Kategoria: Updates for Doctors (EN)

In non-operative treatment of proximal humerus fractures, considerable attention is often paid to how restrictive the immobilisation should be. A randomized trial suggests that choosing between a Velpeau-type bandage, a standard shoulder-arm sling and a sling with additional support does not lead to meaningful differences in functional or radiographic outcomes.

Polish version: Złamanie bliższego końca kości ramiennej — czy rodzaj unieruchomienia ma znaczenie?

Why is this study clinically relevant?

Immobilisation is almost routine when a proximal humerus fracture is selected for conservative management, but the exact device used is often influenced more by local practice than by strong comparative evidence. The practical question is therefore not only whether to immobilise the shoulder, but whether a more elaborate brace produces a real patient benefit.

The study was published online on 21 August 2026 in the Journal of Shoulder and Elbow Surgery.

Study question and methodology

This prospective randomized study included patients with proximal humerus fractures treated non-operatively. Three immobilisation methods were compared:

• a Velpeau-type bandage,
• a standard shoulder-arm sling/brace,
• a shoulder-arm sling with additional support.

The clinical outcome analysis included 34, 39 and 32 patients in the respective groups, for a total of 105. Mean age was 60.6 years and 63.8% of participants were women. Baseline characteristics, including fracture severity according to the Neer classification, were comparable between groups.

Outcomes included pain, range of motion and several measures of shoulder and upper-limb function: SSV, ASES, DASH and Constant score.

Key results

No statistically significant differences were found between the three groups in pain, range of motion or functional scores. Radiographic outcomes likewise did not show an advantage for one immobilisation method over the others.

The authors reported nonunion in 5 of 120 patients (4.1%). Malunion was reported in 22 of 105 patients (21%), with comparable rates across the study groups.

Clinical interpretation for an orthopaedic surgeon

The most practical conclusion is straightforward: once a fracture has appropriately been selected for non-operative management, a more elaborate immobilisation device does not necessarily produce a better result. Choice of device may therefore reasonably take greater account of patient comfort, the ability to use the brace correctly and cost.

This may be particularly relevant in older patients, for whom complex immobilisation devices can be harder to apply, interfere with hygiene and reduce treatment tolerance.

Limitations

The findings need to be interpreted within the population studied. They do not mean that every proximal humerus fracture can be treated with any sling, or that immobilisation method is irrelevant when fracture alignment is unstable, surgery is indicated, or adherence to treatment presents particular problems.

Importantly, the absence of a statistically significant difference is not the same as formally proving equivalence between all three methods. The publicly available abstract does not indicate that the study was designed as an equivalence or non-inferiority trial.

There is also a numerical inconsistency in the abstract: the nonunion denominator is 120 patients, whereas the analysis of the other reported outcomes includes 105. This limits full interpretation without access to the detailed data in the complete article.

What might change in practice — and what does the study not prove?

The trial supports moving away from the assumption that a more restrictive or more expensive immobilisation device inherently improves the outcome of conservative treatment. When fracture stability and the patient’s condition allow, a simple, well-tolerated device that can be used correctly may be a reasonable choice.

The study does not answer which proximal humerus fractures should be treated non-operatively, when exercises should begin, or whether earlier mobilisation is better than longer immobilisation. Those are separate clinical decisions.

Reference

1. Sarıkas M, Kapıcıoğlu M, Şahin K, Bilsel K. Does Immobilization Method Make a Difference in Conservative Management of Proximal Humeral Fractures? A Prospective Randomized Trial. Journal of Shoulder and Elbow Surgery. 2026 Aug 21. DOI: 10.1016/j.jse.2026.08.004. PubMed PMID: 42628626.

← Research Updates, Literature and Evidence-Based Medicine

Content prepared by: Tomasz Poboży - orthopaedic surgeon.

A recent meta-analysis of 13 randomized trials suggests that duloxetine may reduce opioid consumption after total hip and knee arthroplasty. At the same time, the average reduction in pain was small and, at most time points, did not reach the threshold considered clinically important.

Polish version: Duloksetyna po endoprotezoplastyce — mniej opioidów, mały efekt przeciwbólowy

Why is this study clinically relevant?

Duloxetine is a serotonin-noradrenaline reuptake inhibitor (SNRI) used, among other indications, for neuropathic pain and chronic pain syndromes. Interest in arthroplasty comes from its potential role in multimodal analgesia and opioid-sparing strategies. The clinically important question, however, is not only whether a difference is statistically significant, but whether it is large enough for patients to notice.

Study question and methodology

Ibrahim, Abdelnabi and Sheta analysed 13 randomized trials published between 2010 and 2024, including 1,176 patients after total hip or total knee arthroplasty. Outcomes included pain at rest and while walking, opioid consumption converted to morphine equivalents, and adverse events. Random-effects models were used when heterogeneity was substantial, with additional sensitivity analyses for selected outcomes.

Key results

• At 24 hours, duloxetine reduced pain at rest by an average of 5.09 mm and pain while walking by 4.55 mm on a 0–100 mm scale.
• At 2 weeks, the corresponding differences were 6.48 mm and 8.93 mm.
• Most of these differences were statistically significant, but remained below the 10 mm minimal clinically important difference adopted by the authors.
• Opioid use was lower by 22.74 mg morphine equivalents at 48 hours and by 10.58 mg at 72 hours. Heterogeneity for these analyses was extremely high, however: I² 100% and 97%, respectively.
• Nausea or vomiting occurred less often with duloxetine, while somnolence was more frequent.

