Frailty as a Predictor of Adverse Events in Elderly Patients Undergoing Instrumented Cervical and Lumbar Fusion

Introduction: Degenerative spine disease is a leading cause of disability, with increasing prevalence in the elderly. While age has been identified as an independent predictor of outcomes, its predictive value is limited for similarly elderly patients. Here, we aimed to determine the most predictive frailty score of adverse events in patients aged 80 and older undergoing instrumented cervical or lumbar fusion.

Methods: We proceeded with a multi-site (3 tertiary academic centers) retrospective review including patients undergoing instrumented fusion aged 80 and older from January 2010 to present. A composite endpoint encompassing 30-day return to operating room, re-admission and mortality was created. We estimated the area under the ROC curve (AUC-ROC) for frailty scores (mfi-5, mfi-11 and CCI) in relation to that composite score. In addition, we estimated the association between each score and the composite endpoint by means of logistic regression.

Results: A total of 161 patients with an average age of 85 years at the time of surgery were included. We observed a 30-day re-admission rate of 11.4%, re-operation of 3.6% and mortality of 0.6%. The overall rate of the composite endpoint at 30 days was 25 (15.1%). The AUC for mfi-5 was 0.597 (0.501 – 0.693), for mfi-11 0.620 (0.518 – 0.723), and for CCI 0.564 (0.453 – 0.675). The association between the scores and composite endpoint did not reach statistical significance for mfi-5 (OR= 1.45 (0.98 – 2.15), p=0.061), and CCI (OR= 1.13 (0.97 – 1.31), p=0.113) but was statistically significant for mfi-11 (OR= 1.46 (1.07 – 2.00), p=0.018).

Conclusion : This is the largest study comparing frailty index scores in patients undergoing instrumented fusion at 80 years or more. Our findings suggest that, while mfi-11 score correlated with adverse events, the predictive ability of existing scores remains limited, highlighting the need for better approaches to identify select patients at age extremes.