Surgical approaches for inserting hemiarthroplasty of the hip in people with hip fractures

Hip fractures affect over 10 million people annually worldwide and are expected to increase with an ageing population, contributing significantly to morbidity, mortality, and healthcare costs. Hemiarthroplasty, a common treatment for displaced femoral neck fractures, accounts for more than half of hip fracture surgeries in older adults. However, the optimal surgical approach - anterior, lateral, or posterior - remains uncertain, with decisions often based on surgeon preference or institutional protocols.

To assess the effects of different surgical approaches for hemiarthroplasty in the treatment of hip fractures.

We searched CENTRAL, MEDLINE, Embase, and six other databases in November 2024. We also searched two trials registries, nine different conference proceedings, reference lists of included studies, and systematic reviews published within the last five years.

We included randomised controlled trials (RCTs) and quasi-RCTs in adults with hip fractures comparing different surgical approaches for hemiarthroplasty. We excluded studies of participants with high-energy hip fractures, fractures not associated with osteoporosis, or studies comparing hemiarthroplasty with total hip arthroplasty (THA).

We were interested in a primary core outcome set: activities of daily living (ADL), health-related quality-of-life (HRQoL), mobility or functional status, mortality, and pain. Our secondary outcomes were complications, operative details and postoperative care outcomes.

We used the Cochrane RoB 1 tool to assess risk of bias.

We performed meta-analyses using RevMan with a generic inverse-variance approach and random-effects models to calculate risk ratios (RRs), mean differences (MDs) or standardised mean differences (SMDs) with 95% confidence intervals (CIs). We used GRADE to determine the certainty of evidence.

We included 27 studies (23 RCTs, three quasi-RCTs, and one combined RCT/quasi-RCT) with a total of 3369 participants. The studies were conducted in Europe and Asia and published between 1981 and 2024. All studies but one focused on intracapsular fractures. The 'typical' included participant was a geriatric woman with an osteoporotic hip fracture treated with hip hemiarthroplasty who was ambulatory prior to injury and had a varying level of cognitive impairment at baseline.

All but three studies were at high risk of detection bias and had unclear/high risk of bias in at least one domain. We downgraded many outcomes for imprecision, and for risk of bias where sensitivity analysis indicated the estimate was influenced in size or direction by studies with limitations. Anterior versus posterior approaches (7 studies, 455 participants) There was no evidence of a difference in ADL (MD 0.08, 95% CI -0.55 to 0.71; 1 study, 89 participants), mortality (RR 1.0, 95% CI 0.41 to 2.44; 3 studies, 242 participants), or pain (SMD -0.12, 95% CI -0.42 to 0.18; 2 studies, 171 participants) at three-month follow-up, but evidence was of very low-certainty. We found low-certainty evidence of improved early ability to ambulate independently with anterior approach hemiarthroplasty (RR 1.64, 95% CI 1.15 to 2.34; 2 studies, 161 participants). However, no evidence of a difference in measured functional status was shown at three-month follow-up (SMD 0.06, 95% -0.25 to 0.37; 3 studies, 158 participants). No studies reported on early HRQoL. Anterior versus lateral approaches (6 studies, 641 participants) We found no evidence of a difference in ADL (MD 3.08, 95% CI -14.95 to 21.1; 2 studies, 140 participants), or pain (MD -0.29, 95% CI -0.92 to 0.33; 4 studies, 282 participants) at three-month follow-up, but the evidence is very uncertain. There was low-certainty evidence of improved functional status with anterior approach hemiarthroplasty (MD 1.17, 95% CI 0.03 to 2.30; 2 studies, 142 participants). However, this did not reach a clinically important difference. We found that Trendelenburg gait may be reduced slightly with anterior approach hemiarthroplasty at three-month follow-up (RR 0.13, 95% CI 0.04 to 0.40; 1 study, 94 participants). We are unsure of the effect on early HRQoL as no studies reported the outcome, or for early mortality as no events were reported, resulting in a non-estimable effect size. Lateral versus posterior approaches (11 studies, 1840 participants) There was no evidence of a difference in early ADL (MD 0.05, 95% CI -0.33 to 0.43; 1 study, 297 participants), HRQoL (MD -0.03, 95% CI -0.09 to 0.03; 2 studies, 529 participants), functional status (SMD 0.09, 95% CI -0.36 to 0.55; 5 studies, 494 participants), or pain (SMD -0.07, 95% CI -0.41 to 0.27; 6 studies, 752 participants), but evidence was very low-certainty. We found low-certainty evidence of little to no difference between lateral and posterior approaches in mortality (RR 0.88, 95% CI 0.56 to 1.39; 4 studies, 1417 participants).

For people undergoing hemiarthroplasty for intracapsular hip fracture, the evidence is very uncertain regarding the effect of surgical approach on activities of daily living and pain within four months. There is little to no evidence of a difference in health-related quality of life, functional status, or mortality between approaches. There is currently insufficient evidence to determine whether anterior, lateral, or posterior approaches are a more appropriate option for hemiarthroplasty for hip fracture with respect to these outcomes. Further research is needed to improve the certainty of evidence, requiring better-powered trials, adherence to reporting standards, prospective trial registration, involvement of experienced surgeons, and blinded outcome assessment to reduce bias. Ensuring the inclusion of the core outcome set for hip fractures and follow-up of at least four months in all RCTs remains essential.

This Cochrane review had no dedicated funding.

Registration: Prospero CRD42024498914 Previous version available at: https://doi.org/10.1002/14651858.CD001707.