) Sham feedback 25 12 (48) Heymen et al.42 2007 EMG BF (30) 30 21 (70) Changes not assessed PAC QOL and SF-36® Diazepam (30) 30 7 (23) QOL tended to Diazepam placebo (24) 24 9 (38)  improve with BF but this was not statistically significant Reference . Year . Intervention . n . Symptomatic improvement . Depression/anxiety . Quality of life . Bleijenberg et al.36 1994 EMG BF 11 8 (73) Symptom Checklist 90 Not assessed Balloon BF 10 2 (20) Improved depression  and inadequacy  subscales in EMG B  group Koutsomanis et al.37 1995 EMG + balloon BF 30 16 (53) Not assessed Not assessed Balloon BF 30 18 (60) Glia et al.38 1997 EMG BF 13 8 (62) Not assessed Not assessed Manometry BF 13 6 (46) Heymen et al.39* 1999 EMG BF 9 8 (89) Not assessed Not assessed EMG + balloon BF 9 8 (89) EMG BF + EMG HT 8 5 (63) EMG BF + balloon BF + EMG HT 10 6 (60) Chiarioni et al.40 2006 EMG BF 54 43 (80) Not assessed Not assessed Laxatives 55 12 (22) Rao et al.41 2007 Manometry BF 28 21 (75) Not assessed Not assessed Standard treatment 24 15 (63) Sham feedback 25 12 (48) Heymen et al.42 2007 EMG BF (30) 30 21 (70) Changes not assessed PAC QOL and SF-36® Diazepam (30) 30 7 (23) QOL tended to Diazepam placebo (24) 24 9 (38)  improve with BF but this was not statistically significant Values in parentheses are percentages. * Unassisted bowel movement was used to calculate treatment effectiveness as there was no overall patient-reported improvement. EMG, electromyography; BF, biofeedback; HT, home trainer; PAC QOL, patient assessment of constipation quality of life; SF-36®, Short Form 36 (Medical Outcomes Trust, Waltham, Massachusetts, USA), QOL, quality of life. Open in new tab Table 3 Results of included studies Reference . Year . Intervention . n . Symptomatic improvement . Depression/anxiety . Quality of life . Bleijenberg et al.36 1994 EMG BF 11 8 (73) Symptom Checklist 90 Not assessed Balloon BF 10 2 (20) Improved depression  and inadequacy  subscales in EMG B  group Koutsomanis et al.37 1995 EMG + balloon BF 30 16 (53) Not assessed Not assessed Balloon BF 30 18 (60) Glia et al.38 1997 EMG BF 13 8 (62) Not assessed Not assessed Manometry BF 13 6 (46) Heymen et al.39* 1999 EMG BF 9 8 (89) Not assessed Not assessed EMG + balloon BF 9 8 (89) EMG BF + EMG HT 8 5 (63) EMG BF + balloon BF + EMG HT 10 6 (60) Chiarioni et al.40 2006 EMG BF 54 43 (80) Not assessed Not assessed Laxatives 55 12 (22) Rao et al.41 2007 Manometry BF 28 21 (75) Not assessed Not assessed Standard treatment 24 15 (63) Sham feedback 25 12 (48) Heymen et al.42 2007 EMG BF (30) 30 21 (70) Changes not assessed PAC QOL and SF-36® Diazepam (30) 30 7 (23) QOL tended to Diazepam placebo (24) 24 9 (38)  improve with BF but this was not statistically significant Reference . Year . Intervention . n . Symptomatic improvement . Depression/anxiety . Quality of life . Bleijenberg et al.36 1994 EMG BF 11 8 (73) Symptom Checklist 90 Not assessed Balloon BF 10 2 (20) Improved depression  and inadequacy  subscales in EMG B  group Koutsomanis et al.37 1995 EMG + balloon BF 30 16 (53) Not assessed Not assessed Balloon BF 30 18 (60) Glia et al.38 1997 EMG BF 13 8 (62) Not assessed Not assessed Manometry BF 13 6 (46) Heymen et al.39* 1999 EMG BF 9 8 (89) Not assessed Not assessed EMG + balloon BF 9 8 (89) EMG BF + EMG HT 8 5 (63) EMG BF + balloon BF + EMG HT 10 6 (60) Chiarioni et al.40 2006 EMG BF 54 43 (80) Not assessed Not assessed Laxatives 55 12 (22) Rao et al.41 2007 Manometry BF 28 21 (75) Not assessed Not assessed Standard treatment 24 15 (63) Sham feedback 25 12 (48) Heymen et al.42 2007 EMG BF (30) 30 21 (70) Changes not assessed PAC QOL and SF-36® Diazepam (30) 30 7 (23) QOL tended to Diazepam placebo (24) 24 9 (38)  improve with BF but this was not statistically significant Values in parentheses are percentages. * Unassisted bowel movement was used to calculate treatment effectiveness as there was no overall patient-reported improvement. EMG, electromyography; BF, biofeedback; HT, home trainer; PAC QOL, patient assessment of constipation quality of life; SF-36®, Short Form 36 (Medical Outcomes Trust, Waltham, Massachusetts, USA), QOL, quality of life. Open in new tab Two studies evaluated depression or anxiety36,42. However, only pretreatment measures were taken and no post-treatment measures were reported in one of these42, and so changes in depression or anxiety levels with treatment cannot be assessed. In the study of Bleijenberg and colleagues36, patients receiving EMG biofeedback had significantly improved depression and inadequacy scores when measured using Symptom Checklist 90. Only Heymen and co-workers42 assessed changes in QOL with treatment. Although there was a trend for patients in the biofeedback group to improve, this did not achieve statistical significance. Methodological quality of studies Table 4 summarizes the quality of the seven studies. This was poor, with three trials having more than 15 per cent loss to follow-up. It was unclear how randomization was carried out in four trials and whether treatment allocation was concealed adequately. Only two trials achieved allocation concealment and only three studies provided sample size calculations, although the Heymen 2007 trial performed an interim analysis that led to premature termination of enrolment42. None of the trials was blinded adequately and, although there are obvious difficulties in devising a comparable placebo for biofeedback, no trial discussed blinded assessment of outcomes (other than the study by Rao and colleagues41 in which manometry alone was assessed blind) or blinded data analysis. Table 4 Methodological quality of studies included in the review Reference . Year . Randomization . Allocation concealment . Sample size calculation . Blinding . Loss to follow-up . Bleijenberg et al.36 1994 B B No E A Koutsomanis et al.37 1995 A A Yes E A Glia et al.38 1997 A A No E D Heymen et al.39 1999 B B No E A Chiarioni et al.40 2006 A B Yes E A Rao et al.41 2007 B B No B† D Heymen et al.42 2007 B B Yes* B‡ D Reference . Year . Randomization . Allocation concealment . Sample size calculation . Blinding . Loss to follow-up . Bleijenberg et al.36 1994 B B No E A Koutsomanis et al.37 1995 A A Yes E A Glia et al.38 1997 A A No E D Heymen et al.39 1999 B B No E A Chiarioni et al.40 2006 A B Yes E A Rao et al.41 2007 B B No B† D Heymen et al.42 2007 B B Yes* B‡ D Randomization: A, appropriate; B, unclear; C, inappropriate. Allocation concealment: A, concealed; B, unclear; C, not concealed. Blinding: A, double blind; B, blinding of participants or investigators; C, blinding of analysis only; D, unclear; E, not blinded. Loss to follow-up: A, 5 per cent or less; B, 5·1–10·0 per cent; C, 10·1–15·0 per cent; D, more than 15 per cent; E, unclear. * An interim analysis was carried out which led to premature termination of enrolment. † Only manometry was assessed blind. ‡ There was partial blinding of participants who were allocated to medical treatment. Open in new tab Table 4 Methodological quality of studies included in the review Reference . Year . Randomization . Allocation concealment . Sample size calculation . Blinding . Loss to follow-up . Bleijenberg et al.36 1994 B B No E A Koutsomanis et al.37 1995 A A Yes E A Glia et al.38 1997 A A No E D Heymen et al.39 1999 B B No E A Chiarioni et al.40 2006 A B Yes E A Rao et al.41 2007 B B No B† D Heymen et al.42 2007 B B Yes* B‡ D Reference . Year . Randomization . Allocation concealment . Sample size calculation . Blinding . Loss to follow-up . Bleijenberg et al.36 1994 B B No E A Koutsomanis et al.37 1995 A A Yes E A Glia et al.38 1997 A A No E D Heymen et al.39 1999 B B No E A Chiarioni et al.40 2006 A B Yes E A Rao et al.41 2007 B B No B† D Heymen et al.42 2007 B B Yes* B‡ D Randomization: A, appropriate; B, unclear; C, inappropriate. Allocation concealment: A, concealed; B, unclear; C, not concealed. Blinding: A, double blind; B, blinding of participants or investigators; C, blinding of analysis only; D, unclear; E, not blinded. Loss to follow-up: A, 5 per cent or less; B, 5·1–10·0 per cent; C, 10·1–15·0 per cent; D, more than 15 per cent; E, unclear. * An interim analysis was carried out which led to premature termination of enrolment. † Only manometry was assessed blind. ‡ There was partial blinding of participants who were allocated to medical treatment. Open in new tab Heterogeneity of studies All seven included trials were heterogeneous in that they had varied inclusion and exclusion criteria, and different treatment protocols (Table 2). Although all trials required patients to have some features of PFD to be eligible for inclusion, only three studies38,40,42 defined PFD using the criteria set forth by the Rome Foundation. Koutsomanis and co-workers37 selected patients based on history alone whereas Rao et al.41 included patients with both slow-transit constipation and PFD. Two studies used a balloon expulsion test as an inclusion criterion but had different definitions of what constituted a prolonged balloon expulsion time, either more than 5 min40 or more than 1 min41. Although five trials used EMG to provide the necessary feedback, different methodologies were used. Bleijenberg and colleagues36, Chiarioni and co-workers40 and both Heymen trials39,42 used intra-anal acrylic plugs, whereas Koutsomanis et al.37 used adhesive EMG pads. Two trials used manometry for biofeedback, but one used solid-state manometry41 and the other water-perfused manometry38. The number of biofeedback sessions varied between one and ten. Duration of follow-up also varied widely, ranging between 3 and 12 months (Table 2). Biofeedback versus non-biofeedback Three trials compared biofeedback with non-biofeedback treatments40–42. Chiarioni and colleagues40 compared EMG biofeedback with laxatives in 109 patients who had failed conservative management and reported substantial improvement in the biofeedback group. Rao and co-workers41 compared manometry biofeedback with conservative management and sham feedback, and Heymen et al.42 compared EMG biofeedback with diazepam and a diazepam placebo (Table 2). These studies formed the basis for determining the effectiveness of biofeedback, with a meta-analysis showing an OR of 5·861 (95 per cent c.i. 2·175 to 15·794) in favour of biofeedback (P < 0·001) (Fig. 2). Fig. 2 Open in new tabDownload slide Forest plot of studies included in the meta-analysis of biofeedback (BF) versus non-biofeedback treatment. Odds ratios are shown with 95 per cent confidence intervals Comparing different biofeedback techniques Four trials compared different biofeedback techni- ques36–39. Bleijenberg and colleagues36 compared EMG biofeedback with balloon biofeedback, Glia and co-workers38 compared EMG biofeedback with manometry biofeedback, Koutsomanis et al.37 compared balloon training with balloon plus EMG feedback, whereas Heymen et al.39 compared four different types of biofeedback protocol (EMG feedback alone versus EMG feedback + balloon training versus EMG feedback + EMG home trainer versus EMG feedback + balloon training + EMG home trainer) (Table 2). Although an objective of this review was to determine the most effective biofeedback modality, given the heterogeneity of treatment protocols it was not possible to compare treatment effectiveness across different treatment modalities. Instead, a meta-analysis of the most widely used technique, EMG biofeedback, was carried out. Four trials compared EMG biofeedback with non-EMG biofeedback treatments and formed the basis of this meta-analysis. An OR of 6·738 (95 per cent c.i. 2·914 to 15·580) in favour EMG biofeedback was found (P < 0·001) (Fig. 3). Fig. 3 Open in new tabDownload slide Forest plot of studies included in the meta-analysis of electromyography (EMG) biofeedback (EMG BF) versus non-EMG BF. Odds ratios are shown with 95 per cent confidence intervals Discussion Only seven RCTs were included in the review and all had one or more methodological limitations. Meta-analysis of three studies comparing biofeedback with non-biofeedback methods suggested that biofeedback conferred a sixfold increase in likelihood of symptomatic relief. This result, however, must be interpreted with caution owing to the small number of trials, their heterogeneous nature and the methodological issues. Although it was intended to determine which biofeedback modality (balloon, manometry or EMG) was most effective, this was not possible because there were insufficient trials that compared these methods in different treatment arms. A meta-analysis of four trials of EMG biofeedback, the most widely used modality, demonstrated almost a sevenfold increase in the likelihood of symptomatic improvement compared with non-EMG biofeedback. Once again, this result must also be interpreted with caution because of the methodological and heterogeneity issues. The secondary outcomes were depression, anxiety and QOL changes as a result of treatment. Depression and anxiety before and after treatment was assessed in only one study36. Similarly, changes in QOL were assessed in only one trial42. This precluded any further analysis or synthesis of data across studies. Apart from the present study, there has been one meta-analysis25, one systematic review4 and several narrative reviews1,7,33–35 on this topic. The meta-analysis was performed in 1995 and 11 trials were reviewed, of which nine were uncontrolled studies, one was a non-randomized controlled trial43 and only one was a RCT36. The authors concluded that, although it is likely that biofeedback is effective in alleviating symptoms of constipation, the available evidence did not permit firm conclusions to be drawn25. It was also suggested that a strong placebo effect could account for the observed clinical benefit in the absence of well designed placebo-controlled trials25. A more recent systematic review published in 2002 assessed all controlled trials evaluating biofeedback in all gastrointestinal conditions4. However, this review included only one trial containing patients with constipation and so, once again, no definite conclusions could be drawn about the effectiveness of biofeedback in PFD. Despite the methodological limitations, the review suggested that biofeedback was probably more effective than non-biofeedback treatments, which is consistent with the findings of the present study. Although it is encouraging that more RCTs have been conducted in recent years, better designed trials are needed to provide a stronger evidence base for the effectiveness of biofeedback in the treatment of adults with PFD. Future studies should systematically compare different biofeedback modalities to identify the most effective modality. They must employ better randomization methods and study processes to achieve allocation concealment. Outcome measures must be standardized and outcome assessment blinded. These trials must also be adequately powered. Finally, assessment of the impact of treatment on psychopathology and QOL should not be forgotten. Acknowledgements C.E.K. acknowledges the Brian Smith Foundation, St Francis Xavier Cabrini Hospital and the University of Sydney (Colorectal Scholarship) for their support. No funding was received from any commercial sponsor for this systematic review. However, one of the original trials included in the review42 received sponsorship from Sandhill Scientific, Janssen Pharmaceuticals and Milan Pharmaceuticals. References 1 Chiarioni G , Heymen S, Whitehead WE. Biofeedback therapy for dyssynergic defecation . World J Gastroenterol 2006 ; 12 : 7069 – 7074 . Google Scholar Crossref Search ADS PubMed WorldCat 2 Wong P , Kadakia S. How to deal with chronic constipation. A stepwise method of establishing and treating the source of the problem . Postgrad Med 1999 ; 106 : 199 – 210 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Pfeifer J , Agachan F, Wexner SD. Surgery for constipation: a review . Dis Colon Rectum 1996 ; 39 : 444 – 460 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Coulter ID , Favreau JT, Hardy ML, Morton SC, Roth EA, Shekelle P. Biofeedback interventions for gastrointestinal conditions: a systematic review . Altern Ther Health Med 2002 ; 8 : 76 – 83 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 5 Wald A . Chronic constipation: advances in management . Neurogastroenterol Motil 2007 ; 19 : 4 – 10 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Rao SS , Enck P, Loening-Baucke V. Biofeedback therapy for defecation disorders . Dig Dis 1997 ; 15 ( Suppl 1 ): 78 – 92 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Heymen S , Jones KR, Scarlett Y, Whitehead WE. Biofeedback treatment of constipation: a critical review . Dis Colon Rectum 2003 ; 46 : 1208 – 1217 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Battaglia E , Serra AM, Buonafede G, Dughera L, Chistolini F, Morelli A et al. Long-term study on the effects of visual biofeedback and muscle training as a therapeutic modality in pelvic floor dyssynergia and slow transit constipation . Dis Colon Rectum 2004 ; 47 : 90 – 95 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Martelli H , Devroede G, Arhan P, Duguay C. Mechanisms of idiopathic constipation: outlet obstruction . Gastroenterology 1978 ; 75 : 623 – 631 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Preston DM , Lennard-Jones JE. Anismus in chronic constipation . Dig Dis Sci 1985 ; 30 : 413 – 418 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Rome Foundation . Rome III Diagnostic Questionnaires. Constipation Module. http://www.romecriteria.org/pdfs/ConstMode.pdf [accessed 1 November 2007]. 12 Tsiaoussis J , Chrysos E, Glynos M, Vassilakis JS, Xynos E. Pathophysiology and treatment of anterior rectal mucosal prolapse syndrome . Br J Surg 1998 ; 85 : 1699 – 1702 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Burnett C , Whitehead WE, Drossman D. Psychological distress and impaired quality of life in patients with functional anorectal disorders . Gastroenterology 1998 ; 114 ( Suppl 1 ): A729 (Abstract). Google Scholar Crossref Search ADS WorldCat 14 Mason HJ , Serrano-Ikkos E, Kamm MA. Psychological state and quality of life in patients having behavioral treatment (biofeedback) for intractable constipation . Am J Gastroenterol 2002 ; 97 : 3154 – 3159 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Nehra V , Bruce BK, Rath-Harvey DM, Pemberton JH, Camilleri M. Psychological disorders in patients with evacuation disorders and constipation in a tertiary practice . Am J Gastroenterol 2000 ; 95 : 1755 – 1758 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Pinho M , Yoshioka Y, Keighley MR. Long term results of anorectal myectomy for chronic constipation . Br J Surg 1989 ; 76 : 1163 – 1164 . Google Scholar Crossref Search ADS PubMed WorldCat 17 Barnes P , Hawley PR, Preston DM, Lennard-Jones JE. Experience of posterior division of the puborectalis muscle in the management of chronic constipation . Br J Surg 1985 ; 72 : 475 – 477 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Keighley M , Shouler P. Outlet syndrome: is there a surgical option? J R Soc Med 1984 ; 77 : 559 – 563 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Wasserman I . Puborectalis syndrome: rectal stenosis due to anorectal spasm . Dis Colon Rectum 1964 ; 7 : 87 – 98 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Hallan RI , Melling J, Womack NR, Williams NS, Waldron DJ, Morrison JFB. Treatment of anismus in intractable constipation with botulinum A toxin . Lancet 1988 ; 332 : 714 – 717 . Google Scholar Crossref Search ADS WorldCat 21 Maria G , Cadeddu F, Brandara F, Marniga G, Brisinda G. Experience with type A botulinum toxin for treatment of outlet-type constipation . Am J Gastroenterol 2006 ; 101 : 2570 – 2575 . Google Scholar Crossref Search ADS PubMed WorldCat 22 Maria G , Brisinda G, Bentivoglio AR, Cassetta E, Albanese A. Botulinum toxin in the treatment of outlet obstruction constipation caused by puborectalis syndrome . Dis Colon Rectum 2000 ; 43 : 376 – 380 . Google Scholar Crossref Search ADS PubMed WorldCat 23 El-Tawil AM . Reasons for creation of permanent ileostomy for the management of idiopathic chronic constipation . J Gastroenterol Hepatol 2004 ; 19 : 844 – 846 . Google Scholar Crossref Search ADS PubMed WorldCat 24 Keighley M . Surgical management of chronic constipation . Colorectal Dis 1999 ; 1 : 186 – 191 . Google Scholar Crossref Search ADS PubMed WorldCat 25 Ernst E , Resch KL. A meta-analysis of biofeedback treatment for anismus . Eur J Phys Med Rehabil 1995 ; 5 : 157 – 159 . Google Scholar OpenURL Placeholder Text WorldCat 26 Bleijenberg G , Kuijpers H. Treatment of the spastic pelvic floor syndrome with biofeedback . Dis Colon Rectum 1987 ; 30 : 108 – 111 . Google Scholar Crossref Search ADS PubMed WorldCat 27 Dahl JA , Lindquist BL, Tysk C, Leissner P, Phillipson L, Jarnerot G. Behavioral medicine treatment in chronic constipation with paradoxical anal sphincter contraction . Dis Colon Rectum 1991 ; 34 : 769 – 776 . Google Scholar Crossref Search ADS PubMed WorldCat 28 Lestar B , Penninckx F, Kerremans R. Biofeedback defaecation training for anismus . Int J Colorect Dis 1991 ; 6 : 202 – 207 . Google Scholar Crossref Search ADS WorldCat 29 Kawimbe B , Papachrysostomou M, Binnie NR, Clare N, Smith AN. Outlet obstruction constipation (anismus) managed by biofeedback . Gut 1991 ; 32 : 1175 – 1179 . Google Scholar Crossref Search ADS PubMed WorldCat 30 Wexner SD , Cheape JD, Jorge JM, Heymen S, Jagelman DG. Prospective assesment of biofeedback for the treatment of paradoxical puborectalis contraction . Dis Colon Rectum 1992 ; 35 : 145 – 150 . Google Scholar Crossref Search ADS PubMed WorldCat 31 Fleshman JW , Dreznik Z, Meyer K, Fry RD, Carney R, Kodner IJ. Outpatient protocol for biofeedback therapy of pelvic floor outlet obstruction . Dis Colon Rectum 1992 ; 35 : 1 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat 32 Karlbom U , Halden M, Eeg-Olosson KE, Pahlman L, Graf W. Results of biofeedback in constipated patients: a prospective study . Dis Colon Rectum 1997 ; 40 : 1049 – 1155 . Google Scholar Crossref Search ADS WorldCat 33 Enck P . Biofeedback training in disordered defecation. A critical review . Dig Dis Sci 1993 ; 38 : 1953 – 1960 . Google Scholar Crossref Search ADS PubMed WorldCat 34 Jorge JM , Habr-Gama A, Wexner SD. Biofeedback therapy in the colon and rectal practice . Appl Psychophysiol Biofeedback 2003 ; 28 : 47 – 61 . Google Scholar Crossref Search ADS PubMed WorldCat 35 Palsson OS , Heymen S, Whitehead WE. Biofeedback treatment for functional anorectal disorders: a comprehensive efficacy review . Appl Psychophysiol Biofeedback 2004 ; 29 : 153 – 174 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Bleijenberg G , Kuijpers HC. Biofeedback treatment of constipation: a comparison of two methods . Am J Gastroenterol 1994 ; 89 : 1021 – 1026 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 37 Koutsomanis D , Lennard-Jones JE, Roy AJ, Kamm MA. Controlled randomised trial of visual biofeedback versus muscle training without a visual display for intractable constipation . Gut 1995 ; 37 : 95 – 99 . Google Scholar Crossref Search ADS PubMed WorldCat 38 Glia A , Gylin M, Gullberg K, Lindberg G. Biofeedback retraining in patients with functional constipation and paradoxical puborectalis contraction: comparison of anal manometry and sphincter electro- myography for feedback . Dis Colon Rectum 1997 ; 40 : 889 – 895 . Google Scholar Crossref Search ADS PubMed WorldCat 39 Heymen S , Wexner SD, Vickers D, Nogueras JJ, Weiss EG, Pikarsky AJ. Prospective, randomized trial comparing biofeedback techniques for patients with constipation . Dis Colon Rectum 1999 ; 42 : 1388 – 1393 . Google Scholar Crossref Search ADS PubMed WorldCat 40 Chiarioni G , Whitehead WE, Pezza V, Morelli A, Bassotti G. Biofeedback is superior to laxatives for normal transit constipation due to plevic floor dyssynergia . Gastroenterology 2006 ; 130 : 657 – 664 . Google Scholar Crossref Search ADS PubMed WorldCat 41 Rao SS , Seaton K, Miller M, Brown K, Nygaard I, Stumbo P et al. Randomized controlled trial of biofeedback, sham feedback and standard therapy for dyssynergic defecation . Clin Gastroenterol Hepatol 2007 ; 5 : 331 – 338 . Google Scholar Crossref Search ADS PubMed WorldCat 42 Heymen S , Scarlett Y, Jones K, Ringel Y, Drossman D, Whitehead WE. Randomized, controlled trial shows biofeedback to be superior to alternative treatments for patients with pelvic floor dyssynergia type constipation . Dis Colon Rectum 2007 ; 50 : 428 – 441 . Google Scholar Crossref Search ADS PubMed WorldCat 43 Cox DJ , Sutphen J, Borowitz S, Dickens MN, Singles J, Whitehead WE. Simple electromyographic biofeedback treatment for chronic pediatric constipation/encopresis: preliminary report . Appl Psychophysiol Biofeedback 1994 ; 19 : 41 – 50 . Google Scholar OpenURL Placeholder Text WorldCat Copyright © 2008 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) Copyright © 2008 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd.

