Strongyloides stercoralis: Hyperinfection in immunosuppressed dogs

EXPERIMENTAL

PARASITOLOGY

57, 287-296

(1984)

Strongyloides stercoralis: Hyperinfection lmmunosuppressed Dogs GERHARD Laboratory

of Parasitology,

A. SCHAD, School

MARIA

E. HELLMAN,

of Veterinary Medicine, Philadelphia, Pennsylvania

(Accepted for publication

AND DEREK

in

W. MUNCEY

University of Pennsylvania, 19104, U.S.A.

3800 Spruce

Street,

17 February 1984)

SCHAD, G. A., HELLMAN, M. E. AND MUNCEY, D. W. Strongyloides stercoralis: Hyperinfection in immunosuppressed dogs. Experimental Parasitology 57, 287-296. Hyperinfective strongyloidiasis involving the threadworm, Strongyloides stercoralis, is well known in humans and primates. Although this nematode also frequently parasitizes dogs, canine hyperinfective strongyloidiasis has not been reported. To determine whether a fulminant pattern of nematode development can occur in dogs, and to test the S. stercoralisldog system for suitability as a model for human hyperinfective and disseminated strongyloidiasis, five canine infections with a dog-derived strain of S. stercoralis were monitored by the quantitative recovery of larvae from feces. Even 3-month-old pups controlled their initial infections successfully, the number of larvae excreted declining to near zero in 90 days. Immunosuppressive treatment with prednisolone, prednisolone and azathiaprine, or niridazole resulted in a rapid return to former or greater intensities of infection, as judged by larval output. Only first stage (rhabditiform) larvae were passed in the feces, although third stage (Iilariform) larvae occurred in the intestinal contents of dogs when they were examined at necropsy. In 3 of the 5 dogs, the adult worm recovery exceeded the inoculated dose greatly and, in one of these, adults and rhabditiform larvae were found in distant, extraintestinal sites. In the remaining 2 of the 5 dogs, the adult worm population was less than the inoculated dose, but, in both, the infection was terminated by the host’s death before hyperinfection could have developed. The observations demonstrate that autoinfection occurs in dogs infected with S. stercoralis and that, if it is allowed to continue for a sufficiently long time in immunosuppressed hosts, massive hyperinfection, and even disseminated infection, may occur. This spectrum of increasingly invasive parasitism closely resembles strongyloidiasis in humans. Therefore, the S. stercoralisldog system has excellent potential as a model for human hyperinfective and disseminated strongyloidiasis. INDEX DESCRWORS: Strongyloides stercoralis; Nematode, parasitic; Autoinfection; Hyperinfection; Disseminated infection; Dog; Animal model; Immunosuppression; Glucocorticoids; Prednisolone; Azathiaprine; Niridazole.

INTRODUCTION

Strongyloides stercoralis is a common, although often focally distributed, intestinal nematode parasitizing dogs, man, and other primates. It usually exists in exceedingly chronic infections, the diagnosis and treatment of which are difficult. In humans and other primates, these chronic infections are thought to escape immunological control, resulting in autogenous internal reinfection (autoinfection) that may include a massive increase in the number of parasites (hyper-

infection) and migration to abnormal sites (dissemination) (Scowden et al. 1978; Genta et al. 1983). Massive, disseminated infections are often fatal. Until recently, it was uncertain whether autoinfective strongyloidiasis occurred in dogs. There had been only one report of experimentally induced autoinfection in a canine S. stercoralis infection (Nishigori 1928a,b). No cases of naturally occurring autoinfection had been recorded. Indeed, contemporary textbooks of veterinary parasitology indicate that autoinfection is

287 0014-4894/84 $3.00 Copyright Q 1984 by Academic Press, Inc. All rights of reproduction in any form reserved

