Diarrhea

Diarrhea

in HIV Patients

Michael A. Poles, M.D., Douglas T. Dieterich, M.D.

Introduction

As the third decade of our study and treatment of HIV begins, advances in the clinical and basic sciences have given us grounds for significant optimism. Highly active antiretroviral therapy (HAART) significantly suppresses HIV replication allowing for restoration of immune function, and new therapies eradicate many opportunistic infections, leading some optimists to proclaim that the scourge of AIDS will soon be behind us. Still, the reality is that while we have made advances in treating HIV disease in the industrialized world, spread of the virus continues to increase and disease manifestations continue to cause significant morbidity and mortality in the developing world. Even in some cities of the United States, in the year 2000, new HIV cases appear to have doubled compared to each year in the past decade. Any discussion of the clinical manifestations of HIV must now encompass two disparate groups of patients, those with controlled HIV replication and little to no immunosuppression, and those patients with significant immunosuppression and a high risk of opportunistic diseases.

The importance of the gastrointestinal system in the grand scheme of HIV immunopathogenesis and clinical care cannot be underestimated; from mouth to anus, the entire gastrointestinal tract plays an important role. HIV patients experience a wide variety of gastrointestinal manifestations. Of these myriad symptoms, diarrhea is the most predominant. The prevalence of diarrhea in HIV patients, both with and without immunosuppression is high and it results in significant morbidity, increased healthcare expenditures, and decreased quality of life. Physicians caring for patients with HIV require up-to-date information to make rational decisions regarding diarrheal disease to ensure that morbidity and mortality are minimized and quality of life and medical care costs are optimized.

Epidemiology

Before 1996, in the pre-HAART era, at least half of North American and Western European patients with HIV developed diarrhea. More disturbing is the fact that, even today, in developing countries, diarrhea among HIV-infected patients is almost universal (1), and most affected patients continue to die of malnutrition and wasting. Based on pre-HAART data, a pathogen can be identified in 50-85% of patients with diarrhea (2,3).

Little data exists to enumerate the prevalence of diarrhea since the institution of HAART. While the severity of diarrhea appears to have decreased, it is still quite prevalent. One study from the New York Statewide Planning and Research Cooperative System showed that in New York, diarrhea accounted for 2.8% of more than 15,000 hospitalizations of HIV patients in 1998 (4). The most frequent causes of diarrhea in this cohort were Clostridium difficile (51.3%), other protozoal diseases (18.1%), and other organisms, not elsewhere specified (11.7%). Since the incidence of opportunistic infection has been decreased in the face of HAART-induced immune reconstitution, non-opportunistic causes of diarrhea are more likely to be responsible for diarrhea in HAART-treated patients. Although not all patients in his study underwent endoscopy for a diarrhea, Monkemuller et al, found that since the introduction of HAART, the prevalence of opportunistic infectious findings has decreased from 69% to 13% (5).

Whatever the cause of diarrhea, it is significantly associated with diminished quality of life, and is an independent predictor of poor prognosis in HIV patients (4). In addition to substantial work lost, such patients incur annual health costs that are 50% higher than comparable patients without diarrheal symptoms (5).

Differential Diagnosis

Patients with significant immuno suppression

In compiling a list of potential causes of HIV-related diarrhea (Table 1) in immunosuppressed patients the clinician must first and foremost be suspicious of every conceivable disease. New pathogens in common places and common pathogens in new places are the rule and not the exception. If no gastrointestinal condition is identified, look elsewhere. For instance, Pneumocystis carinii pneumonia sometimes presents with diarrhea. Finally, in one's zeal to find an opportunistic infection, the diagnostician should not overlook the fact that HIV-infected persons will also be subject to non-opportunistic, non HIV-related processes typical for their age and sex. Lactose intolerance is common among patients with HIV (8), but many patients consume large quantities of milk or ice cream to gain weight. In addition, most of the drugs used in the treatment of AIDS patients, including reverse transcriptase inhibitors, protease inhibitors and antibiotics are also associated with diarrhea.

The list of diarrhea-inducing pathogens affecting HIV patients is extensive, though the frequency and prevalence of each of these is difficult to judge. Bacterial pathogens are common causes of diarrhea. The most common bacterial cause of diarrhea among severely immunosuppressed HIV patients is MAC, a ubiquitous environmental organism. Studies suggest that approximately 10-40% of AIDS patients are infected with MAC, with most of these infections occurring in the gastrointestinal tract (9,10). MAC-induced diarrhea is typically seen when CD4 cell counts fall to less than 50. The diagnosis of disseminated MAC is readily made by culture of blood or bone marrow, but the diagnosis of gastrointestinal disease may require endoscopy where MAC may appear as granular white nodules with a surrounding rim of erythema. Alternatively, the mucosa may appear erythematous, friable, and occasionally ulcerated, but may appear normal. Fresh stool samples may contain heavy bacterial loads but this may represent colonization, so the definitive diagnosis of MAC infection is usually made histopathologically. The organism is found within foamy macrophages on Ziehl-Neelson staining. Culture of biopsy samples increases the sensitivity of diagnosis. Mycobacterium tuberculosis is a more rare cause of diarrhea in AIDS, but is clinically similar to MAC.

Salmonella species, Shigella flexneri, and Campylobacter jejuni are other common bacterial pathogens responsible for diarrhea, though their incidence has been reduced by the widespread use of antibiotic prophylaxis against opportunistic infections, and in the case of salmonella, AZT (11). Salmonella and Campylobacter are associated with advanced immunosuppression, but Shigella is more commonly diagnosed in the early stages of HIV (12). The absence of a normal humoral and cytotoxic immune response may be associated with failure to clear these organisms from the intestine and lead to a long-term asymptomatic carrier state that often persists despite antibiotic therapy. Poor nutritional status is also a risk factor for salmonellosis (13). Another bacterial pathogen, enteroadherent E coli has been found in over 20% of HIV patients with diarrhea and may play an etiologic role (14,15). Diagnosis is usually made by stool culture or by biopsy where it may be found by light microscopy, in clusters adherent to the small bowel or colonic mucosa brush border.

Clostridium difficile is also quite common in HIV patients, perhaps due to the high prevalence of antibiotic usage, and increased hospitalization rate. Overall, only 5.8% of HIV patients are infected or colonized with C. difficile (16). Still, Clostridium difficile-associated diarrhea is one of the most common hospital-acquired infections in HIV-infected patients during long-term care. (17). In one French Hospital, the annual incidence of clostridium difficile related diarrhea ranged from 1.7 to 6.4 per 100 HIV-infected patients (18). In this study, three independent factors were significantly associated with disease, CD4+ cell counts less than 50, and use of clindamycin or penicillin. C. difficile may also occur without the prior use of antibiotics. One study that reported that C. difficile is more virulent in HIV positive patients (19) has not be substantiated (20).

