Outcomes With Surgery vs Functional Bracing for Patients With Closed, Displaced Humeral Shaft Fractures and the Need for Secondary Surgery: A Prespecified Secondary Analysis of the FISH Randomized Clinical Trial

Abstract

Importance: Existing evidence indicates that surgery fails to provide superior functional outcome over nonoperative care in patients with a closed humeral shaft fracture. However, up to one-third of patients treated nonoperatively may require secondary surgery.

Objective: To compare the 2-year outcomes of patients who required secondary surgery with the outcomes of patients with successful initial treatment.

Design, setting, and participants: This 2-year follow-up of the Finnish Shaft of the Humerus (FISH) randomized clinical trial comparing surgery with nonoperative treatment (functional brace) was completed in January 2020. Enrollment in the original trial was between November 2012 and January 2018 at 2 university hospital trauma centers in Finland. A total of 321 adult patients with closed, displaced humeral shaft fracture were assessed for eligibility. After excluding patients with cognitive disabilities, multimorbidity, or multiple trauma and those refusing randomization, 82 patients were randomized.

Interventions: Interventions were surgery with plate fixation (n = 38; initial surgery group) or functional bracing (n = 44); the latter group was divided into the successful fracture healing group (n = 30; bracing group) and the secondary surgery group (n = 14) with fracture healing problems.

Main outcomes and measures: The primary outcome was Disabilities of Arm, Shoulder and Hand (DASH) score at 2 years (range, 0 to 100 points; 0 denotes no disability, 100 extreme disability; minimal clinically important difference, 10 points).

Results: Of 82 randomized patients, 38 (46%) were female. The mean (SD) age was 48.9 (17.1) years. A total of 74 patients (90%) completed the 2-year follow-up. At 2 years, the mean DASH score was 6.8 (95% CI, 2.3 to 11.4) in the initial surgery group, 6.0 (95% CI, 1.0 to 11.0) in the bracing group, and 17.5 (95% CI, 10.5 to 24.5) in the secondary surgery group. The between-group difference was -10.7 points (95% CI, -19.1 to -2.3; P = .01) between the initial and secondary surgery groups and -11.5 points (95% CI, -20.1 to -2.9; P = .009) between the bracing group and secondary surgery group.

Conclusions and relevance: Patients contemplating treatment for closed humeral shaft fracture should be informed that two-thirds of patients treated with functional bracing may heal successfully while one-third may experience fracture healing problems that require secondary surgery and lead to inferior functional outcomes 2 years after the injury.

Trial registration: ClinicalTrials.gov Identifier: NCT01719887.

Figure 1.. Enrollment and Randomization of Patients in the Finnish Shaft of the Humerus (FISH) Trial at 2-Year Follow-up

In this secondary analysis of the FISH trial, patients were divided into 3 groups: an initial surgery group (patients who were randomized to surgery), a bracing group (patients whose fracture healed with functional bracing), and a secondary surgery group (patients who were randomized to bracing but who later had surgery because of a healing problem). The aim of this secondary analysis was to compare outcomes of patients who underwent secondary surgery with outcomes of patients who healed with the initially allocated treatment method. Patients not willing to participate in randomization but who consented to follow-up were able to choose their preferred treatment. These 42 patients were included in the declined cohort and were analyzed separately according to the same principles as the randomized cohort. We also performed a post hoc analysis of Disabilities of the Arm, Shoulder and Hand scores at 2 years combining all recruited patients from the randomized cohort and declined cohort.

Figure 2.. Disabilities of the Arm, Shoulder and Hand (DASH) Score in Initial Surgery, Bracing, and Secondary Surgery Groups Over Time

Error bars indicate 95% CIs of the point estimates of group means. Boxes indicate 25th and 75th percentiles of observed values; horizontal lines within boxes indicate median DASH scores; error bars indicate the highest and lowest values within 1.5 times the interquartile range; and points beyond the error bars indicate individual values outside of this range. The 14 patients in the bracing group who underwent secondary surgery during the 2-year follow-up period form the secondary surgery group. Before fracture indicates the time point of randomization after sustaining the fracture when patients were asked about their perceived preinjury DASH scores. Number of patients with available DASH scores at each time point is indicated per treatment group. One patient did not report the DASH score at baseline and we were unable to retrieve this information later (eTable 6 in Supplement 2). No imputation was performed for missing data. See eTable 7 in Supplement 2 for data. See eFigure 3 in Supplement 2 for parallel line plot.