Fragility index for extended prophylaxis following abdominopelvic surgery: A methodological survey

Abstract

Background

Fragility Index (FI) is increasingly used to assess robustness of statistically significant p-values reported in randomized controlled trials (RCTs). FI represents the lowest number of non-events changed to events that would make study findings non-significant. This methodological survey was designed to assess the fragility of the evidence for extended VTEp following major abdominopelvic surgery.

Methods

MEDLINE, Embase, and CENTRAL were searched from inception to November 2023. RCTs with parallel, double-armed, superiority design comparing extended VTEp for patients undergoing major abdominopelvic surgery to controls with at least one statistically significant dichotomous outcome were included. Walsh et al.’s method of calculating FI was utilized.

Results

After review of 611 citations, 6 RCTs were identified with 12 statistically significant outcomes between groups. The mean number of patients randomized per RCT was 419 (SD 176). The median FI was 1.5 (range: 1–4). The number of patients lost to follow-up was greater than the FI for 10/12 (83.3 ​%) outcomes.

Conclusions

Statistically significant differences reported in RCTs evaluating extended VTEp following major abdominopelvic surgery are not robust.