Operative versus nonoperative treatment of humeral shaft fractures: a systematic review and meta-analysis.

BACKGROUND

The humeral shaft fracture accounts for 1%-3% of all fractures and occurs in both the young and old population. However, the optimal treatment is still a matter of debate. Even though nonoperative treatment is commonly considered the gold standard, advantages have been described using operative stabilization. This systematic review aims to compare operative and nonoperative treatment in adult patients with humeral shaft fractures.

METHOD: We used the following databases: PubMed, Embase, Cochrane, and CINAHL on October 1, 2018, searching for randomized controlled trials (RCTs) and cohort studies. Two reviewers screened the studies using Covidence, followed by systematic data extraction. The primary outcome was defined as posttreatment complications such as nonunion, radial nerve palsy, malunion, and infections. The secondary outcomes were functional scores and patient-reported outcome measures (PROMs). To assess study quality, the risk of bias in nonrandomized studies of interventions and the Cochrane risk of bias tool were used.

RESULTS

Twelve studies were included: 1 RCT, 1 prospective cohort, and 10 retrospective cohorts with a total of 1406 patients, of whom 835 were treated operatively and 571 nonoperatively. Mean age ranged from 35 to 64, and 54% of the patients were male. The cohort studies had, in general, moderate bias, whereas the RCT had a low bias. There were statistically significant fewer nonunions in the operative treated group with a risk ratio of 0.49 (0.35-0.67), yielding a number needed to treat = 12. There were more deep infections in the operative group with a risk ratio of 2.76 (1.01-7.53) but otherwise no statistical differences concerning malunion or nerve damage. Only 1 study included PROM data.

CONCLUSION

There were fewer nonunions in the operative group but more deep infections. Because of the lack of studies reporting PROMs, the potential positive effect of operative therapy in early aftercare could not be evaluated. Therefore, PROMs should be mandatory in future comparative studies.