Scaling up antiretroviral therapy for HIV-infected children in Côte d'Ivoire: determinants of survival and loss to programme.

OBJECTIVE: To investigate deaths and losses to follow-up in a programme designed to scale up antiretroviral therapy (ART) for HIV-infected children in Côte d'Ivoire. METHODS: Between 2004 and 2007, HIV-exposed children at 19 centres were offered free HIV serum tests (polymerase chain reaction tests in those aged < 18 months) and ART. Computerized monitoring was used to determine: (i) the number of confirmed HIV infections, (ii) losses to the programme (i.e. death or loss to follow-up) before ART, (iii) mortality and loss-to-programme rates during 12 months of ART, and (iv) determinants of mortality and losses to the programme. FINDINGS: The analysis included 3876 ART-naïve children. Of the 1766 with HIV-1 infections (17% aged < 18 months), 124 (7.0%) died, 52 (2.9%) left the programme, 354 (20%) were lost to follow-up before ART, 259 (15%) remained in care without ART, and 977 (55%) started ART (median age: 63 months). The overall mortality rate during ART was significantly higher in the first 3 months than in months 4-12: 32.8 and 6.9 per 100 child-years of follow-up, respectively. Loss-to-programme rates were roughly double mortality rates and followed the same trend with duration of ART. Independent predictors of 12-month mortality on ART were pre-ART weight-for-age z-score < -2, percentage of CD4+ T lymphocytes < 10, World Health Organization HIV/AIDS clinical stage 3 or 4, and blood haemoglobin < 8 g/dl. CONCLUSION: The large-scale programme to scale up paediatric ART in Côte d'Ivoire was effective. However, ART was often given too late, and early mortality and losses to programme before and just after ART initiation were major problems.

Scaling up antiretroviral therapy for HIV-infected children in Côte d'Ivoire: determinants of survival and loss to programme

Objective: To investigate deaths and losses to follow-up in a programme designed to scale up antiretroviral therapy (ART) for HIV- infected children in Cote d'Ivoire. Methods Between 2004 and 2007; HIV-exposed children at 19 centres were offered free HIV serum tests (polymerase chain reaction tests in those aged 18 months) and ART. Computerized monitoring was used to determine: (i) the number of confirmed HIV infections; (ii) losses to the programme (i.e. death or loss to follow-up) before ART; (iii) mortality and loss-to-programme rates during 12 months of ART; and (iv) determinants of mortality and losses to the programme. Findings The analysis included 3876 ART-naive children. Of the 1766 with HIV-1 infections (17aged 18 months); 124 (7.0) died; 52 (2.9) left the programme; 354 (20) were lost to follow-up before ART; 259 (15) remained in care without ART; and 977 (55) started ART (median age: 63 months). The overall mortality rate during ART was significantly higher in the first 3 months than in months 4-12: 32.8 and 6.9 per 100 child-years of follow-up; respectively. Loss-to-programme rates were roughly double mortality rates and followed the same trend with duration of ART. Independent predictors of 12-month mortality on ART were pre-ART weight- for-age z-score -2; percentage of CD4+ T lymphocytes 10; World Health Organization HIV/AIDS clinical stage 3 or 4; and blood haemoglobin 8 g/dl. Conclusion The large-scale programme to scale up paediatric ART in Cote d'Ivoire was effective. However; ART was often given too late; and early mortality and losses to programme before and just after ART initiation were major problems

Impact of opportunistic diseases on chronic mortality in HIV-infected adults in Côte d'Ivoire.

OBJECTIVE: To estimate incidence rates of opportunistic diseases (ODs) and mortality for patients with and without a history of OD among HIV-infected patients in Côte d'Ivoire. METHODS: Using incidence density analysis, we estimated rates of ODs and chronic mortality by CD4 count in patients in a cotrimoxazole prophylaxis trial in Abidjan before the highly active antiretroviral therapy (HAART) era. Chronic mortality was defined as death without a history of OD or death more than 30 days after an OD diagnosis. We used Poisson's regression to examine the effect of OD history on chronic mortality after adjusting for age, gender, and current CD4 count. RESULTS: Two hundred and seventy patients (40% male, mean age 33 years, median baseline CD4 count 261 cells/microl) were followed up for a median of 9.5 months. Bacterial infections and tuberculosis were the most common severe ODs. Of 47 patients who died, 9 (19%) died within 30 days of an OD, 26 (55%) died more than 30 days after an OD, and 12 (26%) died with no OD history. The chronic mortality rate was 31.0/100 person-years for those with an OD history, and 11.1/100 person-years for those with no OD history (rate ratio (RR) 2.81, 95% confidence interval (CI): 1.43 - 5.54). Multivariate analysis revealed that OD history remained an independent predictor of mortality (RR 2.15, 95% CI: 1.07 - 4.33) after adjusting for CD4 count, age and gender. CONCLUSIONS: Before the HAART era, a history of OD was associated with increased chronic HIV mortality in Côte d'Ivoire, even after adjusting for CD4 count. These results provide further evidence supporting OD prophylaxis in HIV-infected patients.

