This Week in Orthopedics — Sep 3, 2026

Welcome to This Week in Orthopedics. This week we're covering 10 notable papers spanning fracture fixation and the timing of surgery, arthroplasty technology and its complications, and the longer view on joint preservation, infection, and spine outcomes. Let's dive in.

We start with trauma, where two papers ask very different questions about how and when we operate. In the Bone and Joint Journal, Larsen and colleagues report a patient- and assessor-blinded multicentre randomized trial of 122 adults with displaced patellar fractures, assigned either to locking plate fixation or to traditional tension-band wiring. At twelve months, the plate group did better on three of the five Knee Injury and Osteoarthritis Outcome Score sub-scales, and the differences were not just statistically significant but clinically meaningful: roughly ten points better for symptoms, about fourteen points better for sport and recreation, and about eleven points better for quality of life. Hardware removal, fixation failure, and reoperation were all more common after tension-band wiring. This is the highest-level evidence we have on a fracture that most of us still treat with a technique essentially unchanged for decades, and it argues that plating should become the default rather than the alternative. The Journal of Bone and Joint Surgery tackles the other half of the equation, timing. King and colleagues studied 2,358 patients aged 65 and over with operatively managed hip fractures at a United Kingdom teaching hospital, and did something previous studies have struggled to do: they excluded everyone whose surgery was delayed for medical optimization, isolating delay caused purely by lack of theatre capacity. More than half of patients waited beyond 36 hours for capacity reasons alone. After propensity-score matching, that capacity delay was associated with roughly a 37 percent higher hazard of death at one year, and the effect was sharpest in the first 30 days, where the hazard was around 70 percent higher. Critically, the harm concentrated in frailer patients — those with a Clinical Frailty Scale score above four had substantially raised short-term mortality with delay, while the effect in fitter patients was not statistically significant. The practical message is about triage: when your list is oversubscribed, the frail patient is the one who cannot afford to wait.

Timing also drives a paper in Arthroscopy on traumatic massive rotator cuff tears. Geng and colleagues retrospectively compared 78 patients repaired either within three weeks of injury or later. Postoperative American Shoulder and Elbow Surgeons scores favoured early repair by about six points, and that advantage held after adjusting for age, sex, comorbidity, and baseline score. Two caveats deserve airtime. Every single patient in both groups exceeded the minimal clinically important difference, so delayed repair still helped substantially. And although retears were numerically more common in the delayed group — just over half versus just under 40 percent — that difference was not statistically significant, and the authors are explicit that the study was underpowered to exclude a clinically relevant difference. Treat this as support for prioritising these patients on the list, not as proof that a delay ruins the tendon.

Turning to arthroplasty, the Bone and Joint Journal publishes a prospective double-blinded randomized trial from Kayani and colleagues comparing conventional jig-based unicompartmental knee arthroplasty with robotic arm-assisted surgery in 107 patients with medial compartment osteoarthritis. The robotic arm delivered more accurate femoral and tibial component positioning, lower inpatient pain scores, less opioid use, and shorter hospital stays. But here is the honest finding: at two years there was no difference in the Oxford Knee Score, the Knee Injury and Osteoarthritis Outcome Score, or the Western Ontario and McMaster index. The only patient-reported measure that separated the groups was the Forgotten Joint Score, better in the robotic group at six months and at two years. There is also a striking methodological note — recruitment fell away as patients increasingly requested robotics and surgeons lost equipoise, which tells you something about how hard randomized evidence in this space is becoming to generate. For now, robotics buys accuracy and a smoother early recovery, not a demonstrably better knee on conventional scores.

