Laboratory Medicine

Adams, Kristen, V

doi: 10.1093/labmed/lmz086pmid: 31786609

The relevancy of autopsy is a cyclical topic in pathology discussions. The 20th Century witnessed the rise and fall of hospital autopsies, which once were performed at a rate of more than 40% and now are performed in less than 5% of deaths in many hospitals.1,2 The essential function of an autopsy is discovery. It serves as the ultimate tool of quality control and performance improvement for the hospital, as well as providing insight into the clinicopathologic correlation of pre- and postmortem findings. Despite all the progress we have made in the modern era in accurate diagnosis of our patients, the major discrepancy rate at autopsy is persistently at 27% to 30%,3–5 in which significant findings are made at the time of autopsy and were not diagnosed before death. After its landmark book about therapeutic errors in medicine in 1999, the National Academy of Medicine published another landmark book in 2015 regarding errors in diagnosis.6 This report calls for a reinvigoration of the autopsy as a major tool for improved diagnosis and patient safety. Despite this call, a recent announcement from the Centers for Medicare and Medicaid Services (CMS) made on September 26, 2019, stated that as part of a process of alleviating burdensome tasks, the organization would “reduce the burden for participating providers” regarding hospital autopsies by “[r]emoving the requirement for a hospital’s medical staff to attempt to secure autopsies in all cases of unusual deaths and of medical-legal and educational interest.” 7 It boggles the mind to imagine that a hospital should not absolutely have the duty or responsibility to assist with securing autopsy in any death that occurs at the facility. All family members of decedents families should be given full disclosure of the option of autopsy, and those cases that require forensic pathologists should be referred to the appropriate medical examiner offices. This is a poor choice by CMS, and all those who are involved in public health should be alarmed by it. The hospital autopsy plays a pivotal role in setting the foundation of human anatomy and understanding disease processes. Currently, pathology residents are required by the Accreditation Council for Graduate Medical Education (ACGME) to demonstrate this competency by performing 50 autopsies. Owing to the decrease in autopsy rates and the expanding role of digital and molecular pathology, in the 2014 meeting of the Association of Pathology Chairs (APC), some pathologists called for abolishing autopsy from pathology training. After 2 years of research and deliberation, the Autopsy Working Group formulated by the APC recommended that autopsy should and will remain an integral part of pathology residency training, and residents must be deemed competent in autopsy practice at the time of their graduation.8 Working through autopsy cases gives residents-in-training valuable opportunities to build their problem-solving skills, to create differential diagnosis, and to craft meaningful reports that accurately describe their findings. Autopsy training prepares pathologists to practice independent examinations, to solve the medical puzzles presented to them, and to provide the correct answers to the families of decedents. Autopsy as a surveillance tool for public health is critically important. Environmental exposures, toxicology specimen gathering, and detection of infectious disease are all part of death investigations when necessary. Regarding workers who may be exposed to particulate matter or other hazardous materials on the job, autopsy is extremely important to determine what chronic injuries may be occurring and the cause of death. The National Coal Workers’ Autopsy Study9 is a good demonstration of this type of surveillance. Outbreaks of significant infections are another example. The autopsy may be the first time a culture specimen is taken and significant results are reported. Reporting of infectious diseases to state health departments is critical because these reports may be sentinel events for emerging infectious agents or pandemic outbreaks such as influenza. New, previously undescribed diseases are still also being discovered at autopsy. Dr Bennet Omalu reported such a discovery with his description of chronic traumatic encephalopathy in a professional American-football player.10 In the medicolegal realm, the autopsy also plays an important role in providing objective information in cases in which the family of a decedent files a lawsuit against the institution or physicians who were providing premortem care. However, autopsy does not, even when a major discrepancy exists between pre- and postmortem diagnosis, lead to an increase in cases in which physicians are deemed guilty of negligence.11 Autopsy is an independent objective evaluation of the body of the decedent and can provide further details of significant disease burden not observed in the premortem clinical setting. Autopsy is the keystone of anatomical pathology residency training. Without a strong foundation in human anatomy, gross anatomy, and histology, one cannot understand the pathology of human disease. This concept is elemental in training competent pathologists: no matter how complex our use of modern digital and molecular pathology, we must know the normal physical structure and anatomy. Residents learn this information in their training on the autopsy service. Performing autopsies gives them the opportunity to build their problem-solving skills, to create differential diagnosess, and to learn to craft meaningful reports that accurately describe the findings they observe. These actions are of great importance in a time when we expect so much from our graduating residents: that they are capable of distinguishing normal from abnormal findings and creating accurate reports that give the correct answers to patients and their families. Perhaps the most critical component of autopsy that is often ignored is the preventive care it provides to the entire generation of the family of the decedent, thereby saving money for preventative care instead of potentially more expensive treatments and diagnostic workups; this information ultimately saves the lives of those who may be impacted by the same familial diseases. Autopsy pathologists also spend a significant amount of time meeting with the families of decedents to review the autopsy findings. At times, they refer a family to genetic counselling or other specialties for further follow-up based on the results of postmortem genetic testing, which has become common practice in the modern era of autopsy. There is so much demand for postmortem genetic testing that many laboratories now offer postmortem testing as part of their laboratory menus. Laboratories that offer testing for cardiomyopathic manifestations, arrhythmias, metabolic diseases, etc, in the postmortem setting understand that patient care does not die with the patient—physicians are still providing answers even after death. There are siblings, children, and parents who become our patients and need follow-up with geneticists to determine their health risks. Also, there are current and prospective parents need to be counseled by maternal-fetal medicine specialists regarding the risks of recurrent pregnancy losses. Imagining the isolated autopsy pathologist as a person who works alone, in a poorly lit morgue, and churns out reports that have little relevance is outdated and erroneous. As an autopsy pathologist at an academic medical center, I meet regularly at multidisciplinary meetings with neonatologists, pediatric intensive-care physicians, obstetricians, and other medical and laboratory professionals to discuss cases. These meetings, in which we present the autopsy findings, will change patient care and improve outcomes for patients with similar problems. The collaborative side of autopsy spreads beyond the walls of our institution. We are routinely asked to investigate medical deaths for the coroners of our state. We work collaboratively with our private hospitals to offer services to the families of their patients. And as one of the last places in our state that actually will perform an autopsy, we also offer these services to the public. The critical role of autopsies in forensic investigation has been brought up more recently: colloquially, several news media outlets have reported on the critical shortage of forensic pathologists nationwide. We are roughly 500 forensic pathologists short of the number we need in the United States; however, the shortage is global. This is the health care crisis that no one wants to talk about, partly because once a patient dies, there is no longer a financial role in insurance reimbursement for further investigation of why that person died. At that point, an investigation of cause of death becomes the responsibility of public health officials, state and local authorities, and families themselves to decide whether they would like to have an autopsy performed—often at great personal financial cost to those families. This is unacceptable—the lack of oversight and lack of investment in supporting the forensic sciences nationwide is a great failing of our government. The patchwork systems in place across the country mean many important and valuable autopsies are not performed, and the need for autopsies, forensic and medical, is neglected and ignored. National societies, such as CAP, the American Society for Clinical Pathology (ASCP), and the American Medical Association (AMA), must support and advocate for more funding and residency programs, to increase the number of fellowship opportunities in forensics. Autopsy is an essential component of competent medicolegal death investigations. We need a revival of forensics because without it, we will experience critical failures in public health, public safety, and epidemiology, as well as the disruption that the shortage brings to the criminal justice system. Autopsy is unsurpassed as a method of quality assurance for assessing the sensitivity and specificity of clinical diagnosis. It is pertinent that we examine innovative ways to make autopsy and forensic pathology attractive career choices. Per CMS and many third-party payers, autopsy is “medically unnecessary”; therefore, hospital autopsy has no direct method for reimbursement. It is time for all of us who practice medicine to rethink this idea and advocate for changing it. The value that autopsy brings to the hospitals and families who need those services goes far beyond identifying the cause of death. Pathologists who perform autopsies are a crucial part of the teams of physicians who identify disease processes and genetic diseases, and who drive future patient management decisions by sharing autopsy findings at multidisciplinary clinicopathological team meetings. In conclusion, sharing of medical knowledge and teaching of dissection skills in autopsy is still and must remain the foundation of resident education, much as it was when pathology began as a discipline. Personal and Financial Conflicts of Interest None reported. References 1. Hasson J , Schneiderman H . Autopsy training programs. To right a wrong . Arch Pathol Lab Med. 1995 ; 119 ( 3 ): 289 – 291 . Google Scholar PubMed WorldCat 2. Landefeld CS , Chren MM , Myers A , Geller R , Robbins S , Goldman L . Diagnostic yield of the autopsy in a university hospital and a community hospital . N Engl J Med. 1988 ; 318 ( 19 ): 1249 – 1254 . Google Scholar Crossref Search ADS PubMed WorldCat 3. Early CA , Gilliland MGF , Kelly KL , Oliver WR , Kragel PJ . Autopsy standardized mortality review: a pilot study offering a methodology for improved patient outcomes . Acad Pathol. 2019 ; 6 : 2374289519826281 . Google Scholar Crossref Search ADS PubMed WorldCat 4. Scordi-Bello IA , Kalb TH , Lento PA . Clinical setting and extent of premortem evaluation do not predict autopsy discrepancy rates . Mod Pathol. 2010 ; 23 ( 9 ): 1225 – 1230 . Google Scholar Crossref Search ADS PubMed WorldCat 5. Gonzalez Franco MV , Ponce Camacho MA , Barboza Quintana O , Ancer Rodriguez J , Cecenas Falcon LA . Discrepancies between clinical and autopsy diagnosis: a study of 331 autopsies performed over a 7 years period . Medicina Universitaria . 2012 ; 14 : 16 – 22 . WorldCat 6. Zeng H , Li R , Hu C , et al. Association of twice-daily radiotherapy with subsequent brain metastases in adults with small cell lung cancer . JAMA Netw Open . 2019 ; 2 ( 5 ): e194308 Google Scholar Crossref Search ADS PubMed WorldCat 7. Fact sheet: Omnibus Burden Reduction (Conditions of Participation) Final Rule CMS-3346-F [Web post]. (2019, September 26) Retrieved October 1, 2019, from https://www.cms.gov/newsroom/fact-sheets/omnibus-burden-reduction-conditions-participation-final-rule-cms-3346-f.Accessed October 22, 2019. 8. Davis GG , Winters GL , Fyfe BS , et al. Report and recommendations of the Association of Pathology Chairs’ Autopsy Working Group . Acad Pathol. 2018 ; 5 : 2374289518793988 . Google Scholar Crossref Search ADS PubMed WorldCat 9. Vallyathan V , Landsittel DP , Petsonk EL , et al. The influence of dust standards on the prevalence and severity of coal worker’s pneumoconiosis at autopsy in the United States of America . Arch Pathol Lab Med. 2011 ; 135 ( 12 ): 1550 – 1556 . Google Scholar Crossref Search ADS PubMed WorldCat 10. Omalu BI , DeKosky ST , Minster RL . Chronic traumatic encephalopathy in a National Football League player . Neurosurgery . 2005 ; 57 ( 1 ): 128 – 134 . Google Scholar Crossref Search ADS PubMed WorldCat 11. Bove KE , Iery C . The role of the autopsy in medical malpractice cases, II . Arch Pathol Lab Med. 2002 ; 126 ( 9 ): 1032 – 1035 . Google Scholar PubMed WorldCat © American Society for Clinical Pathology 2019. All rights reserved. For permissions, please e-mail: [email protected] 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)

Seyyedi,, Negisa;Moghaddasi,, Hamid;Asadi,, Farkhondeh;Hamidpour,, Mohsen;Shoaie,, Kamal