[Compliance in HIV infected adults. Study of opportunistic infection prophylaxis with cotrimoxazole in Ivory Coast]. / Observance chez les adultes infectés par le VIH. Etude d'une prophylaxie des infections opportunistes par le cotrimoxazole en Côte d'Ivoire.

BACKGROUND: The compliance to a daily treatment for illimited duration and the factors that influence it have been rarely studied in sub-Saharian Africa. OBJECTIVE: Describe the compliance to prophylaxis with cotrimoxazole fort (one tablet per day) and its associated factors in patients infected by HIV participating in a clinical trial in Abidjan. METHOD: The tablets packed in individual blisters were provided every month, and the blisters were recuperated the following month. A global compliance ratio (GCR) was established for each patient (empty blisters at the end of the study/follow-up period during the study) and monthly compliance ratio [MCR] (empty blisters during a visit/time lapse since last visit). For each monthly visit foreseen in the protocol, a respect of the appointment ratio (RAR) was described (visits foreseen in the protocol respected that month/visits foreseen in the protocol). The association of GCR with the characteristics on inclusion was studied using logistic regression methods. RESULTS: 530 adults were followed-up for a mean of 10 months. The MCR and the RAR progressed in parallel, decreasing the first 5 months and stabilizing at around 0.80 for the RAR and 0.70 for the MCR. The mean GCR was of 0.77. Three hundred and nine patients (58%) were considered as compliant (0.80

[Bacillary angiomatosis in an adult infected with HIV-1 at an early stage of immunodepression in Abidjan, Côte d'Ivoire]. / Angiomatose bacillaire chez un adulte infecté par le VIH-1 à un stade d'immunodépression peu avancé à Abidjan, Côte d'Ivoire.

Human immunodeficiency virus (HIV)-associated bacillary angiomatosis has rarely been described in Africa. We report here the first case in Côte d'Ivoire. Although in industrialised countries bacillary angiomatosis has been described in patients with low CD4 count, this episode occurred in the first year following HIV-seroconversion in an adult patient with more than 500 CD4 cells per cubic millimetre. Symptoms rapidly and totally disappeared under erythromycin treatment, although with a relapse two years after the end of the first episode. In Africa where people living with HIV often present chronic cutaneous lesions, bacillary angiomatosis may be under-diagnosed. Bacillary angiomatosis must be systematically considered in face of lesions similar to Kaposi's sarcoma. Improving knowledge on symptoms of bacillary angiomatosis in Africa should lead to better treatment and a better estimation of its true frequency which may be underestimated.

HIV-1-related morbidity in adults, Abidjan, Côte d'Ivoire: a nidus for bacterial diseases.

We studied mortality and morbidity in 270 HIV-1-infected adults (60% women, median age 31 years, mean baseline CD4 count 331/mm(3) ) observed in a follow-up that lasted a median 10 months in Côte d'Ivoire. Survival and probability of remaining free from any episode of morbidity at 12 months were 0.80 and 0.50, respectively. Baseline CD4 count <200/mm(3) was the only variable associated with global morbidity and mortality, with hazard ratios of 2.50 and 7.57, respectively. The most frequent causes of morbidity were severe bacterial infections (incidence rate: 26.1 per 100 person-years [py]), followed by oral candidiasis (22.3% py), unexplained weight loss over 10% of baseline body weight (13.3% py), tuberculosis (10.1% py), unexplained chronic diarrhea (9.7% py), and isosporiasis (5.1% py). Nontyphoid Salmonella accounted for 37% of isolated strains during severe bacterial infections, followed by Streptococcus pneumoniae (34%), Escherichia coli (15%), and Shigella species (7%). A significant part of bacterial morbidity occurred in patients with baseline CD4 count > or = 200/mm(3), in whom the incidence rate of bacterial diseases was 21.3% py and the probability of remaining free from any bacterial infection at 12 months was 0.80 (vs. 36.4% py and 0.71 in patients with baseline CD4 count <200/mm(3); p =.07).

