In tropical areas where Plasmodium falciparum malaria is endemic, co-infection with HIV-1 does not lead to a worsening of malaria, raising questions about the immunological interactions between both infections. Alterations of immune response to malaria during HIV-1 infection was investigated in the town of Bobo Dioulasso, Burkina Faso. Sixty-six adults were enrolled, including 37 HIV-1 positive subjects with < 250 CD4+ cells/microliter and clinical AIDS, and 29 HIV-1, negative healthy subjects. In vitro lymphocyte proliferation and cytokine (IFN-gamma, IL-2 and IL-4) production were assessed in isolated mononuclear cells (PBMC) in presence of PHA, PPD or three malarial antigens: the baculovirus-expressed protein from P. falciparum Merozoite Surface Protein-I, a P. falciparum in vitro culture and a crude schizont extract. Compared with healthy subjects. AIDS patients presented with decreased levels of cell proliferation and of IFN-gamma and IL-2 production, in response to all antigens except the schizont extract. Similar levels of IL-4 production were obtained in both groups. Mitogenic stimulation of whole blood cultures was also performed, and revealed similar trends in cytokine production as in PBMC cultures. These results show that some components of the specific human immune responses to falciparum parasites may not be modified during AIDS, in spite of the strong cellular alterations induced by HIV, namely the decrease of the CD4+ lymphocyte subset.

Objectives: (1) To determine the prevalence of sexually transmitted diseases (STDs) in pregnant women in Burkina Faso. (2) To evaluate the potential of clinical management of STDs based on screening with clinical data and urine leucocyte esterase test (LET). Methods: Cross sectional study among antenatal clinic attendees was conducted in 1994 in Ouagadougou and Bobo-Dioulasso, the two largest urban centres in Burkina Faso, where more than 94% of the pregnant women benefit from antenatal care at least twice during their pregnancy. Each woman selected underwent an interview, general and gynaecological examination. Genital samples were collected to confirm the presence of STD pathogens. Logistic regression analysis was done to identify models that predict (a) gonorrhoea and/or chlamydia, (b) trichomoniasis and/or bacterial vaginosis, (c) candidiasis. Sensitivity, specificity and positive and negative predictive values of these models were assessed using standard methods. Results: All 645 consecutive pregnant women were enrolled in the two sites. Among these women 32-4% presented at least one STD. The major STDs were: trichomoniasis (14%), bacterial vaginosis (13%), recent syphilis (3.6%), chlamydial infection (3.1%), genital warts (3%), gonococcal infection (1.6%) and genital ulcer (0.8%). Prevalence of vaginal candidiasis was 14%. The use of a risk marker (length of relationship with regular sexual partner < 3 years), and the positivity (+ + +) of the urine LET provided a sensitivity of 80% and a positive predictive value of 7% for the screening of gonococcal and/or chlamydial infection. If clinical signs and positivity of the urine LET were taken into account sensitivity and positive predictive value of trichomoniasis and/or bacterial vaginosis screening were 77% and 37%, respectively. Clinical signs and positivity of the urine LET showed a low sensitivity (23%) for screening vaginal candidiasis. Conclusions: The prevalence of STDs in pregnant women is high in urban Burkina Faso. Systematic screening combined with effective treatment should be included in antenatal care in the future. Urine LET, if associated with interview and clinical examination offers a simple, rapid and affordable tool for systematic screening of STDs in pregnant women. However, the proportion of overtreatments with proposed strategies will be high. Further studies are needed to develop and validate better algorithms with probably cheap laboratory tests.

SummaryIn the developed word, monitoring HIV-infected patients is routinely determined by CD4 ϩ T lymphocyte absolute counts. The reference procedure, flow cytometry, is expensive, requires sophisticated instrumentation and operators with specific training. Due to these limitations, CD4 counting is often unavailable in developing countries. The Capcellia assay is an enzyme-linked immunoassay for quantitative determination of CD4 and CD8 molecules. We evaluated this method in West Africa on blood samples collected from 39 HIV-uninfected and 44 HIV-infected adult subjects. CD4 concentration ranges were determined according to the clinical stages of the disease. We then studied the relationship between the two methods in the HIVinfected patients. The Spearman's rank correlation was 0.61 (95% confidence interval: 0.38-0.76, P Ͻ 0.0001). Nevertheless, determination of limits of agreement revealed discrepancies between the two methods, especially for CD4 counts Ͼ 0.4 ϫ10 9 /l, which are discussed. We conclude that the Capcellia assay is a convenient means to determine the immunodepression level where flow cytometric instrumentation is unavailable, and can be complementary to CD4 T lymphocyte enumeration.keywords enzyme immunoassay, flow cytometry, CD4 ϩ T lymphocytes counts, HIV correspondence Dr Serge

Our objective was to propose a strategy to screen HIV-infected African people for biological immunodeficiency easily. In a cross-sectional study, we analysed the patterns of diseases and of CD4 counts among 266 HIV-infected adults. Peripheral facial paralysis and chronic cutaneo-mucous diseases were the earlier B-stage diseases. Pulmonary tuberculosis was close to B-stage diseases, and chronic diarrhoea was borderline between B and C stages. Cachexia was the most frequent C-stage symptom (47.8%). Ninety per cent of CDC-C stage people had CD4 counts below 350/microliter, whereas only 75% had CD4 counts below 200/microliter. Regression analysis identified the lymphocyte count, clinical stage and platelet count as predictors of CD4 count below 350/microliter. A simple score (lymphocyte count < or = 2500/microliter and clinical stage > or = B) is proposed to determine this CD4 threshold (positive predictive value: 83%) and to determine those patients needing treatment to prevent wasting and opportunistic infections.

To determine the prevalence of HIV infection among pregnant women in Burkina Faso and to identify factors associated with HIV infection in this population, we performed in 4 antenatal clinics in 4 urban centres located in the centre, the north, the east and the south-west of the country cross-sectional serological surveys between 7 October 1994 and 7 February 1995 using unlinked anonymous HIV screening method. In the towns included in the study more than 94% of pregnant women attend at least 2 antenatal consultations. In each centre all pregnant women attending an antenatal consultation during a 2-week period were invited to take part in a study to measure the prevalence of syphilis. Women with positive syphilis serology were offered treatment free of charge. A total of 1294 pregnant women were recruited into the study. There were no refusals. Thirty-two women (2.5%) had positive syphilis serology. The overall seroprevalence of HIV infection was 8% (95% confidence interval [CI]: 6.6-9.6). There was no evidence that the prevalence of syphilis or HIV infection varied between the 4 towns (P=0.12 and 0.52 respectively). In logistic regression analyses only the presence of syphilis infection was associated with risk of HIV (odds ratio=3.4; 95% CI: 1.4-7.9). The prevalence of syphilis among pregnant women in Burkina Faso is relatively low compared with that of HIV infection. These results suggest that HIV epidemic in Burkina Faso is important and that there is an urgent need to intensify HIV-prevention activities. The need for HIV surveillance in this sentinel population is urgent to follow the course of the epidemic and to measure the impact of interventions. One of the consequences of the high prevalence of HIV infection among pregnant women which requires greater attention in Burkina Faso is the increasing number of children who will be infected with HIV by vertical transmission.