Protozoa are very common causes of opportunistic diarrhea in immunosuppressed, HIV-infected patients. After transmission by fecal oral spread and by contaminated water, Cryptosporidia parvum may inhabit the microvillus border of intestinal epithelial cells. This results in villus blunting with reactive epithelial changes in the duodenum, and cryptitis with epithelial apoptosis in the colon (21). The effect of this is a severe disruption of intestinal barrier function that likely accounts for diarrhea and malabsorption (22). Earlier in the HIV epidemic, infection with Cryptosporidia was responsible for as much as 20% of cases of HIV-related diarrhea (23). While infection in immunocompetent hosts is self-limited with a duration of about 2 weeks, in patients with less than 180 CD4 cells, cryptosporidium may cause chronic, debilitating diarrhea (24). The ileum and jejunum are the predominant sites of infection, but cryptosporidium can be found in the colon and stomach. Infection involving both the large and small intestine is associated with a more severe diarrheal illness. The diagnosis of cryptosporidial infection is usually made on modified acid-fast stain or ELISA of stool. However, mucosal biopsies may be necessary when stool studies are negative.

Microsporidia are very small obligate spore-producing intracellular protozoa found widely in vertebrate and invertebrate hosts. They are considered a significant cause of chronic diarrhea, malabsorption and wasting among HIV-infected patients, though one study found no difference in the frequency of microsporidia among HIV-infected patients with and without chronic diarrhea (25). Patients are usually severely immunodeficient with CD4 cell counts less of than 100. The incidence of this disease in AIDS patients with diarrhea has been reported to be as low as 2% to as high as 44% (26,27). Although there are several species of microsporidia, approximately 90% of human infections are due to Enterocytozoon bieneusi. A second microsporidian, Encephalitozoon intestinalis, is unique in its ability to infect lamina propria macrophages and disseminate to distant sites (28). The diagnosis of intestinal microsporidiosis has traditionally depended on direct visualization of the parasite in stool specimens or intestinal biopsy samples by light and/or electron microscopy. Improved diagnostic methods including modified trichrome stain, fluorescent stains, and immunologic staining of biopsy and stool specimens, as well as PCR of stool specimens (29). Infection is greatest in the proximal jejunum where round to oval plasmodia, sporonts, and mature spores are seen in the supranuclear cytoplasm of infected cells. Microsporidia appear to cause diarrhea, in part by impairing enterocyte function, causing malabsorption (30).

Other presumed protozoal diarrhea-causing pathogens include Giardia lamblia, Isospora belli, Cyclospora cayetanensis, and Entamoeba histolytica. Giardia, the most common is transmitted by contaminated drinking water or through person-to-person contact. The organism is released from cysts in the upper small intestine where it adheres to the brush border, resulting in disrupted crypt architecture. Isospora is a rare cause of diarrhea in North America, but assumes far greater significance in developing countries, occurring in up to 20% of AIDS patients from Haiti or Africa (31). Cyclospora, previously described as blue-green algae, is a coccidian organism that may infect the small bowel and cause watery chronic diarrhea in severely immunosuppressed AIDS patients (32). The mode of transmission appears to be water- or food-borne. Like cryptosporidium, it may be discovered on acid fast staining of the stool, but is larger. Other protozoal infections like Pneumocystis carinii (33), and Toxoplasma gondii (34) may infect the bowel and should be considered in the differential diagnosis.

Cytomegalovirus (CMV) is a ubiquitous herpes virus that often affects the colon (67% in one series (35)) but can cause damage throughout the gastrointestinal tract. Infection is most common in the distal colon. Chronic diarrhea and abdominal pain are the most frequent clinical manifestations of CMV colitis, but patients may also present with lower gastrointestinal hemorrhage, with or without an antecedent history of diarrhea. The colonoscopic abnormalities may be categorized into three main groups: colitis associated with ulcer (39%), ulceration alone (38%), or colitis alone (20%) (36). Subepithelial hemorrhage is also a prominent endoscopic manifestation of disease. Clinical CMV disease usually occurs after the CD4 cell count has fallen to below 100 and increases linearly below this level (37). The diagnosis of CMV colitis is made histologically. Multiple biopsies should be taken of involved mucosa, but since CMV may also be found in normal-appearing mucosa, biopsies should be taken there as well. Histopathological examination reveals characteristic large mononuclear, epithelial, endothelial, or smooth muscle cells containing intranuclear or cytoplasmic inclusions. The inclusion body may be surrounded by a halo of clear space suggestive of the appearance of a classic "owl's eye." In some cases the viral inclusions are atypical or sparse and special staining with immunoperoxidase and DNA in situ hybridization, as well as PCR may increase the sensitivity and specificity of histological examination.

Herpes simplex virus (HSV) can infect the rectum and sigmoid of HIV-infected patients and can result in diarrhea and bleeding. Early infection may appear as small vesicles on the colonic mucosa that may rupture and coalesce to form ulcers. Given the similarity of HSV to CMV infection in endoscopic appearance and symptomatology, they must be distinguished histopathologically. Examination of biopsy specimens will reveal the characteristic Cowdry type A intranuclear inclusions in multinucleated cells. Viral culture or immunohistochemistry also help in the diagnosis.

Adenovirus is another viral cause of diarrhea in immunosuppressed HIV patients. It may also be identified in colon biopsies from AIDS patients with chronic diarrhea in both inflamed and normal appearing mucosa (38). Patients with CD4 cell counts <200 are prone to more prolonged adenovirus infection (39). The results of some studies suggest that adenovirus is not associated with a diarrheal disease in HIV (40,41), while another study suggested that gastrointestinal excretion of adenovirus is associated with worse prognosis in HIV (42). The histopathologic changes of adenovirus are different from that of CMV, but the sensitivity of light microscopy is improved by immunostains (43). Additional studies utilizing stringent techniques have led to the discovery of other viral pathogens that may contribute to diarrhea in patients with AIDS, including astroviruses, caliciviruses, coronavirus, and rotavirus (44). Like adenovirus, the significance of these viruses in the pathogenesis of diarrhea is confounded by their presence in many asymptomatic AIDS patients.

Fungal causes of diarrhea are less common than bacterial, protozoal or viral causes. However, in endemic areas, fungi such as Histoplasma capsulatum may more commonly cause gastrointestinal symptoms. Reactivated Histoplasma may lead to a chronic infection with dissemination to lymphatic tissue. Any segment of the gastrointestinal tract may be affected, but the terminal ileum with its abundant lymphoid tissue is the most common site. Infection may be asymptomatic, but more commonly results in diarrhea, weight loss, fever, and abdominal pain. Endoscopically, the mucosa appears inflamed, occasionally with ulcerations or plaques. The organism may be seen on histopathological examination of Giemsa-stained biopsy material. In many patients the diagnosis is not made until autopsy. Other fungi, such as Candida albicans, Cryptococcus neoformans, North American Blastomycoses, and Coccidiomyces immitis have been reported to cause diarrhea in patients with AIDS, though their significance remains unclear.

The prognosis may actually be better for HIV infected patients with pathogen-negative diarrhea after an extensive evaluation. In one prospective study, the median survival for patients with pathogen-negative diarrhea (48.7 months) was similar to control patients without diarrhea, and significantly greater than those with an enteric pathogen (9.6 months) (45).

Patients with controlled HIV replication

HAART therapy significantly decreases HIV replication and treated patients are more likely to avoid immunosuppression. Despite this, the incidence of diarrhea continues to be high in these patients. Monkemuller showed that among HIV-infected patients undergoing endoscopy between 1995 and 1998, opportunistic gastrointestinal diseases were found in 91% of 23 patients not on antiretroviral therapy, 35% of patients receiving monotherapy, and in only 30% of patients receiving HAART (5). It is unlikely that many patients with near-normal or normal CD4 cell counts will experience opportunistic infectious causes of diarrhea. Unfortunately, little data presently exists to help define the new causes of diarrhea in these patients and many patients are given a diagnosis of either "HIV enteropathy" or "drug-induced diarrhea" without hard evidence.

