Regional Cerebral Oxygen Saturation Level Predicts 30-Day Mortality Rate After Left Ventricular Assist Device Surgery.

OBJECTIVE: Left ventricular assist device (LVAD) surgery is complex, high risk, and expensive. The authors' hypothesis is baseline regional cerebral oxygen saturation (rSO2 ) might be a predictor of postoperative clinical outcomes.

DESIGN: Retrospective review of 210 consecutive continuous flow LVAD patients between 2008 and 2014. The primary measure is 30-day mortality rate and secondary measures include modified major adverse cardiocerebral events (MACE), length of stay (LOS), and intensive care unit (ICU) stay. Multiple logistic regression models were applied to examine if a binary outcome variable, such as 30-day mortality and MACE, is associated with rSO2 at baseline. Log-linear model was used to examine whether LOS or ICU stay hours is associated with rSO2 at baseline.

SETTING: Single institution, academic hospital.

PARTICIPANTS: Patients who received LVAD surgery ​at Jewish Hospital, Louisville, KY.

INTERVENTIONS: All patients received LVAD surgery. Cerebral oximetry monitoring was used in both the preoperative and intraoperative periods.

MEASUREMENTS AND MAIN RESULTS: The authors found that higher rSO2 at baseline is associated with lower 30-day mortality with an odds ratio of 0.94 and 95% confidence interval (0.888, 0.995) for every 1% increase of rSO2 . For secondary outcomes, baseline rSO2 was not significantly associated with MACE, requirement for postoperative renal failure/dialysis, reoperation for bleeding, and LOS or ICU hours.

CONCLUSIONS: Regional cerebral oxygen saturation levels at baseline are significantly associated with 30-day mortality after LVAD surgeries.