Gurusamy, K; Aggarwal, R; Palanivelu, L; Davidson, B R

doi: 10.1002/bjs.6344pmid: 18690637

Abstract Background Surgical training has traditionally been one of apprenticeship. The aim of this review was to determine whether virtual reality (VR) training can supplement and/or replace conventional laparoscopic training in surgical trainees with limited or no laparoscopic experience. Methods Randomized clinical trials addressing this issue were identified from The Cochrane Library trials register, Medline, Embase, Science Citation Index Expanded, grey literature and reference lists. Standardized mean difference was calculated with 95 per cent confidence intervals based on available case analysis. Results Twenty-three trials (mostly with a high risk of bias) involving 622 participants were included in this review. In trainees without surgical experience, VR training decreased the time taken to complete a task, increased accuracy and decreased errors compared with no training. In the same participants, VR training was more accurate than video trainer (VT) training. In participants with limited laparoscopic experience, VR training resulted in a greater reduction in operating time, error and unnecessary movements than standard laparoscopic training. In these participants, the composite performance score was better in the VR group than the VT group. Conclusion VR training can supplement standard laparoscopic surgical training. It is at least as effective as video training in supplementing standard laparoscopic training. Background Surgical training has traditionally been one of apprenticeship, where the surgical trainee learns to perform surgery under the supervision of a trained surgeon. Different procedures have different learning curves1–3. Surgeons experienced in one procedure may not be experienced in another, and results for individual procedures improve with experience1–3. The different methods of laparoscopic surgical training include live animal training, human and animal cadaver training, training using a box trainer (also called a video trainer, VT), and virtual reality (VR) training (training using computer simulation)4. VT is currently being used widely for laparoscopic training and has been shown to be better than standard training5. VR training has been reported to improve learning outcomes in different surgical procedures4,6–9. It also offers an ethical way of assessing the competency of a surgeon in performing a procedure, without risk to a patient10. Other reports suggest that VR training alone is inferior to traditional training for certain procedures11. VR training has been used mainly for development of component skills (such as diathermy, clipping, suturing) and not training of an entire procedure. In contrast to the limited variability of data available during an aeroplane flight on which a pilot is trained using a custom designed simulator, anatomical variations are common in the human body12–14, and skills acquired on a single computer simulation programme may not be applicable in patients11. Although simulators can be expensive, traditional training is not without its costs. Junior surgeons take longer over operations than senior surgeons15–18. Bridges and Diamond19 reported that the average costs of this increased operating time were about US $ 12 000 dollars per year per resident during 1993–1997. The complication rate is also higher for junior surgeons17,18, but the cost of complications was not included in the analysis of Bridges and Diamond19. Thus, the cost of the VR training system has to be balanced against the cost of increased operating time and complication rates during traditional surgical training. There have been previous reviews of VR training in surgery20–23. None of these included a search of educational databases or computer literature databases. None separated the effectiveness of VR training in individuals who have never had any experience in laparoscopic surgery and those who have already performed laparoscopic surgery under supervision. This review included a methodical search of all the relevant medical, educational and computer literature databases including the grey literature24, and included randomized trials that evaluated the effectiveness of VR training. Methods Trial selection and data extraction Only randomized clinical trials (irrespective of language, blinding or publication status) that assessed the effectiveness of VR training compared with video training or no training or standard laparoscopic training (SLT) (irrespective of generic skills or procedure-specific skills) were included in the review. The Cochrane Hepato-Biliary Group Controlled Trials Register, the Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library, Medline, Embase and Science Citation Index Expanded24 were searched until March 2008 using MeSH terms (Therapy, Computer-Assisted OR Surgery, Computer-Assisted OR Computer-Assisted Instruction), free-text terms (virtual real* OR simulat*) combined with the terms ‘train’ and ‘laparoscopy’ (MeSH term and equivalent free-text search terms). A filter for identifying the randomized controlled trials recommended by The Cochrane Collaboration25 was used to filter out non-randomized studies in Medline and Embase. In addition to the above databases, SCOPUS, education indexes (ERIC—Education Resources Information Center, Australian Education Index, British Education Index), computer databases (ACM Digital Library, collection of Computer Science Bibliographies, IEEE/IEE Electronic Library), Meta-register of Controlled Trials and Google Scholar were searched. Grey literature, abstracts and conference proceedings were identified by searching the ISI proceedings, Zetoc General Search, ‘index to thesis’, GrayLIT Network (replaced by Science.gov), abstracts from AMED (Allied and Complementary Medicine Database), biological abstracts, biotechnology abstracts, British Nursing Index, CINAHL database, dissertation abstracts (American institutions) and SIGLE (grey literature in EC member countries and held at The British Library). References of the identified trials were also searched to identify further relevant trials. Two reviewers independently identified the trials for inclusion and extracted the data related to first author, year of publication, inclusion and exclusion criteria, participant characteristics, details of training such as software used, tasks in training, duration of training, outcomes of interest, methodological quality of the studies (without masking the study names), whether intention-to-treat analysis was followed and whether sample size calculations were reported. Outcomes Data for the following outcomes were extracted: patient or animal mortality or morbidity (reported separately in trials where both patients and animals were used for assessing outcome measures); time taken to perform the evaluation task on the simulation model (after training); operating time (if the evaluation task was performed on humans or animals) (after training); error score: the number of undesirable movements, for example, injury to gallbladder or burning non-target tissue8; accuracy; improvement in task performance (objective or subjective); and participant satisfaction. Definitions or scales used by the authors were accepted for the various outcomes. Assessment of methodological quality The instructions given in the Cochrane Handbook for Systematic Reviews of Intervention25 and the Cochrane Hepato-Biliary Group module26 were followed. Owing to the risk overestimation of intervention effects in randomized trials with inadequate methodological quality27–29, the influence of methodological quality on the results was assessed by evaluating the reported randomization and follow-up procedures in each trial. If information was not available in the published trial, the authors were contacted in order to assess the trials correctly. Generation of allocation sequence, allocation concealment, blinding and follow-up were examined. Because participants and trainers cannot be blinded, double blinding was not possible but outcome assessor blinding was feasible and trials were considered to have adequate blinding if outcome assessors were blinded. Trials were considered to be of low risk of bias if the above four methodological qualities were adequate. Statistical methods Meta-analyses were done according to the recommendations of The Cochrane Collaboration25 and the Cochrane Hepato-Biliary Group module26 using the software package Revman 4·230. No dichotomous variables were reported. For the continuous variables, the standardized mean difference (SMD) with confidence interval (c.i.) was calculated (as different authors used different tasks for evaluating the outcomes). These were calculated irrespective of the number of trials included under each outcome in order to obtain a uniform effect estimate. The number of trials and the participants included in the meta-analysis of each outcome has been stated to make it clear that the effect estimate is obtained from two or more trials. A random-effects model31 and a fixed-effect model32 were used when there were two or more trials. In case of discrepancy between the two models, both results were reported; otherwise results of the fixed model were reported. To allow for dropouts and withdrawals between randomization and intervention (or control), the available case analysis25 was adopted. The degree of heterogeneity was measured by I2 value33 and an I2 value over 50 per cent was considered to demonstrate statistical heterogeneity25. Description of studies A total of 2176 references were identified through electronic searches of the Cochrane Hepato-Biliary Group Controlled Trials Register and the Cochrane Central Register of Controlled Trials in The Cochrane Library (n = 43), Medline (n = 96), Embase (n = 72), Science Citation Index Expanded (n = 89) and other databases (n = 1876). One thousand and twenty-two duplicates and 1103 clearly irrelevant references identified through reading abstracts were excluded. The remaining 51 references were retrieved for further assessment. No references were identified through scanning reference lists of the identified randomized trials. It was not possible to identify details concerning one trial identified from the trial register34. No information was available from the authors of this trial, so it was excluded from the review. Twenty-six other references were excluded as they did not meet the inclusion criteria. In total, 23 trials (24 references) were included (Fig. 1). Four trials compared VR and VT training35–38; 12 trials compared VR and no training or SLT6–8,39–48; four trials compared VR training, VT training and no training4,49–51; and three trials compared different methods of VR training52–54. Six trials compared VR training in surgical trainees with limited experience in laparoscopic surgery6,8,35,41,44,46. One trial did not state the experience of the participants53. The other trials included medical students or surgical residents without any experience in laparoscopic surgery. Three trials that had adequate methodological quality in all four components were considered to have a low risk of bias43,45,49. Details regarding the tasks in training and endpoint of training are recorded in Appendix 1 (supporting information). Fig. 1 Open in new tabDownload slide Flow chart of articles identified, included and excluded. *One randomized controlled trial (RCT) had two references. VT, video trainer; VR, virtual reality Outcomes A total of 23 trials involving 622 participants were included in this review. As mentioned in the protocol55, the results were reported separately for trainees with no laparoscopic experience and those with limited laparoscopic experience (such as performing laparoscopic procedures under supervision). The results are summarized in Appendices 2–5 (supporting information). One trial mentioned patient-related outcomes44. None of the remaining trials that used animals or humans to assess the outcomes reported animal or patient morbidity or mortality. No laparoscopic experience Virtual reality versus video trainer training Time taken to perform the job or speed with which the job was completed The meta-analysis showed that there was no statistically significant difference in the time taken to complete the task (SMD − 0·06, 95 per cent c.i. − 0·40 to 0·52), reduction in the time taken to complete the task after compared with before training (SMD − 0·61, 95 per cent c.i. − 1·51 to 0·29), speed in performing the tasks between VR and VT (SMD − 0·33, 95 per cent c.i. − 0·95 to 0·28), or in the change in speed in either hand in performing the task. Error score There was no difference in error scores between the two groups. The error score was assessed by dropping the object, perforation of the object, or by composite error score from the computer. Accuracy The meta-analysis demonstrated statistically significantly better accuracy in the VR group than VT group (SMD 0·68 (95 per cent c.i. 0·05 to 1·31)). Another trial that could not be included in the meta-analysis had a similar result56. Composite score There was no difference in composite score between the two groups in the two trials that reported this outcome4,49. Movements There was a trend for a shorter distance and a smaller number of movements in the VR group than the VT group for both hands. Another trial reported no difference in the economy of movement between the two groups4. Virtual reality versus no training Time taken to perform the job The meta-analysis showed that the VR group performed the tasks more quickly than the group without training (SMD − 1·09 (95 per cent c.i. − 1·50 to − 0·68)). Meta-analysis of the reduction in time between first (before training) and second (after training in the VR group) assessments showed that the reduction was greater in the VR group (SMD − 0·25 (95 per cent c.i. − 0·85 to 0·34)). This difference was not statistically significant. There was no statistically significant difference in the speed or in the improvement in speed between the two groups. Error score All three trials that reported this outcome showed a lower error score in the VR group4,42,49. This was not statistically significant in two trials4,49 and the statistical significance in the third trial42 was not known. The error score was assessed by dropping the object, perforation or tear of the object, slack ligature, or by composite error score from the computer. Accuracy The only trial that reported this outcome showed statistically significantly better accuracy in the VR group40. Composite scores The composite scores in an animal model were reported in three trials7,39,45. One of these trials also reported the composite score in a VR simulator45. There was a trend towards increased composite scores in the VR group compared with the no-training group in the animal model (SMD 0·50 (95 per cent c.i. 0·00 to 0·99)). However, another trial which showed statistically significantly better scores in the VR group could not be included in the review49. In the only trial that reported the composite score in a VR simulator, the VR group had a statistically significantly higher composite score than the no-training group (SMD 1·38 (95 per cent c.i. 0·60 to 2·16))45. Movements There was a statistically significant decrease in the number of movements43,51 and a trend towards a decrease in the distance of movement of the right hand in the VR group compared with the no-training group51. The left hand showed a trend towards a decreased number of movements and distance of movements51. Another trial showed a statistically significant reduction in path tracing of the instruments in the VR group42. Path tracing for individual hands was not reported. Another trial reported no difference in the economy of movement between the two groups4. Limited laparoscopic experience Virtual reality versus video trainer training Composite operative score The VR group had a significantly better composite operative score than the VT group. Only the VR group improved significantly after training. The improvement in composite operative score was also higher in the VR group, although this was not statistically significant35. Improvement in performance on inanimate models Both groups improved significantly on both models. The VR group improved significantly more than the VT group on the VR model. The VT group improved significantly more than the VR group on the VT model35. Supplementary virtual reality versus standard laparoscopic training Patient-oriented outcomes The only patient-oriented outcome reported in all the trials included in this review was the rate conversion to open cholecystectomy, reported in one trial44. In this trial, 20 laparoscopic cholecystectomies performed by seven surgical trainees in the VR group were compared with 17 laparoscopic cholecystectomies performed by six surgical trainees in the SLT group. There was no statistically significant difference in the rate ratio for conversion between the groups (rate ratio 0·12, 95 per cent c.i. 0·01 to 2·43). Operating time All three trials that reported on the duration of operation noted a shorter time in the VR group than the SLT group. This was statistically significant in two trials. There was no statistically significant difference in the time taken to clip and divide the cystic duct or cystic artery in one trial that assessed only this aspect of the procedure46. Error score All five trials that reported this outcome described a statistically significantly lower error score in the VR group than the SLT group6,8,41,44,46. The error score was assessed by two observers based on handling of tissue resulting in tissue damage and keeping the instruments in sight. Composite score The composite score for clipping and division of cystic artery or cystic duct during laparoscopic cholecystectomy in humans was higher in the VR group than the SLT group in the only trial that reported on this outcome46. Economy of movements There was improvement in the economy of movement (avoiding unnecessary movement), but only in the VR group6. Different methods of virtual reality training There were no trials comparing one software package and another. One trial53 compared different durations of VR training. This trial found that the composite score of error, time and path length economy was better when VR training was performed over 5 min with an interval of 2·5 min between the training sessions. Another trial52 compared VR training at easy and difficult levels. The assessment was performed at the same level as the training. The speed was better and blood loss was less at the easy level (statistically significant) than the difficult level. Another trial compared the addition of surgeon's knot tying module in addition to the basic modules54. This trial showed that the time taken to tie a surgeon's knot in a porcine model was shorter for those who underwent training in this module than for those who did not. There was no difference in the time taken to drive the needle through the tissue. The objective error in the additionally trained group was also less, although there was no statistically significant difference in the subjective error score between the two groups. Statistical notes Adopting a random-effects model did not alter the results, although there was statistically significant heterogeneity for some outcomes. Subgroup analysis of trials with a low risk of bias and use of a funnel plot to explore bias57,58 could not be performed because of the lack of trials with a low risk of bias and because few trials had data suitable for meta-analysis under each outcome. Discussion Laparoscopic surgery is different from open surgery because of an increased need for hand–eye co-ordination to perform tasks via a video screen rather than under direct vision, and because of the increased need for manual dexterity to compensate for the use of long instruments, which can amplify any error in movement. There is also a fulcrum effect of the body wall: when the surgeon moves his hand to the patient's right, the operating end of the instrument moves to the patient's left on the monitor40. Tissues must be handled carefully to compensate for the lack of touch sensation and there is also a lack of three-dimensional images. VR training is one of the many methods of laparoscopic surgical training and is currently aimed at improving psychomotor skills40. An increasing number of procedures are being performed laparoscopically. With the decreasing time available to train surgeons as a result of the European Working Time Directive59 and the Modernizing Medical Careers initiative of the Department of Health60, training is necessarily structured to improve surgical skills in the least time with the maximum efficiency. This is applicable to surgical trainees with no experience in laparoscopic surgery and in those who have started their laparoscopic career but have not achieved proficiency. Because of the shortened working hours, the trainees may be exposed to fewer surgical procedures. Thus it is more difficult to develop generic skills, such as suturing or cutting, and also procedure-specific skills, such as cannulating the common bile duct. The trials included in this review mainly assessed the effect of VR training on the development or improvement of generic skills. The evaluation of VR training came from various models. Understandably, the evaluation task (such as laparoscopic cholecystectomy) was performed in humans only in trials in which the participants had some experience in performing laparoscopic surgery under supervision. So, some trials used a porcine model. Although the anatomy is different, the handling of human and porcine tissues is similar. Both live and cadaver porcine tissues are used for training in many courses61–63. Some trials evaluated the tasks using a VT and others used VR models. Both the VT64,65 and VR66 models can discriminate between people with differing surgical capability. In people with no laparoscopic experience, VR training is better than no training in relation to the time taken to perform a task, improving accuracy and decreasing error. Various methods of training were used in the trials but the benefits were consistent. Although only a trend could be shown in certain outcomes, some trials that showed the benefit of VR training could not be included in the meta-analysis. For young surgeons at the beginning of their laparoscopic training, VR training reduced the operating time, error and unnecessary movements during laparoscopic cholecystectomy. No other laparoscopic surgery was evaluated in the clinical trials included, although all laparoscopic operations were included in this review. Thus there is convincing evidence that VR training is a useful supplement to SLT in laparoscopic cholecystectomy in surgical residents with limited laparoscopic experience. However, it must be noted that trainees have to spend extra time on the VR training, which may have to come out of their overall training. Whether it is cost-effective to replace a session of their current schedule with VR training remains unclear. The benefit of VR training over video training is not clear. VR training provides greater accuracy in making incisions and in placing sutures for trainees with no laparoscopic experience. The only trial that assessed the overall operative performance in surgical residents with limited laparoscopic experience after undergoing VR or VT training showed that the composite operative score for laparoscopic cholecystectomy was higher after VR training. There was no comparison between different VR models. A trial that compared different durations of training found that short bursts of training with short intervals in between (5 min followed by 2·5 min rest) produced better results. The models used to evaluate trainee performance were VTs, pigs or humans. In spite of the different models used, and the lack of similarity between the training and evaluation tasks, the VR group performed consistently better in one or more of the measured parameters. This is because of ‘transfer of training’, the effect that the practice of one task has on the learning or performance of a second36. Trainees have different learning curves for different tasks. The time and number of repetitions to achieve proficiency may vary67. A few recent trials43,44,46,54 have used predefined criteria or benchmark levels, rather than repetitions or a fixed time of training. It has been suggested that it is the achievement of these benchmark levels, rather than time taken for training, that determines the completion of training68,69. Using such predefined criteria may help in assessment of the effectiveness of VR training after accounting for intertrainee variability. None of the trials assessed VR training as part of a surgical training curriculum. Hence further trials are necessary to assess the potential role, duration and scope of such training. Only one of the trials used patient-oriented outcomes44. Most were at high risk of bias. Trials with a low risk of bias that use patient-oriented outcomes after VR training as a part of a surgical training curriculum are awaited. Such trials should include cost-effectiveness as one of the outcomes. VR training without haptic feedback is not realistic70. The degree of fidelity or realism does not alter the effectiveness in training71. This might explain the success of VR training despite it being a low-fidelity model. However, improving the fidelity by haptic feedback may increase trainee satisfaction and the enthusiasm to learn on VR models. The advantages of VR over VT training are not as evident as for VR over standard training. However, potential disadvantages of VT training include: an assistant is needed for two-handed tasks; in VTs with a fixed video camera, a distance has to be chosen so that the task can be viewed closely; the introduction of instruments cannot be followed; and VT training requires a trainer. Recent advances in VR technology have made it possible to import images into VR software from external sources72. Three-dimensional images can be constructed if the x, y and z coordinates and colour information of each pixel is available73. In the near future, it might be possible to import these three-dimensional images into VR software. It should then be possible to manipulate the images and train individuals using three-dimensional reconstructions of actual operations rather than train in component skills. Training on numerous such models with anatomical variations could also help with the improvement of decision-making and procedure-specific skills. Supporting information Supporting information may be found in the online version of this article. Acknowledgements This paper is a substantially shortened version of a Cochrane review55 submitted to the Cochrane Hepato-Biliary Group. The authors thank the Cochrane Hepato-Biliary Group for their support in the protocol leading to its publication and for their advice on methodological quality issues related to trials included in this review. Cochrane reviews are regularly updated as new evidence emerges and in response to comments and criticisms, and The Cochrane Library should be consulted for the most recent version of the review. The results of a Cochrane review can be interpreted differently, depending on people's perspectives and circumstances. Please consider the conclusions presented carefully. They are the opinions of the review authors, and are not necessarily shared by The Cochrane Collaboration. References 1 Herrell SD , Smith JA. Robotic-assisted laparoscopic prostatectomy: what is the learning curve? Urology 2005 ; 66 ( Suppl 1 ): 105 – 107 . Google Scholar Crossref Search ADS PubMed WorldCat 2 Tekkis PP , Fazio VW, Lavery IC, Remzi FH, Senagore AJ, Wu JS et al. Evaluation of the learning curve in ileal pouch–anal anastomosis surgery . Ann Surg 2005 ; 241 : 262 – 268 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Tekkis PP , Senagore AJ, Delaney CP, Fazio VW. Evaluation of the learning curve in laparoscopic colorectal surgery: comparison of right-sided and left-sided resections . Ann Surg 2005 ; 242 : 83 – 91 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Munz Y , Kumar BD, Moorthy K, Bann S, Darzi A. Laparoscopic virtual reality and box trainers: is one superior to the other? Surg Endosc 2004 ; 18 : 485 – 494 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Scott DJ , Bergen PC, Rege RV, Laycock R, Tesfay ST, Valentine RJ et al. Laparoscopic training on bench models: better and more cost effective than operating room experience? J Am Coll Surg 2000 ; 191 : 272 – 283 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Grantcharov TP , Kristiansen VB, Bendix J, Bardram L, Rosenberg J, Funch-Jensen P. Randomized clinical trial of virtual reality simulation for laparoscopic skills training . Br J Surg 2004 ; 91 : 146 – 150 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Hyltander A , Liljegren E, Rhodin PH, Lonroth H. The transfer of basic skills learned in a laparoscopic simulator to the operating room . Surg Endosc 2002 ; 16 : 1324 – 1328 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Seymour NE , Gallagher AG, Roman SA, O'Brien MK, Bansal VK, Andersen DK et al. Virtual reality training improves operating room performance: results of a randomized, double-blinded study . Ann Surg 2002 ; 236 : 458 – 463 ; discussion 463–464. Google Scholar Crossref Search ADS PubMed WorldCat 9 Watterson JD , Beiko DT, Kuan JK, Denstedt JD. Randomized prospective blinded study validating acquistion of ureteroscopy skills using computer based virtual reality endourological simulator . J Urol 2002 ; 168 : 1928 – 1932 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Moorthy K , Munz Y, Jiwanji M, Bann S, Chang A, Darzi A. Validity and reliability of a virtual reality upper gastrointestinal simulator and cross validation using structured assessment of individual performance with video playback . Surg Endosc 2004 ; 18 : 328 – 333 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 11 Gerson LB , Van Dam J. A prospective randomized trial comparing a virtual reality simulator to bedside teaching for training in sigmoidoscopy . Endoscopy 2003 ; 35 : 569 – 575 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 12 Heloury Y , Leborgne J, Rogez JM, Robert R, Lehur PA, Pannier M et al. Radiological anatomy of the bile ducts based on intraoperative investigation in 250 cases . Anat Clin 1985 ; 7 : 93 – 102 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Izuishi K , Toyama Y, Nakano S, Goda F, Usuki H, Masaki T et al. Preoperative assessment of the aberrant bile duct using multislice computed tomography cholangiography . Am J Surg 2005 ; 189 : 53 – 55 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Lamah M , Dickson GH. Congenital anatomical abnormalities of the extrahepatic biliary duct: a personal audit . Surg Radiol Anat 1999 ; 21 : 325 – 327 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Babineau TJ , Becker J, Gibbons G, Sentovich S, Hess D, Robertson S et al. The ‘cost’ of operative training for surgical residents . Arch Surg 2004 ; 139 : 366 – 369 ; discussion 369–370. Google Scholar Crossref Search ADS PubMed WorldCat 16 Farnworth LR , Lemay DE, Wooldridge T, Mabrey JD, Blaschak MJ, DeCoster TA et al. A comparison of operative times in arthroscopic ACL reconstruction between orthopaedic faculty and residents: the financial impact of orthopaedic surgical training in the operating room . Iowa Orthop J 2001 ; 21 : 31 – 35 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 17 Kauvar DS , Braswell A, Brown BD, Harnisch M. Influence of resident and attending surgeon seniority on operative performance in laparoscopic cholecystectomy . J Surg Res 2006 ; 132 : 159 – 163 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Wilkiemeyer M , Pappas TN, Giobbie-Hurder A, Itani KM, Jonasson O, Neumayer LA. Does resident post graduate year influence the outcomes of inguinal hernia repair? Ann Surg 2005 ; 241 : 879 – 882 ; discussion 882–884. Google Scholar Crossref Search ADS PubMed WorldCat 19 Bridges M , Diamond DL. The financial impact of teaching surgical residents in the operating room . Am J Surg 1999 ; 177 : 28 – 32 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Carter FJ , Schijven MP, Aggarwal R, Grantcharov T, Francis NK, Hanna GB et al. Consensus guidelines for validation of virtual reality surgical simulators . Surg Endosc 2005 ; 19 : 1523 – 1532 . Google Scholar Crossref Search ADS PubMed WorldCat 21 Gaba DM . The future vision of simulation in health care . Qual Saf Health Care 2004 ; 13 ( Suppl ): i2 – i10 . Google Scholar Crossref Search ADS PubMed WorldCat 22 Haque S , Srinivasan S. A meta-analysis of the training effectiveness of virtual reality surgical simulators . IEEE Trans Inf Technol Biomed 2006 ; 10 : 51 – 58 . Google Scholar Crossref Search ADS PubMed WorldCat 23 Sutherland LM , Middleton PF, Anthony A, Hamdorf J, Cregan P, Scott D et al. Surgical simulation: a systematic review . Ann Surg 2006 ; 243 : 291 – 300 . Google Scholar Crossref Search ADS PubMed WorldCat 24 Royle P , Milne R. Literature searching for randomized controlled trials used in Cochrane reviews: rapid versus exhaustive searches . Int J Technol Assess Health Care 2003 ; 19 : 591 – 603 . Google Scholar Crossref Search ADS PubMed WorldCat 25 Higgins JPT , Green S (eds). Cochrane Handbook for Systematic Reviews of Intervention 4·2·6 [updated September 2006]. The Cochrane Library, Issue 4. John Wiley : Chichester , 2006 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 26 Gluud C , Als-Nielsen B, D'Amico G, Fingerhut A, Gluud LL, Khan S et al. Cochrane Hepato-Biliary Group . About The Cochrane Collaboration (Cochrane Review Groups (CRGs)) . Issue 2. Art. no.: LIVER. John Wiley : Chichester , 2007 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 27 Kjaergard LL , Villumsen J, Gluud C. Reported methodologic quality and discrepancies between large and small randomized trials in meta-analyses . Ann Intern Med 2001 ; 135 : 982 – 989 . Google Scholar Crossref Search ADS PubMed WorldCat 28 Moher D , Pham B, Jones A, Cook DJ, Jadad AR, Moher M et al. Does quality of reports of randomised trials affect estimates of intervention efficacy reported in meta-analyses? Lancet 1998 ; 352 : 609 – 613 . Google Scholar Crossref Search ADS PubMed WorldCat 29 Schulz KF , Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias. Dimensions of methodological quality associated with estimates of treatment effects in controlled trials . JAMA 1995 ; 273 : 408 – 412 . Google Scholar Crossref Search ADS PubMed WorldCat 30 Review Manager (RevMan). Version 4.2 for Windows . The Nordic Cochrane Centre, The Cochrane Collaboration : Copenhagen , 2003 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 31 DerSimonian R , Laird N. Meta-analysis in clinical trials . Control Clin Trials 1986 ; 7 : 177 – 188 . Google Scholar Crossref Search ADS PubMed WorldCat 32 DeMets DL . Methods for combining randomized clinical trials: strengths and limitations . Stat Med 1987 ; 6 : 341 – 350 . Google Scholar Crossref Search ADS PubMed WorldCat 33 Higgins JP , Thompson SG. Quantifying heterogeneity in a meta-analysis . Stat Med 2002 ; 21 : 1539 – 1558 . Google Scholar Crossref Search ADS PubMed WorldCat 34 Buerger T . Comparative study of the effectiveness of training on a Simbionix simulator and training on a box trainer simulator when learning laparoscopic cholecystectomy . 2006 ; http://www.controlled-trials.com/ISRCTN84287327/84287327 [accessed 3 April 2008]. 35 Hamilton EC , Scott DJ, Fleming JB, Rege RV, Laycock R, Bergen PC et al. Comparison of video trainer and virtual reality training systems on acquisition of laparoscopic skills . Surg Endosc 2002 ; 16 : 406 – 411 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Jordan JA , Gallagher AG, McGuigan J, McGlade K, McClure N. A comparison between randomly alternating imaging, normal laparoscopic imaging, and virtual reality training in laparoscopic psychomotor skill acquisition . Am J Surg 2000 ; 180 : 208 – 211 . Google Scholar Crossref Search ADS PubMed WorldCat 37 Kothari SN , Kaplan BJ, DeMaria EJ, Broderick TJ, Merrell RC. Training in laparoscopic suturing skills using a new computer-based virtual reality simulator (MIST-VR) provides results comparable to those with an established pelvic trainer system . J Laparoendosc Adv Surg Tech A 2002 ; 12 : 167 – 173 . Google Scholar Crossref Search ADS PubMed WorldCat 38 Xia Q , Hobarth G, Yanik M, Bretthauer G, Uranus S. Virtual reality training in minimally invasive surgery. In EAES: Proceedings of the 8th International Congress of the European Association for Endoscopic Surgery, Nice , Mouiel J, Mortori A (eds). Monduzzi Editore : Bologna , 2000 ; 547 – 554 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 39 Ahlberg G , Heikkinen T, Iselius L, Leijonmarck CE, Rutqvist J, Arvidsson D. Does training in a virtual reality simulator improve surgical performance? Surg Endosc 2002 ; 16 : 126 – 129 . Google Scholar Crossref Search ADS PubMed WorldCat 40 Gallagher AG , McClure N, McGuigan J, Crothers I, Browning J. Virtual reality training in laparoscopic surgery: a preliminary assessment of minimally invasive surgical trainer virtual reality (MIST VR) . Endoscopy 1999 ; 31 : 310 – 313 . Google Scholar Crossref Search ADS PubMed WorldCat 41 McClusky DA , Gallagher AG, Ritter EM, Lederman AB, Van Sickle KR, Baghai M et al. Virtual reality training improves junior residents' operating room performance: results of a prospective, randomized, double-blinded study of the complete laparoscopic cholecystectomy . J Am Coll Surg 2004 ; 199 ( Suppl 1 ): S73 . Google Scholar Crossref Search ADS WorldCat 42 Tanoue K , Yasunaga T, Konishi K, Okazaki K, Ieiri S, Kawabe Y et al. Effectiveness of training for endoscopic surgery using a simulator with virtual reality: randomized study . Cars 2005: International Congress Series 2005 ; 1281 : 515 – 520 . Google Scholar OpenURL Placeholder Text WorldCat 43 Aggarwal R , Ward J, Balasundaram I, Sains P, Athanasiou T, Darzi A. Proving the effectiveness of virtual reality simulation for training in laparoscopic surgery . Ann Surg 2007 ; 246 : 771 – 779 . Google Scholar Crossref Search ADS PubMed WorldCat 44 Ahlberg G , Enochsson L, Gallagher AG, Hedman L, Hogman C, McClusky DA et al. Proficiency-based virtual reality training significantly reduces the error rate for residents during their first 10 laparoscopic cholecystectomies . Am J Surg 2007 ; 193 : 797 – 804 . Google Scholar Crossref Search ADS PubMed WorldCat 45 Bensalah K , Lucas SM, Zeltser IS, Tuncel A, Jenkins A, Pearle MS et al. Training on the virtual reality laparoscopic simulator improves laparoscopic performance in a virtual enviroment: a randomized controlled trial . J Endourol 2007 ; 21 ( Suppl 1 ): A137 – A138 (Abstract). Google Scholar OpenURL Placeholder Text WorldCat 46 Cosman PH , Hugh TJ, Shearer CJ, Merrett ND, Biankin AV, Cartmill JA. Skills acquired on virtual reality laparoscopic simulators transfer into the operating room in a blinded, randomised, controlled trial . Stud Health Technol Inform 2007 ; 125 : 76 – 81 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 47 Ganai S , Donroe JA, St Louis MR, Lewis GM, Seymour NE. Virtual-reality training improves angled telescope skills in novice laparoscopists . Am J Surg 2007 ; 193 : 260 – 265 . Google Scholar Crossref Search ADS PubMed WorldCat 48 Zeltser IS , Bensalah K, Tuncel A, Lucas S, Jenkins A, Pearle MS et al. Training on the virtual reality laparoscopic simulator improves performance of an unfamiliar live surgical laparoscopic procedure: a randomized, controlled trial . J Endourol 2007 ; 21 ( Suppl 1 ): A137 (Abstract). Google Scholar Crossref Search ADS WorldCat 49 Madan AK , Frantzides CT. Prospective randomized controlled trial of laparoscopic trainers for basic laparoscopic skills acquisition . Surg Endosc 2007 ; 21 : 209 – 213 . Google Scholar Crossref Search ADS PubMed WorldCat 50 Pearson AM , Gallagher AG, Rosser JC, Satava RM. Evaluation of structured and quantitative training methods for teaching intracorporeal knot tying . Surg Endosc 2002 ; 16 : 130 – 137 . Google Scholar Crossref Search ADS PubMed WorldCat 51 Torkington J , Smith SG, Rees BI, Darzi A. Skill transfer from virtual reality to a real laparoscopic task . Surg Endosc 2001 ; 15 : 1076 – 1079 . Google Scholar Crossref Search ADS PubMed WorldCat 52 Hassan I , Alkhawaga M, Gerdes B, Langer P, Kress R, Rothmund M et al. Training of laparoscopic skills with virtual reality simulator: a critical reappraisal of the learning curve . Cars 2005: International Congress Series 2005 ; 1281 : 1370 . Google Scholar OpenURL Placeholder Text WorldCat 53 Mackay S , Morgan P, Datta V, Chang A, Darzi A. Practice distribution in procedural skills training: a randomized controlled trial . Surg Endosc 2002 ; 16 : 957 – 961 . Google Scholar Crossref Search ADS PubMed WorldCat 54 Verdaasdonk EG , Dankelman J, Lange JF, Stassen LP. Transfer validity of laparoscopic knot-tying training on a VR simulator to a realistic environment: a randomized controlled trial . Surg Endosc 2008 ; 22 : 1636 – 1642 . Google Scholar Crossref Search ADS PubMed WorldCat 55 Gurusamy KS , Aggarwal R, Palanivelu L, Davidson BR. Virtual reality training for surgical trainees in laparoscopic surgery . Cochrane Database Syst Rev 2007 ; ( 3 ): CD006575 . Google Scholar OpenURL Placeholder Text WorldCat 56 Jordan JA , Gallagher AG, McGuigan J, McClure N. Virtual reality training leads to faster adaptation to the novel psychomotor restrictions encountered by laparoscopic surgeons . Surg Endosc 2001 ; 15 : 1080 – 1084 . Google Scholar Crossref Search ADS PubMed WorldCat 57 Egger M , Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test . BMJ 1997 ; 315 : 629 – 634 . Google Scholar Crossref Search ADS PubMed WorldCat 58 Macaskill P , Walter SD, Irwig L. A comparison of methods to detect publication bias in meta-analysis . Stat Med 2001 ; 20 : 641 – 654 . Google Scholar Crossref Search ADS PubMed WorldCat 59 Chikwe J , de Souza AC, Pepper JR. No time to train the surgeons . BMJ 2004 ; 328 : 418 – 419 . Google Scholar Crossref Search ADS PubMed WorldCat 60 Payne SR , Shaw MB. What impact will shortened training have on urological service delivery? Ann R Coll Surg Engl 2005 ; 87 : 373 – 378 . Google Scholar Crossref Search ADS PubMed WorldCat 61 Gaarder C , Naess PA, Buanes T, Pillgram-Larsen J. Advanced surgical trauma care training with a live porcine model . Injury 2005 ; 36 : 718 – 724 . Google Scholar Crossref Search ADS PubMed WorldCat 62 Lentz GM , Mandel LS, Goff BA. A six-year study of surgical teaching and skills evaluation for obstetric/gynecologic residents in porcine and inanimate surgical models . Am J Obstet Gynecol 2005 ; 193 : 2056 – 2061 . Google Scholar Crossref Search ADS PubMed WorldCat 63 Smith AM , Forder JA, Annapureddy SR, Reddy KS, Amis AA. The porcine forelimb as a model for human flexor tendon surgery . J Hand Surg [Br] 2005 ; 30 : 307 – 309 . Google Scholar Crossref Search ADS PubMed WorldCat 64 Avgerinos DV , Goodell KH, Waxberg S, Cao CG, Schwaitzberg SD. Comparison of the sensitivity of physical and virtual laparoscopic surgical training simulators to the user's level of experience . Surg Endosc 2005 ; 19 : 1211 – 1215 . Google Scholar Crossref Search ADS PubMed WorldCat 65 Fraser SA , Klassen DR, Feldman LS, Ghitulescu GA, Stanbridge D, Fried GM. Evaluating laparoscopic skills: setting the pass/fail score for the MISTELS system . Surg Endosc 2003 ; 17 : 964 – 967 . Google Scholar Crossref Search ADS PubMed WorldCat 66 Gallagher AG , Satava RM. Virtual reality as a metric for the assessment of laparoscopic psychomotor skills. Learning curves and reliability measures . Surg Endosc 2002 ; 16 : 1746 – 1752 . Google Scholar Crossref Search ADS PubMed WorldCat 67 Brunner WC , Korndorffer JR, Sierra R, Massarweh NN, Dunne JB, Yaul CL et al. Laparoscopic virtual reality training: are 30 repetitions enough? J Surg Res 2004 ; 122 : 150 – 156 . Google Scholar Crossref Search ADS PubMed WorldCat 68 Aggarwal R , Grantcharov T, Moorthy K, Hance J, Darzi A. A competency-based virtual reality training curriculum for the acquisition of laparoscopic psychomotor skill . Am J Surg 2006 ; 191 : 128 – 133 . Google Scholar Crossref Search ADS PubMed WorldCat 69 Aggarwal R , Grantcharov TP, Eriksen JR, Blirup D, Kristiansen VB, Funch-Jensen P et al. An evi- dence-based virtual reality training program for novice laparoscopic surgeons . Ann Surg 2006 ; 244 : 310 – 314 . Google Scholar Crossref Search ADS PubMed WorldCat 70 Madan AK , Frantzides CT, Tebbit C, Quiros RM. Participants' opinions of laparoscopic training devices after a basic laparoscopic training course . Am J Surg 2005 ; 189 : 758 – 761 . Google Scholar Crossref Search ADS PubMed WorldCat 71 Grober ED , Hamstra SJ, Wanzel KR, Reznick RK, Matsumoto ED, Sidhu RS et al. The educational impact of bench model fidelity on the acquisition of technical skill: the use of clinically relevant outcome measures . Ann Surg 2004 ; 240 : 374 – 381 . Google Scholar Crossref Search ADS PubMed WorldCat 72 Jaselskis E , Andrle S, Anderson-Wilk M, Bernard JE, Bryden M, Vance JM. Virtual reality and laser scanning applications. 3D engineering solutions from Iowa State University . http://www.ctre.iastate.edu/pubs/vrls.pdf [accessed 25 July 2006]. 73 Cyberware . Head & Face Color 3D Scanner . http://www.cyberware.com/products/scanners/ps.html [accessed 25 July 2006]. Copyright © 2008 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) Copyright © 2008 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd.

