Healing and complications of periprosthetic humeral fractures after shoulder arthroplasty - 11/11/26

Doi : 10.1016/j.otsr.2026.104990

Joseph Jacquet ^a,⁎ , Laurent Obert ^b, Marc Saab ^c,^{^{1
Associate Editor, Marc Saab had no involvement in the peer-review of this article and has no access to information regarding its peer-review. Full responsibility for the editorial process for this article was delegated to another journal editor.}}, Edouard Bahu ^c, Antoine Poichotte ^d, Clément Tournier ^e, François Sirveaux ^a, Lisa Peduzzi ^a

^a Service de Chirurgie Orthopédique et Traumatologique, Centre Emile Gallé, CHRU de Nancy, 49 Rue Hermite, 54000 Nancy, France

^b Chirurgie Orthopédique et Traumatologique, Chirurgie Plastique, Esthétique et Reconstructrice, Chirurgie de la Main, Centre Hospitalier Universitaire Hôpital Jean Minjoz, CHRU de Besançon, 3 Boulevard Alexandre Fleming, 25030 Besançon, France

^c Service de Chirurgie Orthopédique et Traumatologique, Hôpital Salengro, CHRU de Lille, 2 Avenue Oscar Lambret, 59000 Lille, France

^d Chirurgie Orthopédique et Traumatologique, Centre Hospitalier Loire Vendée Océan, 20 Boulevard Guérin, 85300 Challans, France

^e Chirurgie Orthopédique et Traumatologique, Hôpital Privé Saint Martin, Allée des Tulipes, 33600 Pessac, France

^⁎ Corresponding author at: Service de Chirurgie Orthopédique et Traumatologique, Centre Emile Gallé, CHRU de Nancy, 49 Rue Hermite, 54000 Nancy, France. Service de Chirurgie Orthopédique et Traumatologique Centre Emile Gallé CHRU de Nancy 49 Rue Hermite Nancy 54000 France

Sous presse. Épreuves corrigées par l'auteur. Disponible en ligne depuis le Wednesday 11 November 2026

Abstract

Background

Periprosthetic humeral fractures are an increasing complication of shoulder arthroplasty, and data on their healing remain limited. Three questions were addressed: (1) which fracture, patient and implant characteristics are associated with the time to healing? (2) Are the same characteristics associated with the probability of healing? (3) Does the treatment strategy influence healing and the complication rate?

We hypothesized that (1) fractures below the prosthetic stem (Campbell type 4) would heal more slowly, (2) they would also be less likely to heal, and (3) revision surgery would carry a higher complication rate.

Patients and methods

Retrospective multicentre study of 152 periprosthetic humeral fractures in 147 patients across 14 centres between October 2005 and June 2024, with a minimum six-month follow-up. Healing was defined as bridging on three of four cortices with absence of pain and macromobility. Treatments were nonoperative management, osteosynthesis or implant revision. Time to healing was compared among healed fractures; the probability of healing was analysed in all postoperative fractures by the Kaplan–Meier method and multivariate Cox regression.

Results

Among the 138 postoperative fractures, 80 (58.0%) healed after a median of 3.0 months. Time to healing differed by Campbell type (2.0, 3.0 and 5.0 months for types 1–2, 3 and 4; p = 0.0007) but not by stem length (p = 0.093). In multivariate analysis, healing occurred at a higher rate around short stems (HR = 1.71 [1.09–2.70]; p = 0.020) and in Campbell types 1–2 versus type 4 (HR = 2.19 [1.06–4.52]; p = 0.035). Treatment strategy was associated with neither the speed nor the rate of healing (p = 0.642), but complication rates were 17.8%, 22.8% and 44.0% after nonoperative management, osteosynthesis and revision (p = 0.009), independently of healing (p = 1.000).

Discussion

Campbell type 4 fractures and standard stems were associated with slower healing; whether they also heal less often could not be determined. The higher complication burden after revision may reflect confounding by indication.