Cryptosporidiosis (sometimes called 'crypto' for short - but not to be confused with cryptococcal meningitis) was first reported in humans in 1976. It is caused by the parasite Cryptosporidium parvum, which has been found in 10-15% of people with AIDS in the United States and 30 to 50% in Africa and Haiti (Petersen 1992).

Symptoms and diagnosis

Cryptosporidium lives on the lining of the gastro-intestinal tract, especially the small bowel. It appears to stimulate the secretion of fluids into the intestines and to reduce the absorption of food. Symptoms usually begin from five to fourteen days after infection. These may include voluminous watery diarrhoea, cramping abdominal pain, loss of appetite, weight loss, nausea, vomiting and excessive wind. This causes poor nutrition through the loss of nutrients as well as dehydration.

Cryptosporidium can also attack the tubes from the liver (the biliary tract), causing abdominal pain and jaundice in about 15% of people with cryptosporidiosis. There have also been a few cases of cryptosporidiosis of the lungs in people with AIDS, which resembles PCP, and in the nasal passages.

Most people with intestinal cryptosporidiosis do not develop a fever (although fever can be a symptom of biliary cryptosporidiosis). An individual with both diarrhoea and a fever may be more likely to have MAI.

Cryptosporidiosis is normally diagnosed by testing a sample of stool (faeces) for the organisms or performing a rectal biopsy, a relatively simple and painless procedure. Diagnosis may be made by rectal or intestinal biopsy.

Avoiding infection

Cryptosporidiosis may result from being exposed to the organism for the first time, or from the activation of latent infection. The main routes of transmission are by contact with the faeces of infected people or animals, including during oral-anal sex; by hand-to-mouth spread from surfaces that have been contaminated by infected faeces; by swallowing contaminated water in a swimming pool or lake or drinking contaminated water, or by eating contaminated uncooked food.

It is possible to take steps to avoid infection. However, it is important to remember that a significant proportion of people with AIDS are already latently infected with Cryptosporidium. All the following points are the `counsel of perfection' and need not always be followed to the letter. Also, cryptosporidiosis is, rarely, reported in people with immune restoration on HAART, so these guidelines may be less relevant for people receiving successful HAART.

Domestic water supplies do become contaminated from time to time, and warnings from water companies often come too late. To ensure that it is safe, tap water should be boiled for one minute or passed through a filter capable of removing particles less than one micron in size. Water that has been treated this way should then be kept in the fridge and discarded after 24 hours. Bottled water is not necessarily free from Cryptosporidium.

People with symptomatic HIV infection should avoid farm animals and their excreta, which includes animal manure for gardening. Organically grown vegetables are often grown in raw manure, so it makes sense to brush off as much dirt as possible and then clean any vegetable that isn't peeled with clean water. It may be best to avoid unpasteurised milk and milk products.

It is not clear whether domestic pets might pose a similar risk. Use gloves when cleaning up after them and emptying litter trays.

Children with diarrhoea due to cryptosporidiosis should be kept away from other children until the diarrhoea has been controlled.

Cryptosporidium is notoriously resistant to many disinfectants; for example, it can survive in infectious form in the chlorinated water of swimming pools, and can even resist exposure to full-strength household bleach for two hours. However, heating to at least 65 degrees Centigrade for more than 30 minutes will kill the organisms.

Treatment

In people with healthy immune systems, the organisms are usually eradicated by the immune system after one or two weeks. HIV-positive people with CD4 cell counts above 300 cells/mm³ are also quite likely to have cryptosporidiosis that clears up of its own accord (regardless of treatment) within about four weeks, while people with counts below 150 cells/mm³ almost always have persistent cryptosporidiosis.

There have now been numerous case reports in which people with cryptosporidiosis have experienced dramatic improvement or resolution of symptoms after starting a highly active anti-HIV regimen, such as combination therapy including a protease inhibitor. Since there are no proven specific treatments for cryptosporidiosis, starting or switching to a potent anti-HIV regimen appears to be the best available approach to treating the condition.

