Modified Clavien-Dindo-sink classification system for adolescent idiopathic scoliosis.

PURPOSE: The Clavien-Dindosink (CDS) classification system provides more treatment-focused granularity than subjective methods of describing surgical complications; however, it has not been validated in posterior spinal fusion (PSF) for adolescent idiopathic scoliosis (AIS). The purpose of this study was to modify the CDS system for application in patients with AIS undergoing PSF to assess its inter- and intra-rater reliability for describing complications faced by this population.

METHODS: A review of all complications specific to patients with AIS captured in a large multicenter international database was performed. All complications were classified according to CDS, modified by addition of "prolonged initial hospital stay" as a criterion for Grade II. A survey of this complication list and an additional 20 clinical vignettes (sent out on two occasions) was sent to nine spinal deformity surgeons. Weighted kappa values were used to determine inter- and intra-rater reliability.

RESULTS: The Fleiss κ value for interrater reliability among 5 respondents grading all AIS complications was 0.8 (very good). For each grade, interrater reliability was very good, with an overall range of 0.8-1. The overall kappa value for intrarater reliability among eight respondents grading 20 vignettes was between 0.6 (good) and 0.9 (very good).

CONCLUSION: The modified CDS classification system has very good interrater and intrarater reliability in describing complications following PSF in patients with AIS. This system may be of greater utility for reporting outcomes than a "major" versus "minor" complication system and can serve as a valuable tool for improving surgical practices and patient outcomes in this population.

LEVEL OF EVIDENCE: IV case series.