Kategoria: Updates for Doctors (EN)

Bioresorbable screws may be a feasible alternative to titanium screws for first metatarsophalangeal joint arthrodesis, but the new trial is too small to establish equivalence or superiority. This distinction matters: failure to detect a statistically significant difference in a pilot sample does not mean that the two methods produce identical outcomes.

Polish version · For Doctors and Physiotherapists

Why is this study interesting?

First MTP joint fusion is used for advanced hallux rigidus, severe deformity and selected revision procedures. Fixation has to maintain alignment until union occurs. A bioresorbable implant could theoretically avoid permanent metal hardware, but it still has to provide adequate stability and degrade without clinically important tissue reactions.

Question and methodology

Partio and colleagues conducted a prospective pilot randomised trial in two Finnish hospitals. Thirty patients were allocated 1:1 to fixation with cannulated bioresorbable screws or titanium screws. Joint surfaces were prepared using a cup-and-cone technique, and full weight bearing in a postoperative shoe was allowed after six weeks.

Union, complications and functional outcomes were assessed at 3, 6, 12 and 24 months. The trial was registered as NCT03133039.

Results

• Both groups showed substantial functional improvement, with no demonstrated between-group difference.
• At 24 months, mean AOFAS scores were 89.1 in the titanium group and 90.0 in the bioresorbable group (p = 0.23).
• The overall union rate was 90%.
• Nonunion occurred in one patient with titanium screws and two patients with bioresorbable screws; all three required revision surgery.
• No clinical or radiographic adverse events were attributed to degradation of the bioresorbable screws.
• No patient in either group underwent hardware removal.

Clinical interpretation for an orthopaedic surgeon

The main value of this study is as an early signal of feasibility and safety. At two years there were no reported reactions related to implant degradation, and functional outcomes were good in both groups. That is sufficient to justify a larger study, particularly if bioresorbable fixation is expected to reduce problems associated with retained metalwork.

It does not, however, demonstrate a practical advantage. Hardware was not removed in the titanium group either, so the proposed benefit of avoiding a later removal procedure was not shown here. Three nonunions among 30 participants are also far too few events for a reliable comparison of uncommon complications.

Key limitations

• This was a pilot study with only 30 patients.
• It was not designed as an equivalence trial with a prespecified equivalence margin.
• The small sample provides very limited power for nonunion, implant reactions and reoperations.
• The AOFAS score has methodological limitations, and the bioresorbable group was reported with a mean of 90.0 and a standard deviation of 0.0 at 24 months, making the score distribution difficult to interpret.
• Two years of follow-up does not answer every question about late material degradation.

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

For now, these results support further adequately powered trials of bioresorbable screws. They do not justify treating the technique as standard, safer, or equivalent to titanium fixation. Choice of fixation should still account for construct biomechanics, bone quality, deformity, likelihood of union and the surgeon’s experience.

Source

Partio N, Mattila V, Waris V, Mäenpää H. Bioabsorbable vs. titanium screws in first metatarsophalangeal joint arthrodesis: a pilot randomized controlled trial with 2-year follow-up. Archives of Orthopaedic and Trauma Surgery. Published 9 September 2026;146(1):329. DOI: 10.1007/s00402-026-06501-2. PMID: 42714616.

Content prepared by: Tomasz Poboży - orthopaedic surgeon.

Interactive digital rehabilitation may improve short-term function in some people with knee osteoarthritis and after total knee arthroplasty, but the available evidence is too heterogeneous to treat it as a proven replacement for standard rehabilitation. A recent systematic review provides an encouraging signal while also showing how very different interventions can be grouped under the broad label of “virtual reality”.

Why is this study interesting?

Digital tools are increasingly used in rehabilitation, ranging from screen-based exercises with real-time feedback to immersive virtual-reality headsets. They may make home exercise easier to deliver, increase engagement and provide therapists with information about how tasks are performed. The clinically relevant question, however, is not simply whether patients enjoy the technology, but whether it improves pain and function more than conventional exercise or education.

The review by Kalinowski and colleagues, published on 8 September 2026 in the Journal of Medical Internet Research, evaluated extended-reality (XR) interventions in knee osteoarthritis and recovery after total knee arthroplasty. The authors used a broad XR definition that included both immersive virtual reality and interactive screen-based digital rehabilitation. That distinction is central to interpreting the findings.

