Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Pet Rodents and Fatal Lymphocytic Choriomeningitis in Transplant Patients

Pet Rodents and Fatal Lymphocytic Choriomeningitis in Transplant Patients Pet Rodents and Fatal Lymphocytic Choriomeningitis in Transplant Patients 1 1,2 Brian R. Amman,* Boris I. Pavlin,* Cesar G. Albariño,* James A. Comer,* Bobbie R. Erickson,* Jennifer B. Oliver,* Tara K. Sealy,* Martin J. Vincent,* Stuart T. Nichol,* Christopher D. Paddock,* Abbigail J. Tumpey,* Kent D. Wagoner,* R. David Glauer,† Kathleen A. Smith,‡ Kim A. Winpisinger,‡ Melody S. Parsely,§ Phil Wyrick,¶ Christopher H. Hannafi n,# Utpala Bandy,** Sherif Zaki,* Pierre E. Rollin,* and Thomas G. Ksiazek* In April 2005, 4 transplant recipients became ill after tion is likely greatly underreported as a cause of poor preg- receiving organs infected with lymphocytic choriomenin- nancy outcomes (7). Human infection occurs most com- gitis virus (LCMV); 3 subsequently died. All organs came monly through exposure (by direct contact or inhalation of from a donor who had been exposed to a hamster infected infectious aerosol) to secretions or excretions of infected with LCMV. The hamster was traced back through a Rhode animals (8). To our knowledge, person-to-person transmis- Island pet store to a distribution center in Ohio, and more sion has not been reported, except for transmission from LCMV-infected hamsters were discovered in both. Rodents mother to fetus (7) and 1 previous cluster in December from the Ohio facility and its parent facility in Arkansas 2003 of infection through organ transplantation (9,10). were tested for the same LCMV strain as the 1 involved In early April 2005, 4 recipients of solid-organ trans- in the transplant-associated deaths. Phylogenetic analysis plants in 3 hospitals in Rhode Island and Massachusetts of virus sequences linked the rodents from the Ohio facility became gravely ill shortly after transplantation; 3 subse- to the Rhode Island pet store, the index hamster, and the transplant recipients. This report details the animal trace- quently died (10). All 4 recipients shared a common donor. back and the supporting laboratory investigations. Tissue and blood samples from the donor and recipients were sent from the Rhode Island Department of Health and the Massachusetts Department of Public Health to the ymphocytic choriomeningitis virus (LCMV) is a ro- Centers for Disease Control and Prevention (CDC), where Ldentborne arenavirus endemic in house mice (Mus LCMV was identifi ed as the etiologic agent (10). Viral se- musculus) worldwide (1–3). Lymphocytic choriomeningi- quences from the organ recipients were identical to those tis (LCM) in immunocompetent persons usually is a mild, from a pet hamster acquired by the donor’s household 17 self-limited, viral syndrome or is asymptomatic; aseptic days before organ donation (10). Here we report the results meningitis also can occur, but the infection is rarely fatal of an epidemiologic and environmental investigation to (4–6). In immunocompromised persons, LCM may result identify the origin of the index hamster and the source of in serious systemic infections and death. LCM during preg- the virus. nancy can cause spontaneous abortion or severe birth de- fects, including hydrocephalus, chorioretinitis, blindness, Methods or psychomotor retardation (7). Congenital LCMV infec- Epidemiologic Investigation *Centers for Disease Control and Prevention, Atlanta, Georgia, Thorough epidemiologic investigations were con- USA; †Ohio Department of Agriculture, Reynoldsburg, Ohio, USA; ducted at the Rhode Island pet store where the index ‡Ohio Department of Health, Columbus, Ohio, USA; §Arkansas These authors contributed equally to this article. Department of Health and Human Services, Little Rock, Arkansas, USA; ¶Arkansas Department of Agriculture, Little Rock, Arkansas, Current affi liation: Johns Hopkins Bloomberg School of Public USA; #Rhode Island Department of Environmental Management, Health, Baltimore, Maryland, USA Providence, Rhode Island, USA; and **Rhode Island Department of Public Health, Providence, Rhode Island, USA Current affi liation: Ithaca College, Ithaca, New York, USA Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 13, No. 5, May 2007 719 RESEARCH hamster was purchased, the Ohio distribution facility that says that included serology, immunohistochemistry (IHC), supplied the pet store, and the primary breeding facility in reverse transcription–PCR (RT-PCR), TaqMan (Applied Arkansas. These investigations focused on interviews; re- Biosystems, Foster City, CA, USA), and virus isolation. view of invoices, shipping records, and US Department of Genetic sequences obtained from the respective samples Agriculture inspection reports; and on-site environmental were used in the phylogenetic analysis to identify the assessments. LCMV strain and epidemiologic link leading to transplant- associated deaths. Rodent Sample Collection All available rodent species known to be competent Virus Isolation hosts for LCMV (capable of becoming chronically infected Virus isolation was conducted by using Vero E-6 cells. and shedding virus for up to 9 months) (6,11,12) were col- For blood or serum, 100 μL of sample was used as inocu- lected from the remaining rodent stock at the Rhode Island lum. For tissues, a 10% cell suspension was prepared in a pet store. These species included Syrian hamsters (Mesocri- viral support medium (Hank’s balanced salt solution with cetus auratus), “fancy” mice (M. musculus), and guinea 5% heat-inactivated fetal bovine serum) and clarifi ed by pigs (Cavia porcellus). Although they have not been shown centrifugation. A100-μL aliquot of the supernatant fl uid to be competent reservoirs for LCMV, “fancy” rats (Rattus was used as the inoculum. Flasks were incubated for 1 norvegicus) and gerbils (Meriones unguiculatus) also were hour, fed with maintenance medium, and observed for 2 sampled because of their exposure to the infected rodents. weeks. Cells from fl asks were tested for replicating LCMV Rodents were sampled and euthanized following approved by immunofl uorescent antibody assay (IFA) on 1 of days CDC Animal Care and Use Committee protocols. 4–7 (depending on supplemental information made avail- With a known population size and a LCMV prevalence able through other testing) and again on day 14. If no reac- estimate, the hypergeometric probability distribution was tivity was detected by IFA from days 4 to 7 or on day 14, used to determine the minimum sample size needed to pro- the fl ask was considered negative for virus. vide a 95% chance of detecting at least 1 LCMV infected rodent at each site. The LCMV prevalence was estimated Molecular Detection of LCMV in Rodents to be 4.7% in Ohio and 4.3% in Arkansas. The Ohio preva- Highly sensitive real-time RT-PCR TaqMan assay lence was based on 4 infected of 85 tested at the Rhode was performed as described previously (10). RNA isolated Island pet store; the revised prevalence for Arkansas was from rodent blood, serum specimens, or tissue was sub- based on 9 of 211 positives after data from the Ohio sam- jected to TaqMan real-time assay, and samples with cycle ples were incorporated. threshold (Ct) values <40 were scored as LCMV-positive. The population sizes (Table 1) included only dwarf TaqMan-positive specimens were further analyzed by tra- hamsters and did not distinguish between the Chinese ditional RT-PCR to produce a 232-nt product within the and Roborovsky dwarf hamsters (Cricetulus curtatus and RNA polymerase (L) gene and sequences were obtained by Phodopus roborovskii, respectively). An agreement was using previously described primers (10). The sequences of reached with the owner in which ≈10% of the total popu- LCMV from the transplant recipients, index hamster, and lation of 140 Roborovsky’s dwarf hamsters was sampled. rodents from the Rhode Island pet store and Ohio distri- In this case, the probability of detecting at least 1 positive bution center were then compared with those obtained for rodent was 36.5%. other characterized LCMV strains by using GCG Version 11.1.1 (Accelrys, San Diego, CA, USA) and PAUP (Sinauer Laboratory Investigation Associates Inc., Sunderland, MA, USA). Further evidence The index hamster, the organ recipients, the animals of a genetic link between LCMV detected in the rodents from the pet store, and rodents from the Ohio and Arkansas and the human cases investigated was obtained by analyz- facilities were tested for LCMV with a combination of as- ing the viral S RNA segment. A 611-nt S segment PCR Table 1. Estimated population sizes and samples taken from 2 rodent distribution facilities Projected Probability of detecting Location Species Population size prevalence, % Sample size a positive, % Ohio Syrian hamsters 5,000 5.0 116 99.8 Arkansas Fancy rats >10,000 3.4 125 98.7 Arkansas Fancy mice 200 3.4 75 96.5 Arkansas Gerbils 2,500 3.4 125 98.8 Arkansas Dwarf hamsters 3,750 3.4 113* 98.1 Arkansas Roborovsky dwarf hamster 140 3.4 12† 36.5 *Sample size after removal of 12 Roborovsky hamsters from requested sample size. †Sample size represents an agreed-upon portion of the total population. 