Étude des anomalies de la flore vaginale : facteurs de risque, prédicteurs de la récurrence et association avec l'infection à VIH chez des travailleuses du sexe recrutées dans un essai clinique en Inde et dans deux pays africains

Authors: Guédou, Fernand Aimé
Advisor: Alary, Michel
Abstract: Objectives: The present work consisted in the secondary analysis of data from female sex workers (FSW) recruited at two African and two Indian sites, in the context of a clinical trial on the cellulose sulfate vaginal gel, and had as objectives: 1) to identify risk factors for intermediate vaginal flora (IVF) and bacterial vaginosis (BV), 2) to study the association between vaginal flora abnormalities and HIV infection, 3) to identify predictors of recurrence of BV. Methods: The study included two cross-sectional and one longitudinal analyses. Both cross-sectional analyses used socio-demographic, behavioral and biological data from the screening visit of 1367 participants. The first cross-sectional analysis combined the dichotomous and polytomous logistic regressions to examine risk factors for the IVF in connection with those for BV, while the second analysis used the log-binomial regression to investigate the association between abnormal vaginal flora and HIV. The longitudinal analysis used baseline and follow-up data from 440 FSW enrolled in the clinical trial and used Andersen-Gill proportional hazard model to identify predictors of BV recurrence. Results: IVF shares most of its risk or protective factors with BV, and this supports the idea that both are parts of the same disease. Risk factors common to BV and FVI, and particularly those which are more strongly associated with IVF than with BV, are unlikely to be revealed by dichotomous regressions grouping IVF and normal flora as reference category. IVF was associated with HIV infection as strongly as BV with adjusted prevalence ratios (aPR) = 1.56 [95% confidence interval (CI) = 1.22 to 1.98] and aPR = 1.48 (95% CI = 1.20 to 1.84) respectively. Finally, recent vaginal douching and consistent condom use (CCU) with regular partner proved to be the main predictors of recurrence of BV with adjusted hazard ratio (aHR) = 1.30 (95% CI = 1.02 - 1.64) and aHR = 0.68 (95% CI = 0.49 to 0.93) respectively. Conclusion: The dichotomization of abnormal vaginal flora isolating IVF from BV needs to be reconsidered. IVF should be considered alongside BV and treated as such in preventive strategies which should focus on the adverse effects of vaginal douching and the protective effect of CCU with all types of sexual partners.
Document Type: Thèse de doctorat
Issue Date: 2012
Open Access Date: 19 April 2018
Permalink: http://hdl.handle.net/20.500.11794/23717
Grantor: Université Laval
Collection:Thèses et mémoires

Files in this item:
29589.pdf7.39 MBAdobe PDFView/Open
All documents in CorpusUL are protected by Copyright Act of Canada.