Complications after arthroplasty get two complementary treatments. Also in the Bone and Joint Journal, Swain and colleagues linked United Kingdom primary care and hospital records across 476,620 knee, hip, and shoulder arthroplasties. Overall six-month venous thromboembolism incidence was about one and a half percent, highest after knee and lowest after shoulder replacement. The finding that should change how you read the literature is that more than 30 percent of these events were only visible in the primary care data — meaning hospital-only studies have been systematically undercounting. Rates declined over time for hip and knee, but the successive national guidelines in 2010 and 2012 did not significantly shift them; only shoulder arthroplasty showed a measurable drop after the 2018 guidance. On infection, the Journal of Bone and Joint Surgery offers a 50-year retrospective from Abedi and colleagues charting the shift in periprosthetic joint infection from empiric, procedure-driven care to a multimodal diagnostic framework using validated criteria, synovial and serological biomarkers, and molecular techniques, with surgical strategy now individualised to host, pathogen, and limb. Sitting alongside that review, Weintraub and colleagues in the Bone and Joint Journal report over 300 articulating high-dose antibiotic spacers used in two-stage revision hip arthroplasty. Five-year survival free of reinfection was 89 percent, which is a genuinely good result in a population where 40 percent had already had prior infection-related surgery. Poorer host and limb grades roughly tripled reinfection risk. Spacer fracture and spacer dislocation each occurred in about five percent of patients, and the actionable detail is that a spacer that dislocated predicted dislocation after reimplantation by roughly six-fold — a clear argument for planning a higher-stability bearing in those patients from the outset.

Our last theme is the long view — does what we do actually change natural history? Tanaka and colleagues, in the Bone and Joint Journal, reviewed 683 hips treated with transposition osteotomy of the acetabulum against 353 untreated contralateral hips. In frank dysplasia, twenty-year survival free of severe arthritis or hip replacement was 88 percent after osteotomy versus 69 percent in native hips, supporting a real disease-modifying effect. In borderline dysplasia the picture is different: native borderline hips did well anyway, at 96 percent twenty-year survival, and osteotomy did not significantly improve failure-free survival, although it did slow radiographic arthritis progression. The message is to be considerably more selective before offering periacetabular surgery for borderline morphology. Also in Arthroscopy, Ren and colleagues describe an all-arthroscopic three-tunnel suture-bridge fixation for posterior cruciate ligament tibial avulsion fractures in 22 patients followed at least two years, with posterior tibial translation falling from about 8 millimetres to roughly 3 millimetres, union within three months in all cases, and every patient meeting the clinically important threshold on the International Knee Documentation Committee, Lysholm, and Tegner scores. It is a small Level IV series with no comparator, so read it as a promising technique description rather than evidence of superiority. Finally, Spine carries a national Finnish registry study from Hatakka and colleagues linking three registries across lumbar spinal stenosis surgery. The overall 30-day complication rate was just under 17 percent, driven by dural tears, urinary retention, and wound infection, and rising with older age, higher anaesthetic risk grade, longer operative time, multilevel decompression, and fusion. Reassuringly, disability scores at one year were only modestly worse in patients who had a complication. Less reassuringly, complications raised one-year reoperation risk to about 14 percent overall, and among patients with a spinal haematoma or a wound infection, more than half came back to theatre.

If you only have time for one paper this week, make it the patellar fracture trial in the Bone and Joint Journal. It is a blinded multicentre randomized trial that directly challenges a technique most of us still use by default, and it is immediately actionable in your next trauma list.

Here are the key takeaways from this week in Orthopedics. Locking plate fixation beat tension-band wiring for displaced patellar fractures on patient-reported function at one year, with fewer reoperations and less hardware removal. Capacity-driven delay beyond 36 hours to hip fracture surgery was associated with higher mortality, and frail patients bore almost all of that harm, so prioritise them when theatre time is scarce. Robotic unicompartmental knee arthroplasty improved component accuracy and early recovery but did not improve conventional knee scores at two years. Hospital-only data undercount venous thromboembolism after arthroplasty by roughly a third, and successive guidelines have not clearly moved the needle for hips and knees. And in hip dysplasia, acetabular reorientation appears genuinely disease-modifying in frank dysplasia but not clearly so in borderline hips, which do well left alone.

That's your roundup for This Week in Orthopedics. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And if it saves you time, a quick rating in your podcast app helps other physicians find the show.

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