Clinical interpretation for an orthopaedic surgeon

The most important finding is the gap between statistical significance and effect size. The results do not support the claim that duloxetine provides clearly superior pain control for the average patient after joint replacement. The opioid-sparing signal may nevertheless be clinically useful as one component of a broader multimodal analgesia protocol.

The reductions of 22.74 mg and 10.58 mg morphine equivalents should not be treated as precise, universally expected effects. With I² values of 97–100%, the individual studies differed so markedly that the pooled estimates are better interpreted as indicating a direction of effect than as predicting a specific benefit for a particular protocol.

Limitations

• Substantial heterogeneity affected many pain analyses and virtually all analyses of opioid consumption.
• Two open-label studies were judged to have a high risk of bias for subjective outcomes.
• One study did not use an intention-to-treat analysis.
• Several trials had problems with missing data or lacked a pre-registered analysis plan.
• When the more conservative Hartung–Knapp adjustment was applied, some apparently significant pain results were no longer statistically significant.

What might change in practice — and what does the study not prove?

The meta-analysis strengthens the argument that duloxetine may be considered as one element of an opioid-sparing strategy in appropriately selected patients. It does not justify routine use in every patient after THA or TKA, and it does not demonstrate a large, clinically important analgesic advantage. It also does not establish the optimal dose, timing, duration of treatment or the subgroup most likely to benefit.

Duloxetine has its own adverse-effect and interaction profile. Any perioperative use therefore needs to follow a defined protocol and individual assessment of contraindications rather than the meta-analysis result alone.

Reference

Ibrahim A, Abdelnabi M, Sheta RA. Duloxetine for postoperative pain management in total hip and knee arthroplasty: a systematic review and meta-analysis. Journal of Orthopaedic Surgery and Research. 2026;21:527. DOI: 10.1186/s13018-026-07176-6. Published 2 September 2026.

← Research Updates, Literature and Evidence-Based Medicine

Content prepared by: Tomasz Poboży - orthopaedic surgeon.

A randomized trial in patients with Weber C ankle fractures and deltoid ligament injury found an important difference between early mechanical stability and later clinical outcome. Direct deltoid repair improved early talar stability on weight-bearing or stress imaging, but at 12 months it did not provide an advantage in the radiographic outcomes assessed or in ankle function.

Polish version: Złamanie Weber C — czy naprawiać więzadło trójgraniaste?

Why is this study clinically relevant?

In unstable ankle fractures with injury to the medial ligament complex, a practical question remains after fracture anatomy has been restored: should the deltoid ligament be repaired routinely? In theory, direct repair may improve stability of the medial column, but it also adds another surgical step. Choi and colleagues evaluated this question in a randomized controlled trial.

Study question and methodology

The study included patients with Weber C fractures corresponding to OTA/AO type 44C, produced by a pronation-external rotation mechanism and associated with deltoid ligament injury. After the fracture fixation strategy had been established, patients were randomized to direct repair of the deltoid ligament or to no ligament repair.

Medial clear space and talar tilt were assessed on standing and stress radiographs at 3, 6 and 12 months. Functional outcome was measured with the FAAM. Seventy-one patients completed the 12-month follow-up: 35 in the repair group and 36 in the no-repair group.

Key results

• Medial clear space was comparable between groups across the imaging views and time points assessed.
• At 3 months, the group without deltoid repair showed greater valgus talar tilt on stress imaging. This was a signal of better early stability after direct repair.
• At 12 months, the difference in talar tilt was no longer statistically significant.
• FAAM scores improved in both groups. At 12 months, the activities-of-daily-living component was approximately 89–90% in both groups.

Clinical interpretation for an orthopaedic surgeon

The most useful conclusion is not that the deltoid ligament should never be repaired. Rather, in this specific fracture group, routinely adding medial ligament repair did not translate into better function at one year despite an early improvement in talar stability.

This distinction matters. A better postoperative radiographic parameter is most clinically important when it translates into better function, less late instability, fewer reoperations or lower risk of post-traumatic degeneration. A 12-month follow-up cannot resolve all of those questions.

Limitations

• The trial was conducted at a single centre and included 71 patients with complete one-year follow-up.
• It addressed a narrowly defined group of Weber C fractures with deltoid ligament injury; the findings should not automatically be applied to every unstable ankle fracture.
• One year is adequate for assessing early function but too short for reliable evaluation of late instability, post-traumatic osteoarthritis or revision risk.
• The absence of a statistically significant difference at 12 months is not formal proof that the two strategies are equivalent.
• The publicly available abstract does not answer every subgroup question, including whether particular patterns of medial injury may benefit more from repair.

What might change in practice — and what does the study not prove?

The trial provides an argument against treating direct deltoid ligament repair as an automatic component of every Weber C fracture fixation when the ligament is injured. The decision may still depend on the quality of fracture reduction, residual ankle stability after fixation, the character of the medial injury and the overall fracture pattern.

The study does not prove that repair is unnecessary in every patient, nor does it prove that a transient early difference in stability has no longer-term significance. Larger cohorts and longer follow-up are needed to answer those questions.