There is little consensus as to how to define HIV enteropathy. There are no pathognomonic clinical or endoscopic features that characterize this entity, though it apparently represents diarrhea in an HIV-related patient without other gastrointestinal processes, and with non-specific histologic findings and/or features of malabsorption. Oftentimes, a well-trained gastrointestinal pathologist will identify a mild, nonspecific increase in cellular inflammation in mucosal samples from well-controlled HIV patients. At the present time, there is little strong evidence that HIV is the direct cause, though our current knowledge of the immunopathogenesis of HIV strongly suggests this possibility. Some of the knowledge in support of this assertion includes the presence of HIV in mucosal biopsy specimens, the inflammation induced by this infection, and the decreased mucosal barrier function in HIV-infected patients that may be secondary to the inflammation.

HIV infection should have a profound impact on the gastrointestinal mucosa and its function. After all, the gut is the body's largest lymphoid organ, harboring the majority of the body's HIV-infectable CD4 cell population (46). These cells are more susceptible to HIV infection than are peripheral blood mononuclear cells (PBMCs); the majority of mucosal CD4 cells bear the requisite co-receptors CCR5 and CXCR4 for HIV binding and entry (47). In addition, mucosal lymphocytes exhibit a higher degree of cellular activation than their peripheral counterparts so once HIV gains entry to a cell and has integrated its reverse transcribed genome into the cellular DNA, greater quantities of progeny virus will be produced (48). HIV therefore has a propensity to rapidly infect the gastrointestinal mucosa, regardless of the route of HIV transmission, and spread amongst mucosal CD4 cells, resulting in their rapid depletion (49,50). Mucosal CD4 cell depletion likely has functional consequences resulting in diarrhea since the lymphocyte population of the gut is known to modulate epithelial cell maturation and function, including nutrient, electrolyte and water absorption. There is also some evidence to suggest that despite a lack of CD4 receptors, epithelial cells themselves may be infected by HIV (personal communication). We have shown that both HIV RNA and DNA can almost uniformly be found in small mucosal biopsies of HIV patients with positive plasma viral load and eve in the majority of patients with undetectable plasma viral loads (51).

How can HIV be an inflammatory mucosal disease when infection results in mucosal CD4 depletion and a normo- or hypocellular histological picture? Given the profound depletion of CD4 cells contained therein, the majority of inflammatory cells that compose the cellular infiltrate of the lamina propria are activated CD8 cells (52). While the biopsies are not overtly histopathologically inflamed by virtue of increased cellularity, the presence of large numbers of cytokine-producing effector CD8 cells does, in fact, represent cellular inflammation. Further, inflammation must also encompass what we cannot see, the soluble mediators of inflammation. Kotler et al. was the first to postulate and show that HIV infection, in the absence of other mucosal pathogens, is associated with soluble inflammation (53). Both he and McGowan (54) have shown increased expression of pro-inflammatory cytokines in the mucosa of HIV-infected patients. We have also shown increased pro-inflammatory cytokines as well as increased concentrations of beta chemokines what would function to draw other lymphocytes into the inflamed mucosa (personal communication). The intestinal mucosa can therefore be considered to be a HIV-producing factory. Infection occurs early in infection, resulting in increased production of cytokines and chemokines that initiate greater replication of integrated HIV and result in chemotaxis or more CD4 targets and pro-inflammatory CD8 cells into this environment. Perpetuation of this response results in high mucosal viral load, CD4 depletion and inflammation. It is this inflammation that, as in prototypical inflammatory mucosal diseases, likely contributes to diarrhea. HIV may also alter the function of the gut's intrinsic enteric nervous system or its vast endocrine cell compartment, again resulting in diarrhea (55). Despite the temptation of invoking HIV as a cause of diarrhea in this setting, it is to the patient's advantage to be treated for a specific pathogen, if one exists, than nonspecifically for diarrhea. Therefore, as will be described, an extensive work-up to look for the etiology off diarrhea is worthwhile.

Another etiology for diarrhea in HIV patients that has been largely ignored is steatorrhea, or fat malabsorption. There has been only one pre-HAART prospective study that studied the incidence of fat malabsorption in HIV-infected patients. The authors reported that fat malabsorption was present in 49% of 45 patients (56). Interestingly, in that study, while steatorrhea was seen in 83% of HIV-infected patients who had not been given a diagnosis of AIDS, the incidence was lower in patients with AIDS including those with enteric opportunistic infections. Additionally, steatorrhea was also identified in patients being evaluated for weight loss despite adequate nutritional intake (46%) but without diarrhea. In order to further examine this issue, we studied 33 HIV-infected patients that included patients both pre-HAART and those receiving HAART (57). This study revealed that fat malabsorption was present in 90.9% of patients and was associated with significant diarrhea. Further, fat malabsorption was present regardless of the type of anti-viral regimen or the patients' plasma viral load. While this trial does not prove that fat malabsorption was causative, it is likely that fat malabsorption was playing an important role in the diarrhea experienced by these patients. The mechanism of fat malabsorption is not obvious in these patients. Evidence did not suggest significant pancreatic dysfunction, or overt mucosal damage. The unifying elements in all of these patients include the presence of HIV, with its known predilection for the gastrointestinal mucosa, and the use of nucleoside analogue drugs by these patients. Perhaps these medications induce a subtle biochemical defect in one of these organs resulting in fat malabsorption. Regardless of the cause, the fat malabsorption may be indicative of micronutrient malabsorption or impaired absorption of medications.

Another cause of diarrhea in HAART-treated patients is the medications themselves. According to the Physician's Desk Reference (PDR) the risk of diarrhea associated with anti-retroviral medications ranges from 2% for nevirapine to 28% for nelfinavir. The exact mechanism for the diarrhea remains unknown, but one study suggested that nelfinavir use might result in fat malabsorption, perhaps due to a mucosal receptor or transporter defect.

Another possible etiology for diarrhea in HIV-infected patients receiving HAART is small bowel bacterial overgrowth (SBO), though its presence is controversial. In a study by Wilcox et al, sampling of small bowel fluid did not show a significantly increased incidence of SBO (58). In contrast, in a recent study, we showed, using lactulose breath testing that SBO may be found in around 50% of patients, both those receiving and not-receiving HAART (personal communication). Patients with small bowel overgrowth had a lower mean CD4 cell count than patients without SBO (116 vs. 195 cells/mm³), though statistical significance was not achieved. Other possible etiologies for diarrhea in this population include irritable bowel syndrome and bile salt excess.

Risk Factors

While diarrhea is seen in all groups of HIV patients, those at greater risk include patients with a CD4 cell count below 200-250, and male homosexuals (59,60). Multiple risk factors have been found to be associated with infection by Cryptosporidia including male homosexual contact, consumption of municipal tap water, and animal exposure (61-63). Risk factors that have been reported to be specifically associated with microsporidia-related diarrhea include a CD4 cell count less than 100, male homosexuality, swimming in a pool in the previous 12 months, and recent travel to a tropical country (64). Male homosexual contact may also be an important mode of transmission of Giardia (65). Among HAART-treated patients, the patients most likely to experience diarrhea are those receiving nelfinavir.