doi: 10.1093/labmed/lmz084pmid: 31796957

Abstract Laboratory services form an integral part of medical care in the decision-making of physicians, including those working at ambulatory care centers. Information exchange is essential between ambulatory care centers and laboratories. Inevitable errors have always existed in the exchange of such information on paper, which can be to some extent avoided by developing appropriate computer-based interfaces. Therefore, this review aimed to examine studies conducted to determine the effect of electronic communication between ambulatory care centers and laboratories. This systematic review was conducted on the basis of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. Studies were searched in the PubMed, Embase, Cochrane, and Web of Science, and those written in English and published between 2000 and February 2019 with full texts available were selected. From a total of 3898 papers retrieved from the studied databases, 24 papers were eligible for entering this study after removing similar and nonrelated studies. Electronic exchanges between ambulatory care centers and laboratories can have numerous benefits in terms of financial, organizational, and quality. This evidence for the value of electronic communications is an important factor contributing to its local investment and adoption. lab communication management, lab test request, lab test result, laboratory information exchange, health information exchange, interoperability, ambulatory care centers, laboratories Laboratory services are an important part of medical care and can help the patient, physician, and other care providers to make appropriate decisions. Although diagnostic tests affect more than 70% of healthcare decisions, they are often the least costly option for healthcare and provide objective information about the health of an individual.1–3 Patient referrals to ambulatory care centers are far more frequent than to other healthcare centers.4 Laboratory services in ambulatory care centers can offer continued care through the longitudinal continuity of laboratory tests on patients with chronic illnesses as well as by informing caregivers about increased risks and potential gaps in care. Consequently, laboratory services play a determining role in guiding medical activities in ambulatory care centers.5 Communication is a central issue in terms of the accessibility of laboratory services. Communication here refers to sending a physician’s request for a test and returning the results of the test from the laboratory to the requesting physician, which can be established in a variety of traditional or modern ways. In traditional communication, where the patients themselves sometimes carry the information on paper, patients waste considerable time and energy moving between ambulatory care centers and independent laboratories outside these centers.6,7 Many delays and lags are imposed on the transfer of laboratory results, and physicians cannot always access and review the results on time. This occasionally makes it challenging to track abnormal results in time, jeopardizing the quality of care and patient safety and satisfaction. In some cases, exchange of the traditional kinds of information between physicians and laboratories (both requesting tests and retrieving test results) is performed by e-mail, fax, dedicated telephone lines, and printer setups, which, in addition to the above mentioned problems, are not documented or reliable.1,8–11 Modern or electronic information exchange between ambulatory care centers and laboratories can assist healthcare providers in making informed decisions on ways of improving patient safety and resolving paper-based problems. However, the approaches of developed and developing countries are different in this regard. In developing countries, most primary care centers do not have electronic medical records, and ambulatory care centers and offices are often not digitally connected to their reference laboratory; 12–15 developed countries, which are pioneers in this field, have not completely addressed all the challenges of interoperability.16–19 Due to the limited attention paid to electronic communications between laboratories and ambulatory care centers in proportion to the hospitals,20 as well as the fallibility of traditional communication, the present study attempted to examine the effect of information technology on the information exchange between laboratories and ambulatory care centers by systematically reviewing the literature. Materials and Methods This systematic study was conducted based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement,21 in which a focused text search was performed to examine the effect of information technology on processes related to information transfer between diagnostic laboratories and ambulatory care centers. The study was conducted in collaboration with all authors of this study and the advice of database experts and Internet information science experts to design a search strategy for several electronic bibliographic databases, including PubMed, Web of Science, Embase, and Cochrane, as the primary sources of information. Complementary information was collected through referral to the references of the papers. The search strategy (see the appendix) in the bibliographic databases included searching the title of the papers and selecting keywords with high sensitivity and specificity. For each searched concept, we selected words which, in addition to being closely related in meaning, were used in papers focusing on interoperability and information exchange. After the approval of the keywords, together with appropriate wildcards and truncations, they were combined with OR and AND operators. All types of research studies in English with full texts available that were printed or electronically published from 2000 to February 2019 in journals or conferences indexed in searchable databases were considered. In this systematic review, studies related to the subject were selected with an emphasis on electronic information exchange between systems of diagnostic laboratories and ambulatory care centers, including hospital outpatient wards, primary health centers, emergency departments, and doctors’ offices. In studies not mentioning the research environment, information exchange between laboratory information systems (LIS) and the electronic health record (EHR) system was deemed acceptable. Also, exchanges which were either unidirectional (laboratory to ambulatory centers or ambulatory centers to thelaboratory) or bidirectional (both with laboratory andambulatory centers and with ambulatory centers and the laboratory) containing any health and laboratory information (even alerts) were included in this study. Papers considering the relationship between laboratories and inpatient centers, such as hospitals, with a major focus on the inpatient ward and not discussing ambulatory care services and departments were excluded. A total of 3898 papers were obtained upon searching the mentioned databases by applying the above mentioned restrictions, of which 2091 papers remained after the elimination of duplications. The process of selecting papers was as follows: for the initial evaluation, all authors of this study first examined the titles of papers. In this stage, papers that were not relevant to the searched topic were excluded, and 920 papers remained after the examination of the titles. In the next stage, the abstracts of the papers were reviewed, and 568 papers were excluded from the study because they did not address electronic exchanges specifically between laboratories or ambulatory care centers; 351 papers remained. Then, a complete review of the full text of the papers was performed, and papers examining the transfer of information only between laboratories and ambulatory care centers and precisely addressing the benefits of this transfer were selected. As a result, papers that did not specify that the centers were ambulatory ones or indicated that the studied centers were only inpatient ones were excluded. Eventually, 19 studies remained. In this phase, the references of the papers whose full text had been reviewed (351 papers) were also checked, and 5 papers were added to the study. Upon final evaluation and consensus, a total of 24 eligible papers were included in this study (Figure 1). Figure 1 Open in new tabDownload slide Flow diagram of study selection. *Reference of included studies and relevant reviews. WOS, Web of Science. Figure 1 Open in new tabDownload slide Flow diagram of study selection. *Reference of included studies and relevant reviews. WOS, Web of Science. Results After studying and exploring the 24 papers, and to investigate and identify the effects of laboratories’ electronic communication with ambulatory care centers, relevant information was extracted as detailed in Table 1. There were 2 groups of studies in terms of the type of information exchanged between laboratories and ambulatory care centers: (i) studies based on the exchange of laboratory data (test results and test requests), and (ii) studies examining comments and alarms.22–24 From among these studies, some focused on bidirectional exchanges, including both test results and test requests,11,25–33 whereas some focused on unidirectional exchanges from laboratories to ambulatory care centers9,10,22–24,34–40 or from ambulatory care centers to laboratories.16 This review was limited to the electronic exchanges between laboratories and ambulatory care centers. As a result, the emergency departments,33,35 small physician offices,10,11,36 ambulatory settings,9,27,29 primary health centers,28,38 practices,37 healthcare centers,30,34 acquired immunodeficiency syndrome (AIDS) clinics,27,32 and hospital outpatient departments16,22,39,41 were ambulatory care centers examined in the reviewed papers. Studies on communication with EHR systems24,31 were also examined. Table 1. Included Studies Ref. Research Type - Publication Yr Technology Used For Exchanging Information and the Exchange Parties Effects of Electronic Exchanges 25 Research support - 2005 Electronic exchange between care providers (hospitals and group practices) and ancillary entities, such as laboratories - Increased net and annual profits 10 Cross-sectional study - 2007 Independent Internet-based portal for the exchange of laboratory data between laboratories and offices - High performance in preventive care - Management of chronic illness - Patient satisfaction 22 Observational study - 2008 Implementing computerized alerts for transferring alerts from laboratories to the outpatient section of the hospital - Increased reporting completeness - Improved timeliness of reporting and accountability of physicians - Efficient, complete, and timely management of laboratory information - Increasing workload while improving and creating positive changes in workflow - Improved quality of available data for decision making 23 Comparative study - 2009 Automatic alerts embedded in integrated EHR systems in outpatient centers for sending test results from laboratories - No improvement in timely follow-up of laboratory results using automated alarms 26 Viewpoint paper - 2009 Use of HIE in the field of infant screening information exchange among the hospital, outpatient centers, and childbirth facilities - Improved timing, effectiveness, quality, and efficiency of patient care services in the screening program 24 Validation study - 2009 Electronic exchange of positive lab results from the hospital and outpatient clinics to the laboratory - Timely follow-up of lab test results 9 Case report - 2009 Sending results after generation in internal laboratory information systems using health level 7–compatible interfaces through a commercial interface engine in the integrated ambulatory EHR with the ability to manage lab results - Improved speed of delivering test results - Access to test results at any location where the provider has access to the EHR - Accurate and timely delivery of the test results to a provider - Direct access of patients to their results through a confidential personal health record - Sharing results among multiple providers and routing the results based on conditional logic - Automatic capturing of follow-up operations - Monitoring the results delivery problems with electronic reports 34 RCT - 2010 Sending results to health centers using a lab reporting system called e-Chasqui - Reduced error rate - Reduced errors in reporting the results of cultures and drug susceptibility tests - Preventing loss of results - Reduced number of lost laboratory results at the point of care - Promoting continuous quality improvement 35 Cohort study - 2011 Providing access for the emergency department to laboratory data and other HIE data through a secure Web browser - Cost saving - Reduced requests for laboratory tests 41 Proceedings paper/cohort study - 2011 Sending a test request from the computerized patient record containing the built-in computerized physical order entry called Presco in the hospital and outpatient centers with direct links to the LIS - Increased speed of nursing work 27 Single-arm intervention study - 2012 Exchange of laboratory information through the implementation of a laboratory information exchange interface within the ambulatory EHR in the AIDS clinic - Significant improvements in timely regulation of antiretroviral drugs for patients with AIDS - Timely response to laboratory results - Improved AIDS care both due to the timeliness of responding to important test results and the quality of communication in relation to the tests - Improved care by ensuring that the tests are carried out correctly and follow up of the results of all the requested tests - Improved patient-based communication about laboratory tests 11 Case study -2012 Electronic exchanges between physician offices and laboratories - More advantages - EHR acceptance 36 Retrospective cohort study- 2012 Internet-based HIE portal featuring lab results for small group practices - Moderate improvement in outpatient care quality 37 Research support/ analytical study - 2013 Using HIE to send lab results from laboratories to large offices and hospitals - Significant decrease in the number of laboratory tests - No significant reduction in costs 28 Single-arm intervention -2013 Exchange of information between primary healthcare centers and laboratories by implementing an integrated module for requesting laboratory tests - Improved patient safety by improved patient identification - Improved preanalytical quality - Improved demand management - Improved response time due to the acceleration of the entire analysis process - Improved process tracking, such as sample tracking - Improved administrative work - Improved access to results 29 Simulation study - 2013 Interactions between labs and EHR systems in outpatient centers in a simulated environment - Reduced time consumed by providers - No significant effect on nurses’ use or the total time of visits and waiting time for patients - More time efficiency for administrators than physicians and nurses 30 Qualitative study - 2014 Electronic exchange of laboratory information between LIS and HIS in hospitals and healthcare centers in an integrated environment - Accelerating and managing the workflow of patients - Tracking system to receive samples - Decreased paper work - Minimized errors - Reduced turnaround time - Reduced test redundancy - Improved communication with physicians and decision makers - Improved quality of data collection - Support for quality improvement initiatives 38 Cluster randomized controlled trial - 2014 Communication between laboratories and primary health centers using the Web-based LIS e-Chasqui - Reduced results transfer time between laboratories and health institutions - Reduced culture conversion time - Preventing patients from being ignored (patient follow-up) 31 Qualitative study - 2015 Electronic information exchange between EHR and LIS Potential risks for patient care by undesirable and misleading displaying of results in EHR systems - Created potential for decentralized decision making due to loss, misrepresentation, or misinterpretation of information by changing the flow of information between laboratories and physicians by EHR - Lost ability to control the aspects of the request process by neglecting the precise development of the EHR/LIS interface and the uncertainty about the completeness, accuracy, and sufficiency of laboratory test requests in order to facilitate the efficient processing of laboratory operations 16 Comparative study - 2015 Interchange between the outpatient section of the hospital and laboratories - Reduced transcription errors and illegible handwritten documents - Minimized actions that require human intervention, such as manual data input to LIS - Removed error sources, including losing test results, entering wrong test data, not signing, and wrong labeling of samples - Improved data integrity and accuracy - More accuracy and completeness of electronic patient records than with handwritten records - Facilitated access and sharing of information between physicians, laboratories, and patients. 39 Retrospective observational cross-sectional study - 2015 Education and communication by sending comments from the laboratory to the outpatient and inpatient sections of the hospital - More appropriate request for vitamin D test - Reduced improper test requests and hence laboratory costs - Promoted correct use and self-regulation of vitamin D test requests 40 Qualitative study - 2015 Electronic information exchange is only for laboratory result sending from laboratory to primary care physician, and test ordering is transmitted by carrier - Difficulties in sample handling - Misidentification of samples - Inefficiency in collating and resending misdirected results 32 Longitudinal interventional quasiexperimental - 2017 Exchange of information between the AIDS clinic and the laboratory using laboratory health information exchange - Improved overall treatment of antiretroviral therapy and viral suppression - Reduced black and white differences in the use of antiretroviral therapy and viral suppression 33 Cross-sectional study - 2018 Exchange of laboratory information between the laboratory and the emergency department using various laboratory information exchange systems - Efficiency, quality, and safety of emergency care - Better monitoring of patients’ health - Approval of diagnostic decision making assumptions - Applying evidence-based operations in the medical laboratory Ref. Research Type - Publication Yr Technology Used For Exchanging Information and the Exchange Parties Effects of Electronic Exchanges 25 Research support - 2005 Electronic exchange between care providers (hospitals and group practices) and ancillary entities, such as laboratories - Increased net and annual profits 10 Cross-sectional study - 2007 Independent Internet-based portal for the exchange of laboratory data between laboratories and offices - High performance in preventive care - Management of chronic illness - Patient satisfaction 22 Observational study - 2008 Implementing computerized alerts for transferring alerts from laboratories to the outpatient section of the hospital - Increased reporting completeness - Improved timeliness of reporting and accountability of physicians - Efficient, complete, and timely management of laboratory information - Increasing workload while improving and creating positive changes in workflow - Improved quality of available data for decision making 23 Comparative study - 2009 Automatic alerts embedded in integrated EHR systems in outpatient centers for sending test results from laboratories - No improvement in timely follow-up of laboratory results using automated alarms 26 Viewpoint paper - 2009 Use of HIE in the field of infant screening information exchange among the hospital, outpatient centers, and childbirth facilities - Improved timing, effectiveness, quality, and efficiency of patient care services in the screening program 24 Validation study - 2009 Electronic exchange of positive lab results from the hospital and outpatient clinics to the laboratory - Timely follow-up of lab test results 9 Case report - 2009 Sending results after generation in internal laboratory information systems using health level 7–compatible interfaces through a commercial interface engine in the integrated ambulatory EHR with the ability to manage lab results - Improved speed of delivering test results - Access to test results at any location where the provider has access to the EHR - Accurate and timely delivery of the test results to a provider - Direct access of patients to their results through a confidential personal health record - Sharing results among multiple providers and routing the results based on conditional logic - Automatic capturing of follow-up operations - Monitoring the results delivery problems with electronic reports 34 RCT - 2010 Sending results to health centers using a lab reporting system called e-Chasqui - Reduced error rate - Reduced errors in reporting the results of cultures and drug susceptibility tests - Preventing loss of results - Reduced number of lost laboratory results at the point of care - Promoting continuous quality improvement 35 Cohort study - 2011 Providing access for the emergency department to laboratory data and other HIE data through a secure Web browser - Cost saving - Reduced requests for laboratory tests 41 Proceedings paper/cohort study - 2011 Sending a test request from the computerized patient record containing the built-in computerized physical order entry called Presco in the hospital and outpatient centers with direct links to the LIS - Increased speed of nursing work 27 Single-arm intervention study - 2012 Exchange of laboratory information through the implementation of a laboratory information exchange interface within the ambulatory EHR in the AIDS clinic - Significant improvements in timely regulation of antiretroviral drugs for patients with AIDS - Timely response to laboratory results - Improved AIDS care both due to the timeliness of responding to important test results and the quality of communication in relation to the tests - Improved care by ensuring that the tests are carried out correctly and follow up of the results of all the requested tests - Improved patient-based communication about laboratory tests 11 Case study -2012 Electronic exchanges between physician offices and laboratories - More advantages - EHR acceptance 36 Retrospective cohort study- 2012 Internet-based HIE portal featuring lab results for small group practices - Moderate improvement in outpatient care quality 37 Research support/ analytical study - 2013 Using HIE to send lab results from laboratories to large offices and hospitals - Significant decrease in the number of laboratory tests - No significant reduction in costs 28 Single-arm intervention -2013 Exchange of information between primary healthcare centers and laboratories by implementing an integrated module for requesting laboratory tests - Improved patient safety by improved patient identification - Improved preanalytical quality - Improved demand management - Improved response time due to the acceleration of the entire analysis process - Improved process tracking, such as sample tracking - Improved administrative work - Improved access to results 29 Simulation study - 2013 Interactions between labs and EHR systems in outpatient centers in a simulated environment - Reduced time consumed by providers - No significant effect on nurses’ use or the total time of visits and waiting time for patients - More time efficiency for administrators than physicians and nurses 30 Qualitative study - 2014 Electronic exchange of laboratory information between LIS and HIS in hospitals and healthcare centers in an integrated environment - Accelerating and managing the workflow of patients - Tracking system to receive samples - Decreased paper work - Minimized errors - Reduced turnaround time - Reduced test redundancy - Improved communication with physicians and decision makers - Improved quality of data collection - Support for quality improvement initiatives 38 Cluster randomized controlled trial - 2014 Communication between laboratories and primary health centers using the Web-based LIS e-Chasqui - Reduced results transfer time between laboratories and health institutions - Reduced culture conversion time - Preventing patients from being ignored (patient follow-up) 31 Qualitative study - 2015 Electronic information exchange between EHR and LIS Potential risks for patient care by undesirable and misleading displaying of results in EHR systems - Created potential for decentralized decision making due to loss, misrepresentation, or misinterpretation of information by changing the flow of information between laboratories and physicians by EHR - Lost ability to control the aspects of the request process by neglecting the precise development of the EHR/LIS interface and the uncertainty about the completeness, accuracy, and sufficiency of laboratory test requests in order to facilitate the efficient processing of laboratory operations 16 Comparative study - 2015 Interchange between the outpatient section of the hospital and laboratories - Reduced transcription errors and illegible handwritten documents - Minimized actions that require human intervention, such as manual data input to LIS - Removed error sources, including losing test results, entering wrong test data, not signing, and wrong labeling of samples - Improved data integrity and accuracy - More accuracy and completeness of electronic patient records than with handwritten records - Facilitated access and sharing of information between physicians, laboratories, and patients. 