Early-Onset Ventilator-Associated Pneumonia in Patients with Severe Traumatic Brain Injury: Incidence, Risk Factors, and Consequences in Cerebral Oxygenation and Outcome. Esnault P, Nguyen C, Bordes J, D'Aranda E, Montcriol A, Contagyris C, Cotte J, Goutorbe P, Joubert C, Dagain A, Boret H, Meaudre E. Neurocrit Care. 2017 Oct;27(2):187-198.
Early onset ventilator associated pneumonia (EOVAP) is known to frequently complicate severe traumatic brain injuries (TBI). The available literature describes an incidence as high as 65%. With these occurrences complicating the early post injury period (< 7 days), they can have a significant impact on morbidity and are known to increase ventilator days and incidence of ARDS. Previous literature; however, has not demonstrated an increased risk of mortality or worse neurological outcome.
The authors in this manuscript attempt to describe the incidence and characteristics of EOVAP, specific risk factors associated with EOVAP, and impact on cerebral oxygenation as well as neurological outcome. To this end, they conducted a retrospective, observational review study of prospectively collected data on intubated patients with TBI at a single level 1 trauma center. Subjects >18 years old with TBI were eligible for enrollment over a 7 year period (2007-13). Inclusion criteria were: severe TBI (defined as GCS <8 after initial resuscitation) and ≥ 2 days of mechanical ventilation. Patients were excluded for GCS score >8 to ensure a severely injured cohort and all deaths within 48 hours of admission were excluded to avoid inclusion of patients deemed to have non-survivable injuries. The patients were all managed within acceptable guidelines for VAP prevention (i.e. HOB elevation, routine oral care, etc.) and stress ulcer prophylaxis. Notably, antibiotic prophylaxis was administered for 48h at attending physician discretion. On admission, all patients underwent blind tracheobronchial aspiration with quantitative cultures. TBI management for all patients adhered to a standard progression that is well described in the manuscript. The clinical pulmonary infection score (CPIS) was evaluated daily for the first 7 days as a screen for clinical pneumonia. If score was ≥ 5, a bronchoscopy guided BAL was performed. EOVAP was diagnosed when CPIS was >6 and the quantitative culture showed pathogenic organisms (≥104 CFU/ml). Interestingly, cerebral oxygenation was monitored at the discretion of the attending physician on a subset of patients and included in the analysis.
Of 243 intubated TBI patients admitted to the ICU over the study period, 175 patients met inclusion criteria and were analyzed. The cohort was overwhelmingly male and the most common mechanism of injury was motorcycle crash. The median age was 37, with median ISS=22 (Range: 16-34) and a median GCS score of 6 (Range: 4-8). EOVAP was diagnosed in 106 patients for an overall incidence of 60.6%. Notably, nearly all EOVAP patients had ARDS (92.5%) by Berlin criteria with 70.8% defined as moderate to severe ARDS. Patients with EOVAP were more severely injured (ISS 25 vs. 18; p=0.009) and more frequently underwent therapeutic hypothermia, barbiturate infusion and decompressive craniectomy. After multivariate logistic regression analysis, four risk factors were identified for EOVAP: therapeutic hypothermia use, thoracic AIS score >3, a positive quantitative admission ET aspirate and pre-intubation gastric aspiration events. Interestingly, prophylactic antibiotic use administered during the first 48 hours was found to be a protective factor against EOVAP. For the subset of patients with PbtO2 data available, EOVAP was shown to have deleterious effects on cerebral oxygenation. The PbtO2 was significantly lower around the time of EOVAP diagnosis and there were significantly more brain hypoxia episodes. Finally, EOVAP was found to be an independent risk factor associated with unfavorable neurological outcome at 1 year follow-up as measured by Glasgow outcome scale score (GOS).
This single center study has the inherent limitations on generalizability of any single center study. Nonetheless, Esnault et. al. should be commended on a very rigorous and well-designed study. They effectively characterize the EOVAP risk factors one should consider in caring for these patients as well as the potential effects on morbidity and long term outcome. Importantly, this study posits a potential intervention – antibiotic prophylaxis – that may reduce the impact and incidence of this serious complication.
