Unilateral posterior vertebral column resection versus bilateral approach for post-traumatic thoracic angular kyphosis: a propensity-matched non-inferiority study.

purposePosterior vertebral column resection (PVCR) for severe post-traumatic thoracic angular kyphosis is effective but carries high morbidity. This study compares unilateral PVCR (UPVCR) with traditional bilateral PVCR (BPVCR), hypothesizing that UPVCR is non-inferior to BPVCR in deformity correction while offering improved perioperative outcomes in appropriately selected patients with an intact contralateral osseo-ligamentous tension band.

methodsA retrospective cohort study of patients who underwent PVCR for post-traumatic thoracic angular kyphosis (January 2019 - June 2023) at a tertiary academic center was conducted. Of 58 screened patients, propensity-score matching (1:1) on age, preoperative kyphotic angle, vertebral level, BMI, and preoperative neurological status yielded 40 patients (UPVCR: n = 19; BPVCR: n = 21) with a minimum 24-month follow-up. The primary outcome was correction of the local kyphotic angle. A pre-specified non-inferiority margin of 10° was defined for kyphotic angle correction. Secondary outcomes included operative time, estimated blood loss (EBL), complications, hospital stay, additional radiographic parameters, and patient-reported outcomes (VAS, ODI). All secondary outcome analyses are exploratory and unadjusted for multiplicity. UPVCR was applied exclusively for patients with an intact contralateral tension band and without severe rotational deformity (axial Cobb < 15°); a selection criterion that inherently limits direct comparability and is addressed in detail in the Limitations section.

resultsUPVCR met the pre-specified non-inferiority threshold for kyphotic angle correction (mean difference - 1.9°; 95% CI - 4.3° to + 2.1°; upper bound < 10° margin). UPVCR was associated with significantly shorter operative time (179.7 ± 12.5 vs. 269.3 ± 13.8 min, p < 0.001), lower EBL (740.5 ± 96.4 vs. 1378.6 ± 348.8 mL, p < 0.001), and a shorter hospital stay (5.6 ± 1.2 vs. 9.2 ± 1.6 days, p < 0.001). Complication rates were 10.5% for UPVCR and 19.0% for BPVCR; this numerical difference did not reach statistical significance (Fisher's exact test, p = 0.67), and the study was not powered to detect differences in complication rates. Both techniques achieved comparable correction in sagittal vertical axis (SVA) and pelvic incidence-lumbar lordosis (PI-LL) mismatch, with no differences in correction loss or cage subsidence. Clinical outcomes (VAS, ODI) improved significantly and similarly in both groups.

conclusionsIn appropriately selected patients, specifically those with focal angular post-traumatic kyphosis, an intact contralateral tension band, and no severe rotational component, UPVCR demonstrates radiographic and clinical outcomes that are non-inferior to BPVCR for kyphotic angle correction, while showing statistically significant reductions in operative time, blood loss, and hospital stay. Complication rates, while numerically lower, did not differ significantly between groups, and this study was underpowered to evaluate complications. Confirmation in larger, prospective, adequately powered studies is required before practice recommendations can be made.