American Journal of Primatology 389-98 (1982) Disseminated Strongyloidiasis in Erythrocebus pa tas J.S. HARPER IlIl, J.M. RICE', W.T. LONUON', D.L. SLY', ANDC.MIDDLETON3 IThe National Institutes of Health. Bethesda. Maryland 'Meloy Laboratories, Rookuille, Marylaitd, 3Sinclair Research Farm. University of Missouri Five fatal cases of disseminated strongyloidiasis were identified in Erythrocebus patas caged singly or in groups of two to four in an indoor research facility. This is the first report of fatal hyperinfective strongyloides infection in a species other than great apes and man. Severe pulmonary hemorrhage, duodenitis, and proximal colitis with microscopically demonstrable larvae in affected tissues were the key necropsy findings. E . putus is an available, suitable model for the study of disseminated strongyloidiasis. Key words: Erythrocebus patas, hyperinfective, Strongyloides stercorulis INTRODUCTION Strongyloidiasis has been recognized as a cause of mortality in man and great apes for many years. I t was originally described in French soldiers returning from southeast Asia in the late 19th century. Human strongyloidiasis has reemerged during the last decade as a common clinical disease in children and immunosuppressed adults [Purtilo et al, 1974; Burke, 1978; Rivera et al, 19701. Strongyloidiasis has been spread among severely mentally retarded institutionalized children, whose personal hygiene is poor, when contaminated fingers or other objects come in contact with the oral mucosa [Burke, 1978, Yoeli et al, 19631. Fatal cases have also been reported in the chimpanzee, orangutan, and gibbon. While subclinical intestinal infection has been commonly found in most species of monkeys, systemic infection causing clinical disease and mortality has not been reported in monkeys [DePaoli & Johnsen, 19781. Fatal disease is associated with hyperinfection, an unusual alternative life cycle of Strongyloides stercorulis where first stage larvae change t o infective stage larvae prior to leaving the colon. They then penetrate the wall of the colon and begin their migration through the vasculature to the lungs where they penetrate the alveolar wall, are coughed up, swallowed, and become adults in the proximal small bowel. This self-infecting cycle leads to life-threatening clinical disease. In man, hyperinfection has most commonly been associated with diseases or drugs that cause suppression of cell-mediated immunity [Purtilo, et al, 19741. Hyperinfection has been reported in great apes in the absence of Received February 8, 1482; revision accepted May 26, 1982. Address reprint requests to Dr. James S. Harper 111, NINCDS Building 36, Bethesda. MD 20205. 027~-2565/82/0301-04-0089$03.000 1982 Alan R. Liss, Inc. 90 Harperet a1 TABLE I. Fatal Cases of Disseminated Strongyloidiasis in Patas Monkeys E. patas number Date of death 552 11/24\76 8 months 218 12/13/76 3 years, 5 months 539 04/09/77 11 months 847 03/15/81 6.5 months Age a t death Hematologic abnormalities 9% Eosinophilia 6 months before death; but no eosinophilia 1 week before death Strongyloides infected cagemates Signs of illness Found comatose 11/23/76, no larvae in stool Breeder male number 122 Diarrhea began 12/09/81;larvae in stool; unresponsive to treatment Diarrhea with larvae in stool 04/08/77; unresponsive t o treatment None; found dead; no larvae in stool prior to dealh but many in cecum at P.M. 675 08/26/81 3 years, 11 months Breader male number 647 Lethargic for 1 week; no weight gain for last 8 months arior to death immunosuppression but most cases occurred in juveniles, which are most susceptible to hyperinfection [McClureet d,1973; Uemura et d, 19791. METHODS AND MATERIALS Five spontaneous cases of disseminated strongyloidiasis were identified between 1976 and 1981 in an National Institute of Health [NIII]sponsored Erythrocebuspatas colony held at a contract facility. Clinical records and histopathology reports for all 386 animals that were in the colony during that period were reviewed for evidence of infection with strongyloides. Monkeys that lost weight or had diarrhea were routinely screened for enteropathogenic bacteria and GI parasites as part of the diagnostic workup. Feces were checked for parasites by examining the sediment and the supernatant of a sample placed in saturated zinc-sulfate solution and centrifuged a t 2,000 RPM for ten minutes. Larvae were separated from feces using a Baerman apparatus and grown on charcoal beds to allow identification. A postmortem