Reference

Choi JY, Yu OJ, Jung YH, Suh JS. Deltoid Ligament Repair Does Not Improve Radiographic or Clinical Outcomes in Weber Type C Pronation-External Rotation Ankle Fractures: A Randomized Controlled Trial. Journal of Orthopaedic Trauma. 2026;40(8):399–407. DOI: 10.1097/BOT.0000000000003189. Issue date: 1 August 2026.

← Research Updates, Literature and Evidence-Based Medicine

Content prepared by: Tomasz Poboży - orthopaedic surgeon.

A new formal ESSKA-ESA consensus clarifies how partial-thickness posterosuperior rotator cuff tears should be described and assessed. The most practical message is that these lesions should not be reduced to a single percentage of tendon thickness or one classification. The report should reflect location, depth and the tendon involved, while also distinguishing acute from chronic tears. At the same time, the document makes clear where evidence is still weak — particularly for natural progression and the choice of the best imaging test.

Polish version: Częściowe uszkodzenia stożka rotatorów — nowy konsensus

Why is this consensus useful?

Partial-thickness rotator cuff tears are common, but their clinical significance is often difficult to judge. In practice, it is easy to place too much weight on one ultrasound or MRI measurement of tear depth even though treatment decisions also depend on the lesion’s location, mechanism, tendon involvement, symptoms and the patient’s functional demands.

The consensus from the European Society of Sports Traumatology, Knee Surgery and Arthroscopy and the European Shoulder Associates, published on 7 September 2026, addresses this part of the problem: definitions, classification, aetiology, natural history and diagnosis of partial-thickness posterosuperior rotator cuff tears.

Methodology

The steering group formulated 33 clinical questions, followed by a standardized literature search and evidence assessment. Statements were developed for each question with recommendation grades ranging from A to D, and an independent expert panel assessed them until strong consensus was reached.

This distinction matters. The publication is not a narrative review or an informal collection of expert opinions, but a formal consensus developed using a predefined ESSKA methodology. Its strength nevertheless remains constrained by the quality of the underlying evidence, which is limited in several areas.

Key findings

• One classification is not enough. The strongest recommendations support combining descriptive systems so that location, depth and involvement of individual tendons are captured.
• Acute and chronic tears should be distinguished. This also received a high-grade recommendation and directly affects interpretation and management.
• Articular-sided lesions are more common in overhead and contact athletes. The evidence supporting this finding was weaker than the evidence behind the core classification recommendations.
• No association was demonstrated between partial-thickness posterosuperior cuff tears and calcific tendinopathy.
• Natural history remains difficult to predict. Evidence on the risk and rate of progression is less robust.
• There is no single ideal imaging test. The consensus highlights limitations of current modalities in fully characterising these lesions.

Clinical interpretation for an orthopaedic surgeon

The most useful practical message concerns how these tears are described. A statement such as “50% tendon thickness tear” does not define the whole problem. Two lesions with similar depth may have different locations, mechanisms and tendon involvement, and may occur in patients with very different functional requirements.

The second point concerns imaging. Both ultrasound and MRI can be highly useful, but this consensus does not support treating any single examination as an infallible map of the lesion. Imaging should be interpreted together with the history and clinical examination. This is particularly important in partial-thickness abnormalities, where the distinction between tendinopathy, delamination and a true fibre defect may be less clear than in a large full-thickness tear.

Limitations

• This is a formal consensus, not a randomized trial comparing treatment strategies.
• Some recommendations rely on lower-quality evidence, particularly those concerning natural progression and optimal imaging.
• The document addresses partial-thickness tears of the posterosuperior cuff, so its conclusions should not automatically be applied to every type of rotator cuff lesion.
• This first part of the consensus focuses on definitions, classification, aetiology, natural history and diagnosis; it is not a ready-made algorithm for deciding who should undergo surgery.
• The authors label the publication “Level I, formal consensus”. This should not be equated with the evidence level of a well-conducted randomized trial evaluating treatment efficacy.

What may change in practice — and what does the consensus not prove?

The document may improve communication between radiologists, musculoskeletal ultrasonographers and shoulder surgeons. A report of a partial-thickness tear should ideally provide more than a percentage estimate: location, extent, tendon involvement and whether the lesion appears acute or chronic all add useful clinical context.

The consensus does not show that a particular percentage of tendon involvement automatically requires surgery, that one imaging modality is always superior to the others, or that progression can be predicted reliably in an individual patient. Better prospective evidence is still needed in these areas.

Reference

Kany J, Ruiz Iban MA, Rosso C, et al. The 2026 ESSKA-ESA formal consensus on partial-thickness posterosuperior rotator cuff tears (Part 1): Definitions, classifications, etiologies, natural history and diagnosis. Knee Surgery, Sports Traumatology, Arthroscopy. Published online 7 September 2026. DOI: 10.1002/ksa.70587. PMID: 42704696.

← Research updates, literature and EBM

Content prepared by: Tomasz Poboży - orthopaedic surgeon.

A recent double-blind randomized trial found no significant advantage over placebo for undenatured type II collagen, pharmaceutical-grade crystalline glucosamine sulfate, or diacerein in patients with mild to moderate knee osteoarthritis. The study is clinically useful because all three commonly discussed oral treatments were assessed within the same four-arm, placebo-controlled protocol, with more than 96% of participants completing follow-up.