Hubner, M; Demartines, N; Muller, S; Dindo, D; Clavien, P-A; Hahnloser, D

doi: 10.1002/bjs.6321pmid: 18690630

Abstract Background Many instruments are used for laparoscopic dissection, including monopolar electrosurgery scissors (MES), electrothermal bipolar vessel sealers (BVS) and ultrasonically coagulating shears (UCS). These three devices were compared with regard to dissection time, blood loss, safety and costs. Methods Sixty-one consecutive patients undergoing laparoscopic left-sided colectomy were randomized to MES, BVS or UCS. The primary endpoint was dissection time. Results Patient and operation characteristics did not differ between the groups. Median dissection time was significantly shorter with BVS (105 min) and UCS (90 min) than with MES (137 min) (P < 0·001). With BVS and UCS, significantly fewer additional clips were required (MES 9 versus BVS 0 versus UCS 3; P < 0·001) and there was a trend towards lower blood loss (125 versus 50 versus 50 ml respectively; P = 0·223) and a reduced volume of suction fluid (425 versus 80 versus 110 ml; P = 0·058). Overall satisfaction was similar for the three instruments. Dissection with BVS and UCS was significantly cheaper than with MES, assuming a centre volume of 200 cases per year (P = 0·009). Conclusion BVS and UCS shorten dissection time in laparoscopic left-sided colectomy and are cost-effective compared with MES. Registration number: NCT00517608 (http://www.clinicaltrials.com). Introduction Laparoscopic colorectal surgery has gained wide acceptance for benign and oncological indications1–3. Conventional monopolar electrosurgery scissors (MES) have several shortcomings in this type of surgery, including the risk of thermal injury, difficult haemostasis and smoke production, necessitating the use of additional tools such as bipolar graspers, sutures and clips4. To overcome these problems and to reduce the number of instrument changes, trocars and operating time, several multifunctional tools have been developed. The most popular devices are electrothermal bipolar vessel sealers (BVS)5,6 and ultrasonically coagulating shears (UCS)6–8. Both of these instruments are well established in laparoscopic surgery9–14, but their practicability and cost-effectiveness compared with conventional methods remain undetermined as findings from randomized trials and cohort studies published to date are controversial, and comprehensive cost analyses are lacking9–13,15. The aim of this prospective randomized study was to compare MES, BVS and UCS in laparoscopic colorectal surgery with regard to dissection time, blood loss, technical aspects, surgeon comfort and costs. Methods All consecutive patients admitted for elective laparoscopic left-sided colorectal resection were evaluated for entry into the study. Eligibility criteria included age above 18 years, reasonable communication ability and signed informed consent. Patients requiring a right-sided or total colectomy were excluded. Previous laparotomy was not an exclusion criterion. Sixty-one consecutive eligible patients were randomized to undergo laparoscopic colorectal resection using one of the three dissection devices (no financial support was received from the manufacturers). The procedure was approved by the institutional ethics committee. Randomization was performed by sealed envelopes on the day before surgery to allow nursing staff to prepare the allocated instruments. When using MES (Endo Shears™ 5 mm, Covidien, Mansfield, Massachusetts, USA; trigger switch and cord, Valleylab, Boulder, Colorado, USA), electrical energy is applied between the shears and a reference electrode is placed distantly on the body surface. The advantage of sharp dissection is outweighed by high heat production with thermal spread in surrounding structures and charring. Potential risks are direct coupling to another metal instrument, direct sparking and the passage of current from recently coagulated, electrically isolated tissue4. With the BVS (LigaSure™ 5 mm; Valleylab), high current and low voltage results in the denaturation of collagen and elastin components within the vessel wall and surrounding tissue. According to the manufacturer, secure sealing of vessels for up to 7 mm can be obtained. Thermal spread is minimal compared with conventional electrocautery. The instrument can be used as coagulator, dissector and grasper, thus reducing instrument traffic5,6. For ultrasonic dissection with UCS (Harmonic ACE™; Ethicon Endo-Surgery, Cincinnati, Ohio, USA), piezoelectric transducers transform electrical energy at the functional tip into frictional energy (movement of the tips at 55 kHz), allowing cutting and coagulation of vessels for up to 3–5 mm. Additionally, tissue dissection is eased by a cavitational effect. No surgical smoke is generated during ultrasonic dissection, but visibility may be hindered by dispersed non-viable tissue particles (‘storm’ effect)6–8. Surgical procedures All operations were performed or supervised by two expert laparoscopic colorectal surgeons (D.H., N.D.), who had each performed more than 200 laparoscopic colectomies using any of the three devices. No surgeon had a preference for a particular device. The operative technique was standardized before starting the protocol and differed only with regard to the technology allocated for dissection and haemostasis. For evaluation purposes the colorectal dissection was divided into three phases, each with three or four steps. Phase A involved dissection of the inferior mesenteric vessels, identification of the left ureter, and transection of the inferior mesenteric vessels. Phase B comprised mobilization of the sigmoid colon, opening of the presacral space (preserving the nerves), presacral mobilization (if necessary), and dissection of the distal margin; it ended with laparoscopic stapling and transection of the colon/rectum. Phase C consisted of lateral and medial mobilization of the descending colon, mobilization of the splenic flexure and great omentum (if necessary), and ended with the complete mobilization of the descending colon. The specimen was retrieved through a 4-cm Pfannenstiel incision, and resected. The end-to-end-anastomosis was performed laparoscopically using a circular stapling device. Outcome measures Dissection time (phases A–C), which served as the primary endpoint of the study, was documented by the operation nurses, together with the amount of suction and rinsing fluid used immediately after each phase. At the end of the operation, the surgeon documented the use of the allocated instrument and the need for additional devices in each phase and each step. Intraoperative blood loss was estimated by the surgeon based on his subjective appraisal; perioperative blood transfusions and the difference in haematocrit (value before surgery minus that on the second postoperative day) were recorded. Frequency of lens cleaning and the time-related carbon dioxide consumption were measured as surrogate parameters for the generation of surgical smoke and vapour respectively. Postoperative complications were assessed for up to 30 days after surgery using a validated five-point scale classification16. The operating surgeon graded the practicability of each instrument using a visual analogue scale (VAS) from 0 (poor) to 10 (outstanding) with regard to main features such as dissection capacity, sealing, cutting, management of bleeding, handling and overall satisfaction. Twelve surgeons were equally involved in the evaluation process. Cost analysis Actual costs were calculated in euros as described previously17, including operating theatre time, costs of the allocated device and for the additional instruments used for haemostasis or dissection. Briefly, device-related costs comprised a capital charge (5-year depreciation) for the generator, maintenance charges and disposals. Costs were calculated for four different centre volumes (20, 50, 100 and 200 laparoscopic colorectal resections per year) using the median values for the sum (material costs, costs for operating theatre time and for additional instruments) of each case in each group. Operating theatre costs were €7·77 per min; one red blood cell unit was €127·00. Plastic clips (Hemo-o-lok®; Teleflex® Medical, Research Triangle Park, North Carolina, USA; €4·06 for six clips) or titan clips (Soma Medical, Feusisberg, Switzerland; €66·50 for ten clips) were used for ligation of large vessels and haemostasis. Costs were provided by the institutional accounting department and reflect the actual values for the year 2007. Statistical analysis Descriptive statistics are expressed as median (range). Comparison of continuous variables between the three groups was performed with the Kruskal–Wallis test. The χ2 test was employed for comparison of categorical variables. According to the Bonferroni adjustment for three groups analysed, P < 0·017 was considered to indicate statistical significance. The sample size calculation was based both on the authors' experience with the three devices and on published findings10,13,14. The standard deviation for dissection time was assumed to be 45 min, and a 30-min reduction in dissection time (phases A–C) was considered to be clinically relevant. To find this difference in at least one of the groups with a level of statistical significance of 0·017 (according to the Bonferroni adjustment) and a power of 0·80, calculations included 20 patients in each group. Statistical analysis was performed using standard software package SPSS® version 14.0 (SPSS, Chicago, Illinois, USA). Results Between August 2005 and December 2006, 70 consecutive patients undergoing laparoscopic left-sided colectomy were evaluated. Seven patients refused to participate and two did not meet the inclusion criteria (Fig. 1). The remaining 61 patients (25 men), of median age 62 (range 33–84) years, were randomized to one of the three study groups. No patient was excluded subsequently. Fig. 1 Open in new tabDownload slide Study flow chart of patients admitted for elective laparoscopic colonic surgery. MES, monopolar electrosurgery scissors; BVS, electrothermal bipolar vessel sealer; UCS, ultrasonically coagulating shears The three groups were similar with regard to age, body mass index, sex ratio, American Society of Anesthesiologists grade, co-morbidity (as assessed by the Charlson index18), indication for surgery and type of operation (Table 1). Table 1 Patient demographics and surgical characteristics . MES (n = 20) . BVS (n = 21) . UCS (n = 20) . P . Sex ratio (F : M) 11 : 9 9 : 12 10 : 10 0·736† Age (years)* 62 (33–80) 62 (42–84) 60 (44–80) 0·709‡ ASA grade 0·052†  1 or 2 13 14 16  3 or 4 7 7 4 Charlson score18* 1 (0–6) 2 (0–6) 1 (0–5) 0·440‡ BMI (kg/m2)* 26 (20–40) 26 (16–34) 26 (19–35) 0·674‡ Type of lesion 0·896†  Malignant 9 9 10  Benign 11 12 10 Type of resection 0·871†  Rectosigmoid 15 17 15  Low anterior 5 4 5 Mobilization of splenic flexure 0·926†  Yes 16 16 15  No 4 5 5 . MES (n = 20) . BVS (n = 21) . UCS (n = 20) . P . Sex ratio (F : M) 11 : 9 9 : 12 10 : 10 0·736† Age (years)* 62 (33–80) 62 (42–84) 60 (44–80) 0·709‡ ASA grade 0·052†  1 or 2 13 14 16  3 or 4 7 7 4 Charlson score18* 1 (0–6) 2 (0–6) 1 (0–5) 0·440‡ BMI (kg/m2)* 26 (20–40) 26 (16–34) 26 (19–35) 0·674‡ Type of lesion 0·896†  Malignant 9 9 10  Benign 11 12 10 Type of resection 0·871†  Rectosigmoid 15 17 15  Low anterior 5 4 5 Mobilization of splenic flexure 0·926†  Yes 16 16 15  No 4 5 5 * Values are median (range). MES, monopolar electrosurgery scissors; BVS, electrothermal bipolar vessel sealer; UCS, ultrasonically coagulating shears; ASA, American Society of Anesthesiologists; BMI, body mass index. † χ2 test; ‡ Kruskal–Wallis test. Open in new tab Table 1 Patient demographics and surgical characteristics . MES (n = 20) . BVS (n = 21) . UCS (n = 20) . P . Sex ratio (F : M) 11 : 9 9 : 12 10 : 10 0·736† Age (years)* 62 (33–80) 62 (42–84) 60 (44–80) 0·709‡ ASA grade 0·052†  1 or 2 13 14 16  3 or 4 7 7 4 Charlson score18* 1 (0–6) 2 (0–6) 1 (0–5) 0·440‡ BMI (kg/m2)* 26 (20–40) 26 (16–34) 26 (19–35) 0·674‡ Type of lesion 0·896†  Malignant 9 9 10  Benign 11 12 10 Type of resection 0·871†  Rectosigmoid 15 17 15  Low anterior 5 4 5 Mobilization of splenic flexure 0·926†  Yes 16 16 15  No 4 5 5 . MES (n = 20) . BVS (n = 21) . UCS (n = 20) . P . Sex ratio (F : M) 11 : 9 9 : 12 10 : 10 0·736† Age (years)* 62 (33–80) 62 (42–84) 60 (44–80) 0·709‡ ASA grade 0·052†  1 or 2 13 14 16  3 or 4 7 7 4 Charlson score18* 1 (0–6) 2 (0–6) 1 (0–5) 0·440‡ BMI (kg/m2)* 26 (20–40) 26 (16–34) 26 (19–35) 0·674‡ Type of lesion 0·896†  Malignant 9 9 10  Benign 11 12 10 Type of resection 0·871†  Rectosigmoid 15 17 15  Low anterior 5 4 5 Mobilization of splenic flexure 0·926†  Yes 16 16 15  No 4 5 5 * Values are median (range). MES, monopolar electrosurgery scissors; BVS, electrothermal bipolar vessel sealer; UCS, ultrasonically coagulating shears; ASA, American Society of Anesthesiologists; BMI, body mass index. † χ2 test; ‡ Kruskal–Wallis test. Open in new tab Operative parameters Primary and secondary outcomes are shown in Table 2. The median dissection time was significantly shorter with BVS or UCS than with MES (P < 0·001), due mainly to faster transection of mesenteric fat or omentum (phase C). Similarly, significantly more clips were required in the MES group than in the other two groups (P < 0·001). With BVS and UCS, intraoperative blood loss was lower and less suction fluid was collected than in the MES group, although not significantly so. Relative carbon dioxide consumption was lowest in the BVS group, and frequency of lens cleaning did not differ between the groups (data not shown). No patient received intraoperative or postoperative blood transfusions. Table 2 Operative parameters related to the dissection device used . MES (n = 20) . BVS (n = 21) . UCS (n = 20) . P . Dissection time (min) 137 (65–230) 105 (58–195) 90 (45–177) < 0·001¶  Phase A 30·5 (11–110) 27 (10–95) 22 (10–80) 0·148¶  Phase B 55·5 (20–155) 42 (16–80) 38·5 (15–132) 0·205¶  Phase C 36 (14–95) 23 (6–60) 23 (10–55) 0·022¶ Instrument use (0–11)* 9·5 (1–11) 9 (6–11) 9 (8–11) 0·777¶ Other instrument used† 6 3 5 0·472# No. of sutures or clips 9 (4–28) 0 (0–3) 3 (0–11) < 0·001¶ Estimated blood loss (ml) 125 (0–450) 50 (0–600) 50 (0–500) 0·223¶ Haematocrit (%)‡ 5·5 (0–11) 5 (0–11) 4·5 (0–11) 0·684¶ Suction fluid (ml) 425 (0–1600) 80 (0–8200) 110 (0–1400) 0·058¶ Carbon dioxide/time (l/min)§ 1·18 (0–2·48) 0·54 (0–2·60) 0·99 (0·34–1·74) 0·042¶ . MES (n = 20) . BVS (n = 21) . UCS (n = 20) . P . Dissection time (min) 137 (65–230) 105 (58–195) 90 (45–177) < 0·001¶  Phase A 30·5 (11–110) 27 (10–95) 22 (10–80) 0·148¶  Phase B 55·5 (20–155) 42 (16–80) 38·5 (15–132) 0·205¶  Phase C 36 (14–95) 23 (6–60) 23 (10–55) 0·022¶ Instrument use (0–11)* 9·5 (1–11) 9 (6–11) 9 (8–11) 0·777¶ Other instrument used† 6 3 5 0·472# No. of sutures or clips 9 (4–28) 0 (0–3) 3 (0–11) < 0·001¶ Estimated blood loss (ml) 125 (0–450) 50 (0–600) 50 (0–500) 0·223¶ Haematocrit (%)‡ 5·5 (0–11) 5 (0–11) 4·5 (0–11) 0·684¶ Suction fluid (ml) 425 (0–1600) 80 (0–8200) 110 (0–1400) 0·058¶ Carbon dioxide/time (l/min)§ 1·18 (0–2·48) 0·54 (0–2·60) 0·99 (0·34–1·74) 0·042¶ Values are median (range). * Number of predefined operation steps (of a total of 11) performed using the allocated instrument. † Number of operations where an instrument additional to the allocated one was used. ‡ Haematocrit difference (before minus postoperative day 2). § Time-related carbon dioxide consumption (surrogate parameter for smoke generation). MES, monopolar electrosurgery scissors; BVS, electrothermal bipolar vessel sealer; UCS, ultrasonically coagulating shears ¶ Kruskal–Wallis test; # χ2 test. Open in new tab Table 2 Operative parameters related to the dissection device used . MES (n = 20) . BVS (n = 21) . UCS (n = 20) . P . Dissection time (min) 137 (65–230) 105 (58–195) 90 (45–177) < 0·001¶  Phase A 30·5 (11–110) 27 (10–95) 22 (10–80) 0·148¶  Phase B 55·5 (20–155) 42 (16–80) 38·5 (15–132) 0·205¶  Phase C 36 (14–95) 23 (6–60) 23 (10–55) 0·022¶ Instrument use (0–11)* 9·5 (1–11) 9 (6–11) 9 (8–11) 0·777¶ Other instrument used† 6 3 5 0·472# No. of sutures or clips 9 (4–28) 0 (0–3) 3 (0–11) < 0·001¶ Estimated blood loss (ml) 125 (0–450) 50 (0–600) 50 (0–500) 0·223¶ Haematocrit (%)‡ 5·5 (0–11) 5 (0–11) 4·5 (0–11) 0·684¶ Suction fluid (ml) 425 (0–1600) 80 (0–8200) 110 (0–1400) 0·058¶ Carbon dioxide/time (l/min)§ 1·18 (0–2·48) 0·54 (0–2·60) 0·99 (0·34–1·74) 0·042¶ . MES (n = 20) . BVS (n = 21) . UCS (n = 20) . P . Dissection time (min) 137 (65–230) 105 (58–195) 90 (45–177) < 0·001¶  Phase A 30·5 (11–110) 27 (10–95) 22 (10–80) 0·148¶  Phase B 55·5 (20–155) 42 (16–80) 38·5 (15–132) 0·205¶  Phase C 36 (14–95) 23 (6–60) 23 (10–55) 0·022¶ Instrument use (0–11)* 9·5 (1–11) 9 (6–11) 9 (8–11) 0·777¶ Other instrument used† 6 3 5 0·472# No. of sutures or clips 9 (4–28) 0 (0–3) 3 (0–11) < 0·001¶ Estimated blood loss (ml) 125 (0–450) 50 (0–600) 50 (0–500) 0·223¶ Haematocrit (%)‡ 5·5 (0–11) 5 (0–11) 4·5 (0–11) 0·684¶ Suction fluid (ml) 425 (0–1600) 80 (0–8200) 110 (0–1400) 0·058¶ Carbon dioxide/time (l/min)§ 1·18 (0–2·48) 0·54 (0–2·60) 0·99 (0·34–1·74) 0·042¶ Values are median (range). * Number of predefined operation steps (of a total of 11) performed using the allocated instrument. † Number of operations where an instrument additional to the allocated one was used. ‡ Haematocrit difference (before minus postoperative day 2). § Time-related carbon dioxide consumption (surrogate parameter for smoke generation). MES, monopolar electrosurgery scissors; BVS, electrothermal bipolar vessel sealer; UCS, ultrasonically coagulating shears ¶ Kruskal–Wallis test; # χ2 test. Open in new tab Complications and hospital stay Twenty-six of the 61 patients had at least one postoperative complication (MES, ten; BVS, ten; UCS, six; P = 0·375) (Table 3). Eight patients had complications that required an intervention or stay in the intensive care unit (grade 3 and 4 according to the classification of Dindo et al.16). Two (one each with MES and BVS) were postoperative episodes of ventricular arrhythmia in patients with cardiopathy who needed Intensive care (grade 4a). The remaining six surgical complications were two anastomotic leaks (grade 3b, grade 4a) in the BVS group, one iatrogenic small bowel lesion (grade 3b) and one recurrent colovesical fistula (grade 3b) in a cachectic patient in the MES group, one stoma necrosis following a Hartmann operation (grade 3b; BVS) and one infected intra-abdominal hematoma requiring relaparoscopy (grade 3b, BVS). The overall median (range) hospital stay was 8 (3–31) days, and was similar in the three groups (P = 0·446). Table 3 Postoperative complications and hospital stay . MES (n = 20) . BVS (n = 21) . UCS (n = 20) . P . No. with complications† 10 10 6 0·375‡  Grade 1 3 4 4  Grade 2 4 1 2  Grade 3 2 3 0  Grade 4 1 2 0  Grade 5 0 0 0 Hospital stay (days)* 8·5 (5–29) 7 (3–31) 7 (3–28) 0·446§ . MES (n = 20) . BVS (n = 21) . UCS (n = 20) . P . No. with complications† 10 10 6 0·375‡  Grade 1 3 4 4  Grade 2 4 1 2  Grade 3 2 3 0  Grade 4 1 2 0  Grade 5 0 0 0 Hospital stay (days)* 8·5 (5–29) 7 (3–31) 7 (3–28) 0·446§ * Values are median (range). † According to a validated five-scale classification system16: grade 1, any deviation from normal postoperative course with no need for specific drugs or intervention; grade 2, requiring specific pharmacological treatment; grade 3, requiring surgical, endoscopic or radiological intervention; grade 4, life-threatening complications requiring intensive care; grade 5, death from complication. MES, monopolar electrosurgery scissors; BVS, electrothermal bipolar vessel sealer; UCS, ultrasonically coagulating shears. ‡ χ2 test; § Kruskal–Wallis test. Open in new tab Table 3 Postoperative complications and hospital stay . MES (n = 20) . BVS (n = 21) . UCS (n = 20) . P . No. with complications† 10 10 6 0·375‡  Grade 1 3 4 4  Grade 2 4 1 2  Grade 3 2 3 0  Grade 4 1 2 0  Grade 5 0 0 0 Hospital stay (days)* 8·5 (5–29) 7 (3–31) 7 (3–28) 0·446§ . MES (n = 20) . BVS (n = 21) . UCS (n = 20) . P . No. with complications† 10 10 6 0·375‡  Grade 1 3 4 4  Grade 2 4 1 2  Grade 3 2 3 0  Grade 4 1 2 0  Grade 5 0 0 0 Hospital stay (days)* 8·5 (5–29) 7 (3–31) 7 (3–28) 0·446§ * Values are median (range). † According to a validated five-scale classification system16: grade 1, any deviation from normal postoperative course with no need for specific drugs or intervention; grade 2, requiring specific pharmacological treatment; grade 3, requiring surgical, endoscopic or radiological intervention; grade 4, life-threatening complications requiring intensive care; grade 5, death from complication. MES, monopolar electrosurgery scissors; BVS, electrothermal bipolar vessel sealer; UCS, ultrasonically coagulating shears. ‡ χ2 test; § Kruskal–Wallis test. Open in new tab Influence of the device on costs The factors impacting on overall costs are shown in Table 4. Lower material-related costs in the MES group were outweighed by higher costs for longer operating time and the increased need for additional material. A comprehensive cost analysis that gives the median price per patient according to the device employed is dependent on the centre volume (Table 4). Colonic dissection by BVS and UCS was less expensive than with MES when a centre volume of 200 patients per year was assumed (P = 0·009). Table 4 Material-related costs per patient for each device . MES . BVS . UCS . Instrument costs according to centre volume (€)*  20 cases/year 180 549 620  50 cases/year 180 429 499  100 cases/year 180 389 435  200 cases/year 180 369 403 Theatre time costs (€) 1045 (505–1786) 812 (450–1515) 699 (350–1375) Additional costs (€)† 102 (4–635) 0 (0–127) 13 (0–131) . MES . BVS . UCS . Instrument costs according to centre volume (€)*  20 cases/year 180 549 620  50 cases/year 180 429 499  100 cases/year 180 389 435  200 cases/year 180 369 403 Theatre time costs (€) 1045 (505–1786) 812 (450–1515) 699 (350–1375) Additional costs (€)† 102 (4–635) 0 (0–127) 13 (0–131) Values are median (range). * Calculated on the basis of capital charge for generator (5-year depreciation), and costs for maintenance and disposal of material. † For additional instruments, clips and blood transfusions. MES, monopolar electrosurgery scissors; BVS, electrothermal bipolar vessel sealer; UCS, ultrasonically coagulating shears. Open in new tab Table 4 Material-related costs per patient for each device . MES . BVS . UCS . Instrument costs according to centre volume (€)*  20 cases/year 180 549 620  50 cases/year 180 429 499  100 cases/year 180 389 435  200 cases/year 180 369 403 Theatre time costs (€) 1045 (505–1786) 812 (450–1515) 699 (350–1375) Additional costs (€)† 102 (4–635) 0 (0–127) 13 (0–131) . MES . BVS . UCS . Instrument costs according to centre volume (€)*  20 cases/year 180 549 620  50 cases/year 180 429 499  100 cases/year 180 389 435  200 cases/year 180 369 403 Theatre time costs (€) 1045 (505–1786) 812 (450–1515) 699 (350–1375) Additional costs (€)† 102 (4–635) 0 (0–127) 13 (0–131) Values are median (range). * Calculated on the basis of capital charge for generator (5-year depreciation), and costs for maintenance and disposal of material. † For additional instruments, clips and blood transfusions. MES, monopolar electrosurgery scissors; BVS, electrothermal bipolar vessel sealer; UCS, ultrasonically coagulating shears. Open in new tab Table 5 Comprehensive costs per patient for each device, analysed by centre volume Centre volume (cases/year) . MES . BVS . UCS . P* . 20 1382 (752–2368) 1364 (999–2190) 1323 (969–2003) 0·610 50 1382 (752–2368) 1243 (880–2071) 1202 (849–1882) 0·089 100 1382 (752–2368) 1204 (840–2031) 1138 (784–1818) 0·021 200 1382 (752–2368) 1185 (820–2011) 1105 (752–1786) 0·009 Centre volume (cases/year) . MES . BVS . UCS . P* . 20 1382 (752–2368) 1364 (999–2190) 1323 (969–2003) 0·610 50 1382 (752–2368) 1243 (880–2071) 1202 (849–1882) 0·089 100 1382 (752–2368) 1204 (840–2031) 1138 (784–1818) 0·021 200 1382 (752–2368) 1185 (820–2011) 1105 (752–1786) 0·009 Values are median (range). MES, monopolar electrosurgery scissors; BVS, electrothermal bipolar vessel sealer; UCS, ultrasonically coagulating shears in brackets. * Kruskal–Wallis test. Open in new tab Table 5 Comprehensive costs per patient for each device, analysed by centre volume Centre volume (cases/year) . MES . BVS . UCS . P* . 20 1382 (752–2368) 1364 (999–2190) 1323 (969–2003) 0·610 50 1382 (752–2368) 1243 (880–2071) 1202 (849–1882) 0·089 100 1382 (752–2368) 1204 (840–2031) 1138 (784–1818) 0·021 200 1382 (752–2368) 1185 (820–2011) 1105 (752–1786) 0·009 Centre volume (cases/year) . MES . BVS . UCS . P* . 20 1382 (752–2368) 1364 (999–2190) 1323 (969–2003) 0·610 50 1382 (752–2368) 1243 (880–2071) 1202 (849–1882) 0·089 100 1382 (752–2368) 1204 (840–2031) 1138 (784–1818) 0·021 200 1382 (752–2368) 1185 (820–2011) 1105 (752–1786) 0·009 Values are median (range). MES, monopolar electrosurgery scissors; BVS, electrothermal bipolar vessel sealer; UCS, ultrasonically coagulating shears in brackets. * Kruskal–Wallis test. Open in new tab Surgeon evaluation The subjective evaluation of the three instruments is illustrated in Fig. 2. Overall median satisfaction for the three instruments was equally high, as determined by VAS (MES 7·7 versus BVS 8·0 versus UCS 7·7; P = 0·425). Major advantages of BVS and UCS were seen for bleeding control (MES 5·1, BVS 8·4, UCS 7·9; P < 0·001) and safety aspects (6·0 versus 8·6 versus 7·8 respectively; P < 0·001). Handling was assessed as a significant drawback of UCS (8·8 versus 7·9 versus 4·8; P < 0·001). Fig. 2 Open in new tabDownload slide Surgeon evaluation of the main features of instruments used. 1, Dissection capacity; 2, cutting ability; 3, control of bleeding; 4, handling; 5, safety aspects; 6, overall satisfaction. VAS, visual analogue scale; MES, monopolar electrosurgery scissors; BVS, electrothermal bipolar vessel sealer; UCS, ultrasonically coagulating shears. *P < 0·001 versus BVS and UCS; †P < 0·001 versus MES and BVS (Kruskal–Wallis test) Discussion In this prospective randomized study, the time taken to perform laparoscopic colonic dissection was significantly shorter with the BVS or UCS than with MES. This can be attributed to the type of instrument, as adherence to the surgical protocol was similarly high in all three groups. The difference in dissection time can be explained by reduced instrument traffic, lower clip application and less blood loss, as well as by the need for less rinsing and fluid suction. Another factor may be the reduced smoke production when using BVS, as deduced from a lower relative carbon dioxide consumption in this group. All three instruments can be considered to be safe, as complication rates were comparable. Finally, BVS and UCS were both cost effective. Only a few, heterogeneous, studies of laparoscopic dissection devices in colorectal surgery have been published, making comparison with the present findings difficult. One study of only 38 patients reported reduced operating time and blood loss with BVS and UCS versus MES15, but patient numbers in the three groups were different and the randomization process was not stated. Furthermore, these authors compared BVS 10 mm with UCS 5 mm, and endostaplers were used for vessel control in the UCS group whereas clips were employed in the MES and BVS groups. Two randomized studies10,11 compared BVS alone with MES and stapler or clips respectively, and found a slightly reduced operating time for BVS as well as a lower failure rate regarding haemostasis and vessel sealing, with no significant difference in blood loss. Comparisons with the present study are not easy to make, as these studies evaluated a BVS device for open gastric resection10 or a BVS 10-mm device in laparoscopic colectomy11. One larger trial of 146 patients found a shorter operating time and minimally reduced blood loss for UCS compared with MES12. Two cohort studies comparing BVS with UCS in laparoscopic colectomy reported a slightly reduced operating time and blood loss in the BVS group9,13. All of these studies showed a slight advantage for BVS or UCS compared with MES, but operations and outcome measures were not standardized. The present authors have previously demonstrated the safety and advantages of BVS 5 mm versus 10 mm in a prospective randomized study14. In the present study, all three instruments were compared in one type of laparoscopic colectomy (left-sided). The operation was standardized and divided into specific steps, allowing a detailed analysis of the advantages and disadvantages of each instrument. BVS and UCS are advertised by the manufacturers for their multifunctionality, secure haemostasis, small risk of collateral damage and minimal smoke production. Based on animal studies, vessel sealing with BVS is secure for up to 6–7 mm for arteries and 12 mm for veins6,19–21. Seal strength and failure rate (3–18 per cent) vary considerably between the available BVS products19–21. UCS devices effect secure sealing of vessels for up to 3–5 mm6,20. Thermal spread for BVS and UCS is small (2–3 mm), but should not be ignored6,19–21. Furthermore, ultrasonically induced injury is not visible macroscopically during surgery. Power settings and activation time should therefore be adapted when operating close to important anatomical structures, especially when using UCS6–8. Another advantage of BVS and UCS is the potentially reduced smoke production during dissection, as shown in a recent in vitro study in which the best visibility conditions were achieved with BVS22. This is in accordance with the present findings, although relative carbon dioxide consumption and frequency of lens cleaning are only surrogate parameters for smoke production. The authors are not aware of any other in vivo study that has evaluated smoke production by these devices. Costs for operations depend mainly on operating time and material costs. Despite being expensive devices, BVS and UCS may be cost effective compared with MES, as shown in the present study. The higher material-related costs for BVS and UCS are balanced by a reduced operating time and a decreased need for additional material such as clips. Furthermore, for BVS and UCS the material-related costs per patient decrease with a higher annual caseload, whereas costs for MES (disposals alone) remain the same. With an annual caseload of more than 200 patients, BVS and UCS incur significantly lower costs than MES. In the literature, cost analyses regarding laparoscopic devices are controversial and have been less detailed than in the present study11,12,15. One small study showed no differences between costs for the three devices15, although only operating time and disposals were analysed, with no consideration of the expense of the BVS and UCS generators. Others reported a minor advantage for BVS over stapling devices, or slightly higher costs for UCS compared with MES, without considering operating time11 or providing a detailed cost analysis11,12. As related costs and use of materials differ considerably between countries and even hospitals, cost analyses should be interpreted with caution. To minimize possible bias in the study, each operation was performed in a standard manner by or under supervision of two experienced colorectal surgeons who were familiar with all three devices. The three groups were well matched by randomization and the vast majority of eligible patients were included in the study, minimizing selection bias. However, some points need consideration when interpreting the results. First, the sample size was not sufficiently large to detect or exclude minor differences in the secondary endpoints. Larger studies would be needed to evaluate differences in complication rates and hospital stay. Second, the surgeons' individual preference for one of the instruments might entail a bias concerning the subjective instrument evaluation using a VAS. However, evaluation was performed by 12 different surgeons and overall results were fairly similar for the three devices. In conclusion, BVS and UCS are safe and shorten dissection time in laparoscopic colectomy compared with MES, due mainly to reduced instrument traffic. They can usually be employed without the need for additional devices, and are cost effective. BVS and UCS do not differ significantly regarding dissection time and costs. References 1 Clinical Outcomes of Surgical Therapy Study Group . A comparison of laparoscopically assisted and open colectomy for colon cancer . N Engl J Med 2004 ; 350 : 2050 – 2059 . Crossref Search ADS PubMed WorldCat 2 Veldkamp R , Kuhry E, Hop WC, Jeekel J, Kazemier G, Bonjer HJ et al. Laparoscopic surgery versus open surgery for colon cancer: short-term outcomes of a randomised trial . Lancet Oncol 2005 ; 6 : 477 – 484 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 3 Braga M , Vignali A, Zuliani W, Frasson M, Di Serio C, Di Carlo V. Laparoscopic versus open colorectal surgery: cost–benefit analysis in a single-center randomized trial . Ann Surg 2005 ; 242 : 890 – 895 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Nduka CC , Super PA, Monson JR, Darzi AW. Cause and prevention of electrosurgical injuries in laparoscopy . J Am Coll Surg 1994 ; 179 : 161 – 170 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 5 Heniford BT , Matthews BD, Sing RF, Backus C, Pratt B, Greene FL. Initial results with an electrothermal bipolar vessel sealer . Surg Endosc 2001 ; 15 : 799 – 801 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Harold KL , Pollinger H, Matthews BD, Kercher KW, Sing RF, Heniford BT. Comparison of ultrasonic energy, bipolar thermal energy, and vascular clips for the hemostasis of small-, medium-, and large-sized arteries . Surg Endosc 2003 ; 17 : 1228 – 1230 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 7 Amaral JF . The experimental development of an ultrasonically activated scalpel for laparoscopic use . Surg Laparosc Endosc 1994 ; 4 : 92 – 99 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 8 Emam TA , Cuschieri A. How safe is high-power ultrasonic dissection? Ann Surg 2003 ; 237 : 186 – 191 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 9 Campagnacci R , de Sanctis A, Baldarelli M, Rimini M, Lezoche G, Guerrieri M. Electrothermal bipolar vessel sealing device vs. ultrasonic coagulating shears in laparoscopic colectomies: a comparative study . Surg Endosc 2007 ; 21 : 1526 – 1531 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Lee WJ , Chen TC, Lai IR, Wang W, Huang MT. Randomized clinical trial of Ligasure versus conventional surgery for extended gastric cancer resection . Br J Surg 2003 ; 90 : 1493 – 1496 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Marcello PW , Roberts PL, Rusin LC, Holubkov R, Schoetz DJ. Vascular pedicle ligation techniques during laparoscopic colectomy. A prospective randomized trial . Surg Endosc 2006 ; 20 : 263 – 269 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Morino M , Rimonda R, Allaix ME, Giraudo G, Garrone C. Ultrasonic versus standard electric dissection in laparoscopic colorectal surgery: a prospective randomized clinical trial . Ann Surg 2005 ; 242 : 897 – 901 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Takada M , Ichihara T, Kuroda Y. Comparative study of electrothermal bipolar vessel sealer and ultrasonic coagulating shears in laparoscopic colectomy . Surg Endosc 2005 ; 19 : 226 – 228 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Hubner M , Hahnloser D, Hetzer F, Muller MK, Clavien PA, Demartines N. A prospective randomized comparison of two instruments for dissection and vessel sealing in laparoscopic colorectal surgery . Surg Endosc 2007 ; 21 : 592 – 594 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Targarona EM , Balague C, Marin J, Neto RB, Martinez C, Garriga J et al. Energy sources for laparoscopic colectomy: a prospective randomized comparison of conventional electrosurgery, bipolar computer-controlled electrosurgery and ultrasonic dissection. Operative outcome and costs analysis . Surg Innov 2005 ; 12 : 339 – 344 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Dindo D , Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey . Ann Surg 2004 ; 240 : 205 – 213 . Google Scholar Crossref Search ADS PubMed WorldCat 17 Lesurtel M , Selzner M, Petrowsky H, McCormack L, Clavien PA. How should transection of the liver be performed?: a prospective randomized study in 100 consecutive patients: comparing four different transection strategies . Ann Surg 2005 ; 242 : 814 – 822 . Google Scholar Crossref Search ADS PubMed WorldCat 18 de Groot V , Beckerman H, Lankhorst GJ, Bouter LM. How to measure comorbidity. a critical review of available methods . J Clin Epidemiol 2003 ; 56 : 221 – 229 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Carbonell AM , Joels CS, Kercher KW, Matthews BD, Sing RF, Heniford BT. A comparison of laparoscopic bipolar vessel sealing devices in the hemostasis of small-, medium-, and large-sized arteries . J Laparoendosc Adv Surg Tech A 2003 ; 13 : 377 – 380 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Landman J , Kerbl K, Rehman J, Andreoni C, Humphrey PA, Collyer W et al. Evaluation of a vessel sealing system, bipolar electrosurgery, harmonic scalpel, titanium clips, endoscopic gastrointestinal anastomosis vascular staples and sutures for arterial and venous ligation in a porcine model . J Urol 2003 ; 169 : 697 – 700 . Google Scholar Crossref Search ADS PubMed WorldCat 21 Richter S , Kollmar O, Schilling MK, Pistorius GA, Menger MD. Efficacy and quality of vessel sealing: comparison of a reusable with a disposable device and effects of clamp surface geometry and structure . Surg Endosc 2006 ; 20 : 890 – 894 . Google Scholar Crossref Search ADS PubMed WorldCat 22 Weld KJ , Dryer S, Ames CD, Cho K, Hogan C, Lee M et al. Analysis of surgical smoke produced by various energy-based instruments and effect on laparoscopic visibility . J Endourol 2007 ; 21 : 347 – 351 . Google Scholar Crossref Search ADS PubMed WorldCat Author notes Presented to the Annual Meeting of the Swiss Society of Surgery, Lausanne, July 2007, and published in abstract form as Br J Surg 2007; 94: 766 Copyright © 2008 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) Copyright © 2008 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd.