288

SCHAD,

HELLMAN,

known only in human strongyloidiasis (Georgi 1980; Levine 1968; Soulsby 1982). Considering the common occurrence of S. stercorulis in dogs and the widespread use of the glucocorticoid, prednisolone, in veterinary medicine, the absence of reports of naturally occurring hyperinfective strongyloidiasis provided circumstantial evidence supporting the idea that, in the dog, S. stercoralis is incapable of autoinfection. Indeed, hyperinfection involving Filaroides hirthi, a nematode that is much less common than S. stercoralis, had been reported in immunocompromised dogs (August et al. 1980; Craig et al. 1978), providing additional support for the suggestion that autoinfection does not occur in canine strongyloidiasis . This uncertainty led to the present series of experimental investigations, the purpose of which was to determine whether autoinfection and hyperinfection would occur in canine strongyloidiasis, and, if they did, to explore the potential of this system as a model for human hyperinfective and disseminated strongyloidiasis. These investigations, recently reported in abstract by Schad et al. (1982), are described in detail. In the interim, another preliminary report concerning work toward the establishment of a canine model for disseminated strongyloidiasis has appeared (Grove et al. 1983). MATERIALS

AND

METHODS

The strain of Strongyloides stercoralis used in our investigations was isolated from naturally infected beagles suppled by White Eagle Laboratories, Doylestown, PA, USA. To obtain Blariform larvae (L,) for infecting dogs, the feces of previously infected animals were cultured on filter paper (St011 1963). The resulting larvae were washed in distilled water by repeated sedimentation, aliquots of suspended Ls’s were counted, and the desired inoculum was taken for subcutaneous injection between the shoulders in -1 ml of fluid. Doses varied and are given in the descriptions of the individual experiments. One experimentally infected dog was a basenji. Unless otherwise noted, the other experimental animals were specific-pathogen-free

AND

MUNCEY

(SPF) male beagles from White Eagle Laboratories (Table 1). To minimize the possibility of accidental reinfection, the dogs were housed individually in stainless-steel cages that were washed daily in a cage washer at 82 C. Water and a commercially prepared dry dog food were available ad libitum. One beagle was purchased naturally infected with S. stercoralis. To confirm the SPF status of other dogs, fecal samples were examined on alternate days for 2 weeks using both salt flotation and baermannization. Only animals found helminth free were accepted. Infections were monitored by baermannization of fecal samples weighing at least 1 g, taken on several days per week. The resulting data are presented as weekly means of larvae per gram of feces (LPG). With one exception, infections were allowed to proceed naturally. Then, when the initial infection had run its course, and few, if any, larvae were demonstrable in the feces, immunosuppressive treatment was initiated. In the exceptional case, treatment was begun 44 days before the dog was infected. Additional methodological details and the timing of events in each infection are given in Table I. To seek evidence of autoinfection in the living dog, migrating larvae were sought and recovered from transtracheal saline washes of the distal airways using techniques described by Roudebush (1983). In Experiments 1 and 2, the infected dogs were euthanatized to verify the occurrence of autoinfection soon after a secondary, prednisolone-triggered rise in larval output was evident. Thereafter, dogs were held until death to determine whether hyperinfection and dissemination would occur. Adult worms and larvae were recovered from the intestinal tract by a modified Baermann technique. Migrating larvae were sought in minced pulmonary tissue by peptic digestion (Exp. 1) or by baermannization (Exp. 2-4). To recover migrating larvae from the esophagus, trachea, and bronchi, each organ was slit longitudinally and its lumen was rinsed with a jet of saline. After evidence of hyperinfection with dissemination had been obtained, postmortem examination of dogs in Experiments 4 and 5 included baermannization of the macerated tissues of the brain and saline washes of the cranial and nasal cavities. RESULTS

In Experiment 1, the effect of daily immunosuppressive doses of prednisolone on dogs with chronic low-grade Strongyloides stercoralis infections was examined. An infection, initiated with 1500 subcutaneously injected S. stercoralis L3’s, became patent in 10 days (Table 1:Dog G2). The weekly mean larval output peaked at 135 larvae per

St~0figykks

289

HYPERINFECTION IN DOGS

SterCOrUh.

TABLE I Hosts, Infective Doses, ImmunosuppressiveAgents, and Adult Worm Recoveriesin Five Attempts to Induce Hyperinfection with Strongyloides stercoralis in Dogs From infection Experiment no.

Dog no.

