This week in Orthopedics

This AudioScholar briefing is generated by artificial intelligence for healthcare professionals and trainees. It is not medical advice.

This week: aspirin falls short of anticoagulants after hip and knee replacement, a weekly triple agonist eases knee osteoarthritis pain in obesity, and the anterior approach to hip replacement does not beat the posterior approach on patient-reported recovery.

Welcome to This Week in Orthopedics. We're covering 10 papers this week: 3 in depth, 4 in brief, 3 quick signals.

The New England Journal of Medicine published PEPPER, a comparative-effectiveness noninferiority trial led by Pellegrini [1]. Patients having elective hip or knee replacement were randomised to aspirin 81 milligrams twice daily, warfarin targeted to an international normalized ratio of 2.0, or rivaroxaban 10 milligrams daily, each for 30 days. Of 18,883 patients randomised, 15,317 entered the per-protocol analysis, split fairly evenly between hips and knees. The primary effectiveness outcome was death or symptomatic venous thromboembolism through six months. Event rates were low in every arm: about 1.3 percent with aspirin, 1.2 percent with warfarin and 0.9 percent with rivaroxaban. Aspirin did not meet the prespecified noninferiority margin against either anticoagulant. On the safety composite of bleeding, wound reoperation or myocardial infarction, events ran at about 3 percent with aspirin and about 3.5 percent with the anticoagulants, and neither warfarin nor rivaroxaban was shown noninferior to aspirin. Intention-to-treat results were similar. The main limitation is that the margin was tight and the absolute differences small, so failing to show noninferiority does not prove aspirin worse. The trial adds a large randomised comparison that frames prophylaxis as a trade-off between slightly fewer clots and slightly more wound and bleeding events.

Staying with arthroplasty, Clinical Orthopedics and Related Research published a randomised trial by Rykov and colleagues comparing the direct anterior and posterolateral approaches for total hip replacement [2]. The trial enrolled 128 patients with hip osteoarthritis receiving an uncemented stem and cemented cup, and the abstract does not state the country or number of sites. The primary outcome was the proportion reaching an acceptable symptom state on the Hip Disability and Osteoarthritis Outcome Score at three months, a measure chosen to avoid the ceiling effects of older scores. The approaches did not differ on any subscale. For symptoms, for example, about two thirds of patients in each group reached an acceptable state. There were also no differences in function, participation or quality of life at any point through one year, including the early two-week and six-week visits. Complications occurred in five anterior patients and one posterolateral patient, a difference that was not statistically significant. The trial was nonblinded and powered only to detect a twenty-point difference, so smaller effects and complication differences cannot be excluded. It adds randomised evidence against a faster or larger recovery with the anterior approach, and the authors conclude that this advantage should not be represented to patients.

The New England Journal of Medicine also published two phase 3 trials of retatrutide, a weekly triple agonist of the GIP, GLP-1 and glucagon receptors, reported by Schnitzer and colleagues [3]. TRIUMPH-4 randomised 445 adults and TRIUMPH-1-OA randomised 574, all without diabetes, with painful knee osteoarthritis and a body-mass index of at least 27. In TRIUMPH-4, pain on the ten-point WOMAC scale fell by roughly 1.6 to 1.8 points more than placebo at 68 weeks, and body weight fell by about a fifth or more, against roughly 5 percent with placebo. In TRIUMPH-1-OA, the pain advantage over placebo at 80 weeks ranged from about 1.4 to 1.8 points across doses. Results held under the intention-to-treat analysis. Serious adverse events were more common with retatrutide, and gastrointestinal effects were the most frequent. The trials were industry funded, the placebo groups also improved by about two points, and the abstract reports no structural joint outcomes, so these data concern symptoms and weight rather than disease modification. The trials add phase 3 evidence that a pharmacological weight-loss agent can reduce knee osteoarthritis pain in people with obesity.

Now, four in brief.