Sacchini, V; Beal, K; Goldberg, J; Montgomery, L; Port, E; McCormick, B

doi: 10.1002/bjs.6208pmid: 18690634

Abstract Background Partial breast irradiation has been tested in limited pilot studies and shown to provide acceptable cosmesis, minimal toxicity and adequate local control. The aim of this study was to determine the feasibility of using quadrant high-dose intraoperative radiation therapy (IORT) for the treatment of early-stage breast cancer. Methods Fifty-two women with early-stage breast cancer were treated with breast-conserving therapy and IORT between October 2002 and January 2006. The first 18 women received a radiation dose of 20 Gy. The protocol was then amended and the remaining 34 women were treated with 18 Gy. Each patient was evaluated after surgery, and at 3, 6 and 12 months; complications, toxicity and cosmetic outcomes were recorded by the breast surgeon. Results Women treated with 18 Gy appeared to have a more favourable cosmetic outcome compared with the earlier treatment group. At last follow-up, none of the women treated on the protocol had a breast recurrence. Conclusion Experience suggests that this IORT technique is feasible, although further follow-up is necessary to assess its therapeutic value. Introduction Several large prospective randomized trials have demonstrated the value of breast-conserving therapy (BCT) in the surgical management of breast cancer. It is established that there is no significant difference in disease-free or overall survival rates of women with early-stage breast cancer treated by mastectomy or lumpectomy and radiation therapy1,2. With the widespread adoption of BCT, there has been growing interest in the advancement and refinement of breast irradiation. Over the past decade, much of this attention has focused on the development of partial breast radiotherapy. This form of treatment, with its reduced field and treatment time, should prove beneficial to women with breast cancer. The ability to deliver a single therapeutic dose of radiation to the tumour bed during surgery, or in a few fractions after surgery, spares patients the standard 6-week course of external-beam treatment, alleviating their psychological distress, logistical problems and collateral morbidity. Partial breast radiotherapy also reduces the financial burden to both patient and healthcare system. Partial breast irradiation has been tested in pilot clinical trials, and in selected patients; these preliminary studies have provided good to excellent cosmesis, minimal toxicity and adequate local control. The literature on conservative therapy has suggested that patients who do experience local failure after partial breast irradiation are most likely to develop a local recurrence in the same quadrant as the index cancer3–5. The present study was a phase II clinical trial to investigate the feasibility of treating early-stage invasive breast cancer with intraoperative brachytherapy. Methods Between 8 October 2002 and 10 January 2006, 52 women with early-stage breast cancer were treated with intraoperative radiation therapy (IORT) in accordance with Memorial Sloan-Kettering institutional review board (IRB)-approved protocols. Inclusion criteria for the present protocol were: patients with invasive ductal carcinoma over 60 years old; a primary unicentric lesion of 2 cm or less in diameter; and clinically negative axillary lymph nodes. These were chosen to minimize the risk of multicentricity and the development of remote tumour recurrence. At the beginning of the trial, all patients had magnetic resonance imaging (MRI) to exclude the possibility of occult multicentric or multifocal disease. The low positive predictive value of MRI in the first ten patients caused this to be abandoned—five of the first ten patients were excluded because MRI was suspicious, although only one had multicentric breast cancer on final pathological assessment. Other exclusion criteria included the presence of lobular carcinoma in situ or Paget's disease, a histological type other than ductal carcinoma, visible dimpling of the skin above the tumour, any pre-existing conditions precluding the use of radiotherapy or regular follow-up, and any other contraindication for breast-conserving surgery. After the first 18 women had been treated, the IORT technique and treatment dose were modified because of the unexpectedly high rate of acute toxicity. The initial ten women received a radiation dose of 20 Gy at a depth of 1 cm to the superficial and radial margins, and 0·5 cm to the deep margin. The remaining 34 received a dose of 18 Gy to the lateral margin of the surgical cavity. Patients were treated in a shielded operating room. The surgery consisted of a wide local excision aiming for grossly free margins of 1·5 cm. Attention was paid to ensure negative margins on final pathology. The surgical specimen was affixed with stitches and radio-opaque clips before being positioned on a Plexiglas plate for radiological examination. If the lesion appeared to be close to one of the surgical margins, additional tissue was resected. The surgical specimen was sent to pathology for gross analysis. If the pathologist identified a close or positive margin, the surgeon performed an immediate re-excision of that margin. After surgical excision and sentinel lymph node (SLN) biopsy, the breast parenchyma was detached approximately 1 cm from the underlying skin. The skin was retracted with a long-stat retractor in an effort to avoid skin necrosis secondary to high-dose irradiation (Fig. 1a). Intraoperative brachytherapy was delivered to the tumour bed using a remote afterloading system with an iridium-192 source. Catheters carrying the iridium source were housed in a quadrangular Silastic breast applicator (H.A.M. Applicator; Mick Radio-Nuclear Instruments, Mount Vernon, New York, USA). Ideally, the applicator was placed in the surgical cavity along the pectoralis major muscle and flush with the breast parenchyma (Fig. 1b). Once the applicator was secure in the tumour bed, catheters were connected to the afterloader by source guide tubes. As the applicator was being positioned in the surgical cavity, the optimal isodose curves were calculated by a computer programme. The dose distribution plan determined the optimal isotope dwell times and modulated the appropriate intensity of the radiation treatment dose within the target region (Fig. 1c). Once the computer-generated plan had been approved by the radiation oncologist, information was entered into the treatment computer that controlled the afterloading machine. The surgical staff cleared the operating room during the irradiation treatment period (up to 40 min), although the patient was monitored continuously using a collimated monitor and cameras from outside the operating room. After the radiation had been delivered and the applicator removed from the surgical bed, the breast incision was closed in the usual manner (Fig. 1d). Fig. 1 Open in new tabDownload slide a Skin retracted with a long-stat retractor to avoid skin necrosis secondary to high-dose irradiation; blue discolouration is peritumoural dye injection for sentinel lymph node biopsy. b H.A.M. Silastic applicator placed in the surgical cavity along the pectoralis major muscle and flush with the breast parenchyma. c Optimal isotope dwell times and modulation of the appropriate intensity of the radiation dose within the target region, determined by dose distribution plan. d Breast incision after standard closure, following delivery of radiation and removal of the applicator The minimum follow-up was 12 months in the study protocol. The postoperative cosmetic outcome was evaluated by a breast surgeon at regular intervals in the treatment course: at 3, 6 and 12 months6. At each visit, a cosmetic evaluation based on Radiation Therapy Oncology Group–European Organization for Research and Treatment of Cancer evaluation criteria was completed. Side-effects that occurred in the first 3 months after radiotherapy were considered to be ‘acute toxicity’; those occurring thereafter were classified as ‘intermediate or late toxicity’. The following complications were recorded: skin erythema and desquamation, wound infection, pain and functional impairment, bleeding with haematoma formation, and incidence and duration of infection. Intermediate to late toxicity was evaluated and graded on a scale from 1 to 10 in relation to the pigmentation and consistency of the skin in the treated area, the integrity and cosmetic outcome of the surgical scar, the fibrosis within and outside the index quadrant, and any report of pain or functional impairment. Results In total, 62 women were consented as part of the Memorial Sloan-Kettering Cancer Center IRB-approved protocol. Five of these patients were excluded from treatment because their screening MRI was deemed suspicious, four of the consented women declined IORT before the day of surgery, and one could not be treated with IORT because the radiation oncologist was unable to locate the Silastic breast applicator properly during surgery. Instead, this patient was treated with a wide-excision SLN biopsy and conventional whole-breast radiation therapy. Table 1 shows the demographic and clinical characteristics of the women; their median age was 76·2 (range 45–87) years. Two women under 60 years of age were enrolled as off-protocol subjects: one requested IORT treatment and a second was unable to receive conventional external-beam radiotherapy because of morbid obesity. The inclusion of these patients was approved by the Memorial Sloan-Kettering Cancer Center IRB. All of the study patients underwent SLN biopsy; two proceeded to axillary dissection following the finding of a positive sentinel node on frozen-section examination. None required re-excision for close or positive margins after IORT surgery; however, three patients required additional surgery secondary to wound closure or infectious complications. Median follow-up was 31·4 (range 3·4–46·3) months. Thirty-five of the patients have completed the full 12 months of follow-up as prescribed in the protocol. Table 1 Demographics of 52 patients who underwent lumpectomy and intraoperative radiotherapy . No. of patients . Age (years)* 74·3(8·4) (76·2, 45–87) Follow-up (months)* 28·4(11·5) (31·4, 3·4–48·3) Breast quadrant†  Upper outer 32 (62)  Upper inner 6 (12)  Lower outer 6 (12)  Lower inner 6 (12)  Central 2 (4) Breast cup size†  A 2 (4)  B 12 (23)  C 26 (50)  D 10 (19)  DD 2 (4) . No. of patients . Age (years)* 74·3(8·4) (76·2, 45–87) Follow-up (months)* 28·4(11·5) (31·4, 3·4–48·3) Breast quadrant†  Upper outer 32 (62)  Upper inner 6 (12)  Lower outer 6 (12)  Lower inner 6 (12)  Central 2 (4) Breast cup size†  A 2 (4)  B 12 (23)  C 26 (50)  D 10 (19)  DD 2 (4) * Values are mean(s.d.) (median, range); † values in parentheses are percentages. Open in new tab Table 1 Demographics of 52 patients who underwent lumpectomy and intraoperative radiotherapy . No. of patients . Age (years)* 74·3(8·4) (76·2, 45–87) Follow-up (months)* 28·4(11·5) (31·4, 3·4–48·3) Breast quadrant†  Upper outer 32 (62)  Upper inner 6 (12)  Lower outer 6 (12)  Lower inner 6 (12)  Central 2 (4) Breast cup size†  A 2 (4)  B 12 (23)  C 26 (50)  D 10 (19)  DD 2 (4) . No. of patients . Age (years)* 74·3(8·4) (76·2, 45–87) Follow-up (months)* 28·4(11·5) (31·4, 3·4–48·3) Breast quadrant†  Upper outer 32 (62)  Upper inner 6 (12)  Lower outer 6 (12)  Lower inner 6 (12)  Central 2 (4) Breast cup size†  A 2 (4)  B 12 (23)  C 26 (50)  D 10 (19)  DD 2 (4) * Values are mean(s.d.) (median, range); † values in parentheses are percentages. Open in new tab There were 62 reported complications in the 52 treated patients; two (4 per cent) of the women experienced problems that necessitated reoperation for poor healing, which consisted of dissection of the skin suture and opening of the wound (Table 2). The initial aim was to treat these wounds conservatively by irrigation and packing of the cavity. This method proved unsuccessful and, instead, the patients required a debridement of the lumpectomy cavity with insertion of a drain. In both of these women, the wound healed within 10 days with no further complications. The surgical specimen from the reoperative surgery was sent for pathological examination and was found to contain severe necrosis, thereby preventing granulation. Table 2 Postoperative complications in 52 patients who underwent intraoperative radiotherapy . No. of patients . Skin erythema or desquamation 29 (56) Skin necrosis 0 (0) Wound infection or necrosis 4 (8)  Requiring additional operation 2 (4) Abnormal bleeding or haematoma 3 (6)  Requiring additional operation 0 (0) Seroma 26 (50)  Requiring additional operation 0 (0) . No. of patients . Skin erythema or desquamation 29 (56) Skin necrosis 0 (0) Wound infection or necrosis 4 (8)  Requiring additional operation 2 (4) Abnormal bleeding or haematoma 3 (6)  Requiring additional operation 0 (0) Seroma 26 (50)  Requiring additional operation 0 (0) Values in parentheses are percentages. Open in new tab Table 2 Postoperative complications in 52 patients who underwent intraoperative radiotherapy . No. of patients . Skin erythema or desquamation 29 (56) Skin necrosis 0 (0) Wound infection or necrosis 4 (8)  Requiring additional operation 2 (4) Abnormal bleeding or haematoma 3 (6)  Requiring additional operation 0 (0) Seroma 26 (50)  Requiring additional operation 0 (0) . No. of patients . Skin erythema or desquamation 29 (56) Skin necrosis 0 (0) Wound infection or necrosis 4 (8)  Requiring additional operation 2 (4) Abnormal bleeding or haematoma 3 (6)  Requiring additional operation 0 (0) Seroma 26 (50)  Requiring additional operation 0 (0) Values in parentheses are percentages. Open in new tab Complications reported by women at the 3-, 6- and 12-month follow-up visits are shown in Table 3. The cosmesis scores recorded by the patients at these visits are shown in Table 4. Although the number of patients in the study was too small for formal statistical analysis, the cosmesis scores appeared lower (better) in patients treated with a radiation dose of 18 Gy than with the initial 20 Gy. Table 3 Complications reported 3, 6 and 12 months after intraoperative radiotherapy . Time after surgery (months) . . 3 (n = 46) . 6 (n = 45) . 12 (n = 35) . Seroma 9 (20) 10 (22) 3 (9)  Requiring additional surgery 0 (0) 0 (0) 0 (0) Wound infection or necrosis 3 (7) 4 (9) 1 (3)  Requiring additional surgery 1 (2) 0 (0) 0 (0) Pain 2 (4) 5 (11) 1 (3) . Time after surgery (months) . . 3 (n = 46) . 6 (n = 45) . 12 (n = 35) . Seroma 9 (20) 10 (22) 3 (9)  Requiring additional surgery 0 (0) 0 (0) 0 (0) Wound infection or necrosis 3 (7) 4 (9) 1 (3)  Requiring additional surgery 1 (2) 0 (0) 0 (0) Pain 2 (4) 5 (11) 1 (3) Values in parentheses are percentages. Open in new tab Table 3 Complications reported 3, 6 and 12 months after intraoperative radiotherapy . Time after surgery (months) . . 3 (n = 46) . 6 (n = 45) . 12 (n = 35) . Seroma 9 (20) 10 (22) 3 (9)  Requiring additional surgery 0 (0) 0 (0) 0 (0) Wound infection or necrosis 3 (7) 4 (9) 1 (3)  Requiring additional surgery 1 (2) 0 (0) 0 (0) Pain 2 (4) 5 (11) 1 (3) . Time after surgery (months) . . 3 (n = 46) . 6 (n = 45) . 12 (n = 35) . Seroma 9 (20) 10 (22) 3 (9)  Requiring additional surgery 0 (0) 0 (0) 0 (0) Wound infection or necrosis 3 (7) 4 (9) 1 (3)  Requiring additional surgery 1 (2) 0 (0) 0 (0) Pain 2 (4) 5 (11) 1 (3) Values in parentheses are percentages. Open in new tab Table 4 Mean patient-reported cosmesis scores at 3, 6 and 12 months after surgery and intraoperative radiation therapy of 18 or 20 Gy . Time after surgery (months) . . 3 . 6 . 12 . Symmetry*  20 Gy 4·3 4·7 5·0  18 Gy 3·1 4·0 3·6 Breast oedema†  20 Gy 2·8 3·1 1·7  18 Gy 3·1 1·5 1·6 Skin thickening†  20 Gy 3·9 4·2 3·6  18 Gy 1·9 2·5 2·1 Fibrosis†  20 Gy 4·0 5·3 4·6  18 Gy 3·1 3·3 3·1 Retraction†  20 Gy 4·0 4·9 5·4  18 Gy 3·4 3·4 3·5 Telangiectasia†  20 Gy 2·8 2·2 2·6  18 Gy 3·2 1·5 1·8 Dimpling†  20 Gy 3·7 4·9 5·2  18 Gy 1·7 3·0 3·3 . Time after surgery (months) . . 3 . 6 . 12 . Symmetry*  20 Gy 4·3 4·7 5·0  18 Gy 3·1 4·0 3·6 Breast oedema†  20 Gy 2·8 3·1 1·7  18 Gy 3·1 1·5 1·6 Skin thickening†  20 Gy 3·9 4·2 3·6  18 Gy 1·9 2·5 2·1 Fibrosis†  20 Gy 4·0 5·3 4·6  18 Gy 3·1 3·3 3·1 Retraction†  20 Gy 4·0 4·9 5·4  18 Gy 3·4 3·4 3·5 Telangiectasia†  20 Gy 2·8 2·2 2·6  18 Gy 3·2 1·5 1·8 Dimpling†  20 Gy 3·7 4·9 5·2  18 Gy 1·7 3·0 3·3 * Scale: 1 (symmetrical) to 10 (no symmetry). † Scale: 1 (none) to 10 (most). Open in new tab Table 4 Mean patient-reported cosmesis scores at 3, 6 and 12 months after surgery and intraoperative radiation therapy of 18 or 20 Gy . Time after surgery (months) . . 3 . 6 . 12 . Symmetry*  20 Gy 4·3 4·7 5·0  18 Gy 3·1 4·0 3·6 Breast oedema†  20 Gy 2·8 3·1 1·7  18 Gy 3·1 1·5 1·6 Skin thickening†  20 Gy 3·9 4·2 3·6  18 Gy 1·9 2·5 2·1 Fibrosis†  20 Gy 4·0 5·3 4·6  18 Gy 3·1 3·3 3·1 Retraction†  20 Gy 4·0 4·9 5·4  18 Gy 3·4 3·4 3·5 Telangiectasia†  20 Gy 2·8 2·2 2·6  18 Gy 3·2 1·5 1·8 Dimpling†  20 Gy 3·7 4·9 5·2  18 Gy 1·7 3·0 3·3 . Time after surgery (months) . . 3 . 6 . 12 . Symmetry*  20 Gy 4·3 4·7 5·0  18 Gy 3·1 4·0 3·6 Breast oedema†  20 Gy 2·8 3·1 1·7  18 Gy 3·1 1·5 1·6 Skin thickening†  20 Gy 3·9 4·2 3·6  18 Gy 1·9 2·5 2·1 Fibrosis†  20 Gy 4·0 5·3 4·6  18 Gy 3·1 3·3 3·1 Retraction†  20 Gy 4·0 4·9 5·4  18 Gy 3·4 3·4 3·5 Telangiectasia†  20 Gy 2·8 2·2 2·6  18 Gy 3·2 1·5 1·8 Dimpling†  20 Gy 3·7 4·9 5·2  18 Gy 1·7 3·0 3·3 * Scale: 1 (symmetrical) to 10 (no symmetry). † Scale: 1 (none) to 10 (most). Open in new tab There have been no local or distance recurrences in any of the patients treated by protocol, although the expected number of local recurrences for this population with early breast cancer at a median follow-up of 31·4 months would be less than 1 per cent. One of the patients developed a new primary, non-small-cell lung carcinoma 14 months after IORT surgery for breast carcinoma. The patient subsequently died from metastatic lung cancer. Discussion Despite the evidence supporting the use of BCT for early-stage breast cancer, some patients still elect to undergo mastectomy for non-medical reasons, including age and distance from their radiation site7,8. These observations demonstrate the need for a viable alternative to conventional whole-breast radiation therapy. The rationale for segmental radiotherapy in place of whole-breast irradiation is based on our observation that approximately 85 per cent of local relapses after treatment with BCT are confined to the same quadrant of the breast as the index tumour. The risk of local failure in a distant quadrant is similar to the risk of contralateral cancer in the non-irradiated breast9. More than 90 per cent of recurrent tumour foci are found in close proximity to the original cancer. Residual microscopic disease is believed to be the cause of local disease recurrence10. Some series have reported a higher rate of local recurrence in a breast quadrant remote from the original cancer; extensive ductal carcinoma in situ, lobular histotype and young patient age are generally related to this11,12. Inclusion criteria that took these factors into account were applied in the present study to minimize the risks of local recurrence. Several preliminary pilot studies have demonstrated that single-dose partial breast radiotherapy can be used in selected patients with localized breast cancer, without increasing the normal rate of complications after surgery, but providing good local control and cosmesis13,14. The largest randomized clinical trial to date is now in progress at the Milan Institute. The aim of the trial is to determine equivalence of local recurrence rates between conventional whole-breast radiation therapy and quadrantectomy with IORT. In the present study there was an increased risk of postoperative wound complications in the early patients. Decreasing the dose by 2 Gy appeared to reduce local complications to a more acceptable level. The risk of postoperative complications in IORT is important in evaluating the risks and benefits of this technique. Questions remain concerning the pathway and timing of cell damage and death in the face of irradiation. Critics of segmental breast IORT contend that the treatment cannot ensure an adequate level of cell death in the irradiated breast tissue. It is assumed that in conventional whole-breast radiation therapy, where the breast receives a radiation dose over several weeks, any remaining tumour foci are able to continue cell cycling between treatments. Therefore, it is likely that, with the next treatment dose, the remaining cells will be killed as they have reached a more radiation-sensitive cell phase. In contrast, partial breast irradiation, delivered as a single dose, depends ultimately upon the phase of each cell. Tumour cells in the more radiation-resistant S phase might survive, whereas those in the more sensitive G2 phase will probably be killed15. Alternatively, as time elapses between surgery and the commencement of conventional radiation therapy, there may be exponential repopulation of cells in the remaining breast tissue16. The safety and feasibility of administering a single treatment of high-dose radiation has been called into question. In the absence of long-term follow-up, there remain concerns about lifelong cosmesis, including the possible development of necrosis or fibrosis. However, segmental irradiation is not a novel treatment for carcinoma. It has been successful in the management of several different malignancies, including rectal and brain cancer17,18, where good local control, low toxicity and improved survival have been described19. Future studies are needed to determine whether similar conclusions may be drawn about the use of segmental radiation for the treatment of breast cancer. One caveat is that the oncoplastic technique is becoming more common in the conservative surgical approach to breast cancer. Patients having breast reconstructive surgery may not be suitable for IORT owing to the extent of plastic restoration required for reshaping the breast. References 1 Veronesi U , Cascinelli N, Mariani L, Greco M, Saccozzi R, Luini A et al. Twenty-year follow-up of a randomized study comparing breast-conserving surgery with radical mastectomy for early breast cancer . N Engl J Med 2002 ; 347 : 1227 – 1232 . Google Scholar Crossref Search ADS PubMed WorldCat 2 Fisher B , Anderson S, Bryant J, Margolese RG, Deutsch M, Fisher ER et al. Twenty-year follow-up of a randomized trial comparing total mastectomy, lumpectomy, and lumpectomy plus irradiation for the treatment of invasive breast cancer . N Engl J Med 2002 ; 347 : 1233 – 1241 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Vicini FA , Chen PY, Fraile M, Gustafson GS, Edmundson GK, Jaffray DA et al. Low-dose-rate brachytherapy as the sole radiation modality in the management of patients with early-stage breast cancer treated with breast-conserving therapy: preliminary results of a pilot trial . Int J Radiat Oncol Biol Phys 1997 ; 38 : 301 – 310 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Baglan KL , Martinez AA, Frazier RC, Kini VR, Kestin LL, Chen PY et al. The use of high-dose-rate brachytherapy alone after lumpectomy in patients with early-stage breast cancer treated with breast-conserving therapy . Int J Radiat Oncol Biol Phys 2001 ; 50 : 1003 – 1011 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Veronesi U , Orecchia R, Luini A, Gatti G, Intra M, Zurrida S et al. A preliminary report of intraoperative radiotherapy (IORT) in limited-stage breast cancers that are conservatively treated . Eur J Cancer 2001 ; 37 : 2178 – 2183 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Beal K , McCormick B, Zelefsky MJ, Borgen P, Fey J, Goldberg J et al. Single-fraction intraoperative radiotherapy for breast cancer: early cosmetic results . Int J Radiat Oncol Biol Phys 2007 ; 69 : 19 – 24 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Morrow M , Bucci C, Rademaker A. Medical contraindications are not a major factor in the underutilization of breast conserving therapy . J Am Coll Surg 1998 ; 186 : 269 – 274 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Athas WF , Adams-Cameron M, Hunt WC, Amir-Fazli A, Key CR. Travel distance to radiation therapy and receipt of radiotherapy following breast-conserving surgery . J Natl Cancer Inst 2000 ; 92 : 269 – 271 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Veronesi U , Marubini E, Mariani L, Galimberti V, Luini A, Veronesi P et al. Radiotherapy after breast-conserving surgery in small breast carcinoma: long-term results of a randomized trial . Ann Oncol 2001 ; 12 : 997 – 1003 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Gump FE , Shikora S, Habif DV, Kister S, Logerfo P, Estabrook A. The extent and distribution of cancer in breasts with palpable primary tumors . Ann Surg 1986 ; 204 : 384 – 390 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Pinsky RW , Rebner M, Pierce LJ, Ben-David MA, Vicini F, Hunt KA et al. Recurrent cancer after breast-conserving surgery with radiation therapy for ductal carcinoma in situ: mammographic features, method of detection, and stage of recurrence . AJR Am J Roentgenol 2007 ; 189 : 140 – 144 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Hannoun-Levi JM , Houvenaeghel G, Ellis S, Teissier E, Alzieu C, Lallement M et al. Partial breast irradiation as second conservative treatment for local breast cancer recurrence . Int J Radiat Oncol Biol Phys 2004 ; 60 : 1385 – 1392 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Arthur DW , Koo D, Zwicker RD, Tong S, Bear HD, Kaplan BJ et al. Partial breast brachytherapy after lumpectomy: low-dose-rate and high-dose-rate experience . Int J Radiat Oncol Biol Phys 2003 ; 56 : 681 – 689 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Luini A , Orecchia R, Gatti G, Intra M, Ciocca M, Galimberti V et al. The pilot trial on intraoperative radiotherapy with electrons (ELIOT): update on the results . Breast Cancer Res Treat 2005 ; 93 : 55 – 59 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Rosenstein BS , Lymberis SC, Formenti SC. Biologic comparison of partial breast irradiation protocols . Int J Radiat Oncol Biol Phys 2004 ; 60 : 1393 – 1404 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Vaidya JS , Tobias JS, Baum M, Wenz F, Kraus-Tiefenbacher U, D'souza D et al. Targeted intraoperative radiotherapy (TARGIT): an innovative approach to partial-breast irradiation . Semin Radiat Oncol 2005 ; 15 : 84 – 91 . Google Scholar Crossref Search ADS PubMed WorldCat 17 Kapiteijn E , Marijnen CA, Nagtegaal ID, Putter H, Steup WH, Wiggers T et al. Preoperative radiotherapy combined with total mesorectal excision for resectable rectal cancer . N Engl J Med 2001 ; 345 : 638 – 646 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Flickinger JC , Kondziolka D, Lunsford LD. Radiobiological analysis of tissue responses following radiosurgery . Technol Cancer Res Treat 2003 ; 2 : 87 – 92 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Willett CG . Intraoperative radiation therapy . Int J Clin Oncol 2001 ; 6 : 209 – 214 . Google Scholar Crossref Search ADS PubMed WorldCat Copyright © 2008 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) Copyright © 2008 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd.

Vohra, R S; on behalf of the General Anaesthetic versus Local Anaesthetic for carotid endarterectomy (GALA) trial collaborators; Coughlin, P A; on behalf of the General Anaesthetic versus Local Anaesthetic for carotid endarterectomy (GALA) trial collaborators; McShane, P; on behalf of the General Anaesthetic versus Local Anaesthetic for carotid endarterectomy (GALA) trial collaborators; Bains, M; on behalf of the General Anaesthetic versus Local Anaesthetic for carotid endarterectomy (GALA) trial collaborators;