Age, sex, and breed

1

G2 OS3

basenji db 3 month 6 beagle

2

LE4

3

CO1

4

AE7

3month 6 beagle 3 month d beagle 7 month d beagle

Dose of larvae

until (days)

Patent

Reverting to negative

Immunosuppressive agent

1500 600 + 1500 on Day 54 5cilo

10 11

85 94

104 129

Prednisolone Prednisolone

70(E) 43(E)

400 96cQ

9

-

-44’

Prednisolone

112(D)

13,000

Natural infection 3ooo

-

&Id

203(D)

101,660

11

45

Prednisolonel azathiaprine Niridizole

Suppression initiated

footnote’ 67

a Days after immunosuppressive treatment was initiated; E, euthanatized; D, died. b Less than 1 year, precise age unknown. c Daily prednisolone treatment initiated before dog infected. ’ Days after natural infection was detected. e F’rednisolone, Days 71-l%; Day l%-201, gradual withdrawal of prednisolone; azathiaprine, Days 218-281.

gram of feces (LPG) 3 weeks postinfection @.i.), and then declined gradually to zero at Week 13 (Fig. 1). When no more than 1 LPG had been found for a 19-dayperiod, daily prednisolone treatment (2.2 mg/kg) was begun.Three days later a rise in larval output commenced,but, becauseit peaked at only 40 LPG and declined to about 10 LPG, the daily dose of prednisolone was doubledon the 56thday after treatmentwas initiated. One week later the mean larval

bYS until death”

No. adults recovered

25(D)

Days l%-218;

10

prednisolone,

output rose precipitously to 200 LPG. The dog was sacrificed on the following week, when a single larval count was 3773LPG. Four hundredadult worms were recovered from the small intestine. A nonquantitative examination found rhabditiform larvae in both the small andlargeintestine,and a few filariform larvae in the large intestine. No larvae were found in the trachea or in a peptic digest of the lungs. A secondinfection (OS3), initiated with NECROPSIED

PREDNISOLDNE INCREASED 4 4 mg/kg/day PREDNISDLONE 2.2 mp/kg/day \

1 ‘2 -A-u 0

2

4

6

*

10

12

14

16

-1 22.4 18

WEEKS POST- INFECTION

FIG.

1. The

effect

of daily,

immunosuppressive

doses

of prednisolone

on the course

of an experi-

mentally induced Strongyloides stercoralis infection in a dog (G2). Prednisolonetreatment was initiated after the infection had become chronic and few, if any, larvae were demonstrablein the feces. The weekly mean number of larvae per gram of feces (LPG) was used as a measureof the intensity of infection.

290

SCHAD,

HELLMAN,

AND MUNCEY

600 subcutaneously injected L,‘s, became patent in 11 days (Fig. 2). Fecal output of larvae increased steadily until Week 7, when it peaked at 460 LPG. An attempt to boost the infection with 1500 additional subcutaneously injected larvae failed to increase larval output. Instead of increasing, it decreased to less than 10 LPG by the thirteenth week p.i. When this low chronic level had been maintained for 5 weeks, daily immunosuppressive treatment with prednisolone (4.4 mg/kg) was initiated, whereupon larval output increased rapidly (Fig. 2). After elevated larval counts had been observed for 6 weeks, the dog was sacrificed. At necropsy, 9600 adult worms, along with innumerable rhabditiform larvae, were recovered from the small intestine. The adult worm recovery exceeded the number of larvae in the total inoculum by a factor of 4.6. A nonquantitative examination of the large intestine revealed innumerable rhabditiform larvae, among which a few Ls’s were seen. Sixty four L,‘s were found in the trachea, but none was recovered from the iungs.

h

440400360320280

3 3 r”

In Experiment 2, the effect of daily immunosuppressive doses of prednisolone on dogs with a S. stercorulis infection induced after the beginning of prednisolone treatment was examined. Each of the previously described infections extended over a 6month period. In an attempt to decrease the time until massive hyperinfection and, perhaps, to provoke dissemination as well, immunosuppression (4.4 mg prednisolone/kg/ day) was initiated 6 weeks before a 3month-old beagle (LE4) was infected with 5000 larvae of S. stercoralis (Table I). The infection, as judged by larval output, increased rapidly (Fig. 3). By the eighth week p.i., the diarrheic feces were constituted largely of blood, and transtracheal saline washes demonstrated migrating L,‘s, indicating that autoinfection was occurring. Eosinophil counts declined and, in spite of intensive supportive treatment, the host became severely dehydrated and died 9 weeks p.i. Thirteen thousand adult worms (approximately 2.5 times the number of larvae in the initial inoculum) were recovered from the intestines by use of the Baermann technique.