The development of specific anti-cryptosporidial agents is not easy because it has a waxy membrane which prevents drugs getting in. To date over 70 different drugs have been tested against cryptosporidiosis with only limited success. It is also difficult to be clear from the studies that have taken place whether apparently beneficial drugs simply stopped the diarrhoea without eradicating the Cryptosporidium organisms or whether they actually eradicated it.

Two drugs have clearly failed in the treatment of cryptosporidiosis - oral spiramycin and diclazuril. Other agents that have been studied fall into two categories - antibiotics like azithromycin, paromomycin, roxithromycin, atovaquone and letrazuril, and immune boosters or substitutes, such as interleukin-2, bovine colostrum, or DLE. None of these has shown unequivocal benefits in controlled clinical trials.

There have been reports that people receiving clarithromycin or rifabutin as MAI prophylaxis may be at a reduced risk of developing cryptosporidiosis, and that azithromycin may also have a protective effect. However, other retrospective analyses have found no evidence of any protective effect of azithromycin or clarithromycin, and to date no controlled studies have been conducted.

In the USA, the FDA has approved an expanded access programme for nitazoxanide (NTZ), owned by Romark Laboratories, which has shown promise in alleviating the symptoms of cryptosporidiosis; a trial in Mexico found that 100% of people who took high doses of the drug experienced resolution of their diarrhoea, after an average of 11 days. In phase I studies of several different doses in the USA, between 55 and 75% of participants reported reduction or resolution of diarrhoea.

A study in adults and children without HIV infection in Egypt showed an 80% clearance rate after seven days in those who received nitazoxanide, compared to 40% in the placebo group. In people with AIDS in Mexico, diarrhoea stopped and parasites were eliminated within 14 days in two thirds of the nitazoxanide group.

A placebo-controlled study in children in Zambia found that whilst a three day course of nitazoxanide was generally effective in HIV negative children (52% vs 28% of the placebo group cleared cryptosporidium infection within ten days of the first dose), a single course of treatment was no more effective than a placebo in HIV-infected children. A second open label course of treatment resulted in some cases of clearance, but the results continued to be disappointing (Amadi 2002). Response to treatment in children with HIV was associated with a higher CD4 cell count.

Nitazoxanide is already approved for treatment of protozoal infections in most Latin American countries, and is marketed under the trade names Daxon and Colufase.

A high-dose garlic concentrate called allicin has reported promising results in uncontrolled pilot studies.

There are also options that can help to control the symptoms of cryptosporidiosis, although these do not address the underlying cause. Anti-diarrhoea drugs like imodium, codeine and slow-release morphine may be prescribed. Somatostatin is an anti-diarrhoea treatment given by subcutaneous pump. These are very safe drugs and the art is to take about as much as is needed to improve the diarrhoea without causing constipation and cramps. In addition, pain-relieving agents and anti-spasmodic agents may be used for the colic which may also improve the diarrhoea.

Food supplements that are predigested may be helpful: these are not to everyone's taste but may help prevent weight loss. Fluid and electrolyte replacement is even more important than food replacement in the short term - volumes of diarrhoea as high as 17 litres/day have been reported. Both food and fluid are sometimes given via a central venous line although a PEG tube, inserted directly into the stomach, is more commonly used. Good nursing and general care are especially important.

Research on epidemiology

Vakil reported that contamination of the water supply resulted in 400,000 cases of cryptosporidiosis in Milwaukee in April 1993. 82 HIV-infected people in the area developed cryptosporidiosis. A CD4 count below 50 occurred in 88% of the patients with biliary symptoms versus 63% of those without (p=0.03) indicating that people with CD4 counts below 50 had an increased risk of developing biliary disease. At one year, 4/24 (17%) of patients with bilary symptoms and 30/58 (52%) of those without biliary symptoms were alive (p=0.003).

Samson reported that the use of a water filter with a pore diameter of less than 1 micron (but not filters with larger pores) was associated with a reduced risk of diarrhoea during this outbreak.