Prognostic value of cross-sectional anthropometric indices on short-term risk of mortality in human immunodeficiency virus-infected adults in Abidjan, Côte d'Ivoire.

In sub-Saharan Africa where weight loss is very difficult to estimate, cross-sectional anthropometric indicators could be useful to predict human immunodeficiency virus (HIV)-associated mortality. The study objective was to look for threshold values of baseline body mass index, arm muscle circumference, and fat mass to predict the risk of death in HIV-infected adults included in a 1996-1998 trial of early cotrimoxazole chemoprophylaxis in Abidjan, Côte d'Ivoire (COTRIMO-CI-ANRS 059 trial). The authors graphically determined if consecutive anthropometric categories with the closest hazards ratios of the risk of death could be clustered to obtain a unique threshold that distinctly separated two categories. When the threshold values were determined, the authors estimated the hazards ratio of mortality of this two-category model. A significant increase of mortality was observed for a body mass index of < or =20.3 in men (hazards ratio = 2.6; 95% confidence interval (CI): 1.4, 5.0) and of < or =18.5 in women (hazards ratio = 2.2; 95% CI: 1.05, 4.5) and for a fat mass of < or =6% in men (hazards ratio = 4.6; 95% CI: 2.3, 9.4) and of < or =18% in women (hazards ratio = 2.4; 95% CI: 1.2, 4.9). No simple threshold could be identified for arm muscle circumference. In Côte d'Ivoire where chemoprophylaxis of opportunistic infections has recently been recommended to be widely initiated on clinical criteria, such thresholds may help to screen patients with higher risks of mortality.

Effect of early chemoprophylaxis with co-trimoxazole on nutritional status evolution in HIV-1-infected adults in Abidjan, Côte d'Ivoire.

BACKGROUND: In sub-Saharan Africa, malnutrition is a major complication of HIV disease. Measuring accurately the nutritional benefits of a therapeutic intervention could be an easy-to-monitor secondary outcome. METHODS: Anthropometric data were analysed from patients participating in a placebo-controlled trial of co-trimoxazole prophylaxis in adults recruited at early stages of HIV-1 infection in Côte d'Ivoire (COTRIMO-CI ANRS 059 trial). Body mass index (BMI), arm muscle circumference (AMC) and percentage of fat mass (FM) were measured at baseline and quarterly during the follow up. Percentage of variation from the baseline value was compared between treatment groups and within the groups using Student t-test. RESULTS: An improvement of all anthropometric indicators was observed in the first 3 months of follow up in both treatment groups, significant in the co-trimoxazole group (P < or = 0.0006) but not in the placebo group (P > or = 0.06). In the co-trimoxazole group, this improvement was maintained for up to 24 months for BMI (P = 0.007), 21 months for AMC (P = 0.02) and only up to 12 months for FM (P = 0.04). The placebo group had a stable anthropometric status up to the end of the trial. Differences between treatment groups were significant for up to 15 months for BMI and AMC and 12 months for FM. CONCLUSION: As co-trimoxazole prophylaxis is now recommended in Africa as part of a minimum package of care for HIV-infected symptomatic subjects, the short-term improvement of these anthropometric indicators in adults who start co-trimoxazole prophylaxis should be considered as an effective clinical outcome.

[Clinical, endoscopic and histologic aspects of chronic Helicobacter pylori gastritis in Côte d'Ivoire: study of 102 patients]. / Aspects cliniques, endoscopiques et histologiques des gastrites chroniques à Helicobacter pylori en Côte d'Ivoire: étude de 102 patients.

BACKGROUND: Helicobacter pylori (H. pylori) is the most frequent aetiological factor of chronic gastritis (CG). The relationship between H. pylori gastritis, gastro-duodenal ulcer and some gastric cancers (adenocarcinoma, gastric MALT lymphoma) has now been proven. AIM: Describe clinical, endoscopical and histological aspects of H. pylori gastritis in Côte d'Ivoire. METHODS: Retrospective analysis of 1960 gastroscopy reports carry out between January 1994 and December 1995. Analysis of clinical and gastric histological results in 137 patients. FINDINGS: Among 137 patients with gastric biopsy, 102 had H. pylori gastritis (68 men, 38 women, mean age: 39.3 years) and 35 had chemical gastritis. Epigastric pain was the most frequent symptom. The mucosa was frequently erythematous or exsudative at endoscopy. Histological anomalies were located in the antrum, the fundus or generalised, respectively in 33.3%, 25.5% and 41.2% of cases. Mild atrophic CG was more frequent in various locations. Gastritis activity was present in 81.4%, intestinal metaplasia in 18.6% and follicular lymphoid hyperplasia in 36.3% of cases. CONCLUSION: Clinical and endoscopical aspects of H. pylori gastritis did not present any particularities. Fundic gastritis without antral localisation was not unusual. This situation could be the result of antibiotic and gastric acid secretion inhibitor treatments.

