Review Article

Historically, did Cemented Thompson perform better than uncemented Austin Moore hemiarthroplasty for femoral neck fractures? A meta-analysis of available evidence

¹ Faculty of Medicine, Zagazig University, 44519 Zagazig, Egypt
² Medical Research Group of Egypt, 44523, Cairo, Egypt
³ Faculty of Medicine, Ain Shams University, 11566 Cairo, Egypt
⁴ Faculty of Medicine, Assuit University, 71515 Assuit, Egypt
⁵ Faculty of Medicine, Menoufia University, 32511 Menoufia, Egypt
⁶ Trauma and Orthopaedics, Ashford and St Peters Hospitals NHS Foundation Trust, KT16 0PZ Surrey, UK
⁷ Department of Orthopaedic surgery, Der Balgrist, University of Zurich, 8008 Zurich, Switzerland
⁸ Department of Orthopaedic Surgery and Trauma, Suez Canal University Hospitals, 41522 Ismailia, Egypt
⁹ Trauma and Orthopaedics, Oxford University Hospitals, OX3 9DU Oxford, UK

^* Corresponding author: mohamed.shehata6@outlook.com

Received: 10 May 2019
Accepted: 16 August 2019

Abstract

Introduction: Thompson and Austin Moore prostheses have been commonly used in hemiarthroplasties for displaced femoral neck fractures. There has been considerable debate about which of these prostheses is preferred. The purpose of this meta-analysis was to compare historical data for clinical outcomes of cemented Thompson and uncemented Austin Moore hemiarthroplasty in displaced femoral neck fractures.

Methods: We searched Medline via PubMed, Cochrane Central, Scopus, Ovid and Web of Science for relevant articles up to February 2019. The included outcomes measured were hip function, hip pain, implant-related complications, surgical complications, reoperation rate and hospital stay. The data were pooled as risk ratio (RR) or mean difference (MD) with 95% confidence interval (CI) between the two compared groups in a meta-analysis model.

Results: Ten studies (four RCTs and six observational studies) with a total of 4378 patients were included in the final analysis. The pooled RR showed that the Thompson group was associated with a lower incidence of postoperative hip pain (RR = 0.66, 95% CI [0.54, 0.80]), lesser reoperation rate (RR = 0.46, 95% CI [0.24, 0.88]), lesser intraoperative fractures (RR = 0.15, 95% CI [0.09, 0.25]), but a longer operative time (MD = 12.04 min, 95% CI [2.08, 22.00]) in comparison to the Austin Moore group. The effect estimate did not favour either group in terms of hip function, periprosthetic fractures, prosthetic dislocations, wound infection, mortality and hospital stay.

Conclusion: Evidence shows that Thompson hemiarthroplasty is better than Austin Moore hemiarthroplasty in terms of hip pain, reoperation rate and intraoperative fractures. Whereas the postoperative hip function is equivalent, these results could be considered when assessing the outcomes in modern hips.