Flanigan (1992) reported on 13 out of 47 HIV seropositive people with diarrhoea and the presence of Cryptosporidium oocysts who had self-limiting cryptosporidiosis. All 8 patients with CD4 counts of 180 or more cleared the infection spontaneously without antiretroviral therapy within 7 days to 1 month. 5 out of 39 people with CD4 counts less than 180 had self-limited disease, of whom 4/5 were started on AZT therapy at the time cryptosporidiosis was diagnosed. In patients with CD4 counts of less than 140, 87% (n=34) developed persistent disease.

McGowan reported that 11 out of 38 HIV seropositive patients with diarrhoea found to be shedding Cryptosporidium oocysts had a spontaneous clinical remission. The mean lymphocyte count in patients with remission was significantly higher (119) than in the non-remission group (62). These findings have important implications for the design of anticryptosporidial therapeutic trials, where the possibility of spontaneous clinical remission must be considered.

Response to anti-HIV therapy

Benhamou reported the clearance of Cryptosporidium and Microsporidium parasites in stool of 85% of participants secondary to improvements in immune function during anti-HIV therapy including a protease inhibitor in a pilot study.

Landau reported on the effects of antiretroviral therapy that included a protease inhibitor on 17 people with AIDS who had Cryptosporidium in their stools. 12 had chronic diarrhoea, and five showed cholangitis. Cessation or marked reduction in diarrhoea was observed. All patients with cholangitis showed partial improvement, although four continued to have Cryptosporidium in their stools.

Carr reported resolution of diarrhoea among nine patients with cryptosporidiosis or microsporidiosis treated with combination antiretroviral therapy including at least one protease inhibitor. One person relapsed. During treatment parasites were no longer detectable in stool of 8/9.

Cryptosporidium-specific treatments

Hewitt enrolled 35 people with cryptosporidiosis and CD4 counts below 150 in ACTG 192, comparing paromomycin (500mg four times daily) for 21 days versus placebo. Mean CD4 count at entry was 38. 31 people completed the randomised phase; there was no difference in the proportions achieving a partial or complete response between the treatment arm and the placebo arm. There were no differences between the arms in diarrhoea, anti-diarrhoeal pill use, weight, number of oocysts or other gastro-intestinal symptoms.

White randomized ten people with cryptosporidiosis to receive paromomycin (25-35 mg/kg/day) or placebo for 14 days, at which point participants crossed over to the alternate arm. Significant improvements in stool frequency and character and oocyst secretion were observed while patients received paromomycin.

Kanyok randomized eleven people with AIDS and cryptosporidiosis to receive paromomycin (500 mg every six hours) or placebo. After 14 days of treatment, one of five paromomycin recipients had a complete response, one had a partial response, and three had no response (2/3 of the non-responders were co-infected with Microsporidium). None of six patients randomized to placebo had a response.

Armitage, Clezy, Drake, Fichtenstein (1993), Gathe, Ramratnam and Walmsley all reported complete or partial responses in a majority of patients with cryptosporidiosis treated with paromomycin in small open label studies and case series.

Soave (1998) reported that no difference was noted between oral spiramycin (three million units three times daily for three weeks) or placebo in a randomized, double-blind, placebo-controlled trial in 73 people with AIDS. Poor absorption is thought to have played a role. In a follow-up study, intravenous spiramycin was associated with complete or partial responses in 21 of 31 patients, but at the cost of substantial toxicities.

Connolly (1988) treated 15 patients in a randomized trial of erythromycin (500mg by mouth every day) versus spiramycin (500 mg by mouth every day). 6/10 given erythromycin and all patients given spiramycin showed a reduction in stool volume. Therapy was limited in both groups by nausea, vomiting and abdominal pain.

Sprinz enrolled 24 adults with cryptosporidial diarrhoea in an open label study of roxithromycin (300 mg twice daily for four weeks). 11/23 (50%) had a complete remission, and 7 (29%) had a partial response (diarrhoea improved and/or decreased number of organisms in stool). Roxithromycin was well-tolerated, although three participants had a transient increase in hepatic enzymes that subsided with a 50% dose reduction.