720 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 13, No. 5, May 2007 Rodents and Fatal Lymphocytic Choriomeningitis product was amplifi ed by using 1-step RT-PCR protocols larly due to a lack of employee hand hygiene between han- with a generic primer set (13) capable of amplifying Old dling of individual rodents. As a precautionary measure, World arenaviruses including LCMV. The 1-step RT-PCR all rodents were quarantined at the store and further sales was carried out by using the SuperScript III One-Step RT- were prohibited by the Rhode Island Department of Health PCR System with Platinum Taq High Fidelity as described after LCMV was identifi ed in the organ recipients and the by the manufacturer (Invitrogen, Carlsbad, CA, USA). index hamster. A total of 85 animals (55 hamsters, 8 guinea pigs, 10 Serologic and Immunohistochemical mice, 7 gerbils, and 5 rats) were sampled from the remain- Detection of LCMV in Rodents ing quarantined rodent stock at the Rhode Island pet store. An ELISA was used to evaluate serum samples col- Of these, 1 guinea pig and 2 hamsters were found posi- lected from rodents for immunoglobulin class G (IgG) tive for LCMV by several methods (Table 2). LCMV anti- antibodies that reacted with LCMV antigens produced in bodies were detected in 1 hamster by IFA, but not ELISA. Vero E-6 cells. The assay was run as described in Fischer LCMV isolates were obtained from either blood or kidney et al. (10), except that a protein A/protein G conjugate (Im- and immunohistochemical stains were positive in at least 1 munopure, Pierce Biotechnology Inc., Rockford, IL, USA) organ in each of the 3 rodents. All 3 rodents were positive was used. A subset of the samples was also tested by IFA, for LCMV RNA with the L-gene–specifi c TaqMan primer/ using infected Vero E-6 cells. Immunohistochemical tests probe set. The L-gene sequences obtained from these ro- were carried out on a variety of tissues from the index ham- dents were identical to one another and differed from the ster (blood, adrenal gland, salivary gland, pancreas, liver, index hamster and transplant recipients by only 1 nt (Fig- spleen, kidney, lung, heart, bone marrow, cerebrum, cer- ure, panel A). Further evidence confi rming the presence ebellum, brain stem, spinal cord) as previously described of viruses of the same genetic lineage in this episode was by Fischer et al. (10). gathered by RT-PCR amplifi cation of a product from the S segment. The 611-nt S segment sequences of the index Results hamster and the transplant recipients were 100% identical, thereby reconfi rming the previously established genetic link (10). In addition, the S segment sequences obtained Rhode Island Traceback Investigation from the 2 Rhode Island pet store hamsters were identical, Physical inspection of the pet store where the donor’s and they differed by only 2 nt from the guinea pig sequence hamster was purchased produced no evidence of wild rodent (Figure, panel B). These results indicate the same LCMV infestation. The store maintained live-capture traps in areas virus strain was present in the hamsters and guinea pig in likely to harbor rodents (e.g., near feed bags); the trapping the Rhode Island pet store. log showed no captures in the preceding 3 months. Although other rodents had been housed in the same Ohio Traceback Investigation area of the pet store with the index hamster, detailed re- The Ohio facility served as a distribution/staging area cords were not available and these specifi c rodents were not for rodents destined for sale in the northeastern United identifi ed. Invoices dated from February through March States. Records indicated that it received most of its ham- 2005 confi rmed that all rodents sold at the pet store had sters from its parent breeding facility in Arkansas. Both fa- come from the Ohio facility. cilities, owned by the person, routinely received shipments Biosecurity in the store was limited, with opportunity of rodents from smaller breeders. The Arkansas facility for interspecifi c and intraspecifi c cross-infection, particu- Table 2. Results of laboratory testing on the index hamster and traceback rodents associated with organ transplantation transmission of LCMV* Virus Sequence RT-PCR/ Rodent† IFA ELISA IHC TaqMan isolation (L gene), bp Index hamster ND Neg Pos Pos Pos 232 Pet store hamster 1 Pos Neg Pos Pos Pos 232 Pet store hamster 2 Neg Neg Pos Pos Pos 232 Pet store guinea pig 1 ND Neg Pos Pos Pos 232 Ohio hamster 1 Pos Neg Pos Pos Pos 232 Ohio Hamster 2 Pos Neg Pos Pos Pos NA Ohio hamster 3 Pos Neg ND Neg Neg NA Ohio hamster 4 Neg Neg Pos Pos Pos NA Ohio hamster 5 Neg Neg Neg ND Pos NA *LCMV, lymphocytic choriomeningitis virus; IFA, immunofluorescent antibody assay; IHC, immunohistochemistry; RT-PCR, reverse transcription–PCR; ND, no data; Neg, negative; Pos, positive; NA, no amplification (could not get traditional PCR primers to amplify for sequencing). †The table includes only those rodents that tested positive with 1 test. Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 13, No. 5, May 2007 721 RESEARCH records indicated ongoing AWA violations and poor bio- security practices: escaped mice and wild Mus spp. that entered the facility were routinely captured and added to captive breeding populations; sick rodents shared common airspace with healthy rodents, attending veterinary services were sporadic, rodents were shipped to Ohio without nec- essary permits and veterinary inspections, and rodents from different sources shared cages for breeding purposes with- out adequate quarantine practices. With the assistance of the Ohio Departments of Agri- culture and Health, the Ohio facility was quarantined and inspected on July 18, 2005. During this inspection, several examples of poor biosecurity were found: escaped rodents ran free and entered other holding bins; evidence of a wild rodent infestation was found among the feed sacks; and rodents from disparate sources were housed in adjacent racks, with opportunity for large and small particle cross- contamination. Shipping records were inadequately maintained, mak- ing it diffi cult to accurately account for the individual ship- ments of rodents. However, hamsters arriving at the Ohio facility from either Arkansas or other outside breeders were placed in tubs labeled with the location of origin and the date of arrival. According to employees at the facility, these hamsters were not mixed with other shipments. Figure. Lymphocytic choriomeningitis virus (LCMV) phylogenetic A sample of 126 rodents (116 hamsters, 9 mice, and analysis of L- and S-segment sequence differences. A) Maximum 1 guinea pig) was collected. This total comprised the sta- likelihood analysis of a 232-nt fragment of the L segment was tistically necessary 75 (taken from the general population) completed, and bootstrap numbers were generated based on analysis of 500 replicates. The graphic representation was plus escaped, sick, and dead rodents. Of the specimens ex- outgrouped to the California (CA) LCMV sequence. GenBank nos. amined, 5 hamsters were positive for LCMV by at least for the included sequence are as follows: Rhode Island (RI) and Ohio 1 method (Table 2): 3 were positive by IFA, but not by (OH) transplant recipients strain 200501927 (DQ182703), Rhode ELISA. Three were positive by IHC; these same 3 were Island pet store and Ohio distributor rodents strain 200504261 LCMV RNA–positive as evidenced by L-segment–specifi c (DQ888889), New York (NY) strain WE (AF004519), Wisconsin (WI) transplant recipients strain 810362 (DQ182706), Missouri (MO) TaqMan assay. Virus was isolated from kidney tissues of strain Armstrong (J04331), and the CA congenital infection strain these 3 hamsters, and sequences were obtained from 1 vi- 810366 (DQ182707). B) Maximum likelihood analysis of a 611-nt rus isolate. L-segment sequence comparison of the Ohio fragment of the S segment NP gene was completed as mentioned specimen found exact identity to the Rhode Island pet store above. The GenBank nos. are as follows: RI and OH transplant hamster virus (Figure, panel A). Additionally, sequences recipients strain 200501927 (DQ888890), RI pet store guinea pig strain 200502048 (DQ888891),OH distributor hamster strain obtained from the 611-nt S segment of the Ohio hamster 200504261 (DQ888893),RI pet store hamsters strain 200501966 differed by only 3 nt from the index hamster sequence, thus (DQ888892), NY strain WE (M22138), WI transplant recipient strain showing 99.5% identity (Figure, panel B). These sequences 810362 (DQ182704), MO strain Armstrong (NC_004294), and the were compared to sequences from other previously identi- CA congenital infection strain 810366 (DQ182705). fi ed LCMV strains such as the laboratory strains LCMV- Armstrong and WE, and other isolates from clinical mate- rial (Table 3). Differences in L-segment sequences in the was the largest distributor in North America, with sales to viruses from the index hamster and pet store/distribution many states. center rodents were <0.5%, while LCMV-Armstrong and Breeding operations at the Ohio facility