Question and methodology

The authors performed a systematic review and meta-analyses using a prospectively registered PROSPERO protocol. Five databases were searched, with the search updated through December 2025. Twenty-one studies were included: 12 addressing non-operative management of knee osteoarthritis and 9 evaluating rehabilitation after total knee arthroplasty. Individual study sample sizes ranged from 36 to 306 participants.

The interventions varied substantially. Some used games or movement-tracking exercises displayed on a screen, others used more elaborate feedback systems, and only 6 of the 21 studies actually evaluated immersive virtual reality. No study directly compared immersive headsets with non-immersive digital rehabilitation.

Main findings

In knee osteoarthritis, interactive digital rehabilitation was associated with better function than control interventions. The pooled standardised effect was −0.59, but the 95% prediction interval included both possible benefit and no effect. Certainty of evidence was rated low. For pain at 6–8 weeks, the effect was small and near the threshold of statistical significance; the prediction interval again included no benefit.

After total knee arthroplasty, individual studies suggested possible functional improvement, but at 3–4 months there was no clear advantage for pain. The pooled effect was −0.12, with a wide confidence interval from −0.75 to 0.52. Most studies followed patients for only a short period, generally no longer than three months.

Clinical interpretation for an orthopaedic surgeon

The most reasonable interpretation is that digital rehabilitation can be a useful way to deliver and support exercise, particularly when it improves access, motivation or adherence. The review does not show that XR technology itself is responsible for the benefit. Better outcomes may instead reflect more repetitions, immediate feedback, gamification, greater contact with the rehabilitation team or simply more consistent participation.

The review also does not justify the claim that immersive virtual-reality headsets are more effective than a simpler app or screen-based exercise programme. Intervention categories were inconsistent, and no direct comparison between those approaches was available.

Important limitations

• Most trials were small, and 16 of 21 studies had follow-up shorter than 6 months.
• The XR label covered technologically and therapeutically different programmes.
• Some studies had risk of bias related to missing data and deviations from the planned intervention.
• Only 5 studies reported adverse events, so safety cannot be assessed with confidence.
• The review included English-language publications only, and individual meta-analyses contained relatively few studies.

What could this change in practice?

The review supports using well-designed digital tools as a possible adjunct to rehabilitation when they solve a specific problem—for example, limited access to therapy, difficulty monitoring exercises or poor adherence. Selection should be based on the functional goal, the patient’s abilities, supervision and safety rather than on the novelty of the technology.

The study does not show that XR improves long-term outcomes, reduces revision risk after arthroplasty or can replace individual physiotherapy assessment. It also does not establish superiority of immersive virtual reality over simpler forms of digital rehabilitation. Larger studies with clearly described interventions, longer follow-up and systematic adverse-event reporting are needed.

Source

Kalinowski C, et al. Extended Reality Interventions for Osteoarthritis of the Knee and Recovery After Total Knee Arthroplasty: Systematic Review and Meta-Analyses. J Med Internet Res. 2026;28:e84753. DOI: 10.2196/84753. PubMed · Full text.

Polish version · Research Updates, Literature and Evidence-Based Medicine

Content prepared by: Tomasz Poboży - orthopaedic surgeon.

Colchicine changed several inflammatory biomarkers, but after three months it did not reduce pain, improve function or reduce knee effusion more than placebo. This is an important negative clinical result: a favourable laboratory signal does not necessarily translate into a benefit that patients can feel.

Why is this study interesting?

Synovitis and innate immune activation are involved in the pathophysiology of knee osteoarthritis, making colchicine a plausible drug-repurposing candidate. It affects several inflammatory pathways and is well established in gout and pericarditis, yet previous osteoarthritis studies have been inconsistent.

The CLOAK trial, published online on 23 August 2026 in Seminars in Arthritis and Rheumatism, assessed not only pain and function but also ultrasound-detected effusion and a panel of biological markers. This allowed the investigators to compare clinical outcomes with biochemical effects.

Question and methods

The trial included 120 people with symptomatic Kellgren–Lawrence grade 2 or 3 knee osteoarthritis. Participants were randomly assigned to daily colchicine or placebo for three months. The study used double blinding, and concomitant non-steroidal anti-inflammatory drugs were not permitted.

The primary endpoint was the between-group difference in change in visual analogue scale pain in the index knee. Secondary outcomes included KOOS, ultrasound-measured effusion depth, paracetamol use and changes in markers associated with inflammation and osteoarthritis.

