This retrospective analysis was performed on 94 patients with DFR accompanied by dorsoulnar fragments, spanning the period from October 2018 to November 2022. Mean follow-up was 15.5 (range, 12–20) months. Patients were divided into small- (<5 %, n = 28), middle- (5–15 %, n = 50), and large- (>15 %, n = 16) sized groups according to articular involvement of dorsoulnar fragments determined by three-dimensional (3D) computed tomography (CT) modeling. Subdivision also took place for the presence of postoperative fragment displacement (>2 mm) and fixation methods including volar locking plate (VLP), VLP combined with dorsal hollow compression screw (VDS), and VLP combined with dorsal low-profile mini plate (VDP). The radiographic parameters (volar tilt, radial inclination, and radial height) and functional outcome measures of wrist range of motion, wrist function (DASH, PRWE), and wrist pain (VAS) were evaluated and compared between groups.

Fracture healing was observed in all patients at final follow-up. No instances of dorsoulnar fragment displacement were observed in patients undergoing VDS and VDP treatment and the incidence of the dorsoulnar fragment displacement was 35 % (n = 8) in small-sized group, 21 % (n = 7) in middle-sized group, and 7 % (n = 1) in large-sized group when patients were treated with VLP. In small-sized group, no significant differences were found between patients with and without dorsoulnar fragment displacement in dorsiflexion restriction (10.6 ± 2.8°, 9.1 ± 2.3°, P = 0.159), pronosupination restriction (9.6 ± 2.1°, 8.6 ± 1.7°, P = 0.188), DASH (11.5 ± 4.1, 10.7 ± 3.2, P = 0.562), PRWE (11.9 ± 4.2, 10.6 ± 3.6, P = 0.425), and VAS (1.1 ± 1.1, 0.9 ± 1.0, P = 0.528). In middle-sized combined with large-sized group, the functional outcome measures of dorsiflexion restriction (12.5 ± 3.7°, 9.8 ± 2.9°, P = 0.022), DASH (14.6 ± 5.2, 11.4 ± 3.7, P = 0.030), and PRWE (15.0 ± 4.5, 11.3 ± 3.9, P = 0.016) were superior in patients without dorsoulnar fragment displacement. In patients treated with VLPs, no significant differences were found in dorsiflexion restriction (9.8 ± 2.5°, 10.8 ± 3.5°, 9.4 ± 2.5°, P = 0.299), pronosupination restriction (9.2 ± 1.9°, 10.1 ± 2.8°, 8.9 ± 1.5°, P = 0.200), DASH (11.1 ± 3.5, 12.9 ± 4.3, 11.1 ± 3.6, P = 0.162), PRWE (11.1 ± 3.9, 12.8 ± 4.2, 10.8 ± 3.9, P = 0.188), and VAS (1.0 ± 1.0, 1.4 ± 1.1, 0.9 ± 0.9, P = 0.151) between small-sized, middle-sized, and large-sized groups. In middle-sized group, no significant differences were found in dorsiflexion restriction (10.8 ± 3.5°, 9.4 ± 2.2°, 9.4 ± 2.4°, P = 0.316); pronosupination restriction (10.1 ± 2.8°, 8.8 ± 1.9°, 9.0 ± 2.5°, P = 0.314), DASH (12.9 ± 4.3, 10.3 ± 3.7, 10.5 ± 3.7, P = 0.133), PRWE (12.8 ± 4.2, 10.4 ± 3.8, 10.6 ± 4.1, P = 0.199), and VAS (1.4 ± 1.1, 0.8 ± 0.7, 1.0 ± 1.1, P = 0.201) between subgroups of VLP, VDS, and VDP. No significant differences were found in radiographic parameters between all groups compared.