Impact of Controlled Versus Uncontrolled mFI-5 Frailty on Perioperative Complications After Adult Spinal Deformity Surgery

Study Design: Retrospective review. Objectives: We substratified the mFI-5 frailty index to reflect controlled and uncontrolled conditions and assess their relationship to perioperative complications. Summary of Background Data: Risk assessment before adult spinal deformity (ASD) surgery is critical because the surgery is highly invasive with a high complication rate. Although frailty is associated with risk of surgical complications, current frailty measures do not differentiate between controlled and uncontrolled conditions. Methods: Frailty was calculated using the mFI-5 index for 170 ASD patients with fusion of ≥ 5 levels. Uncontrolled frailty was defined as blood pressure > 140/90 mm Hg, HbA1C > 7% or postprandial glucose > 180 mg/dL, or recent chronic obstructive pulmonary disease (COPD) exacerbation, while on medication. Patients were divided into nonfrailty, controlled frailty, and uncontrolled frailty cohorts. The primary outcome measure was perioperative major and wound complications. Bivariate analysis was performed. Multivariable analysis assessed the relationship between frailty and perioperative complications. Results: The cohorts included 97 nonfrail, 54 controlled frail, and 19 uncontrolled frail patients. Compared with nonfrail patients, patients with uncontrolled frailty were more likely to have age older than 60 years (84% vs. 24%), hyperlipidemia (42% vs. 20%), and Oswestry Disability Index (ODI) score > 42 (84% vs. 52%) (P < 0.05 for all). Controlled frailty was associated with those older than 60 years (41% vs. 24%) and hyperlipidemia (52% vs. 20%) (P < 0.05 for all). On multivariable regression analysis controlling for hyperlipidemia, functional independence, motor weakness, ODI > 42, and age older than 60 years, patients with uncontrolled frailty had greater odds of major complications (OR 4.24, P = 0.03) and wound complications (OR 9.47, P = 0.046) compared with nonfrail patients. Controlled frailty was not associated with increased risk of perioperative complications (P > 0.05 for all). Conclusions: Although patients with uncontrolled frailty had higher risk of perioperative complications compared with nonfrail patients, patients with controlled frailty did not, suggesting the importance of controlling modifiable risk factors before surgery.