Results

At three months, colchicine was not superior to placebo for pain reduction. There was also no advantage in KOOS or ultrasound-detected effusion. No clinically meaningful difference emerged in subgroup analyses of participants with greater pain, more advanced radiographic disease, higher hsCRP or higher uric acid.

At the same time, the colchicine group showed favourable changes in some biomarkers. Statistically significant between-group differences included hsCRP and beta-NGF. For several additional markers the authors reported only trends; these signals should not be treated as proof of clinical benefit.

Clinical interpretation for an orthopaedic surgeon

The most important finding is the absence of improvement in outcomes that matter directly to patients. A biomarker shift may indicate biological activity, but it is not enough to justify treatment when pain, function and effusion do not improve.

The study does not support starting colchicine solely to treat symptomatic knee osteoarthritis. For a patient taking colchicine for another recognised indication, these results of course do not imply that the drug should be stopped; the trial addresses its effectiveness as a treatment for knee osteoarthritis.

Limitations

• A three-month follow-up cannot assess long-term effects or structural progression.
• A sample of 120 may miss small treatment effects and uncommon adverse events.
• Exclusion of concomitant NSAID use limits direct generalisation to some routine clinical settings.
• Multiple biomarker analyses increase the chance of false-positive findings.
• Biochemical change remains a surrogate outcome, not evidence of patient benefit.

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

The findings strengthen the case against routine colchicine use for symptomatic knee osteoarthritis outside another accepted indication or a clinical trial. They also reinforce an important hierarchy of evidence: pain, function and safety should carry more weight in treatment decisions than improvement in an isolated biomarker.

The trial does not prove that colchicine has no anti-inflammatory biological effect or that longer treatment could never produce benefit. It does show that there is currently no clinical basis for waiting for such an effect at the expense of better-supported care. Appropriate exercise and loading, education, weight management, analgesic strategies that respect contraindications, and surgery for properly selected patients remain central to practice.

Source

Samuels J, Tse K, Wei D, et al. CoLchicine for Treatment of OsteoArthritis of the Knee (CLOAK): Clinical and biochemical outcomes from a three-month double-blind, placebo-controlled study. Semin Arthritis Rheum. 2026;153066. DOI: 10.1016/j.semarthrit.2026.153066. PubMed.

Polish version · Research Updates, Literature and Evidence-Based Medicine

Content prepared by: Tomasz Poboży - orthopaedic surgeon.

Tai chi and other traditional Chinese exercise programmes have been studied across many musculoskeletal conditions, but a large number of publications does not automatically mean high-certainty evidence. A recent umbrella review brings together 92 meta-analyses and shows where the signal is relatively robust and where substantial uncertainty remains.

Why is this study interesting?

Research on tai chi, qigong, baduanjin and related exercise forms spans osteoarthritis, spinal pain, fibromyalgia, osteoporosis and other conditions. Individual meta-analyses often end on an optimistic note, but they vary in methodological quality, outcomes and comparators. An umbrella review takes a higher-level view by evaluating the existing meta-analyses rather than focusing on a single trial.

The paper was published online on 17 August 2026 in the British Journal of Sports Medicine.

Question and methods

The authors asked how effective traditional Chinese exercise programmes are for musculoskeletal disorders and how certain the supporting evidence is. Databases were searched through 6 March 2026, and systematic reviews with meta-analyses of randomised trials were included.

• 92 meta-analyses were included.
• The review datasets collectively contained data from 1,744 randomised trials.
• A total of 547 intervention-outcome associations were assessed.
• Methodological quality of the reviews was assessed with AMSTAR 2, while certainty of evidence was graded using GRADE.

The figure of 1,744 should not be interpreted as proof that all 1,744 trials were unique. In umbrella reviews, the same primary studies can appear in more than one meta-analysis, particularly when different outcomes or closely related interventions are assessed.

Main results

Sixty-five of the 92 meta-analyses, or 70.7%, were judged to have high methodological quality. The certainty of individual intervention-outcome associations was much more cautious:

• high-certainty evidence was found for 23 of 547 associations, or 4.2%;
• moderate-certainty evidence was found for 179 associations, or 32.7%;
• the remaining findings were supported by low- or very-low-certainty evidence.

Among the high-certainty findings was improvement in physical function in people with knee osteoarthritis. High-certainty evidence was also reported for selected mental-health outcomes in knee osteoarthritis and neck pain, and for sleep quality in fibromyalgia.

