INCIDENCE AND CAUSE OF TRANSIENT AND PERSISTING DIARRHEA IN RELATION TO CD4 LYMPHOCYTE COUNT IN HIV-INFECTED INDIVIDUALS

Objective: To determine whether there is a relationship between the ca use of diarrhoea in HIV-infected patients and their CD4 lymphocyte cou nt, and to ascertain whether this information can be used to guide inv estigation. Methods: Patients with diarrhoea were retrospectively iden tified from a number of different data sources. All had been investiga ted according to standard protocols. The cause of the diarrhoea was co rrelated with the CD4 count in those patients who had lymphocyte subse t counts performed within 1 month of presentation with diarrhoea. Resu lts: In total, 312 patient-episodes of diarrhoea were studied, in whic h a total of 280 definite and 48 possible pathogens were identified. P atients with a CD4 count of less than 50/mm3 were significantly more l ikely to have diarrhoea due to infection with a definite pathogen. The relative frequency of different pathogens varied with the CD4 count: Cryptosporidium was common in all groups; Microsporidium, cytomegalovi rus and Mycobacterium avium intracellulare only occurred in patients w ith a CD4 count of less than 200/mm3. The latter two were significantl y more common in patients with a CD4 count of less than 50/mm3. Patien ts with CD4 counts above 200/mm3 were significantly more likely to hav e no pathogen found despite extensive investigations. All the pathogen s found in those with a CD4 count of more than 200/mm3 could potential ly be diagnosed on stool examination and blood culture alone, but at l east 16% of those pathogens in patients with a CD4 count of 50-200/mm3 and 24% in patients with a CD4 count of less than 50/mm3 CoUld only b e identified on intestinal biopsy, Conclusions: The range and frequenc y of pathogens encountered varies markedly with CD4 count. Duodenal an d rectal biopsies are required in the investigation of HIV-related dia rrhoea if the CD4 count is less than 200/mm3.