Utilization of the Fragility Index to Assess Randomized Controlled Trials Comparing Cervical Total Disc Arthroplasty to Anterior Cervical Discectomy and Fusion

Introduction: Cervical total disc arthroplasty (CTDA) remains an alternative to anterior cervical discectomy and fusion (ACDF). Randomized controlled trials (RCTs) investigating CTDA are often of varying quality with conflicting conclusions. The fragility index (FI) metric can be used to assess the robustness of statistically significant, dichotomous outcome variables in RCTs. The FI represents the number of patients needed to change an outcome from statistically significant (p ≤ 0.05) to non- significant (p > 0.05). The purpose of this study was to investigate the robustness or fragility of RCTs involving CTDA compared to ACDF.

Methods: A multi-database systematic review was performed to evaluate RCTs comparing CTDA to ACDF. RCTs with two parallel study arms and 1:1 allocation of subjects to treatment or control groups investigating CTDA vs. ACDF with at least one statistically significant, dichotomous outcome were included. Data extracted included total sample size, loss to follow-up, first statistically significant dichotomous outcome variable, and reported P-value. The FI was calculated by individually shifting one patient from the event group to the non-event group with re-calculation of Fisher’s Exact test until the result was no longer statistically significant.

Results: 934 abstracts were identified, with 22 RCTs meeting inclusion criteria spanning from 2008-2021 with 6096 patients. The most commonly extracted variables were Composite Success (n=7) and Rates of Secondary Surgeries (n=7). The mean patient sample size was 277.1 (median 209, range 30-541). The number of patients lost to follow-up was 0 in 1 of the studies and ranged from 0-231 (mean 74.3, median 45). The FI of all studies ranged from 0 to 7, with 3 (13.6%) of the studies having an associated FI of zero. Mean FI was 2.27 (median 1; mode 1; n=7; 31.8%). Loss to follow up exceeded the FI in all but one of the 22 included studies. There was no significant correlation between FI and number of patients in each study (R = 0.33) or the number lost to follow up (R=0.37).

Conclusion : RCTs investigating CTDA vs ACDF are often quite fragile with loss to follow-up often exceeding the FI.