Polish version: Kolagen typu II, glukozamina i diacereina w gonartrozie — bez przewagi nad placebo

Why is this study worth attention?

Oral agents often described as slow-acting treatments for osteoarthritis remain widely used despite inconsistent evidence of benefit. Glucosamine is particularly controversial, but similar questions apply to undenatured type II collagen and diacerein. The trial by Narkbunnam and colleagues, published online on 9 September 2026 in JB & JS Open Access, tested all three strategies under a single protocol against a common placebo group.

Study question and methods

The four-arm, double-blind randomized trial enrolled 216 adults in Thailand with symptomatic Kellgren-Lawrence grade 2 or 3 knee osteoarthritis. For 24 weeks, participants received one of the following:

• undenatured type II collagen, 40 mg once daily,
• pharmaceutical-grade crystalline glucosamine sulfate, 500 mg three times daily,
• diacerein, 50 mg once or twice daily,
• or placebo.

The primary endpoint was total WOMAC score at 24 weeks, with follow-up extended to 36 weeks. The analysis followed the intention-to-treat principle, using imputation and repeated-measures models to account for missing data.

Overall, 208 of 216 participants completed the study, a completion rate of 96.3%. Mean age was 61.7 years, 90.9% were women, and 73.6% had Kellgren-Lawrence grade 2 radiographic disease.

Results

At 24 weeks, none of the three active treatments was superior to placebo for the primary endpoint. The between-group differences in total WOMAC score versus placebo were:

• type II collagen: −4.7 points (95% CI −18.6 to 9.1),
• glucosamine: −8.1 points (95% CI −21.8 to 5.7),
• diacerein: −12.3 points (95% CI −25.9 to 1.3).

All confidence intervals crossed zero, and none of the comparisons demonstrated a clear difference exceeding the study’s threshold for minimal clinical importance. At 36 weeks, the pattern remained similar, with differences versus placebo of −1.8, −4.8 and −1.3 points, respectively.

No significant between-group differences were found for pain measures, joint-space width or performance tests. Adverse events were mild and similar across groups, and no serious adverse events were reported.

Clinical interpretation for an orthopaedic surgeon

The important point is not simply that the p values were above 0.05, but the size and precision of the observed effects. Point estimates for all three interventions favoured improvement, yet the effects were neither sufficiently precise nor sufficiently large to establish a reliable advantage over placebo. There was also no consistent supportive signal across secondary outcomes.

Clinically, these findings weaken the rationale for routinely using any of the three products as stand-alone treatment with the expectation of a substantial improvement in pain and function. The results may be especially useful when discussing realistic treatment benefit, cost and alternatives with better-established efficacy.

Limitations

The study was conducted in a Thai population, with a marked predominance of women and mostly Kellgren-Lawrence grade 2 disease. The findings therefore cannot automatically be assumed to apply identically to populations with different demographics or more advanced osteoarthritis.

Follow-up to 36 weeks is adequate for short-term symptomatic assessment but too short to support conclusions about long-term structural effects. The trial was also not designed to detect rare adverse events.

What might this change in practice — and what does it not prove?

The trial provides a strong argument against treating undenatured type II collagen, crystalline glucosamine sulfate or diacerein as routine monotherapy expected to deliver a clinically important improvement in knee osteoarthritis. It does not prove that no individual patient can improve, nor can the result be extrapolated automatically to every commercial formulation, different doses, combination therapy or patients with a different disease stage.

It also does not show that these interventions increase or reduce the long-term likelihood of joint replacement. Answering that question would require a substantially longer follow-up and a different study design.

Reference

1. Narkbunnam R, Awirotananon K, Chareancholvanich K, Achawakulthep C, Adulkasem N, Chatmaitri S, Hirunyachoke P, Pornrattanamaneewong C. A Comparison of Clinical Outcomes Among Undenatured Collagen Type II, Glucosamine Sulfate, and Diacerein on Osteoarthritic Knee Treatment: A Randomized Clinical Trial. JB & JS Open Access. 2026;11(3). DOI: 10.2106/JBJS.OA.26.00225. PMID: 42698930.

← Research updates, literature and EBM

Content prepared by: Tomasz Poboży - orthopaedic surgeon.

Polish version

For an acute humeral shaft fracture, applying a functional brace immediately after the injury may not be the most comfortable or safest option during the first few days. The multicentre randomised COBRAS trial, published in the Journal of Orthopaedic Trauma, found that immediate Sarmiento bracing was associated with more pain and more skin complications than initial coaptation splinting followed by conversion to a brace after the acute swelling had settled. The study addresses only the first 10 days of treatment and does not answer questions about fracture union or long-term function.

Why is this study relevant to an orthopaedic surgeon?

Non-operative treatment of selected humeral shaft fractures is based on functional bracing. A practical question remains, however: should the Sarmiento brace be applied immediately after injury, or is it better to use a splint during the period of greatest swelling and move to functional bracing later?

This may look like a technical detail, but it affects pain, treatment tolerance and the condition of the skin. The value of the COBRAS trial is that it compared the two strategies in a randomised design rather than relying only on local preference.

Study question and methods

The multicentre, parallel-group randomised trial included adults with an acute closed AO/OTA 12A, 12B or 12C humeral shaft fracture selected for non-operative treatment. The analysis included 58 patients.