Clinical Manifestations

Signs and Symptoms

The history and physical examination may offer clues to direct the diagnostician to the possible site of disease involvement (colon or small bowel). While it is not uncommon to have pathogens affecting both sites, narrowing the work-up in this way may decrease the depth and expense of evaluation. Diarrhea of small intestine etiology (enteritis) is typically large volume and watery, resulting in dehydration and electrolyte abnormalities. Stool outputs greater than 2-3 liters may be seen. The greater the degree of mucosal architectural disturbance caused by the pathogen, the more likely the patient is to have other signs of malabsorption. Patients with enteritis may also experience periumbilical abdominal pain that is usually described as being cramping.

In comparison to small bowel-related diarrhea, diarrhea of colonic origin (colitis) is usually described as frequent, small volume, occasionally bloody bowel movements that often contain pus and mucous. Left, and less commonly right lower quadrant pain is a common accompaniment and may be associated with pain on defecation, and tenesmus, a sense of incomplete evacuation of rectal contents. Patients with rebound tenderness are also more likely to have a colon process.

Elements in the history and physical examination may also suggest the presence of specific pathogens. When fever over 101° F and abdominal pain accompanies the diarrhea bacterial pathogens may be the cause. The acute febrile diarrheal illness due to Salmonella species, Shigella flexneri, and Campylobacter jejuni may include cramping, bloating, nausea, and possibly tenesmus with passage of multiple small volume stools containing blood, pus and mucous. These organisms may rarely be associated with toxic megacolon or pseudomembranous colitis. Enterocolitis due to Clostridium difficile is no different than in immunocompetent patients, and typically results in a bloody, mucoid diarrhea with abdominal pain and fever. The diarrheal disease due to MAC is typically protracted and associated with abdominal pain, fever, anemia, weight loss, and night sweats. Complications of gastrointestinal MAC infection include obstruction, fistula formation, bleeding, and perforation.

Protozoal diarrhea is also protracted. It is non bloody, voluminous, frequently occurring at night, and may be associated with nausea, vomiting, severe weight loss and cramps. Fever is not a common accompaniment. Patients with less than 50 CD4 cells may experience fulminant disease, with passage of greater than 2 liters of stool daily, which is often refractory to therapy. These patients often are anorectic and experience significant weight loss and malnutrition.

Most patients with CMV colitis experience intermittent or persistent diarrhea, crampy lower abdominal pain, tenesmus and weight loss. Rebound tenderness, in a patient with compatible medical history, is strongly suggestive of peritonitis from CMV-induced perforation or inflammation (66). HSV can cause diarrhea or constipation, hematochezia, and tenesmus due to painful perianal ulcers. Gastrointestinal symptoms may also be accompanied by dysuria, paresthesias and urinary or fecal incontinence due to sacral nerve involvement.

Coincident with the decrease in opportunistic causes of diarrhea with HAART, the characteristics of the diarrhea have also changed. Whereas opportunistic causes of diarrhea are commonly small bowel in origin and therefore more watery and voluminous, resulting in dehydration, HAART-associated diarrhea is commonly less voluminous and less commonly results in dehydration. This type of diarrhea commonly defies localization to a small or large bowel source. Further, there are no pathognomonic signs or symptoms that distinguish medication-related diarrhea from steatorrhea, small bowel bacterial overgrowth, or HIV-related inflammation.

Radiographic Manifestations

The role of radiography in the evaluation of HIV-related diarrhea is not well defined at present. The majority of findings on routine abdominal radiographs, barium enema and upper gastrointestinal series with small bowel follow-through including flocculation of the contrast due to high volumes of liquid, and bowel wall thickening will require further endoscopic evaluation. Abdominal CT scan may offer few clues to assist in diagnosis. Mass lesions or intraperitoneal lymphadenopathy may suggest mycobacterial infection or lymphoma which can cause diarrhea through induction of a protein-losing enteropathy.

Diagnosis

Approach to Diagnosis (Figure 1)

An extensive work-up will reveal a diagnosis in the majority of immunosuppressed HIV-infected patients, but the it is more likely to be fruitful in homosexual men, patients with significant weight loss, those with prolonged diarrhea, those with a CD4 cell count less than 100 and those not taking HAART (60).

When initiating a diarrhea workup on HIV-infected patients the most important point to remember is that for each patient the workup should be individualized. Patient characteristics must be carefully evaluated with particular attention paid to the patient's CD4 cell count. After all, as mentioned, newly diagnosed patients with high HIV viral loads and suppressed CD4 cell counts are vastly different than HAART-treated patients whose viral loads are significantly suppressed and whose CD4 cell count has never been below 200.

The intensity of the diagnostic investigation of in both immunosuppressed and non-immunosuppressed HIV patients with chronic diarrhea remains controversial though little has been published post-HAART. The strongest evidence for a minimalist approach to the workup comes from an initial cost-benefit analysis that was published in 1990 (67). The authors concluded that evaluation with stool culture alone was associated with the same diarrhea remission rate and was more cost effective than a more extensive evaluation that included stool culture, ova and parasite exam, blood cultures, upper endoscopy and colonoscopy with biopsies. They suggested that endoscopy be reserved for patients who fail to respond to non-specific antidiarrheal therapy. This early study has been criticized because it assumed that diarrhea would resolve in 67% of patients treated with diphenoxylate regardless of the cause, did not take into account many treatable enteropathogens, and the effectiveness of many therapies. Subsequent studies have proved that more extensive workups are more effective, and most experts recommend an intensive approach including endoscopic examination in most patients with advanced HIV disease (60,68,69). When the workup results in a diagnosis there is no guarantee that the etiology discovered is etiologic in the patient's symptoms, nor does it guarantee a response to treatment. Even when an enteropathogen is found and treated, the benefit may not outweigh the cost and discomfort of an extensive evaluation (70).

Regardless of the extent of evaluation, when an HIV-positive patient presents with diarrhea, standard medical practice is the best way to approach the problem. A thorough history and physical examination are first, paying particular attention to travel history, medications that can cause diarrhea, and diet (e.g. lactose). In some patients despite intensive investigations, no cause will be identified. Patients in whom no diagnosis can be established after an extensive workup often have spontaneous resolution of their symptoms and better survival than patients who have a pathogen identified (45,60). It is not uncommon to find more than one pathogen at a time. Therefore if a cause of diarrhea is identified and treated, but the patient is not responding, another etiology is likely and the clinician should keep looking. Further, it is important to note that non-infectious etiologies such as fat malabsorption and medication-induced diarrhea may contribute to the symptoms in this patient group as well. Treatment of both infectious and non-infectious contributants will improve response rates.

In the case of non-immunosuppressed patient with controlled HIV replication, there is no published data upon which to base the intensity of evaluation. It is still likely that multiple etiologies for diarrhea will be identified by the evaluation, but it is less likely that these etiologies will be opportunistic infections. It is also not entirely clear whether performing the same type of evaluation, as the immunosuppressed patients will result in improved outcome, or decreased health care costs. Cost-effective analysis must be repeated on this patient subset. With this in mind, repeated use of expensive endoscopic procedures may represent an ineffective use of resources.