examination was made of every monkey that died. Tissues were fixed in 10% buffered formalin, stained with hematoxylin and eosin, and examined by light microscopy. RESULTS Five fatal strongyloides cases occurred in monkeys born in this colony. Table I lists the pertinent clinical history of each case. After the initial cluster of fatalities, every Disseminated Strongyloidiasis in E patas 91 FECAL POST MORTEM DEATHS r- 1980 1981 Isolation of Fecal Positives &Treatment YEAR INFECTION IDENTIFIED Fig. 1. Number of monkeys infected grouped by year infection was identified. monkey in the rooms where deaths occurred was checked for strongyloides larvae. Eight fecal positive monkeys, their cagemates, and the cagemates of the three initial fatal cases were put in hepa filtered isolator chambers, given thiabendazole (TBZ)100 mgikg once daily for 2 consecutive days and then once weekly for 1month. This stopped the daily shed of larvae. Two monkeys reshed larvae and were treated for an additional 5 weeks. Monkeys were removed from the chambers when they had at least three negative fecal exams over a period of at least 2 months. Single, prophylactic annual doses of TBZ were given to the entire colony after the initial cluster of cases. Since the final two fatal cases were noted in 1981, the frequency of thiabendazole prophylaxis has been increased to one dose every 3 months. Figure 1 shows the number of infected monkeys grouped by the year when the cases were identified. In 1978 strongyloides were detected by fecal examination in one monkey, and another had incidental histopathologic evidence of duodental infection. Duodenal strongyloidiasis was an incidental histopathologic finding in two postmortems in 1979. Five fecal exams were positive in 1980 and six intestinal infections were identified as incidental findings at postmortem examination. A third infant and second subadult female died of disseminated strongyloidiasis in 1981. In the first 10 months of 1981 one monkey was found incidentally to be infected upon histopathologic examination of the duodenum at postmortem and nine animals that were shedding larvae in their stools were identified. Eosinophilia was noted in five monkeys who were shedding larvae in their stools but did not have diarrhea. No eosinophilia was noted in 21 others who did have diarrhea when they were sheding larvae. An enzyme-linked immunosorbent assay for S. stercoralis antibody became available in 1981. Forty-two monkeys in the colony were found to have significant levels of antibody ( > 1/32).Between January, 1976, and December, 1981, seventy-five monkeys with serologic, fecal exam or histopathologic evidence of strongyloidiasis have been identified from a colony of 386. Larvae from a number of 92 Harperet a1 Fig. 2. Duodenum @OX)numerous adult and larval strongyloides present in mucosa fecal sheders were collected and later identified as S. stercoralis [Neva, personal communication]. Pulmonary hemorrhage was noted on the gross postmortem examination of all five fetal cases of disseminated strongyloidiasis. Bright red petechia or ecchymoses were seen throughout all lung lobes. Thickening of the wall of the large intestine and enlargement of mesenteric lymph nodes were observed in four cases. Hyperemia of the mucosal and serosal surfaces of the proximal duodenum, cecum, and proximal colon was noted in all cases. All animals lost weight during their final month of life and three progressed to the point of emaciation. Numerous strongyloides adults and larvae were evident in all histologic sections from the small intestine in the fatal cases. They were found free in the lumen between villi, burrowing into mucosa and invading the crypt epithelium. A few had penetrated to the Brunner’s glands. Infiltration of the villi by mononuclear cells and lyniphocytes was minimal and limited to the immediate area of penetration by a parasite (Figs. 2 and 3). Mucosal ulcer formation and neutrophil infiltration were not noted in any animal. There was significant shortening of villi in the duodenum in only one case. I n two monkeys, a larva was found in a submucosal lymph nodule. This was the only evidence of larval penetration through the mucosa of the small intestine. Numerous larvae were present on Disseminated Strongyloidiasis in Epatas 93 Fig. 3. Duodenum (220X higher magnification of Figure 2) to show strongyloides within lumen and intraepithelial tunnels. the serosa of the small intestine, but their point of exit from the small or large intestine