Griffin, S M; Lamb, P J; Shenfine, J; Richardson, D L; Karat, D; Hayes, N

doi: 10.1002/bjs.6294pmid: 18655213

Abstract Background The aim of this study was to evaluate the diagnosis, management and outcome of patients with spontaneous rupture of the oesophagus in a single centre. Methods Between October 1993 and May 2007, 51 consecutive patients with spontaneous oesophageal rupture were evaluated with contrast radiology and flexible endoscopy. Patients with limited contamination who fulfilled specific criteria were managed by a non-operative approach, whereas the remainder underwent thoracotomy. Results The median time to diagnosis was 24 (range 4–604) h. Initial diagnosis was by contrast swallow in 18 of 24 patients, computed tomography in 15 of 17 and endoscopy in 18 of 18. There were no deaths among 17 patients who were managed non-operatively with targeted drainage, intravenous antimicrobials, nasogastric decompression and enteral nutrition. Of 31 patients who underwent primary thoracotomy and oesophageal repair (over a Τ tube in 29), 11 died in hospital. Three patients could not be resuscitated adequately and did not have surgical intervention. Conclusion Spontaneous oesophageal rupture represents a spectrum of disease. Accurate radiological and endoscopic evaluation can identify those suitable for radical non-operative treatment and those who require thoracotomy. Introduction Spontaneous rupture of the oesophagus, first described by Boerhaave in 17241, is characterized by barogenic disruption of the distal oesophagus and contamination of the mediastinum and pleural cavity with gastric contents. Surgery has traditionally been the treatment of choice, although it is now recognized that non-operative management can be employed according to the degree of contamination2–7. The rarity of this condition means that even specialist centres have limited exposure to the problem and current evidence is based on small case series. Some series have inappropriately included patients with iatrogenic perforation or focused only on those who survive to surgery7–9. The aim of this study was to evaluate the diagnosis, management and outcome of all patients presenting with spontaneous rupture of the oesophagus over a 13-year period with a consistent approach to management. Methods A prospective database was analysed to identify all patients with spontaneous rupture of the oesophagus treated at the Northern Oesophago-gastric Unit between October 1993 and May 2007. Clinical presentation, diagnosis, management and outcome were evaluated. Patients with iatrogenic rupture of the oesophagus, perforation from swallowed foreign bodies or other barogenic trauma (for example blast injuries) were excluded from the study. Spontaneous rupture was defined as complete disruption of the oesophageal wall occurring in the absence of pre-existing pathology1,3,10. The series included tertiary referrals from all regional hospitals with a suspected diagnosis of spontaneous oesophageal rupture. The diagnosis of perforation was made by the appearance of a full-thickness defect on flexible upper gastrointestinal videoendoscopy or extravasation of contrast during a radiological swallow (performed with water-soluble contrast followed by barium if no leak was demonstrated) or contrast-enhanced computed tomography (CT). Following initial resuscitation, all patients were investigated in the specialist unit to confirm the diagnosis and further evaluate the site of perforation and degree of contamination. Stable patients underwent a combination of a contrast swallow (to determine the site, extent of leakage and communication with the pleural cavity), endoscopy (to determine the site of leakage and length of tear) and CT (to determine the site and extent of contamination). If the patient had undergone a contrast swallow or CT before transfer these investigations were repeated if the imaging was suboptimal. Endoscopy was performed only in the specialist unit. Unstable or ventilated patients underwent initial endoscopic evaluation in the critical care unit to confirm the diagnosis, followed by CT if non-operative management was contemplated. It was recorded whether patients met predefined criteria for systemic inflammatory response syndrome (SIRS) at the time of admission11. A multidisciplinary approach was adopted, involving oesophagogastric surgeon, intensivist, interventional radiologist, dietician, physiotherapist and nursing staff. Patients with limited contamination or who had demonstrated tolerance were managed using an intensive non-operative approach if they met the criteria outlined in Table 1, whereas those with more extensive contamination underwent thoracotomy (Fig. 1). Fig. 1 Open in new tabDownload slide Management of spontaneous rupture of the oesophagus Table 1 Criteria for non-operative management Perforation contained by the mediastinal pleura Flow of contrast back into oesophagus on contrast swallow No clinical evidence of mediastinitis No solid food contamination of mediastinum or Demonstrated tolerance to pleural contamination (radiologically drained) for > 72 h before transfer Perforation contained by the mediastinal pleura Flow of contrast back into oesophagus on contrast swallow No clinical evidence of mediastinitis No solid food contamination of mediastinum or Demonstrated tolerance to pleural contamination (radiologically drained) for > 72 h before transfer Open in new tab Table 1 Criteria for non-operative management Perforation contained by the mediastinal pleura Flow of contrast back into oesophagus on contrast swallow No clinical evidence of mediastinitis No solid food contamination of mediastinum or Demonstrated tolerance to pleural contamination (radiologically drained) for > 72 h before transfer Perforation contained by the mediastinal pleura Flow of contrast back into oesophagus on contrast swallow No clinical evidence of mediastinitis No solid food contamination of mediastinum or Demonstrated tolerance to pleural contamination (radiologically drained) for > 72 h before transfer Open in new tab Patients managed non-operatively received supportive care in a critical care facility. They were kept nil by mouth with intravenous broad-spectrum antimicrobials, antisecretory agents, nasogastric decompression, enteral nutrition and targeted mediastinal/pleural drainage. Contrast radiology, endoscopy and CT were used to monitor the perforation, with a low threshold for re-evaluation if there was evidence of sepsis or failure to improve. Subsequent collections were drained under radiological guidance and surgical intervention considered only if this failed. Patients who did not satisfy these criteria underwent thoracotomy. A left posterolateral (eighth intercostal space) or right posterolateral (sixth intercostal space) thoracotomy was used for access depending on the site of perforation. Solid food and contamination were removed, with debridement of necrotic tissue. The defect was opened to allow repair of all oesophageal layers, including mucosa, over an 18-G Τ tube8,10. Drains were placed in the mediastinum adjacent to the repair and in the pleural cavity. Patients were monitored in the same way as those receiving non-operative treatment. The Τ tube was left in situ for a minimum of 3 weeks or until thoracic sepsis was fully controlled and a formal oesophagocutaneous fistula had been established. Statistical analysis Statistical analysis was by χ2 test for categorical data and Mann–Whitney U test for continuous data. P < 0·050 was considered statistically significant. Results Fifty-one patients with confirmed spontaneous rupture of the oesophagus were identified, 37 men and 14 women with a median age of 64 (range 18–88) years. Thirty-eight patients were referred from outlying hospitals and 13 patients from other units within the hospital. All patients had a history of vomiting before presentation. Thirty-eight complained of chest pain and 22 had abdominal pain. Clinical signs of peritonitis had led to a negative laparotomy in three patients before referral. Only seven patients had Mackler's triad of vomiting, chest pain and surgical emphysema at the time of presentation12. Forty-two patients exhibited SIRS at presentation. Seventeen patients were suspected of having spontaneous rupture of the oesophagus at the time of initial hospital admission. A variety of provisional diagnoses were made for other patients (Table 2). Forty-six patients had an abnormal chest radiograph, and a pneumomediastinum or hydropneumothorax was present in 36. The emergency interpretation of the chest radiograph differed from the reported result in 27 instances owing to failure to identify mediastinal, thoracic or subcutaneous air. The median delay from admission to confirmation of the diagnosis was 24 (range 4–604) h. Delay in diagnosis of more than 24 h had no effect on mortality rate (six of 23 versus eight of 28 patients with a shorter delay; χ2 = 0·039, 1 d.f., P = 0·843). Table 2 Initial diagnosis and admitting specialty Specialty and initial diagnosis . No. of patients . Medicine 20  Pneumonia 11  Spontaneous pneumothorax 4  Myocardial infarction 2  Oesophageal varices 1  Mallory–Weiss tear 1  Gastroenteritis 1 General surgery 15  Spontaneous oesophageal rupture 6  Peritonitis 2  Food bolus obstruction 1  Mesenteric ischaemia 1  Pancreatitis 1  Oesophagitis 1  Perforated duodenal ulcer 2  Leaking aortic aneurysm 1 Intensive care 15  Spontaneous oesophageal rupture 11  Sepsis of unknown origin 4 Cardiothoracic surgery 1  Thoracic aortic dissection 1 Specialty and initial diagnosis . No. of patients . Medicine 20  Pneumonia 11  Spontaneous pneumothorax 4  Myocardial infarction 2  Oesophageal varices 1  Mallory–Weiss tear 1  Gastroenteritis 1 General surgery 15  Spontaneous oesophageal rupture 6  Peritonitis 2  Food bolus obstruction 1  Mesenteric ischaemia 1  Pancreatitis 1  Oesophagitis 1  Perforated duodenal ulcer 2  Leaking aortic aneurysm 1 Intensive care 15  Spontaneous oesophageal rupture 11  Sepsis of unknown origin 4 Cardiothoracic surgery 1  Thoracic aortic dissection 1 Open in new tab Table 2 Initial diagnosis and admitting specialty Specialty and initial diagnosis . No. of patients . Medicine 20  Pneumonia 11  Spontaneous pneumothorax 4  Myocardial infarction 2  Oesophageal varices 1  Mallory–Weiss tear 1  Gastroenteritis 1 General surgery 15  Spontaneous oesophageal rupture 6  Peritonitis 2  Food bolus obstruction 1  Mesenteric ischaemia 1  Pancreatitis 1  Oesophagitis 1  Perforated duodenal ulcer 2  Leaking aortic aneurysm 1 Intensive care 15  Spontaneous oesophageal rupture 11  Sepsis of unknown origin 4 Cardiothoracic surgery 1  Thoracic aortic dissection 1 Specialty and initial diagnosis . No. of patients . Medicine 20  Pneumonia 11  Spontaneous pneumothorax 4  Myocardial infarction 2  Oesophageal varices 1  Mallory–Weiss tear 1  Gastroenteritis 1 General surgery 15  Spontaneous oesophageal rupture 6  Peritonitis 2  Food bolus obstruction 1  Mesenteric ischaemia 1  Pancreatitis 1  Oesophagitis 1  Perforated duodenal ulcer 2  Leaking aortic aneurysm 1 Intensive care 15  Spontaneous oesophageal rupture 11  Sepsis of unknown origin 4 Cardiothoracic surgery 1  Thoracic aortic dissection 1 Open in new tab The diagnosis of oesophageal perforation was made initially by a contrast swallow in 18 patients, contrast-enhanced CT in 15 and flexible videoendoscopy in 18. Eight of 18 patients diagnosed by endoscopy had undergone previous radiological investigations (contrast swallow in six, CT in two) that had failed to identify a perforation. Endoscopy was ultimately performed in all 51 patients, and identified the oesophageal tear in every patient. The median length of mucosal tear at endoscopy was 3 (range 1–11) cm. The perforation was left sided in 39 patients and right sided in 12. At endoscopy a nasogastric tube was placed across the perforation to decompress the stomach. Three patients with established multiorgan failure could not be resuscitated adequately for surgical intervention and all died within 24 h. Seventeen patients who met the criteria for non-operative management were managed without surgery. Sixteen had limited contamination and one patient with a delayed diagnosis had demonstrated tolerance to prolonged pleural drainage. Nutrition was maintained enterally via a nasojejunal tube placed at the time of endoscopy (12 patients) or via a surgically placed jejunostomy (five). No patient required surgery, although eight required percutaneous drainage of collections under radiological guidance. One complex, loculated collection required repeated drainage facilitated by instillation of streptokinase. Two patients ultimately had a Τ tube placed across the perforation and brought out to establish a controlled oesophagocutaneous fistula using a percutaneous–endoscopic approach, avoiding thoracotomy13. The median hospital stay was 14 (range 6–113) days. There were no in-hospital deaths in this group (Table 3). Table 3 Comparison of clinical data and outcome in patients who had non-operative or surgical management . Management . All patients (n = 48) . P . . Non-operative (n = 17) . Surgical (n = 31) . Age (years)* 64 (18–87) 64 (47–78) 64 (18–87) 0·382† Delay to diagnosis (h)* 22 (4–500) 24 (4–604) 24 (4–604) 0·657† SIRS on admission 9 30 39 < 0·001‡ Hospital stay (days)* 14 (6–113) 44 (3–147) 41 (3–147) 0·033† ICU stay (days)* 0 (0–30) 19 (2–75) 9 (0–75) < 0·001† Death 0 11 11 0·005‡ . Management . All patients (n = 48) . P . . Non-operative (n = 17) . Surgical (n = 31) . Age (years)* 64 (18–87) 64 (47–78) 64 (18–87) 0·382† Delay to diagnosis (h)* 22 (4–500) 24 (4–604) 24 (4–604) 0·657† SIRS on admission 9 30 39 < 0·001‡ Hospital stay (days)* 14 (6–113) 44 (3–147) 41 (3–147) 0·033† ICU stay (days)* 0 (0–30) 19 (2–75) 9 (0–75) < 0·001† Death 0 11 11 0·005‡ * Values are median (range). SIRS, systemic inflammatory response syndrome; ICU, intensive care unit. † Mann–Whitney U test; ‡ χ2 test. Open in new tab Table 3 Comparison of clinical data and outcome in patients who had non-operative or surgical management . Management . All patients (n = 48) . P . . Non-operative (n = 17) . Surgical (n = 31) . Age (years)* 64 (18–87) 64 (47–78) 64 (18–87) 0·382† Delay to diagnosis (h)* 22 (4–500) 24 (4–604) 24 (4–604) 0·657† SIRS on admission 9 30 39 < 0·001‡ Hospital stay (days)* 14 (6–113) 44 (3–147) 41 (3–147) 0·033† ICU stay (days)* 0 (0–30) 19 (2–75) 9 (0–75) < 0·001† Death 0 11 11 0·005‡ . Management . All patients (n = 48) . P . . Non-operative (n = 17) . Surgical (n = 31) . Age (years)* 64 (18–87) 64 (47–78) 64 (18–87) 0·382† Delay to diagnosis (h)* 22 (4–500) 24 (4–604) 24 (4–604) 0·657† SIRS on admission 9 30 39 < 0·001‡ Hospital stay (days)* 14 (6–113) 44 (3–147) 41 (3–147) 0·033† ICU stay (days)* 0 (0–30) 19 (2–75) 9 (0–75) < 0·001† Death 0 11 11 0·005‡ * Values are median (range). SIRS, systemic inflammatory response syndrome; ICU, intensive care unit. † Mann–Whitney U test; ‡ χ2 test. Open in new tab Thirty-one patients were managed surgically, 21 via a left and ten via a right thoracotomy. Two patients early in the series underwent primary oesophageal repair and the remaining patients had repair over an 18-G Τ tube. All had a surgical jejunostomy inserted via a minilaparotomy. One patient required repeat thoracotomy to drain a complex collection 12 days after Τ tube repair. There were 11 in-hospital deaths, a median of 23 (range 2–71) days after surgery, including one patient who had a primary oesophageal repair. Five early deaths (within 5 days of surgery) were from multiorgan failure. The main factors contributing to the six later deaths were sepsis and multiorgan failure (two patients), cardiac failure, bronchopleural fistula, bleeding duodenal ulcer and small bowel obstruction. Patients managed surgically had a median hospital stay of 44 (range 3–147) days. For those who left hospital this was 57 (range 37–147) days. All patients required intensive care support, with a median stay of 19 (range 2–75) days. The proportion of patients with SIRS on admission was higher in the surgical group (P < 0·001), but there was no difference in the delay to diagnosis between groups (Table 3). The length of intensive care stay (P < 0·001), overall hospital stay (P = 0·033) and the in-hospital mortality rate (P = 0·005) were greater in the surgical group than in the non-operative group. The overall in-hospital mortality rate for the entire series was 27 per cent (14 of 51). Discussion A management algorithm for spontaneous oesophageal rupture based on radiological and endoscopic findings was used to select patients for radical non-operative management or thoracotomy. The diagnosis of spontaneous oesophageal rupture was frequently delayed, probably reflecting its rarity and varied presentation. This condition represents a spectrum of disease from contained leaks that can be managed by an intensive non-operative strategy to widespread mediastinal and pleural contamination requiring thoracotomy. Atypical presentations were common; although all patients in this series had a history of vomiting, only seven had chest pain and subcutaneous emphysema as described by Mackler12. Subclinical air escape is a common early feature on chest radiography but seems frequently to be missed by admitting staff, resulting in admission to the wrong specialty and delayed diagnosis as seen here and in other series14. Rapid diagnosis and treatment has repeatedly been associated with a better outcome in spontaneous oesophageal perforation, with 24 h being arbitrarily set as a divide for diagnostic delay4,6,15–17. In this series, the median delay to diagnosis was 24 h and a longer delay was not associated with increased mortality. Although unexpected, this finding is likely to reflect selection bias; some patients with a lesser degree of contamination and those who demonstrated tolerance experienced longer delays yet had a better outcome than those who were moribund at presentation but whose diagnosis had been made rapidly. Although the clinical picture and fluid from thoracocentesis may suggest oesophageal rupture, further investigations are needed to confirm the diagnosis and plan management. A contrast swallow was the most common initial investigation for stable patients in this series. It is important to note that there were a number of false-negative results, a limitation identified in previous studies18,19. Advances in contrast-enhanced CT mean that it is increasingly effective for patients who are unstable or in whom the diagnosis is in doubt20. Flexible endoscopy was used to confirm the diagnosis in all patients in the present series, including those with negative radiological examinations, and allowed further evaluation of the tear. When the perforation was contained by mediastinal pleura it allowed insertion of nasogastric and nasojejunal feeding tubes to facilitate non-operative management. For ventilated patients it was vital to confirm the diagnosis and provide information required to plan surgical management, including the site, level and extent of the mucosal tear. In the authors' experience the mucosal tear, which must be included in any repair, is invariably longer than the muscular injury. Endoscopy can also exclude oesophageal perforation and prevent unnecessary thoracotomy for patients with a similar presentation resulting from spontaneous mediastinum from ruptured bullae21, mediastinal haematoma and primary chest sepsis. Historically there were concerns that endoscopy caused further damage but there is no evidence to support this belief. It has also been used without additional morbidity to examine the oesophagus in penetrating thoracic trauma and following oesophagogastric surgery22,23. This study has confirmed that there is a spectrum of contamination2–7 that can be classified by radiological, endoscopic and clinical findings, and used to tailor management. The concept of a rupture contained by the mediastinal pleura was introduced by Barrett24 in 1947. The intensive non-operative strategy described seems ideal for this situation and was successful for all such patients. This non-operative approach depends on a rapid, accurate initial assessment and focused multidisciplinary approach to managing complications. Continuous reassessment and reinvestigation is required with a low threshold for intervention. Two patients required further control of the oesophageal leak. This was achieved by a combined endoscopic–radiological technique, which has been reported separately13. Those with a delayed diagnosis who demonstrate tolerance to pleural drainage may also be suitable for non-operative management7,25–27. Surgery remains the mainstay of treatment for those who present early with widespread pleural contamination. These patients invariably exhibit a systemic inflammatory response and are acutely unwell. If there has been a long delay before diagnosis, a primary oesophageal repair is prone to leak owing to local sepsis and tissue oedema15,17. The preferred technique in this study was repair over a Τ tube to create a controlled oesophagocutaneous fistula8,10. This was successful in minimizing mediastinal and pleural sepsis, with only one patient requiring repeat thoracotomy. Of equal importance to the type of repair are thorough lavage and debridement, and careful placement of mediastinal and pleural drains. No patient in this series required oesophageal resection, which would have been considered as a salvage procedure if oesophageal repair had not been possible. Good results have been reported in small series using oesophagectomy and exclusion for selected patients with severe mediastinal sepsis28,29, although such patients subsequently require further major surgery for oesophageal reconstruction. The overall hospital stay, intensive care stay and in-hospital mortality rate in the operative group were higher than those in patients managed non-operatively, and in some smaller series30. This may be partly because all the surgical patients were at the severe end of the disease spectrum, reflected by five early deaths from multiorgan failure despite aggressive supportive treatment. The overall mortality rate of 27 per cent is comparable with published values3,6,27 despite there being no case selection in this series, which included even moribund patients in whom supportive care only was instituted. Recent studies have proposed the insertion of self-expandable stents to treat oesophageal disruption31,32. Although immediate stent insertion following iatrogenic perforation of advanced oesophageal carcinoma may be appropriate33, a delay to stent insertion and the presence of contamination are inevitable with spontaneous rupture and may explain why such patients had poor outcomes in recent reports31,32. In this situation the stent may prevent adequate drainage of sepsis, delay healing and potentially erode adjacent structures if it cannot be removed. There is also a high risk of stent migration in the absence of a mechanical stricture to hold it in place. The authors strongly recommend avoiding the use of stents in these situations. Although surgery remains the mainstay of treatment, intensive non-operative management is possible in carefully selected patients with spontaneous oesophageal rupture. Success depends on a multidisciplinary approach with continuous reassessment and a low threshold for intervention. Acknowledgements The authors acknowledge Dr Alastair Gascoigne (Consultant in Intensive Care and Respiratory Medicine) and Mr Shaun Preston (Consultant Upper Gastrointestinal Surgeon) for their help with the clinical management of patients in this series. References 1 Boerhaave H . Atrocis, nec descripti pruis, morbi historia. Secundum medicae artis leges consripta. Lugduni Batavorum Boutesteniana 1724 (English translation: Derbes VJ, Mitchell RE . Bull Med Libr Assoc 1955 ; 43 : 217 – 240 ). Google Scholar PubMed OpenURL Placeholder Text WorldCat 2 Bufkin BL , Miller JI, Mansour KA. Esophageal perforation: emphasis on management . Ann Thorac Surg 1996 ; 61 : 1447 – 1451 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Walker WS , Cameron EW, Walbaum PR. Diagnosis and management of spontaneous transmural rupture of the oesophagus (Boerhaave's syndrome) . Br J Surg 1985 ; 72 : 204 – 207 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Curci JJ , Horman MJ. Boerhaave's syndrome: the importance of early diagnosis and treatment . Ann Surg 1976 ; 183 : 401 – 408 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Steadman C , Kerlin P, Crimmins F, Bell J, Robinson D, Dorrington L et al. Spontaneous intramural rupture of the oesophagus . Gut 1990 ; 31 : 845 – 849 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Khan AZ , Strauss D, Mason RC. Boerhaave's syndrome: diagnosis and surgical management . Surgeon 2007 ; 5 : 39 – 44 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Vogel SB , Rout WR, Martin TD, Abbitt PL. Esophageal perforation in adults: aggressive conservative treatment lowers morbidity and mortality . Ann Surg 2005 ; 241 : 1016 – 1021 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Naylor AR , Walker WS, Dark J, Cameron EW. Τ tube intubation in the management of seriously ill patients with oesophagopleural fistulae . Br J Surg 1990 ; 77 : 40 – 42 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Gupta NM , Kaman L. Personal management of 57 consecutive patients with esophageal perforation . Am J Surg 2004 ; 187 : 58 – 63 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Shenfine J , Griffin SM. Oesophageal emergencies. In A Companion to Specialist Surgical Practice: Oesophageal Surgery (3rd edn), Griffin SM, Raimes SA (eds). Elsevier-Saunders : Philadelphia , 2006 ; 365 – 392 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 11 Bone RC , Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine . Chest 1992 ; 101 : 1644 – 1655 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Mackler S . Spontaneous rupture of the esophagus; an experimental and clinical study . Surg Gynecol Obstet 1952 ; 95 : 345 – 356 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 13 Shenfine J , Hayes N, Richardson DL, Griffin SM. Combined percutaneous–endoscopic management of a perforated esophagus: a novel technique . Gastrointest Endosc 2001 ; 54 : 649 – 651 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Han SY , McElvein RB, Aldrete JS, Tishler JM. Perforation of the esophagus: correlation of site and cause with plain film findings . AJR Am J Roentgenol 1985 ; 145 : 537 – 540 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Keighley MR , Girdwood RW, Ionescu MI, Wooler JH. Spontaneous rupture of the oesophagus. Avoidance of postoperative morbidity . Br J Surg 1972 ; 59 : 649 – 652 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Reeder LB , DeFilippi VJ, Ferguson MK. Current results of therapy for esophageal perforation . Am J Surg 1995 ; 169 : 615 – 617 . Google Scholar Crossref Search ADS PubMed WorldCat 17 Skinner DB , Little AG, DeMeester TR. Manage- ment of esophageal perforation . Am J Surg 1980 ; 139 : 760 – 764 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Fan ST , Lau WY, Yip WC, Poon GP, Yeung C, Wong KK. Limitations and dangers of Gastrografin swallow after esophageal and upper gastric operations . Am J Surg 1988 ; 155 : 495 – 497 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Lamb PJ , Griffin SM, Chandrashekar MV, Richardson DL, Karat D, Hayes N. Prospective study of routine contrast radiology after total gastrectomy . Br J Surg 2004 ; 91 : 1015 – 1019 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Fadoo F , Ruiz DE, Dawn SK, Webb WR, Gotway MB. Helical CT esophagography for the evaluation of suspected esophageal perforation or rupture . AJR Am J Roentgenol 2004 ; 182 : 1177 – 1179 . Google Scholar Crossref Search ADS PubMed WorldCat 21 Langweiler TE , Steffani KD, Bogoevski DP, Mann O, Izbicki JR. Spontaneous pneumomediastinum . Ann Thorac Surg 2004 ; 78 : 711 – 713 . Google Scholar Crossref Search ADS PubMed WorldCat 22 Griffin SM , Lamb PJ, Dresner SM, Richardson DL, Hayes N. Diagnosis and management of a mediastinal leak following radical oesophagectomy . Br J Surg 2001 ; 88 : 1346 – 1351 . Google Scholar Crossref Search ADS PubMed WorldCat 23 Horwitz B , Krevsky B, Buckman RF, Fisher RS, Dabezies MA. Endoscopic evaluation of penetrating esophageal injuries . Am J Gastroenterol 1993 ; 88 : 1249 – 1253 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 24 Barrett NR . Report of a case of spontaneous perforation of the oesophagus successfully treated by operation . Br J Surg 1947 ; 35 : 216 – 218 . Google Scholar Crossref Search ADS PubMed WorldCat 25 Cameron JL , Kieffer RF, Hendrix TR, Mehigan DG, Baker RR. Selective non-operative management of contained intrathoracic esophageal disruptions . Ann Thorac Surg 1979 ; 27 : 404 – 408 . Google Scholar Crossref Search ADS PubMed WorldCat 26 Andersen OS , Giustra PE. Nonoperative management of contained esophageal perforation . Arch Surg 1981 ; 116 : 1214 – 1217 . Google Scholar Crossref Search ADS PubMed WorldCat 27 Altorjay A , Kiss J, Voros A, Bohak A. Nonoperative management of esophageal perforations. Is it justified? Ann Surg 1997 ; 225 : 415 – 421 . Google Scholar Crossref Search ADS PubMed WorldCat 28 Altorjay A , Kiss J, Voros A, Sziranyi E. The role of esophagectomy in the management of esophageal perforations . Ann Thorac Surg 1998 ; 65 : 1433 – 1436 . Google Scholar Crossref Search ADS PubMed WorldCat 29 Lundell L , Liedman B, Hyltander A. Emergency oesophagectomy and proximal deviating oesophagostomy for fulminent mediastinal sepsis . Eur J Surg 2001 ; 167 : 675 – 678 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 30 Brauer RB , Liebermann-Meffert D, Stein HJ, Bartels H, Siewert JR. Boerhaave's syndrome: analysis of the literature and report of 18 new cases . Dis Esophagus 1997 ; 10 : 64 – 68 . Google Scholar Crossref Search ADS PubMed WorldCat 31 Johnsson E , Lundell L, Liedman B. Sealing of esophageal perforation or ruptures with expandable metallic stents: a prospective controlled study on treatment efficacy and limitations . Dis Esophagus 2005 ; 18 : 262 – 266 . Google Scholar Crossref Search ADS PubMed WorldCat 32 Fischer A , Thomusch O, Benz S, von Dobschuet Z, Baier P, Hopt U. Non-operative treatment of 15 benign esophageal perforations with self-expandable covered metal stents . Ann Thorac Surg 2006 ; 81 : 467 – 472 . Google Scholar Crossref Search ADS PubMed WorldCat 33 Di Franco F , Lamb PJ, Karat D, Hayes N, Griffin SM. Iatrogenic perforation of localized oesophageal cancer . Br J Surg 2008 ; 95 : 837 – 839 . Google Scholar Crossref Search ADS PubMed WorldCat Copyright © 2008 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) Copyright © 2008 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd.

