This Week in Physical Medicine & Rehabilitation — Sep 10, 2026

Welcome to This Week in Physical Medicine and Rehabilitation. This week we're covering 10 notable papers spanning exercise prescription and mobility screening in ageing adults, stroke rehabilitation from dose to orthotics to dysphagia, spinal cord injury prognosis and research priorities, and a set of studies asking a hard question about what we can safely stop doing or move online. Let's dive in.

We start with the biggest paper of the week, in the BMJ, where Lu and colleagues pooled 124 randomised trials and more than eighteen thousand adults aged forty and over in a hierarchical Bayesian network meta-analysis of exercise for bone health [1]. Brisk walking or jogging, and combined aerobic plus resistance training, probably improve lumbar spine bone mineral density compared with no exercise, and brisk walking or jogging was the standout for total hip. Mind-body exercise — think tai chi and yoga — showed moderate certainty benefit at the femoral neck, which is not where most of us would have placed our bets. Two things matter clinically. First, the dose-response was non-linear and inverted U-shaped, meaning more is not indefinitely better; the threshold for a clinically meaningful gain sat at roughly six hundred metabolic equivalent minutes per week, which the authors translate as about two to three hours of brisk walking or jogging weekly. That is a prescription you can write in clinic. Second, in the fracture analysis, covering twenty-six trials and about eleven thousand participants, mixed aerobic exercise cut fracture odds by roughly two thirds and mind-body exercise by about forty percent, while other modalities were neutral. The certainty for most of these estimates was low to moderate, so this is cautious support rather than a mandate, and the absolute bone density differences are small — but the direction and the dose ceiling are useful.

Alongside that, the American Journal of Physical Medicine and Rehabilitation published work from Pua and colleagues on how we should actually screen for preclinical mobility limitation — that pre-disability stage where people are still independent but have started modifying how they climb stairs [2]. In 235 community-dwelling adults with a mean age of sixty-eight, fast gait speed discriminated better than habitual gait speed, and the empirically derived thresholds were about 1.24 metres per second for habitual speed and 1.67 metres per second for fast speed. Those fast-walk numbers are considerably higher than the frailty cut-points most of us carry in our heads, which is the point: if you only measure comfortable walking speed, you will miss people who are compensating. Put these two papers together and you have a workable pathway — screen with a four-metre fast walk, and prescribe a dose-specific, modality-tailored exercise programme to those who fall below threshold.

Turning to stroke, three papers this week address dose, orthotics, and dysphagia. In the American Journal of Physical Medicine and Rehabilitation, Park analysed a single-centre retrospective cohort of 1,036 adults after a first-ever ischaemic stroke, using Modified Barthel Index gain per inpatient day as the outcome [3]. About forty percent of patients had poor functional efficiency. More therapy helped — each additional thirty minutes per day was associated with better efficiency — and better nutritional status, measured by the Geriatric Nutritional Risk Index, was also associated with better efficiency. The more interesting signal is the interaction: in patients meeting criteria for sarcopenia, defined by low muscle mass index plus low grip strength in the non-paretic hand, the benefit of adding therapy time was blunted. This is observational and single-centre, so causality is unproven, but it argues that piling on more therapy minutes in an undernourished, sarcopenic patient may underperform unless you address the substrate at the same time.

On orthotics, Topics in Stroke Rehabilitation published a systematic review and meta-analysis by Yoon covering ten studies and 931 participants examining ankle-foot orthosis conditions [4]. Pooled results showed modest reductions in lower-limb electromyographic activity and in spasticity, both small-to-moderate in size with low heterogeneity, while the effect on psychosocial outcomes was essentially nil and highly heterogeneous. The author's interpretation deserves attention: reduced muscle activity may reflect welcome mechanical assistance, or it may reflect a patient who has stopped recruiting their own dorsiflexors. In someone with residual voluntary ankle function, that second interpretation should give you pause, and it reinforces individualising the prescription rather than treating an ankle-foot orthosis as a default.

The third stroke paper, also in Topics in Stroke Rehabilitation, comes from Liang and colleagues, who compared transcranial direct current stimulation alone against tongue-nape acupuncture combined with stimulation in 85 patients with post-stroke dysphagia [5]. The combination group had greater improvement in swallowing scores and a higher overall clinical response rate — about ninety-one percent versus just under eighty percent — with no difference in adverse reactions, plus reductions in serum neural injury markers and increased oxyhaemoglobin in sensorimotor and prefrontal cortex on near-infrared imaging. It's a single-centre study of modest size with group assignment rather than described randomisation and no sham control, so treat the mechanistic biomarker story as hypothesis-generating rather than established neuroprotection.

