Portal de Periódicos da CAPES

Indirect markers to initiate highly active antiretroviral therapy in a rural African setting

2004; Lippincott Williams & Wilkins; Volume: 18; Issue: 8 Linguagem: Inglês

10.1097/00002030-200405210-00025

1473-5571

Jean-Michel Tassie, Tonia Marquardt, Henri Damisoni, Oola Didakus Odhiambo, Maryline Mulemba, Elisabeth Szumilin, Dominique Legros,

HIV-related health complications and treatments

In 2002, the World Health Organization (WHO) estimated that six million people are in need of life-sustaining highly active antiretroviral therapy (HAART) in the developing world [1]. Laboratory and economic constraints require the consideration of simple markers in initiating and monitoring HAART [2]. WHO recommends initiating HAART on either WHO clinical stage 4 or a CD4 cell count of less than 200 cells/μl. The WHO simplified criteria consider stage 2 or 3 with a total lymphocyte count (TLC) of less than 1200 cells/μl, or stage 4 [1]. We evaluated the value of WHO clinical staging, body mass index (BMI), haemoglobinemia and TLC as markers for initiating HAART in a rural African project. The study was conducted in Chiradzulu District Hospital, Malawi. HIV-positive adults seen at post-test counselling or follow-up consultation were included. The WHO clinical staging was assessed by a medical interview and examination [3]. Diagnoses were mainly presumptive. TLC was obtained by manual differential count on whole blood cells and CD4 cell counts using the Facscount system (Becton Dickinson, Mountain View, CA, USA). Different cut-offs and combinations of markers were tested to reach the highest sensitivity to identify patients with a CD4 cell count of less than 200 cells/μl. A total of 206 patients were included, 135 (66%) following post-test counselling and 71 in median 5 months after testing. The median age was 32 years and 121 (59%) were women. Overall, 125 patients (61%) were eligible for HAART (stage 4 or CD4 cell count < 200 cells/μl). Forty-three patients (21%) were asymptomatic, 27 (13%) at clinical stage 2, 81 (39%) at stage 3, and 55 (27%) at stage 4. Overall, 106 patients (52%) had a CD4 cell count of less than 200 cells/μl [18 (25%), 52 (64%) and 36 (66%) of patients at stages 1–2, 3 and 4, respectively]. The median CD4 and total lymphocyte counts were 336 and 1800 cells/μl at stages 1–2, 151 and 1400 cells/μl at stage 3, and 103 and 1400 at stage 4. The CD4 cell count was less than 200 cells/μl in 38% (8/21) of patients with Kaposi's sarcoma and in 61% (17/28) of patients with pulmonary tuberculosis. The CD4 cell count was weakly correlated with BMI (r = 0.33), haemoglobinemia (r = 0.38) and TLC (r = 0.51) (P < 0.0001 for each). To identify patients with less than 200 CD4 cells/μl, stage 3 or stage 4 presented with similar predictive values (Table 1). The algorithm stage 3 or 4 or TLC of less than 1200 cells/μl had a sensitivity of 93% and a specificity of 49%. There was no significant difference by sex in sensitivity (P = 0.8) or specificity (P = 0.3). The algorithm stage 4 or stages 2–3 with a TLC of less than 1200 cells/μl presented with a sensitivity of 61%. Combined with clinical stage and TLC, BMI and haemoglobinemia added little in detecting severe immunodeficiency.Table 1: Sensitivity, specificity, positive and negative predictive values of World Health Organization clinical stage and total lymphocyte count for identifying patients with less than 200 CD4 cells/μl.This study was conducted in a rural setting in Africa with very limited access to laboratory facilities. Adults were included shortly after learning their HIV-positive status and 61% were in urgent need of HAART. WHO clinical stages 3 or 4 or a TLC of less than 1200 cells/μl was the most sensitive algorithm to identify severe immunodeficiency. The WHO simplified criteria appeared poorly sensitive, as it considers stage 3 if the TLC is below 1200 cells/μl. It would have delayed treatment in a third of patients. In our context, stages 3 or 4 presented the same predictive values for severe immunodeficiency, and therefore for initiating treatment. As a result of the limited investigations available, we certainly underestimated stage 4, which is defined by accurate diagnosis, whereas stage 3 is defined on symptoms easily identified [3]. This situation reflects daily clinical practice in peripheral health facilities. Our results, based on a small sample of patients, require further validation. However, in a cohort in Rwanda, survival was as similar at stages 3 or 4 [4]. Recently, international guidelines to initiate HAART were assessed in a cohort of 155 Ethiopians followed without treatment. With WHO simplified criteria, 41% of the adults who died never met the treatment indication [5]. TLC is known to be correlated with CD4 cell count [6,7] and is of special interest in deciding when to initiate HAART [8]. For now, manual counting is only occasionally available in rural settings and is time consuming. Therefore, it seems difficult to consider this investigation systematically before initiating HAART. BMI or haemoglobinemia added little to the decision to treat in our sample. How these markers perform for treatment monitoring requires further research. In rural settings, the scarcity of resources and the massive need for treatment require markers as simple as possible to initiate HAART. WHO clinical stages 3 or 4 thus appear to be a valuable universal marker. When technically feasible, TLC could improve the screening for treatment in patients at stages 1 or 2. In any case, the lack of a laboratory should not be an obstacle in accessing treatment.