Also in knee osteoarthritis, the American Journal of Sports Medicine published a multicentre placebo-controlled trial by Garza and colleagues testing a single injection of autologous adipose-derived stromal vascular fraction, prepared at the point of care, in 114 adults with moderate-to-severe disease [4]. At twelve months, about half of treated patients met the composite responder definition for pain and function, against 29 percent with placebo, and no serious device-related adverse events were reported. The placebo arm had only 31 patients, the analysis used a one-sided test, and the abstract reports no imaging results despite MRI at baseline and twelve months, so the evidence is early.

Clinical Orthopedics and Related Research published a propensity-weighted cohort study by Randall and colleagues using a Florida statewide inpatient database of 3545 admissions for native hip or knee septic arthritis treated with debridement [5]. In septic knees, debridement on hospital day two or later was linked to more than double the risk of repeat debridement and about a 50 percent higher one-year mortality than debridement on day zero or one. In septic hips, neither outcome differed significantly by timing. This is retrospective administrative data without information on antibiotic timing or diagnostic delay, so confounding remains possible.

Arthroscopy published a meta-analysis by Giai Via and colleagues of five randomised trials, 615 patients in total, comparing capsular repair with no repair after hip arthroscopy for femoroacetabular impingement [6]. Most patient-reported scores, including the modified Harris Hip Score and the International Hip Outcome Tool, showed no significant difference. The Hip Outcome Score favoured repair, and revisions were fewer with repair, at 8 versus 15 cases. The benefit appeared mainly in trials with unrepaired T-capsulotomy, while interportal capsulotomy trials showed similar outcomes. With five trials and small revision counts, the signal is modest.

Knee Surgery Sports Traumatology Arthroscopy published a retrospective comparison by Maione and colleagues of monoplanar and biplanar medial closing-wedge distal femoral osteotomy in 86 valgus knees [7]. Coronal correction was equivalent, and noninferiority of the biplanar cut was formally shown. Lateral hinge fractures occurred in three quarters of monoplanar knees against about a fifth of biplanar knees. Both groups improved beyond clinically important thresholds on patient-reported scores. The cohort spans fourteen years, so technique changes over time may confound the comparison, and the authors call for prospective study.

Three quick signals.

The Journal of Shoulder and Elbow Surgery published the 2025 International Consensus Meeting recommendations on preventing shoulder periprosthetic infection, led by Namdari [8]. Consensus was reached on all 25 questions, with moderate-strength evidence against corticosteroid injection within three months of arthroplasty. Evidence was insufficient for every intraoperative measure reviewed, and many recommendations rest on extrapolated hip and knee data.

Arthroscopy published an international Delphi consensus led by Chahla, in which 22 experts retained 35 of 39 statements on meniscal centralization [9]. The panel supported centralization for root tears with more than 3 millimetres of extrusion on MRI and discouraged it in advanced osteoarthritis. This is level five expert opinion, and the panel calls for long-term data.

Also in Arthroscopy, a commentary by Lädermann and Vukanic argues that isolated arthroscopic Bankart repair is no longer a default first-line procedure in higher-risk shoulders, citing higher recurrence than with remplissage or Latarjet [10]. It is an opinion piece drawing on existing literature and presents no new data.

Stepping back across this week's papers, the strongest designs mostly returned null or noninferiority-failing results on the outcomes patients feel, while the more confident claims came from consensus panels, retrospective databases and a commentary. Several consensus documents openly rest on extrapolated or limited evidence. Where positive randomised data did appear, the endpoints were symptom scores and weight over a year or so, with structural and long-term joint outcomes still unreported.

If you only have time for one paper this week, make it PEPPER [1]. It reopens the question of whether low-dose aspirin is an adequate default for thromboprophylaxis after hip and knee replacement, and it puts the bleeding and wound cost of anticoagulants into the same randomised frame.

That's your roundup for This Week in Orthopedics. The full transcript and references are available on the episode page. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And a quick rating on Apple Podcasts or Spotify helps other physicians discover the show.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.