1

I t

240s zoo-

I60IZO“-

DOSED

0

2

:’

DOSED

\

PREDNISOLONE /

6 6 IO

b

NECROPSIED

12

14

16 16

20

22

24

WEEKS POST-INFECTION

2. The effect of daily, immunosuppressive doses of prednisolone on the course of an experimentally induced Strongyloides stercoralis infection in a dog (OS3). Prednisolone treatment was initiated after the infection had become chronic and few, if any, larvae were demonstrable in the feces. The weekly mean number of larvae per gram of feces (LPG) was used as a measure of the intensity of infection. FIG.

StrongJ’l0ides

SterCOrah:

HYPERINFECTION

IN DOGS

291

in an attempt to elicit a new increase,daily azathiaprinetreatment (2 mglkg) was initiated while prednisolone was withdrawn. However, no new rise in larval output was provoked. The dose of azathiaprine was thereforeincreasedto 3 mg/kg/day.Larval output declined sharply to zero, with many dead larvae being found in the feces. A transtracheal wash did not recover migrating larvae. Because it appeared that azathiaprine might be killing the larvae, adults, or both, this treatment was terminated and daily administration of predniWEEKS POST- INFECTION solone (4.4 mg/kg) was resumed. This reFIG. 3. The course of an experimentally induced sumption was followed by a dramatic reStrongyloides stercoralis infection in a dog (LE4) crudescence of larval output, which peaked treated with daily immunosuppressive doses of prednisolone beginning 6 weeks before infection. The at 35,000LPG. weekly mean number of larvae excreted per gram of Beginningabout 2 weeks after prednisofeces (LPG) was used as a measure of the intensity of lone treatment was resumed (when larval infection. output reached 1049LPG), the dog developedan intermittent bloody diarrheaandits Innumerable rhabditiform larvae (LJ appetite declined. After an additional 5 were again present in the gastrointestinal weeks, a transtrachealwash recovered 34 tract. Among theselarvae, the longer, more L,‘s, suggestingthat numerouslarvae were active L,‘s proved readily identifiable and migrating, Thereafter, despite intensive were counted.-0 hundredand thirty L3’s supportive therapy, the dog became inwererecoveredfrom the small intestineand creasingly dehydrated and died 8 weeks 22 from the cecum and colon, the small after prednisolonetreatment was resumed. number found in the large intestine sug- Eosinophil counts ranged from 1 to 3% gestingthat autoinfection was occurring in during the terminal phase of the infection the lower bowel. Eight other Ls’s, presum- and only declined to zero immediately beably in the processof migrating back to the fore death (Fig. 4). Postmortem examinasmall intestine, were recovered from the tion revealed a massive infection of over lungs (l), trachea (6), and stomach(1). 100,000 adult worms, with 1500 in the In Experiment 3, the effect of sequential stomachand 10in the trachea. daily treatment with prednisoloneand aza- Because the dog was found dead, possible vomiting and aspiration of adult thiaprine on the course of a S. stercoralis infection in a dog was examined. This ex- worms or postmortem wanderingcould experiment useda naturally infected dog with plain their presence in the trachea and a preinvestigationlarval output of - 17LPG stomach. Nevertheless,these observations (TableI). As in Experiment 1, the courseof suggestedthe possibility that the infection infection (Fig. 4) was monitored until no had, in fact, disseminated. Therefore, a larvae were found in the feces. Thereafter, more extensive and intensive examination daily prednisolone treatment (4.4 mg/kg) of the distribution of the parasitewas conwas initiated and the fecal excretion of ducted, the results of which are summalarvae increasedmarkedly, peaking at 6100 rized in Table II. In Experiment 4, the effect of niridazole LPG 12 weeks after treatment was begun. infecLarval output failed to increasefurther, and on the courseof canineS. stercoralis DOG DIED