This distinction matters. The review does not show that every traditional exercise programme improves every musculoskeletal symptom. It identifies a relatively small group of more reliable findings and a much larger body of evidence that still requires caution.

Clinical interpretation for an orthopaedic surgeon

The most practical finding concerns knee osteoarthritis. Tai chi can reasonably be considered one exercise option for patients who are comfortable with slower training involving weight transfer, postural control and coordination. It should not be presented as a separate treatment that “repairs cartilage”, but as one way of delivering exercise.

Exercise selection should take into account safe standing ability, fall risk, comorbidities, pain severity and the likelihood of regular participation. For some patients, a simpler strengthening programme, lower-load exercise or individually supervised rehabilitation may be a better starting point.

Limitations

• An umbrella review inherits the limitations of the included meta-analyses and primary trials.
• The interventions differed in style, intensity, duration, delivery and instructor experience.
• Control groups ranged from no exercise to other active interventions, so effect sizes do not always answer the same clinical question.
• Subjective outcomes are vulnerable to expectation effects, and full participant blinding is generally not feasible in exercise trials.
• Overlap of primary studies across meta-analyses may make the apparent evidence base look larger than the number of unique trials.

What might this change in practice?

The review supports considering tai chi as a legitimate exercise option rather than dismissing it as an intervention without scientific support. The strongest practical argument concerns physical function in knee osteoarthritis. For other indications, patients should be told when certainty is only moderate, low or very low.

What does the study not prove?

• It does not prove that tai chi is superior to a well-designed strength, aerobic or neuromuscular programme.
• It does not justify one universal exercise programme or dose for all conditions.
• It does not show that every form of traditional exercise is effective for every type of musculoskeletal pain.
• It does not replace assessment of red-flag symptoms or treatment that requires a different clinical approach.

Polish version · Research Updates, Literature and Evidence-Based Medicine

Source

Chen WX et al. Efficacy and evidence certainty of traditional Chinese exercises for musculoskeletal disorders: an umbrella review of meta-analyses. Br J Sports Med. Published online 17 August 2026. DOI: 10.1136/bjsports-2026-111792. PubMed.

Content prepared by: Tomasz Poboży - orthopaedic surgeon.

Preserving both cruciate ligaments and more native bone did not translate into greater everyday physical activity after robotic bi-unicompartmental knee arthroplasty than after total knee arthroplasty. This secondary analysis of a randomised trial used accelerometers for up to two years. The result is clinically interesting, but the small sample does not establish equivalence between the procedures.

Polish version · For Doctors and Physiotherapists

Why is this study interesting?

Bi-unicompartmental knee arthroplasty replaces the medial and lateral tibiofemoral compartments with two separate unicompartmental implants. It preserves the cruciate ligaments and more native structures than conventional total knee arthroplasty. Biomechanically, that is attractive, but the clinically relevant question is whether the patient actually walks more and becomes more active in everyday life.

The study was published on 10 September 2026 in Bone & Joint Research.

Question and methods

This was an analysis of secondary outcomes from the TRUCK randomised trial. Patients were allocated to manual total knee arthroplasty or robotic arm-assisted bi-unicompartmental knee arthroplasty. Sixty-nine participants had usable accelerometer data for this analysis: 38 after total knee arthroplasty and 31 after bi-unicompartmental arthroplasty.

A triaxial accelerometer was worn on the operated thigh for seven days, with at least three valid recording days required for analysis. Measurements were obtained before surgery and at one and two years.

• total activity-related energy expenditure;
• time spent in moderate-to-vigorous physical activity;
• time spent in light activity;
• sedentary time;
• the relationship between objectively measured activity and patient-reported outcome measures.

The trial involved a selected population with disease in the medial and lateral compartments, preserved cruciate and collateral ligaments, no major deformity and no advanced patellofemoral disease.

Results

No statistically significant between-group differences were found in any objectively measured activity parameter at either one or two years. This included total activity energy expenditure, moderate-to-vigorous activity, light activity and sedentary time.

At one year, the bi-unicompartmental group showed numerically greater total activity and somewhat more moderate-to-vigorous movement, but the differences were not statistically significant and were not maintained at two years.

Preoperative activity was a stronger predictor of postoperative activity. Higher baseline energy expenditure and more moderate-to-vigorous activity were associated with greater activity after surgery.

Both groups improved similarly on clinical questionnaires. However, the association between patient-reported outcomes and objectively measured movement was weak and inconsistent. A patient may therefore report substantial improvement in pain and function without becoming more active in daily life.

