Date of Award


Document Type

Open Access Thesis

Degree Name

Medical Doctor (MD)

First Advisor

Richard A Martinello


BACKGROUND: As part of rapidly expanding HIV-1 Care Programs in sub-Saharan Africa, many HIV-1 infected individuals are being identified who do not yet qualify for ART. These individuals may benefit from treatment of co-infections, such as soil-transmitted helminths. Treatment of helminth infection in HIV-1 infected individuals may improve health and nutrition and may slow progression to AIDS. In this setting, access to both medical care and diagnostic testing is limited. Determining correlates of helminth infection in HIV-I infected individuals may allow clinicians to better target empiric anti-helminth therapy in HIV-1 infected individuals. METHODS: To describe the prevalence and correlates of helminth infection in HIV-1 infected individuals in Nairobi, Kenya, we screened stool samples of HIV-1 infected adults in Kenya. Stool samples were collected from consenting participants. Helminth infection was diagnosed by wet preparation, Kato-Katz technique, and Formol Ether Concentration. Sociodemographic data and medical history were collected using standardized questionnaires. RESULTS: 912 HIV-1 infected adults were screened. The median age was 34 years (IQR 28 - 40) and 74% of those screened were female. Most (75.6%) did not have access to adequate sanitation, as defined by a flush toilet and piped running water. Most participants had received primary education only (52.9%), followed by secondary schooling (32.4%), college (7.7%) and no education (4.2%). Dirt floor was reported by 35.8% of participants. The mean number of rooms in the home was 1.7 (50+/- 1.1). The mean number of children per household was 1.6 (SD+/- 1.3). 113.7% reported current diarrhea, and 6.3% current blood in stool. Mean CD4 count was 439 and median Cb4 count 389 (range 4-1726). Among those screened, 16.1 % had helminths detected in stool specimens. Hookworm was the most commonly found helminth, representing 56.5% of positive samples. All subjects had light helminth burden by WHO criteria except one patient who had moderate infection with ascaris. Helminth co-infection was associated with educational level. For every increase in level of education attained (none, primary, secondary, or college) there was a decrease in the odds of helminth infection OR=0.41, Cl0.26 - 0.65, p=0.01 ). Helminth infection was also significantly associated with lack of flush toilets (OR 1.90, C1 1.07 - 3.39). Age, gender, having a dirt floor, number of children, current diarrhea, or blood in stool did not significantly correlate with the risk of helminth infection. In addition, there was no significant association between absolute CD4 count and presence of helminth infection. DISCUSSION: In this cohort of HIV-1 seropositive individuals, there was a lower prevalence of helminth co-infection (10-15%) than found in some other areas of Africa. Lower educational levels and lack of access to flush toilets were associated with helminth co-infection. These co-factors may be useful for identifying helminth co-infected HIV-I infected individuals in urban centers to target for either stool testing or empiric anti-helminthics.