Laboratory tests, serologies, culture, imaging

Stool studies initiate the diagnostic evaluation. Three sets of stool samples should be tested for standard bacterial pathogens, ova and parasites, Clostridium difficile toxin, and special stains for Cryptosporidium (modified acid fast stain, direct fluorescent antibody (DFA), enzyme-linked immunosorbent assay (ELISA)) and Microsporidia (chromotrope-based techniques, "Fungi-fluor" stains and a stool PCR assay that also allows for speciation (71). If Giardia is a consideration then DFA staining or detection of Giardia antigens in the stool by ELISA should be undertaken. While sending three sets of stool samples is routine, examination of a single stool specimen was found to detect 96% of cases of cryptosporidiosis among hospitalized HIV patients (72). Acid fast staining, or mycobacterial culture for diagnosis of MAC are generally not recommended as positive findings may indicate colonization in the aerodigestive tree and not true infection. Testing for fecal leukocytes, using methylene blue may be helpful in localizing the diarrhea to a colonic origin as small bowel pathogens due not usually result in significant mucosal inflammation. In addition, 24-hour stool samples should be submitted for determination of stool weight and quantification of stool fat. Optimally, stool fat should be measured on the third day after a 100-gram fat diet for two days. In order to ensure that the patient follows this diet, he should eat a fast food hamburger and French fries twice daily for the two days prior to stool collection. Another early diagnostic step should be testing for small bowel bacterial overgrowth. This may be accomplished with a lactulose or similar breath test.

Ancillary serologic testing such as albumin or prealbumin will help in the assessment of the patient's nutritional status. In addition, the serum carotene may be examined to assess the degree of malabsorption.

Invasive Diagnostic tests

In the immunosuppressed HIV-infected patient, endoscopy plays a vital role in the evaluation of chronic diarrhea. The medical history will help guide the endoscopist in determining the workup to be undertaken. If it is not clear whether investigation of the upper or lower gastrointestinal tract should be performed first, both may be performed on the same day. Patients with classic small bowel diarrhea should have upper endoscopy first with small bowel biopsy from the distal duodenum or proximal jejunum. Aspirates and cytological brushings as well as biopsy may be performed, though the routine performance of aspiration of duodenal contents is not substantiated in the literature (73). Whether or not the cost of performance of electron microscopy on duodenal biopsies for improved detection of Microsporidia is debatable and I personally feel is not necessary (70). It is unclear whether enteroscopy with more distal jejunal biopsies will result in greater diagnostic yield, though some pathogens such as microsporidia are more prevalent in the jejunum.

If the history suggests colitis, then endoscopy of the lower tract should be performed. The choice of lower endoscopic procedure is controversial. Recent studies suggest that the yield of flexible sigmoidoscopy is adequate for detection of enteric pathogens, with colonoscopy reserved for patients in whom flexible sigmoidoscopy fails to reveal the etiology but in whom colitis is likely (60, 74). The authors have published data (75) that suggest sigmoidoscopy alone would miss at least 33% of CMV colitis. Instead of subjecting a patient to two preparations, as might occur if initial performance of a flexible sigmoidoscopy fails to reveal a diagnosis, colonoscopy may be preferable.

Regardless of the endoscopic procedure performed, multiple biopsies should be performed for histopathology, and possibly mycobacterial culture, even if mucosa appears normal, because MAC and CMV can be found in grossly normal-appearing mucosa. Bacterial and viral culturing of biopsy material does not appear to be worthwhile as the overall yield is low (76). Biopsies should be examined assiduously by a pathologist trained to identify microsporidia and CMV to avoid the need for expensive diagnostic techniques such as electron microscopy (77).

Management

Empiric therapy

Since diarrhea exacts an enormous toll on the patient's quality of life, rapid alleviation of their symptoms should be a primary concern. Immediately after stool studies have been obtained, the patient should be vigorously treated for their diarrhea. In addition to use of non-specific antidiarrheals (see below), it may be helpful to initiate empiric therapy for bacterial pathogens, if small bowel bacterial overgrowth or bacterial colitis is possible. Given the high incidence of steatorrhea found among HIV-infected patients, it seems worthwhile to consider initiating a low fat diet and perhaps use of pancreatic enzyme replacement, though this is not supported by any literature at this time. Lactose-free diets can also help. If the patient has suffered from malabsorption and has experienced significant weight loss, a restrictive diet may not be ideal.

Non-antimicrobial agent therapy

The most effective therapy for HIV-related diarrhea in the severely immunosuppressed patient is control of their HIV by HAART (78). At this time it is not possible to determine whether the benefit is achieved by suppression of HIV itself or the subsequent immune reconstitution, but use of protease inhibitor-based therapy is associated with a significant decrease in stool frequency and diarrhea recurrence (79).

If diarrhea remains with or without improved HIV control and whether or not an etiology for diarrhea is established, symptomatic treatment is necessary for the patients' quality of life. There are three categories of antidiarrheal agents: luminal agents, antimotility agents, and hormones. Luminal agents may be considered, but are seldom effective enough to be used alone. Fiber, one such agent, seemingly paradoxically, takes up water and builds bulk in the stool when there is diarrhea. Another luminal agent, cholestyramine, binds bile acids which can stimulate fluid secretion. Most antimotility agents reduce diarrhea by increasing transit time, thus prolonging contact with the absorptive surface. Loperamide is an over-the-counter drug that is useful as a first line agent. Lomotil, a combination of diphenoxylate and atropine, can also be helpful, though high doses may result in unwanted anticholinergic effects. If stronger remedies are required, opiates can be utilized, and are usually quite effective. Codeine, morphine, methadone, paregoric, and deodorized tincture of opium are all appropriate. Higher doses than are recommended in the package insert may be necessary to control devastating HIV-related diarrhea. If these do not provide adequate relief, then hormonal therapy can be attempted. Octreotide is the synthetic version of the natural somatostatin. It slows transit time, decreases active secretion, and inhibits all gastrointestinal hormones. In one study (80) it was shown to significantly decrease stool frequency and volume compared with placebo. More recent studies have shown it to be no more effective than placebo, whether or not an enteric pathogen was present (81). It must be administered subcutaneously and can be associated with a number of side effects, but in some patients appears to be quite useful. Idiopathic colonic inflammation or ulceration in HIV infection may respond to corticosteroid therapy (82) or to a trial of mesalamine therapy (personal communication).

Equally important is ensuring adequate nutritional status. Fluid replacement should be started early to prevent excessive dehydration. This should be combined with nutritional support to prevent malnutrition. If only colonic disease is present and diarrhea is manageable, oral supplementation is both possible and effective. Dietary counseling is mandatory, however, at any level of disease. Psychosocial support, including counseling, for both the patient and the caring relatives, is required to alleviate anxiety, particularly when the diarrhea becomes intractable.