had been sus- WE differed by 18.1% and 13.4%, respectively. Compari- pended in February 2005 by order of the United States De- son of the S-segment sequences between the index hamster partment of Agriculture (USDA) for multiple violations and pet store/distribution center viruses showed <1% dif- of the Animal Welfare Act (AWA) (14); the suspension ference. LCMV-Armstrong and WE differed by 14.1% and included all rodent species other than mice and rats. In- 14.7%, respectively, from the index hamster virus. terviews and a review of prior USDA facility inspection 722 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 13, No. 5, May 2007 Rodents and Fatal Lymphocytic Choriomeningitis Table 3. Comparison of nucleotide identity differences among LCMV strains and isolates* State/sample 1 2 3 4 5 6 7 8 9 10 11 12 13 Rhode Island Lung recipient 1 0.0 0.0 0.0 0.0 0.4 0.4 0.4 0.4 13.4 15.8 18.1 22.9 Liver recipient 2 0.0 0.0 0.0 0.0 0.4 0.4 0.4 0.4 13.4 15.8 18.1 22.9 Kidney recipient A 3 0.0 0.0 0.0 0.0 0.4 0.4 0.4 0.4 13.4 15.8 18.1 22.9 Kidney recipient B 4 0.0 0.0 0.0 0.0 0.4 0.4 0.4 0.4 13.4 15.8 18.1 22.9 Donor's hamster 5 0.0 0.0 0.0 0.0 0.4 0.4 0.4 0.4 13.4 15.8 18.1 22.9 Pet store hamster 1 6 0.3 0.3 0.3 0.3 0.3 0.0 0.0 0.0 12.9 15.3 18.5 23.4 Pet store hamster 2 7 0.3 0.3 0.3 0.3 0.3 0.0 0.0 0.0 12.9 15.3 18.5 23.4 Pet store guinea pig 8 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.0 12.9 15.3 18.5 23.4 Ohio Distributor hamster 9 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.2 12.9 15.3 18.5 23.4 New York WE 10 12.4 12.4 12.4 12.4 12.4 12.1 12.1 12.3 12.3 16.2 18.5 24.1 Wisconsin Kidney recipients 11 14.4 14.4 14.4 14.4 14.4 14.4 14.4 14.2 14.4 14.6 19.5 22.5 Missouri Armstrong 12 14.1 14.1 14.1 14.1 14.1 13.7 13.7 13.9 14.1 15.2 13.7 21.7 California Congenital infection 13 14.7 14.7 14.7 14.7 14.7 14.4 14.4 14.6 14.7 13.3 14.6 14.4 *LCMV, lymphocytic choriomeningitis virus. The percentage of differences for each pair of sequences was calculated with PAUP as uncorrected distances. Values on the upper diagonal represent differences in the L fragment (232 nt) and values on the lower diagonal represent differences on the S fragment (611 nt). Arkansas Traceback Investigation supports the hypothesis that the index hamster’s infection Several attempts were made to sample the rodents in came from the rodent distribution center in Ohio, rather the Arkansas facility. The initial sample was to include only than from wild M. musculus populations around the home Syrian hamsters and guinea pigs, but these species were de- of the donor or pet store. stroyed by the owner when LCMV was found in the Ohio Sequence and phylogenetic data provided strong sup- facility. Other remaining rodent species were then sampled port for the presence of the same LCMV lineage in ham- as a proxy measure. With the assistance of the Arkansas sters and guinea pigs in the Rhode Island pet store and the Department of Health, 450 rodents were sampled, includ- Ohio distribution center and established the epidemiologic ing 125 fancy rats, 125 gerbils, 75 fancy mice, 113 Chinese link of that particular lineage of LCMV to transplant-as- dwarf hamsters, and 12 Roborovski dwarf hamsters. One sociated deaths. Comparison of LCMV genotypes obtained fancy rat was IgG positive for LCMV by ELISA. All other from this investigation with previously identifi ed strains test results, including virus isolation, were negative. LCMV-Armstrong and WE and other isolates from clini- cal material found considerable differences (Figure, Table Discussion 3). While the differences in S-segment sequences between This report documents the animal traceback investiga- the index hamster and pet store/distribution center viruses tion that linked a major pet rodent distribution operation to were <1% (<3-nt difference), LCMV-Armstrong and WE the recent outbreak of lymphocytic choriomeningitis in 4 differed from index hamster sequences by 14.1% (76-nt organ transplant recipients in Rhode Island and Massachu- difference) and 14.7% (86-nt difference), respectively. setts. This investigation demonstrates the ways in which Similarly, the differences in L-segment sequences in the classic epidemiology, laboratory diagnostics, and molecu- viruses from the index hamster and pet store/distribution lar biology can complement one another in the investiga- center rodents were <0.5% (3-nt difference), while LCMV- tion of disease clusters. LCMV was not found in the organ Armstrong and WE differed by 18.1% and 13.4% (41 nt donor’s tissues; however, the viral isolate from the pet store and 30 nt), respectively. These analyses indicate an overall hamster was sequenced and matched to the sequences of close identity among LCMV strains implicated in this in- the isolates from the recipients (10). The near-complete se- vestigation, and wide differences from previously published quence match between the virus found in the index hamster sequences of strains Armstrong and WE. This molecular and the virus sequenced from the Ohio hamster indicates evidence corroborates the epidemiologic data implicating that the genotypes share a common lineage that is dis- LCMV transmission within the commercial pet trade. tinct from previously identifi ed strains. It is unlikely that After the identifi cation of LCMV in the Ohio facility, this genotype would be as similar to a genotype found in all states that had received animals from this facility in the wild house mice in Rhode Island. Thus, the animal trace- previous 5 months were notifi ed. Many potentially infected back, coupled with the molecular phylogenetic evidence, animals still remained in stores; their disposition was deter- Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 13, No. 5, May 2007 723 RESEARCH mined by individual states. Actions taken ranged from sale with infected wild mice, and can pass the infection to hu- or adoption with an information leafl et or informed consent mans. Most human LCMV infections are associated with to issuing stop sale orders on specifi c rodent species from exposure to wild house mice (6,15); however, several out- the Ohio facility. breaks have been attributed to laboratory and pet mice and Upon notifi cation of the Ohio sample test results, the hamsters (5,6,16,17). One example is the 1974 outbreak Ohio Department of Agriculture informed the proprietor of associated with pet hamsters sold by a single distributor. A the LCMV contamination in his facility and requested that total of 181 symptomatic cases (46 requiring hospitaliza- a written plan for decontaminating the facility and a sec- tion) in persons with hamster contact were identifi ed in 12 ond plan for keeping the facility LCMV-free in the future states; no deaths occurred (5). The outbreak was brought be provided. The proprietor responded by depopulating the under control by voluntary cessation of sale and destruction Ohio facility. The quarantine was lifted after the building of the infected breeding stock. was disinfected, but the facility was never reopened. A direct link between the Arkansas and Ohio facili- Prevention ties was established by the discovery of the marked tubs LCMV surveillance should be a primary concern in the in Ohio. This enabled the Arkansas Departments of Health pet rodent industry to avoid entry of this virus into pet trade and Agriculture (Livestock and Poultry) to issue a Joint populations. Because of the ubiquitous distribution of the Quarantine and Inspection Order. However, several days house mouse, eliminating the natural reservoir of LCMV after depopulating the Ohio facility, the proprietor also de- is not practical. Steps can be taken, however, to exclude stroyed all the Syrian hamsters and guinea pigs at the Ar- wild house mice from homes and businesses. Immunocom- kansas facility. Although efforts were made to sample the promised persons and pregnant women should be advised remaining rodents at the Arkansas facility in an attempt to to avoid close contact with all rodents and infested areas. pinpoint the source of the virus found in Ohio, the owner Educational materials should address the risk from expo- refused to allow access to the rodents on several occasions sure to wild mice as well as pet rodents. The virus is not and >4 months elapsed between the initial and the ultimate- naturally present in pet rodent species and the ease of trans- ly successful attempts to conduct sampling. Virus isolation mission of the virus from pet rodents to humans may be on all samples was unsuccessful. greater than from wild mice when one considers the nature Several factors may have contributed to the lack of vi- of the relationship between pet rodents and their owners rus in the Arkansas breeding facility, including the follow- (i.e., close physical contact). Therefore, every effort should ing: 1) the virus was never there, 2) the destruction of the be made to eliminate the virus from pet populations when Syrian hamsters and guinea pigs eliminated the virus from it is discovered. the facility, 3) the elapsed time allowed for the removal LCMV is already actively excluded from laboratory of infected animals