39 Retrospective observational cross-sectional study - 2015 Education and communication by sending comments from the laboratory to the outpatient and inpatient sections of the hospital - More appropriate request for vitamin D test - Reduced improper test requests and hence laboratory costs - Promoted correct use and self-regulation of vitamin D test requests 40 Qualitative study - 2015 Electronic information exchange is only for laboratory result sending from laboratory to primary care physician, and test ordering is transmitted by carrier - Difficulties in sample handling - Misidentification of samples - Inefficiency in collating and resending misdirected results 32 Longitudinal interventional quasiexperimental - 2017 Exchange of information between the AIDS clinic and the laboratory using laboratory health information exchange - Improved overall treatment of antiretroviral therapy and viral suppression - Reduced black and white differences in the use of antiretroviral therapy and viral suppression 33 Cross-sectional study - 2018 Exchange of laboratory information between the laboratory and the emergency department using various laboratory information exchange systems - Efficiency, quality, and safety of emergency care - Better monitoring of patients’ health - Approval of diagnostic decision making assumptions - Applying evidence-based operations in the medical laboratory HIE, health information exchange;LIS, laboratory information systems; EHR, electronic health record; HIS, hospital information systems. Open in new tab Table 1. Included Studies Ref. Research Type - Publication Yr Technology Used For Exchanging Information and the Exchange Parties Effects of Electronic Exchanges 25 Research support - 2005 Electronic exchange between care providers (hospitals and group practices) and ancillary entities, such as laboratories - Increased net and annual profits 10 Cross-sectional study - 2007 Independent Internet-based portal for the exchange of laboratory data between laboratories and offices - High performance in preventive care - Management of chronic illness - Patient satisfaction 22 Observational study - 2008 Implementing computerized alerts for transferring alerts from laboratories to the outpatient section of the hospital - Increased reporting completeness - Improved timeliness of reporting and accountability of physicians - Efficient, complete, and timely management of laboratory information - Increasing workload while improving and creating positive changes in workflow - Improved quality of available data for decision making 23 Comparative study - 2009 Automatic alerts embedded in integrated EHR systems in outpatient centers for sending test results from laboratories - No improvement in timely follow-up of laboratory results using automated alarms 26 Viewpoint paper - 2009 Use of HIE in the field of infant screening information exchange among the hospital, outpatient centers, and childbirth facilities - Improved timing, effectiveness, quality, and efficiency of patient care services in the screening program 24 Validation study - 2009 Electronic exchange of positive lab results from the hospital and outpatient clinics to the laboratory - Timely follow-up of lab test results 9 Case report - 2009 Sending results after generation in internal laboratory information systems using health level 7–compatible interfaces through a commercial interface engine in the integrated ambulatory EHR with the ability to manage lab results - Improved speed of delivering test results - Access to test results at any location where the provider has access to the EHR - Accurate and timely delivery of the test results to a provider - Direct access of patients to their results through a confidential personal health record - Sharing results among multiple providers and routing the results based on conditional logic - Automatic capturing of follow-up operations - Monitoring the results delivery problems with electronic reports 34 RCT - 2010 Sending results to health centers using a lab reporting system called e-Chasqui - Reduced error rate - Reduced errors in reporting the results of cultures and drug susceptibility tests - Preventing loss of results - Reduced number of lost laboratory results at the point of care - Promoting continuous quality improvement 35 Cohort study - 2011 Providing access for the emergency department to laboratory data and other HIE data through a secure Web browser - Cost saving - Reduced requests for laboratory tests 41 Proceedings paper/cohort study - 2011 Sending a test request from the computerized patient record containing the built-in computerized physical order entry called Presco in the hospital and outpatient centers with direct links to the LIS - Increased speed of nursing work 27 Single-arm intervention study - 2012 Exchange of laboratory information through the implementation of a laboratory information exchange interface within the ambulatory EHR in the AIDS clinic - Significant improvements in timely regulation of antiretroviral drugs for patients with AIDS - Timely response to laboratory results - Improved AIDS care both due to the timeliness of responding to important test results and the quality of communication in relation to the tests - Improved care by ensuring that the tests are carried out correctly and follow up of the results of all the requested tests - Improved patient-based communication about laboratory tests 11 Case study -2012 Electronic exchanges between physician offices and laboratories - More advantages - EHR acceptance 36 Retrospective cohort study- 2012 Internet-based HIE portal featuring lab results for small group practices - Moderate improvement in outpatient care quality 37 Research support/ analytical study - 2013 Using HIE to send lab results from laboratories to large offices and hospitals - Significant decrease in the number of laboratory tests - No significant reduction in costs 28 Single-arm intervention -2013 Exchange of information between primary healthcare centers and laboratories by implementing an integrated module for requesting laboratory tests - Improved patient safety by improved patient identification - Improved preanalytical quality - Improved demand management - Improved response time due to the acceleration of the entire analysis process - Improved process tracking, such as sample tracking - Improved administrative work - Improved access to results 29 Simulation study - 2013 Interactions between labs and EHR systems in outpatient centers in a simulated environment - Reduced time consumed by providers - No significant effect on nurses’ use or the total time of visits and waiting time for patients - More time efficiency for administrators than physicians and nurses 30 Qualitative study - 2014 Electronic exchange of laboratory information between LIS and HIS in hospitals and healthcare centers in an integrated environment - Accelerating and managing the workflow of patients - Tracking system to receive samples - Decreased paper work - Minimized errors - Reduced turnaround time - Reduced test redundancy - Improved communication with physicians and decision makers - Improved quality of data collection - Support for quality improvement initiatives 38 Cluster randomized controlled trial - 2014 Communication between laboratories and primary health centers using the Web-based LIS e-Chasqui - Reduced results transfer time between laboratories and health institutions - Reduced culture conversion time - Preventing patients from being ignored (patient follow-up) 31 Qualitative study - 2015 Electronic information exchange between EHR and LIS Potential risks for patient care by undesirable and misleading displaying of results in EHR systems - Created potential for decentralized decision making due to loss, misrepresentation, or misinterpretation of information by changing the flow of information between laboratories and physicians by EHR - Lost ability to control the aspects of the request process by neglecting the precise development of the EHR/LIS interface and the uncertainty about the completeness, accuracy, and sufficiency of laboratory test requests in order to facilitate the efficient processing of laboratory operations 16 Comparative study - 2015 Interchange between the outpatient section of the hospital and laboratories - Reduced transcription errors and illegible handwritten documents - Minimized actions that require human intervention, such as manual data input to LIS - Removed error sources, including losing test results, entering wrong test data, not signing, and wrong labeling of samples - Improved data integrity and accuracy - More accuracy and completeness of electronic patient records than with handwritten records - Facilitated access and sharing of information between physicians, laboratories, and patients. 39 Retrospective observational cross-sectional study - 2015 Education and communication by sending comments from the laboratory to the outpatient and inpatient sections of the hospital - More appropriate request for vitamin D test - Reduced improper test requests and hence laboratory costs - Promoted correct use and self-regulation of vitamin D test requests 40 Qualitative study - 2015 Electronic information exchange is only for laboratory result sending from laboratory to primary care physician, and test ordering is transmitted by carrier - Difficulties in sample handling - Misidentification of samples - Inefficiency in collating and resending misdirected results 32 Longitudinal interventional quasiexperimental - 2017 Exchange of information between the AIDS clinic and the laboratory using laboratory health information exchange - Improved overall treatment of antiretroviral therapy and viral suppression - Reduced black and white differences in the use of antiretroviral therapy and viral suppression 33 Cross-sectional study - 2018 Exchange of laboratory information between the laboratory and the emergency department using various laboratory information exchange systems - Efficiency, quality, and safety of emergency care - Better monitoring of patients’ health - Approval of diagnostic decision making assumptions - Applying evidence-based operations in the medical laboratory Ref. Research Type - Publication Yr Technology Used For Exchanging Information and the Exchange Parties Effects of Electronic Exchanges 25 Research support - 2005 Electronic exchange between care providers (hospitals and group practices) and ancillary entities, such as laboratories - Increased net and annual profits 10 Cross-sectional study - 2007 Independent Internet-based portal for the exchange of laboratory data between laboratories and offices - High performance in preventive care - Management of chronic illness - Patient satisfaction 22 Observational study - 2008 Implementing computerized alerts for transferring alerts from laboratories to the outpatient section of the hospital - Increased reporting completeness - Improved timeliness of reporting and accountability of physicians - Efficient, complete, and timely management of laboratory information - Increasing workload while improving and creating positive changes in workflow - Improved quality of available data for decision making 23 Comparative study - 2009 Automatic alerts embedded in integrated EHR systems in outpatient centers for sending test results from laboratories - No improvement in timely follow-up of laboratory results using automated alarms 26 Viewpoint paper - 2009 Use of HIE in the field of infant screening information exchange among the hospital, outpatient centers, and childbirth facilities - Improved timing, effectiveness, quality, and efficiency of patient care services in the screening program 24 Validation study - 2009 Electronic exchange of positive lab results from the hospital and outpatient clinics to the laboratory - Timely follow-up of lab test results 9 Case report - 2009 Sending results after generation in internal laboratory information systems using health level 7–compatible interfaces through a commercial interface engine in the integrated ambulatory EHR with the ability to manage lab results - Improved speed of delivering test results - Access to test results at any location where the provider has access to the EHR - Accurate and timely delivery of the test results to a provider - Direct access of patients to their results through a confidential personal health record - Sharing results among multiple providers and routing the results based on conditional logic - Automatic capturing of follow-up operations - Monitoring the results delivery problems with electronic reports 34 RCT - 2010 Sending results to health centers using a lab reporting system called e-Chasqui - Reduced error rate - Reduced errors in reporting the results of cultures and drug susceptibility tests - Preventing loss of results - Reduced number of lost laboratory results at the point of care - Promoting continuous quality improvement 35 Cohort study - 2011 Providing access for the emergency department to laboratory data and other HIE data through a secure Web browser - Cost saving - Reduced requests for laboratory tests 41 Proceedings paper/cohort study - 2011 Sending a test request from the computerized patient record containing the built-in computerized physical order entry called Presco in the hospital and outpatient centers with direct links to the LIS - Increased speed of nursing work 27 Single-arm intervention study - 2012 Exchange of laboratory information through the implementation of a laboratory information exchange interface within the ambulatory EHR in the AIDS clinic - Significant improvements in timely regulation of antiretroviral drugs for patients with AIDS - Timely response to laboratory results - Improved AIDS care both due to the timeliness of responding to important test results and the quality of communication in relation to the tests - Improved care by ensuring that the tests are carried out correctly and follow up of the results of all the requested tests - Improved patient-based communication about laboratory tests 11 Case study -2012 Electronic exchanges between physician offices and laboratories - More advantages - EHR acceptance 36 Retrospective cohort study- 2012 Internet-based HIE portal featuring lab results for small group practices - Moderate improvement in outpatient care quality 37 Research support/ analytical study - 2013 Using HIE to send lab results from laboratories to large offices and hospitals - Significant decrease in the number of laboratory tests - No significant reduction in costs 28 Single-arm intervention -2013 Exchange of information between primary healthcare centers and laboratories by implementing an integrated module for requesting laboratory tests - Improved patient safety by improved patient identification - Improved preanalytical quality - Improved demand management - Improved response time due to the acceleration of the entire analysis process - Improved process tracking, such as sample tracking - Improved administrative work - Improved access to results 29 Simulation study - 2013 Interactions between labs and EHR systems in outpatient centers in a simulated environment - Reduced time consumed by providers - No significant effect on nurses’ use or the total time of visits and waiting time for patients - More time efficiency for administrators than physicians and nurses 30 Qualitative study - 2014 Electronic exchange of laboratory information between LIS and HIS in hospitals and healthcare centers in an integrated environment - Accelerating and managing the workflow of patients - Tracking system to receive samples - Decreased paper work - Minimized errors - Reduced turnaround time - Reduced test redundancy - Improved communication with physicians and decision makers - Improved quality of data collection - Support for quality improvement initiatives 38 Cluster randomized controlled trial - 2014 Communication between laboratories and primary health centers using the Web-based LIS e-Chasqui - Reduced results transfer time between laboratories and health institutions - Reduced culture conversion time - Preventing patients from being ignored (patient follow-up) 31 Qualitative study - 2015 Electronic information exchange between EHR and LIS Potential risks for patient care by undesirable and misleading displaying of results in EHR systems - Created potential for decentralized decision making due to loss, misrepresentation, or misinterpretation of information by changing the flow of information between laboratories and physicians by EHR - Lost ability to control the aspects of the request process by neglecting the precise development of the EHR/LIS interface and the uncertainty about the completeness, accuracy, and sufficiency of laboratory test requests in order to facilitate the efficient processing of laboratory operations 16 Comparative study - 2015 Interchange between the outpatient section of the hospital and laboratories - Reduced transcription errors and illegible handwritten documents - Minimized actions that require human intervention, such as manual data input to LIS - Removed error sources, including losing test results, entering wrong test data, not signing, and wrong labeling of samples - Improved data integrity and accuracy - More accuracy and completeness of electronic patient records than with handwritten records - Facilitated access and sharing of information between physicians, laboratories, and patients. 39 Retrospective observational cross-sectional study - 2015 Education and communication by sending comments from the laboratory to the outpatient and inpatient sections of the hospital - More appropriate request for vitamin D test - Reduced improper test requests and hence laboratory costs - Promoted correct use and self-regulation of vitamin D test requests 40 Qualitative study - 2015 Electronic information exchange is only for laboratory result sending from laboratory to primary care physician, and test ordering is transmitted by carrier - Difficulties in sample handling - Misidentification of samples - Inefficiency in collating and resending misdirected results 32 Longitudinal interventional quasiexperimental - 2017 Exchange of information between the AIDS clinic and the laboratory using laboratory health information exchange - Improved overall treatment of antiretroviral therapy and viral suppression - Reduced black and white differences in the use of antiretroviral therapy and viral suppression 33 Cross-sectional study - 2018 Exchange of laboratory information between the laboratory and the emergency department using various laboratory information exchange systems - Efficiency, quality, and safety of emergency care - Better monitoring of patients’ health - Approval of diagnostic decision making assumptions - Applying evidence-based operations in the medical laboratory HIE, health information exchange;LIS, laboratory information systems; EHR, electronic health record; HIS, hospital information systems. Open in new tab Four studies focused on negative effects,9,28,31,40 and others evaluated the positive effects. The investigations led to the classification of the benefits of electronic communication between laboratories and ambulatory care centers in the following 3 domains: financial benefits, organizational benefits, and quality benefits. The frequency distribution of studies with respect to the benefits they mentioned is presented in Table 2. Table 2. Frequency Distribution of the Positive Effects of Electronic Exchanges between Ambulatory Care Centers and Laboratories Publication date (reference) Positive Impacts Seen (%) Financial (19) Organizational (52) Quality (76) Time Saving (48) Improved workflow (14) Improved Communication (24) Improved Care Quality (29) Improved Medical Record (19) Error Reduction (48) 200525 √ 200710 √ 200822 √ √ 200926 √ √ 200924 √ 20099 √ √ √ 201034 √ √ 201135 √ √ 201141 √ 201227 √ √ √ √ 201211 √ 201236 √ 201337 √ √ 201328 √ √ √ √ 201329 √ 201430 √ √ √ √ √ 201438 √ √ 201516 √ √ √ √ 201539 √ √ 201732 √ 201833 √ Publication date (reference) Positive Impacts Seen (%) Financial (19) Organizational (52) Quality (76) Time Saving (48) Improved workflow (14) Improved Communication (24) Improved Care Quality (29) Improved Medical Record (19) Error Reduction (48) 200525 √ 200710 √ 200822 √ √ 200926 √ √ 200924 √ 20099 √ √ √ 201034 √ √ 201135 √ √ 201141 √ 201227 √ √ √ √ 201211 √ 201236 √ 201337 √ √ 201328 √ √ √ √ 201329 √ 201430 √ √ √ √ √ 201438 √ √ 201516 √ √ √ √ 201539 √ √ 201732 √ 201833 √ Open in new tab Table 2. Frequency Distribution of the Positive Effects of Electronic Exchanges between Ambulatory Care Centers and Laboratories Publication date (reference) Positive Impacts Seen (%) Financial (19) Organizational (52) Quality (76) Time Saving (48) Improved workflow (14) Improved Communication (24) Improved Care Quality (29) Improved Medical Record (19) Error Reduction (48) 200525 √ 200710 √ 200822 √ √ 200926 √ √ 200924 √ 20099 √ √ √ 201034 √ √ 201135 √ √ 201141 √ 201227 √ √ √ √ 201211 √ 201236 √ 201337 √ √ 201328 √ √ √ √ 201329 √ 201430 √ √ √ √ √ 201438 √ √ 201516 √ √ √ √ 201539 √ √ 201732 √ 201833 √ Publication date (reference) Positive Impacts Seen (%) Financial (19) Organizational (52) Quality (76) Time Saving (48) Improved workflow (14) Improved Communication (24) Improved Care Quality (29) Improved Medical Record (19) Error Reduction (48) 200525 √ 200710 √ 200822 √ √ 200926 √ √ 200924 √ 20099 √ √ √ 201034 √ √ 201135 √ √ 201141 √ 201227 √ √ √ √ 201211 √ 201236 √ 201337 √ √ 201328 √ √ √ √ 201329 √ 201430 √ √ √ √ √ 201438 √ √ 201516 √ √ √ √ 201539 √ √ 201732 √ 201833 √ Open in new tab Financial Benefits In most cases, ambulatory care centers are administered privately by physicians and therefore need to be sure of the cost effectiveness of the systems to create the infrastructure and software needed to exchange their information with laboratories. According to the studies, electronic communication between ambulatory care centers and laboratories can lead to net societal savings at the national level,25,35 suggesting that patients and providers are likely to gain more than other stakeholders.25 However, in a study by Ross et al,37 electronic information exchanges did not significantly alter the cost of laboratory tests. Organizational Benefits Organizational benefits are those that are organizationally beneficial to e-communication partners, including time saving, workflow improvement, and improved communication. Time Saving Time saving is an organizational benefit mentioned in most studies (48%).22,24,27,29,30,38,41 Electronic exchanges can save time in different ways in different parts of the workflow of these exchanges. These exchanges initially lead to the timely delivery of results to physicians and reduce the time spent on exchanging results between laboratories and ambulatory care centers.9,22,38 They also improve the response time and timely follow-up of the patients by the physician.22–24,27,28 Decreased response time can be critical in some cases. It has been shown that electronic exchanges are helpful for patients with AIDS and tuberculosis,27 for whom timely responses of physicians to laboratory results is vital, both for quicker patient recovery and reduced risk of virus transmission. However, Singh et al did not report a similar result.23 They stated that automatic alerts would not lead to timely follow-up of laboratory results. Another effect of electronic exchanges is the reduction in turnaround time, that is, the time from the moment the test is performed to the time of delivery of the results to the physician.30 Electronic exchanges can also be effective in providing timely care to patients. In some cases, electronic exchanges significantly improve the timing of antiretroviral drugs for patients with human immunodeficiency virus,27 and occasionally, they enhance the timeliness of the screening program.26 Since various stakeholders are involved in electronic exchanges, time saving can be considered from different perspectives. Consequently, electronic exchanges can be effective in accelerating nurses’ task performance41 and reducing the time consumed by providers.29 Electronic exchanges can even provide more time efficiency for administrators than for physicians and nurses.29 Nevertheless, electronic exchanges do not have a significant effect for patients as a potential stakeholder, and the time spent on visits and waiting time is not reduced.29 Improved Cobas Workflow Electronic exchanges will alter the workflow. They can reduce paperwork and tasks requiring human intervention and thus improve workflow.16,22,30 The