could not be determined. A few larvae were noted in the bile duct in two individuals. Filariform larvae were found in all layers of the colon and cecum. In the mucosa they caused a moderate to severe neutrophil and mononuclear response. Shortening of villi was evident in some areas of severe cellular response and heavy concentrations of larvae. Clusters of larvae creating tracts throughout the muscularis t o the serosa were common. In contrast to leukocytic infiltration provoked by their presence in the mucosa, larvae induced a minimal cellular response in the muscularis (Fig. 4). Mesenteric lymph nodes were hypercellular with mononuclear cells and occasional giant cells at the periphery of germinal centers. Larvae were present in some lymph nodes as well as the surrounding connective tissue. Larvae were rarely seen in the liver but occasionally granulomas with giant cells were present. Marked congestion of hepatic sinusoids was a consistent finding. In the lung, hemorrhage filled most alveoli, many bronchioles, and larger bronchi. Pigment-filled macrophages were common in the pulmonary parenchyma as were scattered neutrophiles and giant cells. Larvae were not numerous in the lung. Most appeared to be in alveoli, attracting little or no cellular response. They were also occasionally 94 Harperet a1 Fig. 4. Colon (330X) cluster of strongyloides larvae following a common tract through the muscularis noted in bronchi and on the visceral surface of the pleura. The severity of the hemorrhage did not appear to correlate with the number of larvae present in the lung (Fig. 5 and 5a). Both of the subadult females that died had been housed with breeder males that are now known to be infected with strongyloides (Table I). The mothers of the three juvenile cases do not have evidence of strongyloidiasis but other monkeys inhabiting the room where the mothers, infants, and weanlings were housed are known t o be infected. Monkeys in the rooms where infections occurred were housed together for harem breeding. Females were separated when they became pregnant. Infants were kept with their mothers until weaning (at 6 months of age),when two or three juveniles were caged together. This movement pattern was the probable method by which the parasite was spread. The source of the initial patas strongyloides infections could not be determined. No cases of intestinal strongyloidiasis were identified prior to the first disseminated case in November 1976. All of the monkeys with proven strongyloidiasis were born in the colony except one. This one (Patas 443)was acquired as an adult in August, 1975, and strongyloidiasis was diagnosed in March, 1977. A stool examination during quarantine was negative. Single stool examinations detect less than one third of human infections Disseminated Strongyloidiasis in E patas 95 Fig. 5 . Lung (130X cross section of a strongyloides larva. Note the severe alevolar hemorrhage. Insert. Lung with 330X higher magnification of the larvae in Figure 5. [Burke, 19781. Pastas infections were at least as difficult to detect. In many cases of postmortem proven strongyloidiasis, we had not detected larvae despite repeated examinations of concentrated stool samples. Patas 443 could have been the source of the infection . There is another possible mode of entry of the parasite into the colony. Chimpanzees and patas were housed in open cages in the same room for a number of months during 1976. Later some of the chimpanzees were shown t o be infected with strongyloides. While there was no direct contact between the species, some of the chimpanzees had a habit of throwing their feces. This could have resulted in indirect spread of the parasite. Blood from 16 personnel in contact with chimpanzees or E. patus was examined for the presence of S. stereoralis antibody. One individual had significant ELISA antibody and larvae were found in one stool after multiple samplings with a Baerman apparatus (Neva, personal communication). This individual had daily contact with chimpanzees and patas monkeys. We are unable to determine if this individual acquired the infection from monkey contact. Multiple instances of transmission of strongyloides between man and ape have been reported [Flynn, 19731. 