Objective: To compare early operative treatment with nonoperative treatment of fragility fractures of the pelvis regarding mortality and functional outcome.
Design: Retrospective.
Setting: Two trauma centers.
Patients and methods: Two hundred thirty consecutive patients 60 years of age or older with an isolated low-energy fracture of the pelvis and with a follow-up of at least 24 months. In center 1, treatment consisted of a nonoperative attempt and early operative fixation if mobilization was not possible. In center 2, all patients were treated nonoperatively.
Main outcome measurements: Primary outcome was mortality. Secondary outcomes were in-hospital complications. Patients who survived were contacted by phone, and a modified Majeed score was obtained to assess functional outcome at the final follow-up.
Results: At the final follow-up (mean 61 months, SD 24), 105/230 (45.7%) patients had died. One year after the initial hospitalization, 34/148 patients [23%, 95% confidence interval (CI): 17%-31%] of the early operative group and 14/82 patients (17%, 95% CI: 10%-27%) of the nonoperative group had died (P = 0.294). Nonoperative treatment had a protective effect on survival during the first 2 years (hazard ratio of the nonlinear effect: 2.86, 95% CI: 1.38-5.94, P < 0.001). Patients in the early operative treatment group who survived the first 2 years had a better long-term survival. The functional outcome at the end of follow-up as measured by a modified Majeed score was not different between the 2 groups (early operative: 66.1, SD 12.6 vs. nonoperative: 65.7, SD 12.5, P = 0.910).
Conclusion: Early operative fixation of patients who cannot be mobilized within 3-5 days was associated with a higher mortality rate and complication rate at 1 year but with a better long-term survival after more than 2 years. Hence, patients with a life expectancy of less than 2 years may not benefit from surgery with regard to survival.
Level of evidence: Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.