[Causes of fever in adults infected by HIV-1. Ambulatory follow-up in the ANRS 059 trial in Abidjan, Ivory Coast]. / Causes de fièvre chez des adultes infectés par le VIH-1. Suivi en ambulatoire dans le cadre de l'essai ANRS 059 à Abidjan, Côte d'Ivoire.

OBJECTIVE: Describe the causes of fever in HIV-1 infected adults in Abidjan, Ivory Coast. METHODS: Exhaustive analysis of all the morbid episodes with raise in temperature to above 37.5 degrees C in patients followed-up prospectively, within the framework of the ANRS 059 study from April 1996 to March 1998. RESULTS: One hundred and four patients presented 269 episodes of fever. At the start of these episodes, the mean CD4 count was of 311/mm3, fever had lasted a mean of 3.4 days and mean body temperature was 38.7 degrees C. The 269 episodes lead to 288 diagnoses: 152 specific etiologic diagnoses and 136 non-specific syndrome diagnoses. Community bacterial infections represented 55% of the specific diagnoses, followed by malaria (16%) and tuberculosis (12%). The mean CD4 count during the bacterial episodes was 208/mm3, in malaria 384/mm3 and in tuberculosis 245/mm3. Non-typhi salmonella, pneumococci and Escherischia coli represented 37%, 32%, and 15% respectively of the bacteria isolated. The mean duration between the first and last day of fever was 8.4 days. This time lapse was superior or equal to 30 days in 22 episodes (8%), 50% of which were mycobacterioses (36% tuberculosis and 14% atypic mycobacterioses). Nineteen episodes (7%) lead to death within a mean delay of 58 days. The first cause of death was atypic mycobacteriosis (26%). Death was significantly associated with a CD4 count < 200/mm3 and to prolongation of fever for more than 30 days. CONCLUSION: Other than the frequently described role of tuberculosis in HIV morbidity in sub-Saharian Africa, the role of bacterial diseases, responsible for early death, potentially severe, but curable should be underlined. The diffusion of antibiotic treatment algorithms adapted to the principle clinical syndromes encountered, might improve the treatment of adults infected by HIV consulting in sub-Saharian Africa.

[Spontaneous ascitic infection in cirrhotic Africans. Descriptive study apropos of 12 cases]. / L'infection spontanée du liquide d'ascite chez le cirrhotique africain. Etude descriptive à propos de 12 cas.

BACKGROUND: Spontaneous ascitic infection (SAI) is a frequent and serious complication of cirrhosis. OBJECTIVE: In a retrospective study, the authors report clinical and biological data associated with SAI for cirrhotic patients in an African medical centre. METHODS: Twenty-two cirrhotic patients with ascites were included in a one-year study (November 1996 to October 1997). Clinical and biological data were obtained through medical files. FINDINGS: The mean age of the 22 cirrhotic patients with ascites (12 men, 10 women) was 48.9 years. Twelve cases of SAI were found. In a univariate analysis, the more frequent data in patients with SAI when compared to patients without SAI were: fever or hypothermia (91.7% versus 10%, p = 0.002), abdominal pain (83.3% versus 40%, p = 0.046), cloudy ascitic fluid (66.7% versus 10%, p = 0.003), medium albuminemia (18.2 g/l versus 23 g/l, p = 0.02), medium prothrombin rate (42.8% versus 58.3%; p = 0.04) and ascitic fluid protein level < or = 10 g/l (91.7% versus 30%, p = 0.01). The protein level in ascitic fluid cirrhotic patients was significantly lower in SAI than in patients without SAI (7.6 g/l versus 11 g/l; p = 0.005). In a multivariate analysis, protein levels in ascitic fluid were the only factor associated with SAI (p = 0.024).

[Nutrition problems experienced by adults infected with the human immunodeficiency virus in Abidjan (Ivory Coast)]. / Troubles alimentaires ressentis par des adults infectés par le virus de l'immunodéficience humaine à Abidjan (Côte d'Ivoire).

