Internal fixation versus arthroplasty for hip fracture in patients with dementia or cognitive impairment: a systematic review and meta-analysis

<p><strong>Background: </strong>Hip fractures in patients with dementia or cognitive impairment are associated with increased morbidity, mortality, complications, and functional decline. Arthroplasty is generally preferred to internal fixation in older adults with a displaced femoral neck fracture, but the optimal strategy in this population remains uncertain because of frailty, limited rehabilitation potential, and the frequent exclusion of patients with dementia from comparative trials. We compared the outcomes of internal fixation and arthroplasty after hip fracture surgery in patients with dementia or cognitive impairment.</p> <p><strong>Methods: </strong>PubMed/MEDLINE, Web of Science, and Google Scholar were searched from inception to July 2026 for comparative studies of internal fixation (IF), hemiarthroplasty (HA), or total hip arthroplasty (THA) in patients with dementia or cognitive impairment who sustained a hip fracture. Randomized, quasi-randomized, and observational studies were eligible. Studies comparing two arthroplasty constructs were analysed separately from internal fixation versus arthroplasty comparisons. Dichotomous outcomes were pooled as risk ratios (RR) using Mantel-Haenszel fixed-effect models and continuous outcomes as mean differences (MD) using inverse-variance random-effects models, with 95% confidence intervals (CI) and, where estimable, 95% prediction intervals (PI). Certainty of evidence was rated with GRADE. The protocol was prospectively registered in PROSPERO (CRD420261468833).</p> <p><strong>Results: </strong>Eight studies were included. Six enrolled patients with displaced femoral neck fracture exclusively; two registry or claims cohorts covered hip fracture more broadly. One quasi-randomized trial compared two arthroplasty constructs and was analysed separately. In the internal fixation versus arthroplasty comparisons, operative time was numerically shorter with IF but the difference was not significant (MD &minus;29.41 minutes, 95% CI &minus;110.82 to 52.01; I&sup2; = 78.6%; two trials). Reoperation or revision was more frequent after IF (RR 3.36, 95% CI 1.33 to 8.48; I&sup2; = 0%; two trials). Mortality was higher after IF before one year (RR 1.20, 95% CI 1.07 to 1.36), at one year (RR 1.23, 95% CI 1.16 to 1.29), and at two years (RR 1.21, 95% CI 1.14 to 1.28), with I&sup2; = 0% throughout; the prediction interval excluded the null value of 1 at one and two years but not at the short-term landmark. The EQ-5D index at 12 months did not differ significantly between strategies (MD &minus;0.07, 95% CI &minus;0.25 to 0.10). Certainty of evidence was very low for all outcomes.</p> <p><strong>Conclusions: </strong>In patients with dementia or cognitive impairment, arthroplasty was associated with fewer reoperations and lower mortality than internal fixation, although the reoperation estimate rests on two small trials and differences in health-related quality of life remained uncertain. Given the very low certainty of the evidence, the surgical choice should be individualized according to fracture characteristics, frailty, functional status, and instability risk.</p> <p>&nbsp;</p>