Complications in humeral shaft fractures - non-union, iatrogenic radial nerve palsy, and postoperative infection: a systematic review and meta-analysis.

The aim of this systematic review and meta-analysis was to assess risk for iatrogenic radial nerve palsy (iRNP), non-union, and post-operative infection in humeral shaft fractures. A PubMed search including original articles comparing different treatments for humeral shaft fractures published since January 2000 was performed. Random effect models with relative risks (RR) and 95% CIs were calculated for treatment groups and outcomes. Of the 841 results, 43 studies were included in the meta-analysis (11 level II, 5 level III, 27 level IV). Twenty-seven compared intramedullary nailing (IM) with ORIF, nine conservative with operative treatment, four ORIF with minimally invasive plate osteosynthesis (MIPO), and three anterior/anterolateral with posterior approach. iRNP risk was higher for ORIF vs IM (18 studies; RR: 1.80; P = 0.047), ORIF vs MIPO (4 studies; RR: 5.60; P = 0.011), and posterior vs anterior/anterolateral approach (3 studies; RR: 2.68; P = 0.005). Non-union risk was lower for operative vs conservative therapy (six studies; RR: 0.37; P < 0.001), but not significantly different between ORIF and IM (21 studies; RR: 1.00; P = 0.997), or approaches (two studies; RR: 0.36; P = 0.369). Post-operative infection risk was higher for ORIF vs IM (14 studies; RR: 1.84; P = 0.004) but not different between approaches (2 studies; RR: 0.95; P = 0.960). Surgery appears to be the method of choice when aiming to secure bony union, albeit risk for iRNP has to be considered, particularly in case of ORIF vs IM or MIPO, and posterior approach. Due to the limited number of randomised studies, evidence on the best treatment option remains moderate, though.