Clinical interpretation for an orthopaedic surgeon

Tissue preservation and more physiological knee kinematics do not guarantee a change in real-world behaviour. Postoperative activity is also shaped by habits, preoperative activity, body mass, comorbidities, patient goals, environment and rehabilitation.

The study also illustrates why questionnaires and accelerometry should not be treated as interchangeable outcomes. Questionnaires assess symptoms and perceived capability; accelerometers record how much movement the patient actually undertakes. These tools answer different questions.

Limitations

• Physical activity was a secondary outcome, and the trial was not powered specifically to detect differences in these measures.
• The sample was small and the number of valid recordings declined over time. At two years, accelerometer data were available for 24 patients after total knee arthroplasty and 15 after bi-unicompartmental arthroplasty.
• Three valid days of recording were sufficient for analysis and may not capture full week-to-week variability.
• The comparison was robotic bi-unicompartmental arthroplasty versus manual total knee arthroplasty, so the effects of implant strategy and robotic technology cannot be separated.
• The findings apply to carefully selected patients with preserved ligaments and a specific pattern of osteoarthritis.

What might change in practice?

When counselling patients, greater everyday activity should not be promised simply because a procedure preserves the cruciate ligaments and more native knee structures. It may be more useful to define realistic activity goals before surgery and to plan rehabilitation not only around regaining the ability to walk, but also around progressively changing physical-activity behaviour.

The study also supports complementing conventional clinical outcomes with real-world activity measures in trials of techniques that are intended to improve higher-level function.

What does the study not prove?

• It does not establish equivalence between the two operations; absence of statistical significance in a small, underpowered analysis is not evidence of identical effects.
• It does not exclude differences in maximal performance, kinematics, gait speed or other outcomes that this analysis was not designed to assess reliably.
• It does not justify broadening indications for bi-unicompartmental arthroplasty beyond patients who meet the trial criteria.
• It cannot separate the effect of robotic assistance from the effect of preserving ligaments and reducing the extent of bone resection.

Source

Kanakala S et al. Effect of robotic arm-assisted bi-unicompartmental knee arthroplasty versus total knee arthroplasty on objectively measured physical activity: secondary outcomes of a randomized controlled trial. Bone Joint Res. 2026;15(9):1136–1146. DOI: 10.1302/2046-3758.159.BJR-2026-0018.R1. PubMed.

Content prepared by: Tomasz Poboży - orthopaedic surgeon.

The trial is striking because of the size of the WOMAC difference, but it does not yet establish high-intensity laser therapy as standard treatment for knee osteoarthritis. The comparison was with therapeutic ultrasound, not with a sham procedure or exercise alone, which substantially limits what can be concluded.

Why is this study interesting?

Physical modalities are often added to exercise programmes for knee osteoarthritis even though the benefit of individual techniques remains uncertain. A randomised trial from Thailand, published on 8 September 2026, compared high-intensity laser therapy (HILT) with therapeutic ultrasound. The reported between-group difference was much larger than the authors’ prespecified estimate of a minimally important change, making the result clinically interesting—but also worth examining carefully in light of the study design.

Question and methods

The single-centre trial enrolled 66 adults with knee osteoarthritis. Participants were randomised 1:1 to HILT or therapeutic ultrasound. Both modalities were delivered as an adjunct to exercise twice weekly for six weeks. The outcome assessor was blinded to treatment allocation.

The primary endpoint was change in the total WOMAC score 12 weeks after completion of treatment. The authors used 10 points as an approximate threshold for a minimally important change. Analysis followed the intention-to-treat principle. The trial was registered retrospectively, after recruitment had begun.

Results

Mean improvement in WOMAC was 40.0 points in the HILT group and 8.2 points in the therapeutic-ultrasound group. The adjusted between-group difference was −27.7 points (95% CI −39.0 to −16.5; p<0.001). Secondary analyses also favoured HILT for pain, function and quality of life, with the reported effect persisting up to 12 weeks after treatment ended.

Clinical interpretation for an orthopaedic surgeon

The magnitude of the difference is difficult to ignore, but the trial answers a narrower question than the headline might suggest: HILT plus exercise performed better than therapeutic ultrasound plus exercise. It does not show how much of the observed difference was due specifically to the laser, and how much may have reflected differences in expectations, treatment experience or a relatively weak response in the comparison group.