workflow in electronic information exchanges between offices and laboratories can have managerial and clinical trends which may involve various stakeholders, including physicians, patients, laboratory staff, nurses, and even nonclinical staff. In some cases, the positive effect of electronic exchanges is evident in accelerating and managing the workflow of patients.29 Electronic exchanges also improve administrative work.28 Of course, the impact of electronic exchanges on the workflow efficiency of clinical and nonclinical staff may differ, and in some cases, efficiency is improved more for managers than for physicians and nurses.30 Improved Communication Communication is the main part of the transfer of information. Therefore, improving communication plays an important role in the timely and accurate transmission of information to the correct destination. Various studies9,16,27,28,30 have reported improved communication as a benefit of electronic exchanges. Improved communication can be considered from the perspective of any party, including physicians and care providers,9,16,28,30 laboratories,16 and, occasionally, patients.9,16,27 Quality Benefits Some of the benefits of electronic interchange between offices and laboratories which increase the quality of the information exchanged are improved quality of care, reduction in error sources, and improved patient medical cords. These are detailed below. Improved Quality of Care Improvement of care is another benefit of electronic exchanges mentioned in the reviewed studies.10,26,27,33,42 Regarding the care area of the health centers examined in these studies, electronic exchanges enhance emergency care,33 screening programs,26 preventive care,10 ambulatory care,36 and for patients with AIDS.27 Improving care for people with AIDS can also have a positive impact on wider public health goals, such as preventing the transmission of the virus.27 Electronic exchanges can improve care in several ways. Improvement of care can be the result of ensuring the reception of accurate test results, that the results are appropriate to the tests requested, and that the results can be responded to more quickly.27 Therefore, improving the quality of AIDS care can both depend on timely responses to important laboratory results and the quality of exchanges. In some cases, electronic exchanges can improve care by fostering the better monitoring of patients’ health.33 Furthermore, the physicians’ better assessment of their diagnostic hypotheses and the application of evidence-based methods in laboratories enhance care.33 Finally, patient satisfaction can be indicative of improved care.10 Reduction in Error Sources Poor communication is a common cause of ambulatory center-to-laboratory-to-ambulatory centers problems. Information technology can reduce errors occurring during the transmission of information between ambulatory care centers and laboratories and resolve communication problems to some extent.9,16,27,28,30,34,35,37,39 Errors during information transfer can occur when requesting and performing a test or when returning the test results; electronic exchange can help reduce errors in both aspects. With the help of electronic exchange in the requesting and performing phases of a test, the request rate of a test,35,37 inappropriate requests,39 and redundancy of the tests are decreased,30 and accurate and appropriate requests are sent.27,39 In some cases, electronic exchanges can prevent the loss of test results and reduce errors in the entry of test data as well as errors that may occur in manual systems due to transcription or ineligible handwritings.16 Electronic exchanges can also improve sample transfer during the test request phase in which errors may occur. Improved process tracking, especially sample tracking (because all sample specifications are inserted correctly),28,30 as well as prevention of mislabeling16 for laboratory samples, are other benefits of electronic exchanges. According to the studies we examined, electronic exchanges have positive effects on reducing return phase errors of laboratory test results. Electronic exchanges ensure that the tests are correct and that the results are relevant to them.27 They also reduce reporting errors,34 and electronic reporting can be used to investigate the problems in result delivery.9 Moreover, the physician’s follow-up can be recorded automatically and will be measurable.9 Avoiding the loss of results is yet another benefit of electronic exchanges. Improved Patient Medical Records Increasing the integrity and completeness of information,16,22 enhancing data accuracy,16 improving data quality,22,30 and promoting data integrity16 are some of the aspects which ultimately lead to improved patient medical records. Another positive aspect of electronic exchanges is that the EHR is accepted by its users and users are willing to use the EHR.11 Challenges of Electronic Exchange between Laboratories and Offices Apart from numerous studies focusing on the benefits of electronic exchange between laboratories and ambulatory care centers, some have also reported its negative effects.9,28,31,40 Despite all the benefits mentioned, electronic exchange can be challenging due to inappropriate development of interfaces and inaccurate workflow analysis. The problems mentioned in the studies we examined include provider record issues,9 logical errors of routing results and interfaces,9 the problem of collecting and resending lost results in the transmission path,40 the configuration and maintenance problems of the EHR system,9 creating potential hazards for patient care, creating the potential for decentralized decisions by displaying undesirable and misleading results in the system,31 loss of the ability to control the aspects of the requesting process, and the uncertainty of completeness, accuracy, and adequacy of laboratory test requests for facilitating the efficient processing of laboratory operations.31 In one study, the failures of the established model were criticized by laboratories and clinicians in terms of problems and difficulties in resolving errors and the need to adapt to new nomenclature, respectively.28 Therefore, in order to prevent problems, achieve high productivity, and enjoy the benefits mentioned here, various aspects of the design and development of electronic exchanges should be taken into consideration. Discussion Ambulatory care centers centers are one of the most important healthcare organizations, as they resolve most ambulatory care needs. Physicians working in ambulatory care centers often require laboratory results for making therapeutic decisions. Therefore, there is an inevitable need for the exchange of requests and laboratory results between offices and laboratories, where electronic exchanges are proposed to avoid errors in paper-based exchanges.11,16 In practice, in order to convince ambulatory care centers, offices, and laboratories to use information technology for electronic exchanges, it is essential to make the benefits of these exchanges available to them. However, there has not yet been a specific review of the effects of electronic health information exchanges (HIEs) between ambulatory care centers and laboratories. Wu and LaRue (2017) reviewed studies published between 2010 and 201543 on the barriers and facilitators of accepting electronic exchange of health information between and within hospitals, health centers, private clinics, and nursing care facilities, but they did not focus on sharing information with secondary care centers such as laboratories. In a systematic review, Fontaine et al44 mentioned the benefits of electronic exchange of information as including improved workflow, enhanced patient quality and safety, cost savings, and increased productivity. In their review, although the information exchanges of primary care centers with laboratories were not specifically investigated, improved access to test results was considered a reliable benefit. In the present study review, we focus on references that can reflect the benefits of electronic exchanges between ambulatory care centers and laboratories, the 2 health centers that need to exchange information with each other to provide patient care. Peer-reviewed papers were categorized into 2 groups, conference papers (only 1 case41) and journal papers. The benefits derived from all studies are classified into 3 categories of financial, organizational, and quality benefits. The significance of the benefits for deciding on the importance of electronic exchanges between ambulatory care centers and laboratories can be helpful to ministry and government officials and serves as a stimulus to encourage health centers to accept the use of these electronic exchanges. Moreover, in a study by Fontaine et al45 on motivational factors (perceived and expected benefits) for participation in HIEs, internal factors included cost saving, efficiency, quality of care, and patient safety. All of these factors have somehow been addressed in the studies examined in this study review. Cost savings is an incentive for electronic exchange acceptance which may, among other things, be due to the electronic transfer of test results directly to the EHR. That is, electronic exchanges of results reduces employee time consumption and paper consumption and ultimately saves costs.45 The positive effects of electronic exchanges on costs are an issue mentioned in the literature25,35,39; Frisse et al35 specifically addressed this in the emergency department. The researchers cited a total annual societal savings of nearly $1.95 million. Annual operating costs during the study period were approximately $880,000, and the net societal savings was approximately $1.07 million. The reduction in admissions from the emergency department contributed to 97.6% of the total savings. In 1 study, the annual profit derived from the communication of external laboratories with independent outpatient centers was $18.8 billion for electronic communications based on nonstandard data and $31.8 billion for electronic communications based on standardized and coded data (full interoperability), and its net profit equaled $13.9 billion.25 Limitations There are limitations to this study review, for example, excluding papers that were not available in English. Also, studies that did not clearly mention that their studied environment was an outpatient or inpatient one were excluded from this study. Conclusion The most important contribution of this systematic review is its focus on ambulatory care centers, which provide the majority of healthcare services and, therefore, must exchange information with external institutions, including laboratories. Ambulatory care centers are often run by nonaffiliated organizations and individuals, and convincing these people to adopt and use information technology requires further knowledge about the benefits and effects of this technology. In this review, we discuss financial, organizational, and quality benefits of electronic exchanges between ambulatory care centers and laboratories which can enhance the acceptability and investability of this technology. Abbreviations Abbreviations PRISMA Preferred Reporting Items for Systematic Reviews and Meta-Analyses LIS laboratory information systems EHR electronic health record AIDS acquired immunodeficiency syndrome HIEs health information exchanges HIS hospital information systems. Appendix: Search Strategy The search strategy in 4 bibliographic databases (PubMed, Embase, Web of Science, Cochrane) included primary domains (information exchange, ambulatory care centers, laboratories) and alternative words. In each category, such words were selected that, in addition to being closely related in concept, were used in papers whose main subject was interoperability and information exchange. Using OR operator, the words in each category, along with the appropriate wildcards and truncation, formed larger sets of words. These sets were then combined with the AND and OR operators. A sample work done in the PubMed database is as follows. Table A1. Search Strategy #1 (Electronic[ti] OR computer*[ti] OR “order entry”[ti] OR exchang*[ti] OR interchang*[ti] OR electronic link*[ti] OR communication[ti] OR interoperability) #2 (laborator*[ti] OR laboratory test*[ti] OR lab test*[ti] OR order test*[ti] OR diagnostic test*[ti] OR test order*[ti] OR request*[ti] OR laboratory result*[ti] OR lab result*[ti] OR test result*[ti] OR laboratory order*[ti]) #3 (physician office*[ti] OR ambulatory[ti] OR outpatient*[ti] OR “family doctor”[ti] OR “primary healthcare”[ti] OR “primary care”[ti] OR practice*[ti]) #4 (#1 AND #2)OR(#1 AND #3) #1 (Electronic[ti] OR computer*[ti] OR “order entry”[ti] OR exchang*[ti] OR interchang*[ti] OR electronic link*[ti] OR communication[ti] OR interoperability) #2 (laborator*[ti] OR laboratory test*[ti] OR lab test*[ti] OR order test*[ti] OR diagnostic test*[ti] OR test order*[ti] OR request*[ti] OR laboratory result*[ti] OR lab result*[ti] OR test result*[ti] OR laboratory order*[ti]) #3 (physician office*[ti] OR ambulatory[ti] OR outpatient*[ti] OR “family doctor”[ti] OR “primary healthcare”[ti] OR “primary care”[ti] OR practice*[ti]) #4 (#1 AND #2)OR(#1 AND #3) Open in new tab Table A1. Search Strategy #1 (Electronic[ti] OR computer*[ti] OR “order entry”[ti] OR exchang*[ti] OR interchang*[ti] OR electronic link*[ti] OR communication[ti] OR interoperability) #2 (laborator*[ti] OR laboratory test*[ti] OR lab test*[ti] OR order test*[ti] OR diagnostic test*[ti] OR test order*[ti] OR request*[ti] OR laboratory result*[ti] OR lab result*[ti] OR test result*[ti] OR laboratory order*[ti]) #3 (physician office*[ti] OR ambulatory[ti] OR outpatient*[ti] OR “family doctor”[ti] OR “primary healthcare”[ti] OR “primary care”[ti] OR practice*[ti]) #4 (#1 AND #2)OR(#1 AND #3) #1 (Electronic[ti] OR computer*[ti] OR “order entry”[ti] OR exchang*[ti] OR interchang*[ti] OR electronic link*[ti] OR communication[ti] OR interoperability) #2 (laborator*[ti] OR laboratory test*[ti] OR lab test*[ti] OR order test*[ti] OR diagnostic test*[ti] OR test order*[ti] OR request*[ti] OR laboratory result*[ti] OR lab result*[ti] OR test result*[ti] OR laboratory order*[ti]) #3 (physician office*[ti] OR ambulatory[ti] OR outpatient*[ti] OR “family doctor”[ti] OR “primary healthcare”[ti] OR “primary care”[ti] OR practice*[ti]) #4 (#1 AND #2)OR(#1 AND #3) Open in new tab References 1. Oral B , Cullen RM , Diaz DL , Hod EA , Kratz A . Downtime procedures for the 21st century: using a fully integrated health record for uninterrupted electronic reporting of laboratory results during laboratory information system downtimes . Am J Clin Pathol. 2015 ; 143 ( 1 ): 100 – 4 . Google Scholar Crossref Search ADS PubMed WorldCat 2. The Lewin Group . The value of laboratory screening and diagnostic tests for prevention and health care improvement . Falls Church, VA : The Lewin Group ; 2009 . WorldCat COPAC 3. Carter JY , Lema OE , Wangai MW , Munafu CG , Rees PH , Nyamongo JA . Laboratory testing improves diagnosis and treatment outcomes in primary health care facilities . Afr J Lab Med. 2012 ; 1 ( 1 ): 8 . Google Scholar PubMed WorldCat 4. Noren J , Frazier T , Altman I , DeLozier J . Ambulatory medical care: a comparison of internists and family-general practitioners . N Engl J Med. 1980 ; 302 ( 1 ): 11 – 16 . Google Scholar Crossref Search ADS PubMed WorldCat 5. Crawford JM , Shotorbani K , Sharma G , et al. Improving American healthcare through “Clinical Lab 2.0”: a project Santa Fe report . Acad Pathol. 2017 ; 4 : 2374289517701067 . Google Scholar Crossref Search ADS PubMed WorldCat 6. Elder NC , McEwen TR , Flach JM , Gallimore JJ . Management of test results in family medicine offices . Ann Fam Med. 2009 ; 7 ( 4 ): 343 – 351 . Google Scholar Crossref Search ADS PubMed WorldCat 7. Elder NC , McEwen TR , Flach J , Gallimore J , Pallerla H . The management of test results in primary care: does an electronic medical record make a difference? Fam Med. 2010 ; 42 ( 5 ): 327 – 333 . Google Scholar PubMed WorldCat 8. Eisenberg JM , Williams SV , Garner L , Viale R , Smits H . Computer-based audit to detect and correct overutilization of laboratory tests . Med Care. 1977 ; 15 ( 11 ): 915 – 921 . Google Scholar Crossref Search ADS PubMed WorldCat 9. Yackel TR , Embi PJ . Unintended errors with EHR-based result management: a case series . J Am Med Inform Assoc. 2010 ; 17 ( 1 ): 104 – 107 . Google Scholar Crossref Search ADS PubMed WorldCat 10. Kern LM , Barrón Y , Blair AJ III , et al. Electronic result viewing and quality of care in small group practices . J Gen Intern Med. 2008 ; 23 ( 4 ): 405 – 410 . Google Scholar Crossref Search ADS PubMed WorldCat 11. Ramaiah M , Subrahmanian E , Sriram RD , Lide BB . Workflow and electronic health records in small medical practices . Perspect Health Inf Manag. 2012 ; 9 : 1d . Google Scholar PubMed WorldCat 12. Akhlaq A , McKinstry B , Muhammad KB , Sheikh A . Barriers and facilitators to health information exchange in low- and middle-income country settings: a systematic review . Health Policy Plan. 2016 ; 31 ( 9 ): 1310 – 1325 . Google Scholar Crossref Search ADS PubMed WorldCat 13. Yi M . Major issues in adoption of electronic health records . J Digit Inf Manag . 2018 ; 16 ( 4 ): 180 – 191 . WorldCat 14. Gambo I , Oluwagbemi O , Achimugu P . Lack of interoperable health information systems in developing countries: an impact analysis . J Health Inform Dev Ctries . 2011 ; 5 ( 1 ): 185 – 196 . WorldCat 15. Msukwa MK. User perceptions on electronic medical record system (EMR) in Malawi : Citeseer ; 2011 . Google Preview WorldCat COPAC 16. Vecellio E , Malley MW , Toouli G , Georgiou A , Westbrook JI . Data quality associated with handwritten laboratory test requests: classification and frequency of data-entry errors for outpatient serology tests . Health Inf Manag. 2015 ; 44 ( 3 ): 7 – 12 . Google Scholar PubMed WorldCat 17. Henricks WH . “Meaningful use” of electronic health records and its relevance to laboratories and pathologists . J Pathol Inform. 2011 ; 2 : 7 . Google Scholar Crossref Search ADS PubMed WorldCat 18. Lehmann CU , O’Connor KG , Shorte VA , Johnson TD . Use of electronic health record systems by office-based pediatricians . Pediatrics. 2015 ; 135 ( 1 ): e7 – e15 . Google Scholar Crossref Search ADS PubMed WorldCat 19. Reisman M . EHRs: the challenge of making electronic data usable and interoperable . P T. 2017 ; 42 ( 9 ): 572 – 575 . Google Scholar PubMed WorldCat 20. Beckwith BA , Aller RD , Brassel JH , Brodsky VB , de Baca ME . Laboratory interoperability best practices: ten mistakes to avoid . Northfield, IL : College of American Pathologists ; 2013 . COPAC 21. Moher D , Liberati A , Tetzlaff J , Altman DG ; PRISMA Group . Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement . Ann Intern Med. 2009 ; 151 ( 4 ): 264 – 269 . Google Scholar Crossref Search ADS PubMed WorldCat 22. Staes CJ , Evans RS , Rocha BH , et al. Computerized alerts improve outpatient laboratory monitoring of transplant patients . J Am Med Inform Assoc. 2008 ; 15 ( 3 ): 324 – 332 . Google Scholar Crossref Search ADS PubMed WorldCat 23. Singh H , Thomas EJ , Sittig DF , et al. Notification of abnormal lab test results in an electronic medical record: do any safety concerns remain? Am J Med. 2010 ; 123 ( 3 ): 238 – 244 . Google Scholar Crossref Search ADS PubMed WorldCat 24. Singh H , Wilson L , Petersen LA , et al. Improving follow-up of abnormal cancer screens using electronic health records: trust but verify test result communication . BMC Med Inform Decis Mak. 2009 ; 9 : 49 . Google Scholar Crossref Search ADS PubMed WorldCat 25. Walker J , Pan E , Johnston D , Adler-Milstein J , Bates DW , Middleton B . The value of health care information exchange and interoperability . Health Aff (Millwood). 2005 ; Suppl Web Exclusives : W5-10 – W5-18 . Google Scholar Crossref Search ADS WorldCat 26. Downs SM , van Dyck PC , Rinaldo P , et al. Improving newborn screening laboratory test ordering and result reporting using health information exchange . J Am Med Inform Assoc. 2010 ; 17 ( 1 ): 13 – 18 . Google Scholar Crossref Search ADS PubMed WorldCat 27. Bell DS , Cima L , Seiden DS , Nakazono TT , Alcouloumre MS , Cunningham WE . Effects of laboratory data exchange in the care of patients with HIV . Int J Med Inform. 2012 ; 81 ( 10 ): e74 – e82 . Google Scholar Crossref Search ADS PubMed WorldCat 28. Gascón F , Herrera I , Vázquez C , et al. Electronic health record: design and implementation of a lab test request module . Int J Med Inform. 2013 ; 82 ( 6 ): 514 – 521 . Google Scholar Crossref Search ADS PubMed WorldCat 29. Zhou Y , Ancker JS , Upadhye M , et al. The impact of interoperability of electronic health records on ambulatory physician practices: a discrete-event simulation study . Inform Prim Care. 2013 ; 21 ( 1 ): 21 – 29 . Google Scholar Crossref Search ADS PubMed WorldCat 30. Al Saleem N , El Metwally A , Househ MS . Electronic lab information exchange (ELIE) in Saudi Arabia. In: Mantas J , Househ MS , Hasman A , eds. Integrating Information Technology and Management for Quality of Care . Amsterdam, the Netherlands : IOS Press ; 2014 : 134 – 137 . Google Preview WorldCat COPAC 31. Wilkerson ML , Henricks WH , Castellani WJ , Whitsitt MS , Sinard JH . Management of laboratory data and information exchange in the electronic health record . Arch Pathol Lab Med. 2015 ; 139 ( 3 ): 319 – 327 . Google Scholar Crossref Search ADS PubMed WorldCat 32. Cunningham WE , Ford CL , Kinsler JJ , et al. Effects of a laboratory health information exchange intervention on antiretroviral therapy use, viral suppression, and racial/ethnic disparities . J Acquir Immune Defic Syndr. 2017 ; 75 ( 3 ): 290 – 298 . Google Scholar Crossref Search ADS PubMed WorldCat 33. Raymond L , Paré G , Maillet É , Ortiz de Guinea A , Trudel MC , Marsan J . Improving performance in the ED through laboratory information exchange systems . Int J Emerg Med. 2018 ; 11 ( 1 ): 15 . Google Scholar Crossref Search ADS PubMed WorldCat 34. Blaya JA , Shin SS , Yale G , et al. Electronic laboratory system reduces errors in National Tuberculosis Program: a cluster randomized controlled trial . Int J Tuberc Lung Dis. 2010 ; 14 ( 8 ): 1009 – 1015 . Google Scholar PubMed WorldCat 35. Frisse ME , Johnson KB , Nian H , et al. The financial impact of health information exchange on emergency department care . J Am Med Inform Assoc. 2012 ; 19 ( 3 ): 328 – 333 . Google Scholar Crossref Search ADS PubMed WorldCat 36. Kern LM , Barrón Y , Dhopeshwarkar RV , Kaushal R . Health information exchange and ambulatory quality of care . Appl Clin Inform. 2012 ; 3 ( 2 ): 197 – 209 . Google Scholar Crossref Search ADS PubMed WorldCat 37. Ross SE , Radcliff TA , Leblanc WG , Dickinson LM , Libby AM , Nease DE Jr . Effects of health information exchange adoption on ambulatory testing rates . J Am Med Inform Assoc. 2013 ; 20 ( 6 ): 1137 – 1142 . Google Scholar Crossref Search ADS PubMed WorldCat 38. Blaya JA , Shin SS , Yagui M , et al. Reducing communication delays and improving quality of care with a tuberculosis laboratory information system in resource poor environments: a cluster randomized controlled trial . PLoS One. 2014 ; 9 ( 4 ): e90110 . Google Scholar Crossref Search ADS PubMed WorldCat 39. Salinas M , López-Garrigós M , Flores E , Leiva-Salinas M , Ahumada M , Leiva-Salinas C . Education and communication is the key for the successful management of vitamin D test requesting . Biochem Med (Zagreb). 2015 ; 25 ( 2 ): 237 – 241 . Google Scholar Crossref Search ADS PubMed WorldCat 40. Litchfield I , Bentham L , Lilford R , McManus RJ , Hill A , Greenfield S . Test result communication in primary care: a survey of current practice . BMJ Qual Saf. 2015 ; 24 ( 11 ): 691 – 699 . Google Scholar Crossref Search ADS PubMed WorldCat 41. Meyer R , Lovis C . Interoperability in hospital information systems: a return-on-investment study comparing CPOE with and without laboratory integration . Stud Health Technol Inform. 2011 ; 169 : 320 – 324 . Google Scholar PubMed WorldCat 42. Dullabh P , Moiduddin A. White paper: Electronic exchange of clinical laboratory: information issues and opportunities . Washington, DC : Office of the Assistant Secretary for Planning and Evaluation, US Department of Health and Human Services ; 2008 . Google Preview WorldCat COPAC 43. Wu H , LaRue EM . Linking the health data system in the U.S.: Challenges to the benefits . Int J Nurs Sci. 2017 ; 4 ( 4 ): 410 – 417 . Google Scholar PubMed WorldCat 44. Fontaine P , Ross SE , Zink T , Schilling LM . Systematic review of health information exchange in primary care practices . J Am Board Fam Med. 2010 ; 23 ( 5 ): 655 – 670 . Google Scholar Crossref Search ADS PubMed WorldCat 45. Fontaine P , Zink T , Boyle RG , Kralewski J . Health information exchange: participation by Minnesota primary care practices . Arch Intern Med. 2010 ; 170 ( 7 ): 622 – 629 . Google Scholar Crossref Search ADS PubMed WorldCat © American Society for Clinical Pathology 2019. All rights reserved. For permissions, please e-mail: [email protected] 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)

