Early-Onset Ventilator-Associated Pneumonia in Patients with Severe Traumatic Brain Injury: Incidence, Risk Factors, and Consequences in Cerebral Oxygenation and Outcome.

BACKGROUND: Early-onset ventilator-associated pneumonia (EOVAP) occurs frequently in severe traumatic brain-injured patients, but potential consequences on cerebral oxygenation and outcome have been poorly studied. The objective of this study was to describe the incidence, risk factors for, and consequences on cerebral oxygenation and outcome of EOVAP after severe traumatic brain injury (TBI).

METHODS: We conducted a retrospective, observational study including all intubated TBI admitted in the trauma center. An EOVAP was defined as a clinical pulmonary infection score >6, and then confirmed by an invasive method. Patient characteristics, computed tomography (CT) scan results, and outcome were extracted from a prospective register of all intubated TBI admitted in the intensive care unit (ICU). Data concerning the cerebral oxygenation monitoring by PbtO2 and characteristics of EOVAP were retrieved from patient files. Multivariate logistic regression models were developed to determine the risk factors of EOVAP and to describe the factors independently associated with poor outcome at 1-year follow-up.

RESULTS: During 7 years, 175 patients with severe TBI were included. The overall incidence of EOVAP was 60.6% (47.4/1000 days of ventilation). Significant risk factors of EOVAP were: therapeutic hypothermia (OR 3.4; 95% CI [1.2-10.0]), thoracic AIS score ≥3 (OR 2.4; 95% CI [1.1-5.7]), and gastric aspiration (OR 5.2, 95% CI [1.7-15.9]). Prophylactic antibiotics administration was a protective factor against EOVAP (OR 0.3, 95% CI [0.1-0.8]). EOVAP had negative consequences on cerebral oxygenation. The PbtO2 was lower during EOVAP: 23.5 versus 26.4 mmHg (p <0.0001), and there were more brain hypoxia episodes: 32 versus 27% (p = 0.03). Finally, after adjusting for confounders, an EOVAP was an independent factor associated with unfavorable neurologic functional outcome at the 1-year follow-up (OR 2.71; 95% CI [1.01-7.25]).

CONCLUSIONS: EOVAP is frequent after a severe TBI (overall rate: 61%), with therapeutic hypothermia, severe thoracic lesion, and gastric aspiration as main risk factors. EOVAP had a negative impact on cerebral oxygenation measured by PbtO2 and was independently associated with unfavorable outcome at 1-year follow-up. This suggests that all precautions available should be taken to prevent EOVAP in this population.