268. Frailty is a better predictor than age of mortality and perioperative complications after surgery for Degenerative Cervical Myelopathy: An analysis of 41,369 patients from the NSQIP database 2010-2018

Abstract

<h3>BACKGROUND CONTEXT</h3> The effect of frailty on the perioperative outcomes for patients undergoing surgery for degenerative cervical myelopathy (DCM) patients is poorly understood. <h3>PURPOSE</h3> The objective of the current study was to model the effect of various models of frailty on the perioperative outcomes for surgical DCM patients. <h3>STUDY DESIGN/SETTING</h3> Retrospective cohort study. <h3>PATIENT SAMPLE</h3> Patients under going surgery for DCM registered in the National Safety and Quality Improvment Database (NSQIP). <h3>OUTCOME MEASURES</h3> Mortality, adverse events, readmission, reoperation, length of hospital stay and discharge to nonhome destination. <h3>METHODS</h3> The modified frailty index 11- and 5-point (MFI-11, MFI-5), modified Charlson comorbidity index (mCCI) and ASA grade were calculated for patients with a diagnosis of DCM from the NSQIP database 2010-2018. The MFI-11 and MFI-5 scores were stratified into "Not Frail," "Prefrail," "Frail" and "Severely Frail." Univariate analysis was performed to assess the risk of mortality and adverse events by index. Multivariable modelling of age and frailty with a base model (type of surgical approach, number of operated levels and gender) was performed, with receiver operating characteristic area under the curve analysis (AUC) to define discriminative ability of each measure. <h3>RESULTS</h3> A total of 41,369 patients were included. Increasing frailty correlated significantly with increased risk of mortality, adverse events, longer hospital stay and risk of nonhome discharge destination. Comparing age and the other indices, the MFI-5 had the largest effect size for all variables (odds ratio 8.37 [Frail] - 27.70 [Severely Frail]). MFI-5 and mCCI have the best discriminative ability at predicting adverse events with multivariable modelling, but the base model with MFI-5 and age had the best predictive value (AUC=0.84). The effect size of frailty was greater than age alone for all outcomes, and MFI-5 had a categorical assessment correlation with MFI-11 of 0.988 (p<0.001). <h3>CONCLUSIONS</h3> Frailty has a larger effect size and better discriminative value compared to age alone to predict adverse events after surgery for DCM. A multivariable regression model using MFI-5 produced the most robust predictive tool, which substantiates the use of the MFI-5 in preference to the MFI-11 in clinical practice. <h3>FDA DEVICE/DRUG STATUS</h3> This abstract does not discuss or include any applicable devices or drugs.