Yang,, Lingmin;Cui,, Ranliang;Li,, Yueguo;Liang,, Kai;Ni,, Min;Gu,, Yajun

doi: 10.1093/labmed/lmz063pmid: 31626700

Abstract Objective To explore novel biomarkers for patients with pancreatic ductal adenocarcinoma (PDAC), from the perspective of tumor hypoxia. Methods We screened 29 differentially expressed and hypoxia-upregulated genes from the Oncomine database. A total of 12 secretory proteins that interact with hypoxia-inducible factor 1 (HIF-1A) were selected by STRING (protein-protein interaction networks). After excluding enzymes and collagens, insulin-like growth factor-binding protein 3 (IGFBP3), glycoprotein NBM (GPNMB), transforming growth factor–β-induced (TGFBI), and biglycan (BGN) were detected by sandwich enzyme-linked immunosorbent assay (ELISA) in patients with cancer and healthy control individuals. Results The serum level of TGFBI was significantly elevated in patients with PDAC, compared with healthy controls; the assay could discriminate among cases of PDAC in different clinical stages. The amount of TGFBI was significantly decreased after treatment. The combination of TGFBI and cancer antigen (CA) 19-9 was more accurate than TGFBI or CA 19-9 alone as diagnostic markers. Also, TGFBI might be used as a prognostic marker according to the PROGgeneV2 Pan Cancer Prognostics Database. Conclusions Serum TGFBI, combined with CA 19-9, offers higher diagnostic value than other methods for patients with PDAC. Also, TGFBI might be used as a prognostic marker. biomarker, hypoxia, pancreatic ductal adenocarcinoma, TGFBI, HIF-1A, CA 19-9 Pancreatic ductal adenocarcinoma (PDAC), the main subtype of malignant neoplasm in the pancreas, accounts for approximately 85% of new diagnosed cases.1 According to the GLOBOCAN 2018 online database, pancreatic cancer ranks as the seventh-leading cause of cancer deaths worldwide, and the incidence of and death rate from pancreatic cancer has been rising for 10 years.2 In addition, it has been estimated3 that pancreatic cancer will replace breast cancer as the third-leading cause of cancer deaths in the near future. In clinical practice, pancreatic cancer is extremely difficult to detect and diagnose early because of the lack of reliable, sensitive biomarkers or other approaches, which leads to the estimated high death rate from the disease. Cancer antigen (CA) 19-9 is regarded as the most common biomarker for clinical routine use in patients with pancreatic cancer; it is used as a laboratory diagnostic and prognostic tool. However, CA 19-9 by itself is not recommended for screening of pancreatic cancer by the American Society of Clinical Oncology (ASCO) in recent decades, mainly because of the relatively low sensitivity in detecting neoplasms at early stages.1,4 Still, previous reports have found that an elevated serum CA 19-9 level may predict disease progression, recurrence, or metastasis in patients with pancreatic cancer. There is insufficient evidence to persuade us to believe that CA 19-9 measurement by itself could be a perfect marker for prognosis without confirmation by imaging or pathologic testing, according to ASCO guidelines.4 Thus, novel blood biomarkers of early detection or clinical prognosis are urgently needed for diagnostic research. Hypoxia, a common feature of solid tumors, has been associated with malignant migration, distant metastasis, resistance to therapy, and poor outcomes.5 Clinical investigators6 have discovered that tissues harboring pancreatic-cancer cells have been deprived of oxygen without exception. It is well accepted that hypoxia plays a pivotal role in driving metastatic spread. Therefore, focusing on hypoxia-inducible factors and related markers may offer an alternative, and possibly more efficient, way to identify novel biomarkers for patients with pancreatic cancer. Materials and Methods Sample Size Calculation Serum specimens from 5 patients with PDAC and 10 healthy people (as determined by their checkup results) were collected via preliminary enzyme-linked immunosorbent assay (ELISA) testing for sample size evaluation. The G* Power online tool, version 3.1, was used to calculate statistical power.7 Clinical Specimens In our study, we prospectively enrolled 28 patients with PDAC and 56 healthy individuals at Tianjin Medical University Cancer Institute and Hospital, Tianjin, China, from October 2017 through April 2018. Enrolled patients underwent the curative resection, and their specimens were confirmed to be cancerous by using pathologic testing. Patients younger than 18 years or who were pregnant were excluded from the project. The study was specifically approved by the institutional review board; written informed consent was obtained from all participants. All examinations were performed in accordance with the ethical standards presented in the Declaration of Helsinki, as revised in 2013. In each patient, we obtained 5 mL of whole blood, using a serum separator tube, before and after the operation. After clotting, which usually took 20 minutes at room temperature, the blood was centrifuged at 1000×g for 15 minutes, and the supernatant was immediately transferred into a clean test tube. Finally, the serum was aliquoted into a 0.5-mL tube, stored, and transported at −20°C or lower. Serum biomarkers for patients with PDAC were measured using quantitative ELISA. ELISA Serum levels of CA 19-9, transforming growth factor–β-induced (TGFBI), glycoprotein NBM (GPNMB), biglycan (BGN), and insulin-like growth factor-binding protein 3 (IGFBP3) were detected using a commercial sandwich ELISA kit according to manufacturer-provided instructions. For detection of CA 19-9, 50 μL of standard or specimen solution (1:10 dilution) was added into each well. After pipetting 50 μL of horseradish peroxidase (HRP)–conjugate antibody, the microtiter plate, which contained antibody-antigen complex, was incubated for 1 hour at 37°C. Then, the microplate washer (Wellwash, Thermo Fisher Scientific Inc) was used for automated washing, aspiration, dispensing, and shaking of 96-well plates. Next, 50 μL of substrate A and 50 μL of substrate B were added into each well, followed by incubation in a dark room for 15 minutes at 37°C. Finally, the optical density was analyzed by a microplate reader (Synergy HT; Biotek Instruments, Inc.) at 450 nm within 10 minutess, after adding 50 μL of stop solution. When detecting TGFBI, 100 μL of standard or specimen solution (1:2000 dilution) was added into each well, followed by incubation at 37°C for 90 minutes. After washing, 100 μL of anti-TGFBI antibody was added and incubated for 1 hour at 37°C. Repeat washing, 100 μL of substrate A was transferred into each well and then incubated at 37°C for 30 minutes. Next, 90 μL of TMB solution was added after carefully washing, and incubated in the dark room for 30 minutes at 37°C. Finally, the optical density was analyzed by a microplate reader (SynergyHT) at 450 nm, after adding 100 μL of stop solution. Measurements of serum GPNMB (1:50 dilution), BGN (1:50 dilution), and IGFBP3 (1:1000 dilution) were conducted in accordance with the protocol of TGFBI. Statistics The paired t test was used to compare differences in the same patients before and after surgical procedures. For 2 unpaired groups, the Mann-Whitney test was conducted to analyze differences with a non-normal distribution or unequal variance. The unpaired t test was suitable for a normal distribution and equal variance. A 2-sided P value of less than .05 was considered statistically significant. We calculated receiver operating characteristic (ROC) curves to determine the diagnostic value and optimal cutoff of candidate biomarkers. According to the area under the curve (AUC), the serum marker could be divided into 5 groups, including noninformative (AUC ≤0.5), less accurate (0.5 < AUC ≤ 0.7), moderately accurate (0.7 < AUC ≤ 0.9), highly accurate (0.9 < AUC < 1), and prefect tests (AUC = 1).8 Statistical analysis was performed with the SPSS software, version 22.0 (IBM Corporation). Results Differentially Expressed and Hypoxia-Upregulated Genes from the Online Database We searched the Oncomine database for the messenger RNA (mRNA) expression signatures in pancreatic cancer with the following keywords: “Concept: Upregulated genes in response to hypoxia and in response to HIF-1 expression–Literature-defined Concepts”, “Primary Filters—Analysis Type-Differential analysis-Cancer vs. normal analysis-Pancreatic Cancer vs. Normal Analysis”, “Dataset Filters-Data Type-mRNA”, and “Sample Filters-Sample Type: Tissue Specimen”. The search returned 5 records in the online database (Table S1). According to the literature-defined concepts, 191 genes were upregulated in response to hypoxia and in response to HIF-1 expression (Table S2). Combined with the differentially expressed genes between serum from patients with pancreatic cancer and healthy patients, there were 48, 55, 28, 37, and 26 potential targets in the Badea, Segara, Logsdon, Iacobuzio, and Pei Pancreas datasets, respectively (Tables S3–S7). The Venn diagram showed that 29 genes were differentially expressed and hypoxia-upregulated in at least 3 cohorts among 5 pancreatic studies (Figure 1). In the 29 genes listed earlier herein, 14 candidate proteins were found to be directly or indirectly interacted with HIF-1A by the online protein-protein network analyzer STRING V10.5 (Figure 2).9,10 By searching the Human Protein Atlas,11 we found 12 secretory proteins that could be used as blood biomarkers for pancreatic cancer. Considering the stability of proteins in the blood, 8 of 12 nonenzyme proteins were selected for the next step. Then, 4 kinds of collagens were excluded from the biomarker-discovery process for the alteration of serum levels after the operations. Finally, IGFBP3, GPNMB, TGFBI, and BGN were found to be potential biomarkers for patients with pancreatic cancer (Figure 3). Figure 1 Open in new tabDownload slide The differentially expressed and hypoxia-upregulated genes in 5 pancreatic studies from the Oncomine database. Figure 1 Open in new tabDownload slide The differentially expressed and hypoxia-upregulated genes in 5 pancreatic studies from the Oncomine database. Figure 2 Open in new tabDownload slide The interactions of candidate proteins with hypoxia-inducible factor 1 (HIF-1A) by the online protein-protein network analyzer STRING V10.5. Figure 2 Open in new tabDownload slide The interactions of candidate proteins with hypoxia-inducible factor 1 (HIF-1A) by the online protein-protein network analyzer STRING V10.5. Figure 3 Open in new tabDownload slide The rootmap for the biomarker discovery. Insulin-like growth factor–binding protein 3 (IGFBP3), glycoprotein NBM (GPNMB), transforming growth factor–β-induced (TGFBI), and biglycan (BGN) were found to be potential biomarkers for patients with pancreatic cancer. Figure 3 Open in new tabDownload slide The rootmap for the biomarker discovery. Insulin-like growth factor–binding protein 3 (IGFBP3), glycoprotein NBM (GPNMB), transforming growth factor–β-induced (TGFBI), and biglycan (BGN) were found to be potential biomarkers for patients with pancreatic cancer. Sample Size Determination To reach the desired power of a statistical test, the sample size was estimated by G*Power 3.1 software before detection of IGFBP3, GPNMB, TGFBI, and BGN in human serum.7 The protocol for sample-size calculation was summarized in Figure 4. In the preliminary test, serum levels of TGFBI, GPNMB, BGN, and IGFBP3 were measured in 5 patients with PDAC (group 1; n = 5) and 10 healthy individuals (group 2; n = 10). The sample size was computed by given α, power (1-β), and effect size in 2 independent groups according to the results of t testing. Serum concentrations of 4 candidate markers were shown as mean (SD), which determined the effect size. α error is generally established beforehand as 0.05, and 0.9 for power (1-β). The minimal sample size was 25 and 49 for groups 1 and 2, respectively. Therefore, serum levels of 4 candidate markers were detected in more than 25 patients with PDAC and 49 healthy controls. Figure 4 Open in new tabDownload slide The procedures for sample-size calculation by G*Power 3.1. By given α, power (1-β), and effect size in 2 independent groups, the sample size was estimated to be more than 25 patients with pancreatic adenocarcinoma and 49 healthy control individuals. Figure 4 Open in new tabDownload slide The procedures for sample-size calculation by G*Power 3.1. By given α, power (1-β), and effect size in 2 independent groups, the sample size was estimated to be more than 25 patients with pancreatic adenocarcinoma and 49 healthy control individuals. Serum Concentration of TGFBI, GPNMB, BGN, and IGFBP3 Of the 4 proteins differentially expressed and hypoxia-upregulated in patients with pancreatic cancer, serum levels of TGFBI, GPNMB, BGN, and IGFBP3 were detected by sandwich ELISA in patients with cancer (n = 28) and healthy individuals (n = 56). The demographic and clinical characteristics of patients with PDAC and healthy individuals were summarized in Table S8. The serum concentrations of TGFBI were significantly higher in patients with PDAC than those in healthy controls (median [minimum–maximum]; 2141.94 ng/mL [616.91–8071.47] vs 1023.28 ng/mL [535.94–2115.65]; P <.001;, Mann-Whitney testing; Figure 5A). Similarly, the serum levels of GPNMB were significantly higher in patients with PDAC than in healthy controls (median [minimum–maximum]; 3747.85 pg/mL [2063.05–6505.00] vs 2542.95 pg/mL [1233.45–4654.25]; P <.001; Mann-Whitney test; Figure 5B). Serum levels of BGN were significantly higher in patients with PDAC than normal controls (median [minimum–maximum]; 2865.48 pg/mL [1771.20–6007.00] vs 2426.78 pg/mL [979.75–5072.50]; P = 0.01; Mann-Whitney test; Figure 5C). In contrast, serum levels of IGFBP3 were significantly lower in patients with PDAC than those observed in controls (median [minimum–maximum]; 3261.13 ng/mL [1321.09–10856.08] vs 4902.25 ng/mL [634.05–9518.67]; P <.001; Mann-Whitney test; Figure 5D). Figure 5 Open in new tabDownload slide Serum concentration of transforming growth factor–β-induced (TGFBI), glycoprotein NBM (GPNMB), biglycan (BGN), and insulin-like growth factor–binding protein 3 (IGFBP3) in patients with cancer (n = 28) and healthy control individuals (n = 56), as detected by a commercial enzyme-linked immunosorbent assay (ELISA) kit. A, Serum concentration of TGFBI in patients with pancreatic ductal adenocarcinoma (PDAC) and controls. B, Serum concentration of GPNMB in patients with PDAC and controls. C, Serum concentration of BGN in patients with PDAC and controls. D, Serum IGFBP3 levels were significantly reduced in cancer cases. E, Serum TGFBI concentrations were significantly increased in patients with cancer at stages I or II, compared with healthy controls, as well as in stages III + IV vs I + II. F, Serum GPNMB assay results among healthy controls vs patients with stages I + II, or I + II and III + IV cancer. G, Serum BGN assay results among healthy controls vs and patients with stages I and II, or I + II, III + IV. H, In 6 patients with PDAC, serum levels of TGFBI were found to be significantly decreased after treatment. * indicates P <.05; **, P <.01; ***, P <.001; and ****, P <.001. Figure 5 Open in new tabDownload slide Serum concentration of transforming growth factor–β-induced (TGFBI), glycoprotein NBM (GPNMB), biglycan (BGN), and insulin-like growth factor–binding protein 3 (IGFBP3) in patients with cancer (n = 28) and healthy control individuals (n = 56), as detected by a commercial enzyme-linked immunosorbent assay (ELISA) kit. A, Serum concentration of TGFBI in patients with pancreatic ductal adenocarcinoma (PDAC) and controls. B, Serum concentration of GPNMB in patients with PDAC and controls. C, Serum concentration of BGN in patients with PDAC and controls. D, Serum IGFBP3 levels were significantly reduced in cancer cases. E, Serum TGFBI concentrations were significantly increased in patients with cancer at stages I or II, compared with healthy controls, as well as in stages III + IV vs I + II. F, Serum GPNMB assay results among healthy controls vs patients with stages I + II, or I + II and III + IV cancer. G, Serum BGN assay results among healthy controls vs and patients with stages I and II, or I + II, III + IV. H, In 6 patients with PDAC, serum levels of TGFBI were found to be significantly decreased after treatment. * indicates P <.05; **, P <.01; ***, P <.001; and ****, P <.001. Because serum levels of TGFBI, GPNMB, and BGN were significantly elevated in patients with PDAC, they were more suitable for the candidate biomarker. For cancer screening and early detection, sensitivity is the top priority; at this point, genes expressed at high levels are superior to genes expressed at low levels, which may lead to false-negative results because of instrument detection limits or human operation error. Next, the serum concentrations of the 3 potential targets mentioned earlier herein were analyzed in patients whose cancer was at different clinical stages. As shown in Figure 5E, serum TGFBI could discriminate between healthy controls and patients with stage I + II cancer or I + II and III + IV cancer (P <.05). However, no statistical difference was found for the other 2 secreted proteins, GPNMB and BGN, or between healthy controls and patients with stage I + II cancer or I + II and III + IV cancer(Figure 5F, Figure 5G). In 6 patients with PDAC, serum levels of TGFBI were found to be significantly decreased after treatment (Figure 5H), which indicated that TGFBI might be used as a potential diagnostic or prognostic biomarker for patients with cancer. TGFBI Combined with CA 19-9 Offers Higher Diagnostic Value Because TGFBI might offer higher diagnostic value in the 4 potential biomarkers mentioned earlier herein, we analyzed the clinical performance of TGFBI by the ROC curve (Figure 6). Based on the protein levels of TGFBI in serum specimens, the optimal cutoff value of 1173.04 ng per mL led to sensitivity of 89.30% and specificity of 76.80%. The AUC of TGFBI was determined to be 0.872, with the 95% confidence interval (CI) from 0.774 to 0.970. As a moderately accurate marker, the AUC of TGFBI was slightly lower than the CA 19-9 value of 0.932 (95% CI, 0.861–1.000). However, the combination of TGFBI and CA 19-9 proved to be more accurate than TGFBI or CA 19-9 alone (AUC = 0.975; 95% CI, 0.943–1.000). Figure 6 Open in new tabDownload slide The receiver operating characteristic (ROC) analysis of the diagnostic performance of serum transforming growth factor–β-induced (TGFBI). TGFBI combined with cancer antigen (CA) 19-9 offers higher diagnostic value for patients with pancreatic cancer (area under the curve = 0.975; 95% confidence interval, 0.943–1.000). Figure 6 Open in new tabDownload slide The receiver operating characteristic (ROC) analysis of the diagnostic performance of serum transforming growth factor–β-induced (TGFBI). TGFBI combined with cancer antigen (CA) 19-9 offers higher diagnostic value for patients with pancreatic cancer (area under the curve = 0.975; 95% confidence interval, 0.943–1.000). Prognostic Value of Serum TGFBI For assessment of the prognostic application of TGFBI, we searched the online pancancer prognostics database PROGgeneV2.12 The input gene was defined as “TGFBI”, survival measure was selected to be “death”, and the cancer type as “pancreas”. The low-expression and high-expression groups were divided by the median of TGFBI gene expression. There were 5 pancreatic studies containing survival information in patients (GSE21501, GSE28735, GSE50827, GSE57495, and TCGA data pancreatic adenocarcinoma), excluding 1 study with cell lines (GSE71729). In 3 cohorts (GSE28735, GSE57495, and TCGA data pancreatic adenocarcinoma), patients with pancreatic cancer with a high TGFBI expression level had significantly shorter overall survival than those with a low TGFBI expression level (P <.05), which indicated that serum TGFBI might be used as a prognostic biomarker for patients with pancreatic cancer (Figure S1–S5). Discussion As a common characteristic of human carcinomas, especially malignant neoplasms, tumor hypoxia plays an important role in pancreatic-cancer pathogenesis, mainly because the insufficient oxygen supply and abnormal vascularization could not provide adequate blood to the ever-increasing tumor tissues.5,13 Two research groups14,15 have concluded that hypoxia has a negative impact on the effectiveness of clinical treatment and that hypoxic tumors may lead to adverse outcomes. Thus, we propose that exploring hypoxic markers might shed some light on the laboratory diagnosis or prognosis for patients with pancreatic cancer. Recently, the criterion standard for oxygen measurements is the polarographic electrode, which is inserted into the primary or metastatic tumor tissues. Thus, there is an urgent need for noninvasive methods to detect and characterize hypoxia as tumor biomarkers. We preliminarily found 29 differentially expressed and hypoxia-upregulated genes in at least 3 cohorts among 5 pancreatic studies from the Oncomine database. Then, 12 secretory proteins that interact with HIF-1A were selected for the next round. After excluding the enzymes and collagens, IGFBP3, GPNMB, TGFBI, and BGN were found to be potential biomarkers for patients with pancreatic cancer. Then, serum levels of the 4 proteins listed earlier herein were detected by sandwich ELISA in patients with cancer (n = 28) and healthy individuals (n = 56). Serum levels of TGFBI, GPNMB, and BGN were significantly elevated in patients with PDAC. Serum TGFBI could discriminate between healthy controls and patients with I + II, or I + II and III + IV (P <.05); also, serum levels of TGFBI were found to be significantly reduced after curative treatment. According to the ROC curve, the combination of TGFBI and CA 19-9 proved to be more accurate than TGFBI or CA 19-9 alone (AUC = 0.975; 95% CI, 0.943–1.000), which indicate that serum TGFBI, combined with CA 19-9, offers higher diagnostic value for patients with PDAC. By searching PROGgeneV2, we found that serum TGFBI might be used as an additional prognostic biomarker. TGFBI (or BIGH3/BIG-H3) is a 68-kD extracellular matrix protein consisting of 683 amino acids with 4 evolutionarily conserved fasciclin-1 domains and a C-terminal Arg-Gly-Asp motif.16 As early as 1992, TGFBI was first reported in a human lung adenocarcinoma cell line that was a downstream protein of transforming growth factor β (TGF-β)17; later, it was confirmed in other neoplasms, including pancreatic cancer and insulinoma.18,19 As a secreted protein, TGFBI was directly involved in several clinical conditions, such as malignant neoplasms, diabetes, and corneal dystrophy.20 High expression of TGFBI was observed in several solid tumors, such as pancreatic carcinoma and oral squamous-cell carcinoma,18,21 and it was identified as a promising prognostic factor in renal cell carcinoma and gastric cancer.22,23 Schneider et al18 found that there was a 32.4-fold increase in TGFBI mRNA levels in human pancreatic tissues, compared with those of healthy controls, and TGFBI upregulation was induced by TGF-β in pancreatic-cancer cells. Turtoi et al24 reported that TGFBI, LTBP2, and asporin (ASPN) were novel, overexpressed, and potentially accessible proteins in human PDAC, which had the potential to be of clinical value for diagnostic and therapeutic applications by using 2D-nano–high-performance liquid chromatography (HPLC)–combination of 2 mass spectrometry analyzers (MS/MS) in fresh tissue with PDAC and fresh healthy tissue. Sato et al25 conducted an immunohistochemical analysis with anti-TGFBI antibody in 75 case individuals with PDAC, which showed a significant association between strong TGFBI staining and poor patient outcome by univariate analyses and log-rank tests. Han et al20 demonstrated that TGFBI acted as a promoter in several gastrointestinal-tract cancers, including cholangiocarcinomas, hepatic carcinomas, and gastric carcinomas, which induced the FAK/AKT/AKT1S1/PRS6/EIF4EBP pathway to modulate cell survival and proliferation. Previous studies mainly focus on tissue levels of TGFBI and the clinical utility of TGFBI as a prognostic biomarker in patients with PDAC. We found that serum TGFBI could be more useful in diagnosis and prognosis of PDAC. Also, we report that upregulation of TGFBI may be attributed to tumor hypoxia, as well as hypoxia-induced factors such as HIF-1; this latter finding needs to be further elucidated. Conclusions Reliable, sensitive markers are urgently needed for laboratory diagnostics and stratification of patients with pancreatic ductal adenocarcinoma (PDAC) who will benefit from clinical treatment. By focusing on hypoxia-related markers, we have found an alternative, noninvasive, and possibly more efficient way to identify new candidates for patients with pancreatic cancer. Serum TGFBI, combined with CA 19-9, offers higher diagnostic value for patients with PDAC. More importantly, TGFBI might be used as a prognostic biomarker as well; this idea requires more experimental validation. Supplementary Materials Supplementary tables and figures can be found online at www.labmedicine.com. Funding This work was supported by National Natural Science Foundation of China [grant numbers 81601820; 21705160]. Abbreviations Abbreviations PDAC pancreatic ductal adenocarcinoma CA cancer antigen ASCO American Society of Clinical Oncology ELISA enzyme-linked immunosorbent assay TGFBI transforming growth factor–β-induced GPNMB glycoprotein NBM BGN biglycan IGFBP3 insulin-like growth factor-binding protein 3 HRP horseradish peroxidase ROC receiver operating characteristic AUC area under the curve mRNA messenger RNA CI confidence interval TGF-β transforming growth factor β ASPN aspirin HPLC high-performance liquid chromatography MS/MS combination of 2 mass spectrometry analyzers PDAC pancreatic ductal adenocarcinoma HIF-1A hypoxia-inducible factor 1 References 1. Ilic M , Ilic I . Epidemiology of pancreatic cancer . World J Gastroenterol. 2016 ; 22 ( 44 ): 9694 – 9705 . Google Scholar Crossref Search ADS PubMed WorldCat 2. Bray F , Ferlay J , Soerjomataram I , Siegel RL , Torre LA , Jemal A . Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries . CA Cancer J Clin. 2018 ; 68 ( 6 ): 394 – 424 . Google Scholar Crossref Search ADS PubMed WorldCat 3. Ferlay J , Partensky C , Bray F . More deaths from pancreatic cancer than breast cancer in the EU by 2017 . Acta Oncol. 2016 ; 55 ( 9–10 ): 1158 – 1160 . Google Scholar Crossref Search ADS PubMed WorldCat 4. Locker GY , Hamilton S , Harris J , et al. ; ASCO . ASCO 2006 update of recommendations for the use of tumor markers in gastrointestinal cancer . J Clin Oncol. 2006 ; 24 ( 33 ): 5313 – 5327 . Google Scholar Crossref Search ADS PubMed WorldCat 5. Erkan M , Kurtoglu M , Kleeff J . The role of hypoxia in pancreatic cancer: a potential therapeutic target? Expert Rev Gastroenterol Hepatol. 2016 ; 10 ( 3 ): 301 – 316 . Google Scholar Crossref Search ADS PubMed WorldCat 6. Koong AC , Mehta VK , Le QT , et al. Pancreatic tumors show high levels of hypoxia . Int J Radiat Oncol Biol Phys. 2000 ; 48 ( 4 ): 919 – 922 . Google Scholar Crossref Search ADS PubMed WorldCat 7. Faul F , Erdfelder E , Buchner A , Lang AG . Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses . Behav Res Methods. 2009 ; 41 ( 4 ): 1149 – 1160 . Google Scholar Crossref Search ADS PubMed WorldCat 8. Greiner M , Pfeiffer D , Smith RD . Principles and practical application of the receiver-operating characteristic analysis for diagnostic tests . Prev Vet Med. 2000 ; 45 ( 1-2 ): 23 – 41 . Google Scholar Crossref Search ADS PubMed WorldCat 9. Szklarczyk D , Morris JH , Cook H , et al. The STRING database in 2017: quality-controlled protein-protein association networks, made broadly accessible . Nucleic Acids Res. 2017 ; 45 ( D1 ): D362 – D368 . Google Scholar Crossref Search ADS PubMed WorldCat 10. Snel B , Lehmann G , Bork P , Huynen MA . STRING: a web-server to retrieve and display the repeatedly occurring neighbourhood of a gene . Nucleic Acids Res. 2000 ; 28 ( 18 ): 3442 – 3444 . Google Scholar Crossref Search ADS PubMed WorldCat 11. Uhlén M , Fagerberg L , Hallström BM , et al. Proteomics. Tissue-based map of the human proteome . Science. 2015 ; 347 ( 6220 ): 1260419 . Google Scholar Crossref Search ADS PubMed WorldCat 12. Goswami CP , Nakshatri H . PROGgene: gene expression based survival analysis web application for multiple cancers . J Clin Bioinforma. 2013 ; 3 ( 1 ): 22 . Google Scholar Crossref Search ADS PubMed WorldCat 13. Sormendi S , Wielockx B . Hypoxia pathway proteins as central mediators of metabolism in the tumor cells and their microenvironment . Front Immunol. 2018 ; 9 : 40 . Google Scholar Crossref Search ADS PubMed WorldCat 14. Vaupel P , Mayer A . The clinical importance of assessing tumor hypoxia: relationship of tumor hypoxia to prognosis and therapeutic opportunities . Antioxid Redox Signal. 2015 ; 22 ( 10 ): 878 – 880 . Google Scholar Crossref Search ADS PubMed WorldCat 15. Walsh JC , Lebedev A , Aten E , Madsen K , Marciano L , Kolb HC . The clinical importance of assessing tumor hypoxia: relationship of tumor hypoxia to prognosis and therapeutic opportunities . Antioxid Redox Signal. 2014 ; 21 ( 10 ): 1516 – 1554 . Google Scholar Crossref Search ADS PubMed WorldCat 16. Kawamoto T , Noshiro M , Shen M , et al. Structural and phylogenetic analyses of RGD-CAP/beta ig-h3, a fasciclin-like adhesion protein expressed in chick chondrocytes . Biochim Biophys Acta. 1998 ; 1395 ( 3 ): 288 – 292 . Google Scholar Crossref Search ADS PubMed WorldCat 17. Skonier J , Neubauer M , Madisen L , Bennett K , Plowman GD , Purchio AF . cDNA cloning and sequence analysis of beta ig-h3, a novel gene induced in a human adenocarcinoma cell line after treatment with transforming growth factor-beta . DNA Cell Biol. 1992 ; 11 ( 7 ): 511 – 522 . Google Scholar Crossref Search ADS PubMed WorldCat 18. Schneider D , Kleeff J , Berberat PO , et al. Induction and expression of βig-h3 in pancreatic cancer cells . Biochim Biophys Acta. 2002 ; 1588 ( 1 ): 1 – 6 . Google Scholar Crossref Search ADS PubMed WorldCat 19. Nabokikh A , Ilhan A , Bilban M , et al. Reduced TGF-β1 expression and its target genes in human insulinomas . Exp Clin Endocrinol Diabetes. 2007 ; 115 ( 10 ): 674 – 682 . Google Scholar Crossref Search ADS PubMed WorldCat 20. Han B , Cai H , Chen Y , et al. The role of TGFBI (βig-H3) in gastrointestinal tract tumorigenesis . Mol Cancer. 2015 ; 14 : 64 . Google Scholar Crossref Search ADS PubMed WorldCat 21. Tomioka H , Morita K , Hasegawa S , Omura K . Gene expression analysis by cDNA microarray in oral squamous cell carcinoma . J Oral Pathol Med. 2006 ; 35 ( 4 ): 206 – 211 . Google Scholar Crossref Search ADS PubMed WorldCat 22. Lebdai S , Verhoest G , Parikh H , et al. Identification and validation of TGFBI as a promising prognosis marker of clear cell renal cell carcinoma . Urol Oncol. 2015 ; 33 ( 2 ): 69.e11 – 69.e18 . Google Scholar Crossref Search ADS WorldCat 23. Suzuki M , Yokobori T , Gombodorj N , et al. High stromal transforming growth factor β-induced expression is a novel marker of progression and poor prognosis in gastric cancer . J Surg Oncol. 2018 ; 118 ( 6 ): 966 – 974 . Google Scholar Crossref Search ADS PubMed WorldCat 24. Turtoi A , Musmeci D , Wang Y , et al. Identification of novel accessible proteins bearing diagnostic and therapeutic potential in human pancreatic ductal adenocarcinoma . J Proteome Res. 2011 ; 10 ( 9 ): 4302 – 4313 . Google Scholar Crossref Search ADS PubMed WorldCat 25. Sato T , Muramatsu T , Tanabe M , Inazawa J . Identification and characterization of transforming growth factor beta-induced in circulating tumor cell subline from pancreatic cancer cell line . Cancer Sci. 2018 ; 109 ( 11 ): 3623 – 3633 . Google Scholar Crossref Search ADS PubMed WorldCat Author notes These coauthors contributed equally to this manuscript. © American Society for Clinical Pathology 2019. All rights reserved. For permissions, please e-mail: [email protected] 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)

