Intramedullary nailing versus sliding hip screw for AO/OTA 31-A2 and 31-A3 trochanteric fractures: a systematic review and meta-analysis of randomized controlled trials

BACKGROUND: Trochanteric femoral fractures (TFFs) are frequent injuries in older adults, with unstable patterns (AO/OTA 31-A2 and A3) carrying a higher risk of complications. Intramedullary nailing (IMN) and sliding hip screw (SHS) are the two main surgical options. However, guideline recommendations differ, and prior meta-analyses are limited. This study aimed to systematically compare the efficacy and safety of IMN versus SHS in treating adult patients with unstable TFFs based on randomized controlled trials (RCTs). METHODS: MEDLINE, Embase, and CENTRAL (January 2008-March 2025) were searched for eligible RCTs, which included adults with 31-A2 or A3 fractures randomized to IMN or SHS. Primary outcomes were mortality and reoperation. Secondary outcomes included implant failures, nonunion, surgical parameters, and postoperative mobility, pain, and function. Risk of bias (RoB) was assessed using RoB 2, and the certainty of evidence (CoE) with GRADE. RESULTS: Eighteen RCTs (n = 3237 patients) were included. No significant differences were found between IMN and SHS in three-month (low CoE) and 12-month mortality (moderate CoE). Reoperation rates trended higher with SHS, but not significantly (pooled OR = 1.70; 95% CI, 0.97-2.97; low CoE). SHS was associated with higher rates of arthroplasty conversion (pooled OR 1.92; 95% CI 1.00-3.68; low CoE), nonunion (pooled OR 1.93; 95% CI 1.12-3.34; low CoE), and infection (pooled OR 2.20; 95% CI 1.29-3.74; low CoE), while implant failure did not differ significantly (pooled OR 1.35; 95% CI 0.91 to 2.01; low CoE). IMN was associated with higher functional scores, less pain (within three months), and a greater likelihood of regaining pre-fracture mobility (CoE low- moderate). CONCLUSIONS: IMN demonstrated comparable mortality to SHS. Although overall reoperation rates did not differ significantly, IMN was associated with lower odds of arthroplasty conversion, nonunion, and infection, as well as reduced early postoperative pain and improved early postoperative function. However, given the overall low to moderate CoE, these findings should be interpreted with caution. Implant selection should remain individualized, taking into account patient characteristics, fracture morphology, and surgeon experience.