Short-term complications of open versus arthroscopic osteochondral allograft transplantation in the knee: a national...

Osteochondral allograft (OCA) transplantation can be performed through an open or arthroscopic approach, yet comparative data on early safety are scarce. This study compares 30-day complications after open versus arthroscopic knee OCA using a national database and identifies independent risk factors for adverse events.

The American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database was queried for all patients undergoing elective knee OCA procedures from 2010 to 2023. Adults with complete perioperative data were included. One-to-one nearest-neighbor propensity score matching was used to control for baseline differences between cohorts. Of 2087 cases meeting inclusion criteria, 2078 (1165 open, 913 arthroscopic) remained after 9 patients with concurrent billing for both codes were excluded, and 1826 of these patients were included in the final propensity-matched analysis (913 per group). Thirty-day complications, defined by any adverse event (AAE) and resource-utilization metrics, were compared with Wilcoxon rank-sum or chi-square testing. Separate multivariable logistic regressions identified predictors of AAE within each cohort.

Two thousand seventy-eight patients were initially identified, with 1165 (56%) undergoing open OCA and 913 (44%) undergoing arthroscopic OCA. After matching, 913 patients remained in each group. Open surgery was associated with a longer mean operative time (124.9 min vs.101.9 min, p < 0.001; median [IQR] 108 [77-152] vs.90 [61-126] minutes) and a greater mean length of stay (0.60 days vs.0.28 days, p < 0.001; median [IQR] 0 [0-1] vs.0 [0-0] days), differences that probably reflected the fact that larger or more complex lesions were more often treated with the open approach, as was the case for the greater overall burden of concomitant procedures, rather than being attributable to the approach itself. Open OCA was also associated with higher odds of overnight hospital admission compared with arthroscopic OCA (OR 2.99, 95% CI 2.34-3.81, p < 0.001). Thirty-day AAE rates were similarly low between cohorts (2.3% for open vs 2.2% for arthroscopic). No deaths, myocardial infarctions, strokes, or pneumonia occurred. Rates of individual complications (surgical-site infection, thromboembolism, transfusion, unplanned reoperation) were <1% with no between-group differences.

Knee OCA transplantation is exceptionally safe in the early postoperative period, with complication rates below 3% and no statistically significant difference in complications between open and arthroscopic techniques. Open procedures are associated with greater resource utilization, specifically longer operative time and length of stay, most likely reflecting their use for larger or anatomically complex lesions rather than an inherent difference in safety between approaches. In the open cohort, smoking is the principal modifiable risk factor for short-term morbidity and should be a focus of perioperative optimization. These findings support selecting surgical approaches based on lesion size, location, and technical considerations rather than concerns about early complications.