Systemic antibiotics for preventing ventilator-associated pneumonia in comatose patients: a systematic review and meta-analysis. Righy C, Brasil P, Valles J, Bozza FA, Loeches IM. Ann Intensive Care. 2017 Dec;7(1):67.
As described in the previous article by Esnault et. al., early onset ventilator associated pneumonia (EOVAP) is prevalent in severe TBI patients. Similar observations have been made in other critically ill, comatose patient populations. Additionally, EOVAP is known to be a leading cause of mortality in these patients. While VAP prevention bundles are an important set of practices for preventing late onset VAP, they are likely less effective with prevention of EOVAP in comatose patients. Given the high prevalence of EOVAP in these compromised patients, more effective interventions are warranted. Prophylactic systemic antibiotics have been described as a potential intervention for the prevention of EOVAP.
The authors undertook a review of the available published literature through December 2015 to evaluate the potential efficacy of prophylactic systemic antibiotics in preventing EOVAP in comatose ICU patients. Two authors queried PubMed & Cochrane databases to identify and evaluate full length articles of randomized clinical trials or prospective cohorts that studied the effect of systemic antibiotics at the time of intubation (or just before) on EOVAP compared with placebo or no prophylaxis.
After an exhaustive search that met standard literature search and systematic review standards (PRISMA), 11,340 citations were identified. 26 manuscripts received detailed analysis by two authors. Inclusion criteria were based on PICO approach. Inclusion criteria were as follows: 1) Clinical trials or prospective cohorts 2) Population-adults comatose patients 3) Intervention- systemic antibiotic prophylaxis at time of intubation or just prior 4) Control group- patients with no intervention or placebo 5) Outcome- VAP incidence, ICU and hospital mortality, hospital LOS and ventilator days. Excluded studies did not report enough data to determine odds ratio or relative risks. Of the 26 manuscripts that were selected for detailed analysis, only 3 met inclusion criteria. These 3 studies included a total of 267 patients which were the basis of the systematic review and statistical meta-analysis. The authors report detailed data from these three studies in the manuscript. Notably, only one of the studies was exclusively TBI trauma patients. Two studies were organized as randomized clinical trials. The other had a prospective cohort, but was compared to historic controls.
The statistical meta-analysis using a fixed effect model showed a relative risk reduction of EOVAP of 0.32 (C.I. 0.19-0.54) in favor of systemic antibiotic prophylaxis. In effect, the meta-analysis demonstrated 1/3 lower occurrence of EOVAP in the intervention group. ICU mortality; however, was not improved (RR 1.03, CI 0.7-1.53) with the prophylactic treatment. This finding is consistent with past EOVAP literature. Similarly, antibiotic administration did not affect duration of mechanical ventilation. The analysis did show a modest, but statistically significant decrease in ICU LOS of 9 hours. This result is likely not clinically significant though.
The overall meta-analysis – though statistically sound – suffers from a lack of included studies and the resultant low number of patients complicates interpretation of the data and may artificially amplify the hypothetical impact of prophylaxis. Also, only 2 of the studies have similar study design with one using historical controls. Thus it is impossible to make strong recommendations on this potential intervention based on this meta-analysis alone. Additionally, the antibiotic regimens used were all different which introduces additional heterogeneity to the results. However, this study does argue for more research on prophylactic antibiotics as an EOVAP prevention strategy given the paucity of available, quality data. A large randomized trial is needed to further evaluate this question for in hospital morbidity of this complication as well as long term functional outcomes.
Hemodynamic Assessment of Patients With Septic Shock Using Transpulmonary Thermodilution and Critical Care Echocardiography: A Comparative Study. Vignon P, Begot E, Mari A, Silva S, Chimot L, Delour P, Vargas F, Filloux B, Vandroux D, Jabot J, François B, Pichon N, Clavel M, Levy B, Slama M, Riu-Poulenc B. Chest. 2018 Jan;153(1):55-64.
