Objective Scaphoid fractures are associated with high rates of late- or nonunion after conservative treatment. Nonunion is reported to occur in approximately 10% of all scaphoid fractures. It is known that the union of scaphoid fractures is affected by factors such as location at proximal pole, tobacco smoking, and the time from injury to treatment. Same factors seem to affect the healing after surgery for scaphoid nonunion. While the impact of preoperative humpback deformity on the functional outcome after surgery has been previously reported, the impact of humpback deformity, displacement, and the presence of bony cysts on union rate and time to healing after surgery has not been studied.

Purpose The primary purpose of this study is to assess the association of humpback deformity, fragment displacement, and the size of cysts along the fracture line with the union rate and union time, following surgery of scaphoid nonunion. The second purpose of the study is to investigate the interobserver reliability in the evaluation of computed tomography (CT) scans of scaphoid nonunion.

Patients and Methods From January 2008 to December 2018, 178 patients were surgically treated in our institution. After exclusion criteria were met, 63 patients with scaphoid delayed- or established nonunion, and preoperative CT scans of high quality (<2mm./ slice), were retrospectively analyzed. There was 58 men and 5 women with a mean age of 30 years (range: 16–72 years). Four orthopaedic surgeons and one radiologist independently analyzed the CT scans. The dorsal cortical angle (DCA), lateral intrascaphoid angle (LISA), the height-to-length ratio, the size of the cysts, and displacement of the fragments were measured. Healing was defined by CT scan, or by conventional X-ray, and status of no pain at clinical examination. Thirty-two of the patients had developed nonunion (>6 months postinjury), while 31 were in a stage of delayed union (3–6 months postinjury).

Results Open surgery with cancellous or structural bone graft was the treatment of choice in 49 patients, 8 patients were treated with arthroscopic bone grafting, and 6 patients with delayed union were operated with percutaneous screw fixation, without bone graft. Overall union rate was 86% (54/63) and was achieved after 84 days (12 weeks) (mean). The failure rate and time to healing were not associated with the degree of the humpback deformity, size of the cysts, or displacement of the nonunion in general. However, greater dislocation, and the localization of the nonunion at the scaphoid waist, showed significant influence on the union rate. Dislocation at nonunion site, in the group of the patients who united after surgery, was 2.7 mm (95% confidence interval [CI]: 1.5–3.7), and in the group who did not unite was 4.2 mm (95% CI: 2.9–5.7); p = 0.048). Time from injury to surgery was significantly correlated with time to union (p < 0.05), but not associated with the union rate (p < 0.4). Patients treated arthroscopically achieved faster healing (42 days), (standard deviation [SD]: 22.27) as compared with patients treated by open techniques (92 days; SD: 70.86). Agreement among five observers calculated as intraclass correlation coefficient was for LISA: 0.92; for height-to-length ratio: 0.73; for DCA: 0.65; for size of cysts: 0.61; and for displacement in millimeters: 0.24, respectively.

Conclusions The degree of humpback deformity and the size of cysts along the fracture line of scaphoid nonunion have no predictive value for the result, neither for the union rate nor the union time after surgery for the scaphoid nonunion. However, larger dislocation of the fragments measured at the scaphoid waist showed lower union rate. Time to healing following surgery is mainly influenced by the time from injury to the surgical treatment and may be influenced by the choice of the surgical technique. Interrater reliability calculation was best with LISA measurements, and worse with the measurements of the dislocation.