Holmberg reported that people receiving clarithromycin or rifabutin as MAI prophylaxis had a lower rate of cryptosporidiosis than those receiving no MAI prophylaxis, suggesting that these drugs may have a prophylactic effect. No protective effect was seen among people receiving azithromycin.

Fichtenbaum (2001) reported a retrospective analysis of the use of clarithromycin and rifabutin among 2288 participants in two American trials which were designed to study the impact of rifabutin and clarithromycin in the prevention of MAI and CMV. Median CD4 count was 29 at entry and median follow-up was 463 days. The rate of cryptosporidiosis (2-3% per year) did The relative risk of cryptosporidiosis was 0.50 among those receiving rifabutin while clarithromycin had no impact of risk of cryptosporidiosis (RR=0.98).

Dunne reported on the use of azithromycin in the open-label treatment of 151 people with cryptosporidiosis. Among 63 evaluable participants who received 600 mg/day for 14 or 28 days, 14% experienced reduction in oocyst counts. By day 28 the frequency of bowel movements had fallen to 3 or less per day in 48% of participants and a clinical response was seen in 68%.

Blanshard conducted pilot studies of azithromycin, letrazuril and paromomycin in the treatment of cryptosporidiosis. Azithromycin at a dose of 500 mg/day was ineffective. Letrazuril (150 to 200mg/day) was associated with an improvement in symptoms in 40% of recipients and cessation of excretion of oocysts in stool in 70%, although biopsies remained positive. Paromomycin was associated with complete remission of diarrhoea in 60% and some improvement in symptoms in an additional 5% but did not eliminate the infection.

Smith reported reductions in still volume and oocysts shedding in 11 patients with cryptosporidiosis treated with azithromycin (600mg/day) plus paromomycin (1g twice daily) for 4 weeks, followed by paromomycin alone.

Amadi randomised 50 HIV-positive and 50 HIV-negative children with cryptosporidiosis to receive 100mg of nitazanoxide twice daily for three days or placebo. Cryptosporidial infections were established by the presence of cryptosporidial oocysts in stool samples, and children with Giardia and/or Entamoeba histolytica were excluded. After seven days diarrhoea had stopped in 56% of HIV-negative children who received nitazoxanide compared to 23% of the HIV-negative placebo group (p=0.037), and at day 10 cryptosporidium was not present in the stools of 52% of the HIV-negative treated group, compared with 38% of the placebo group. Seventeen per cent of the HIV-negative placebo group had died by day 8. However, only 8% of the HIV-positive children treated with nitazoxanide experienced resolution of diarrhoea by day 7, compared to 25% of the placebo group (a non-significant difference). No benefit in terms of parasitological clearance or mortality was observed in HIV-positive children; 20% had died by day 8 of the study, compared to none of the HIV-negative children treated with nitazoxanide. Eradication of cryptosporidium was associated with a higher CD4 cell count 1387 cells/mm³ versus 554 cells/mm³. After the initial ten-day randomisation phase, mothers were offered the option of a further three day course of open-label nitazozanide treatment. Twenty-four HIV-positive children entered this phase of the study; four days after ceasing treatment, 16 had no evidence of diarrhoea. Ten out of 13 of the HIV-positive children on their second course of nitazoxanide treatment had a clinical response at this stage. However, only three of twelve achieved eradication of cryptosporidium after a second phase of treatment.

David treated 30 people with AIDS and cryptosporidial diarrhoea with nitazoxanide (500, 1000, 1500 or 2000mg/day) for four weeks. Preliminary analysis from 22 participants who completed at least four weeks of treatment revealed that 15 (68%) had a reduction in bowel movement frequency; four had complete resolution of diarrhoea. 9 (41%) had parasitologic improvement. There were no significant toxicities.

Rossignol randomised 66 individuals with cryptosporidial diarrhoea and HIV to receive either 500mg bid nitazoxanide or 1000mg bid, or placebo, with crossover to one of the randomised doses by the placebo group after 14 days. Parasite clearance was reported in 63 and 67% of the nitazanoxide groups, and this was statistically superior to the placebo group response.