Denutrition is frequent among HIV-infected (HIV+) adults in sub-Saharan Africa. One of the risk factors for denutrition is a reduction in dietary intake. Eating disorders may be partly responsible for such decreases in food intake. We prospectively analyzed the frequency, associated factors and progression of anorexia, dysphagia and food aversion in a cohort of 330 HIV-infected adults included in a trial of early chemoprophylaxis with cotrimoxazole in Abidjan, Ivory Coast. Patients were followed-up by means of scheduled monthly visits. Eating disorders were assessed using a standardized questionnaire after 6, 12 and 18 months of follow-up. After six months of follow-up, 28% of the patients reported anorexia, 9% dysphagia and 28% food aversion. Multivariate analysis showed that anorexia was significantly more frequent in women than in men (odds ratio (OR) = 2.0 [95% confidence interval: 1.2-3.5]) and in patients with a CD4+ lymphocyte count < 200/mm3 (OR = 1.8 [1.0-3.5]). The risk of dysphagia was also higher for women than for men (OR = 1.8 [1.0-3.5]). The risk of dysphagia was also higher for women than for men (OR = 3.3 [1.3-8.4]). Patients with < 200 CD4+ lymphocytes/mm3 were more likely than those with CD4+ lymphocyte counts of over 200 to suffer food aversion (OR = 1.8 [1.1-3.0]). We analyzed the progression of dietary problems during follow-up and found that anorexia and dysphygia tended to disappear from one evaluation to the next whereas the number of patients reporting food aversion tended to increase. For patients reporting anorexia at the 6-month evaluation, significantly more women than men reported the persistence of anorexia at the 12-month evaluation (16% versus 5%; p = 0.03). Among patients with dysphagia at six months, those with a CD4+ lymphocyte count below 200/mm3 were much more likely than those with a CD4+ count above 200 to report persistent dysphagia at the 12-month evaluation (7% versus 0%; p = 0. 02, Fischer's exact test). For patients with no dietary problems after six months, those taking cotrimoxazole were significantly more likely than those of the placebo group to report food aversion at the 12-month evaluation (21% versus 8%; p = 0.01). We found that dietary problems were associated more with the stage of immunodeficiency that with socioeconomic factors, with the exception of sex, which was associated with several outcomes. These data stress the importance of detecting these frequent dietary problems as part of the overall clinical management of HIV+ adults in Africa, and of providing affected individuals with early nutritional counseling.

Early chemoprophylaxis with trimethoprim-sulphamethoxazole for HIV-1-infected adults in Abidjan, Côte d'Ivoire: a randomised trial. Cotrimo-CI Study Group.

BACKGROUND: In sub-Saharan Africa, various bacterial diseases occur before pneumocystosis or toxoplasmosis in the course of HIV-1 infection, and are major causes of morbidity and mortality. We did a randomised, double blind, placebo-controlled clinical trial at community-health centres in Abidjan, Côte d'Ivoire, to assess the efficacy of trimethoprim-sulphamethoxazole (co-trimoxazole) chemoprophylaxis at early stages of HIV-1 infection. METHOD: 843 HIV-infected patients were screened and 545 enrolled in the study. Eligible adults (with HIV-1 or HIV-1 and HIV-2 dual seropositivity at stages 2 or 3 of the WHO staging system) received co-trimoxazole chemoprophylaxis (trimethoprim 160 mg, sulphamethoxazole 800 mg) daily or a matching placebo. The primary outcome was the occurrence of severe clinical events, defined as death or hospital admission irrespective of the cause. Analyses were by intention to treat. FINDINGS: Four of the randomised patients were excluded (positive for HIV-2 only). 120 severe events occurred among 271 patients in the co-trimoxazole group and 198 among 270 in the placebo group. Significantly fewer patients in the co-trimoxazole group than in the placebo group had at least one severe event (84 vs 124); the probability of remaining free of severe events was 63.7% versus 45.8% (hazard ratio 0.57 [95% CI 0.43-0.75], p=0.0001) and the benefit was apparent in all subgroups of initial CD4-cell count. Survival did not differ between the groups (41 vs 46 deaths, p=0.51). Co-trimoxazole was generally well tolerated though moderate neutropenia occurred in 62 patients (vs 26 in the placebo group). INTERPRETATION: Patients who might benefit from co-trimoxazole could be recruited on clinical criteria in community clinics without knowing the patients CD4-cell count. This affordable measure will enable quick public-health intervention, while monitoring bacterial susceptibility and haematological tolerance.