Conclusions

In the new millennium, we can no longer consider all HIV patients equal. There is a vast dichotomy between those severely immunosuppressed AIDS patients, more commonly seen in areas where HAART therapy is either not used or nor taken correctly, and those patients with suppressed HIV viral replication and no significant immunosuppression. While diarrhea is seen in both of these patient groups, the causes differ. Opportunistic bacterial, protozoal and viral causes of diarrhea predominate in the immunosuppressed patients, as well as non-infectious processes such as steatorrhea, effects of medications, and small bowel bacterial overgrowth. A vigorous workup that relies heavily on endoscopic biopsies is necessary in this group. Opportunistic causes of diarrhea are less likely in patients with controlled HIV replication and non-infectious etiologies play a more important part. The role of endoscopy in this group has not been adequately defined. Regardless of the patient group or the etiology of diarrhea aggressive treatment is necessary to ensure maintenance of quality of life and reduce health care utilization.

Tables and Figures

Table 1. Common Etiologies for Diarrhea in HIV-infected Patients [Download PDF]

Figure 1. Proposed Diagnostic Workup

REFERENCES:

1. Colebunders R, Francis H, Mann JM, et al Persistent diarrhea, strongly associated with HIV infection in Kinshasa, Zaire. Am J Gastroenterol 1987;82:859-64 [Pub Med]

2. Smith PD, Lane HC, Gill VJ, Manischewitz JF, Quinnan GV, Fauci AS, Masur H. Intestinal infections in patients with the acquired immunodeficiency syndrome (AIDS). Etiology and response to therapy. Ann Intern Med 1988;108:328-33. [Pub Med]

3. Rene E, Marche C, Regnier B, et al. Intestinal infections in patients with acquired immunodeficiency syndrome. A prospective study in 132 patients. Dig Dis Sci 1989;34:773-80. [Pub Med]

4. Anastasi JK, Capili B. HIV and diarrhea in the era of HAART: 1998 New York state hospitalizations. Am J Infect Control 2000;28:262-6. [Pub Med]

5. Monkemuller KE, Call SA, Lazenby AJ, Wilcox CM Declining prevalence of opportunistic gastrointestinal disease in the era of combination antiretroviral therapy. Am J Gastroenterol 2000;95:457-62. [Pub Med]

6. Weber R, Ledergerber B, Zbinden R, et al. Enteric infections and diarrhea in human immunodeficiency virus-infected persons: prospective community-based cohort study. Swiss HIV Cohort Study. Arch Intern Med 1999;159:1473-80. [Pub Med]

7. Lubeck DP, Bennett CL, Mazonson PD, Fifer SK, Fries JF. Quality of life and health service use among HIV-infected patients with chronic diarrhea. J Acquir Immune Defic Syndr 1993;6:478-84. [Pub Med]

8. Corazza GR, Ginaldi L, Furia N, Marani-Toro G, Di Giammartino D, Quaglino D. The impact of HIV infection on lactose absorptive capacity. J Infect 1997;35:31-5. [Pub Med]

9. Ristola MA, von Reyn CF, Arbeit RD, et al. High rates of disseminated infection due to non-tuberculous mycobacteria among AIDS patients in Finland. J Infect 1999;39:61-7. [Pub Med]

10. Chan IS, Neaton JD, Saravolatz LD, Crane LR, Osterberger J. Frequencies of opportunistic diseases prior to death among HIV-infected persons. Community Programs for Clinical Research on AIDS. AIDS 1995;9:1145-51. [Pub Med]

11. Casado JL, Valdezate S, Calderon C, Navas E, Frutos B, Guerrero A, Martinez-Beltran J. Zidovudine therapy protects against Salmonella bacteremia recurrence in human immunodeficiency virus-infected patients. J Infect Dis 1999;179:1553-6. [Pub Med]

12. Nelson MR, Shanson DC, Hawkins DA, et al. Salmonella, Campylobacter and Shigella in HIV-seropositive patients. AIDS 1992;6:1495. [Pub Med]

13. Tacconelli E, Tumbarello M, Ventura G, Leone F, Cauda R, Ortona L. Risk factors, nutritional status, and quality of life in HIV-infected patients with enteric salmonellosis. Ital J Gastroenterol Hepatol 1998;30:167-72. [Pub Med]

14. Mathewson JJ, Salameh BM, DuPont HL, Jiang ZD, Nelson AC, Arduino R, Smith MA, Masozera N. HEp-2 cell-adherent Escherichia coli and intestinal secretory immune response to human immunodeficiency virus (HIV) in outpatients with HIV-associated diarrhea. Clin Diagn Lab Immunol 1998;:87-90. [Pub Med]

15. Schultsz C, Snijders F, Dankert J. No evidence for a role of enteroadherent Escherichia coli in diarrhea in human immunodeficiency virus-infected patients. J Infect Dis 1996;174:246. [Pub Med]

16. Mainardi JL, Lacassin F, Guilloy Y, Goldstein FW, Leport C, Acar JF, Vilde JL. Low rate of Clostridium difficile colonization in ambulatory and hospitalized HIV-infected patients in a hospital unit: a prospective survey. J Infect 1998;37:108-11. [Pub Med]

17. DeMarais PL, Gertzen J, Weinstein RA. Nosocomial infections in human immunodeficiency virus-infected patients in a long-term-care setting. Clin Infect Dis 1997;25:1230-2. [Pub Med]

18. Barbut F, Meynard JL, Guiguet M, Avesani V, Bochet MV, Meyohas MC, Delmee M, Tilleul P, Frottier J, Petit JC. Clostridium difficile-associated diarrhea in HIV-infected patients: epidemiology and risk factors. J Acquir Immune Defic Syndr Hum Retrovirol 1997;16:176-81. [Pub Med]

19. Cappell MS, Philogene C. Clostridium difficile infection is a treatable cause of diarrhea in patients with advanced human immunodeficiency virus infection: a study of seven consecutive patients admitted from 1986 to 1992 to a university teaching hospital. Am J Gastroenterol 1993;88:891-7.84. [Pub Med]

20. Lu SS, Schwartz JM, Simon DM, Brandt LJ. Clostridium difficile-associated diarrhea in patients with HIV positivity and AIDS: a prospective controlled study. Am J Gastroenterol 1994;89:1226-9 [Pub Med]

21. Lumadue JA, Manabe YC, Moore RD, Belitsos PC, Sears CL, Clark DP. A clinicopathologic analysis of AIDS-related cryptosporidiosis. AIDS 1998 Dec 24;12(18):2459-66. [Pub Med]

22. Lima AA, Silva TM, Gifoni AM, Barrett LJ, McAuliffe IT, Bao Y, Fox JW, Fedorko DP, Guerrant RL. Mucosal injury and disruption of intestinal barrier function in HIV-infected individuals with and without diarrhea and cryptosporidiosis in northeast Brazil. Am J Gastroenterol 1997;92:1861-6. [Pub Med]

23. Connolly GM, Dryden MS, Shanson DC, Gazzard BG. Cryptosporidial diarrhoea in AIDS and its treatment. Gut 1988;29:593-7. [Pub Med]

24. Flanigan T, Whalen C, Turner J, Soave R, Toerner J, Havlir D, Kotler D. Cryptosporidium infection and CD4 counts. Ann Intern Med 1992;116:840-2. [Pub Med]

25. Rabeneck L, Gyorkey F, Genta RM, Gyorkey P, Foote LW, Risser JM. The role of Microsporidia in the pathogenesis of HIV-related chronic diarrhea. Ann Intern Med 1993 1;119:895-9. [Pub Med]