Two strategies were compared:

• initial coaptation splinting, followed by transition to a Sarmiento brace after the acute swelling had decreased;
• immediate application of a Sarmiento brace.

The investigators recorded daily pain on a 0–10 scale for the first 10 days, opioid use converted to morphine equivalents, PROMIS function, patient satisfaction and skin complications related to pressure from the splint or brace.

Results

Patients who received an immediate Sarmiento brace reported more pain during the first 10 days. The mean treatment effect was 0.97 points on the 0–10 pain scale against immediate bracing (95% CI 0.26–1.69; p=0.010).

The skin-complication difference was more pronounced: complications occurred in 33.3% of patients treated with an immediate brace compared with 7.1% of those who started in a coaptation splint (p=0.022).

There was no significant difference in cumulative opioid use over 10 days: the median was 20.0 mg morphine equivalents in the splint group and 32.5 mg in the immediate-brace group (p=0.308). There was also no significant difference in function measured on day 7 or in patient satisfaction.

How should the findings be interpreted?

The most practical conclusion is straightforward: there is no clear reason to rush the transition to a functional brace at the expense of poorer tolerance during the acute phase of injury. During the period of greatest swelling, a conventional coaptation splint may be better tolerated, with conversion to a functional brace once the soft tissues have settled.

A pain difference of about one point is not large in isolation, but it occurred alongside more than a fourfold difference in skin complications. That makes the overall result more clinically persuasive than the pain score alone.

Importantly, immediate bracing did not produce a detectable advantage in early function, opioid use or satisfaction. The trial therefore supports a staged approach during the first days after injury.

Limitations

• The analysis included only 58 patients, so the sample was small.
• The main observation period covered only the first 10 days after injury.
• The study focused on pain, tolerance and skin complications rather than fracture union.
• It does not show whether either strategy changes the risk of nonunion, secondary displacement or later surgery.
• The results apply to closed fractures selected for non-operative treatment and should not be extrapolated to fractures that require surgical fixation.

What could this change in practice?

The study does not introduce a new technology. Instead, it clarifies a simple element of everyday fracture care. For a patient with a closed humeral shaft fracture being treated non-operatively, it is reasonable to accept an initial period of splinting and move to functional bracing once swelling decreases rather than treating immediate Sarmiento bracing as a goal in itself.

What does the study not prove?

It does not prove that coaptation splinting leads to better union or better function after several months. It also does not determine which humeral shaft fractures should be treated operatively rather than non-operatively. This is a study of the initial phase of non-operative treatment, not of the entire management strategy.

Back to: Trauma — Literature for Clinicians

Source

Kingery MT, Ganta A, Egol KA. Comparison of Early Bracing Versus Splinting for the Initial Treatment of Acute Humeral Shaft Fractures (The COBRAS Trial). Journal of Orthopaedic Trauma. 2026;40(8):408–415. DOI: 10.1097/BOT.0000000000003196. PubMed: PMID 42466769.

Content prepared by: Tomasz Poboży - orthopaedic surgeon.

Polish version

A randomised trial with a mean follow-up of 10.8 years found no advantage to preserving the first tarsometatarsal joint (TMT1) over primary arthrodesis in selected patients with unstable Lisfranc injury. This is important because it is easy to assume that preserving a mobile joint must lead to better long-term function. In this specific patient group, that advantage was not demonstrated.

Why is this study clinically interesting?

Lisfranc injuries have long generated debate about when joint preservation should be prioritised and when primary arthrodesis is the more reasonable strategy. Most published studies have relatively short follow-up. The report by Grün and colleagues, published online on 9 July 2026 in The Journal of Bone and Joint Surgery, is a long-term continuation of an earlier randomised trial and provides a rare comparison of the two approaches more than a decade later.

Study question and methodology

The original trial enrolled 48 adults with unstable Lisfranc injuries involving the three medial tarsometatarsal joints. An important detail is that patients with a fracture involving TMT1 were not included.

• One group underwent primary arthrodesis of TMT1, TMT2 and TMT3.
• In the other group, TMT1 was preserved and temporarily stabilised with a dorsal bridge plate, while TMT2 and TMT3 were also fused.

After a mean of 10.8 years, 41 of the 48 originally randomised patients—85% of the cohort—were assessed. Outcomes included the AOFAS midfoot score, Manchester–Oxford Foot Questionnaire (MOxFQ), pain, SF-36, return to activity, use of orthoses or inserts, and degenerative changes on computed tomography.

Results

At approximately 10 years, there was no significant difference in AOFAS scores: 89.3 points after primary TMT1 arthrodesis versus 92.6 points after temporary bridge plating (p = 0.66). No significant differences were found in MOxFQ, pain, SF-36, return to activity or the need for orthotic support.

The frequency of degenerative changes in adjacent joints was similar. After TMT1 preservation, degenerative changes within TMT1 itself were common, but only one patient subsequently required arthrodesis. The bridge-plating strategy, however, involved planned removal of the fixation hardware, meaning an additional procedure that was not required after primary arthrodesis.

Clinical interpretation for an orthopaedic surgeon

The practical conclusion is not that every patient should undergo arthrodesis. The study shows something narrower and more useful: in this selected group, preserving TMT1 did not translate into a detectable functional advantage after more than 10 years. At the same time, primary fusion of the first tarsometatarsal joint did not lead to poorer function or a higher frequency of degenerative change in adjacent joints.

