This Week in Orthopedics — Aug 20, 2026

Welcome to This Week in Orthopedics. This week we're covering 10 notable papers spanning shoulder and elbow surgery, bone health and spinal deformity, and the durability of joint preservation and biologic treatments. Let's dive in.

We'll start with the shoulder, where three papers approach the same joint from very different angles. In the Journal of Shoulder and Elbow Surgery, Harkin and colleagues tackled one of the most contentious decisions in shoulder arthroplasty — what to do with the biconcave or severely retroverted glenoid in a patient who still has an intact rotator cuff [1]. They compared anatomic total shoulder arthroplasty using an all-polyethylene posterior augmented glenoid against reverse total shoulder arthroplasty in Walch B2 and B3 glenoids, drawing on prospectively collected data from just over two hundred shoulders and then propensity matching thirty-six patients per group on age, sex and retroversion. At two years, the patient-reported outcomes were essentially superimposable — American Shoulder and Elbow Surgeons scores of about 88 in both arms, identical pain scores, and the same rates of achieving minimal clinically important difference. Revision rates were low in both groups, around three percent. Where the groups diverged was motion and radiographs. The anatomic group gained roughly eight more degrees of external rotation and about thirteen more degrees of forward flexion, but radiolucency around the glenoid component was seen in about four in ten anatomic shoulders compared with fewer than one in ten reverse shoulders. That's a Level III retrospective comparison with small matched groups, and two years is early for a polyethylene glenoid. The practical reading is that both operations work, the anatomic augmented glenoid buys motion in a younger patient with an intact cuff, and the radiolucency signal is a reason to keep these patients on the radar rather than a reason to abandon the implant.

Staying in the shoulder but moving to instability, the American Journal of Sports Medicine published five-year follow-up on arthroscopic Latarjet from Berge and colleagues in Oslo [5]. Eighty-five patients, mostly young men, and a genuinely hard population — two thirds of the cohort had more than ten dislocations before surgery, and over four in ten were revision procedures. At a median of just over five years, only two patients had recurrent dislocation and eight reported subluxations. Western Ontario Shoulder Instability Index scores improved substantially and, importantly, were stable between one year and five years, so the early result is the durable result. Quality of life on the EQ-5D matched general population norms, and 87 percent of patients were satisfied. Ten patients did require reoperation for complications, which is the number to quote in consent. Set against that, JAMA Internal Medicine published a review by Haas and colleagues on shoulder pain in primary care that pushes hard in the opposite direction for the far commoner nontraumatic presentation [6]. They argue for the umbrella term subacromial pain over the tangle of impingement, bursitis and tendinopathy labels, and they make the case that early imaging is not indicated without trauma or red flags, because structural findings often don't correlate with symptoms and generate overtreatment. Their strongest statement is that high-certainty evidence shows subacromial pain does not benefit from surgical intervention. Read together, these two papers draw a clear line — structural instability with bone loss is a surgical problem with excellent durable results, while atraumatic subacromial pain is a watchful-waiting problem, and the harm comes from confusing the two.

Rounding out the upper limb, a systematic review in the Journal of Shoulder and Elbow Surgery from Nadeem and colleagues pooled ten comparative studies and more than seventeen hundred patients undergoing distal biceps tendon repair with either cortical button or suture anchor fixation [10]. Function, range of motion and strength recovery — around 85 to 95 percent of the contralateral limb — were comparable, and pooled complication rates showed no significant difference in re-rupture, lateral antebrachial cutaneous nerve neuropraxia, or posterior interosseous nerve palsy. The striking number is that nerve events dominated the complication profile, with lateral antebrachial cutaneous neuropraxia accounting for about two thirds of all complications and occurring in roughly a quarter of patients in both groups. So the implant debate looks settled by equipoise, and the consent conversation should centre on sensory nerve symptoms rather than on which device you plan to use.

The second theme this week is bone quality as a modifiable target in spine surgery, and here two papers in Spine sit in productive tension. Maayan and colleagues performed a systematic review and meta-analysis of perioperative anabolic osteoporosis pharmacotherapy in adult spinal deformity surgery — thirteen studies, over two thousand patients, with teriparatide the agent in nine of them [3]. Anabolic therapy around the time of surgery was associated with roughly half the odds of proximal junctional kyphosis and about a third of the odds of reoperation for mechanical failure. Heterogeneity for the kyphosis outcome was zero, and leave-one-out analysis preserved the direction of effect. The honest caveat, which the authors state plainly, is that only one of thirteen studies was randomised and the certainty of evidence is low. Alongside it, Lim and Kang published a synthesis-without-meta-analysis review focused specifically on degenerative lumbar scoliosis and long-segment fusion [7]. They found that most core reports linked lower bone quality to mechanical complications — with osteoporosis associated with several-fold higher odds of screw loosening and of mechanical complications — but they declined to pool anything, citing retrospective design, probable cohort overlap, inconsistent regression tables and measure-dependent null findings, and they graded certainty as very low. Their verdict on perioperative anabolic therapy is that it is promising but not definitive, and that no class effect specific to degenerative lumbar scoliosis has been established. The clinical signal is consistent even where the statistical confidence is not — screen bone density before long fusions, and treat osteoporosis as something you act on preoperatively rather than merely document.