Soave treated 68 patients in a placebo-controlled, dose-escalating study (with crossover to open-label treatment for non-responders) of diclazuril (50-600 mg by mouth for 7 days); at the highest dose only, limited efficacy was observed. Poor bioavailability may limit diclazuril's efficacy.

Connolly (1990) treated nine critically ill AIDS patients with cryptosporidial diarrhoea with diclazuril (200mg or 400mg per day). Five died during treatment but no direct toxic effects of the drug were apparent. Stool volumes remained unchanged throughout treatment. In the five patients who completed the treatment Cryptosporidia were still identified in stools. Comparison of rectal biopsies of 6 patients before and after treatment (one patient was initially negative) showed clearance of oocysts in one case.

Victor reported that letrazuril (50mg daily for seven days) eliminated Cryptosporidium oocysts from the stool of a man with advanced HIV disease and cryptosporidial diarrhoea.

Harris conducted a Phase I study of letrazuril (50 - 100mg/day orally for six weeks). Of 14 evaluable recipients with CD4 counts ranging from 5 to 99, five had symptomatic improvement and eradication of Cryptosporidium oocysts from the stool, two had symptomatic improvement without eradication of oocysts, and seven did not respond. Seven developed a transient drug-related rash.

Fanning suggested that octreotide may be less effective in controlling cryptosporidiosis than other causes of diarrhoea in people with AIDS.

Fareed treated 18 adults with CD4 counts below 100 and cryptosporidial diarrhoea with a high-dose garlic concentrate, allicin, administered in two 30mg doses (dissolved in 90cc distilled water) per day, the first taken orally, the second as a rectal retention enema. During the first three weeks, 8/18 observed a reduction in number of bowel movements or a mild increase in body weight. After six weeks, 10/16 continued to show these improvements. Stool examinations have been consistently negative for Cryptosporidium among all eight participants who have completed eight weeks of treatment. The major reported side-effect was a strong garlic smell and taste.

Fries reported two trials of hyperimmune bovine colostrum. In the first, 40 participants received either colostrum (20g/day) or placebo for one week; all participants received colostrum during the second week. The treatment had no effect on stool frequency, volume or mean oocyst counts, but the group which received two weeks treatment experienced a significant fall in median oocyst count during the second week. No adverse events were noted. In the second, trial participants received colostrum at 20g/day escalating to 40g/day and 80g/day at weekly intervals if diarrhoea persisted. The 80g/day dose caused unacceptable bloating and/or increased diarrhoea in some participants. No effect on stool volume or frequency was noted, although there was a trend to reduction in median oocyst counts.

McMeeking treated 14 patients with cryptosporidiosis with either an immunised bovine dialyzable leukocyte extract (immune DLE) or non-immune DLE as placebo. 6/7 patients treated with immune DLE gained weight and had a decrease in bowel movement frequency, with eradication of oocysts from stool in 5/7 patients. 6/7 patients given placebo failed to gain weight, and 5/7 continued to lose weight. Four showed no clearing of oocysts from stool. Five original placebo recipients were switched to immune DLE; four had a decrease in bowel-movement frequency and significant weight gain, with eradication of oocysts from stool in two patients.

ImmuCell Corp. reported the results of four phase I/II trial of a bovine-derived specific polyclonal antibody against Cryptosporidium called CryptoGAM, which enrolled 64 people with AIDS. Reduction in the number of parasites but no reduction in clinical symptoms was observed. ImmuCell announced that development of CryptoGAM has been discontinued due to difficulty recruiting to trials.

Cook treated an AIDS patient for cryptosporidiosis diarrhoea with a somatostatin analogue, SMS-201-995 (100 µg subcutaneously three times a day). The patient had reduced stool volume and frequency and remained diarrhoea-free for eight months with continued daily treatment. Cryptosporidial infection remained unchanged. Simon treated another cryptosporidiosis patient with a somatostatin analogue. Cryptosporidiosis resolved in this patient, which the investigators attributed to improved immune functioning resulting from improved nutritional status. Somatostatin is thought to inhibit intestinal secretion and enhance water and electrolyte absorption but not control the parasite.