Imaging findings should also be separated from symptoms. Degenerative change in a preserved TMT1 was common on imaging, but this did not automatically lead to another operation; only one patient required secondary arthrodesis.

Limitations

• The study was randomised but small: 48 patients were enrolled initially and 41 were assessed at approximately 10 years.
• A non-significant statistical difference is not formal proof that the procedures are equivalent; the sample may have been too small to detect subtle differences or uncommon complications.
• TMT2 and TMT3 were fused in both groups. The findings therefore cannot be directly extrapolated to a comparison between completely joint-preserving fixation and arthrodesis of the entire medial Lisfranc complex.
• Fractures involving TMT1 were excluded, so the conclusions apply to a defined injury pattern.
• The AOFAS score is an older measure combining patient-reported elements with clinical examination. It is therefore useful that the investigators also included MOxFQ, pain and SF-36.

What may change in practice—and what the study does not prove

The study strengthens the argument that primary TMT1 arthrodesis in an appropriately selected patient with an unstable Lisfranc injury does not necessarily produce a worse long-term result simply because motion at that joint is sacrificed. Conversely, preserving TMT1 can also produce a good result despite frequent degenerative changes on imaging.

The study does not prove that arthrodesis is universally superior, does not resolve every type of Lisfranc injury and does not allow every fixation technique to be treated as equivalent to temporary dorsal bridge plating. The decision must still account for the injury pattern, condition of the articular surfaces, ability to restore the Lisfranc complex anatomically, patient activity and the consequences of a possible additional procedure.

Back to: Trauma — Literature for Clinicians

References

Grün W, Poulsen M, Riiser MØ, Hvaal K, Husebye EE, Stødle AH. A Randomized Controlled Trial of Primary Arthrodesis Versus Temporary Bridge Plating in Unstable Lisfranc Injuries: A Concise Follow-up, at 10 Years, of a Previous Report. J Bone Joint Surg Am. Published online July 9, 2026. DOI: 10.2106/JBJS.26.00020. PubMed: PMID 42424434.

Original trial: Stødle AH, Hvaal KH, Brøgger HM, Madsen JE, Husebye EE. Temporary Bridge Plating vs Primary Arthrodesis of the First Tarsometatarsal Joint in Lisfranc Injuries: Randomized Controlled Trial. Foot Ankle Int. 2020;41(8). DOI: 10.1177/1071100720925815.

Content prepared by: Tomasz Poboży - orthopaedic surgeon.

Polish version

Fracture nonunion is usually analysed in terms of mechanical stability, blood supply, infection, bone quality and systemic patient factors. A study published on 2 September 2026 in Science Translational Medicine adds a compelling immunological component: sustained neutrophil extracellular traps (NETs) may actively impair bone regeneration.

Why is this study clinically interesting?

NETs are structures released by neutrophils in response to tissue injury and infection. During acute trauma they are part of the normal inflammatory response. The study suggests, however, that the important issue is not simply whether NETs appear at a fracture site, but how long they persist. A short-lived response may form part of physiological healing, whereas prolonged activation may become detrimental.

The study is particularly interesting because it goes beyond a simple association. It combines observations in humans with mechanistic experiments in animal models and identifies a specific biological pathway linking persistent inflammation with impaired vascularisation of the fracture callus.

Study question and methodology

The investigators asked whether excessive and sustained NET formation could be causally related to fracture nonunion and whether suppressing this process might improve bone regeneration.

The study used mouse models with genetic alterations affecting NET formation or clearance, pharmacological inhibition of selected components of the pathway, and assessment of biomarkers in patients with normally healing fractures and in patients with nonunion. Particular attention was paid to type H vessels, which are involved in coupling angiogenesis with osteogenesis within the fracture callus.

Key findings

Bone injury produced a rapid increase in NET formation in both mice and humans. In normal healing, however, this response was transient.

• Genetically impaired NET clearance caused by combined deficiency of DNase1 and DNase1-like-3 initially favoured callus mineralisation, but later led to persistent NETs, impaired regeneration and more frequent nonunion.
• Additional deletion of PAD4, an enzyme important for NET formation, improved bone regeneration and reduced nonunion in the experimental model.
• Persistent NETs increased activation of the cGAS-STING pathway and reduced formation of type H vessels within the callus.
• Pharmacological inhibition of cGAS-STING restored type H vessel formation and improved healing in the relevant animal model.
• In patients with normal healing, serum NET markers declined over time. In nonunion, they remained elevated and were accompanied by greater accumulation of NETs and STING within the callus.
• In preclinical models, both the PAD4 inhibitor GSK484 and dornase alfa, a recombinant DNase I that enhances NET clearance, accelerated bone repair and reduced nonunion.

Clinical interpretation for an orthopaedic surgeon

One of the most important messages is that the inflammatory response after fracture is strongly time-dependent. The conclusion is not simply that “less inflammation means better healing”. Early neutrophil activity is part of normal fracture repair; the problem appears when mechanisms that should resolve that response fail.

A second important finding is the proposed connection between persistent immune activation and impaired angiogenesis. Type H vessels are associated with areas of active osteogenesis, so their loss provides a biologically plausible explanation for how abnormally prolonged neutrophil activation could translate into poorer bone regeneration.