Djärv, T; Lagergren, J; Blazeby, J M; Lagergren, P

doi: 10.1002/bjs.6293pmid: 18581441

Abstract Background The aim of the study was to assess health-related quality of life (HRQL) in patients with surgically cured oesophageal cancer. Methods A Swedish nationwide cohort of patients undergoing oesophagectomy for cancer between April 2001 and January 2004 was studied prospectively, and compared with a Swedish age- and sex-adjusted reference population. Validated European Organisation for Research and Treatment of Cancer quality of life questionnaires were used to assess HRQL at 6 months and 3 years after surgery. A mean score difference of 10 or more between groups was considered clinically relevant and tested further for statistical significance. Results Of 358 patients, 117 (32·7 per cent) survived for at least 3 years. Of these, 87 patients (74·4 per cent) responded to the questionnaires. Six months after surgery, most aspects of HRQL were substantially worse than in the reference population with no improvement at 3 years. Patients alive at 3 years reported significantly poorer role and social function, and significantly more problems with fatigue, diarrhoea, appetite loss, nausea and vomiting, than in the reference population. Conclusion HRQL in long-term survivors after oesophagectomy does not improve between 6 months and 3 years after surgery, and is worse than that in a comparable reference population. Introduction The effects of oesophagectomy on long-term health-related quality of life (HRQL) in patients cured surgically of oesophageal cancer deserve attention. Although surgery offers a chance of cure, it is associated with a substantial risk of tumour recurrence, typically occurring within 1 year, and more than 60 per cent of patients undergoing oesophagectomy die within 3 years1. Patients who survive for 3 years, however, may be considered cured, as late tumour recurrences are rare2. Oesophageal resection often entails high morbidity and a greatly reduced HRQL in the early postoperative years3,4. Population-based studies have shown that 6 months after surgery patients have a significantly poorer HRQL than a background population5. Little is known, however, about long-term HRQL in patients who are considered cured, with a long remaining life expectancy and a HRQL that is not affected by tumour recurrence. Specific symptoms, functions or problems related to oesophageal surgery may be transient or persistent. To address the hypothesis that HRQL improves gradually over time in patients with oesophageal cancer cured by surgery, a prospective population-based study of the postoperative development of HRQL among long-term survivors was undertaken, and their HRQL scores were compared with those of the background population. Methods A prospective, nationwide, population-based cohort study was carried out in Sweden between 2 April 2001 and 31 January 2007. All patients in Sweden with oesophageal cancer who had survived for at least 3 years after curative surgery (performed between 2 April 2001 and 31 January 2004) were eligible for inclusion in the study. Data were collected prospectively via the Swedish Oesophageal and Cardia Cancer Register, which included almost all eligible patients6. The means by which rapid case ascertainment and prospective data collection among surgically treated patients were achieved have been presented elsewhere5,7. Details of patient and tumour characteristics, surgical procedures and complications were collected from surgical, histopathological, endoscopic, radiological and intensive care unit records, and assessed using a standard protocol. Tumour location and stage were classified according to established definitions8,9. Mortality was determined by linkage to the complete Swedish Register of the Total Population. Informed consent was obtained from all patients before inclusion in the study, which was approved by the ethics committee of Karolinska University Hospital and the Karolinska Institute. Health-related quality of life measurements HRQL was assessed 6 months and 3 years after surgery by means of mailed self-administered questionnaires, developed and validated by the European Organisation for Research and Treatment of Cancer (EORTC). The EORTC Quality of Life Questionnaire (QLQ)-C30 version 3.010 was used to measure core aspects of HRQL, and site-specific issues were addressed by the oesophageal-specific module, QLQ-OES1811. The QLQ-C30 consisted of one global HRQL scale, five functional scales (physical, role, emotional, cognitive and social), three symptom scales (fatigue, nausea or vomiting, and pain) and six single items (sleeping disorders, appetite loss, dyspnoea, diarrhoea, constipation and financial problems). Each item had four response alternatives (1, not at all; 2, a little; 3, quite a bit; 4, very much), and the global quality of life scale had a seven-step scale ranging from very poor to excellent. The QLQ-OES18 included four symptom scales (dysphagia, eating difficulties, reflux and pain) and six single items (trouble swallowing saliva, choking, dry mouth, taste, cough and speech); response alternatives were as for the QLQ-C30. Statistical analysis All questionnaire responses were transformed linearly into scores on a 0–100 scale according to the EORTC scoring manual12. A high score on functional scales and the global HRQL scale represented a high level of function and better quality of life respectively, whereas a high score on symptom scales and items indicated increased symptoms. Mean scores with 95 per cent confidence intervals were calculated. To provide a surrogate baseline representing HRQL before the development of oesophageal cancer, a Swedish reference population for the QLQ-C30 was used13. This reference was matched with regard to age and all comparisons with this reference group were stratified by sex. No such reference population was available for the QLQ-OES18. Based on previous research14, a difference of at least 10 mean score points between time points or comparison groups was considered clinically relevant, and a difference of 5–10 was considered weak. When the mean scores differed by 10 points or more, a two-sample t test was used to examine statistical significance at the 0·050 level. Missing responses were handled according to the EORTC scoring manual12. All analyses were performed with the statistical software Stata® 9·2 (Stata Corporation, College Station, Texas, USA). Results During the study, 358 patients underwent potentially curative surgery for oesophageal cancer. Seventy-eight patients (21·8 per cent) died before the 6-month assessment and a further 163 (45·5 per cent) died within 3 years of oesophagectomy. The remaining 117 patients (32·7 per cent), who survived for at least 3 years after surgery, were eligible for inclusion in the questionnaire study. Of these, 93 (79·5 per cent) returned the 6-month questionnaires and 87 (74·4 per cent) the 3-year questionnaires. Reasons for non-participation at 6 months were administrative error (18 patients) or patient choice (six). Non-participation at 3 years was due to administrative error (20), patient choice (nine) and emigration (one). Only 78 of the 93 patients who responded at 6 months also completed the 3-year questionnaires and were eligible for analysis. Patient, tumour and treatment characteristics are shown in Table 1. Table 1 Patient and tumour characteristics for 87 patients with surgically cured oesophageal cancer . No. of patients . Sex  M 62 (71)  F 25 (29) Age (years)  < 60 27 (31)  60–69 28 (31)  70–84 32 (37) Co-morbidity  No 30 (34)  Yes 57 (66) Histological type of tumour  Adenocarcinoma and dysplasia 68 (78)  Squamous cell carcinoma 19 (22) Tumour location  Upper and middle oesophagus 12 (14)  Lower oesophagus 39 (45)  Cardia (Siewert II–III) 36 (41) Tumour stage  0–I 34 (39)  II 36 (41)  III 13 (15)  IV 3 (3)  Unknown 1 (1) Surgical approach  Transthoracic 68 (78)  Transhiatal (abdominal only) 19 (22) Substitute  Stomach 69 (79)  Jejunum 16 (18)  Colon 2 (2) Anastomosis  Stapled 56 (64)  Handsewn 31 (36) Complications  No 51 (59)  Yes 36 (41) . No. of patients . Sex  M 62 (71)  F 25 (29) Age (years)  < 60 27 (31)  60–69 28 (31)  70–84 32 (37) Co-morbidity  No 30 (34)  Yes 57 (66) Histological type of tumour  Adenocarcinoma and dysplasia 68 (78)  Squamous cell carcinoma 19 (22) Tumour location  Upper and middle oesophagus 12 (14)  Lower oesophagus 39 (45)  Cardia (Siewert II–III) 36 (41) Tumour stage  0–I 34 (39)  II 36 (41)  III 13 (15)  IV 3 (3)  Unknown 1 (1) Surgical approach  Transthoracic 68 (78)  Transhiatal (abdominal only) 19 (22) Substitute  Stomach 69 (79)  Jejunum 16 (18)  Colon 2 (2) Anastomosis  Stapled 56 (64)  Handsewn 31 (36) Complications  No 51 (59)  Yes 36 (41) Values in parentheses are percentages. Open in new tab Table 1 Patient and tumour characteristics for 87 patients with surgically cured oesophageal cancer . No. of patients . Sex  M 62 (71)  F 25 (29) Age (years)  < 60 27 (31)  60–69 28 (31)  70–84 32 (37) Co-morbidity  No 30 (34)  Yes 57 (66) Histological type of tumour  Adenocarcinoma and dysplasia 68 (78)  Squamous cell carcinoma 19 (22) Tumour location  Upper and middle oesophagus 12 (14)  Lower oesophagus 39 (45)  Cardia (Siewert II–III) 36 (41) Tumour stage  0–I 34 (39)  II 36 (41)  III 13 (15)  IV 3 (3)  Unknown 1 (1) Surgical approach  Transthoracic 68 (78)  Transhiatal (abdominal only) 19 (22) Substitute  Stomach 69 (79)  Jejunum 16 (18)  Colon 2 (2) Anastomosis  Stapled 56 (64)  Handsewn 31 (36) Complications  No 51 (59)  Yes 36 (41) . No. of patients . Sex  M 62 (71)  F 25 (29) Age (years)  < 60 27 (31)  60–69 28 (31)  70–84 32 (37) Co-morbidity  No 30 (34)  Yes 57 (66) Histological type of tumour  Adenocarcinoma and dysplasia 68 (78)  Squamous cell carcinoma 19 (22) Tumour location  Upper and middle oesophagus 12 (14)  Lower oesophagus 39 (45)  Cardia (Siewert II–III) 36 (41) Tumour stage  0–I 34 (39)  II 36 (41)  III 13 (15)  IV 3 (3)  Unknown 1 (1) Surgical approach  Transthoracic 68 (78)  Transhiatal (abdominal only) 19 (22) Substitute  Stomach 69 (79)  Jejunum 16 (18)  Colon 2 (2) Anastomosis  Stapled 56 (64)  Handsewn 31 (36) Complications  No 51 (59)  Yes 36 (41) Values in parentheses are percentages. Open in new tab There were no clinically relevant differences (score of 10 or more) in mean scores at 6 months and 3 years. In general, mean scores were similar between the two assessments (Table 2). For weak mean score differences (score 5–10), improvements in the general symptoms of fatigue, appetite loss, diarrhoea and financial difficulties were suggested, along with worsening of the oesophageal-specific symptoms of dysphagia, reflux and dryness of the mouth (Table 2). Between 6 months and 3 years after surgery, most patients remained at the same level of HRQL (mean score change within ± 10 points); the proportions of patients who improved or deteriorated were similar (Table 3). In terms of physical function, about half of the patients (50 per cent) stayed at the same level, and improvement or deterioration was seen in 30 and 18 per cent of patients respectively. Dysphagia improved in 29 per cent and deteriorated in 32 per cent. The greatest improvement was seen for fatigue (47 per cent) and oesophageal pain (45 per cent); 42 per cent of patients reported increased problems with reflux (Table 3). Table 2 Health-related quality of life scores in 87 patients with surgically cured oesophageal cancer . Mean score . . 6 months (n = 78) . 3 years (n = 87) . Global quality of life scale 66 (61, 71) 69 (64, 73) Functioning scales  Physical function 80 (75, 85) 81 (77, 85)  Role function 73 (67, 80) 74 (67, 80)  Emotional function 75 (69, 80) 77 (72, 82)  Cognitive function 82 (78, 87) 81 (77, 86)  Social function 76 (70, 82) 78 (72, 84) General symptom scales  Fatigue 40 (34, 47) 35 (29, 41)*  Nausea and vomiting 16 (11, 22) 15 (11, 19)  Pain 18 (12, 23) 16 (11, 21) General symptom items  Dyspnoea 25 (19, 31) 27 (21, 33)  Insomnia 22 (16, 28) 25 (18, 32)  Appetite loss 27 (20, 34) 18 (12, 25)*  Constipation 6 (2, 11) 10 (6, 15)  Diarrhoea 35 (28, 43) 29 (23, 35)*  Financial difficulties 15 (9, 21) 10 (5, 15)* Oesophageal-specific symptom scales  Dysphagia 16 (12, 21) 23 (16, 29)*  Eating difficulties 31 (25, 36) 28 (22, 33)  Reflux 29 (21, 36) 35 (28, 41)*  Oesophageal pain 23 (18, 28) 20 (15, 25) Oesophageal-specific items  Trouble swallowing saliva 11 (5, 17) 14 (9, 20)  Choking 18 (13, 24) 15 (10, 20)  Dry mouth 23 (16, 31) 32 (24, 39)*  Taste problems 17 (10, 24) 16 (11, 23)  Cough 24 (18, 31) 23 (17, 28)  Speech difficulties 10 (4, 15) 7 (3, 10) . Mean score . . 6 months (n = 78) . 3 years (n = 87) . Global quality of life scale 66 (61, 71) 69 (64, 73) Functioning scales  Physical function 80 (75, 85) 81 (77, 85)  Role function 73 (67, 80) 74 (67, 80)  Emotional function 75 (69, 80) 77 (72, 82)  Cognitive function 82 (78, 87) 81 (77, 86)  Social function 76 (70, 82) 78 (72, 84) General symptom scales  Fatigue 40 (34, 47) 35 (29, 41)*  Nausea and vomiting 16 (11, 22) 15 (11, 19)  Pain 18 (12, 23) 16 (11, 21) General symptom items  Dyspnoea 25 (19, 31) 27 (21, 33)  Insomnia 22 (16, 28) 25 (18, 32)  Appetite loss 27 (20, 34) 18 (12, 25)*  Constipation 6 (2, 11) 10 (6, 15)  Diarrhoea 35 (28, 43) 29 (23, 35)*  Financial difficulties 15 (9, 21) 10 (5, 15)* Oesophageal-specific symptom scales  Dysphagia 16 (12, 21) 23 (16, 29)*  Eating difficulties 31 (25, 36) 28 (22, 33)  Reflux 29 (21, 36) 35 (28, 41)*  Oesophageal pain 23 (18, 28) 20 (15, 25) Oesophageal-specific items  Trouble swallowing saliva 11 (5, 17) 14 (9, 20)  Choking 18 (13, 24) 15 (10, 20)  Dry mouth 23 (16, 31) 32 (24, 39)*  Taste problems 17 (10, 24) 16 (11, 23)  Cough 24 (18, 31) 23 (17, 28)  Speech difficulties 10 (4, 15) 7 (3, 10) Values in parentheses are 95 per cent confidence intervals. * Weak mean score difference (5–10). Open in new tab Table 2 Health-related quality of life scores in 87 patients with surgically cured oesophageal cancer . Mean score . . 6 months (n = 78) . 3 years (n = 87) . Global quality of life scale 66 (61, 71) 69 (64, 73) Functioning scales  Physical function 80 (75, 85) 81 (77, 85)  Role function 73 (67, 80) 74 (67, 80)  Emotional function 75 (69, 80) 77 (72, 82)  Cognitive function 82 (78, 87) 81 (77, 86)  Social function 76 (70, 82) 78 (72, 84) General symptom scales  Fatigue 40 (34, 47) 35 (29, 41)*  Nausea and vomiting 16 (11, 22) 15 (11, 19)  Pain 18 (12, 23) 16 (11, 21) General symptom items  Dyspnoea 25 (19, 31) 27 (21, 33)  Insomnia 22 (16, 28) 25 (18, 32)  Appetite loss 27 (20, 34) 18 (12, 25)*  Constipation 6 (2, 11) 10 (6, 15)  Diarrhoea 35 (28, 43) 29 (23, 35)*  Financial difficulties 15 (9, 21) 10 (5, 15)* Oesophageal-specific symptom scales  Dysphagia 16 (12, 21) 23 (16, 29)*  Eating difficulties 31 (25, 36) 28 (22, 33)  Reflux 29 (21, 36) 35 (28, 41)*  Oesophageal pain 23 (18, 28) 20 (15, 25) Oesophageal-specific items  Trouble swallowing saliva 11 (5, 17) 14 (9, 20)  Choking 18 (13, 24) 15 (10, 20)  Dry mouth 23 (16, 31) 32 (24, 39)*  Taste problems 17 (10, 24) 16 (11, 23)  Cough 24 (18, 31) 23 (17, 28)  Speech difficulties 10 (4, 15) 7 (3, 10) . Mean score . . 6 months (n = 78) . 3 years (n = 87) . Global quality of life scale 66 (61, 71) 69 (64, 73) Functioning scales  Physical function 80 (75, 85) 81 (77, 85)  Role function 73 (67, 80) 74 (67, 80)  Emotional function 75 (69, 80) 77 (72, 82)  Cognitive function 82 (78, 87) 81 (77, 86)  Social function 76 (70, 82) 78 (72, 84) General symptom scales  Fatigue 40 (34, 47) 35 (29, 41)*  Nausea and vomiting 16 (11, 22) 15 (11, 19)  Pain 18 (12, 23) 16 (11, 21) General symptom items  Dyspnoea 25 (19, 31) 27 (21, 33)  Insomnia 22 (16, 28) 25 (18, 32)  Appetite loss 27 (20, 34) 18 (12, 25)*  Constipation 6 (2, 11) 10 (6, 15)  Diarrhoea 35 (28, 43) 29 (23, 35)*  Financial difficulties 15 (9, 21) 10 (5, 15)* Oesophageal-specific symptom scales  Dysphagia 16 (12, 21) 23 (16, 29)*  Eating difficulties 31 (25, 36) 28 (22, 33)  Reflux 29 (21, 36) 35 (28, 41)*  Oesophageal pain 23 (18, 28) 20 (15, 25) Oesophageal-specific items  Trouble swallowing saliva 11 (5, 17) 14 (9, 20)  Choking 18 (13, 24) 15 (10, 20)  Dry mouth 23 (16, 31) 32 (24, 39)*  Taste problems 17 (10, 24) 16 (11, 23)  Cough 24 (18, 31) 23 (17, 28)  Speech difficulties 10 (4, 15) 7 (3, 10) Values in parentheses are 95 per cent confidence intervals. * Weak mean score difference (5–10). Open in new tab Table 3 Postoperative change in health-related quality of life between 6 months and 3 years in the 76 patients with oesophageal cancer who responded to both assessments . Worse* . Stable† . Improved‡ . Global quality of life scale 18 (24) 31 (41) 25 (33) Functioning scales  Physical function 14 (18) 38 (50) 23 (30)  Role function 19 (25) 26 (34) 28 (37)  Emotional function 13 (17) 35 (46) 25 (33)  Cognitive function 16 (21) 38 (50) 20 (26)  Social function 21 (28) 31 (41) 22 (29) General symptom scales  Fatigue 22 (29) 16 (21) 36 (47)  Nausea and vomiting 22 (29) 32 (42) 22 (29)  Pain 16 (21) 35 (46) 24 (32) General symptom items  Dyspnoea 16 (21) 46 (61) 12 (16)  Insomnia 22 (29) 30 (39) 22 (29)  Appetite loss 12 (16) 36 (47) 26 (34)  Constipation 11 (14) 55 (72) 6 (8)  Diarrhoea 14 (18) 34 (45) 25 (33)  Financial difficulties 8 (11) 52 (68) 14 (18) Oesophageal-specific symptom scales  Dysphagia 24 (32) 30 (39) 22 (29)  Eating difficulties 13 (17) 30 (39) 29 (38)  Reflux 32 (42) 16 (21) 25 (33)  Oesophageal pain 23 (30) 17 (22) 34 (45) Oesophageal-specific items  Trouble swallowing saliva 13 (17) 50 (66) 8 (11)  Choking 10 (13) 44 (58) 16 (21)  Dry mouth 18 (24) 39 (51) 14 (18)  Taste problems 10 (13) 45 (59) 15 (20)  Cough 14 (18) 42 (55) 17 (22)  Speech difficulties 5 (7) 58 (76) 10 (13) . Worse* . Stable† . Improved‡ . Global quality of life scale 18 (24) 31 (41) 25 (33) Functioning scales  Physical function 14 (18) 38 (50) 23 (30)  Role function 19 (25) 26 (34) 28 (37)  Emotional function 13 (17) 35 (46) 25 (33)  Cognitive function 16 (21) 38 (50) 20 (26)  Social function 21 (28) 31 (41) 22 (29) General symptom scales  Fatigue 22 (29) 16 (21) 36 (47)  Nausea and vomiting 22 (29) 32 (42) 22 (29)  Pain 16 (21) 35 (46) 24 (32) General symptom items  Dyspnoea 16 (21) 46 (61) 12 (16)  Insomnia 22 (29) 30 (39) 22 (29)  Appetite loss 12 (16) 36 (47) 26 (34)  Constipation 11 (14) 55 (72) 6 (8)  Diarrhoea 14 (18) 34 (45) 25 (33)  Financial difficulties 8 (11) 52 (68) 14 (18) Oesophageal-specific symptom scales  Dysphagia 24 (32) 30 (39) 22 (29)  Eating difficulties 13 (17) 30 (39) 29 (38)  Reflux 32 (42) 16 (21) 25 (33)  Oesophageal