Surgical costs in adult cervical Deformity: Do higher cost surgeries lead to better Outcomes?

Abstract

Background: As our focus on delivering cost effective healthcare increases, interventions like cervical deformity surgery, which are associated with high resource utilization, have received greater scrutiny. The purpose of this study was to assess relationship between surgical costs, deformity correction, and patient reported outcomes in ACD surgery.

Methods: ACD Patients ≥ 18 years with baseline (BL) and 2-year (2Y) data were included. Cost of surgery was calculated by applying average Medicare reimbursement rates by CPT code to surgical details of each patient in the cohort. CPT codes for corpectomy, ACDF, osteotomy, decompression, levels fused, and instrumentation were considered in the analysis. Costs of complications and reoperations were intentionally excluded from the cost analysis. Patients were ranked into two groups by surgical cost: lowest cost (LC) and highest cost (HC). ANCOVA assessed differences in outcomes while accounting for covariates as appropriate.

Results: 113 met inclusion criteria. While mean age, frailty, BMI and gender composition were similar between cost groups, mean CCI was significantly higher in the HC group compared to that of the LC group (p=.014). At baseline, LC and HC groups had similar HRQLs and radiographic deformity (all p >.05). Logistic regression accounting for baseline age, deformity and CCI found that HC patients had significantly lower odds of undergoing reoperation within 2-years (OR: 0.309, 95 % CI: 0.193 - 0.493, p <.001). Furthermore, logistic regression accounting for baseline age, deformity and CCI found odds of DJF were significantly lower for those in the HC group (OR: 0.163, 95 % CI: 0.083 - 0.323, p <.001). At 2-years, logistic regression accounting for age and baseline TS-CL found HC patients still had significantly higher odds of reaching a "0″ TS-CL modifier at 2-years (OR: 3.353, 95 % CI: 1.081 - 10.402, p=.036). Logistic regression accounting for age and baseline NDI score found HC patients had significantly higher odds of reaching MCID in NDI at 2-years (OR: 4.477, 95 % CI: 1.507 - 13.297, p=.007). A similar logistic regression accounting for age and baseline mJOA score found odds of reaching MCID in mJOA significantly higher for high-cost patients (OR: 2.942, 95 % CI: 1.101 - 7.864, p=.031).

Conclusions: While patient presentation influences surgical planning and costs, this study attempted to control for such variations to assess impact of surgical costs on outcomes. Despite continued scrutiny over healthcare costs, we found that more costly surgical interventions can produce superior radiographic alignment as well as patient reported outcomes for patients with cervical deformity.