Diarréia nosocomial em unidade de terapia intensiva: incidência e fatores de risco

Abstract

Background - Nosocomial diarrhea is one of the most common infection diseases among patients admitted at Intensive Care Units (ICUs), although its epidemiology as well as its causes and consequences are rarely documented in hospital infections surveillance studies. Objective - To study the incidence and the risk factors for diarrhea among adult patients admitted at a clinical-surgical ICU for a 12-months period. Patient and methods – Four-hundred-fifty-seven patients were studied prospectively, after application of exclusion criteria, to determine diarrhea incidence and risk factors from October 2005 to October 2006. Nosocomial diarrhea was defined as clinical picture of diarrhea beginning 72 hours from hospital admission, with at least two watery or pasty bowel movements for more than two consecutive days. Data acquisition was done on a daily basis at ICU until patient discharge or death in the unit. Variables analyzed were the following: age, sex, days of hospitalization before admission at the ICU, days of hospitalization at the ICU, APACHE II score, comorbidities, organic insufficiency, surgical history for the last 30 days, current use of antimicrobials (including number of days and total number of antibiotics), antisecretories, enteral nutrition, parenteral, fleet enemas, total duration of diarrhea (in days), period of time between patient admission and beginning of diarrhea, presence of similar case of diarrhea in ICU during the last week or at the same time, serum albumin and hemoglobin levels in at least one occasion during admission. For data analysis, the patients were divided into groups with and without nosocomial diarrhea. The level of significance was 5% (P <0.05). The significant variables in univariate analysis were analyzed, in a multivariate form, through estimation of a logistic regression model. Results - Diarrhea occurred in 135 (29.5%) patients and lasted for a mean of 5.4 days. The period of time between patient admission and beginning of diarrhea was 17.8 days and similar cases during the period were recorded in 113 (83.7%) patients. In multivariate analysis using logistic regression model only the number of antibiotics (OR 1.65; C.I. 95% = 1.39 – 1.95) and the number of days of antibiotic therapy (OR 1.16; C.I. 95% = 1.12 – 1.20) were statistically associated with the occurrence of diarrhea (P <0.05). Each day added of antibiotic therapy increased in 16% the risk of diarrhea (C.I. 12% to 20%), while the

10 addition of one more antibiotic to the scheme increased the chances of occurring diarrhea in 65% (C.I. 39% to 95%). Conclusion - The incidence of nosocomial diarrhea in ICU is high (29.5%). The main risk factors were number of prescribed antibiotics and duration of antibiotic therapy. Besides the enteric precautions, judicious and limited prescription of antimicrobials will probably decrease the incidence of diarrhea at this unit.