RESUMEN
Current literature does not definitely demonstrate the superiority of any particular scaphoid reconstruction method. The primary goal of this retrospective single center study was to evaluate the influence of reconstruction techniques and other factors on the union rate after reconstruction of scaphoid nonunions. In the study, 370 patients with nonunions classified as stable (Mayo 1) or unstable (Mayo 2) depending on carpal alignment were included. Minimal radiological follow-up was 3 months. Bone healing after scaphoid reconstruction was evaluated using computer tomography scans in 294 and radiograms in 76 patients. Univariate and multivariate logistic regressions were applied in the analysis. Eight statistically significant factors that influenced the union rate were fixation method, type of previous operations, number of all operations, number of reconstructions, smoking, postoperative pathological lateral intrascaphoid angle (LISA>45°), avascular necrosis (AVN), and stability of nonunion. Smoking and postoperative LISA>45° significantly reduced the probability of bone fusion by approximately 4.4 and 9.5 times, respectively. Patients with reduced vascularity in the nonunion site had a 5.2 times lower chance of bone fusion. Our multivariate logistic regression model can explain 32% of failures after scaphoid reconstruction, including postoperative LISA>45°, patients' present record of smoking, and reduced vascularity in the nonunion site. However, bone graft type does not impact the union rate in general; however, in case of AVN, the tendency toward higher union rates was observed for medial femoral condyle free flaps when compared to other types of graft (non-vascularized and pedicled bone grafts considered as one group, p = 0.09).