For trauma surgeons, the work is also interesting because it may eventually support biological risk stratification. At present, nonunion risk is assessed mainly from the injury pattern, mechanical stability, soft-tissue condition, vascularity, infection and patient-related factors. Markers of sustained NET activity might one day complement these variables, but clinical use remains a distant prospect.

Limitations

The principal limitation is fundamental: the therapeutic interventions were evaluated preclinically. The human data show an association between sustained NET markers and nonunion, but they do not provide randomised evidence that inhibiting NETs in patients improves fracture healing.

Genetic models with impaired NET clearance deliberately create an exaggerated biological phenotype and may not reflect the most common causes of nonunion in routine trauma practice. The optimal timing, dose and duration of any therapy directed against NETs are also unknown. This matters because suppressing a normal inflammatory response too early or too strongly could theoretically have the opposite effect.

It is also unclear whether the mechanism will have similar importance across different fracture settings, including high-energy injuries, open fractures, infected fractures, osteoporotic bone or substantial mechanical instability.

What may change in practice—and what the study does not prove

At present, this study does not justify using dornase alfa, PAD4 inhibitors or cGAS-STING inhibitors to treat fracture nonunion. There are no clinical data defining efficacy, safety or a treatment regimen for orthopaedic patients.

The work does, however, change the biological framework in which nonunion can be considered. It describes a coherent sequence: sustained NETs → cGAS-STING activation → impaired type H vessel formation → poorer bone regeneration. If this is confirmed prospectively in humans, it could lead both to a new risk biomarker and to a new class of adjunctive therapies aimed at fracture healing.

A particularly valuable next step would be clinical research testing whether NET biomarker profiles can distinguish, early in the course of healing, fractures likely to unite normally from those at risk of nonunion. Controlled interventional trials would logically come later.

Back to: Trauma — Literature for Clinicians

Reference

Xie W, Albertsen L-C, Eis-Janzyk G, et al. Sustained neutrophil extracellular traps cause fracture nonunion. Science Translational Medicine. 2026;18(865):eadx9869. DOI: 10.1126/scitranslmed.adx9869. PubMed PMID: 42685150.

Content prepared by: Tomasz Poboży - orthopaedic surgeon.

Polish version

Adding 10 mg of morphine to an intraosseous injection during primary total knee arthroplasty did not reduce pain or opioid consumption during the first 14 days after surgery. This is the practical finding of a double-blind randomised trial published in the 19 August 2026 issue of The Journal of Bone and Joint Surgery; the paper had been available online since 21 April 2026.

Why is this study clinically interesting?

Multimodal analgesic protocols are increasingly used in total knee arthroplasty to control pain while limiting opioid exposure. One approach is intraosseous drug delivery, intended to produce a high local concentration in the operative region. Intraosseous administration has established uses for selected antibiotics, but that does not mean every drug given by this route will provide a clinically meaningful additional effect.

The study by Pekas and colleagues addressed a specific question: does adding morphine to an intraosseous injection improve pain control after TKA beyond a contemporary multimodal analgesic protocol?

Study question and methodology

The prospective randomised trial enrolled 100 patients undergoing elective primary total knee arthroplasty and used double blinding. The final analysis included 88 patients; mean age was 69.1 years.

All patients received spinal anaesthesia with intravenous sedation and a surgeon-performed adductor canal block. Both groups received 500 mg of vancomycin in 110 mL of normal saline by the intraosseous route. In the intervention group, 10 mg of morphine was added to that solution; the control group received an otherwise identical injection without morphine.

For 14 days after surgery, patients repeatedly reported pain intensity, opioid use converted to morphine equivalents, and nausea and vomiting. Immediate postoperative pain and opioid requirements, as well as patient-reported outcomes, were also assessed.

Results

There were no between-group differences in daily pain at any time point during the 14-day follow-up (p = 0.969). Daily opioid consumption also did not differ (p = 0.377), and total and weekly opioid consumption remained similar (p ≥ 0.878).

In practical terms, within the protocol studied, 10 mg of intraosseous morphine provided no measurable additional analgesic benefit.

Clinical interpretation for an orthopaedic surgeon

The negative result is clinically relevant. Adding another component to a perioperative pathway is justified only if the benefit outweighs the additional cost, risk and complexity. In this trial, no such benefit was demonstrated for intraosseous morphine.

The finding should nevertheless be interpreted in the context of effective baseline analgesia. Patients received spinal anaesthesia, intravenous sedation and an adductor canal block. With such a regimen, the scope for demonstrating a small incremental benefit may have been limited. That does not change the principal practical conclusion: this study provides no basis for routinely adding morphine to the intraosseous injection in TKA.

Limitations

• The final analysis included 88 of the 100 randomised patients.
• Pain and opioid use were followed for only the first 14 postoperative days.
• The trial evaluated one specific 10 mg morphine dose within one perioperative protocol; it does not answer questions about other doses or drug combinations.
• The study population was relatively demographically homogeneous, which may limit generalisability.
• Both groups received intraosseous vancomycin, so the study isolates the effect of added morphine rather than comparing the intraosseous route itself with other routes of drug administration.

