Perioperative peripheral nerve blocks for elective below-knee amputations and phantom limb syndrome: an exploratory multicenter retrospective cohort study.

Nguyen, Eric Hoang; Hoffman, Kevin; Gotewal, Sunny; Wetzig, Ashley; Wetzig, Alexander; Kwater, Andrzej; Owolabi, Adebukola; Mensah, Cassius et al. · Reg Anesth Pain Med · 2026

retrospective_cohort · Level III

Where this comes from

Abstract

Phantom limb syndrome (PLS) is a common complication following below-knee amputations (BKA) causing painful and/or nonpainful sensations in the absent limb. Peripheral nerve blocks (PNBs) are used for perioperative analgesia, but their association with PLS risk remains unclear. This study sought to evaluate potential associations regarding PLS incidence after BKA with and without PNBs. This retrospective cohort study was conducted using Epic Cosmos data from 2016 to 2025. Adults undergoing elective BKA were identified using Current Procedural Terminology codes 27880-27889 and categorized by PNBs versus none. The primary outcome was PLS diagnosis within 12 months of surgery using International Classification of Diseases, 10th Revision, Clinical Modification codes (G54.6, G54.7). Relative risks (RRs) were estimated using Mantel-Haenszel stratification adjusted for age, sex, and race only, with subgroup analysis by comorbidity burden. Secondary outcomes included length of stay (LOS). Among 38 433 patients, 7430 (19.3%) received perioperative PNBs and 31 003 (80.7%) did not. PLS diagnosis coding occurred in 18.8% of the PNB group versus 15.1% of the non-block group (absolute difference 3.7%). Analysis adjusted for age, sex, and race only did not change the association (Mantel-Haenszel RR 1.24, 95% CI 1.20 to 1.28). PNBs were associated with a shorter LOS. All estimates should be considered exploratory associations. In this multicenter electronic health record-based cohort, exploratory analysis showed that perioperative PNB use was associated with increased rates of documented PLS coding across subgroups. However, the overall RR increase is modest. The observed association may reflect residual confounding by indication and differential surveillance rather than a causal effect. PNBs still remain appropriate for improving acute pain and reducing LOS. Prospective studies are needed to clarify causality and inform prevention strategies.