Preoperative Left Ventricular Ejection Fraction is Associated With Delayed Extubation After Surgical Correction of Neuromuscular Scoliosis

BACKGROUND: Surgical correction of neuromuscular scoliosis (NMS) carries a high risk of perioperative cardiopulmonary complications, including delayed extubation. This study aims to determine preoperative risk factors of delayed extubation after operative correction of neuromuscular scoliosis. METHODS: A retrospective chart review identified consecutive patients with NMS surgically treated at an urban academic institution. In accordance with institutional protocol, all patients received preoperative cardiac clearance, including echocardiography. Delayed extubation was defined as intubation for >1 day (24 h). Preoperative and intraoperative variables were compared between patients intubated for ≤24 hours and those intubated for >24 hours. Because several diagnoses were represented by small numbers, individual diagnoses were not included as separate variables; instead, univariate analyses were stratified by cerebral palsy (CP) and non-CP cohorts. Multivariate linear regression analyses were then performed to determine whether preoperative and intraoperative variables were associated with the number of days intubated (NDI). A receiver operating characteristic (ROC) curve was constructed to assess the ability of left ventricular ejection fraction (LVEF) to predict delayed extubation and identify the optimal cutoff. RESULTS: Forty-three patients met the inclusion criteria. In subgroup univariate analyses, greater blood loss and blood loss per level fused were associated with delayed extubation in both CP and non-CP cohorts. LVEF was negatively associated with the number of days intubated in the CP cohort on univariate analysis (r=-0.442, P=0.035), and the non-CP cohort showed a similar trend (r=-0.529, P=0.063). In the overall cohort, lower LVEF remained independently associated with prolonged intubation on multivariable analysis (β=-0.11, 95% CI: -0.18 to -0.04; P=0.003). In adjusted subgroup models, LVEF was not independently significant in either cohort. ROC analysis demonstrated an area under the curve of 0.78 for CP and 0.57 for non-CP. CONCLUSIONS: Lower LVEF appears to be associated with delayed extubation in the overall cohort, and subgroup univariate analyses suggest a similar association in the CP cohort. Preoperative echocardiography may help identify patients at risk for delayed extubation and allow for proper preoperative optimization, preparation, and counseling. LEVEL OF EVIDENCE: level III.