What may change in practice—and what the study does not prove

The results argue against routinely including intraosseous morphine as an additional analgesic component in primary TKA when a similar multimodal protocol is already being used. This supports simplifying pathways in which an added drug does not provide a demonstrable benefit.

The trial does not show that the intraosseous route is ineffective in general. It did not assess the efficacy of intraosseous antibiotics or other drugs, and it does not resolve the value of other local analgesic strategies. Nor can it be used to conclude that morphine has no role in pain management after TKA—the intervention tested was specifically intraoperative intraosseous administration at the stated dose.

Back to: Joint Replacement — Literature for Clinicians

Reference

Pekas DR, Adrados M, Lee MM, et al. Effectiveness of Intraosseous Morphine for Pain Control in Total Knee Arthroplasty: A Double-Blinded, Randomized Trial. J Bone Joint Surg Am. 2026;108(16):1249-1255. DOI: 10.2106/JBJS.25.01037. PubMed: PMID 42013190.

Content prepared by: Tomasz Poboży - orthopaedic surgeon.

Polish version

Does manipulation for frozen shoulder have to be performed under general anaesthesia? A new randomised trial suggests that, in carefully selected adults with primary frozen shoulder persisting despite at least 6 months of non-operative treatment, an outpatient procedure under local anaesthesia may be feasible. The study did not demonstrate superiority in its primary endpoint, so the findings are better viewed as evidence of feasibility than as proof of a new standard of treatment.

Why is this study interesting?

Manipulation under anaesthesia is one option for refractory frozen shoulder, but the conventional approach uses an operating theatre and general anaesthesia. The investigators asked whether a similar short-term result could be achieved in an outpatient setting using an ultrasound-guided WALANT protocol—wide-awake local anaesthesia without general anaesthesia.

Question and methodology

This single-centre prospective randomised trial enrolled 72 adults with primary frozen shoulder that persisted despite at least 6 months of conservative treatment. Participants were allocated 1:1 to outpatient manipulation using an ultrasound-guided WALANT protocol or to manipulation under general anaesthesia in the operating theatre.

The primary endpoint was forward shoulder flexion at 4 weeks. The investigators also assessed other ranges of motion, pain, UCLA Shoulder, DASH and Constant–Murley scores, pre-procedure anxiety and early complications. Six participants in the general-anaesthesia group did not complete follow-up; complete-case analysis therefore included 36 patients in the WALANT group and 30 in the general-anaesthesia group.

Results

At 4 weeks, mean forward flexion was 170.6° in the WALANT group and 159.6° after manipulation under general anaesthesia. The between-group difference was 11.0°, but the 95% confidence interval crossed no difference (−0.8° to 22.7°; p=0.068). The prespecified primary endpoint therefore did not show a statistically significant advantage for either strategy.

In secondary analyses, pain, selected motion measures and the Constant–Murley score transiently favoured WALANT during the earlier weeks, but these differences were no longer present at week 4. UCLA Shoulder and DASH scores did not differ between groups. Pre-procedure anxiety was lower in the outpatient group. During the short follow-up, the investigators reported no procedure-related complications, infections, early recurrences or repeat interventions.

Clinical interpretation for an orthopaedic surgeon

The clinically useful message is not that “WALANT is better”. Rather, the trial suggests that, in a specialised setting, ultrasound-guided outpatient manipulation can be technically feasible and can produce short-term outcomes broadly similar to manipulation under general anaesthesia. A potential organisational advantage would be reduced reliance on operating-theatre and anaesthetic resources, but the study did not formally evaluate costs or health-system efficiency.

Patient selection remains central in frozen shoulder. Manipulation is not a first-line treatment for every patient with shoulder stiffness. The decision must take into account symptom duration, response to previous treatment, the clinical diagnosis and the risk of injury during manipulation.

Limitations

• The study was single-centre and included a relatively small number of patients.
• The main follow-up lasted only 4 weeks, so durability of benefit, recurrence and late complications cannot be assessed.
• Six patients were lost to follow-up, all from the general-anaesthesia group, and the reported analysis was based on complete cases.
• The trial was registered retrospectively, an important methodological limitation.
• The prespecified primary endpoint was not statistically significant; favourable differences at earlier time points were secondary findings.

What might this change in practice — and what does the study not prove?

The study supports further evaluation of outpatient frozen-shoulder manipulation without general anaesthesia and may be particularly relevant to centres combining shoulder treatment with interventional ultrasound. It does not establish statistical equivalence between the two methods, does not show a WALANT advantage at 4 weeks and is far too small to assess rare complications reliably.

Back to: Research Updates, Literature and Evidence-Based Medicine

See also: Outpatient and Ultrasound-Guided Procedures

References

1. Konarski W, Poboży T, Hordowicz M, Poboży K, Domańska J. Current concepts of natural course and in management of frozen shoulder: A clinical overview. Orthopedic Reviews. 2020;12(4):8832. DOI: 10.4081/or.2020.8832.
2. Kurtulus T, Bacaksiz T, Maden M, et al. WALANT technique versus general anesthesia in frozen shoulder manipulation: a prospective randomized controlled trial. Journal of Orthopaedic Surgery and Research. Published 6 September 2026. DOI: 10.1186/s13018-026-07169-5.

Educational material for healthcare professionals. It does not constitute an individual therapeutic recommendation.

Content prepared by: Tomasz Poboży - orthopaedic surgeon.