Search

To determine the component costs of care to optimize treatment with limited resources.

Background Our goal was to determine the incidence rate and risk factors for loss to follow-up (LTFU) of HIV-infected patients in Northern France.Methods We estimated the incidence rate of LTFU in 1,007 HIV-infected patients under care from January 1997 to December 2006. We then investigated potential risk factors for LTFU at inclusion and during follow-up.Results The incidence of LTFU was estimated to be 3.5 per 100 person-years. Risk factors for LTFU at enrolment in a multivariate Cox model were age <30 years (hazard ratio [HR] 1.66 versus >40 years, 95% confidence interval [CI] 1.04–2.64), transmission by injection drug use (HR 5.26 versus men who have sex with men, 95% CI 2.90– 9.52), no phone number provided (HR 5.4, 95% CI 3.6–8.2), no primary care physician (HR 2.10, 95% CI 1.25–3.52) and sub-Saharan African origin (HR 2.09, 95% CI 1.36–3.22). Patients with CD4+T-cell counts <200 cells/mm3(HR 0.49 versus =350 cells/mm3, 95% CI 0.32– 0.76) and 200– 349 cells/mm3at baseline (HR 0.63 versus =350 cells/mm3, 95% CI 0.41–0.98) had a decreased risk of LTFU. During follow-up, the risk of LTFU increased when the most recent CD4+T-cell count was <200 cells/ mm3(HR 2.06, 95% CI 1.16–3.66), the patient was not on highly active antiretroviral therapy (HAART; HR 4.20, 95% CI 2.66–6.61) and the patient was on HAART but had a detectable viral load (HR 1.92, 95% CI 1.19–3.01).Conclusions Our findings will help clinicians recognize patients who require additional support for retention in care, including younger patients, injection drug users, people of sub-Saharan African origin, patients who are healthier at enrolment and patients who do not adhere to HAART during follow-up.

Objective The effectiveness of antiretroviral treatment (ART) was compared in 416 naive patients from a French clinical cohort infected with B and non-B HIV-1 subtypes.Methods Time to HIV viral load (VL) undetectability was calculated for each subtype group. Three other parameters were estimated 3, 6 and 12 months after enrolment: clinical progression (that is, AIDS-defining events or death), changes in CD4 cell counts from baseline and proportion of patients achieving an undetectable VL (<400 HIV-RNA copies/ml).Results In this cohort, 317 patients (76%) were infected with a B subtype and 99 (24%) with a non-B subtype. Median time to VL undetectability was similar in the B subtype group [147 days, 95% confidence interval (CI) 119–165] and non-B subtype group (168 days, 95% CI: 105–234; P=0.16). After adjusting for AIDS-defining events at enrolment, baseline CD4 cell counts and VL, and for the treatment on which patients were initiated, no association was found between HIV subtypes and time to VL undetectability (B subtype vs non-B subtype: hazard ratio=0.80, 95% CI: 0.62–1.02, P=0.07). In the 3, 6 and 12 months after enrolment, subtype had no impact on clinical progression, CD4 cell count or VL responses to ART. This suggests that B and non-B subtypes do not affect first-line therapy efficacy, which is encouraging in view of the worldwide spread of non-B HIV-1 subtypes and the increasing availability of ART in developing countries. However, in this study we did not take into account individual non-B subtype species, therefore further studies should be designed to evaluate the efficacy of these regimens in patients with particular non-B subtypes.

En France, une prothèse articulaire (PA) est implantée chez 5 à 7 % des sujets de plus de 60ans. Le taux global d’infection de ces PA (IPA) est estimé à 1–2 %, cependant l’incidence des IPA d’origine hématogène (IPA-H) n’est pas connue. Notre étude avait pour but d’évaluer l’incidence des IPA-H au décours d’une bactériémie.