26. Coyle CM, Wittner M, Kotler DP, Noyer C, Orenstein JM, Tanowitz HB, Weiss LM. Prevalence of microsporidiosis due to Enterocytozoon bieneusi and Encephalitozoon (Septata) intestinalis among patients with AIDS-related diarrhea: determination by polymerase chain reaction to the microsporidian small-subunit rRNA gene. Clin Infect Dis.1996;23:1002-6. [Pub Med]

27. Conteas CN, Berlin OG, Lariviere MJ, Pandhumas SS, Speck CE, Porschen R, Nakaya T. Examination of the prevalence and seasonal variation of intestinal microsporidiosis in the stools of persons with chronic diarrhea and human immunodeficiency virus infection. Am J Trop Med Hyg 1998;58:559-61. [Pub Med]

28. Orenstein JM, Dieterich DT, Kotler DP. Systemic dissemination by a newly recognized intestinal microsporidia species in AIDS. AIDS 1992;6:1143-50. [Pub Med]

29. Muller A, Stellermann K, Hartmann P, Schrappe M, Fatkenheuer G, Salzberger B, Diehl V, Franzen C. A powerful DNA extraction method and PCR for detection of microsporidia in clinical stool specimens. Clin Diagn Lab Immunol 1999;6:243-6. [Pub Med]

30. Schmidt W, Schneider T, Heise W, Schulzke JD, Weinke T, Ignatius R, Owen RL, Zeitz M, Riecken EO, Ullrich R. Mucosal abnormalities in microsporidiosis. AIDS 1997;11:1589-94. [Pub Med]

31. DeHovitz JA, Pape JW, Boncy M, Johnson WD Jr. Clinical manifestations and therapy of Isospora belli infection in patients with the acquired immunodeficiency syndrome. N Engl J Med 1986 ;315:87-90. [Pub Med]

32. Curry A, Smith HV. Emerging pathogens: Isospora, Cyclospora and microsporidia. Parasitology 1998;117 Suppl:S143-59. [Pub Med]

33. Dieterich DT, Lew EA, Bacon DJ, Pearlman KI, Scholes JV. Gastrointestinal pneumocystosis in HIV-infected patients on aerosolized pentamidine: report of five cases and literature review. Am J Gastroenterol 1992;87:1763-70. [Pub Med]

34. Bonacini M, Kanel G, Alamy M. Duodenal and hepatic toxoplasmosis in a patient with HIV infection: review of the literature. Am J Gastroenterol 1996;91:1838-40. [Pub Med]

35. Chachoua A, Dieterich D, Krasinski K, Greene J, Laubenstein L, Wernz J, Buhles W, Koretz S. 9-(1,3-Dihydroxy-2-propoxymethyl)guanine (ganciclovir) in the treatment of cytomegalovirus gastrointestinal disease with the acquired immunodeficiency syndrome. Ann Intern Med 1987;107:133-7. [Pub Med]

36. Wilcox CM, Chalasani N, Lazenby A, Schwartz DA. Cytomegalovirus colitis in acquired immunodeficiency syndrome: a clinical and endoscopic study. Gastrointest Endosc 1998;48:39-43. [Pub Med]

37. Gallant JE, Moore RD, Richman DD, Keruly J, Chaisson RE. Incidence and natural history of cytomegalovirus disease in patients with advanced human immunodeficiency virus disease treated with zidovudine. The Zidovudine Epidemiology Study Group. J Infect Dis 1992;166:1223-7. [Pub Med]

38. Janoff EN, Orenstein JM, Manischewitz JF, Smith PD. Adenovirus colitis in the acquired immunodeficiency syndrome. Gastroenterology 1991;100:976-9. [Pub Med]

39. Laughon BE, Druckman DA, Vernon A, Quinn TC, Polk BF, Modlin JF, Yolken RH, Bartlett JG. Prevalence of enteric pathogens in homosexual men with and without acquired immunodeficiency syndrome. Gastroenterology 1988;94:984-93. [Pub Med]

40. Giordano MO, Martinez LC, Rinaldi D, Espul C, Martinez N, Isa MB, Depetris AR, Medeot SI, Nates SV. Diarrhea and enteric emerging viruses in HIV-infected patients. AIDS Res Hum Retroviruses 1999 1;15:1427-32. [Pub Med]

41. Khoo SH, Bailey AS, de Jong JC, Mandal BK. Adenovirus infections in human immunodeficiency virus-positive patients: clinical features and molecular epidemiology. J Infect Dis 1995;172:629-37. [Pub Med]

42. Sabin CA, Clewley GS, Deayton JR, Mocroft A, Johnson MA, Lee CA, McLaughlin JE, Griffiths PD Shorter survival in HIV-positive patients with diarrhoea who excrete adenovirus from the GI tract. J Med Virol 1999;58:280-5. [Pub Med]

43. Yan Z, Nguyen S, Poles M, Melamed J, Scholes JV. Adenovirus colitis in human immunodeficiency virus infection: an underdiagnosed entity. Am J Surg Pathol 1998;22:1101-6. [Pub Med]

44. Grohmann GS, Glass RI, Pereira HG, Monroe SS, Hightower AW, Weber R, Bryan RT. Enteric viruses and diarrhea in HIV-infected patients. Enteric Opportunistic Infections Working Group. N Engl J Med 1993 1;329:14-20. [Pub Med]

45. Blanshard C, Gazzard BG. Natural history and prognosis of diarrhoea of unknown cause in patients with acquired immunodeficiency syndrome (AIDS). Gut 1995;36:283-6. [Pub Med]

46. Cerf-Bensussan N, Guy-Grand D. Intestinal intraepithelial lymphocytes. Gastroenterol Clin North Am 1991; 20:549-576. [Pub Med]

47. Anton PA, Elliott J, Poles MA, et al. Enhanced levels of functional HIV-1 co-receptors on human mucosal T cells demonstrated using intestinal biopsy tissue. AIDS. 2000 (in press). [Pub Med]

48. Ullrich R, Schieferdecker HL, Ziegler K, Riecken EO, Zeitz M. Gamma delta T cells in the human intestine express surface markers of activation and are preferentially located in the epithelium. Cell Immunol 1990; 128:619-627. [Pub Med]

49. Veazey RS, DeMaria M, Chalifoux LV, et al. Gastrointestinal tract as a major site of CD4+ T cell depletion and viral replication in SIV infection. Science 1998; 280:427-431. [Pub Med]

50. Heise C, Miller CJ, Lackner A, Dandekar S. Primary acute simian immunodeficiency virus infection of intestinal lymphoid tissue is associated with gastrointestinal dysfunction. J.Infect.Dis 1994; 169:1116-1120. [Pub Med]

51. Poles M, Elliott J, Brown S, et al. HIV-1 is detectable in mucosal biopsies in patients with undetectable plasma viral loads. Digestive Disease Week. Orlando, May 1999 [abstract G3461]. [Pub Med]

52. Olsson J, Poles M, Spetz A,L, et al. HIV infection is associated with significant mucosal inflammation, characterized by increased CCR5, CXCR4 receptor and b?chemokine expression (in press). [Pub Med]

53. Kotler DP, Reka S, Clayton F. Intestinal mucosal inflammation associated with human immunodeficiency virus infection. Dig Dis Sci 1993; 38:1119-1127. [Pub Med]

