Ventilator-associated pneumonia due to acinetobacter baumannii in intensive care units: Secondary bacteremia risk factors and predictors of mortality

Acinetobacter baumannii is a multidrug resistant (MDR), gram-negative bacterium commonly implicated in ventilator associated pneumonia (VAP) in critically ill patients hospitalized in intensive care units (ICUs). In recent years, ventilator-associated pneumonia and hospital-acquired pneumonia (HAP) in ICUs, despite significant progress in diagnosis, treatment and protection measures, also continues to be the most common and lethal nosocomial infections. Patients in the intensive care unit with VAP often subsequently develop Acinetobacter bacteremia, which may significantly worsen outcomes. It is not clear which patients will develop bacteremia or which patients will have a mortal outcome. In our study, we aimed to determine bacteremia risk factors and mortality predictors by grouping Acinetobacter VAP patients according to bacteremia variable and mortality variable. Thus, patients will be predicted before diagnosis and prognosis will be improved with early empirical treatment. The study was carried out retrospectively on the patients diagnosed with Acinetobacter spp. VAP between 1st of January 2014 and 30th of June 2019 in the tertiary intensive care unit of our hospital. In our study 231 attacks of 214 patients which recieved our inclusion and exclusion criteria were assessed. Patients were grouped according to bacteremia and mortality variables. In our study, 125 attacks in the VAP group (control) and 106 attacks in the group bacteremia secondary to VAP (case) were evaluated. All attacks were VAP with late onset (starting at least 4 days after intubation). A total of 231 attacks were included in the study: 125 attacks in group VAP (VAP with positive sputum for Acinetobacter spp.) and 106 attacks in group secondary bacteremia (VAP with positive sputum and blood culture for Acinetobacter spp.). In all 231 attacks, VAP with positive sputum for Acinetobacter spp. was defined as late onset (developed more than 4 days after initiation of mechanical ventilation). Accordingly, in the comparison of bacteremic and nonbacteremic attacks, 54.1 % (n = 125) were found to be nonbacteremic, 45.9 % (n = 106) were found to be bacteremia secondary to VAP. The median age and gender distribution of the attacks were similar. Peripheral venous catheter usage rate and on total parenteral nutrition rate were significantly higher in the bacteremia group than the group nonbacteremia (p = 0.001, p = 0.000). The median values of APACHE-II score in the nonbacteremia group were found to be significantly lower than in the bacteremia group (24 vs. 28, p < 0.001). The median SOFA score on admission to the ICUs of the patients in both groups were mostly statistically similar (6 vs. 5 p = 0.173), as well as charlson index distribution was similar. In the analyzes performed according to the mortality variable, 112 attacks in nonsurviving patients and 119 attacks in surviving patients were evaluated. The median age of the nonsurviving patient group was significantly higher than that of the surviving group (78 vs. 69 years, p < 0.001), similarly, the male gender ratio was higher in the nonsurvival group (72.3 % vs. 53.6 %, p = 0.003). The median hospital length of stay in the nonsurviving group was found to be significantly higher than the surviving group (59 vs. 40 days, p < 0.001). Similarly, the median ICU length of stay was significantly higher (39 vs. 29.5 days, p = 0.006). The rate of peripheral arterial catheter usage in the nonsurviving group was found to be significantly lower than that of surviving patients (83 % vs. 94.1 %, p = 0.008). The rate of percutaneous endoscopic gastrostomy (PEG) usage was significantly higher in the nonsurviving patient group (10.1 % vs. 1.8 %, p = 0.008). Similarly, the rate of steroid usage was found to be significantly higher in nonsurviving patients (8.5 % vs. 0 %, p = 0.002). The sensitivity of APACHE-II score (on first positive sputum culture for Acinetobacter spp. culture in VAP group and on first positive blood culture for Acinetobacter spp. culture in bacteremic group) above 22 for mortality sensitivity was 97.3 %, specificity was 41.2 %, positive predictive value (PPV) was 60.9 % and negative predictive value (NPD) was 94.2 %. The sensitivity of SOFA 2 score (on first positive sputum culture for Acinetobacter spp. culture in VAP group and on first positive blood culture for Acinetobacter spp. culture in bacteremic group) above 8 for mortality sensitivity was 83 %, specificity was 70.6 %, PPV was 54 % and NPD was 52.1 %. The Charlson score of 4 and above had sensitivity of 74.1 %, 52.1 % specificity, 58.7 % PPV and 64.9 % NPV for mortality. As a result, total parenteral nutrition and venous catheter were identified as risk factors for the development of bacteremia in Acinetobacter VAP patients. While the presence of bacteremia, immunosupression, age and ICU length of stay were found to be risk factors for ICU mortality, peripheral arterial catheterization was evaluated as a protective factor. In addition, severity of disease (APACHE > 22 and SOFA > 8 and the Charlson score ≥ 4) was thought to be independent predictors for mortality in our critically ill population.