In spinal cord injury, three papers span technology, prognosis, and priorities. The Journal of Spinal Cord Medicine published a meta-analysis by Hu and colleagues of twenty randomised trials and 662 patients using virtual reality, two thirds of which used non-immersive systems [6]. Compared with conventional therapy, virtual reality improved lower-limb motor scores, walking category, balance, timed walking and timed up-and-go, and Barthel Index. But upper-limb outcomes showed no significant benefit across four separate measures, and the Spinal Cord Independence Measure total score was not statistically significant. So the honest reading is that virtual reality is a reasonable adjunct for lower-limb and balance work, and not yet justified as a substitute for hands-on upper-limb therapy.

Also in the Journal of Spinal Cord Medicine, Murotani and colleagues looked at seventy patients with sensorimotor complete cervical injury — grade A on the ASIA Impairment Scale — assessed within seventy-two hours [7]. Sixteen patients converted to an incomplete grade by three months. A long sensory zone of partial preservation, and particularly an MRI-based sensory zone spanning ten or more segments below the presumptive injured segment, raised the odds of conversion to grade C or D roughly eighteen-fold, and predicted greater motor score gain. This is a retrospective single-institution case series with few events, so the estimate is imprecise, but anchoring the zone of partial preservation to the MRI-defined injury level is a cheap refinement to early prognostic conversations with families.

That prognostic optimism sits uneasily beside a commentary in Spinal Cord from Sorani and Rabchevsky on the unmet burden of secondary complications [8]. Their funding analysis notes that spinal cord injury receives roughly 0.2 percent of National Institutes of Health funding against approximately one percent of United States disease burden, and that within spinal cord injury research, movement and mobility studies attracted nearly twice the funding given to pain, bladder, bowel, and pressure injury research combined — despite two decades of survey data showing those are precisely what patients rank as their greatest daily burden. They also flag the shortage of trained spinal cord injury physiatrists as a structural barrier. It is a call to realign, and it is a useful mirror for our own clinic priorities.

Finally, two papers in Archives of Physical Medicine and Rehabilitation ask what we can stop doing, or do remotely. Duygu-Yildiz and colleagues ran a single-blind randomised non-inferiority trial in thirty-two limbs with stage two or three lower extremity lymphoedema, comparing compression bandaging alone against bandaging plus manual lymphatic drainage, five days a week for four weeks [9]. Both arms achieved large reductions in limb volume — about twenty percent with compression alone and twenty-two percent with the combination — and compression alone met the pre-specified non-inferiority margin, with no between-group differences in tissue stiffness or ultrasound-measured subcutaneous thickness. This is a small trial in a single setting and it does not settle the question for every phenotype, but it supports prioritising high-quality compression when therapist time or cost is the limiting factor. And Martin and colleagues systematically reviewed thirteen randomised trials of telerehabilitation-delivered pelvic floor muscle training for stress urinary incontinence in women [10]. Digital delivery consistently reduced symptoms and improved condition-specific quality of life, clearly beating no treatment or passive management, but performing comparably rather than better than face-to-face physiotherapy or paper booklets. Adherence was the mechanism of success and technological complexity was the main barrier. No serious adverse events. Heterogeneity prevented meta-analysis, so this is narrative synthesis — but as an access solution for women with geographic or logistical barriers, it holds up.

If you only have time for one paper this week, make it the BMJ network meta-analysis of exercise and bone health [1]. It converts a vague recommendation to stay active into a specific, dose-anchored, modality-specific prescription you can give tomorrow to every patient over forty in your clinic.

Here are the key takeaways from this week in Physical Medicine and Rehabilitation. Aim for roughly two to three hours a week of brisk walking or jogging, or combined aerobic and resistance work, for bone density in adults over forty — and remember the benefit curve flattens rather than climbing forever. Screen for preclinical mobility limitation with a fast four-metre walk rather than comfortable speed, using thresholds near 1.67 metres per second. After stroke, more therapy minutes help, but sarcopenia blunts that return, so assess muscle mass, grip strength, and nutrition before simply adding sessions. Individualise ankle-foot orthosis prescription, because reduced muscle activity can mean assistance or it can mean disuse. And in lymphoedema, compression is doing the heavy lifting — manual lymphatic drainage may be an optional extra rather than an essential component.

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

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And if you have a paper you have been meaning to read, upload the PDF, or paste any link, at audioscholar dot C C. We will turn it into audio like this one, in any of thirty-one languages.

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