96 Harperetal DISCUSSION Strongyloides can be commonly found in the small intestine of most species of monkeys in the wild [De Paoli & Johnsen; 1978;Flynn, 19731.This is the first report of hyperinfective strongyloidiasis in a primate other than a great ape. Subclinical patas strongyloidiasis has proven to be an infection that is quite difficult to eradicate. Human cases of disseminated strongyloidiasis have occured more than 30 years after the last known exposure to an endemic area [Higenbottom & Heard, 19761. Sampling of duodenal contents is much more effective than fecal examination in man for identifying asymptomatic carriers, but the technique is difficult to use with uncooperative patients [Grove et al, 19751. These subclinical patas infections persisted despite the repetitive screening of all monkeys in the infected rooms and the isolation and treatment of cases and their cagemates for a prolonged period. The number of strongyloides positive diarrhea cases rose dramatically toward the end of 1980 and the first few months of 1981. (Fig. 1).While clinical disease was greatly diminished for almost 3 years after treating all identifiable cases, some monkeys still harbored parasites in their small intestine. The presence of undetected carriers combined with the movement of individuals for breeding and weaning allowed the parasite to infect a significant number of monkeys in the colony once again. Some patas infected with strongyloides initially show signs of gastrointestinal disturbance, as do great apes and man. Frequently no history of illness preceded death in patas; similar observations have been made in gibbons [DePaoli & Johnsen, 19781. Respiratory distress has been the initial sign of illness in disseminated strongyloidiasis cases in great apes and man, but it was not noted in any of the spontaneous patas cases. Disseminated strongyloidiasis mortality, 5 deaths among 75 with evidence of infection in this colony of 386 individuals, is much lower than in gibbons where almost all of the animals that were infected eventually died. Unlike gibbons, patas die infrequently from strongyloidiasis alone. In patas, as in man, some cases are asymptomatic, the majority of clinical cases present as diarrhea, and a few cases become disseminated. None of the pat as that developed disseminated strongyloidiasis were receiving immunosuppressive agents at the time of diagnosis, while almost all adult humans with systemic disease were immunosuppressed [Purtilo et al, 19741. Eosinophilia may occur transiently but it is not a reliable indicator of disease. It has been shown to to be a common finding in experimental gibbon infections but not in the naturally occurring disease of great apes or man [De Paoli & Johnsen, 19781. In man it may be an uncommon finding because human immunosuppressive therapy tends to cause eosinopenia [Purtilo, et al, 19741. Patas intestinal lesions did not appear to be as severe as those of the great apes [DePaoli & Johnsen, 19781. Extensive ulcer formation in the small or large bowel was not noted in patas monkeys. The critical similarity is that intestinal penetration by filariform larvae can occur with sufficient frequency to cause death in patas as well as the great apes and man. Pulmonary hemorrhage is the most consistent pathologic finding in all species in which fatal disseminated disease occurs. The severity of the hemorrhage does not appear to correlate with the number of larvae present in any species affected. Defective cell-mediated immunity due to disease andlor chemotherapy has been associated with most human cases of fatal strongyloidiasis, but some human cases have occurred in children with no history of immune deficiency due to disease or chemotherapy [Carvalho-Filho,1978; Smith et al, 1977; Burke, 1978, Lintermans, 19751. An inherent immunologic deficiency or immaturity may predispose young gibbons and orangutans to disseminated strongyloidiasis [McClureet al, 19731. Many adult patas in our colony have antibody to strongyloides, but few infants have antibody before l?h Disseminated Strongyloidiasis in E patas 97 years of age. Infant patas may have an immunologic aberration similar to the great apes and man which predisposes them to fatal disseminated disease. We suggest that E. patas is a suitable model for strongyloidiasis research. Cost, lack of availability, and their endangered status limit the use of great apes for experimental strongyloidiasis research. The low rate of fatal strongyloidiasis in patas more closely resembles the spectrum of disease in man than does the spectrum of disease seen in great apes. The most commonly reported model of experimental strongyloidiasis is the rat (S. ratti). However, this species does not produce hyperinfective disease even in immunosuppressed rats and the rats frequently spontaneously self-cure [Flynn, 19731. While patas are more expensive to work with than rats, the production of hyperinfective disease and the greater similarity of various immunologic reactions between patas and humans make patas an attractive alternate model. Strongyloidiasis is a zoonotic disease. Careful serologic screening in combination with multiple fecal parasite concentration examinations identified an infected animal caretaker and allowed the treatment of the individual before clinical disease occurred. CONCLUSIONS 1)E. patas is susceptible to fatal disseminated strongyloidiasis. 