Philip,, Joseph;Bajaj, Anantpreet, Kaur;Sharma,, Sanjeevan;Kushwaha,, Neerja;Kumar,, Sudeep;Kumar Biswas,, Amit

doi: 10.1093/labmed/lmz069pmid: 31758694

Abstract Background Donor factors have a variable correlation with cluster of differentiation (CD)34+ cell dose in allogeneic peripheral blood stem cell (PBSC) harvests. CD34+ cell dose affects the speed of hematopoietic recovery and percentage of donor chimerism in the recipient. Methods A total of 25 allogeneic PBSC transplants performed during a 3-year period were included. All donors underwent mobilization with filgrastim. Leukapheresis, flowcytometric CD34+ cell enumeration, and chimerism analysis were performed and correlated with recipient outcome. Results Besides age, all other donor parameters had a positive correlation with CD34+ cell count. Engraftment kinetics and chimerism had a positive correlation with the CD34+ yield of the PBSC product. Acute graft-vs-host disease (GVHD) was observed in patients with complete chimerism at day 30 after transplantation. Conclusion Adequate CD34+ cell yield happens in healthy donors, independent of donor demographic patterns with G-CSF only. A diverse population of donors can thus be approached for Matched Unrelated Donor (MUD) transplants. An accurate quantitative analysis of early donor chimerism in the recipient (at day 30) is an excellent tool for post-transplant monitoring for acute GvHD. allogeneic stem cell transplant, donor parameters, CD34+ cell enumeration, engraftment kinetics, chimerism analysis, acute GVHD Allogeneic hematopoietic stem cell transplantation (HSCT) emerged as a treatment modality for life-threatening hematological disorders in the 1990s.1 Various sources (mobilized peripheral blood stem cells [PBSC], bone marrow or umbilical cord blood) have been used for allogeneic HSCT.2,3 PBSC donation has potential benefits compared with the other methods.3 Administration of granulocyte colony-stimulating factor (G-CSF) at a dose of 7.5 to 10 μg per kg subcutaneously consecutively for 4 to 5 days, followed by leukapheresis on day 5, allows the collection of adequate number of hematopoietic progenitor cells (HPCs) in nearly all the healthy individuals if 2 to 4 times the blood volume of the donor is processed. 4A target value of 4 to 6 × 106 cluster of differentiation (CD)34+ cells (surrogate marker for human stem cells) per kg body weight of the recipient is recommended, to ensure an adequate engraftment without increasing the chances of GVHD.5 An adequate yield is important for successful transplant outcome. Donor characteristics such as age, sex, body weight, and preprocedure white blood cell (WBC) count have been identified as possible predictors of good PBSC mobilization in peripheral blood stem-cell donors and have been found to correlate with stem-cell yield.4,6 The dose of CD34+ cells infused in a PBSC-transplant recipient is also an important predictor of neutrophil and platelet recovery. This variable serves as a biomarker of stem-cell engraftment kinetics.7 Currently, chimerism monitoring is being performed routinely at centers that perform allogeneic stem-cell transplantation to evaluate the stability of donor engraftment.8 Early recipient chimerism patterns can predict graft rejection, persistent disease, or failure.8The development of full chimerism in a patient who has received a transplant means that more than 95% of the cells are derived from the donor.8 Acute GVHD occurs due to a complex interaction of T cells in the PBSC product against the host tissues within the first few weeks after transplantation.9 Chronic GVHD occurs later and is arbitrarily defined as the presence or persistence of GVHD beyond 100 days after transplantation.9 The present study analyses the donor factors: age, sex, body weight, preprocedure WBC count and platelet count, and the impact of these factors on the CD34+ yield. Further, we analyzed the impact of the yield on the time to neutrophil and platelet recovery. Finally, chimerism analysis was performed in the recipient at day 30 to correlate the percentage of chimerism in the recipient at day 30 with the outcome at 100 days after transplantation. Materials and Methods Patients A total of 25 allogeneic stem-cell transplants during a period of 3 years (January 2016–December 2018) performed at our center were included in the study. All patients had a human leukocyte antigen (HLA)–matched related donor (10/10 match using next-generation sequencing [HiSeq 2000; Illumina, Inc.]). The diagnosis in this group mostly included leukemia (acute myeloid leukemia [AML], acute lymphoblastic leukemia [ALL], juvenile myelomonocytic leukemia [JMML]), aplastic anemia, and thalassemia. PBCS Mobilization Regimen The PBSC mobilization technique was performed according to the institutional protocol. Cells from all donors were mobilized by administering filgrastim injections at a dose of 16 μg per kg once or twice daily, followed by collection on day 5. Collection Leukapheresis was performed with a continuous-flow blood-cell separator (COM.TEC, Fresenius Kabi, AG). Venous access was obtained by a double lumen femoral catheter. Anticoagulant citrate dextrose solution A (ACD-A) was used as the anticoagulant and was infused at a ratio of 1 mL of anticoagulant to 9 mL of whole blood. The inlet draw rate was 50 mL to 80 mL per minute in small-volume apheresis procedures and 80 mL to 100 mL in large-volume apheresis procedures. The blood volume processed was 2 to 4 times (average, 3 times) the total blood volume for all patients who underwent large-volume apheresis. The apheresis kit was primed with cross-matched, leukodepleted, and irradiated packed red blood cells (PRBCs) in cases of pediatric donors who weighed less than 15 kg. Adequate PBSC collection was defined as a dose of 2 × 106 CD34+ cells per kg weight of the recipient. PBSC Product Yield: CD34 Enumeration The enumeration of CD34+ cells was performed by flow-cytometric analysis using a single-platform flowcytometer, the BD FACSCalibur (Beckon, Dickinson and Company), using the lyse/no wash (LNW) technique. We added 20 μL of FITC/PE (fluorescein isothicyanate/phycoerythrin) and 20 μL of AAD (amino actinomycin d) dye to the BD Trucount tube (Becton, Dickinson and Company). Then, we added 100 μL of diluted specimen material and allowed the mixture to incubate at room temperature for 20 minutes in the dark. Next, we added 2 mL of lysing ammonium chloride solution to this tube and incubated the resulting mixture for 10 more minutes in the dark. After that, we tested the specimen using a flow cytometer. The list-mode date (LMD) was analyzed using the software BD CellQuest Pro, version 5.2 (Beckon, Dickinson and Company). The CD34 was obtained as total number of CD34+ cells per μL. This value was further analyzed in terms of total cell dose per kg of weight of the recipient, per International Society on Hemotherapy and Graft Engineering (ISHAGE) guidelines. Time To Engraftment We performed routine peripheral blood-cell counts on a daily basis after the day of transplantation. Absolute neutrophil count (ANC) and platelet count were analyzed using the Sysmex KX-21 automated hematology analyzer (Sysmex Corporation) and by peripheral blood smear. Neutrophil engraftment was defined as the first of the 3 consecutive days of ANC greater than 500 per μL, unsupported; platelet engraftment was defined as the first of the 3 consecutive days in which a value was recorded of greater than 20,000 platelets per μL, unsupported. All patients received GVHD prophylaxis with methotrexate injection, cyclosporin injection, and folinic acid (taken orally). Antibiotics were administered per institutional protocol. Chimerism Analysis The chimerism analysis, using Polymerase Chain Reaction (PCR) was alone to determine donor DNA and recipient DNA, and to to express donor DNA as a percentage of recipient DNA. Chimerism of greater than 95% donor DNA in the recipient was interpreted as complete chimerism. We performed chimerism analysis on day 30 and correlated it with the recipient outcome on day 100. Statistical Analysis Statistical analysis was performed using SPSS statistical software, version 23.0 (IBM Corporation). The relationship of donor factors such as age, body weight, preprocedure platelet count, and preprocedure WBC count, with viable CD34+ count, were analyzed using Spearman correlation coefficient. We also analyzed the relationship of dose (per kg body weight of recipient) of CD34+ cells to time of neutrophil and platelet engraftment and percentage chimerism of donor DNA in the recipient using the Spearman correlation coefficient. Statistical significance was set at P <.05. Results A total of 26 peripheral blood stem-cell transplantation (PBSCT) procedures in 25 patients were performed at our center during a 3-year period. We performed 1 apheresis procedure per person, with the exception of 1 donor. A yield of greater than 2 × 106 per kg body weight of recipient was obtained from this donor on day 1 of collection; however, an optimum dose that would be appropriate for greater than the 4 × 106 per kg body weight of the recipient was targeted. The distribution of donor sex, age, and body weight is shown in Figure 1, Figure 2, and Figure 3, respectively. The minimum CD34+ collection yield was 2.1 × 106 per kg body weight of the recipient; the maximum CD34+ collection yield was 19.6 × 106 per kg body weight of the recipient. Per the institutional protocol, a minimum preprocedure peripheral blood WBC count of 20,000 per μL was considered to be the cutoff value for starting the collection procedure. Per past experience at our center, this correlates to a CD34+ count of 10 per μL or greater in autologous donors. Figure 1 Open in new tabDownload slide Distribution of donor sex values. Figure 1 Open in new tabDownload slide Distribution of donor sex values. Figure 2 Open in new tabDownload slide Distribution of donor age values. Figure 2 Open in new tabDownload slide Distribution of donor age values. Figure 3 Open in new tabDownload slide Distribution of donor body weight values. Figure 3 Open in new tabDownload slide Distribution of donor body weight values. Relationship between Donor Factors and CD 34+ Cell Yield The mean CD34+ cell dose for the age of 20 years or greater was 8.95 × 106 per kg body weight of the recipient; for ages 20 years to 40 years, the dose was 9.04 × 106 per kg body weight of recipient; for ages 40 years and older, the dose was 7.93 × 106 per kg body weight of recipient. No statistically significant correlation was present between the CD34+ cell dose and the age of the donor (correlation coefficient, −0.072; P = .73). The mean CD34+ cell counts for donors weighing less than 40 kg was 9.99 × 106 per kg body weight of the recipient; for those weighing 41 kg to 60 kg, the count was 5.78 × 106 per kg body weight of the recipient. For those weighing 61 kg to 80 kg, the count was was 10.38 × 106 per kg body weight of the recipient; for those weighing more than 80 kg, the count was 8.75 × 106 kg body weight of the recipient. Weight had a positive correlation with CD34+ count, which did not attain statistical significance (correlation coefficient of 0.104; P = .28). Figure 4 and Figure 5 show the correlation between the CD34+ cell dose and the preprocedure platelet count and preprocedure WBC count, respectively. The preprocedure platelet and WBC counts had a positive correlation (correlation coefficient, 0.260 and 0.227, respectively) with the postprocedure CD34+ count but also did not approach statistical significance (P = .20 and P = .26, respectively). Figure 4 Open in new tabDownload slide Correlation between the cluster of differentiation (CD)34+ cell dose and the preprocedure platelet count. Figure 4 Open in new tabDownload slide Correlation between the cluster of differentiation (CD)34+ cell dose and the preprocedure platelet count. Figure 5 Open in new tabDownload slide Correlation between the cluster of differentiation (CD)34+ cell dose and the preprocedure white blood cell count. Figure 5 Open in new tabDownload slide Correlation between the cluster of differentiation (CD)34+ cell dose and the preprocedure white blood cell count. Relationship between Postprocedure CD34+ Yield and Time To Neutrophil and Platelet Engraftment The PBSC product was transplanted immediately or after storage at 2°C to 6°C for 24 hours after collection, and was not cryopreserved. Table 1 shows the mean time to neutrophil and platelet engraftment in the subset of recipients who received a CD34+ cell dose of less than 4 × 106 per kg body weight of recipient, 4 to 6 × 106 per kg body weight of recipient, and greater than 6 × 106 per kg body weight of recipient. Table 1. Mean Time to Neutrophil and Platelet Engraftment in Subset of Recipients with Variable CD34+ Count CD34+ Count (106/kg Body Weight of Recipient) Patients, No. Mean Time to Neutrophil Engraftment (d) Mean Time to Platelet Engraftment (d) <4 3 13 11.5 4–6 3 12.7 10.3 >6 19 11.6 11.2 CD34+ Count (106/kg Body Weight of Recipient) Patients, No. Mean Time to Neutrophil Engraftment (d) Mean Time to Platelet Engraftment (d) <4 3 13 11.5 4–6 3 12.7 10.3 >6 19 11.6 11.2 CD, cluster of differentiation. Open in new tab Table 1. Mean Time to Neutrophil and Platelet Engraftment in Subset of Recipients with Variable CD34+ Count CD34+ Count (106/kg Body Weight of Recipient) Patients, No. Mean Time to Neutrophil Engraftment (d) Mean Time to Platelet Engraftment (d) <4 3 13 11.5 4–6 3 12.7 10.3 >6 19 11.6 11.2 CD34+ Count (106/kg Body Weight of Recipient) Patients, No. Mean Time to Neutrophil Engraftment (d) Mean Time to Platelet Engraftment (d) <4 3 13 11.5 4–6 3 12.7 10.3 >6 19 11.6 11.2 CD, cluster of differentiation. Open in new tab The correlations between the CD34+ cell dose infused and the time to neutrophil and platelet engraftment are shown in Figure 6 and Figure 7, respectively. The time to neutrophil engraftment had a positive correlation with the CD34+ dose; however, this value was not statistically significant (correlation coefficient, 0.111; P = .61). The time to platelet engraftment had a positive correlation with CD34+ dose; this value also did not approach statistical significance (correlation coefficient, 0.085; P = .70). Figure 6 Open in new tabDownload slide Correlation between the cluster of differentiation (CD)34+ cell dose infused and the time to neutrophil engraftment. Figure 6 Open in new tabDownload slide Correlation between the cluster of differentiation (CD)34+ cell dose infused and the time to neutrophil engraftment. Figure 7 Open in new tabDownload slide Correlation between the cluster of differentiation (CD)34+ cell dose infused and the time to platelet engraftment. Figure 7 Open in new tabDownload slide Correlation between the cluster of differentiation (CD)34+ cell dose infused and the time to platelet engraftment. Realtionship between CD34+ Yield, Early Donor Chimerism (at Day 30 after Transplantation), And Engratment Kinetics Chimerism analysis was available for 23 patients at day 30 after transplantation because 2 patients had died from transplant-related causes, at day 2 and day 14, respectively. In total, 82.6% of the recipients had a chimerism of greater than 95% at day 30 (n = 19) and 17.4% of the recipients had a chimerism of 95% of greater (n = 4) at that time. No patient had donor chimerism of less than 50% in the recipient at day 30 after transplantation. The percentage of donor chimerism in the recipient had a strong negative correlation with the days to neutrophil engraftment; however, this value was not statistically significant (P = .29). The chimerism analyzed had a negative correlation with the days to platelet engraftment (correlation coefficient, −0.089; P = .69) and neutrophil engraftment (correlation coefficient, −0.231; P = .29) as well as a mild positive correlation with the CD34+ yield of the stem-cell product (correlation coefficient, 0.018; P = .93). Still, no correlation was statistically significant. Chimerism Analysis at Day 30 (Early Donor Chimerism) and Patient Outcome at Day 100 After Transplantation The patients were followed up until day 100 after transplantation for any evidence of acute GVHD and its relationship with the percentage of donor chimerism in the recipient. A total of 82.6% patients had chimerism greater than 95% at day 30. Also, 10.5% of these patients developed grade II–IV actute GVHD (n = 2). Both groups had fatal outcomes. In addition, 17.4% of the recipients had donor chimerism of less than 95% at day 30. No acute GVHD was observed in any of these patients. Discussion In our study, donor factors did not correlate with the CD34+ yield of the PBSC product. The results of earlier studies6,7,10 have reported a correlation between the preprocedure peripheral blood CD34+ count and the CD34+ yield of the PBSC product. However, due to a resource-poor setting and to ensure cost-effectiveness, a preprocedure CD34+ count was not obtained for allogeneic donors. Donor age did not affect the CD34+ yield in our study. Similar results have been reported in other studies, including Miflin et al.11 However, certain other authorship groups12,13 report a decreased yield in donors with increasing age. In a study by Suzuya et al,14 the most important predictor of a good PBSC collection was donor age, with increased counts being observed in younger donors. These variable results can be attributed to the sample-size variation in these studies. Sex and donor body weight did not have an effect on the CD34+ yield in the donors in our study.6,14 Earlier studies have yielded similar findings. Although some studies11,13 have shown a higher CD34+ count in male donors, such studies are few: most articles in the literature shows no effect of sex on CD34+ yield in the setting of an allogeneic PBSC transplant.6,14 Preprocedure platelet and WBC count did not significantly correlate with CD34+ yield in our study. The results of earlier studies14,15 have shown a positive correlation between the preprocedure platelet and leukocyte count and the number of CD34+ cells collected in healthy donors. Tomblyn et al15 reported in their study findings that healthy donors who had a preprocedure WBC count of 25 × 109 per L or greater and a preprocedure platelet count greater than 100 × 109 per L were likely to have a peripheral blood CD34+ value of 20 per μL or greater. The results of this study also showed that if the hematological parameters (preprocedure WBC and platelet count) are favorable in a healthy donor, leukapheresis can be performed without necessarily awaiting the preprocedure peripheral blood CD34+ count of the donor.15 Similar results were reported by Suzuya et al.14 A significant positive correlation was found among preprocedure leukocyte count, preprocedure platelet count, and CD34+ yield.14 In our study results, the small sample size might explain a lack of correlation between the hematological parameters and the CD34+ dose. Variable results are reported in earlier studies,16–18 correlating the donor parameters with the CD34+ yield in the setting of an allogeneic peripheral blood stem-cell transplant. Timely engraftment is an area of concern immediately after transplantation. When the time to hematopoeitic recovery was correlated with the CD34+ dose in the PBSC product, the time to neutrophil and platelet engraftment did not signifcantly correlate with the CD34+ cell dose obtained. The same results have been reported by Perez-Simon et al19: the CD34+ cell dose did not significantly influence the speed of hematopoietic recovery. The authors have commented that although, in autologous transplantation, infusion of a higher cell dose leads to quicker engraftment (as high as between 2.2 and 2.4 × 106/kg body weight of recipient), in the allogeneic setting, increasing the dose above an optimal range does not affect the speed of engraftment.20 It has been observed19 that the CD34+ cell subpopulation is a stronger predictor of the speed of hematopoeitic recovery. In their study, Kamel et al7 found no correlation between the CD34+ dose in the PBSC product and the time to platelet engraftment. They also discovered a fair correlation between the CD34+ dose and the time to neutrophil engraftment. The results of 2 other studies7,21 have shown that the higher the CD34+ cell dose, the quicker the hematopoeitic recovery. In our study, no such correlation was observed which can, again, be attributed to the small sample size of the study. With allogeneic PBSCT, the most important concern was the possibility of an increased risk of GVHD due to a higher number of T cells in the PBSC product (compared with stem cells harvested from the bone marrow).4 The major challenge of posttransplantation monitoring is to prevent such potential negative events. In this context, chimerism monitoring is performed to predict disease relapse, GVHD, or graft rejection after an allogeneic PBSCT.22–24 Many randomized control trials and cohort studies have examined the relative risk of acute and chronic GVHD after PBSCT. In allogeneic PBSCT, an increased trend towards chronic GVHD is well documented.4,19,21 The CD34+ cell dose may affect the percentage of donor chimerism in the recipient because the results of certain studies21,22 have shown that patients who receive a higher dose of CD34+ cells reach complete chimerism more rapidly. In our study, the chimersim analyzed at day 30 did not significantly correlate with the speed of neutrophil and platelet recovery or with the CD34+ dose. The number of individuals who developed acute GVHD was 2 of 25. Both of the affected patients had greater than 95% donor chimerism at day 30. None of the 4 patients who harbored chimerism of 95% or greater at day 30 developed acute GVHD. The results of earlier studies25–27 have shown that lymphoid chimerism (full donor chimerism) at day 30 had a significant relationship with the development of acute GVHD.The same result was also found in a study by Mohty et al,26 in which patients with full donor T-cell chimerism at day 30 had a higher incidence of grade II through IV acute GVHD, compared with patients with mixed T-cell chimerism.25,26 Although our findings are consistent with these study results, a limitation of our study was its small sample size, resulting from the limited number of individuals opting for an allogeneic transplant due to economic constraints. A larger sample size is needed to further validate the findings of this study. Conclusion PBSCs are currently the most commonly used stem-cell source for allogeneic transplantation. These cells can be collected easily after mobilization with hematopoeitic growth factors. Adequate yield in healthy donors happens independent of donor demographic parameters. Thus, a diverse population of donors can be approached for matched unrelated stem cell transplants. Also, healthy donors tend to have an adequate yield with granulocyte colony-stimulating factor (G-CSF) only. None of the donors in our study required plerixafor for mobilization and collection of the PBSC product. Time to engraftment appears to not depend on the CD34+ cell yield in an allogeneic PBSC setting. However, further studies with a bigger sample size might provide conclusive evidence on this subject. In conclusion, full donor chimerism at day 30 can be studied as a useful tool to predict acute GVHD. An accurate quantitative analysis of early donor chimerism in the recipient is an excellent tool for posttransplantation monitoring. Also, early detection of engraftment, disease relapse, graft rejection, and GVHD can help in achieving favorable patient outcomes. Table 1 shows the mean time to neutrophil and platelet engraftment in the subset of recipients who received a CD34+ cell dose of less than 4 × 106 per kg body weight of recipient, 4 to 6 × 106 per kg body weight of recipient, and greater than 6 × 106 per kg body weight of recipient. Abbreviations HSCT hematopoietic stem cell transplantation PBSC peripheral blood stem cells G-CSF granulocyte colony-stimulating factor HPCs hematopoietic progenitor cells CD cluster of differentiation WBC white blood cell HLA human leukocyte antigen AML acute myeloid leukemia ALL acute lymphoblastic leukemia JMML juvenile myelomonocytic leukemia ACD-A anticoagulant citrate dextrose solution A PRBCs packed red blood cells LNW lyse/no wash FITC/PE fluorescein isothicyanate/phycoerythrin AAD amino actinomycin d LMD list-mode date ISHAGE International Society on Hemotherapy and Graft Engineering ANC absolute neutrophil count PCR polymerase chain reaction PBSCT peripheral blood stem-cell transplantation G-CSF granulocyte colony-stimulating factor References 1. Passweg JR , Baldomero H , Bader P , et al. Hematopoietic stem cell transplantation in Europe 2014: more than 40 000 transplants annually . Bone Marrow Transplant. 2016 ; 51 ( 6 ): 786 – 792 . Google Scholar Crossref Search ADS PubMed WorldCat 2. Russell NH , Hunter A , Rogers S , Hanley J , Anderson D . Peripheral blood stem cells as an alternative to marrow for allogeneic transplantation . Lancet. 1993 ; 341 ( 8858 ): 1482 . Google Scholar Crossref Search ADS PubMed WorldCat 3. Bensinger WI . Allogeneic transplantation: peripheral blood vs. bone marrow . Curr Opin Oncol. 2012 ; 24 ( 2 ): 191 – 196 . Google Scholar Crossref Search ADS PubMed WorldCat 4. Forman SJ , Negrin RS , Antin JH , Appelbaum FR , eds. Thomas’ Hematopoietic Cell Transplantation. 5th edn. Hoboken, NJ: John Wiley & Sons, Ltd.; 2016 . Google Preview WorldCat COPAC 5. Martino M , Fedele R , Massara E , Recchia AG , Irrera G , Morabito F . Long-term safety of granulocyte colony-stimulating factor in normal donors: is it all clear? Expert Opin Biol Ther. 2012 ; 12 ( 5 ): 609 – 621 . Google Scholar Crossref Search ADS PubMed WorldCat 6. Lysák D , Koza V , Jindra P . Factors affecting PBSC mobilization and collection in healthy donors . Transfus Apher Sci. 2005 ; 33 ( 3 ): 275 – 283 . Google Scholar Crossref Search ADS PubMed WorldCat 7. Kamel AM , El-Sharkawy N , Mahmoud HK , et al. Impact of CD34 subsets on engraftment kinetics in allogeneic peripheral blood stem cell transplantation . Bone Marrow Transplant. 2005 ; 35 ( 2 ): 129 – 136 . Google Scholar Crossref Search ADS PubMed WorldCat 8. Reshef R , Hexner EO , Loren AW , et al. Early donor chimerism levels predict relapse and survival after allogeneic stem cell transplantation with reduced intensity conditioning . Blood Marrow Transplant . 2014 ; 20 ( 11 ): 1758 – 1766 . Google Scholar Crossref Search ADS WorldCat 9. Sung AD , Chao NJ . Concise review: acute graft-versus-host disease: immunobiology, prevention and treatment . Stem Cells Transl Med. 2013 ; 2 ( 1 ): 25 – 32 . Google Scholar Crossref Search ADS PubMed WorldCat 10. Yu J , Leisenring W , Bensinger WI , Holmberg LA , Rowley SD . The predictive value of white cell or CD34+ cell count in the peripheral blood for timing apheresis and maximizing yield . Transfusion. 1999 ; 39 ( 5 ): 442 – 450 . Google Scholar Crossref Search ADS PubMed WorldCat 11. Miflin G , Charley C , Stainer C , Anderson S , Hunter A , Russell N . Stem cell mobilization in normal donors for allogeneic transplantation: analysis of safety and factors affecting efficacy . Br J Haematol. 1996 ; 95 ( 2 ): 345 – 348 . Google Scholar Crossref Search ADS PubMed WorldCat 12. Shimizu N , Asai T , Hashimoto S , et al. Mobilization factors of peripheral blood stem cells in healthy donors . Ther Apher. 2002 ; 6 ( 6 ): 413 – 418 . Google Scholar Crossref Search ADS PubMed WorldCat 13. Martino M , Callea I , Condemi A , et al. Predictive factors that affect the mobilization of CD34(+) cells in healthy donors treated with recombinant granulocyte colony-stimulating factor (G-CSF) . J Clin Apher. 2006 ; 21 ( 3 ): 169 – 175 . Google Scholar Crossref Search ADS PubMed WorldCat 14. Suzuya H , Watanabe T , Nakagawa R , et al. Factors associated with granulocyte colony-stimulating factor-induced peripheral blood stem cell yield in healthy donors . Vox Sang. 2005 ; 89 ( 4 ): 229 – 235 . Google Scholar Crossref Search ADS PubMed WorldCat 15. Tomblyn M , Gordon LI , Singhal S , et al. Use of total leukocyte and platelet counts to guide stem cell apheresis in healthy allogeneic donors treated with G-CSF . Bone Marrow Transplant. 2005 ; 36 ( 8 ): 663 – 666 . Google Scholar Crossref Search ADS PubMed WorldCat 16. Anderlini P , Przepiorka D , Seong C , et al. Factors affecting mobilization of CD34+ cells in normal donors treated with filgrastim . Transfusion. 1997 ; 37 ( 5 ): 507 – 512 . Google Scholar Crossref Search ADS PubMed WorldCat 17. Grigg AP , Roberts AW , Raunow H , et al. Optimizing dose and scheduling of filgrastim (granulocyte colony-stimulating factor) for mobilization and collection of peripheral blood progenitor cells in normal volunteers . Blood. 1995 ; 86 ( 12 ): 4437 – 4445 . Google Scholar Crossref Search ADS PubMed WorldCat 18. De la Rubia J , Diaz MA , Verdeguer A , et al. Donor-age related differences in PBPC mobilization with rHuG-CSF . Transfusion. 2001 ; 41 ( 2 ): 201 – 205 . Google Scholar Crossref Search ADS PubMed WorldCat 19. Perez-Simon JA , Diez-Campelo M , Martino R , et al. Impact of CD34+ cell dose on the outcome of patients undergoing reduced-intensity-conditioning allogeneic peripheral blood stem cell transplantation . Blood. 2003 ; 102 ( 3 ): 1108 – 1113 . Google Scholar Crossref Search ADS PubMed WorldCat 20. Pérez-Simón JA , Martín A , Caballero D , et al. Clinical significance of CD34+ cell dose in long-term engraftment following autologous peripheral blood stem cell transplantation . Bone Marrow Transplant . 1992 ; 24 ( 12 ): 1279 – 1283 . Google Scholar Crossref Search ADS WorldCat 21. Zaucha JM , Gooley T , Bensinger WI , et al. CD34 cell dose in granulocyte colony-stimulating factor-mobilized peripheral blood mononuclear cell grafts affects engraftment kinetics and development of extensive chronic graft-versus-host disease after human leukocyte antigen-identical sibling transplantation . Blood. 2001 ; 98 ( 12 ): 3221 – 3227 . Google Scholar Crossref Search ADS PubMed WorldCat 22. Urbano-Ispizua A , Carreras E , Marín P , et al. Allogeneic transplantation of CD34+ selected cells from peripheral blood from human leukocyte antigen–identical siblings: detrimental effect of a high number of donor CD34+ cells? Blood . 2001 ; 98 ( 8 ): 2352 – 2357 . Google Scholar Crossref Search ADS PubMed WorldCat 23. Bader P , Niethammer D , Willasch A , Kreyenberg H , Klingebiel T . How and when should we monitor chimerism after allogeneic stem cell transplantation? Bone Marrow Transplant. 2005 ; 35 ( 2 ): 107 – 119 . Google Scholar Crossref Search ADS PubMed WorldCat 24. Childs R , Clave E , Contentin N , et al. Engraftment kinetics after nonmyeloablative allogeneic peripheral blood stem cell transplantation: full donor T-cell chimerism precedes alloimmune responses . Blood. 1999 ; 94 ( 9 ): 3234 – 3241 . Google Scholar Crossref Search ADS PubMed WorldCat 25. Mohty M , Bilger K , Jourdan E , et al. Higher doses of CD34+ peripheral blood stem cells are associated with increased mortality from chronic graft-versus-host disease after allogeneic HLA-identical sibling transplantation . Leukemia 2003 ; 17 : 869 – 875 . Google Scholar Crossref Search ADS PubMed WorldCat 26. Mohty M , Avinens O , Faucher C , Viens P , Blaise D , Eliaou JF . Predictive factors and impact of full donor T-cell chimerism after reduced intensity conditioning allogeneic stem cell transplantation . Haematologica. 2007 ; 92 ( 7 ): 1004 – 1006 . Google Scholar Crossref Search ADS PubMed WorldCat 27. Mohty M , Jacot W , Faucher C , et al. Infectious complications following allogeneic HLA-identical sibling transplantation with antithymocyteglobulin-based reduced intensity preparative regimen . Leukemia. 2003 ; 17 ( 11 ): 2168 – 2177 . Google Scholar Crossref Search ADS PubMed WorldCat © American Society for Clinical Pathology 2019. All rights reserved. For permissions, please e-mail: [email protected] 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)