292

SCHAD,

HELLMAN,

AND MUNCEY DOG

DIED

PRECNISOLCNE

PREDNISOLONE

0

2

4 6 8 IO

12

14

16 18

20

22

24

26

20

30

32

34

36

36

40

WEEKS POST- INFECTION

FIG. 4. The effect of a sequence of attempted immunosuppressive treatments with prednisolone and azathiaprine on the course of a naturally acquired Strongyloides stercoralis infection in a dog (COl). The weekly mean number of larvae excreted per gram of feces (LPG) was used as a measure of the intensity of infection.

tion was studied. Experiments 1 and 3 suggested that immunosuppressivetreatment of dogs chronically infected with S. stercoralis could elicit hyperinfective development of the parasite. However, since the action of prednisoloneis complex (Miller 1983),theseexperimentsprovided little insight into the immune mechanisms that Distribution

and Abundance

might control the distribution and abundance of S. stercoralis. For this reason, it was of interest to determinewhether niridazole, a drug which inhibits various cellmediatedimmunological responsesbut has little effect on antibody production (Mahmoud et al. 1975;Pelley et al. 1975;Grove and Warren 1976),would cause a recru-

TABLE II of Strongyloides stercoralis Larvae and Adults in a Dog Treated with Prednisolone and AzathiaprineO Larvae recovered

Organ and method of examinationb

Total

Bhabditiform

Filariform 03)

Adults recovered

Oesophagus Stomach Small intestine Large intestine

W) (B) (B) (B)

100 2,800 125,000 23,700

100 2,800 123,750 23,300

0 0 1,250 400

0 1,500 100,000 150

Trachea Lungs

W) (B)

700 60

560 30

140 30

10 0

Cranial cavity Nasal cavities Brain

w (W) W

4 408 0

4 400 0

0 0 0

0 0 0

152,764

150,944

1,820

101,660

Totals

0 Prednisolone, Days 71-l%; Days 196-201, gradual withdrawal 218; prednisolone, Days 218-281. b (B), organ baermannized; (W), lumen or space washed.

of prednisolone;

azathiaprine,

Days 196-

~tr0ngJ’~oides

HYPERINFECTION

SferCOrdiS:

descenceof a chronic S. stercorulis infection in dogs. A 7-month-oldbeagle(AE7) was infected with 3000S. stercoralis larvae (TableI). In this somewhat older dog, the infection reverted to apparent negativity within 7 weeks, and, 2 weeks later, daily treatment with niridazole (20 mg/kg) was initiated. A modest increasein larval output occurred (Fig. 5), which was not sustained.The dog died 3 weeksafter treatmentwith niridazole had begun,due to the drug’s side effects. Although larval output had declined to zero at the time of death, postmortem examination revealed a light infection of 10 adult worms, 130Ls’s in the small intestine and 15 LX’s in the large intestine. No parasiteswere recoveredfrom washingsof the tracheaor from the baermannizationof the lungs. The apparent absence of rhabditiform larvae was unexpectedand is inexplicable. DISCUSSION

Attempts have been made to establisha laboratory model of human hyperinfective strongyloidiasisusing Strongyloides rutti in glucocorticoid-treated rodents (Olsen and Schiller 1978;Moqbel and Denham 1978; Grove and Dawkins 1981).In no instance, however, did the number of adult worms recoveredat necropsy exceed the number of larvae given at infection. Furthermore, spontaneoushyperinfection is unknown in rodent strongyloidiasis. Spontaneouslyoccurring hyperinfective and disseminated strongyloidiasishas beendescribedin several species of primates, particularly an-