A phase I cross-over trial of atovaquone (750mg by mouth four times a day) for cryptosporidiosis was conducted at the NIH in the USA. No toxicity was observed, although, since most of these patients suffer from severe malabsorption, the lack of toxicity might reflect low systemic drug levels. This trial was halted when no difference was seen between the treatment and placebo arms.

References

Amadi B et al. Effect of nitazoxanide on morbidity and mortality in Zambian children with cryptosporidiosis: a randomised trial. The Lancet 360(9343): 1375-80, 2002.

Armitage K et al. Treatment of cryptosporidiosis with paromomycin. Archives of Internal Medicine 152(12): 2497-2499, 1992.

Benhamou Y et al. Effects of triple antiretroviral therapies including a HIV protease inhibitor on chronic intestinal cryptosporidiosis and microsporidiosis in HIV-infected patients. Fourth Conference on Retroviruses and Opportunistic Infections, Washington, abstract 357, 1997.

Blanshard C et al. Pilot studies of azithromycin, letrazuril and paromomycin in the treatment of cryptosporidiosis. International Journal of STDs and AIDS 8(2): 124-129, 1997.

Carr A et al. Treatment of HIV-1-associated microsporidiosis and cryptosporidiosis with combination antiretroviral therapy. Lancet 351(9098): 256-261, 1998.

Clezy K et al. Paromomycin for the treatment of cryptosporidial diarrhoea in AIDS patients. AIDS 5(9): 1146-1147, 1991.

Connolly GM et al. Cryptosporidial diarrhoea in AIDS and its treatment. Gut 29(5): 593-597, 1988.

Connolly GM et al. Diclazuril in the treatment of severe cryptosporidial diarrhoea (letter). AIDS 4(7): 700-701, 1990.

Cook DJ et al. Somatostatin treatment for cryptosporidial diarrhoea in a patient with AIDS. Annals of Internal Medicine 108(5): 708-709, 1988.

Davis LJ et al. Nitazoxanide for AIDS-related cryptosporidial diarrhea: an open-label safety, efficacy and pharmacokinetic study. 36^th Interscience Conference on Antimicrobial Agents and Chemotherapy, New Orleans, abstract L050, 1996.

Drake J et al. Efficacy and tolerability of oral aminosidine [paromomycin] (Gabbroral TM) for the treatment of cryptosporidial diarrhea in patients with AIDS. First International Congress on Drug Therapy in HIV Infection, Glasgow, abstract P45, 1992.

Dunne MW et al. Open label azithromycin in the treatment of cryptosporidiosis. Ninth International Conference on AIDS, Berlin, abstract PO-B10-1500, 1993.

Fanning M et al. Pilot study of Sandostatin (octreotide) therapy of refractory HIV-associated diarrhea. Digestive Diseases and Sciences 36: 476-80, 1991.

Fareed G et al. The use of a high-dose garlic preparation for the treatment of Cryptosporidium parvum diarrhea. Eleventh International Conference on AIDS, Vancouver, abstract Th.B.4215, 1996.

Fichtenbaum CJ et al. Use of paromomycin for treatment of cryptosporidiosis in patients. Clinical Infectious Diseases 16(2): 298-300, 1993.

Fichtenbaum C et al. Rifabutin but not clarithromycin prevents cryptosporidiosis in persons with advanced HIV infection. AIDS 14(18): 2889-2893, 2000.

Flanigan T et al. Cryptosporidium infection and CD4 counts. Annals of Internal Medicine 116(10): 840-842, 1992.

Flanigan TP et al. Prospective trial of paromomycin for cryptosporidiosis in AIDS. American Journal of Medicine 100(3): 370-372, 1996.

Fries L et al. Clinical and microbiologic effects of bovine anti-cryptosporidium immunoglobulin (BACI) on cryptosporidial diarrhea in AIDS. 34th Interscience Conference on Antimicrobial Agents and Chemotherapy, Orlando, abstract M21, 1994.