and subsequent decontamination of the rodent populations, because the infection can be an occu- facility, and 4) a complete replacement of the rodent stock pational hazard to laboratory workers who work around was accomplished within the facility. Although the facility infected rodents (5), and because inapparent infection can was under quarantine for 3 of the 4 months between the fi rst interfere with experimental results in rodent studies (18). and last attempts at sampling, only sporadic surveillance of Economic considerations may prohibit such rigorous bi- the facility was carried out by a governmental authority. osecurity measures like those used for laboratory animals; Elimination of the virus from the population did not likely however, sentinel surveillance (19), adequate veterinary occur naturally because LCMV can chronically infect mice care, exclusion of wild rodents (20), and good infection and will lead to persistent colonial transmission (6,11). control practices can substantially reduce the opportu- Since the time of our investigation, the proprietor’s nity for introduction and spread of LCMV and other pet license has been suspended by the USDA for 5 years for rodent pathogens in commercial pet populations. Efforts violations of the AWA unrelated to this investigation. The to increase such practices within the pet trade are under Rhode Island pet retailer who sold the index hamster report- way. Adherence to regulations that are already in place for edly ceased business relations with the distributor shortly obtaining permits and veterinary inspection of commercial after the infection was linked to his facilities. rodent populations can also reduce the likelihood of infec- To our knowledge, the Rhode Island outbreak repre- tion and improve animal welfare. sents the fi rst documented case of fatal LCMV infection Further efforts to reduce risk for LCM in pet owners are involving a pet animal (10). In the previous cluster of trans- ongoing. Education is critical in preventing LCM and other plant-related LCMV deaths, no rodent exposure was identi- pet-related infections. Potential pet owners should choose fi ed (9). Several rodent species that are sold as pets, includ- pets appropriate to their household (21). Pregnant women ing hamsters, mice, and guinea pigs, can be incidental hosts and immunocompromised persons should avoid pet rodents of LCMV. These species become infected through contact altogether (22). Additionally, pet owners should be advised 724 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 13, No. 5, May 2007 Rodents and Fatal Lymphocytic Choriomeningitis 9. Paddock C, Ksiazek T, Comer JA, Rollin P, Nichol S, Shieh WJ, et of the possibility of acquiring zoonotic diseases from any al. Pathology of fatal lymphocytic choriomeningitis virus infection pet and of precautions that should be taken to prevent ac- in multiple organ transplant recipients from a common donor. Mod- quiring pet-related infections such as LCM, tularemia (23), ern Pathology. 2005;18(Suppl 1):263A–4A. salmonellosis (24), and others. Persons can reduce risk for 10. Fischer SA, Graham MB, Kuehnert MJ, Kotton CN, Srinivasan A, Marty FM, et al. Transmission of lymphocytic choriomeningitis vi- infection from pet rodents by being attentive to proper hand rus via organ transplantation. N Engl J Med. 2006;354:2235–49. hygiene and environmental cleaning. Additional informa- 11. Traub E. The epidemiology of lymphocytic choriomeningitis in tion on LCMV is available from the CDC website (25). white mice. J Exp Med. 1936;64:183–200. 12. Lehmann-Grube F. Lymphocytic choriomeningitis virus. New York: Springer-Verlag; 1971. Acknowledgments 13. Bowen MD, Rollin P, Ksiazek TG, Hustad HL, Bausch DG, Dem- We wish to acknowledge the following persons and agen- byet AH, et al. Genetic diversity among Lassa virus strains. J Virol. 2000;74:6992–7004. cies for their contributions to this investigation: James Mills, 14. Animal Welfare Act 1990 (7 USC, 2131–2156). Beltsville (MD): Herminia Alva, Archer Miller, Thomas Stevens, Michael Bell, US Department of Agriculture Animal Welfare Information Center. Joel Montgomery, Mitesh Patel, Jeannette Guarner, Kimberly [cited 2006 June 1]. Available from http://www.nal.usda.gov/awic/ Slaughter, Deborah Cannon, Rita Helfand, Joe Foster, Heather legislat/awa.htm 15. Childs JE, Peters CJ. Ecology and epidemiology of arenaviruses and Horton; Frank Wilson, Altino McKelvey, Randy Owens, Sherry their hosts. In: Salvo MS, editor. The Arenaviridae. New York: Ple- Langley, Joe Bates, Rick Hogan, Reggie Rogers, Gail Kusturin, num Press; 1993. p. 331–84. and other staff in the Arkansas Department of Health and Hu- 16. Baum SG, Lewis AM, Rowe WP, Huebner RJ. Epidemic nonmen- man Services; Jeff Hayes, David Frew, and other staff in the Ohio ingitic lymphocytic-choriomeningitis-virus infection: an outbreak in a population of laboratory personnel. N Engl J Med. 1966;274: Departments of Agriculture and Health; Bernadette Juarez, Clara 934–6. Markin, Randy Coleman, Carl LaLonde, Betty Goldentyer, and 17. Ackermann R, Stille W, Blumenthal K. Syrische hamster als unber- additional staff in the US Department of Agriculture Animal and trager von lymphozytärer choriomeningitis. Dtsch Med Wochenschr. Plant Inspection Service. 1972;97:1725–31. 18. Nicklas W, Homberger FR, Illgen-Wilcke B, Jacobi K, Kraft V, Dr Amman is a mammalogist for the Special Pathogens Kunstyr I, et al. Implications of infectious agents on results of ani- mal experiments. Report of the Working Group on Hygiene of the Branch at CDC in Atlanta, Georgia. His research interests include Gesellschaft fur Versuchstierkunde–Society for Laboratory Animal investigating the ecology of, and relationships between, emerging Science (GV-SOLAS). Lab Anim. 1999;33(Suppl 1):S39–87. viruses and their mammalian reservoirs. 19. Lipman NS, Homberger FR. Rodent quality assurance testing: use of sentinel animal systems. Lab Animal. 2003;32:36–43. 20. Centers for Disease Control and Prevention. Rodent control: seal up! References Trap up! Clean up! [cited 2006 June 1]. Available from http://www. cdc.gov/rodents 1. Childs JE, Glass GE, Ksiazek TG, Rossi CA, Oro JG, LeDuc JW. 21. Centers for Disease Control and Prevention. Healthy pets, healthy Human-rodent contact and infection with lymphocytic choriomen- people. [cited 2006 June 1]. Available from http://www.cdc.gov/ ingitis and Seoul viruses in an inner-city population. Am J Trop Med healthypets Hyg. 1991;44:117–21. 22. Centers for Disease Control and Prevention. Update: interim guid- 2. Childs JE, Glass GE, Korch GW, Ksiazek TG, LeDuc JW. Lym- ance for minimizing risk for human lymphocytic choriomeningitis phocytic choriomeningitis virus infection and house mouse (Mus virus infection associated with pet rodents. MMWR Morb Mortal musculus) distribution in urban Baltimore. Am J Trop Med Hyg. Wkly Rep. 2005;54:799–801. 1992;47:27–34. 23. Centers for Disease Control and Prevention. Brief report: tularemia 3. Park JY, Peters CJ, Rollin PE, Ksiazek TG, Katholi CR, Waites KB, associated with a hamster bite—Colorado, 2004. MMWR Morb et al. Age distribution of lymphocytic choriomeningitis virus serum Mortal Wkly Rep. 2005;53:1202–3. antibody in Birmingham, Alabama: evidence of a decreased risk of 24. Centers for Disease Control and Prevention. Outbreak of multidrug- infection. Am J Trop Med Hyg. 1997;57:37–41. resistant Salmonella Typhimurium associated with rodents pur- 4. Lewis AM Jr, Rowe WP, Turner HC, Huebner RJ. Lymphocytic- chased at retail pet stores—United States, December 2003–October choriomeningitis virus in hamster tumor: spread to hamsters and 2004. MMWR Morb Mortal Wkly Rep. 2005;54:429–33. humans. Science. 1965;150:363–4. 25. Centers for Disease Control and Prevention. Lymphocytic chorio- 5. Gregg MB. Recent outbreaks of lymphocytic choriomeningitis in the meningitis (LCMV). [cited 2006 June 1]. Available from http:// United States of America. Bull World Health Organ. 1975;52:549– www.cdc.gov/ncidod/dvrd/spb/mnpages/dispages/lcmv.htm 6. Rousseau MC, Saron MF, Brouqui P, Bourgeade A. Lymphocytic Address for correspondence: Brian R. Amman, Centers for Disease choriomeningitis virus in southern France: four case reports and a review of the literature. Eur J Epidemiol. 1997;13:817–23. Control and Prevention, Mailstop A26, Atlanta, GA 30333, USA; email: 7. Barton LL, Mets MB, Beauchamp CL. Lymphocytic choriomen- [email protected] ingitis virus: emerging fetal teratogen. Am J Obstet Gynecol. 2002;187:1715–6. Use of trade names is for identifi cation only and does not imply 8. Richmond JY, McKinney RW, editors. Biosafety in microbiological endorsement by the Public Health Service or by the U.S. and biomedical laboratories. 4th ed. Washington: US Government Department of Health and Human Services. Printing Offi ce; 1999. Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 13, No. 5, May 2007 725 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Emerging Infectious Diseases Pubmed Central

Pet Rodents and Fatal Lymphocytic Choriomeningitis in Transplant Patients

Download PDF
7 pages

Loading next page...