IN DOGS

thropoid apes infected with S. stercorulis (DePaoliand Johnson1978;McClure et al. 1973). More recently, experimentally induced S. stercorulis infections in the patas monkey (Erythrocebus patus) have been shown to undergo intense dissemination (Gentaet al. 1983;Harper et al. 1984).But, in primates, dissemination occurs so readily that theseinfections cannot be considered to represent human disseminated strongyloidiasis with much fidelity. For these reasons, we chose to investigate S. stercorulis in prednisolone-treated dogs with the hope that this system might provide a good model of human-disseminated strongyloidiasis. Also, dogs are becoming relatively well-known immunologically and, therefore,provideopportunitiesfor research not afforded by primates (Shift-me and Wilson 1980). In each of our 5 dogs, the adult worm population had declined to a nadir scarcely demonstrableby fecal examination,but recrudescedafter immunosuppressivetreatment. Under conditions designed to prevent accidental reinfection from the environment, these observations suggestthat autoinfection occurred. This deduction is supported by the recovery of filariform larvae (L3) from the lungs of 2 living dogs by transtracheallavage. Furthermore, L,‘s were recoveredat necropsyfrom the intestines and from the circuitous migratory route that returns the parasitesto the small intestinevia lungs, trachea,esophagus,and stomach. Autoinfection without a great increasein adult worm density can occur in immuno-

8 !i z

DOSED L3

NIRIOAZOLE 20 mg/kq/doy

.1 3ooo

WEEKS

293

POST-INFECTION

FIG. 5. The effect of daily niridazole treatment on the course of an experimentally induced Strongyloides stercoralis infection in a dog (AE7). The weekly mean number of larvae excreted per gram of feces (LPG) was used as a measure of the intensity of infection.

294

SCHAD,

HELLMAN,

logically intact dogs (Nishigori 1928 a,b). Investigations in humans indicate that the parasite populations increase to hyperinfective levels in immunocompromised hosts. In 3 of the 4 prednisolone-treated dogs, hypet-infection occurred; the fourth dog was probably sacrificed before the parasite population had time to increase markedly. Our results suggest that, although individual variation is marked, when dogs are strongly immunosuppressed and when suppression is continued for sufftciently long, hyperinfection can be provoked. While no attempt was made to monitor the dogs’ immune responses to S. stercoralis infection, we did attempt to verify that the prednisolonetreatment had an immunosuppressive effect. Blood samples were collected periodically, before and after treatment, for lymphocyte transformation tests, which were done by the Clinical Immunology Laboratory of the Veterinary Hospital of the University of Pennsylvania. These showed a decreased post-treatment responsiveness to phytohemagglutinin (PHA) or to PHA and conconavalin A in 4 of our 5 dogs; the fifth (AE7) was not tested. Thus, independent evidence indicated that the immunosuppressive treatments were successful. Cost precluded the use of control animals, but it is unlikely that the adult worm populations found in our dogs are attributable to reinfection from the external environment. We have emphasized that helminthlogically sterile cages were provided daily. Additionally, we never observed L,‘s, i.e., the infective filariform larvae, in fecal samples taken from the cages, indicating that the L,‘s occurring in the intestine are not expelled in the feces but, instead, must penetrate the host’s tissues. The failure to find L,‘s in these samples also shows that the rhabditiform larvae did not have time to attain infectivity at the cool temperatures of our climate-controlled animal colony. Furthermore, if reinfection is assumed to account for the recrudes-

AND MUNCEY

cence of infection, then it becomes difficult to explain the very sharp rise that often followed the initiation of prednisolone treatment, when the feces were practically devoid of larvae and reinfection from the outside could not have been intense (Figs. ~24). In the absence of a control animal to confirm that the period of depressed larval output preceding immunosuppressive treatment did reflect a corresponding nadir in adult worm abundance, it is important to note that the dog (G2), sacrificed soon after its larval output began to increase, harbored only a few adult worms (i.e., 400, or 20% of the dose). This contrasts sharply with the large adult worm population (i.e., 9600 worms, or 400% of the dose) of OS3, the dog allowed to survive for 7 weeks after its post-treatment increase in larval output began. Indeed, the adult worm population of G2 also contrasts sharply with that of the other two dogs (LE4 and COl) that were held for several weeks after their larval excretion began to increase markedly, and only resembles that of one dog (AE7), whose larval output had declined to an undetectable level by the time of its death. Both of these dogs with low rates of larval output harbored adult worm populations that were substantially smaller than the dose of larvae given (Table I). Thus, although these observations cannot substitute fully for a control animal sacrificed at precisely the appropriate time, they do indicate that low adult worm populations were associated with low, or relatively low, levels of larval output, and that larval output probably was an adequate measure of the adult worm populations. These observations are in agreement with those of Grove and Northern (1982), who found that canine S. stercoralis infections were selflimiting, with a decline in adult worm populations, and a corresponding decline in larval output, occurring as resistance developed. It is possible that arrested larvae sur-