Gathe J et al. Treatment of gastro-intestinal cryptosporidiosis with paromomycin. Sixth International Conference on AIDS, San Francisco, abstract 2121, 1990.

Giang TT et al. Cryptosporidiosis of the nasal mucosa in a patient with AIDS (letter). AIDS 8:555-556, 1994.

Harris M et al. A phase 1 study of letrazuril in AIDS-related cryptosporidiosis. AIDS 8: 1109-1113, 1994.

Hewitt RG et al. Paromomycin: no more effective than placebo for treatment of cryptosporidiosis in patients with advanced human immunodeficiency virus infection. AIDS Clinical Trial Group. Clinical Infectious Diseases 31(4): 1084-92, 2000.

Holmberg SD et al. Possible effectiveness of clarithromycin and rifabutin for cryptosporidiosis chemoprophylaxis in HIV disease. HIV Outpatient Study (HOPS) Investigators. Journal of the American Medical Association 279(5): 384-386, 1998.

Kanyok TP et al. Preliminary results of a randomized, blinded, control study of paromomycin vs. placebo for the treatment of Cryptosporidium diarrhea in AIDS patients. Ninth International Conference on AIDS, Berlin, abstract B10-1508, 1993.

Katz MD et al. Treatment of severe cryptosporidium related diarrhea with octreotide in a patient with AIDS. Drug Intelligence and Clinical Pharmacology 22: 134-6, 1988.

Landau A et al. Impact of antiretroviral therapy (ART) on cryptosporidiosis outcome and factors of clinical resistance. Fifth Conference on Retroviruses and Opportunistic Infections, Chicago, abstract 480, 1998.

McGowan I et al. The natural history of cryptosporidial diarrhoea in HIV-infected patients. AIDS 7(3): 349-354, 1993.

McMeeking A et al. A controlled trial of bovine dialyzable leukocyte extract for cryptosporidiosis in patients with AIDS. Journal of Infectious Diseases 161(1): 108-112, 1990.

Petersen C. Cryptosporidiosis in patients infected with the human immunodeficiency virus. Clinical Infectious Diseases 15(6): 903-909, 1992.

Rossignol JF et al. A double blind placebo controlled study of nitazoxanide in the treatment of cryptosporidial diarrhoea in AIDS patients in Mexico. Transactions of the Royal Society for Tropical Medicine and Hygiene 92 (6): 663-666, 1998.

Simon D et al. Resolution of Cryptosporidium infection in an AIDS patient after improvement of nutritional and immune status with octreotide. Case report. American Journal of Gastroenterology 5: 615-618, 1991.

Smith NS et al. Combination paromomycin (P) and azithromycin (A) for cryptosporidiosis. Fifth Conference on Retroviruses and Opportunistic Infections, Chicago, abstract 481, 1998.

Soave R et al. Oral diclazuril therapy for cryptosporidiosis. Sixth International Conference on AIDS, San Francisco, abstract ThB 520, l990.

Soave R et al. Treatment strategies for cryptosporidiosis. Annals of the New York Academy of Sciences 616: 442-451, 1990.

Soave R. Personal communication reported in Marco M et al. The OI Report. Treatment Action Group, New York, 1998.

Sprinz E et al. AIDS-related cryptosporidial diarrhoea: an open study with roxithromycin. Journal of Antimicrobial Chemotherapy 41 (suppl B): 85-91, 1998.

Vakil N et al. Biliary cryptosporidiosis in HIV-infected people after the waterborne outbreak of cryptosporidiosis in Milwaukee. New England Journal of Medicine 334(1): 19-23, 1996.

Victor GH et al. Letrazuril therapy for cryptosporidiosis: clinical response and pharmacokinetics. AIDS 7(3): 438-439, 1993.

Walmsley S et al. Effectiveness of paromomycin in cryptosporidiosis in AIDS. Ninth International Conference on AIDS, Berlin, abstract B10-1473, 1993.

White AC et al. Paromomycin for cryptosporidiosis in AIDS: a prospective double-blind trial. Journal of Infectious Diseases 170:4 19-424, 1994.