 
/lp/pubmed-central/pet-rodents-and-fatal-lymphocytic-choriomeningitis-in-transplant-zlj00Lo6VQ

References (27)

Publisher
Pubmed Central
ISSN
1080-6040
eISSN
1080-6059
DOI
10.3201/eid1305.061269
Publisher site
See Article on Publisher Site

Abstract

Pet Rodents and Fatal Lymphocytic Choriomeningitis in Transplant Patients 1 1,2 Brian R. Amman,* Boris I. Pavlin,* Cesar G. Albariño,* James A. Comer,* Bobbie R. Erickson,* Jennifer B. Oliver,* Tara K. Sealy,* Martin J. Vincent,* Stuart T. Nichol,* Christopher D. Paddock,* Abbigail J. Tumpey,* Kent D. Wagoner,* R. David Glauer,† Kathleen A. Smith,‡ Kim A. Winpisinger,‡ Melody S. Parsely,§ Phil Wyrick,¶ Christopher H. Hannafi n,# Utpala Bandy,** Sherif Zaki,* Pierre E. Rollin,* and Thomas G. Ksiazek* In April 2005, 4 transplant recipients became ill after tion is likely greatly underreported as a cause of poor preg- receiving organs infected with lymphocytic choriomenin- nancy outcomes (7). Human infection occurs most com- gitis virus (LCMV); 3 subsequently died. All organs came monly through exposure (by direct contact or inhalation of from a donor who had been exposed to a hamster infected infectious aerosol) to secretions or excretions of infected with LCMV. The hamster was traced back through a Rhode animals (8). To our knowledge, person-to-person transmis- Island pet store to a distribution center in Ohio, and more sion has not been reported, except for transmission from LCMV-infected hamsters were discovered in both. Rodents mother to fetus (7) and 1 previous cluster in December from the Ohio facility and its parent facility in Arkansas 2003 of infection through organ transplantation (9,10). were tested for the same LCMV strain as the 1 involved In early April 2005, 4 recipients of solid-organ trans- in the transplant-associated deaths. Phylogenetic analysis plants in 3 hospitals in Rhode Island and Massachusetts of virus sequences linked the rodents from the Ohio facility became gravely ill shortly after transplantation; 3 subse- to the Rhode Island pet store, the index hamster, and the transplant recipients. This report details the animal trace- quently died (10). All 4 recipients shared a common donor. back and the supporting laboratory investigations. Tissue and blood samples from the donor and recipients were sent from the Rhode Island Department of Health and the Massachusetts Department of Public Health to the ymphocytic choriomeningitis virus (LCMV) is a ro- Centers for Disease Control and Prevention (CDC), where Ldentborne arenavirus endemic in house mice (Mus LCMV was identifi ed as the etiologic agent (10). Viral se- musculus) worldwide (1–3). Lymphocytic choriomeningi- quences from the organ recipients were identical to those tis (LCM) in immunocompetent persons usually is a mild, from a pet hamster acquired by the donor’s household 17 self-limited, viral syndrome or is asymptomatic; aseptic days before organ donation (10). Here we report the results meningitis also can occur, but the infection is rarely fatal of an epidemiologic and environmental investigation to (4–6). In immunocompromised persons, LCM may result identify the origin of the index hamster and the source of in serious systemic infections and death. LCM during preg- the virus. nancy can cause spontaneous abortion or severe birth de- fects, including hydrocephalus, chorioretinitis, blindness, Methods or psychomotor retardation (7). Congenital LCMV infec- Epidemiologic Investigation *Centers for Disease Control and Prevention, Atlanta, Georgia, Thorough epidemiologic investigations were con- USA; †Ohio Department of Agriculture, Reynoldsburg, Ohio, USA; ducted at the Rhode Island pet store where the index ‡Ohio Department of Health, Columbus, Ohio, USA; §Arkansas These authors contributed equally to this article. Department of Health and Human Services, Little Rock, Arkansas, USA; ¶Arkansas Department of Agriculture, Little Rock, Arkansas, Current affi liation: Johns Hopkins Bloomberg School of Public USA; #Rhode Island Department of Environmental Management, Health, Baltimore, Maryland, USA Providence, Rhode Island, USA; and **Rhode Island Department of Public Health, Providence, Rhode Island, USA Current affi liation: Ithaca College, Ithaca, New York, USA Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 13, No. 5, May 2007 719 RESEARCH hamster was purchased, the Ohio distribution facility that says that included serology, immunohistochemistry (IHC), supplied the pet store, and the primary breeding facility in reverse transcription–PCR (RT-PCR), TaqMan (Applied Arkansas. These investigations focused on interviews; re- Biosystems, Foster City, CA, USA), and virus isolation. view of invoices, shipping records, and US Department of Genetic sequences obtained from the respective samples Agriculture inspection reports; and on-site environmental were used in the phylogenetic analysis to identify the assessments. LCMV strain and epidemiologic link leading to transplant- associated deaths. Rodent Sample Collection All available rodent species known to be competent Virus Isolation hosts for LCMV (capable of becoming chronically infected Virus isolation was conducted by using Vero E-6 cells. and shedding virus for up to 9 months) (6,11,12) were col- For blood or serum, 100 μL of sample was used as inocu- lected from the remaining rodent stock at the Rhode Island lum. For tissues, a 10% cell suspension was prepared in a pet store. These species included Syrian hamsters (Mesocri- viral support medium (Hank’s balanced salt solution with cetus auratus), “fancy” mice (M. musculus), and guinea 5% heat-inactivated fetal bovine serum) and clarifi ed by pigs (Cavia porcellus). Although they have not been shown centrifugation. A100-μL aliquot of the supernatant fl uid to be competent reservoirs for LCMV, “fancy” rats (Rattus was used as the inoculum. Flasks were incubated for 1 norvegicus) and gerbils (Meriones unguiculatus) also were hour, fed with maintenance medium, and observed for 2 sampled because of their exposure to the infected rodents. weeks. Cells from fl asks were tested for replicating LCMV Rodents were sampled and euthanized following approved by immunofl uorescent antibody assay (IFA) on 1 of days CDC Animal Care and Use Committee protocols. 4–7 (depending on supplemental information made avail- With a known population size and a LCMV prevalence able through other testing) and again on day 14. If no reac- estimate, the hypergeometric probability distribution was tivity was detected by IFA from days 4 to 7 or on day 14, used to determine the minimum sample size needed to pro- the fl ask was considered negative for virus. vide a 95% chance of detecting at least 1 LCMV infected rodent at each site. The LCMV prevalence was estimated Molecular Detection of LCMV in Rodents to be 4.7% in Ohio and 4.3% in Arkansas. The Ohio preva- Highly sensitive real-time RT-PCR TaqMan assay lence was based on 4 infected of 85 tested at the Rhode was performed as described previously (10). RNA isolated Island pet store; the revised prevalence for Arkansas was from rodent blood, serum specimens, or tissue was sub- based on 9 of 211 positives after data from the Ohio sam- jected to TaqMan real-time assay, and samples with cycle ples were incorporated. threshold (Ct) values <40 were scored as LCMV-positive. The population sizes (Table 1) included only dwarf TaqMan-positive specimens were further analyzed by tra- hamsters and did not distinguish between the Chinese ditional RT-PCR to produce a 232-nt product within the and Roborovsky dwarf hamsters (Cricetulus curtatus and RNA polymerase (L) gene and sequences were obtained by Phodopus roborovskii, respectively). An agreement was using previously described primers (10). The sequences of reached with the owner in which ≈10% of the total popu- LCMV from the transplant recipients, index hamster, and lation of 140 Roborovsky’s dwarf hamsters was sampled. rodents from the Rhode Island pet store and Ohio distri- In this case, the probability of detecting at least 1 positive bution center were then compared with those obtained for rodent was 36.5%. other characterized LCMV strains by using GCG Version 11.1.1 (Accelrys, San Diego, CA, USA) and PAUP (Sinauer Laboratory Investigation Associates Inc., Sunderland, MA, USA). Further evidence The index hamster, the organ recipients, the animals of a genetic link between LCMV detected in the rodents from the pet store, and rodents from the Ohio and Arkansas and the human cases investigated was obtained by analyz- facilities were tested for LCMV with a combination of as- ing the viral S RNA segment. A 611-nt S segment PCR Table 1. Estimated population sizes and samples taken from 2 rodent distribution facilities Projected Probability of detecting Location Species Population size prevalence, % Sample size a positive, % Ohio Syrian hamsters 5,000 5.0 116 99.8 Arkansas Fancy rats >10,000 3.4 125 98.7 Arkansas Fancy mice 200 3.4 75 96.5 Arkansas Gerbils 2,500 3.4 125 98.8 Arkansas Dwarf hamsters 3,750 3.4 113* 98.1 Arkansas Roborovsky dwarf hamster 140 3.4 12† 36.5 *Sample size after removal of 12 Roborovsky hamsters from requested sample size. †Sample size represents an agreed-upon portion of the total population. 