Slrongyloides

stercora/is.’

viving in the tissues from the time of the initial infection could have resumeddevelopment, matured, and contributed to the repopulation of the intestinal tract. However, becausethe final adult worm population in at least 2, and probably 3 dogsexceededthe number of larvae in the initial exposure greatly, the putative arrested larvae could haveplayed only a minor role, at most, in these recrudescences. Disseminated infection was observed in one dog (COl), in which abnormally distributed adult worms occurred in the stomach and, particularly, in the trachea. Furthermore, rhabditiform larvae were found in many abnormal extraintestinal sites, including the esophagus, stomach, trachea, and lungs, as well as in the nasal and cranial cavities. This severe dissemination, not previously reported to occur in canine strongyloidiasis, is an important attribute of a complete model of human hyperinfective, disseminatedstrongyloidiasis. Although the investigations reported herein were only preliminary in nature, using single dogs serving as their own controls, and with experimental protocols varying from dog to dog, they do provide persuasiveevidence for the occurrenceof autoinfection,hyperinfection, and disseminated infection in canine strongyloidiasis. Furthermore, they indicate that S. stercoralis infection in the immunosuppressed dogs will prove to be a faithful model of human hyperinfective and disseminated strongyloidiasis.CanineStrongyloides stercoralis infections, like those of man, tend to becomechronic and may be reactivated by immunosuppressivetreatment, but the ease, timing, and degree of reactivation shows considerable variability. In these characteristics, the model reproduces human strongyloidiasiswith remarkablefidelity. ACKNOWLEDGMENTS We thank Dr. Michael Page, Dr. Hassan M. S. El Nagger, and Dr. Hamish Robertson for their generous assistance with various aspects of this research, and

HYPERINFECTION

295

IN DOGS

Dr. Peter Felsburg and Ms. Linda Aikens who conducted the lymphocyte transformation tests. We also thank Dr. K. D. Murrell, Dr. W. M. Hominick and two anonymous reviewers for criticism of the manuscript. Expert technical assistance was provided by Michael Lindsay, Louise Andrews, David Leiby, Charles Duffy, and Karen Blumrick. The contributions of Harriet Hill and Marc Franz are gratefully acknowledged. We thank Susan Pharaoh for typing several drafts of the manuscript. Revisions of this paper were made while the senior author was supported by a Fellowship from the Wellcome Trust. REFERENCES POWERS, R. D., BAILEY, W. S., AND D. L. 1980. Filaroides hirthi in a dog: Fatal hyperinfection suggestive of autoinfection.

AUGUST,

J. R.,

DIAMOND,

Journal ciation

of the American

Veterinary

Medical

Asso-

176, 331-334. CRAIG, T. W., BROWN, T. W., SHEFOTAD, D. K., AND WILLIAMS, G. D. 1978. Fatal Filaroides hirthi: Infection in a dog. Journal of the American Veterinary Medical Association 172, 1096-98. DEPAOLI, A., AND JOHNSON,D. 0. 1978. Fatal strongyloidiasis in gibbons (Hylobates lar). Veterinary Pathology 15, 31-39. GENTA, R. M., HARPER, J. S., GAM, A. A., LONDON, W. T., AND NEVA, E A. 1983. Experimental disseminated strongyloidiasis in Erythrocebus patas monkeys. Part II: Immunology. American Journal of Tropical Medicine and Hygiene, in press. GEORGI, J. R. 1980. “Parasitology for Veterinarians.” Saunders, Philadelphia. GROVE, D. I., AND DAWKINS, H. J. S. 1981. Effects of prednisolone on murine strongyloidiasis. Parasitology 83, 401-409. GROVE, D. I., HEENAN, P. J., AND NORTHERN,C. 1983. Persistent and disseminated infections with Strongyloides stercoralis in immunosuppressed dogs. International Journal of Parasitology 13, 483-490. GROVE, D. I., AND NORTHERN, C. 1982. Infection and immunity in dogs infected with a human strain of Strongyloides stercoralis. Transactions of the Royal Society of Tropical Medicine and Hygiene 76, 833-

