Background: In developing countries, access to laboratory tests remains limited, and the use of simple tools such as weight to monitor HIV-infected patients treated with antiretroviral therapy should be evaluated.
Methods: Cohort study of 2451 Cambodian and 2618 Kenyan adults who initiated antiretroviral therapy between 2001 and 2007. The prognostic value of weight gain at 3 months of antiretroviral therapy on 3–6 months mortality, and at 6 months on 6–12 months mortality, was investigated using Poisson regression.
Results: Mortality rates [95% confidence interval (CI)] between 3 and 6 months of antiretroviral therapy were 9.9 (7.6–12.7) and 13.5 (11.0–16.7) per 100 person-years in Cambodia and Kenya, respectively. At 3 months, among patients with initial body mass index less than or equal to 18.5 kg/m2 (43% of the study population), mortality rate ratios (95% CI) were 6.3 (3.0–13.1) and 3.4 (1.4–8.3) for those with weight gain less than or equal to 5 and 5–10%, respectively, compared with those with weight gain of more than 10%. At 6 months, weight gain was also predictive of subsequent mortality: mortality rate ratio (95% CI) was 7.3 (4.0–13.3) for those with weight gain less than or equal to 5% compared with those with weight gain of more than 10%.
Conclusion: Weight gain at 3 months is strongly associated with survival. Poor compliance or undiagnosed opportunistic infections should be investigated in patients with initial body mass index less than or equal to 18.5 and achieving weight gain less than or equal to 10%.
aUnité d'Epidémiologie des Maladies Emergentes, Institut Pasteur, 25-28 rue du Docteur Roux, France
bMédecins Sans Frontières, 8 rue Saint Sabin, Paris, France
cMédecins Sans Frontières, Nairobi, Kenya
dInfectious Diseases Department, Khmero-Soviet Friendship Hospital, St. 134 Sangkat Vealvong Khan 7 Makara, Phnom Penh, Cambodia
eEpicentre, 42 bis boulevard Richard Lenoir, Paris, France.
Received 11 September, 2008
Revised 18 December, 2008
Accepted 23 December, 2008
Correspondence to Yoann Madec, Unité d'Epidémiologie des Maladies Emergentes, Institut Pasteur, 25-28 rue du Docteur Roux, 75015 Paris, France. E-mail: email@example.com