Place de l'impédancemétrie dans le dépistage de la dénutrition à l'hôpital

The prevalence of malnutrition is high in patients and tends to worsen during the hospital stay. In the absence of one reliable method to evaluate patients, the assessment of nutritional status is based on a global approach. Body composition measurement by bio-impedance analysis (BIA) is one of these approaches. Body composition measurements can detect malnutrition or abnormal hydration. Fat free mass, fat mass, and total body water are the main body compartments that are evaluated. Determination of abnormal body composition can then guide nutritional support. The reliability of BIA depends on the equation used to predict body composition and the parameters included in the formula (weight, height, sex, age, race, etc.). These parameters allow to minimize measurement errors. Thus, formula developed for specific populations allow to evaluate the nutritional status with reasonable error rates. BIA has been found to be inaccurate with abnormal distribution of body compartments (ascites, dialysis, lypodystrophy, etc.) or extreme weights (cachexia, obesity). Multi-frequency or segmental BIA was developed to overcome hydration abnormalities and variations in body geometry. However, these techniques require further validation. The BIA seems to have some limitations. This review aims to assess the reliability of BIA to detect protein-calorie malnutrition at hospital admission or during nutritional follow-up of patients.     

Les déterminants de la survie des patients vivant avec le VIH sous thérapie antirétrovirale dans la ville de Goma,RD-Congo

BackgroundThe study aimed to identify factors associated with the survival of patients receiving antiretroviral therapy.MethodsA historic cohort of HIV patients from two major hospitals in Goma (Democratic Republic of Congo) was followed from 2004 to 2012. The Kaplan-Meier method was used to describe the probability of survival as a function of time since inclusion into the cohort. The log-rank test was used to compare survival curves based on determinants. The Cox regression model identified the determinants of survival since treatment induction.ResultsThe median follow-up time was 3.56 years (IQR = 2.22–5.39). The mortality rate was 40 deaths per 1000 person-years. Male gender (RR: 2.56; 95 %CI 1.66–4.83), advanced clinical stage (RR: 2.12; 95 %CI 1.15–3.90), low CD4 count (CD4 < 50) (RR: 2.05; 95 %CI : 1.22–3.45), anemia (RR: 3.95; 95 %CI 2.60–6.01), chemoprophylaxis with cotrimoxazole (RR: 4.29, 95 % CI 2.69–6.86) and period of treatment initiation (2010–2011) (RR: 3.34; 95 %CI 1.24–8.98) were statistically associated with short survival.ConclusionInitiation of treatment at an early stage of the disease with use of less toxic molecules and an increased surveillance especially of male patients are recommended to reduce mortality.     

Intérêt de la détection et de la prise en compte d’une surdispersion dans un modèle de Poisson : illustration à partir d’un exemple

BackgroundResearchers often use the Poisson regression model to analyze count data. Overdispersion can occur when a Poisson regression model is used, resulting in an underestimation of variance of the regression model parameters. Our objective was to take overdispersion into account and assess its impact with an illustration based on the data of a study investigating the relationship between use of the Internet to seek health information and number of primary care consultations.MethodsThree methods, overdispersed Poisson, a robust estimator, and negative binomial regression, were performed to take overdispersion into account in explaining variation in the number (Y) of primary care consultations. We tested overdispersion in the Poisson regression model using the ratio of the sum of Pearson residuals over the number of degrees of freedom (χ²/df). We then fitted the three models and compared parameter estimation to the estimations given by Poisson regression model.ResultsVariance of the number of primary care consultations (Var[Y] = 21.03) was greater than the mean (E[Y] = 5.93) and the χ²/df ratio was 3.26, which confirmed overdispersion. Standard errors of the parameters varied greatly between the Poisson regression model and the three other regression models. Interpretation of estimates from two variables (using the Internet to seek health information and single parent family) would have changed according to the model retained, with significant levels of 0.06 and 0.002 (Poisson), 0.29 and 0.09 (overdispersed Poisson), 0.29 and 0.13 (use of a robust estimator) and 0.45 and 0.13 (negative binomial) respectively.ConclusionDifferent methods exist to solve the problem of underestimating variance in the Poisson regression model when overdispersion is present. The negative binomial regression model seems to be particularly accurate because of its theorical distribution ; in addition this regression is easy to perform with ordinary statistical software packages.