720 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 13, No. 5, May 2007 Rodents and Fatal Lymphocytic Choriomeningitis product was amplifi ed by using 1-step RT-PCR protocols larly due to a lack of employee hand hygiene between han- with a generic primer set (13) capable of amplifying Old dling of individual rodents. As a precautionary measure, World arenaviruses including LCMV. The 1-step RT-PCR all rodents were quarantined at the store and further sales was carried out by using the SuperScript III One-Step RT- were prohibited by the Rhode Island Department of Health PCR System with Platinum Taq High Fidelity as described after LCMV was identifi ed in the organ recipients and the by the manufacturer (Invitrogen, Carlsbad, CA, USA). index hamster. A total of 85 animals (55 hamsters, 8 guinea pigs, 10 Serologic and Immunohistochemical mice, 7 gerbils, and 5 rats) were sampled from the remain- Detection of LCMV in Rodents ing quarantined rodent stock at the Rhode Island pet store. An ELISA was used to evaluate serum samples col- Of these, 1 guinea pig and 2 hamsters were found posi- lected from rodents for immunoglobulin class G (IgG) tive for LCMV by several methods (Table 2). LCMV anti- antibodies that reacted with LCMV antigens produced in bodies were detected in 1 hamster by IFA, but not ELISA. Vero E-6 cells. The assay was run as described in Fischer LCMV isolates were obtained from either blood or kidney et al. (10), except that a protein A/protein G conjugate (Im- and immunohistochemical stains were positive in at least 1 munopure, Pierce Biotechnology Inc., Rockford, IL, USA) organ in each of the 3 rodents. All 3 rodents were positive was used. A subset of the samples was also tested by IFA, for LCMV RNA with the L-gene–specifi c TaqMan primer/ using infected Vero E-6 cells. Immunohistochemical tests probe set. The L-gene sequences obtained from these ro- were carried out on a variety of tissues from the index ham- dents were identical to one another and differed from the ster (blood, adrenal gland, salivary gland, pancreas, liver, index hamster and transplant recipients by only 1 nt (Fig- spleen, kidney, lung, heart, bone marrow, cerebrum, cer- ure, panel A). Further evidence confi rming the presence ebellum, brain stem, spinal cord) as previously described of viruses of the same genetic lineage in this episode was by Fischer et al. (10). gathered by RT-PCR amplifi cation of a product from the S segment. The 611-nt S segment sequences of the index Results hamster and the transplant recipients were 100% identical, thereby reconfi rming the previously established genetic link (10). In addition, the S segment sequences obtained Rhode Island Traceback Investigation from the 2 Rhode Island pet store hamsters were identical, Physical inspection of the pet store where the donor’s and they differed by only 2 nt from the guinea pig sequence hamster was purchased produced no evidence of wild rodent (Figure, panel B). These results indicate the same LCMV infestation. The store maintained live-capture traps in areas virus strain was present in the hamsters and guinea pig in likely to harbor rodents (e.g., near feed bags); the trapping the Rhode Island pet store. log showed no captures in the preceding 3 months. Although other rodents had been housed in the same Ohio Traceback Investigation area of the pet store with the index hamster, detailed re- The Ohio facility served as a distribution/staging area cords were not available and these specifi c rodents were not for rodents destined for sale in the northeastern United identifi ed. Invoices dated from February through March States. Records indicated that it received most of its ham- 2005 confi rmed that all rodents sold at the pet store had sters from its parent breeding facility in Arkansas. Both fa- come from the Ohio facility. cilities, owned by the person, routinely received shipments Biosecurity in the store was limited, with opportunity of rodents from smaller breeders. The Arkansas facility for interspecifi c and intraspecifi c cross-infection, particu- Table 2. Results of laboratory testing on the index hamster and traceback rodents associated with organ transplantation transmission of LCMV* Virus Sequence RT-PCR/ Rodent† IFA ELISA IHC TaqMan isolation (L gene), bp Index hamster ND Neg Pos Pos Pos 232 Pet store hamster 1 Pos Neg Pos Pos Pos 232 Pet store hamster 2 Neg Neg Pos Pos Pos 232 Pet store guinea pig 1 ND Neg Pos Pos Pos 232 Ohio hamster 1 Pos Neg Pos Pos Pos 232 Ohio Hamster 2 Pos Neg Pos Pos Pos NA Ohio hamster 3 Pos Neg ND Neg Neg NA Ohio hamster 4 Neg Neg Pos Pos Pos NA Ohio hamster 5 Neg Neg Neg ND Pos NA *LCMV, lymphocytic choriomeningitis virus; IFA, immunofluorescent antibody assay; IHC, immunohistochemistry; RT-PCR, reverse transcription–PCR; ND, no data; Neg, negative; Pos, positive; NA, no amplification (could not get traditional PCR primers to amplify for sequencing). †The table includes only those rodents that tested positive with 1 test. Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 13, No. 5, May 2007 721 RESEARCH records indicated ongoing AWA violations and poor bio- security practices: escaped mice and wild Mus spp. that entered the facility were routinely captured and added to captive breeding populations; sick rodents shared common airspace with healthy rodents, attending veterinary services were sporadic, rodents were shipped to Ohio without nec- essary permits and veterinary inspections, and rodents from different sources shared cages for breeding purposes with- out adequate quarantine practices. With the assistance of the Ohio Departments of Agri- culture and Health, the Ohio facility was quarantined and inspected on July 18, 2005. During this inspection, several examples of poor biosecurity were found: escaped rodents ran free and entered other holding bins; evidence of a wild rodent infestation was found among the feed sacks; and rodents from disparate sources were housed in adjacent racks, with opportunity for large and small particle cross- contamination. Shipping records were inadequately maintained, mak- ing it diffi cult to accurately account for the individual ship- ments of rodents. However, hamsters arriving at the Ohio facility from either Arkansas or other outside breeders were placed in tubs labeled with the location of origin and the date of arrival. According to employees at the facility, these hamsters were not mixed with other shipments. Figure. Lymphocytic choriomeningitis virus (LCMV) phylogenetic A sample of 126 rodents (116 hamsters, 9 mice, and analysis of L- and S-segment sequence differences. A) Maximum 1 guinea pig) was collected. This total comprised the sta- likelihood analysis of a 232-nt fragment of the L segment was tistically necessary 75 (taken from the general population) completed, and bootstrap numbers were generated based on analysis of 500 replicates. The graphic representation was plus escaped, sick, and dead rodents. Of the specimens ex- outgrouped to the California (CA) LCMV sequence. GenBank nos. amined, 5 hamsters were positive for LCMV by at least for the included sequence are as follows: Rhode Island (RI) and Ohio 1 method (Table 2): 3 were positive by IFA, but not by (OH) transplant recipients strain 200501927 (DQ182703), Rhode ELISA. Three were positive by IHC; these same 3 were Island pet store and Ohio distributor rodents strain 200504261 LCMV RNA–positive as evidenced by L-segment–specifi c (DQ888889), New York (NY) strain WE (AF004519), Wisconsin (WI) transplant recipients strain 810362 (DQ182706), Missouri (MO) TaqMan assay. Virus was isolated from kidney tissues of strain Armstrong (J04331), and the CA congenital infection strain these 3 hamsters, and sequences were obtained from 1 vi- 810366 (DQ182707). B) Maximum likelihood analysis of a 611-nt rus isolate. L-segment sequence comparison of the Ohio fragment of the S segment NP gene was completed as mentioned specimen found exact identity to the Rhode Island pet store above. The GenBank nos. are as follows: RI and OH transplant hamster virus (Figure, panel A). Additionally, sequences recipients strain 200501927 (DQ888890), RI pet store guinea pig strain 200502048 (DQ888891),OH distributor hamster strain obtained from the 611-nt S segment of the Ohio hamster 200504261 (DQ888893),RI pet store hamsters strain 200501966 differed by only 3 nt from the index hamster sequence, thus (DQ888892), NY strain WE (M22138), WI transplant recipient strain showing 99.5% identity (Figure, panel B). These sequences 810362 (DQ182704), MO strain Armstrong (NC_004294), and the were compared to sequences from other previously identi- CA congenital infection strain 810366 (DQ182705). fi ed LCMV strains such as the laboratory strains LCMV- Armstrong and WE, and other isolates from clinical mate- rial (Table 3). Differences in L-segment sequences in the was the largest distributor in North America, with sales to viruses from the index hamster and pet store/distribution many states. center rodents