A prospective, multi-institutional (5 French hospitals) observational study was performed of 137 critically ill patients with septic shock who underwent concurrent Transpulmonary Thermodilution (TPT) and Critical Care Echocardiography (CCE) in order to determine the concordance between the two modalities in ventilated patients. The inclusion criteria allowed for mechanically ventilated patients with arterial and central venous catheters. Septic shock was defined and treated according to the Surviving Sepsis guidelines. Subjects were excluded if not in normal sinus rhythm or less than 18 years of age, pregnant, or had contra-indications for TEE or catheter placement. TPT and CCE were performed successively in random order; two assessments of each were performed within 19±17 minutes. The mean age of the patients was 61±15 years; 52% male. Mean SOFA score was 10±3. 8 patients had acute cor pulmonale (ACP) The TEP and CCE interpretation agreed in 66% of patients with a kappa agreement at a moderate level (0.48; 95% CI 0.37-0.60) in patients with ACP but improved to a good level of agreement if ACP was not present (0.66; 95% CI 0.55-0.77). The serum lactate at 6 hours was similar regardless of the TPT and CCE results. Mortality was the same whether the CCE and TPT interpretations were concordant or discordant at 36% each (p=0.99). There were no complications related to either technique. The authors concluded that there was reasonable concordance between the modalities but that either method could not predict lactate clearance or mortality. There is currently no role for use as an endpoint of resuscitation, particularly in light of the current lack of validated algorithms to drive resuscitation.
Respiratory System Mechanics During Low Versus High Positive End-Expiratory Pressure in Open Abdominal Surgery: A Substudy of PROVHILO Randomized Controlled Trial. D'Antini D, Huhle R, Herrmann J, Sulemanji DS, Oto J, Raimondo P, Mirabella L, Hemmes SNT, Schultz MJ, Pelosi P, Kaczka DW, Vidal Melo MF, Gama de Abreu M, Cinnella G; European Society of Anaesthesiology and the PROtective VEntilation Network. Anesth Analg. 2018 Jan;126(1):143-149.
The PROtective Ventilation using High versus Low positive end-expiratory pressure (PROVHILO) study was a randomized, double-blind, controlled clinical trial compared intra-operative low tidal volume/high PEEP (PEEP) with recruitment maneuver ventilation (12 cm H2O) to low PEEP (0-2 cm H2O) without recruitment maneuver ventilation in 900 subjects at two sites. There was no difference in post-operative pulmonary complications despite the ventilator strategy utilized but high PEEP had a greater rate of intra-operative hemodynamic instability. The current paper performed a subset analysis of these patients who underwent elective, open abdominal procedures under general anesthesia who were at high risk for post-op pulmonary complications based on the Assess Respiratory Risk in Surgical Patients in Catalonia (ARISCAT) score ≤26. The primary aim was to determine the incidence of intra-tidal recruitment/decruitment (R/D) and alveolar over-distension between these strategies with 18 patients in each cohort (36 total patients). Patients were excluded if they were less than 18 years old, had a BMI greater than 40, were pregnant, prior mechanical ventilation more than 30 minutes within 30 days, and severe cardiopulmonary comorbidities.The primary endpoints were measured using the volume-dependent elastance index (%E2) which is a non-invasive, radiation free assessment of dominant overdistension and intratidal R/D. The elastance was greater lower PEEP cohort (18.6 [16…22] vs 13.4 [11.0…17.0] cm H2O·L−1; P < .01) as was the %E2 (−15.4 [−28.0…6.5] vs 6.2 [−0.8…14.0] %; P < .05). The intratidal R/D was greater in the lower PEEP group (61% vs 22%; P = .037) but there was no difference in intratidal overdistension between cohorts (6% vs 6% p=1.0). The authors concluded that lung recruitment and a PEEP strategy of 12 cm H2O decreased intratidal R/D without affecting overdistension.