General Session: Cervical-1
Presented by: D. Ohnmeiss - View Audio/Video Presentation (Members Only)
D. Ohnmeiss(1), M. Hisey(2), H. Bae(3)
(1) Texas Back Institute Research Foundation, Plano, TX, United States
(2) Texas Back Institute, Denton, TX, United States
(3) Cedars-Sinai Spine Center, Los Angeles, CA, United States
Introduction: In FDA-regulated trials, adverse events (AEs) are documented and assessed as a primary metric evaluating safety of an investigational treatment. AEs are typically defined as any clinically adverse sign, symptom, syndrome, or illness that occurred or worsened during or after treatment, regardless of cause. The purpose of this study was to determine if AEs were related clinical outcomes in patients undergoing cervical spine surgery.
Methods: The study was based on a post hoc analysis of data collected as part of an FDA IDE, prospective, randomized clinical trial conducted across 24 sites. The study included 186 anterior cervical fusion and 389 total disc replacement patients treated at one or two contiguous cervical levels. Clinical outcome measures were the Neck Disability Index (NDI), visual analog scales (VAS) assessing pain, and SF-12 scores. AEs were evaluated and classified by a clinical events committee composed of two orthopedic surgeons and one neurosurgeon. Patients were categorized as demonstrating: 1) at least one definitely or possibly device-related AE, 2) unrelated AEs only, or 3) no AEs that were ongoing at the given time point. Instances of device-related AE's included neck pain, device complications, headache, neurological and cervical spine disorder events. Unrelated AEs included cardiovascular, gastrointestinal, trauma, respiratory, and non-cervical events. ANOVA and Tukey's test for multiple comparisons were used to determine statistically significant differences between AE groups.
Results: At 60 months, 55 patients were classified as having at least one device-related AE, 363 patients had unrelated AEs, and 77 patients had no AEs. A significant difference was observed between groups for each outcome at 24, 36, 48 and 60 months (p< 0.0001). At each time point, the mean NDI, VAS neck pain, and SF-12 mental component and physical component scores were significantly worse for both the device-related AE and the unrelated AE groups compared to patients with no AEs (Table 1; p< 0.0001). Patients with device-related AE's demonstrated the poorest clinical outcomes. Results were similar at the earlier annual follow-up periods.
Discussion: Patients with treatment-related AEs had poorer clinical outcomes compared to patients without AEs. Patients with AEs unrelated to study treatment also exhibited poorer clinical outcomes, including disease-specific outcomes, compared to patients without AEs. Results through 60 months suggest that AEs unrelated to the investigational treatment may negatively influence patient pain and function scores. The occurrence of AEs, even those not related to the study treatment, significantly impacted commonly used outcome measures and thus may explain some of the outcome variation seen in clinical trials.
Table 1. Comparison of scores in AE groups.