were <0.5%, while LCMV-Armstrong and Breeding operations at the Ohio facility had been sus- WE differed by 18.1% and 13.4%, respectively. Compari- pended in February 2005 by order of the United States De- son of the S-segment sequences between the index hamster partment of Agriculture (USDA) for multiple violations and pet store/distribution center viruses showed <1% dif- of the Animal Welfare Act (AWA) (14); the suspension ference. LCMV-Armstrong and WE differed by 14.1% and included all rodent species other than mice and rats. In- 14.7%, respectively, from the index hamster virus. terviews and a review of prior USDA facility inspection 722 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 13, No. 5, May 2007 Rodents and Fatal Lymphocytic Choriomeningitis Table 3. Comparison of nucleotide identity differences among LCMV strains and isolates* State/sample 1 2 3 4 5 6 7 8 9 10 11 12 13 Rhode Island Lung recipient 1 0.0 0.0 0.0 0.0 0.4 0.4 0.4 0.4 13.4 15.8 18.1 22.9 Liver recipient 2 0.0 0.0 0.0 0.0 0.4 0.4 0.4 0.4 13.4 15.8 18.1 22.9 Kidney recipient A 3 0.0 0.0 0.0 0.0 0.4 0.4 0.4 0.4 13.4 15.8 18.1 22.9 Kidney recipient B 4 0.0 0.0 0.0 0.0 0.4 0.4 0.4 0.4 13.4 15.8 18.1 22.9 Donor's hamster 5 0.0 0.0 0.0 0.0 0.4 0.4 0.4 0.4 13.4 15.8 18.1 22.9 Pet store hamster 1 6 0.3 0.3 0.3 0.3 0.3 0.0 0.0 0.0 12.9 15.3 18.5 23.4 Pet store hamster 2 7 0.3 0.3 0.3 0.3 0.3 0.0 0.0 0.0 12.9 15.3 18.5 23.4 Pet store guinea pig 8 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.0 12.9 15.3 18.5 23.4 Ohio Distributor hamster 9 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.2 12.9 15.3 18.5 23.4 New York WE 10 12.4 12.4 12.4 12.4 12.4 12.1 12.1 12.3 12.3 16.2 18.5 24.1 Wisconsin Kidney recipients 11 14.4 14.4 14.4 14.4 14.4 14.4 14.4 14.2 14.4 14.6 19.5 22.5 Missouri Armstrong 12 14.1 14.1 14.1 14.1 14.1 13.7 13.7 13.9 14.1 15.2 13.7 21.7 California Congenital infection 13 14.7 14.7 14.7 14.7 14.7 14.4 14.4 14.6 14.7 13.3 14.6 14.4 *LCMV, lymphocytic choriomeningitis virus. The percentage of differences for each pair of sequences was calculated with PAUP as uncorrected distances. Values on the upper diagonal represent differences in the L fragment (232 nt) and values on the lower diagonal represent differences on the S fragment (611 nt). Arkansas Traceback Investigation supports the hypothesis that the index hamster’s infection Several attempts were made to sample the rodents in came from the rodent distribution center in Ohio, rather the Arkansas facility. The initial sample was to include only than from wild M. musculus populations around the home Syrian hamsters and guinea pigs, but these species were de- of the donor or pet store. stroyed by the owner when LCMV was found in the Ohio Sequence and phylogenetic data provided strong sup- facility. Other remaining rodent species were then sampled port for the presence of the same LCMV lineage in ham- as a proxy measure. With the assistance of the Arkansas sters and guinea pigs in the Rhode Island pet store and the Department of Health, 450 rodents were sampled, includ- Ohio distribution center and established the epidemiologic ing 125 fancy rats, 125 gerbils, 75 fancy mice, 113 Chinese link of that particular lineage of LCMV to transplant-as- dwarf hamsters, and 12 Roborovski dwarf hamsters. One sociated deaths. Comparison of LCMV genotypes obtained fancy rat was IgG positive for LCMV by ELISA. All other from this investigation with previously identifi ed strains test results, including virus isolation, were negative. LCMV-Armstrong and WE and other isolates from clini- cal material found considerable differences (Figure, Table Discussion 3). While the differences in S-segment sequences between This report documents the animal traceback investiga- the index hamster and pet store/distribution center viruses tion that linked a major pet rodent distribution operation to were <1% (<3-nt difference), LCMV-Armstrong and WE the recent outbreak of lymphocytic choriomeningitis in 4 differed from index hamster sequences by 14.1% (76-nt organ transplant recipients in Rhode Island and Massachu- difference) and 14.7% (86-nt difference), respectively. setts. This investigation demonstrates the ways in which Similarly, the differences in L-segment sequences in the classic epidemiology, laboratory diagnostics, and molecu- viruses from the index hamster and pet store/distribution lar biology can complement one another in the investiga- center rodents were <0.5% (3-nt difference), while LCMV- tion of disease clusters. LCMV was not found in the organ Armstrong and WE differed by 18.1% and 13.4% (41 nt donor’s tissues; however, the viral isolate from the pet store and 30 nt), respectively. These analyses indicate an overall hamster was sequenced and matched to the sequences of close identity among LCMV strains implicated in this in- the isolates from the recipients (10). The near-complete se- vestigation, and wide differences from previously published quence match between the virus found in the index hamster sequences of strains Armstrong and WE. This molecular and the virus sequenced from the Ohio hamster indicates evidence corroborates the epidemiologic data implicating that the genotypes share a common lineage that is dis- LCMV transmission within the commercial pet trade. tinct from previously identifi ed strains. It is unlikely that After the identifi cation of LCMV in the Ohio facility, this genotype would be as similar to a genotype found in all states that had received animals from this facility in the wild house mice in Rhode Island. Thus, the animal trace- previous 5 months were notifi ed. Many potentially infected back, coupled with the molecular phylogenetic evidence, animals still remained in stores; their disposition was deter- Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 13, No. 5, May 2007 723 RESEARCH mined by individual states. Actions taken ranged from sale with infected wild mice, and can pass the infection to hu- or adoption with an information leafl et or informed consent mans. Most human LCMV infections are associated with to issuing stop sale orders on specifi c rodent species from exposure to wild house mice (6,15); however, several out- the Ohio facility. breaks have been attributed to laboratory and pet mice and Upon notifi cation of the Ohio sample test results, the hamsters (5,6,16,17). One example is the 1974 outbreak Ohio Department of Agriculture informed the proprietor of associated with pet hamsters sold by a single distributor. A the LCMV contamination in his facility and requested that total of 181 symptomatic cases (46 requiring hospitaliza- a written plan for decontaminating the facility and a sec- tion) in persons with hamster contact were identifi ed in 12 ond plan for keeping the facility LCMV-free in the future states; no deaths occurred (5). The outbreak was brought be provided. The proprietor responded by depopulating the under control by voluntary cessation of sale and destruction Ohio facility. The quarantine was lifted after the building of the infected breeding stock. was disinfected, but the facility was never reopened. A direct link between the Arkansas and Ohio facili- Prevention ties was established by the discovery of the marked tubs LCMV surveillance should be a primary concern in the in Ohio. This enabled the Arkansas Departments of Health pet rodent industry to avoid entry of this virus into pet trade and Agriculture (Livestock and Poultry) to issue a Joint populations. Because of the ubiquitous distribution of the Quarantine and Inspection Order. However, several days house mouse, eliminating the natural reservoir of LCMV after depopulating the Ohio facility, the proprietor also de- is not practical. Steps can be taken, however, to exclude stroyed all the Syrian hamsters and guinea pigs at the Ar- wild house mice from homes and businesses. Immunocom- kansas facility. Although efforts were made to sample the promised persons and pregnant women should be advised remaining rodents at the Arkansas facility in an attempt to to avoid close contact with all rodents and infested areas. pinpoint the source of the virus found in Ohio, the owner Educational materials should address the risk from expo- refused to allow access to the rodents on several occasions sure to wild mice as well as pet rodents. The virus is not and >4 months elapsed between the initial and the ultimate- naturally present in pet rodent species and the ease of trans- ly successful attempts to conduct sampling. Virus isolation mission of the virus from pet rodents to humans may be on all samples was unsuccessful. greater than from wild mice when one considers the nature Several factors may have contributed to the lack of vi- of the relationship between pet rodents and their owners rus in the Arkansas breeding facility, including the follow- (i.e., close physical contact). Therefore, every effort should ing: 1) the virus was never there, 2) the destruction of the be made to eliminate the virus from pet populations when Syrian hamsters and guinea pigs eliminated the virus from it is discovered. the facility, 3) the elapsed time allowed