2) Clinical signs of disease are nonspecific and include diarrhea, weight loss, and sudden death. 3) Larvae are not necessarily demonstrable in the stool prior to death. 4) Pulmonary hemorrhage and enterocolitis are consistent pathologic features. 5) Strongyloidiasis is a zoonosis and personnel in contact with E. patas must be informed of the risk of disease and should be periodically screened serologically if strongyloidiasis is present in the colony. ACKNOWLEDGMENTS Four of the five cases were transplacentally exposed to ethylnitrosourea (ENU)at a low dose that has not been shown to cause demonstrable immunosuppression after birth (Larchian, W.; Bennett, M.; Elkort, R.; Rice, J. Immunological effects of l-ethyl-lnitrosourea (ENU)on Erythrocebus patas monkeys. FEDERATION PROCEEDINGS 39: 1139, 1980).While slightly more monkeys exposed to the transplacental carcinogen ENU intrautero developed fatal disease ( 4 of 218 ENU vs 1 of 84 not exposed),we do not feel that much weight can be placed on that small difference since we have no way of knowing if equal numbers of monkeys were exposed in each group, if they were exposed to similar numbers of parasites, or if the groups were age matched at the time of exposure. We thank Steve and Jeff Harbaugh for dedicated technical support and Louise Harris and Betty Finch for patiently editing the many revisions of the manuscript. We are grateful to Dr. Franklin Neva, Dr. Robert Genta, and Mr. A1 Gam of the Laboratory of Parasitic Diseases, NIAID, for identification of the parasite and serologic survey of the colony. NINCDS contract N01-NS-7-2375, NCI Contract N01-CP-15766 to Meloy Laboratories in part, supported this work. These studies were performed in an animal facility accredited by AAALAC. This facility follows the Public Health Service Guidelines for the handling of class I11 etiologic agents. REFERENCES Burke, J.A. Strongyloidiasis in childhood. Carvalho-Filho,E. Strongyloidiasis.CLINICS AMERICAN JOURNAL OF DISEASES OF I N GASTROENTEROLOGY 71179-200, CHILDHOOD 138:1130-1136,1978, 1978. 98 Harperetal DePaoli, A,; Johnsen, D.O. Fatal strongyloidiasis in gibbons (Hylobatses Lar). VETERINARY PATHOLOGY 15:31-39, 1978. Flynn, R.J. Strongyloidids. pp. 203-207 in PARASITES OF LABORATORY ANIMALS. Ames, Iowa State University Press, 1973. Grove, D.I.; Warren, K.S. Mahmoud, A.A. Algorithms in the diagnosis and management of exotic diseases Ill strongyloidiasis. JOURNAL OF INFECTIOUS DISEASE 131~775778,1975. Higenbottom, T.W.; Heard, B.E. Opportunistic pulmonary strongyloidiasis complicating asthma treated with steroids. THORAX 31:226-'233,1976. Lintermans, J.P. Fatal periotonitis an unusual complication of strongyloides stercoralis infection. CLINICAL P E D l A T R I C S 14:974-975, 1975. McClure, H.M.; Strozier, L.M.; Keeling, M.E.; Healy, G.R. Strongyloidosis in two infant orangutans. JOURNAL OF THE AMERICAN VETERINARY h4EDICAL ASSOCIATION 163:629-632,1973. Neva, F., Chief, Laboratory of Parasitic Disease, NIAID, NIH. Personal communication, 1981. Purtilo, D.T.; Meyers, W.M.; Connor. D.H. Fatal strongyloidiasis in immunosuppressed patients. THE AMEKICAN JOURNAL OF MEDICINE 56:488-493,1974. Rivera, E.; Maldonado, N.;Velez-Garcia, E.; Grillo, A,; Malaret, G. Hyperinfection syndrome with strongyloides stercoralis. ANNALS OF INTERNAL MEDICINE 72:199204,1970. Smith, S.B.; Schwartzman, M.; Mencia, F.; Blum, E.B.; Krogstad, D.; Nitzkin, J.; Healy, G.R. Fatal strongyloidiasis presenting as acute abdominal distress in an urban child. JOURNAL OF PEDIATRICS 91:607-609, 1977. Uemura, E.; Houser, W.; Cupp, C.J. Strongyyloidiasis in an infant orangutan. JOURNAL OF MEDICAL PRIMATOLOGY 8:282-288, 1979. Yoeli, M.; M o s t , H.; Berman, H.H.; Scheinesson, G.P. The clinical picture of a massive strongyloides infection in a child. TRANSACTIONS OF THE ROYAL SOCIETY OF TROPICAL MEDICINE AND HYGIENE 57:346-352,1963.