The impact of durotomy on short-term morbidity and long-term outcomes in adult spinal deformity surgery: a multicenter propensity score-matched observational study.

Abstract

Objective: Incidental durotomy is a common complication in adult spinal deformity (ASD) correction, but its long-term impact remains unclear. This study aimed to evaluate the long-term outcomes associated with incidental durotomy in ASD surgery, focusing on hospital length of stay (LOS), thromboembolic events, and patient-reported outcome measures (PROMs).

Methods: The authors conducted a retrospective cohort study using data from their institutional spine deformity registry, including ASD surgeries from January 1, 2013, to December 13, 2021. Adult patients who underwent elective thoracolumbosacral posterior spinal fusion involving ≥ 6 vertebrae were included. Primary outcomes were hospital LOS, postoperative thromboembolic events, 30-day readmission rate, reoperation rate, and PROMs. Secondary outcomes involved preoperative risk factors, surgical details, and perioperative complications and outcomes. Unadjusted and adjusted analyses were performed. Propensity scores were calculated using logistic regression, adjusting for relevant covariates, and propensity score matching (PSM) with an attempted 1:5 ratio was applied. The p values were adjusted using the Benjamini-Hochberg correction.

Results: A total of 646 patients were included, with 53 (8.2%) experiencing durotomy and 593 controls. After PSM, 51 patients with durotomy and 242 controls were well balanced across baseline characteristics. The durotomy group had significantly longer hospital LOS (9.1 ± 5.0 days vs 7.0 ± 3.4 days; median difference 2 days, 95% CI 0.6-3.5 days; p = 0.044) and increased thromboembolic events (10 [19.6%] vs 13 [5.4%]; proportion difference 14.2%, 95% CI 3.0%-25.5%; p = 0.022) compared with the control group. No significant differences were found in 30-day readmissions, reoperation rates, and PROMs. Surgical details revealed increased operative time (median difference 64.2 minutes, 95% CI 25.7-102.8 minutes, p = 0.012), estimated blood loss (median difference 300 mL, 95% CI 60-800 mL, p = 0.035), and intraoperative total red blood cell transfusion volume (median difference 250 mL, 95% CI 25-775 mL, p = 0.035) in the durotomy group.

Conclusions: Incidental durotomy was associated with increased hospital LOS and higher thromboembolic event rates in ASD surgery, but did not significantly affect long-term patient-reported outcomes or reoperation rates. These findings suggest that while durotomy may increase short-term morbidity, it does not worsen long-term outcomes. Future studies should explore strategies for minimizing in-hospital complications in this patient population.