It is also important that WOMAC and several other key outcomes were patient-reported. Blinding the outcome assessor does not remove the risk introduced when participants and therapists are likely to recognise which physical modality is being delivered.

Limitations

• small, single-centre sample;
• short follow-up;
• no sham-treatment group and no exercise-only group;
• retrospective trial registration;
• the minimally important change threshold was used as an approximation rather than a value validated specifically for this population and comparison;
• secondary outcomes were exploratory;
• a large effect from one centre requires independent replication.

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

The study provides a credible preliminary signal that HILT may be a useful adjunct to exercise for some patients with knee osteoarthritis. It justifies a larger trial with sham treatment, longer follow-up and an economic evaluation. It does not prove superiority over exercise alone, a disease-modifying effect, or that routine funding of HILT is justified. It also does not identify which patients, if any, are most likely to benefit.

Source

Ninlerd C, Benjakulsap N, Amornsupak C, et al. Efficacy of high-intensity laser therapy versus ultrasound therapy in patients with knee osteoarthritis: a randomized controlled trial. Lasers Med Sci. 2026;41(1):219. DOI: 10.1007/s10103-026-05019-8. PubMed.

---

Polish version · For Doctors and Physiotherapists

Content prepared by: Tomasz Poboży - orthopaedic surgeon.

Does a third day of prophylactic antibiotics reduce infection after severe open tibial fractures compared with two days? The PROACT trial did not show such a benefit. It also does not establish that the two regimens are equivalent, because the confidence intervals remain wide.

Why is this study clinically interesting?

The duration of antibiotic prophylaxis in open fractures remains debated. A regimen that is too short raises concern about infection, whereas unnecessary extension increases exposure to adverse effects and antimicrobial selection pressure. In Gustilo-Anderson IIIA and IIIB open tibial fractures, this question is particularly relevant because the soft-tissue injury is extensive and the baseline risk of infection is substantial.

PROACT was a pragmatic randomised trial conducted in a centre with integrated orthopaedic and reconstructive care. It did not compare antibiotics with no antibiotics. Instead, it tested two closely related durations: 48 versus 72 hours.

Question and methods

Adults with Gustilo-Anderson type IIIA or IIIB open tibial fractures were eligible. Of 241 patients assessed for initial eligibility, 204 were randomised 1:1 to either two or three days of intravenous cefuroxime and amikacin prophylaxis. Metronidazole was added when contamination was related to agricultural exposure.

The two-day group included 104 patients and the three-day group 100. All were analysed according to their original allocation. Outcome assessors were unaware of treatment allocation, whereas the nature of the intervention did not allow full blinding of the treating team. Orthoplastic care was standardised.

The primary endpoint was surgical-site infection within 90 days. Secondary outcomes included unplanned reoperation within one year and secondary amputation.

Results

• Infection within 90 days occurred in 15 of 104 patients after the two-day regimen (14%) and in 18 of 100 after the three-day regimen (18%).
• The risk difference was −3.6 percentage points, with a 95% confidence interval from −14.0 to 7.0 percentage points; p=0.60.
• Time-to-infection analysis likewise showed no statistically significant difference: HR 1.22; 95% CI 0.60–2.46.
• Unplanned reoperation within 12 months occurred in 24 of 104 patients in the two-day group (23%) and 19 of 100 in the three-day group (19%).
• Each group had one secondary amputation because of failure of soft-tissue coverage.

Clinical interpretation for an orthopaedic surgeon

The most defensible conclusion is that in this centre, extending prophylaxis from two to three days was not shown to reduce infections. This is an important signal against routinely adding another day simply “to be safe”.

However, a non-significant result must not be converted into proof of equivalence. The confidence interval remains compatible with clinically relevant effects in either direction. The trial also answers a narrow question: one additional day within a specific antibiotic protocol and within a well-organised orthoplastic pathway.

Limitations

• The study was single-centre and conducted in India, so care pathways, infection epidemiology and antimicrobial susceptibility may differ in other healthcare systems.
• The sample was too small to exclude clinically important differences with high precision.
• Only 48 and 72 hours of one antibiotic regimen were compared; the findings cannot be extrapolated to other agents or very short prophylaxis.
• Only IIIA and IIIB fractures were included. The trial does not resolve management in IIIC injuries, other anatomical sites, delayed presentation or unusual contamination.
• It does not replace decisions about rapid administration of the first antibiotic dose, radical debridement, fracture stabilisation and timely soft-tissue coverage.