Bai,, Wenjing;Chen,, Jiangnan;Mao,, Yijian;Wang,, Zhihui;Qian,, Xiaohong;Hu,, Xingzhong;Xu,, Ke;Pan,, Yong

doi: 10.1093/labmed/lmz072pmid: 31746342

Abstract Objectives The application of Light’s criteria misidentifies approximately 30% of transudates as exudates, particularly in patients on diuretics with cardiac effusions. The purpose of this study was to establish a predictive model to effectively identify cardiac effusions misclassified by Light’s criteria. Methods We retrospectively studied 675 consecutive patients with pleural effusion diagnosed by Light’s criteria as exudates, of which 43 were heart failure patients. A multivariate logistic model was developed to predict cardiac effusions. The performance of the predictive model was assessed by receiver operating characteristic (ROC) curves, as well as by examining the calibration. Results It was found that protein gradient of >23 g/L, pleural fluid lactate dehydrogenase (PF-LDH) levels, ratio of pleural fluid LDH to serum LDH level (P/S LDH), pleural fluid adenosine deaminase (PF-ADA) levels, and N-terminal pro–brain natriuretic peptide (NT-pro-BNP) levels had a significant impact on the identification of cardiac effusions, and those were simultaneously analyzed by multivariate regression analysis. The area under the curve (AUC) value of the model was 0.953. The model also had higher discriminatory properties than protein gradients (AUC, 0.760) and NT-pro-BNP (AUC, 0.906), all at a P value of <.01. Conclusion In cases of suspected cardiac effusion, or where clinicians cannot identify the cause of an exudative effusion, this model may assist in the correct identification of exudative effusions as cardiac effusions. pleural effusion, protein gradient, heart failure, transudate, diagnosis, natriuretic peptides The most common cause of pleural effusions is heart failure (37.4%).1 However, the diagnosis of cardiac effusions is not straightforward. The first step in the diagnosis of the cause of pleural effusion is to use Light’s criteria to determine whether it is exudate or transudate.2 Light’s criteria are sensitive to the identification of exudates. Traditionally, if any of the following Light’s criteria2 are met, pleural fluid is considered an exudate: (1) pleural fluid protein-to-serum protein ratio of >0.5; (2) pleural fluid LDH-to-serum LDH ratio of >0.6; or (3) pleural fluid LDH level of >2/3 the upper normal reference serum value. However, their specificity is low.3 As a result, some transudates are incorrectly classified as exudates, particularly those due to heart failure when diuretics are given.4,5 A proposed method for decreasing the misclassification of exudates involves protein gradient, albumin gradient, and pleural fluid lactate dehydrogenase (PF-LDH). In a previous study, the protein gradient and albumin gradient were correctly identified as high in 55% and 83% of the misclassified exudates, respectively, due to congestive heart failure.6 In another study, PF-LDH was correctly identified as 79%.7 However, in general, pleural fluid albumin is not routinely carried out in the laboratory. Brain natriuretic peptide (BNP) is a 108-amino-acid precursor molecule synthesized by myocytes. When the myocytes are stimulated, this precursor (pro-BNP) is enzymatically cleaved into a metabolically inactive N-terminal (NT) protein consisting of 76 amino acids (NT-pro-BNP) and an active 32-amino-acid protein (BNP hormone).8 It has previously been demonstrated that NT-pro-BNP is mainly used in the diagnosis of heart failure.9–12 Therefore, the aim of the present study was to develop a practical model based on the combination of NT-pro-BNP and conventional pleural fluid (PF) tests in order to effectively identify cardiac effusions misclassified by Light’s criteria. Materials and Methods Patients A retrospective cohort study was conducted on consecutive patients who underwent thoracentesis at Wenzhou Central Hospital (Wenzhou, China) from January 2012 to October 2018, and whose pleural effusion met Light’s criteria as exudates. This study was approved by the Clinical Ethics Committee of Wenzhou Central Hospital. Basic patient demographic information (sex, age), serum chemistries (protein, LDH, NT-pro-BNP), and pleural fluid biochemistries (leukocytes, the proportion of lymphocytes and neutrophils, LDH, protein, adenosine deaminase [ADA]) were collected. In cases of repeated thoracentesis, only the first procedure was considered. Diagnostic Criteria The diagnosis of heart failure was undertaken based on medical history, physical examination, chest radiographs, echocardiogram, and response to diuretics. Hepatic hydrothorax was diagnosed based on predefined criteria.13 Malignancy was defined as the presence of malignant cells from pleural effusion cytology or biopsy specimens. Probable malignancy was defined as a pleural effusion associated with a known malignancy, where malignant cells could not be detected in pleural effusion or pleural tissue, and effusion produced by another mechanism could be ruled out. Tuberculous was defined as caseating granulomatous inflammation present in a biopsy specimen or a positive acid-fast bacilli stain or mycobacterial culture of pleural effusion. Parapneumonic was defined as that associated with bacterial pneumonia, lung abscess, or bronchiectasis. Empyema was defined as purulence with pleural effusion. Other causes of pleural effusion were determined by well-established clinical criteria.14 Cases for which data was missing for pleural fluid, or for which there was more than 1 potential cause of the effusion and the final diagnosis was not clear, were excluded. After, parapneumonics, empyemas, hepatic hydrothorax, and patients with hypoalbuminemia were also excluded. Laboratory Measurements Biochemical measurements were analyzed using an AU5400 analyzer (Olympus, Tokyo, Japan). NT-pro-BNP was determined using electrochemiluminescence (ECL) technology (Roche Modular Analytics E170; Roche AG, Basel, Switzerland). A CX 21 Olympus binocular microscope was used for classification of cells. Statistical Analysis Statistical analysis was undertaken using the SPSS 23.0 (IBM, Armonk, NY, USA) and R software programs. R is freely available at https://cran.r-project.org/. Continuous variables were expressed as the mean (standard deviation [SD]) or median (P25, P75) values and were compared using an unpaired Student’s t-test or the Mann-Whitney U test. Categorical variables were presented as counts and percentages and were compared using a chi-square test or Fisher exact test. Candidate variables with a P value of <.2 after performing univariate analysis were included. Then, logistic regression analysis was performed. The PF-LDH, P/S LDH, PF-ADA, and NT-pro-BNP were log10 transformed, as these variables showed skewed distributions. A predictive model was established to assess the probability of cardiac effusions. Then, the performance of the model was evaluated using receiver operating characteristic (ROC) curves, as well as calibration.15 The Hosmer–Lemeshow goodness-of-fit test was used to assess the calibration. A P value of <.05 was considered statistically significant. Results This retrospective cohort study consisted of 675 patients from January 2012 to October 2018 with pleural effusion diagnosed as exudate by Light’s criteria, of which 43 (6.4%) were cardiac effusions and 638 (93.6%) were noncardiac effusions. The basic characteristics and laboratory analysis of patients with cardiac and noncardiac effusions are shown in Table 1. Compared with noncardiac effusion, the proportion of age ≥75 years, serum-PF-protein gradient (protein gradient), and concentration of NT-pro-BNP in cardiac effusions were significantly higher (all P <.001) for cardiac effusions. However, the concentrations of PF-ADA, PF-LDH, and P/S LDH in cardiac effusions were lower. Table 1. Baseline Characteristics of the Study Population Characteristica Cardiac Effusions (n = 43) Noncardiac Effusions (n = 632)b P Value Male 25 (58) 379 (60) .813 Age, y 77.9 ± 13.3 57.4 ± 20.6 <.001  <75 9 (21) 467 (74)  ≥75 34 (79) 165 (26) Pleural fluid  Leukocytes, ×μL 696.0 (277–1328.0) 1917 (965–4083) <.001  Lymphocytes, % 63 (44-75) 65 (49–81) .333  Neutrophils, % 8 (4–13) 10 (3–17) .446  LDH, U/L 44 (110–189) 393 (238–611) .001  Protein, g/L 36.1 ± 9.6 46.7 ± 9.0 <.001  ADA, U/L 6.0 (4.1–8.2) 14.5 (8.9–44.2) <.001 Serum  Protein, g/L 61.2 ± 6.7 63.7 ± 6.8 .02  LDH, U/L 207 (186–274) 204 (175–281) .679  NT-pro-BNP, ng/L 3431 (1224–8886) 204 (62–555)c <.001 Protein gradient, g/L 25.1 ± 11.3 16.9 ± 8.2 <.001 PF/S LDH 0.66 (0.51–0.87) 1.7 (1.1–2.7) <.001 Characteristica Cardiac Effusions (n = 43) Noncardiac Effusions (n = 632)b P Value Male 25 (58) 379 (60) .813 Age, y 77.9 ± 13.3 57.4 ± 20.6 <.001  <75 9 (21) 467 (74)  ≥75 34 (79) 165 (26) Pleural fluid  Leukocytes, ×μL 696.0 (277–1328.0) 1917 (965–4083) <.001  Lymphocytes, % 63 (44-75) 65 (49–81) .333  Neutrophils, % 8 (4–13) 10 (3–17) .446  LDH, U/L 44 (110–189) 393 (238–611) .001  Protein, g/L 36.1 ± 9.6 46.7 ± 9.0 <.001  ADA, U/L 6.0 (4.1–8.2) 14.5 (8.9–44.2) <.001 Serum  Protein, g/L 61.2 ± 6.7 63.7 ± 6.8 .02  LDH, U/L 207 (186–274) 204 (175–281) .679  NT-pro-BNP, ng/L 3431 (1224–8886) 204 (62–555)c <.001 Protein gradient, g/L 25.1 ± 11.3 16.9 ± 8.2 <.001 PF/S LDH 0.66 (0.51–0.87) 1.7 (1.1–2.7) <.001 aData are expressed as mean (SD) or medians (P25, P75) or number (percentage). bThe etiologies of noncardiac effusions: 367 were malignancies (of which 96 were probable diagnoses), 250 were tuberculosis, 9 were surgery, 5 were trauma, and 1 was a pulmonary embolism. cThe proportion of NT-pro-BNP is 331/632 (52%) because of missing observations. LDH, lactate dehydrogenase; ADA, adenosine deaminase; NT-pro-BNP, N-terminal fragment of pro–brain natriuretic peptide; protein gradient, serum-pleural fluid protein gradients; P/S LDH, ratio of pleural fluid LDH to serum LDH level. Open in new tab Table 1. Baseline Characteristics of the Study Population Characteristica Cardiac Effusions (n = 43) Noncardiac Effusions (n = 632)b P Value Male 25 (58) 379 (60) .813 Age, y 77.9 ± 13.3 57.4 ± 20.6 <.001  <75 9 (21) 467 (74)  ≥75 34 (79) 165 (26) Pleural fluid  Leukocytes, ×μL 696.0 (277–1328.0) 1917 (965–4083) <.001  Lymphocytes, % 63 (44-75) 65 (49–81) .333  Neutrophils, % 8 (4–13) 10 (3–17) .446  LDH, U/L 44 (110–189) 393 (238–611) .001  Protein, g/L 36.1 ± 9.6 46.7 ± 9.0 <.001  ADA, U/L 6.0 (4.1–8.2) 14.5 (8.9–44.2) <.001 Serum  Protein, g/L 61.2 ± 6.7 63.7 ± 6.8 .02  LDH, U/L 207 (186–274) 204 (175–281) .679  NT-pro-BNP, ng/L 3431 (1224–8886) 204 (62–555)c <.001 Protein gradient, g/L 25.1 ± 11.3 16.9 ± 8.2 <.001 PF/S LDH 0.66 (0.51–0.87) 1.7 (1.1–2.7) <.001 Characteristica Cardiac Effusions (n = 43) Noncardiac Effusions (n = 632)b P Value Male 25 (58) 379 (60) .813 Age, y 77.9 ± 13.3 57.4 ± 20.6 <.001  <75 9 (21) 467 (74)  ≥75 34 (79) 165 (26) Pleural fluid  Leukocytes, ×μL 696.0 (277–1328.0) 1917 (965–4083) <.001  Lymphocytes, % 63 (44-75) 65 (49–81) .333  Neutrophils, % 8 (4–13) 10 (3–17) .446  LDH, U/L 44 (110–189) 393 (238–611) .001  Protein, g/L 36.1 ± 9.6 46.7 ± 9.0 <.001  ADA, U/L 6.0 (4.1–8.2) 14.5 (8.9–44.2) <.001 Serum  Protein, g/L 61.2 ± 6.7 63.7 ± 6.8 .02  LDH, U/L 207 (186–274) 204 (175–281) .679  NT-pro-BNP, ng/L 3431 (1224–8886) 204 (62–555)c <.001 Protein gradient, g/L 25.1 ± 11.3 16.9 ± 8.2 <.001 PF/S LDH 0.66 (0.51–0.87) 1.7 (1.1–2.7) <.001 aData are expressed as mean (SD) or medians (P25, P75) or number (percentage). bThe etiologies of noncardiac effusions: 367 were malignancies (of which 96 were probable diagnoses), 250 were tuberculosis, 9 were surgery, 5 were trauma, and 1 was a pulmonary embolism. cThe proportion of NT-pro-BNP is 331/632 (52%) because of missing observations. LDH, lactate dehydrogenase; ADA, adenosine deaminase; NT-pro-BNP, N-terminal fragment of pro–brain natriuretic peptide; protein gradient, serum-pleural fluid protein gradients; P/S LDH, ratio of pleural fluid LDH to serum LDH level. Open in new tab The sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and AUC values of each predictor are shown in Table 2. This study showed that an increased probability of cardiac effusions was associated with the presence of age ≥75 years (likelihood ratio [LR] = 3.0), and protein gradient of >31 g/L (LR = 6.1). Conversely, the study showed that a decreased probability of cardiac effusions was associated with age <75 years (LR = 0.28) and protein gradient of ≤23 g/L (LR = 0.38). Table 2. Measures of Diagnostic Accuracy for Parameters that Identity a Cardiac Effusion among Exudates Defined by Light’s Criteria Parameter Sensitivity,% (95% CI) Specificity, % (95% CI) Positive LR (95% CI) Negative LR (95% CI) AUC (95% CI) Age ≥75, y 79.1 (64.0–90.0) 73.9 (70.3–77.3) 3.0 (2.6–3.6) 0.28 (0.2–0.5) 0.77 (0.73–0.80) Protein gradient >23 g/L 69.8 (53.9–82.8) 79.1 (75.7–82.2) 3.3 (2.7–4.1) 0.38 (0.2–0.6) 0.74 (0.71–0.78) Protein gradient >31 g/L 32.5 (19.1–48.5) 94.6 (92.6–96.2) 6.1 (3.9–9.3) 0.71 (0.5–1.0) 0.64 (0.60–0.67) Parameter Sensitivity,% (95% CI) Specificity, % (95% CI) Positive LR (95% CI) Negative LR (95% CI) AUC (95% CI) Age ≥75, y 79.1 (64.0–90.0) 73.9 (70.3–77.3) 3.0 (2.6–3.6) 0.28 (0.2–0.5) 0.77 (0.73–0.80) Protein gradient >23 g/L 69.8 (53.9–82.8) 79.1 (75.7–82.2) 3.3 (2.7–4.1) 0.38 (0.2–0.6) 0.74 (0.71–0.78) Protein gradient >31 g/L 32.5 (19.1–48.5) 94.6 (92.6–96.2) 6.1 (3.9–9.3) 0.71 (0.5–1.0) 0.64 (0.60–0.67) CI = confidence interval; protein gradient, serum-pleural fluid protein gradients. Open in new tab Table 2. Measures of Diagnostic Accuracy for Parameters that Identity a Cardiac Effusion among Exudates Defined by Light’s Criteria Parameter Sensitivity,% (95% CI) Specificity, % (95% CI) Positive LR (95% CI) Negative LR (95% CI) AUC (95% CI) Age ≥75, y 79.1 (64.0–90.0) 73.9 (70.3–77.3) 3.0 (2.6–3.6) 0.28 (0.2–0.5) 0.77 (0.73–0.80) Protein gradient >23 g/L 69.8 (53.9–82.8) 79.1 (75.7–82.2) 3.3 (2.7–4.1) 0.38 (0.2–0.6) 0.74 (0.71–0.78) Protein gradient >31 g/L 32.5 (19.1–48.5) 94.6 (92.6–96.2) 6.1 (3.9–9.3) 0.71 (0.5–1.0) 0.64 (0.60–0.67) Parameter Sensitivity,% (95% CI) Specificity, % (95% CI) Positive LR (95% CI) Negative LR (95% CI) AUC (95% CI) Age ≥75, y 79.1 (64.0–90.0) 73.9 (70.3–77.3) 3.0 (2.6–3.6) 0.28 (0.2–0.5) 0.77 (0.73–0.80) Protein gradient >23 g/L 69.8 (53.9–82.8) 79.1 (75.7–82.2) 3.3 (2.7–4.1) 0.38 (0.2–0.6) 0.74 (0.71–0.78) Protein gradient >31 g/L 32.5 (19.1–48.5) 94.6 (92.6–96.2) 6.1 (3.9–9.3) 0.71 (0.5–1.0) 0.64 (0.60–0.67) CI = confidence interval; protein gradient, serum-pleural fluid protein gradients. Open in new tab The coefficients obtained by the multivariate logistic regression analysis are shown in Table 3. The variables finally selected were as follows: protein gradient of >23 g/L, PF-LDH levels, P/S LDH, PF-ADA levels, and NT-pro-BNP levels. The AUC value of the model was excellent (AUC, 0.953). The positive and negative likelihood ratios were 6.98 and 0.08, respectively. The bootstrap-corrected AUC (0.954) was slightly higher than the apparent AUC. Table 3. Logistic Regression Models for Diagnosing Cardiac Effusionsa Parameter Coefficient (SE) OR (95% CI) P value Protein gradient >23 g/L 1.13 (0.49) 3.09 (1.18–8.11) .022 Log10 (PF-LDH) −4.31 (1.46) 0.013 (0.001–0.236) .003 Log10 (P/S LDH) 1.01 (1.70) 2.75 (0.10–76.3) .551 Log10 (PF-ADA) −1.94 (1.07) 0.14 (0.02–1.17) .07 Log10 (NT-pro-BNP) 2.53 (0.48) 12.57 (4.91–32.18) <.001 Parameter Coefficient (SE) OR (95% CI) P value Protein gradient >23 g/L 1.13 (0.49) 3.09 (1.18–8.11) .022 Log10 (PF-LDH) −4.31 (1.46) 0.013 (0.001–0.236) .003 Log10 (P/S LDH) 1.01 (1.70) 2.75 (0.10–76.3) .551 Log10 (PF-ADA) −1.94 (1.07) 0.14 (0.02–1.17) .07 Log10 (NT-pro-BNP) 2.53 (0.48) 12.57 (4.91–32.18) <.001 aThe model included a protein gradient of >23 g/L, PF-LDH, P/S LDH, PF-ADA, and NT-pro-BNP. Intercept = 1.822; Hosmer-Lemeshow, P = .119, R2 = 0.636, AUC = 0.953. SE, standard error; OR, odds ratio; CI, confidence interval; protein gradient, serum-pleural fluid protein gradients; PF-LDH, pleural fluid lactate dehydrogenase; P/S LDH, ratio of pleural fluid LDH to serum LDH level; PF-ADA, pleural fluid adenosine deaminase; NT-pro-BNP, N-terminal fragment of pro–brain natriuretic peptide. Open in new tab Table 3. Logistic Regression Models for Diagnosing Cardiac Effusionsa Parameter Coefficient (SE) OR (95% CI) P value Protein gradient >23 g/L 1.13 (0.49) 3.09 (1.18–8.11) .022 Log10 (PF-LDH) −4.31 (1.46) 0.013 (0.001–0.236) .003 Log10 (P/S LDH) 1.01 (1.70) 2.75 (0.10–76.3) .551 Log10 (PF-ADA) −1.94 (1.07) 0.14 (0.02–1.17) .07 Log10 (NT-pro-BNP) 2.53 (0.48) 12.57 (4.91–32.18) <.001 Parameter Coefficient (SE) OR (95% CI) P value Protein gradient >23 g/L 1.13 (0.49) 3.09 (1.18–8.11) .022 Log10 (PF-LDH) −4.31 (1.46) 0.013 (0.001–0.236) .003 Log10 (P/S LDH) 1.01 (1.70) 2.75 (0.10–76.3) .551 Log10 (PF-ADA) −1.94 (1.07) 0.14 (0.02–1.17) .07 Log10 (NT-pro-BNP) 2.53 (0.48) 12.57 (4.91–32.18) <.001 aThe model included a protein gradient of >23 g/L, PF-LDH, P/S LDH, PF-ADA, and NT-pro-BNP. Intercept = 1.822; Hosmer-Lemeshow, P = .119, R2 = 0.636, AUC = 0.953. SE, standard error; OR, odds ratio; CI, confidence interval; protein gradient, serum-pleural fluid protein gradients; PF-LDH, pleural fluid lactate dehydrogenase; P/S LDH, ratio of pleural fluid LDH to serum LDH level; PF-ADA, pleural fluid adenosine deaminase; NT-pro-BNP, N-terminal fragment of pro–brain natriuretic peptide. Open in new tab The ROC curves and calibration data are shown in Figure 1 and Figure 2, respectively. A plot along the 45° line corresponds to the model in which the predicted probability is identical with the observed probability, indicating perfect calibration. Figure 1 Open in new tabDownload slide Receiver operating characteristics (ROC) of the model studied. Figure 1 Open in new tabDownload slide Receiver operating characteristics (ROC) of the model studied. Figure 2 Open in new tabDownload slide Calibration of the model studied. Figure 2 Open in new tabDownload slide Calibration of the model studied. Figure 3 illustrates the nomogram corresponding to the model. The weight of each of the variables selected can be observed in these. The probability of a cardiac effusion for a determined score is shown in the lower part of the picture. Figure 3 Open in new tabDownload slide Nomogram corresponding to model. Figure 3 Open in new tabDownload slide Nomogram corresponding to model. The estimated probability (Pr) of classifying an individual as having a cardiac effusion using the model was as follows: Pr (cardiac effusions) = eLP/ (1 + eLP), where e = 2.718282, and, LP(linear predictor)=1.822+1.13∗(Protein gradient>23 g/L)−4.31∗log10(PF−LDH)+1.01∗log10(P/S LDH)−1.94∗log10(PF−ADA)+2.53∗log10(NT−pro−BNP)for the Model. The model correctly classified a higher proportion of patients with cardiac effusions (93.6%) (data not shown). Discussion This study confirmed that the predictive model established with analytical parameters possesses excellent discriminant ability for the identification of cardiac effusions misclassified by Light’s criteria. Indeed, it is sometimes difficult to distinguish a cardiac effusion from a noncardiac effusion, which may necessitate additional tests for patients. Therefore, the proposed model can assist clinicians by providing additional evidence to indicate whether an effusion is cardiac or not. The gold standard for distinguishing transudates from exudates is on the basis of Light’s criteria.2 Light’s criteria correctly classify almost all exudates; however, there are some limitations, such as misclassifying some cardiac effusions as exudates. For these cases, an albumin gradient of >12 g/L or a protein gradient of >31 g/L has been proposed to indicate possible transudates.6,16 However, albumin gradient was not included in this study due to the lack of PF–albumin data; in fact, PF–albumin is not routinely examined in most clinical laboratories. Porcel 17 found that a protein gradient of >31 g/L revealed that 121/196 (62%) of transudates were miscategorized as exudates by the standard criteria. In the current investigation, 43 patients with cardiac effusions were misclassified, and it was found that the best cutoff level of the protein gradient was 23 g/L, a threshold that enabled reclassification of 30 (70%) as cardiac effusions, which was slightly lower than the rate obtained by Porcel et al. (79%).18 In contrast, only 14 (33%) were reclassified when the cutoff was set at 31 g/L. Lactate dehydrogenase (LDH) is a reliable indicator of the degree of pleural inflammation, as it will rise in exudate (e.g., malignancies, tuberculosis, trauma, and pulmonary embolism in our study). However, serum LDH shows no significant difference between cardiac effusions and noncardiac effusions. As a result, PF-LDH and P/S LDH were selected by multivariate regression analysis. Adenosine deaminase (ADA) tends to be higher in tuberculosis effusions than in other exudates, and it was an extensively studied predictor of tuberculosis effusion,19,20 which is why PF-ADA is high in noncardiac effusions. Discrimination and calibration were performed and diagnostic accuracy was determined to quantify the predictions of the model performance. Bootstrap resampling was also performed to correct overfitting, revealing that there was no overfitting in the obtained results. The predictive model for cardiac effusion was established based on the following variables by multivariate logistic regression analysis: protein gradient of >23 g/L, PF-LDH levels, P/S LDH, PF-ADA levels, and NT-pro-BNP levels. The variable with the greatest discriminatory capacity was the NT-pro-BNP (Table 3). This model correctly identified 93.6% of patients, with an AUC value of 0.953. Therefore, the diagnostic yield of this model for the diagnosis of cardiac effusions was superior to that using a protein gradient of >23 g/L (AUC, 0.76) and NT-pro-BNP (AUC, 0.906), all with a P value of <.01. Therefore, it may assist in the reclassification of cardiac effusions as transudates. The present study contains several limitations. First, it was conducted retrospectively, and missing data may cause some biases. Second, patients were collected from a single center in a designated tuberculosis medical hospital. Therefore, a significant percentage of patients with tuberculous effusions may also cause biases. Third, the study had a small sample size; further studies with larger sample sizes need to be conducted to confirm these results. Conclusion In summary, this model resulted in a high diagnostic accuracy for the prediction of cardiac effusions misclassified as exudates by Light’s criteria. LM Abbreviations Abbreviations ROC receiver operating characteristic PF-LDH pleural fluid lactate dehydrogenase P/S LDH ratio of pleural fluid LDH to serum LDH level PF-ADA pleural fluid adenosine deaminase NT-pro-BNP N-terminal pro–brain natriuretic peptide PF pleural fluid ECL electrochemiluminescence SD standard deviation CI confidence interval SE standard error OR odds ratio References 1. Korczyński P , Górska K , Konopka D , Al-Haj D , Filipiak KJ , Krenke R . Significance of congestive heart failure as a cause of pleural effusion: pilot data from a large multidisciplinary teaching hospital . Cardiol J. 2018 . doi: . WorldCat Crossref 2. Light RW , Macgregor MI , Luchsinger PC , Ball WC Jr . Pleural effusions: the diagnostic separation of transudates and exudates . Ann Intern Med. 1972 ; 77 ( 4 ): 507 – 513 . doi: . Google Scholar Crossref Search ADS PubMed WorldCat Crossref 3. Heffner JE . Discriminating between transudates and exudates . Clin Chest Med. 2006 ; 27 ( 2 ): 241 – 252 . doi: . Google Scholar Crossref Search ADS PubMed WorldCat Crossref 4. Porcel JM , Martínez-Alonso M , Cao G , Bielsa S , Sopena A , Esquerda A . Biomarkers of heart failure in pleural fluid . Chest. 2009 ; 136 ( 3 ): 671 – 677 . doi: . Google Scholar Crossref Search ADS PubMed WorldCat Crossref 5. Roth BJ , O’Meara TF , Cragun WH . The serum-effusion albumin gradient in the evaluation of pleural effusions . Chest. 1990 ; 98 ( 3 ): 546 – 549 . doi: . Google Scholar Crossref Search ADS PubMed WorldCat Crossref 6. Bielsa S , Porcel JM , Castellote J , Mas E , Esquerda A , Light RW . Solving the Light’s criteria misclassification rate of cardiac and hepatic transudates . Respirology. 2012 ; 17 ( 4 ): 721 – 726 . doi: . Google Scholar Crossref Search ADS PubMed WorldCat Crossref 7. Kummerfeldt CE , Chiuzan CC , Huggins JT , et al. Improving the predictive accuracy of identifying exudative effusions . Chest. 2014 ; 145 ( 3 ): 586 – 592 . doi: . Google Scholar Crossref Search ADS PubMed WorldCat Crossref 8. Porcel JM . The use of probrain natriuretic peptide in pleural fluid for the diagnosis of pleural effusions resulting from heart failure . Curr Opin Pulm Med. 2005 ; 11 ( 4 ): 329 – 333 . doi: . Google Scholar Crossref Search ADS PubMed WorldCat Crossref 9. Porcel JM , Chorda J , Cao G , Esquerda A , Ruiz-González A , Vives M . Comparing serum and pleural fluid pro-brain natriuretic peptide (NT-proBNP) levels with pleural-to-serum albumin gradient for the identification of cardiac effusions misclassified by Light’s criteria . Respirology. 2007 ; 12 ( 5 ): 654 – 659 . doi: . Google Scholar Crossref Search ADS PubMed WorldCat Crossref 10. Porcel JM , Vives M , Cao G , Esquerda A , Rubio M , Rivas MC . Measurement of pro-brain natriuretic peptide in pleural fluid for the diagnosis of pleural effusions due to heart failure . Am J Med. 2004 ; 116 ( 6 ): 417 – 420 . doi: . Google Scholar Crossref Search ADS PubMed WorldCat Crossref 11. Long AC , O’Neal HR Jr , Peng S , Lane KB , Light RW . Comparison of pleural fluid N-terminal pro-brain natriuretic peptide and brain natriuretic-32 peptide levels . Chest. 2010 ; 137 ( 6 ): 1369 – 1374 . doi: . Google Scholar Crossref Search ADS PubMed WorldCat Crossref 12. Porcel JM , Bielsa S , Morales-Rull JL , et al. Comparison of pleural N-terminal pro-B-type natriuretic peptide, midregion pro-atrial natriuretic peptide and mid-region pro-adrenomedullin for the diagnosis of pleural effusions associated with cardiac failure . Respirology. 2013 ; 18 ( 3 ): 540 – 545 . doi: . Google Scholar Crossref Search ADS PubMed WorldCat Crossref 13. Xiol X , Guardiola J . Hepatic hydrothorax . Curr Opin Pulm Med. 1998 ; 4 ( 4 ): 239 – 242 . doi: . Google Scholar Crossref Search ADS PubMed WorldCat Crossref 14. Garrido VV , Viedma EC , Villar AF , et al. Recommendations of diagnosis and treatment of pleural effusion. Update . Arch Bronconeumol. 2014 ; 50 : 235 – 249 . doi: . Google Scholar PubMed WorldCat Crossref 15. Rao SJ . Regression modeling strategies: with applications to linear models, logistic regression, and survival analysis . Taylor & Francis. 2003 ; 98 : 257 – 258 . doi: . WorldCat Crossref 16. Light RW . The Light criteria: the beginning and why they are useful 40 years later . Clin Chest Med. 2013 ; 34 ( 1 ): 21 – 26 . doi: . Google Scholar Crossref Search ADS PubMed WorldCat Crossref 17. Porcel JM . Identifying transudates misclassified by Light’s criteria . Curr Opin Pulm Med. 2013 ; 19 ( 4 ): 362 – 367 . doi: . Google Scholar Crossref Search ADS PubMed WorldCat Crossref 18. Porcel JM , Ferreiro L , Civit C , et al. Development and validation of a scoring system for the identification of pleural exudates of cardiac origin . Eur J Intern Med. 2018 ; 50 : 60 – 64 . doi: . Google Scholar Crossref Search ADS PubMed WorldCat Crossref 19. Burgess LJ , Maritz FJ , Le Roux I , Taljaard JJ . Combined use of pleural adenosine deaminase with lymphocyte/neutrophil ratio. Increased specificity for the diagnosis of tuberculous pleuritis . Chest. 1996 ; 109 ( 2 ): 414 – 419 . doi: . Google Scholar Crossref Search ADS PubMed WorldCat Crossref 20. Goto M , Noguchi Y , Koyama H , Hira K , Shimbo T , Fukui T . Diagnostic value of adenosine deaminase in tuberculous pleural effusion: a meta-analysis . Ann Clin Biochem. 2003 ; 40 ( Pt 4 ): 374 – 381 . doi: . Google Scholar Crossref Search ADS PubMed WorldCat Crossref © American Society for Clinical Pathology 2019. All rights reserved. For permissions, please e-mail: [email protected] 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)