pain 23 (30) 17 (22) 34 (45) Oesophageal-specific items  Trouble swallowing saliva 13 (17) 50 (66) 8 (11)  Choking 10 (13) 44 (58) 16 (21)  Dry mouth 18 (24) 39 (51) 14 (18)  Taste problems 10 (13) 45 (59) 15 (20)  Cough 14 (18) 42 (55) 17 (22)  Speech difficulties 5 (7) 58 (76) 10 (13) Values in parentheses are percentages. Owing to missing data total for each variable does not add up to 100 per cent. * Mean score deterioration of at least 10; † mean score change within ± 10; ‡ mean score improvement of at least 10 or within 10 points of maximum score. Open in new tab Table 3 Postoperative change in health-related quality of life between 6 months and 3 years in the 76 patients with oesophageal cancer who responded to both assessments . Worse* . Stable† . Improved‡ . Global quality of life scale 18 (24) 31 (41) 25 (33) Functioning scales  Physical function 14 (18) 38 (50) 23 (30)  Role function 19 (25) 26 (34) 28 (37)  Emotional function 13 (17) 35 (46) 25 (33)  Cognitive function 16 (21) 38 (50) 20 (26)  Social function 21 (28) 31 (41) 22 (29) General symptom scales  Fatigue 22 (29) 16 (21) 36 (47)  Nausea and vomiting 22 (29) 32 (42) 22 (29)  Pain 16 (21) 35 (46) 24 (32) General symptom items  Dyspnoea 16 (21) 46 (61) 12 (16)  Insomnia 22 (29) 30 (39) 22 (29)  Appetite loss 12 (16) 36 (47) 26 (34)  Constipation 11 (14) 55 (72) 6 (8)  Diarrhoea 14 (18) 34 (45) 25 (33)  Financial difficulties 8 (11) 52 (68) 14 (18) Oesophageal-specific symptom scales  Dysphagia 24 (32) 30 (39) 22 (29)  Eating difficulties 13 (17) 30 (39) 29 (38)  Reflux 32 (42) 16 (21) 25 (33)  Oesophageal pain 23 (30) 17 (22) 34 (45) Oesophageal-specific items  Trouble swallowing saliva 13 (17) 50 (66) 8 (11)  Choking 10 (13) 44 (58) 16 (21)  Dry mouth 18 (24) 39 (51) 14 (18)  Taste problems 10 (13) 45 (59) 15 (20)  Cough 14 (18) 42 (55) 17 (22)  Speech difficulties 5 (7) 58 (76) 10 (13) . Worse* . Stable† . Improved‡ . Global quality of life scale 18 (24) 31 (41) 25 (33) Functioning scales  Physical function 14 (18) 38 (50) 23 (30)  Role function 19 (25) 26 (34) 28 (37)  Emotional function 13 (17) 35 (46) 25 (33)  Cognitive function 16 (21) 38 (50) 20 (26)  Social function 21 (28) 31 (41) 22 (29) General symptom scales  Fatigue 22 (29) 16 (21) 36 (47)  Nausea and vomiting 22 (29) 32 (42) 22 (29)  Pain 16 (21) 35 (46) 24 (32) General symptom items  Dyspnoea 16 (21) 46 (61) 12 (16)  Insomnia 22 (29) 30 (39) 22 (29)  Appetite loss 12 (16) 36 (47) 26 (34)  Constipation 11 (14) 55 (72) 6 (8)  Diarrhoea 14 (18) 34 (45) 25 (33)  Financial difficulties 8 (11) 52 (68) 14 (18) Oesophageal-specific symptom scales  Dysphagia 24 (32) 30 (39) 22 (29)  Eating difficulties 13 (17) 30 (39) 29 (38)  Reflux 32 (42) 16 (21) 25 (33)  Oesophageal pain 23 (30) 17 (22) 34 (45) Oesophageal-specific items  Trouble swallowing saliva 13 (17) 50 (66) 8 (11)  Choking 10 (13) 44 (58) 16 (21)  Dry mouth 18 (24) 39 (51) 14 (18)  Taste problems 10 (13) 45 (59) 15 (20)  Cough 14 (18) 42 (55) 17 (22)  Speech difficulties 5 (7) 58 (76) 10 (13) Values in parentheses are percentages. Owing to missing data total for each variable does not add up to 100 per cent. * Mean score deterioration of at least 10; † mean score change within ± 10; ‡ mean score improvement of at least 10 or within 10 points of maximum score. Open in new tab Compared with the overall results, stratification into sex and age groups (less than 60, 60–69, and 70 years or more), and separate analyses of patients with tumour stage 0–I or of those who had total gastrectomy alone for cardia cancer, did not reveal any clinically relevant differences (data not shown). When mean HRQL scores from the QLQ-C30 questionnaire at 3 years after oesophagectomy were compared with those in an age- and sex-matched random sample of the Swedish general population, both male and female cohort participants showed a clinically relevant and statistically poorer HRQL (Table 4). Scores were worse for role and social functions, fatigue, nausea and vomiting, appetite loss and diarrhoea in both sexes. Women in the cohort also reported a poorer global quality of life, and physical, emotional and cognitive function scores than the female reference population. Table 4 Health-related quality of life scores at 3 years in 87 patients with surgically cured oesophageal cancer compared with a reference Swedish population . Mean score . Mean score . . Male study patients (n = 62) . Male reference 60–69 years (n = 278) . P* . Female study patients (n = 25) . Female reference 60–69 years (n = 271) . P* . Global quality of life scale 70 (64, 75) 77 64 (56, 73) 78 < 0·001 Functioning scales  Physical function 83 (78, 88) 88 77 (67, 85) 87 0·032  Role function 75 (67, 83) 87 0·003 71 (56, 85) 87 0·035  Emotional function 80 (74, 85) 86 70 (57, 82) 84 0·028  Cognitive function 82 (77, 87) 87 80 (70, 90) 90 0·056  Social function 77 (70, 84) 91 < 0·001 80 (68, 92) 92 0·015 General symptom scales  Fatigue 33 (26, 40) 20 < 0·001 39 (28, 51) 20 0·002  Nausea and vomiting 13 (8, 18) 2 < 0·001 21 (11, 32) 3 0·002  Pain 15 (9, 21) 18 21 (11, 31) 21 General symptom items  Dyspnoea 28 (21, 35) 19 25 (10, 40) 16  Insomnia 22 (14, 29) 17 32 (18, 49) 23  Appetite loss 15 (9, 21) 2 < 0·001 22 (8, 37) 4 0·015  Constipation 8 (4, 13) 4 16 (6, 26) 9  Diarrhoea 28 (20, 35) 5 < 0·001 31 (18, 44) 4 < 0·001  Financial difficulties 11 (5, 17) 5 9 (1, 19) 7 . Mean score . Mean score . . Male study patients (n = 62) . Male reference 60–69 years (n = 278) . P* . Female study patients (n = 25) . Female reference 60–69 years (n = 271) . P* . Global quality of life scale 70 (64, 75) 77 64 (56, 73) 78 < 0·001 Functioning scales  Physical function 83 (78, 88) 88 77 (67, 85) 87 0·032  Role function 75 (67, 83) 87 0·003 71 (56, 85) 87 0·035  Emotional function 80 (74, 85) 86 70 (57, 82) 84 0·028  Cognitive function 82 (77, 87) 87 80 (70, 90) 90 0·056  Social function 77 (70, 84) 91 < 0·001 80 (68, 92) 92 0·015 General symptom scales  Fatigue 33 (26, 40) 20 < 0·001 39 (28, 51) 20 0·002  Nausea and vomiting 13 (8, 18) 2 < 0·001 21 (11, 32) 3 0·002  Pain 15 (9, 21) 18 21 (11, 31) 21 General symptom items  Dyspnoea 28 (21, 35) 19 25 (10, 40) 16  Insomnia 22 (14, 29) 17 32 (18, 49) 23  Appetite loss 15 (9, 21) 2 < 0·001 22 (8, 37) 4 0·015  Constipation 8 (4, 13) 4 16 (6, 26) 9  Diarrhoea 28 (20, 35) 5 < 0·001 31 (18, 44) 4 < 0·001  Financial difficulties 11 (5, 17) 5 9 (1, 19) 7 Values in parentheses are 95 per cent confidence intervals. * Two-sample t test performed when mean scores differed by 10 points or more. Open in new tab Table 4 Health-related quality of life scores at 3 years in 87 patients with surgically cured oesophageal cancer compared with a reference Swedish population . Mean score . Mean score . . Male study patients (n = 62) . Male reference 60–69 years (n = 278) . P* . Female study patients (n = 25) . Female reference 60–69 years (n = 271) . P* . Global quality of life scale 70 (64, 75) 77 64 (56, 73) 78 < 0·001 Functioning scales  Physical function 83 (78, 88) 88 77 (67, 85) 87 0·032  Role function 75 (67, 83) 87 0·003 71 (56, 85) 87 0·035  Emotional function 80 (74, 85) 86 70 (57, 82) 84 0·028  Cognitive function 82 (77, 87) 87 80 (70, 90) 90 0·056  Social function 77 (70, 84) 91 < 0·001 80 (68, 92) 92 0·015 General symptom scales  Fatigue 33 (26, 40) 20 < 0·001 39 (28, 51) 20 0·002  Nausea and vomiting 13 (8, 18) 2 < 0·001 21 (11, 32) 3 0·002  Pain 15 (9, 21) 18 21 (11, 31) 21 General symptom items  Dyspnoea 28 (21, 35) 19 25 (10, 40) 16  Insomnia 22 (14, 29) 17 32 (18, 49) 23  Appetite loss 15 (9, 21) 2 < 0·001 22 (8, 37) 4 0·015  Constipation 8 (4, 13) 4 16 (6, 26) 9  Diarrhoea 28 (20, 35) 5 < 0·001 31 (18, 44) 4 < 0·001  Financial difficulties 11 (5, 17) 5 9 (1, 19) 7 . Mean score . Mean score . . Male study patients (n = 62) . Male reference 60–69 years (n = 278) . P* . Female study patients (n = 25) . Female reference 60–69 years (n = 271) . P* . Global quality of life scale 70 (64, 75) 77 64 (56, 73) 78 < 0·001 Functioning scales  Physical function 83 (78, 88) 88 77 (67, 85) 87 0·032  Role function 75 (67, 83) 87 0·003 71 (56, 85) 87 0·035  Emotional function 80 (74, 85) 86 70 (57, 82) 84 0·028  Cognitive function 82 (77, 87) 87 80 (70, 90) 90 0·056  Social function 77 (70, 84) 91 < 0·001 80 (68, 92) 92 0·015 General symptom scales  Fatigue 33 (26, 40) 20 < 0·001 39 (28, 51) 20 0·002  Nausea and vomiting 13 (8, 18) 2 < 0·001 21 (11, 32) 3 0·002  Pain 15 (9, 21) 18 21 (11, 31) 21 General symptom items  Dyspnoea 28 (21, 35) 19 25 (10, 40) 16  Insomnia 22 (14, 29) 17 32 (18, 49) 23  Appetite loss 15 (9, 21) 2 < 0·001 22 (8, 37) 4 0·015  Constipation 8 (4, 13) 4 16 (6, 26) 9  Diarrhoea 28 (20, 35) 5 < 0·001 31 (18, 44) 4 < 0·001  Financial difficulties 11 (5, 17) 5 9 (1, 19) 7 Values in parentheses are 95 per cent confidence intervals. * Two-sample t test performed when mean scores differed by 10 points or more. Open in new tab Discussion In this population-based study of patients undergoing surgery for oesophageal cancer, HRQL was similar at 6 months and 3 years after surgery. HRQL for patients surviving for 3 years was also significantly poorer than that of the age- and sex-matched background population. The population-based and prospective design, in combination with a high participation frequency, serves to reduce the risk of selection bias and facilitates generalizability. Selection bias is probably not the explanation for the lack of improvement between 6 months and 3 years, as previous research has indicated that non-responders are more likely to have a poorer HRQL15. Moreover, there were no differences in clinical variables between patients who did and those who did not participate in the 3-year follow-up (data not shown). Other possible sources of selection bias are incomplete registration and non-participation; however, the number of registered patients showed good correspondence with the incidence and resection rate of oesophageal cancer in Sweden1, and the total participation rate was high. The risk of misclassification of HRQL was reduced by the use of questionnaires with documented good reliability and validity10,11. The lack of data for individual baseline HRQL is a disadvantage, but would not affect the main finding of the study—a lack of improvement between 6 months and 3 years after surgery. In addition, a large and randomly selected reference population, adjusted for age and sex, was used as a surrogate for individual baseline values. Such an approach may in fact be better than using preoperative HRQL as baseline, as patients' preoperative HRQL is likely to be affected substantially by the newly detected oesophageal cancer and impending major surgery. The use of HRQL reported by sex- and age-matched controls from the background population has been suggested by others to be a better alternative than individual preoperative HRQL data16. The 6-month time window was chosen carefully on the basis of previous research indicating that the acute postoperative phase will have subsided by 6 months4,17–19. The 3-year assessment was carried out at a time when the relative (disease specific) survival is similar to the survival of the background population in Sweden2. Finally, to minimize the risks arising from multiple testing, statistical analysis was performed only when differences between mean scores in the comparison groups were clinically relevant14. The finding of worsening of HRQL measures 6 months after oesophageal cancer surgery is in accordance with the available literature17,20–22. A smaller study has, however, provided conflicting results23. A novel finding is that this level of HRQL persists over time, and is lower than in the general population, notably among patients who have had curative surgery. Although HRQL usually deteriorates rapidly in patients with recurrent disease, patients with surgical cure of oesophageal cancer are thought to regain their baseline HRQL within 1–2 years of surgery3. In one UK study3, many aspects of HRQL had recovered to baseline levels by 3 years, but patients still had significantly poorer physical function, and more problems with diarrhoea, breathlessness and reflux than before surgery. Other studies have indicated quicker recovery after oesophageal cancer surgery, although methodological differences may explain these results20–22. There is no improvement in HRQL between 6 months and 3 years after surgery for oesophageal cancer, and patients report worse HRQL than an age- and sex-matched background population. This finding should help patients to be better informed, improve clinical decision making and follow-up, and facilitate the treatment of symptoms after surgery. Acknowledgements This study was funded by the Swedish Cancer Society. T.D. was supported by the Karolinska Institute, J.L. by the Swedish Research Council, and P.L. by the Swedish Cancer Society, Karolinska Institute and Karolinska University Hospital. References 1 Rouvelas I , Zeng W, Lindblad M, Viklund P, Ye W, Lagergren J. Survival after surgery for oesophageal cancer: a population-based study . Lancet Oncol 2005 ; 6 : 864 – 870 . Google Scholar Crossref Search ADS PubMed WorldCat 2 Sundelof M , Ye W, Dickman PW, Lagergren J. Improved survival in both histologic types of oesophageal cancer in Sweden . Int J Cancer 2002 ; 99 : 751 – 754 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Lagergren P , Avery KN, Hughes R, Barham CP, Alderson D, Falk SJ et al. Health-related quality of life among patients cured by surgery for esophageal cancer . Cancer 2007 ; 110 : 686 – 693 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Blazeby JM , Sanford E, Falk SJ, Alderson D, Donovan JL. Health-related quality of life during neoadjuvant treatment and surgery for localized esophageal carcinoma . Cancer 2005 ; 103 : 1791 – 1799 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Viklund P , Lindblad M, Lu M, Ye W, Johansson J, Lagergren J. Risk factors for complications after esophageal cancer resection: a prospective population-based study in Sweden . Ann Surg 2006 ; 243 : 204 – 211 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Viklund P , Lindblad M, Lagergren J. Influence of surgery-related factors on quality of life after esophageal or cardia cancer resection . World J Surg 2005 ; 29 : 841 – 848 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Lagergren J , Bergstrom R, Lindgren A, Nyren O. Symptomatic gastroesophageal reflux as a risk factor for esophageal adenocarcinoma . N Engl J Med 1999 ; 340 : 825 – 831 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Siewert JR , Stein HJ. Classification of adenocarcinoma of the oesophagogastric junction . Br J Surg 1998 ; 85 : 1457 – 1459 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 9 Sobin LH , Wittekind C. International Union Against Cancer: TNM Classification of Malignant Tumours (6th edn). Wiley-Liss : New York , 2002 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 10 Aaronson NK , Ahmedzai S, Bergman B, Bullinger M, Cull A, Duez NJ et al. The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology . J Natl Cancer Inst 1993 ; 85 : 365 – 376 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Blazeby JM , Conroy T, Hammerlid E, Fayers P, Sezer O, Koller M et al. Clinical and psychometric validation of an EORTC questionnaire module, the EORTC QLQ-OES18, to assess quality of life in patients with oesophageal cancer . Eur J Cancer 2003 ; 39 : 1384 – 1394 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Fayers PM Aaronson N, Bjordal K, Groenvold M, Curran D, Bottomley A, on behalf of the EORTC Quality of Life Group . The EORTC QLQ-C30 Scoring Manual (3rd edn). European Organisation for Research and Treatment of Cancer : Brussels , 2001 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 13 Michelson H , Bolund C, Nilsson B, Brandberg Y. Health-related quality of life measured by the EORTC QLQ-C30—reference values from a large sample of Swedish population . Acta Oncol 2000 ; 39 : 477 – 484 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 14 Osoba D , Rodrigues G, Myles J, Zee B, Pater J. Interpreting the significance of changes in health-related quality-of-life scores . J Clin Oncol 1998 ; 16 : 139 – 144 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Blazeby JM , Nicklin J, Brookes ST, Winstone K, Alderson D. Feasibility of quality of life assessment in patients with upper gastrointestinal tract cancer . Br J Cancer 2003 ; 89 : 497 – 501 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Moraca RJ , Low DE. Outcomes and health-related quality of life after esophagectomy for high-grade dysplasia and intramucosal cancer . Arch Surg 2006 ; 141 : 545 – 549 . Google Scholar Crossref Search ADS PubMed WorldCat 17 Blazeby JM , Farndon JR, Donovan J, Alderson D. A prospective longitudinal study examining the quality of life of patients with esophageal carcinoma . Cancer 2000 ; 88 : 1781 – 1787 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Zieren HU , Jacobi CA, Zieren J, Muller JM. Quality of life following resection of oesophageal carcinoma . Br J Surg 1996 ; 83 : 1772 – 1775 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Brooks JA , Kesler KA, Johnson CS, Ciaccia D, Brown JW. Prospective analysis of quality of life after surgical resection for esophageal cancer: preliminary results . J Surg Oncol 2002 ; 81 : 185 – 194 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Avery KN , Metcalfe C, Barham CP, Alderson D, Falk SJ, Blazeby JM. Quality of life during potentially curative treatment for locally advanced oesophageal cancer . Br J Surg 2007 ; 94 : 1369 – 1376 . Google Scholar Crossref Search ADS PubMed WorldCat 21 Viklund P , Wengstrom Y, Rouvelas I, Lindblad M, Lagergren J. Quality of life and persisting symptoms after oesophageal cancer surgery . Eur J Cancer 2006 ; 42 : 1407 – 1414 . Google Scholar Crossref Search ADS PubMed WorldCat 22 de Boer AG , van Lanschot JJ, van Sandick JW, Hulscher JB, Stalmeier PF, de Haes JC et al. Quality of life after transhiatal compared with extended transthoracic resection for adenocarcinoma of the esophagus . J Clin Oncol 2004 ; 22 : 4202 – 4208 . Google Scholar Crossref Search ADS PubMed WorldCat 23 De Boer AG , Genovesi PI, Sprangers MA, Van Sandick JW, Obertop H, Van Lanschot JJ. Quality of life in long- term survivors after curative transhiatal oesophagectomy for oesophageal carcinoma . Br J Surg 2000 ; 87 : 1716 – 1721 . Google Scholar Crossref Search ADS PubMed WorldCat Copyright © 2008 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) Copyright © 2008 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd.