838. GROVE, D. I., AND WARREN, K. S. 1976. Effects on murine trichinosis of niridazole, a suppressant of cellular but not humoral immunological responses. Annals

of Tropical

Medicine

and Parasitology

70,

449-453. HARPER, J. S., GENTA, R. M., GAM, A. A., LONDON, W. J., AND NEVA, F. A. 1984. Experimental disseminated strongyloidiasis in Etythrocebus patas monkeys. Part I: Pathology. American Journal of Tropical Medicine and Hygiene, in press. LEVINE, N. D. 1968. “Nematode Parasites of Damestic Animals and of Man.” Burgess, Minneapolis, MN.

296

SCHAD,

HELLMAN,

A. A. F., MONDEL, M. A., WARREN, K. S., AND WEBSTER, L. T. 1975. Niridazole II. A potent long-acting suppressant of cellular hypersensitivity. Journal of Immunology 114, 279-283. MCCLURE, H. M., STROZIER,L. M., KEELING, M. E., AND HEALY, G. R. 1973. Strongyloidiasis in two infant orangutans. Journal of American Veterinary MAHMOUD,

Medical

Association

163, 629-632.

MILLER, E. M. 1983. Glucocorticoidinduced immunosuppression. Veterinary Medicine and Small Animal Clinician 78, 1199-1204. MOQBEL, R., AND DENHAM, D. A. 1978. Strongyloides ratti: the effect of betamethasone on the course of infection in rats. Parasitology 76, 289-298. NISHIGORI, M. 1928a. The factors which influence the external development of Strongyloides stercoralis and on autoinfection with this parasite. Taiwan Igakkai Zasshi 277, 31-33 (English Summary). NISHIGORI, M. 1928b. On various factors influencing the development of Strongyloides stercoralis and autoinfection. Taiwan Igakkai Zassi 27, l-56. Translated in “Tropical Medicine and Parasitology, Classic Investigations” (B. H. Kean, K. E. Mott, and A. J. Russell, eds.), Vol. 2, pp. 340-345. Cornell Univ. Press, Ithaca, NY. OLSEN, C. E., AND SCHILLER, E. L. 1978. Strongyloides ratti infections in rats II. Effects of cortisone treatment. American Journal of Tropical Medicine and Hygiene

27, 527-531.

PELLEY, R. P., PELLEY, R. J., STAVITSKY, A. B.,

AND

MUNCEY

MAHMOUD, A. A. F., AND WARREN, K. S. 1975. Niridazole, a potent long-acting suppressant of cellular hypersensitivity III. Minimal suppression of antibody responses. Journal of Immunology 115, 14771482.

ROUDEBUSH, P. 1983. Diagnostics for respiritory diseases. In “Current Veterinary Therapy” (R. W. Kirk, ed.) Vol. 8, pp. 222-224. Saunders, Philadelphia. SCHAD, G. A., MUNCEY, D. W., AND HELLMAN, M. 1982. Hyperinfective canine strongyloidiasis. In “Molecular and Biochemical Parasitology, Abstracts of the Fiih International Congress of Parasitology (Toronto)” (D. Mettrick ed.), p. 682. Elsevier, Amsterdam/New York. SCOWDEN, E. B., SCHAFFNER, W., AND STONE W. J. 1978. Overwhelming strongyloidiasis; an unappreciated opportunistic infection. Medicine (Baltimore) 57, 527-544. SHIFR~NE,M., AND WILSON, E D. 1980. In “The canine as a biological model: immunological, hematological and oncological aspects,” p. 425. National Information Service, U. S. Dept. of Commerce, Springfield, VA. SOULSBY, E. J. L. 1982. “Helminths, Arthropods and Protozoa of Domesticated Animals.” Balliere, Tindall and Cassell, London. STOLL, N. R. 1962. For hookworm diagnosis, is finding an egg enough? Annals of the New York Academy

of Sciences

98, 712-724.