54. McGowan I, Radford-Smith G, Jewell DP. Cytokine gene expression in HIV-infected intestinal mucosa. AIDS 1994;8:1569-75. [Pub Med]

55. Sharkey KA, Sutherland LR, Davison JS, Zwiers H, Gill MJ, Church DL. Peptides in the gastrointestinal tract in human immunodeficiency virus infection. The GI/HIV Study Group of the University of Calgary. Gastroenterology 1992;103:18-28. [Pub Med]

56. Koch J, Garcia-Shelton YL, Neal EA, et al. Steatorrhea: A common manifestation in patients with HIV/AIDS. Nutrition 1996;12:507-510. [Pub Med]

57. Poles MA, Fuerst M, McGowan I, et al. HIV-Related Diarrhea is Multifactorial and Fat Malabsorption is a Leading Cause, Independent of HAART (in press). [Pub Med]

58. Wilcox CM, Waites KB, Smith PD. No relationship between gastric pH, small bowel bacterial colonisation, and diarrhoea in HIV-1 infected patients. Gut 1999;44:101-5. [Pub Med]

59. Rabeneck L, Crane MM, Risser JM, Lacke CE, Wray NP. Effect of HIV transmission category and CD4 count on the occurrence of diarrhea in HIV-infected patients. Am J Gastroenterol 1993;88:1720-3. [Pub Med]

60. Wilcox CM, Schwartz DA, Cotsonis G, Thompson SE 3rd. Chronic unexplained diarrhea in human immunodeficiency virus infection: determination of the best diagnostic approach. Gastroenterology1996;110:30-7 [Pub Med]

61. Caputo C, Forbes A, Frost F, Sinclair MI, Kunde TR, Hoy JF, Fairley CK. Determinants of antibodies to Cryptosporidium infection among gay and bisexual men with HIV infection. Epidemiol Infect 1999;122:291-7. [Pub Med]

62. Davis LJ, Roberts HL, Juranek DD, Framm SR, Soave R. A survey of risk factors for cryptosporidiosis in New York City: drinking water and other exposures. Epidemiol Infect 1998;121:357-67. [Pub Med]

63. Glaser CA, Safrin S, Reingold A, Newman TB. Association between Cryptosporidium infection and animal exposure in HIV-infected individuals. J Acquir Immune Defic Syndr Hum Retrovirol 1998 1;17:79-82. [Pub Med]

64 . Hutin YJ, Sombardier MN, Liguory O, Sarfati C, Derouin F, Modai J, Molina JM. Risk factors for intestinal microsporidiosis in patients with human immunodeficiency virus infection: a case-control study. J Infect Dis 1998;178:904-7. [Pub Med]

65. Esfandiari A, Swartz J, Teklehaimanot S. Clustering of giardiosis among AIDS patients in Los Angeles County. Cell Mol Biol 1997;43:1077-83. [Pub Med]

66. Dieterich DT. Ganciclovir treatment of cytomegalovirus gastrointestinal disease in patients with AIDS. In: Spector S, ed. Ganciclovir therapy for cytomegalovirus infection. New York: Marcel Dekker, 1991:129-144. [Pub Med]

67. Johanson JF, Sonnenberg A. Efficient management of diarrhea in the acquired immunodeficiency syndrome (AIDS). A medical decision analysis. Ann Intern Med 1990 15;112:942-8. [Pub Med]

68. Blanshard C, Francis N, Gazzard BG. Investigation of chronic diarrhoea in acquired immunodeficiency syndrome. A prospective study of 155 patients. Gut 1996;39:824-32. [Pub Med]

69. Wilcox CM, Rabeneck L, Friedman S. AGA technical review: malnutrition and cachexia, chronic diarrhea, and hepatobiliary disease in patients with human immunodeficiency virus infection. Gastroenterology 1996;111:1724-52. [Pub Med]

70. Bini EJ, Kornacki S. Electron microscopic examination of small bowel biopsies in HIV-positive patients with chronic diarrhea: Is it worth the added cost? Gastroenterology 1997; 112; A935. [Pub Med]

71. Liguory O, David F, Sarfati C, Schuitema ARJ, Hartskeerl RA, Derouin F, Modai J, Molina JM. Diagnosis of infections caused by Enterocytozoon bieneusi and Encephalitozoon intestinalis using polymerase chain reaction in stool specimens. AIDS 1997;11:723-726. [Pub Med]

72. Blackman E, Binder S, Gaultier C, Benveniste R, Cecilio M. Cryptosporidiosis in HIV-infected patients: diagnostic sensitivity of stool examination, based on number of specimens submitted. Am J Gastroenterol 1997 Mar;92(3):451-3. [Pub Med]

73. Bown JW, Savides TJ, Mathews C, Isenberg J, Behling C, Lyche KD. Diagnostic yield of duodenal biopsy and aspirate in AIDS-associated diarrhea. Am J Gastroenterol 1996;91:2289-92. [Pub Med]

74. Kearney DJ, Steuerwald M, Koch J, Cello JP. A prospective study of endoscopy in HIV-associated diarrhea. Am J Gastroenterol 1999;94:596-602. [Pub Med]

75. Dieterich DT, Rahmin M. Cytomegalovirus colitis in AIDS: presentation in 44 patients and a review of the literature. J Acquir Immmune Defic Syndr. 1991;4(Suppl 1):S29-S35. [Pub Med]

76. Beaugerie L, Salauze B, Bure A, Deluol AM, Hoyeau-Idrissi N, Carbonnel F, Ngo Y, Cosnes J, Rozenbaum W, Nicolas JC, Gendre JP. Results of culture form colonoscopically obtained specimens for bacteria and fungi in HIV-infected patients with diarrhea. Gastrointest Endosc 1996;44:663-6. [Pub Med]

77. Olofinlade O, Adeonigbagbe O, Karowe M, Robilotti J, Gualtieri N, Cacciarelli A, Berlin I, Guerrieri C. The diagnostic value of electron microscopy in human immunodeficiency virus-positive patients with gastrointestinal disease. Scand J Gastroenterol 2000;35:329-32. [Pub Med]

78. Maggi P, Larocca AM, Quarto M, Serio G, Brandonisio O, Angarano G, Pastore G. Effect of antiretroviral therapy on cryptosporidiosis and microsporidiosis in patients infected with human immunodeficiency virus type 1. Eur J Clin Microbiol Infect Dis 2000;19:213-7. [Pub Med]

79. Bini EJ, Cohen J. Impact of protease inhibitors on the outcome of human immunodeficiency virus-infected patients with chronic diarrhea. Am J Gastroenterol 1999;94:3553-9. [Pub Med]

80. Cello JP, Grendell JH, Basuk et al. Effect of octreotide on refractory AIDS-associated diarrhea. Ann Intern Med. 1991;115:705-710. [Pub Med]

81. Simon DM, Cello JP, Valenzuela J, et al. Multicenter trial of octreotide in patients with refractory acquired immunodeficiency syndrome-associated diarrhea. Gastroenterology 1995;108:1753-60. [Pub Med]

82. Gopal DV, Hassaram S, Marcon NE, Kandel G. Idiopathic colonic inflammation in AIDS: an open trial of prednisone. Am J Gastroenterol 1997;92:2237-40.[Pub Med]