Fatores de risco de letalidade precoce em pacientes com alta mortalidade em unidade de terapia intensiva com bacteremia por acinetobacter bacmannii resistente a carbapenêmicos

Abstract

Acinetobacter baumannii carbapenem-resistant (CRAB) bacteremias in intensive care units (ICU) present a crude mortality varying from 30 to 79.8% and attributed mortality 58.2%. In the first 72 hours, the lethality can reach 50%. Objective: To identify the risk factors of early lethality in patients with high mortality in intensive care unit with Acinetobacter baumannii carbapenem-resistant bacteremia. Casuistics and Methods: This is a retrospective cohort study with a nested control case study performed in a tertiary hospital with ICU patients who developed CRAB bacteremias from January 2011 to December 2014. To determine the risk factors associated to lethality, two groups were compared: those that survived and those that died, called population 1. In the sequence it was analyzed separately the patients who died; population 2, a subgroup being the patients who died in the first 72 hours, after the isolation of the bacteria, called cases and controls those that survived after 72 hours. A 21-day survival analysis was performed. Lethality risk factors were evaluated by univariate analysis and sequentially performed the logistic regression. Results: Out of the 72 patients, 57 (79.2%) died and 15 (20.8%) survived. In population 1, no significant difference was observed in relation to variables: sex, age, comorbidities, origin of bacteremias, clinical diagnoses and treatment used. The only significant variable was APACHEII in the univariate analysis of this population at the time of blood culture (P = 0.007). In the logistic regression, cardiovascular disease (OR = 9.65, 95% CI: 1.00-93.01, P = 0.049) and APACHEII at the time of blood culture collection (OR = 11.24, CI 95%: 1.94-65.04, P = 0.006). Sepsis with septic shock was a protective factor for the surviving group (OR = 0.12, 95% CI: 0.02-0.55, P = 0.005). In the population 2, the significant variables for the univariate analysis were: undetermined origin of bacteremia (P = 0.039), pneumonia (P = 0.045), previous use of polymyxin (P = 0.001), carbapenemics and previous polymyxin use (P = 0.001), specific treatment with polymyxin (P = 0.023), hemodialysis and polymyxin (P = 0.032), APACHEII at the time of blood culture (P = 0.0016). In the logistic regression, the use of previous polymyxin (OR = 0.01, 95% CI: 0.00-0.12, P <0.001) and the specific treatment (OR = 0.03, 95% CI: 0.00 (OR = 4.99, 95% CI: 1.05-23.66, P = 0.042, P <0.001) were factors that have increased survival and APACHEII at the time of blood culture collection was associated with a higher lethality (OR = 4.99, 95% CI: 1.05-23.66, P = 0.042). Conclusion: Crude mortality rate was 79.2% and the early lethality of the third day was 54.4%. Severity was an independent predictor of mortality in both populations. Septic shock was a protective factor in population 1, we believe that hemodynamic and advanced life support in this population may have increased survival. Treatment was a protective factor in the population 2. Maintaining strict antimicrobial control and intensifying preventive measures in the ICU is a priority. Further studies assessing the hemodynamic support and therapeutic efficacy of polymyxin in this population are required.