for the removal LCMV is already actively excluded from laboratory of infected animals and subsequent decontamination of the rodent populations, because the infection can be an occu- facility, and 4) a complete replacement of the rodent stock pational hazard to laboratory workers who work around was accomplished within the facility. Although the facility infected rodents (5), and because inapparent infection can was under quarantine for 3 of the 4 months between the fi rst interfere with experimental results in rodent studies (18). and last attempts at sampling, only sporadic surveillance of Economic considerations may prohibit such rigorous bi- the facility was carried out by a governmental authority. osecurity measures like those used for laboratory animals; Elimination of the virus from the population did not likely however, sentinel surveillance (19), adequate veterinary occur naturally because LCMV can chronically infect mice care, exclusion of wild rodents (20), and good infection and will lead to persistent colonial transmission (6,11). control practices can substantially reduce the opportu- Since the time of our investigation, the proprietor’s nity for introduction and spread of LCMV and other pet license has been suspended by the USDA for 5 years for rodent pathogens in commercial pet populations. Efforts violations of the AWA unrelated to this investigation. The to increase such practices within the pet trade are under Rhode Island pet retailer who sold the index hamster report- way. Adherence to regulations that are already in place for edly ceased business relations with the distributor shortly obtaining permits and veterinary inspection of commercial after the infection was linked to his facilities. rodent populations can also reduce the likelihood of infec- To our knowledge, the Rhode Island outbreak repre- tion and improve animal welfare. sents the fi rst documented case of fatal LCMV infection Further efforts to reduce risk for LCM in pet owners are involving a pet animal (10). In the previous cluster of trans- ongoing. Education is critical in preventing LCM and other plant-related LCMV deaths, no rodent exposure was identi- pet-related infections. Potential pet owners should choose fi ed (9). Several rodent species that are sold as pets, includ- pets appropriate to their household (21). Pregnant women ing hamsters, mice, and guinea pigs, can be incidental hosts and immunocompromised persons should avoid pet rodents of LCMV. These species become infected through contact altogether (22). Additionally, pet owners should be advised 724 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 13, No. 5, May 2007 Rodents and Fatal Lymphocytic Choriomeningitis 9. Paddock C, Ksiazek T, Comer JA, Rollin P, Nichol S, Shieh WJ, et of the possibility of acquiring zoonotic diseases from any al. Pathology of fatal lymphocytic choriomeningitis virus infection pet and of precautions that should be taken to prevent ac- in multiple organ transplant recipients from a common donor. Mod- quiring pet-related infections such as LCM, tularemia (23), ern Pathology. 2005;18(Suppl 1):263A–4A. salmonellosis (24), and others. Persons can reduce risk for 10. Fischer SA, Graham MB, Kuehnert MJ, Kotton CN, Srinivasan A, Marty FM, et al. Transmission of lymphocytic choriomeningitis vi- infection from pet rodents by being attentive to proper hand rus via organ transplantation. N Engl J Med. 2006;354:2235–49. hygiene and environmental cleaning. Additional informa- 11. Traub E. The epidemiology of lymphocytic choriomeningitis in tion on LCMV is available from the CDC website (25). white mice. J Exp Med. 1936;64:183–200. 12. Lehmann-Grube F. Lymphocytic choriomeningitis virus. New York: Springer-Verlag; 1971. Acknowledgments 13. Bowen MD, Rollin P, Ksiazek TG, Hustad HL, Bausch DG, Dem- We wish to acknowledge the following persons and agen- byet AH, et al. Genetic diversity among Lassa virus strains. J Virol. 2000;74:6992–7004. cies for their contributions to this investigation: James Mills, 14. Animal Welfare Act 1990 (7 USC, 2131–2156). Beltsville (MD): Herminia Alva, Archer Miller, Thomas Stevens, Michael Bell, US Department of Agriculture Animal Welfare Information Center. Joel Montgomery, Mitesh Patel, Jeannette Guarner, Kimberly [cited 2006 June 1]. Available from http://www.nal.usda.gov/awic/ Slaughter, Deborah Cannon, Rita Helfand, Joe Foster, Heather legislat/awa.htm 15. Childs JE, Peters CJ. Ecology and epidemiology of arenaviruses and Horton; Frank Wilson, Altino McKelvey, Randy Owens, Sherry their hosts. In: Salvo MS, editor. The Arenaviridae. New York: Ple- Langley, Joe Bates, Rick Hogan, Reggie Rogers, Gail Kusturin, num Press; 1993. p. 331–84. and other staff in the Arkansas Department of Health and Hu- 16. Baum SG, Lewis AM, Rowe WP, Huebner RJ. Epidemic nonmen- man Services; Jeff Hayes, David Frew, and other staff in the Ohio ingitic lymphocytic-choriomeningitis-virus infection: an outbreak in a population of laboratory personnel. N Engl J Med. 1966;274: Departments of Agriculture and Health; Bernadette Juarez, Clara 934–6. Markin, Randy Coleman, Carl LaLonde, Betty Goldentyer, and 17. Ackermann R, Stille W, Blumenthal K. Syrische hamster als unber- additional staff in the US Department of Agriculture Animal and trager von lymphozytärer choriomeningitis. Dtsch Med Wochenschr. Plant Inspection Service. 1972;97:1725–31. 18. Nicklas W, Homberger FR, Illgen-Wilcke B, Jacobi K, Kraft V, Dr Amman is a mammalogist for the Special Pathogens Kunstyr I, et al. Implications of infectious agents on results of ani- mal experiments. Report of the Working Group on Hygiene of the Branch at CDC in Atlanta, Georgia. His research interests include Gesellschaft fur Versuchstierkunde–Society for Laboratory Animal investigating the ecology of, and relationships between, emerging Science (GV-SOLAS). Lab Anim. 1999;33(Suppl 1):S39–87. viruses and their mammalian reservoirs. 19. Lipman NS, Homberger FR. Rodent quality assurance testing: use of sentinel animal systems. Lab Animal. 2003;32:36–43. 20. Centers for Disease Control and Prevention. Rodent control: seal up! References Trap up! Clean up! [cited 2006 June 1]. Available from http://www. cdc.gov/rodents 1. Childs JE, Glass GE, Ksiazek TG, Rossi CA, Oro JG, LeDuc JW. 21. Centers for Disease Control and Prevention. Healthy pets, healthy Human-rodent contact and infection with lymphocytic choriomen- people. [cited 2006 June 1]. Available from http://www.cdc.gov/ ingitis and Seoul viruses in an inner-city population. Am J Trop Med healthypets Hyg. 1991;44:117–21. 22. Centers for Disease Control and Prevention. Update: interim guid- 2. Childs JE, Glass GE, Korch GW, Ksiazek TG, LeDuc JW. Lym- ance for minimizing risk for human lymphocytic choriomeningitis phocytic choriomeningitis virus infection and house mouse (Mus virus infection associated with pet rodents. MMWR Morb Mortal musculus) distribution in urban Baltimore. Am J Trop Med Hyg. Wkly Rep. 2005;54:799–801. 1992;47:27–34. 23. Centers for Disease Control and Prevention. Brief report: tularemia 3. Park JY, Peters CJ, Rollin PE, Ksiazek TG, Katholi CR, Waites KB, associated with a hamster bite—Colorado, 2004. MMWR Morb et al. Age distribution of lymphocytic choriomeningitis virus serum Mortal Wkly Rep. 2005;53:1202–3. antibody in Birmingham, Alabama: evidence of a decreased risk of 24. Centers for Disease Control and Prevention. Outbreak of multidrug- infection. Am J Trop Med Hyg. 1997;57:37–41. resistant Salmonella Typhimurium associated with rodents pur- 4. Lewis AM Jr, Rowe WP, Turner HC, Huebner RJ. Lymphocytic- chased at retail pet stores—United States, December 2003–October choriomeningitis virus in hamster tumor: spread to hamsters and 2004. MMWR Morb Mortal Wkly Rep. 2005;54:429–33. humans. Science. 1965;150:363–4. 25. Centers for Disease Control and Prevention. Lymphocytic chorio- 5. Gregg MB. Recent outbreaks of lymphocytic choriomeningitis in the meningitis (LCMV). [cited 2006 June 1]. Available from http:// United States of America. Bull World Health Organ. 1975;52:549– www.cdc.gov/ncidod/dvrd/spb/mnpages/dispages/lcmv.htm 6. Rousseau MC, Saron MF, Brouqui P, Bourgeade A. Lymphocytic Address for correspondence: Brian R. Amman, Centers for Disease choriomeningitis virus in southern France: four case reports and a review of the literature. Eur J Epidemiol. 1997;13:817–23. Control and Prevention, Mailstop A26, Atlanta, GA 30333, USA; email: 7. Barton LL, Mets MB, Beauchamp CL. Lymphocytic choriomen- [email protected] ingitis virus: emerging fetal teratogen. Am J Obstet Gynecol. 2002;187:1715–6. Use of trade names is for identifi cation only and does not imply 8. Richmond JY, McKinney RW, editors. Biosafety in microbiological endorsement by the Public Health Service or by the U.S. and biomedical laboratories. 4th ed. Washington: US Government Department of Health and Human Services. Printing Offi ce; 1999. Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 13, No. 5, May 2007 725

Journal

Emerging Infectious Diseases – Pubmed Central

Published: May 1, 2007

There are no references for this article.