Sultan, J; Robinson, S; Hayes, N; Griffin, S M; Richardson, D L; Preston, S R

doi: 10.1002/bjs.6299pmid: 18655220

Abstract Background Endoscopic ultrasonography (EUS) can detect low-volume ascites (LVA) not apparent on computed tomography. The aim of this study was to assess the importance of LVA for management of patients with oesophagogastric (OG) cancer. Methods Patients with LVA were identified from a prospective OG cancer unit database between January 2002 and January 2006. Results Of 1118 patients staged with OG cancer, 802 had EUS. The incidence of LVA was 8·4 per cent overall but fell to 6·5 per cent when those with metastases on computed tomography were excluded. Only patients with gastric and OG junction carcinoma had LVA. Staging laparoscopy in the 21 patients with LVA revealed that 11 (52 per cent) were inoperable. The remainder had laparotomy and complete (R0) resection was possible in only five (50 per cent). In 106 patients who had staging laparoscopy after EUS without LVA, 37 (34·9 per cent) were inoperable and 56 of the remaining 69 (81 per cent) had R0 resection. Conclusion The presence of LVA on EUS is uncommon in patients with OG cancer but very important, being indicative of incurable disease in 76 per cent. This information will be helpful in counselling patients regarding management options and the low likelihood of potentially curative treatment. Introduction Accurate staging of oesophageal and gastric (OG) cancer is essential to aid appropriate management and provision of information for consent. Contrast-enhanced multislice computed tomography (CT) is the primary investigation to exclude distant metastatic disease. Endoscopic ultrasonography (EUS) offers the best preoperative locoregional staging for oesophageal and gastric cancer1–3. It has also been shown to influence management and increase concordance of decision making in oesophageal and junctional cancers4. Laparoscopy can detect metastatic disease not apparent on CT and avoid unnecessary laparotomy for up to a third of patients with newly diagnosed gastric adenocarcinoma5. Precise stage-dependent management limits the number of unnecessary exploratory surgical interventions6,7. EUS has been reported to be more sensitive than transabdominal ultrasonography, CT, the operative findings of laparoscopy and laparotomy in the detection of intraperitoneal fluid7,8. In the process of staging OG cancers, EUS detects ascites in a number of patients (Fig. 1) in whom multislice CT does not. These patients are described as having low-volume ascites (LVA). There has been little published on the clinical relevance of LVA detected by EUS. The aim of this study was to assess the frequency of LVA when staging OG cancer, and its clinical importance. Fig. 1 Open in new tabDownload slide Endoscopic ultrasonography demonstrating low-volume ascites (black arrows) using a 7·5-MHz radial probe at the tip of the left lobe of the liver and in the lesser sac of a patient with linitis plastica T3 N0 diffuse-type gastric adenocarcinoma Methods All patients referred to a single, dedicated OG unit for cancer staging between January 2002 and January 2006 were included. Data were collected on a prospectively compiled database, with EUS reports including the term LVA (present or absent). During the study period, the unit's standard protocol of staging for all OG cancers, except those referred with overt metastases on CT, was to perform endoscopy, helical multislice CT (oral water plus intravenous contrast enhancement) and EUS. EUS examination was carried out under conscious sedation using a radial scanning echoendoscope (GFUM20, GFUM240 or MH908 oesophagoprobe for stenotic oesophageal tumours; KeyMed (Medical and Industrial Equipment), Southend-on-Sea, UK). Patients who had potentially curable T3 disease with an intra-abdominal component also had staging laparoscopy if the disease was still considered potentially curable after CT, EUS and fitness assessments. The staging protocol ensured that patients had CT and fitness assessments on one day, with endoscopy and EUS within 7 days of CT. The CT results were not reported by the time of EUS. The EUS staging classification was based on the sixth tumour node metastasis (TNM) classification system of the International Union Against cancer9, and contributed to an overall clinical (cTNM) stage for each patient. The results of staging investigations for all patients were discussed at a specialist OG cancer multidisciplinary meeting to determine subsequent management. Results Of 1118 patients with OG cancer, 802 underwent EUS. The remainder had metastatic disease, clear invasion of adjacent structures or severe co-morbid disease evident on information provided by the referring team. Of the 802 patients, 67 (8·4 per cent) demonstrated LVA (Table 1; Fig. 2), of whom 46 had no further staging (four were deemed inoperable because of medical co-morbidity, 24 had metastases on CT, and 18 had advanced locoregional disease on EUS). Fig. 2 Open in new tabDownload slide Outcome of 67 patients with oesophagogastric cancer and low-volume ascites (LVA). CT, computed tomography; EUS, endoscopic ultrasonography Table 1 Demographic data of the 67 patients with low-volume ascites . No. of patients (n = 67) . Age (years)* 70·5 (32–89) Sex ratio (M : F) 2·5 : 1 Site  Gastric 47  Oesophagogastric junction 15  Oesophagus 5 Histology  Adenocarcinoma 63  Squamous cell carcinoma 4 Lauren subtype  Adenocarcinoma Intestinal 37 Mixed 6 Diffuse 20 . No. of patients (n = 67) . Age (years)* 70·5 (32–89) Sex ratio (M : F) 2·5 : 1 Site  Gastric 47  Oesophagogastric junction 15  Oesophagus 5 Histology  Adenocarcinoma 63  Squamous cell carcinoma 4 Lauren subtype  Adenocarcinoma Intestinal 37 Mixed 6 Diffuse 20 * Values are median (range). Open in new tab Table 1 Demographic data of the 67 patients with low-volume ascites . No. of patients (n = 67) . Age (years)* 70·5 (32–89) Sex ratio (M : F) 2·5 : 1 Site  Gastric 47  Oesophagogastric junction 15  Oesophagus 5 Histology  Adenocarcinoma 63  Squamous cell carcinoma 4 Lauren subtype  Adenocarcinoma Intestinal 37 Mixed 6 Diffuse 20 . No. of patients (n = 67) . Age (years)* 70·5 (32–89) Sex ratio (M : F) 2·5 : 1 Site  Gastric 47  Oesophagogastric junction 15  Oesophagus 5 Histology  Adenocarcinoma 63  Squamous cell carcinoma 4 Lauren subtype  Adenocarcinoma Intestinal 37 Mixed 6 Diffuse 20 * Values are median (range). Open in new tab The rapid staging process meant that many patients (143) with metastases on CT still had EUS, of whom 24 had LVA. If CT results had been available before EUS, then only 659 patients would have had EUS, detecting LVA in 43 (6·5 per cent). A total of 21 patients with LVA and potentially curable disease had staging laparoscopy, of which three were cancers of the OG junction (Siewert type II or III) and the rest were true gastric cancers. Eleven (52 per cent) had positive findings at laparoscopy, which precluded them from potentially curative treatment (omental or peritoneal metastases in six patients, hepatic invasion in two, fixed tumours in two and fixed coeliac nodes in one). Laparoscopy detected ascites in only four of these 11 patients. Of the ten remaining patients, laparoscopy detected ascites in only one. Peritoneal cytology was performed in eight of the ten patients and was negative in all. Of ten patients with LVA still deemed operable, only five had potentially curative treatment. Two had surgery alone. Histology demonstrated pathological (p) TNM stage pT3 N0 M0 and pT2b N1 M0 gastric adenocarcinoma. Three received neoadjuvant chemotherapy and surgery. The clinical and pathological stages of disease following preoperative chemotherapy (yp) for these patients were cT3 N2 M0 ypT2b N2 M0 with extranodal spread, cT3 N1 M0 ypT2a N0 M0 and cT3 N1 M0 ypT2a N0 M0. Five patients (50 per cent) were incurable at laparotomy. Of these, two with low-volume omental deposits had palliative resections. The three others had ‘open and close’ procedures (crural and hepatic invasion with fixed retroperitoneal lymphadenopathy in one, fixation to pancreas and colonic mesentery alone in one and with peritoneal metastases in the lesser sac in one). A further 106 patients who did not have LVA also had staging laparoscopy as per protocol. Of these, 33 (31·1 per cent) had positive findings that precluded potentially curative resection and four were considered unfit for resection after laparoscopy. Of the remaining 69 patients, 56 had a complete (R0) resection. Of 21 patients with LVA who proceeded to full staging, therefore, only five (24 per cent) had potentially curative resections. In contrast, of 106 patients without LVA who had full staging, 56 (52·8 per cent) had potentially curative resections (Table 2). Table 2 Outcome data of patients undergoing staging laparoscopy after endoscopic ultrasonography Staging laparoscopy . Procedure . LVA on EUS (n = 21) . No LVA on EUS (n = 106) . Positive None 11 (52) 37 (34·9) Negative Total 10 (48) 69 (65·1) R0 resection 5 (50) 56 (81) R1 resection 0 (0) 4 (6) R2 resection 2 (20) 4 (6) Open and close 3 (30) 5 (7) Staging laparoscopy . Procedure . LVA on EUS (n = 21) . No LVA on EUS (n = 106) . Positive None 11 (52) 37 (34·9) Negative Total 10 (48) 69 (65·1) R0 resection 5 (50) 56 (81) R1 resection 0 (0) 4 (6) R2 resection 2 (20) 4 (6) Open and close 3 (30) 5 (7) Values in parentheses are percentages. LVA, low-volume ascites; EUS, endoscopic ultrasonography; R0, in which all margins are histologically free of tumour; R1, with microscopic residual disease; R2, with gross residual disease. Open in new tab Table 2 Outcome data of patients undergoing staging laparoscopy after endoscopic ultrasonography Staging laparoscopy . Procedure . LVA on EUS (n = 21) . No LVA on EUS (n = 106) . Positive None 11 (52) 37 (34·9) Negative Total 10 (48) 69 (65·1) R0 resection 5 (50) 56 (81) R1 resection 0 (0) 4 (6) R2 resection 2 (20) 4 (6) Open and close 3 (30) 5 (7) Staging laparoscopy . Procedure . LVA on EUS (n = 21) . No LVA on EUS (n = 106) . Positive None 11 (52) 37 (34·9) Negative Total 10 (48) 69 (65·1) R0 resection 5 (50) 56 (81) R1 resection 0 (0) 4 (6) R2 resection 2 (20) 4 (6) Open and close 3 (30) 5 (7) Values in parentheses are percentages. LVA, low-volume ascites; EUS, endoscopic ultrasonography; R0, in which all margins are histologically free of tumour; R1, with microscopic residual disease; R2, with gross residual disease. Open in new tab Discussion This study found that LVA was identified only in tumours that had a gastric component, being present in gastric cancer and Siewert type II and III OG junctional carcinomas; LVA was not detected in any oesophageal cancer without a gastric component of the primary disease. It also found that, of 21 patients with EUS-detected LVA deemed potentially curable before laparoscopy, 16 (76 per cent) turned out to be inoperable or incurable at surgery. EUS performed as part of the staging protocol detected LVA in 8·4 per cent of patients and 6·5 per cent after excluding EUS evaluations deemed unnecessary in the light of CT results. This is lower than in previous studies, where the frequency of ascites detected on EUS has been reported as varying between 9 and 39 per cent7,8,10,11. Two studies have reported the incidence of peritoneal metastases in the presence of EUS-detected ascites (64 per cent in both studies)8,11. In a prospective study of 301 consecutive patients with gastric cancer8, all patients had EUS, CT and transcutaneous ultrasonography before laparoscopy, laparotomy or both. Ascites was detected by CT or ultrasonography in 6 per cent, compared with 32 per cent by EUS. Some 87 per cent of those with CT- or ultrasonography-detected ascites had peritoneal metastases compared with 64 per cent of those identified by EUS. The only study to look specifically at the value of EUS in the detection of ascites not visible on CT-examined patients with gastric cancer in Hong Kong reported ascites in 9 per cent, of which 64 per cent had peritoneal metastases11. This study used a 12-MHz EUS catheter probe passed through a standard diagnostic endoscope. Chen and colleagues7 reported that only 14 per cent of patients with gastric cancer had macroscopic peritoneal carcinomatosis in the presence of ascites. They noted that EUS-detected ascites correlated with depth of tumour invasion, lymph node metastasis, poor cellular differentiation, and with poorer survival rates after surgery. In the present study, nine of 21 patients with LVA and potentially operable disease staged by CT and EUS ultimately had peritoneal disease (six detected at laparoscopy and a further three at laparotomy after a negative laparoscopy) (Fig. 2). EUS identification of LVA is therefore an infrequent but important finding. There should be heightened awareness of inoperability as LVA is indicative of incurable disease in three-quarters of patients. Even in the presence of a normal laparoscopy, half of those with EUS-detected LVA are incurable at the time of laparotomy or surgery. This information may help with preoperative counselling and influence patient consent by preparing them for the low likelihood of potentially curative therapy. Acknowledgements The authors thank Mrs S. Johnson, Data Manager on the Northern Oesophago-Gastric Unit, for her tireless support of the unit database and skill in data retrieval. References 1 Tsendsuren T , Jun SM, Mian XH. Usefulness of endoscopic ultrasonography in preoperative TNM staging of gastric cancer . World J Gastroenterol 2006 ; 12 : 43 – 47 . Google Scholar Crossref Search ADS PubMed WorldCat 2 Lightdale CJ , Kulkarni G. Role of endoscopic ultrasonography in the staging and follow-up of esophageal cancer . J Clin Oncol 2005 ; 23 : 4483 – 4489 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Kelly S , Harris KM, Berry E, Hutton J, Roderick P, Cullingworth J et al. A systematic review of the staging performance of endoscopic ultrasound in gastro-oesophageal carcinoma . Gut 2001 ; 49 : 534 – 539 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Preston SR , Clark GWB, Martin IG, Sue Ling HM, Harris KM. Effect of endoscopic ultrasonography on the management of 100 consecutive cases of oesophageal and junctional carcinoma . Br J Surg 2003 ; 90 : 1220 – 1224 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Burke EC , Karpeh MS, Conlon KC, Brennan MF. Laparoscopy in the management of gastric adenocarcinoma . Ann Surg 1997 ; 225 : 262 – 267 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Davies J , Chalmers AG, Sue-Ling HM, May J, Miller GV, Martin IG et al. Spiral computed tomography and operative staging of gastric carcinoma: a comparison with histopathological staging . Gut 1997 ; 41 : 314 – 319 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Chen CH , Yang CC, Yeh YH. Preoperative staging of gastric cancer by endoscopic ultrasound: the prognostic usefulness of ascites detected by endoscopic ultrasound . J Clin Gastroenterol 2002 ; 35 : 321 – 327 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Lee YT , Ng EK, Hung LC, Chung SC, Ching JY, Chan WY et al. Accuracy of endoscopic ultrasonography in diagnosing ascites and predicting peritoneal metastases in gastric cancer patients . Gut 2005 ; 54 : 1541 – 1545 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Sobin LH , Wittekind C. TNM Classification of Malignant Tumours (6th edn). Wiley–Liss : New York , 2002 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 10 Nguyen PT , Chang KJ. EUS in the detection of ascites and EUS-guided paracentesis . Gastrointest Endosc 2001 ; 54 : 336 – 339 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Chu KM , Kwok KF, Law S, Wong KH. A prospective evaluation of catheter probe EUS for the detection of ascites in patients with gastric carcinoma . Gastrointest Endosc 2004 ; 59 : 471 – 474 . Google Scholar Crossref Search ADS PubMed WorldCat Author notes Presented in part to a meeting of the Association of Surgeons of Great Britain and Ireland, Edinburgh, UK, May 2006, to the International Symposium on Endoscopic Ultrasonography, Amsterdam, The Netherlands, June 2006, and to a joint meeting of the Association of Upper Gastrointestinal Surgeons of Great Britain and Ireland and the Spanish Association of Surgeons, Madrid, Spain, November 2006, and published in abstract form as Br J Surg 2006; 93(Suppl 1): 74, Endoscopy 2006; 38(Suppl 1): P2 and Cir Esp 2006; 80(Suppl 1): 1 Copyright © 2008 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) Copyright © 2008 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd.