Diarrhea in a Pancreatic Transplant
Sheldon Campbell M.D., Ph.D., and Drew Olsen, M.D.
Dr. Olsen is a resident in the Department of Pathology, Yale University School of Medicine. Drs. Olsen and Campbell report no commercial conflict of interest.
Release Date: 04/02/1999
Termination Date: 04/02/2002
Estimated time to complete: 1 hour(s).
Albert Einstein College of Medicine designates this enduring material for a maximum of 1.0 AMA PRA Category 1 Credit(s)™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.
Albert Einstein College of Medicine is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians.
Learning ObjectivesUpon completion of this Cyberounds®, you should be able to:
A 36-year-old man presented with watery diarrhea and red urine.
The patient had a renal transplant from a living, related donor two years prior to presentation, and a pancreatic transplant one year prior to presentation, both secondary to refractory, complicated diabetes mellitus. There was no history of rejection.
The patient had begun to have watery diarrhea one day ago. Then, on the day of presentation, his urine became red and he sought medical attention. He denied fever, chills, night sweats, or urinary frequency or urgency. He denied significant travel or ill contacts.
Hypertension, multiple laser treatments for diabetic retinopathy.
Lungs clear, no abdominal findings.
The patient's illness was considered a possible urinary tract infection (UTI), with some element of dehydration. He was sent out on antimicrobial therapy and asked to return in two days for repeat labs.
Six days later, the patient developed a blanching erythematous rash which Dermatology considered consistent with a viral exanthem, most likely of enteroviral origin. The patient was begun on acyclovir 800 mg t.i.d. The diarrhea was treated with Imodium® and improved.
Two days later, the patient re-presented with two days of nausea and vomiting. Further labs, including stool culture, blood cultures and stool Ova & Parasites (O&P) were obtained and the patient admitted to the hospital. His dehydration was attributed to a probable viral gastroenteritis. Renal biopsy showed no rejection and pancreatic enzymes were within the reference range. However, the patient's diarrhea returned and there was persistent nausea.
On the fifth hospital day, a laboratory result was obtained.
Q. What would be on your differential diagnosis list? How would you use the laboratory to approach the various possibilities?
A. An immunocompromised transplant patient with persistent diarrhea has a broad differential. Infectious possibilities include the usual bacterial pathogens, though the patient's course is somewhat prolonged for the Salmonella/Shigella/Campylobacter/E. coli O157 group of organisms which are normally detected on routine stool cultures.
Clostridium difficile toxin mediated enterocolitis should be considered strongly in this patient as well. It is diagnosed by detecting the toxin in the stool by EIA, latex agglutination, or antibody-inhibited cytopathic effect in cell culture. Patients need not be on antibacterial therapy (though this one was) to be at risk for Clostridium difficile enterocolitis, which has been associated with cancer chemotherapeutic drugs and with a variety of immunocompromised hosts.
Viral causes of gastroenteritis include the common rotavirus, typically detected by any of several EIA procedures, as well as a variety of difficult-to-diagnose viruses. These infections are usually self-limited in normal hosts, but may have a prolonged, severe course in immunocompromised patients.
In addition, an important cause of colitis in immunocompromised hosts is cytomegalovirus (CMV). CMV antigen testing would be a primary approach to assess the likelihood of a CMV-associated colitis.
A variety of protozoa; Giardia, Cryptosporidium and Cyclospora can cause chronic diarrheal illness in normal and compromised hosts. A helminth, Tricuris trichiura, can cause diarrheal illness, as can Strongyloides stercoralis, which is a particularly significant pathogen in the compromised host.
Many rhabditiform larvae of Strongyloides stercoralis were observed in the stool examination for ova and parasites.
Strongyloides is usually diagnosed by demonstrating the rhabditiform larvae in the stool. The larvae are 180-360 mm long by 14-20 mm wide. The only other nematode larvae commonly seen in the stool are hookworm larvae (Necator americanus or Ancylostoma duodenale). The hookworm larvae are rarely seen in fresh stool; the embryonated ova of hookworms typically hatch into larvae only 12+ hours after the stool is passed. When specimens are collected and placed rapidly in preservatives (as should be done with all stool O&P), hookworm larvae should not be present. In comparison with hookworm larvae, Strongyloides larvae have a short buccal canal and a genital primordium that indents the intestine about halfway down the length of the organism.
The patient was a resident of Connecticut and walked barefoot in a small Connecticut town. There was no history of travel, no visits to psychiatric institutions, no contacts with foreign-born persons.
Q. Where did this infection come from?
A. Strongyloides stercoralis is an organism that is widely distributed in moist tropical and subtropical areas of the world, but not Connecticut. Barring a travel exposure, the organism can only have come from the transplanted organ. Pancreatic transplants include a segment of duodenum, which is anastomosed to the urinary bladder to provide a conduit for excretion of pancreatic enzymes from the transplanted organ. Strongyloides resides in the duodenal mucosa and, thus, may be transmitted via a pancreatic transplant.
The donor of the transplanted pancreas was a young male from Brazil who died from a ruptured berry (intracranial) aneurysm. A saved serum from the donor was positive for Strongyloides antibodies.
Strongyloides stercoralis is a nematode parasite of humans that has a wide global distribution and an estimated global prevalence of 100 million cases.(1) Infective filariform larvae develop from the rhabditiform larvae after passage, reside in the soil and infect the host by penetrating intact skin. The larvae migrate through the tissues to the small intestine, either via the lung or directly, where they invade the mucosa. Ova hatch during their passage through the GI tract. Filariform larvae can develop during GI passage and can penetrate the colonic wall or perianal skin to maintain a chronic infection.
Symptoms are typically mild; dermatological and pulmonary symptoms may transiently occur at the time of infection. Chronically infected patients may be asymptomatic or have mild GI complaints but, often, asymptomatic eosinophilia is the only manifestation. Chronic infections lasting up to 60 years from the last reported exposure have been described.
Pancreatic transplants include the segment of duodenum into which the exocrine pancreas drains. In this case, the donor almost certainly harbored Strongyloides, which, when introduced into a host who was immunosuppressed to prevent transplant rejection, proliferated and caused a GI presentation. The patient may also have had disseminated strongyloidiasis, though parasites were not observed in the skin lesions or elsewhere. Transmission of Strongyloides has also been described from renal allografts.(2),(3) Presumably, in these cases, the recipient acquired worms that were migrating through the organ.
In normal hosts, the immune system typically keeps the infection in check. In steroid-treated patients, patients with profound leukopenia from chemotherapy and other immunocompromised hosts, this autoinfective cycle can spiral out of control. Patients with this hyperinfection syndrome can have multisystem disease, with peritonitis, respiratory failure and CNS involvement. In addition to direct and immune-mediated tissue damage, the migrating larvae can carry GI flora with them, resulting in unusual and persistent Gram negative bacterial infections.
In immunocompetent hosts, eosinophilia is often a clue to the presence of occult worms. Unfortunately, because steroid therapy, the primary risk factor for disseminated disease, blunts the eosinophilic response so that less than 50% of persons with disseminated strongyloidiasis have eosinophil counts above the normal range, the lack of eosinophilia in the case above is, therefore, not surprising. HIV-infected patients, interestingly, do not seem to be at increased risk for either primary Strongyloides infection or for hyperinfection.
In a review of 37 renal allograft recipients with Strongyloides infections, deVault et al(4) found that most patients had history of travel to or residence in an endemic area but the period elapsed since probable exposure was over 30 years in some cases. Only eight patients exhibited any sign of infection prior to transplantation. Two primary clinical presentations were observed. The first, a GI presentation with pain, diarrhea, nausea and vomiting, contrasted with a pulmonary presentation resembling progressive pneumonia or ARDS. Eosinophilia was observed in only 45% of cases. Interestingly, the patient described above had tacrolimus rather than cyclosporine A in his immunosuppressive regimen; cyclosporine has antiparasitic activity in animal models and there are case reports which suggest activity in patients as well.
Particularly in immunocompetent patients, Strongyloides larvae are shed in small numbers. Persistence in performing multiple O&P exams over days to weeks is critical to making the diagnosis.(5) In patients with a high index of clinical suspicion sampling of the duodenal contents to recover larvae may facilitate the detection of Strongyloides infections. Special procedures for concentration and culture of larvae have been described but few laboratories in the US are experienced in their use. Diagnosis may be facilitated by Strongyloides serology,(6),(7) especially in non-endemic areas; seropositivity in US patients has a high predictive value for infection. In hyperinfection syndromes, the larvae may be found in Gram stains of the sputum or detected by 'trailing' effects on culture plates.(8) Patients have been described with hyperinfection syndrome and larvae seen in sputum but not stool.
An increasing number of laboratories are instituting protocols for limiting O&P exams(9) and screening for a limited number of protozoa, rather than routinely performing comprehensive examination of all stools for rare parasites. These protocols make sense from a resource management perspective, given the comparative rarity of parasitic diseases in the industrialized world. This case, however, is a reminder that global travel and migration can spread exotic diseases to novel areas and unusual hosts. Laboratory directors should be aware of the clinical situations which require a more complete examination of stool for evidence of unusual parasitic infestations.