Fadeyi, Emmanuel, A;Naal,, Tawfeq;Green,, Mary;Simmons, Julie, H;Jones, Mary, Rose;Pomper, Gregory, J

doi: 10.1093/labmed/lmz078pmid: 31756244

Abstract Background Anti-M is most often assumed to be naturally occurring and can be comprised of a mixture of predominantly immunoglobulin(Ig)M with a lesser IgG component. Anti-M-antibodies usually do not react at 37°C and therefore are considered to be of little clinical significance. Methods A 28-year-old man presented with hemorrhagic shock from numerous injuries sustained in a motor vehicle collision. The patient received several units of red blood cells (RBCs). The antibody screen, the direct antiglobulin test (DAT), and the RBC genotype were sent for laboratory evaluation. Results A total of 12 days after the first antibody screening result was negative (7 days after transfusion), the lowest hemoglobin value was 5.5 g per dL, and we observed a positive antibody screening result and DAT with immunoglobulin (Ig)G anti-M identified. After transfusion of 4 units of M antigen–negative RBC, the post-transfusion hemoglobin level increased to 8.9 g per dL. Conclusion Obtaining M antigen–negative compatible RBCs is necessary in patients demonstrating IgG anti-M in plasma. anti-IgG, anti-IgM, anti-M, direct antiglobulin test, delayed hemolytic transfusion reaction, red blood cell transfusions The observation by Landsteiner in 1900 that blood plasma from some individuals will agglutinate red blood cells (RBCs) from others led to the discovery of the ABO blood groups. This typing system is one of the most important factors in making blood transfusion a safe clinical practice. After the identification of the A and B blood group antigens, blood group serology blossomed throughout the 20th century,1 such that the International Society of Blood Transfusion now recognizes 360 blood-group antigens, most of which belong to 1 of 36 genetically discrete blood-group systems.2,3 Antibodies against many of these 360 antigens have the potential to be clinically significant—that is, they can facilitate accelerated destruction of RBCs carrying the corresponding antigen. Delayed hemolytic transfusion reaction (DHTR) usually occurs in patients who have previously been immunized to the offending antigen. Before a conventional DHTR, antibody titers have dropped to a level too low to cause significant intravascular or extravascular hemolysis and often to a level too low to be detected serologically. Consequently, transfusion of RBCs expressing that antigen will initiate a secondary or anamnestic response which, after a period of several days, will initiate clearance of the transfused RBCs. Typically, the HTR occurs approximately 5 to 7 days after the transfusion; however, in extreme cases, the reaction may be as early as 3 days and as late as 14 to 23 days.4,5 Clinical features often involve fever, jaundice, hemoglobinuria, and a drop in hemoglobin level; renal failure is rare.4 Delayed reactions that are only detectable by serological tests (especially a positive direct antiglobulin test [DAT] result) and cause no significant morbidity may be referred to as delayed serological transfusion reactions.5 Antibodies most frequently involved in DHTRs are those of the Rh, Kidd, Duffy, and Kell systems, although other blood-group antibodies are implicated occasionally.4 M and N determinants are carried on glycophorin A (GPA), a major RBC sialic acid–rich glycoprotein. M differs from N in the amino-acid composition of the extracellular tip of GPA. M is a commonly occurring RBC antigen. Approximately 75.4% of blood donors test homozygous or heterozygous for the M antigen. Anti-M is usually assumed to occur naturally and to consist of immunoglobulin (Ig)M, often with an IgG component. Anti-M antibodies usually do not react at 37°C and therefore are considered to be of little clinical significance when performing transfusion testing. Cases of hemolytic transfusion reaction due to anti-M antibodies are extremely rare but have been reported. Herein, we present a case of DHTR due to anti-M antibodies. Case Report and Serological Findings A 28-year-old man with no previous transfusion history presented with hemorrhagic shock from numerous injuries sustained in a pedestrian vs motor vehicle collision. Massive transfusion protocol was initiated; the patient received 10 units of packed RBCs, 6 of which were M-antigen positive. The patient continued to experience bleeding and complications from his injuries. After the results of 2 antibody screening tests were negative, the patient received 1 additional unit of packed RBCs. Three days after arrival at the emergency department, the patient was found to have anemia, requiring additional blood products; as a result, a third antibody screen was performed. A total of 12 days after the initial negative antibody screening result, the patient had anemia, with lowest hemoglobin value of 5.5 g per dL and no reported bleeding from any source. An antibody screening test was requested 12 days after the initial negative result; the latter test yielded a positive result in the indirect antiglobulin phase, with anti-M identified in plasma. The predicted MN phenotype of the patient from the RBC genotype was M-N+. RBC genotyping is performed using PCR 5’-hydrolysis probe assay. (Versiti). The results of a DAT (Standard Tube DAT, Immucor, Inc) were positive with anti-IgG and negative with anti-C3d. IgG DAT using a Gel Micro Typing System card (Thermo Fisher Scientific Inc) was also positive. Anti-M was detected in the eluate using the ELU-KIT (Immucor, Inc), consistent with a DHTR. The only pretransfusion hemolysis-related laboratory value available was total bilirubin, of 0.9 mg per dL (reference range, 0.3–1.2 mg/DL). Post-transfusion hemolysis values were total bilirubin, 4.1 mg per dL; direct bilirubin, 1.0 mg per dL (reference range, 0.1–0.2 mg/dL); lactic acid dehydrogenase (LDH), 493 IU per L (reference range, 90–271 IU/L); haptoglobin, <30 mg per dL (reference range, 30–200 mg/dL); and reticulocyte count, 3.2% (reference range, <1.5%). These laboratory values were consistent with hemolysis from RBC transfusion. The patient continued to have no fever, with all vital signs within normal limits. A follow-up examination 2 days later revealed no clinical evidence of hemolysis. The patient was transfused with 4 units of M antigen–negative packed RBCs. Next-day post-transfusion hemoglobin level increased to 8.9 g per dL and remained stable. Seven days later, the hemodynamic condition of the patient deteriorated, secondary to sequela of recent trauma, sepsis, and multiorgan failure, and the patient eventually died of cardiorespiratory collapse. Discussion Anti-M antibodies in humans are mostly of the IgM class; however, an IgG component can also be present with IgM. Also, albeit rarely, the naturally occurring anti-M can be solely of the IgG class.4 When IgM anti-M antibodies are present but not reactive at 37°C, then the selection of M antigen–negative blood for transfusion is unnecessary. However, M antigen–negative RBCs are necessary for transfusion when the antibody is reactive at 37°C or is of IgG class. Anti-M can show various clinical presentations, as discussed by the 3 cases presented by Das et al.6 The first case was an 18-month-old child admitted for a surgical procedure. There was an ABO grouping discrepancy, and on the results of further work-up, it was revealed that the patient had the naturally occurring IgM type of anti-M antibody in his serum. The ABO grouping discrepancy occurred due to IgM anti-M. The second case was a 20-year-old woman admitted for heart valve replacement surgery. As with the first case individual, this patient also had ABO grouping discrepancy which, on the results of further work-up, demonstrated IgM anti-M in plasma, which caused the ABO discrepancy. The third case individual was a 2-year-old girl with choledochal cysts and chronic liver disease, who was scheduled to undergo a surgical procedure. Her blood group was confirmed to be O RhD-positive, without any obvious grouping discrepancy. However, cross-matching with the O-group packed RBCs (PRBCs) showed incompatibility with both units. Further work-up with antibody panel identification revealed IgG anti-M in plasma. Tondon et al7 also reported 2 cases of anti-M, 1 of which presented as cross-match incompatibility and the other as a blood group discrepancy. Alperin et al8 first reported on a 52-year-old woman who had carried a healthy pregnancy to term—she had M antigen–negative RBCs, experienced a DHTR, and exhibited an anti-M antibody after the infusion of 4 units of M antigen–positive RBCs. Measurements of erythrocyte survival using 51Cr-labeled donor M+ and M− RBCs and in vitro studies of monocyte-macrophage phagocytosis of sensitized reagent RBCs implicated anti-M in the pathogenesis of hemolysis. When IgG anti-M is detected in the indirect antiglobulin phase, which is performed at or near 37°C, this antibody is capable of causing hemolytic transfusion reactions and hemolytic disease of the newborn; therefore, it is clinically significant. A recently published report9 demonstrates anti-M causing hemolytic disease of the fetus and newborn (HDFN). High-titer IgG anti-M was detected in the cord blood of 3 fetuses with titers ranging from 1:1 to 1:128, which confirmed IgG anti-M to be a clinically significant alloantibody. The immunoglobulin class, if anti-M is detected, should require a prewarmed technique to evaluate the potential clinical significance of the antibody. Also, if significant, antigen-negative RBCs should be given to avoid a transfusion reaction. Parry-Jones et al10 reported anti-M–induced DHTR caused by interleukin (IL)-2 in a 33-year-old man with renal cell carcinoma after right radical nephrectomy, which was later treated with combination of IL‐2 and interferons (IFNs). IL‐2 is a growth factor produced by T cells and can activate B cells to produce antibodies. Parry Jones et al postulated that the exogenous cytokine therapy was responsible for stimulating the production of clinically significant IgG antibodies against the M antigen present on transfused RBCs, causing their accelerated destruction. Also, complement activation by anti-M may be a factor in the case, reported by Sancho et al,11 of a 61‐year‐old woman with no history of transfusion who received 4 units of RBC after an operation and 4 more units 7 days after the previous transfusion, without signs of starting to bleed again; her hemoglobin values decreased to 6.5 from 10.8 g per dL with positive hemolytic markers. Hemolysis was not clinically suspected until blood was requested for further transfusion and anti‐M was identified in compatibility testing. The patient had a positive direct antiglobulin test result with IgG and complement C3d. Due to the complicated clinical course of the patient, the authors could not evaluate clinical signs associated with complement‐mediated intravascular hemolysis. However, the presence of C3d on RBCs may support the role of complement activation. The present case demonstrates the importance of IgG anti-M antibodies in pretransfusion testing. Although hemolysis that occurs due to this antibody is rare, DHTR and HDFN do occur. Obtaining M antigen–negative compatible RBCs is necessary in patients demonstrating IgG anti-M in plasma. The M antigen–positive RBC received by the patient contributed to his acute anemia in the setting of ongoing high oxygen demand due to his injuries. Twelve days after the receipt of antigen-positive RBC, his hemoglobin level dropped. He was eventually transfused with antigen-negative, cross-matched, compatible RBC with adequate increment in hemoglobin. The DHTR may have contributed to his acute anemia and eventual multiorgan failure. However, due to his injuries, the patient eventually developed a hemodynamic instability and cardiorespiratory collapse, which led to his death. In this article, our intention is to increase the awareness of pathologists and physicians to this rare and potentially clinically significant antibody, as well as emphasizing the importance of obtaining antigen-negative RBC to patients presenting with IgG anti-M in plasma. This information may help identify those patients receiving blood transfusions who are at increased risk for DHTR. Abbreviations Abbreviations RBCs red blood cells DHTF delayed hemolytic transfusion reaction DAT direct antiglobulin test GPA glycophorin A Ig immunoglobulin LDH lactic acid dehydrogenase PRBCs packed red blood cells HDFN hemolytic disease of the fetus and newborn IL interleukin IFNs interferons Acknowledgments We thank the dedicated clinical laboratory scientists of the Wake Forest Baptist Health Blood Bank Laboratory, Winston-Salem, NC, for their assistance in our technical work. References 1. Poole J , Daniels G . Blood group antibodies and their significance in transfusion medicine . Transfus Med Rev. 2007 ; 21 ( 1 ): 58 – 71 . Google Scholar Crossref Search ADS PubMed WorldCat 2. Storry JR , Clausen FB , Castilho L , et al. International Society of Blood Transfusion working party on red cell immunogenetics and blood group terminology: report of the Dubai, Copenhagen and Toronto meetings . Vox Sang. 2019 ; 114 ( 1 ): 95 – 102 . Google Scholar Crossref Search ADS PubMed WorldCat 3. Fung MK , Eder AF , Spitalnik SL , Westhoff CM , eds. Technical Manual . 19th ed. Bethesda, MD : AABB ; 2017 : 233 – 235 . Google Preview WorldCat COPAC 4. Klein HG , Anstee DJ. Blood Transfusion in Clinical Medicine . 11th ed. Oxford, England : Blackwell Publishing ; 2005 . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC 5. Ness PM , Shirey RS , Thoman SK , Buck SA . The differentiation of delayed serologic and delayed hemolytic transfusion reactions: incidence, long-term serologic findings, and clinical significance . Transfusion. 1990 ; 30 ( 8 ): 688 – 693 . Google Scholar Crossref Search ADS PubMed WorldCat 6. Das R , Dubey A , Agrawal P , Chaudhary RK . Spectrum of anti-M: a report of three unusual cases . Blood Transfus. 2014 ; 12 ( 1 ): 99 – 102 . Google Scholar PubMed WorldCat 7. Tondon R , Kataria R , Chaudhry R . Anti-M: report of two cases and review of literature . Asian J Transfus Sci. 2008 ; 2 ( 2 ): 81 – 83 . Google Scholar Crossref Search ADS PubMed WorldCat 8. Alperin JB , Riglin H , Branch DR , Gallagher MT , Petz LD . Anti-M causing delayed hemolytic transfusion reaction . Transfusion. 1983 ; 23 ( 4 ): 322 – 324 . Google Scholar Crossref Search ADS PubMed WorldCat 9. Li S , Mo C , Huang L , et al. Hemolytic disease of the fetus and newborn due to alloanti‐M: three Chinese case reports and a review of the literature . Transfusion . 2019 ; 59 ( 1 ): 385 – 395 . Google Scholar Crossref Search ADS PubMed WorldCat 10. Parry-Jones N , Gore ME , Taylor J , Treleaven JG . Delayed haemolytic transfusion reaction caused by anti-M antibody in a patient receiving interleukin-2 and interferon for metastatic renal cell cancer . Clin Lab Haematol. 1999 ; 21 ( 6 ): 407 – 408 . Google Scholar Crossref Search ADS PubMed WorldCat 11. Sancho JM , Pujol M , Fernández F , Soler M , Manzano P , Feliu E . Delayed haemolytic transfusion reaction due to anti-M antibody . Br J Haematol. 1998 ; 103 ( 1 ): 268 – 269 . Google Scholar Crossref Search ADS PubMed WorldCat © American Society for Clinical Pathology 2019. All rights reserved. For permissions, please e-mail: [email protected] 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)