The modifiable-risk theme continues in the third Spine paper, an observational study from the Canadian Spine Outcomes and Research Network reported by Chan and colleagues, covering more than six thousand patients undergoing surgery for degenerative lumbar conditions [9]. Thirty-seven percent had obesity, and about five percent had a body mass index of forty or above. Patients with any class of obesity were less likely to reach the minimal clinically important difference on the Oswestry Disability Index at one year, with the effect most pronounced in severe obesity. The complication gradient was steeper — in Class III obesity, incidental durotomy was modestly more common, surgical site infection about half again as likely, and wound dehiscence more than doubled. These patients still improve; they simply improve less reliably and with more perioperative trouble, which is a shared-decision-making conversation, not an automatic exclusion.

Finally, two papers on durability and biologics. In Knee Surgery, Sports Traumatology, Arthroscopy, Gosnell and colleagues reported minimum ten-year outcomes after primary hip arthroscopy for structurally stable femoroacetabular impingement performed with interportal capsulotomy and no capsular repair [4]. In 150 patients followed a mean of nearly twelve years, survivorship was about 91 percent, with roughly five percent converting to hip arthroplasty and another five percent undergoing revision arthroscopy. Modified Harris Hip Score improved by around 36 points, and about three quarters of patients reached an acceptable symptom state. Female sex and higher body mass index predicted worse ten-year scores, while Tönnis grade 1 changes and lower preoperative function predicted outright failure. This is a single-surgeon Level IV series without a comparison group, so it doesn't resolve the capsular repair debate — but it does give you decade-long benchmarking numbers for counselling a well-selected, structurally stable hip. In the same journal, Delgado and colleagues published a meta-analysis with meta-regression of placebo-controlled trials of platelet-rich plasma in knee osteoarthritis [8]. Across twenty studies and roughly thirty-three hundred patients, platelet-rich plasma beat placebo for pain and function at both six and twelve months, but the certainty of that evidence was rated very low. Younger age, a higher platelet increase factor, leucocyte-rich preparations and calcium chloride activation were all associated with better six-month function — yet these variables were interrelated, none of them explained results at twelve months, and the authors are explicit that this is hypothesis-generating, not a mandate for a particular formulation.

One more paper deserves attention for what it dismantles. In the Journal of Bone and Joint Surgery, Berthelsen and colleagues tested the assumption that antibiotic-loaded ceramic bone grafts deliver four to six weeks of local protection [2]. Using a murine implantation model with mass spectrometry quantification, they found that both commercial gentamicin-loaded products — the hydroxyapatite–calcium sulfate composite and the pure calcium sulfate — released more than 98 percent of their antibiotic within six hours and more than 99 percent within twenty-four hours in vivo. Their reanalysis suggests the apparent prolonged elution in earlier laboratory studies was an artifact of serial dilution in the assay rather than sustained release. If you are relying on these grafts for weeks of local antimicrobial cover in bone infection, that assumption is not supported.

If you only have time for one paper this week, make it the Journal of Bone and Joint Surgery study on antibiotic-loaded ceramic bone grafts [2]. It directly contradicts a widely held belief that shapes how many surgeons plan systemic antibiotic duration and dead-space management in bone infection, and the practice implication is immediate.

Here are the key takeaways from this week in Orthopedics. Antibiotic-loaded ceramic grafts give a burst, not a depot — don't shorten systemic therapy on the assumption of weeks of local cover. In Walch B2 and B3 glenoids with an intact cuff, anatomic arthroplasty with a posterior augmented glenoid and reverse arthroplasty give equivalent two-year patient-reported outcomes; anatomic gives more motion, reverse gives cleaner radiographs. Arthroscopic Latarjet holds up at five years in a hard instability population, while atraumatic subacromial pain should be managed without imaging and without surgery. Bone health is emerging as a modifiable target before long spinal fusions, with anabolic therapy associated with fewer proximal junctional kyphosis events, though the evidence remains low certainty. And in both lumbar fusion and hip arthroscopy, elevated body mass index consistently predicts less improvement — worth quantifying in the preoperative conversation.

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 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.