What could this change in practice?

PROACT supports re-examining protocols in which a third day of prophylaxis is added automatically despite adequate debridement and standardised care. It does not justify independently shortening treatment outside local protocols. A larger multicentre trial, ideally designed to assess non-inferiority or equivalence of the shorter regimen, is still needed.

The study does not show that antibiotics can compensate for inadequate surgical debridement, nor that prophylactic dosing is sufficient once infection has developed. It also does not answer whether an even shorter regimen would be safe.

Source

Devendra A, Imran A, Irfan Pc M, et al. Two-day versus three-day systemic antibiotic prophylaxis for Gustilo-Anderson type IIIA and IIIB open tibial fractures (PROACT): a randomised controlled trial. Lancet Regional Health – Southeast Asia. 2026;53:100830. DOI: 10.1016/j.lansea.2026.100830. PubMed.

Polish version · More research updates for doctors

Content prepared by: Tomasz Poboży - orthopaedic surgeon.

Robotic assistance can improve planning and execution of component positioning, but it does not answer whether a total knee arthroplasty should be cemented or cementless. A recent meta-analysis found no reliable clinical advantage for either fixation method in revision risk, manipulation for stiffness, operative time or length of stay. Certainty of evidence was very low for all pooled outcomes.

Why is this study interesting?

Use of robotic-assisted total knee arthroplasty is increasing, as is interest in cementless fixation. The two technologies are often presented as complementary: robotic assistance may improve bone preparation and implant positioning, while biological ingrowth may provide durable fixation. The clinically relevant question, however, is whether this combination actually reduces revision risk or improves recovery.

Research question and methods

The authors performed a prospectively registered systematic review and meta-analysis comparing cementless and cemented fixation in robotic-assisted total knee arthroplasty. PubMed, Embase, Web of Science and CENTRAL were searched through 29 July 2026. Ten studies, reported in 11 publications, were included. The evidence was observational rather than randomized.

A random-effects model with Hartung–Knapp adjustment was used. Risk of bias was assessed with ROBINS-I and certainty of evidence with GRADE. Adjusted registry and institutional estimates were considered separately from simple unadjusted event rates.

Results

• All-cause revision: RR 0.63; 95% CI 0.09–4.53; three studies. The confidence interval is compatible with both substantial benefit and clinically important harm.
• Manipulation under anaesthesia for stiffness: RR 1.03; 95% CI 0.09–11.74; three studies. The estimate is extremely imprecise.
• Operative time: mean difference −12.91 minutes; 95% CI −36.38 to 10.55. Heterogeneity was very high (I² 94.4%).
• Length of stay: mean difference −0.24 days; 95% CI −3.67 to 3.19.

Adjusted analyses of implant survival were inconsistent. Certainty of evidence for all pooled outcomes was rated very low.

Clinical interpretation for an orthopaedic surgeon

The most defensible conclusion is a negative one: current evidence does not allow fixation to be selected on the expectation of superior implant survival or a lower risk of stiffness. Absence of a statistically significant difference is not evidence of equivalence. Event numbers were small, follow-up and patient characteristics varied, and the confidence intervals remained wide.

In practice, fixation choice should still take account of bone quality, age and activity level, implant design, surgeon and team experience, and credible long-term survivorship data. Robotic assistance does not remove these variables.

Limitations

• no randomized trials and potential selection of younger or healthier patients for cementless fixation;
• few clinically important events;
• different robotic systems, implants, perioperative pathways and follow-up durations;
• very high heterogeneity for operative time;
• insufficient evidence for firm conclusions about long-term osseointegration and implant survival.

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

The study is useful as a reminder that new technology should not automatically be equated with a better clinical result. It supports individualized fixation choice and the need for registries and comparative studies with longer follow-up.

The meta-analysis does not prove that the two fixation methods are equivalent, that cementless fixation shortens surgery, or that robotic assistance improves implant survival. It evaluated fixation within robotic TKA, not robotic versus conventional TKA.

Source

Lu X, et al. Cementless versus cemented fixation in robotic-assisted total knee arthroplasty: a systematic review and meta-analysis. Journal of Robotic Surgery. Published online 10 September 2026. DOI: 10.1007/s11701-026-03858-9. PubMed PMID 42720876.

PROSPERO: CRD420261446055.

For Doctors and Physiotherapists · Polish version

Content prepared by: Tomasz Poboży - orthopaedic surgeon.