Virtual Table Tennis Helps Stroke Patients Recover Arm Function — and Works Best After Age 56

DRAFT ARTICLE - Ready for Review

When a stroke damages the brain’s motor pathways, the arm is often the last thing to come back. Roughly 69% of stroke patients with one-sided weakness have upper-limb dysfunction, and about half still live with a limited arm six months later. Standard physical therapy helps, but gains tend to plateau. A new meta-analysis published in BMC Neurology suggests a familiar, low-cost addition to the toolkit: motion-sensing video games built around racket swings — with table tennis and tennis titles the two most common types tested (Cai et al. 2026, PMID 42104238).

Ten trials, 426 patients

The review, registered in PROSPERO (CRD420251141175), combed PubMed, Web of Science, Scopus, and the Cochrane Library for randomized controlled trials published between January 1999 and September 2025. Ten trials made the cut, enrolling 426 stroke survivors — 215 assigned to game-based training and 211 to conventional therapy.

The participants were mostly in their fifties, with a wide range of impairment: about 80% had mild to moderate disability. Programs ran 2 to 12 weeks, with sessions 2 to 5 times per week lasting 20 to 60 minutes. The games were motion-controlled titles from the Nintendo Wii and Xbox Kinect era. Categorized by sport, tennis studies were the most common (3 trials) and table tennis studies the second most common (2 trials) — together making up half the evidence base, ahead of baseball, golf, and generic swing games.

The headline number

Every trial measured arm function with the Fugl-Meyer Upper Extremity assessment (FMA-UE), the standard clinical scale for post-stroke arm recovery. Pooled across all ten trials, patients who trained with swing-based games improved significantly more than those receiving conventional therapy alone:

• Overall effect: SMD 0.64 (95% CI 0.44–0.84) — a moderate, clinically meaningful effect. In plain terms, the average game-trained patient recovered about two-thirds of a standard deviation more arm function than the average conventionally treated patient.
• Age mattered most. Patients aged 56 and older showed an effect of SMD 0.83 (95% CI 0.56–1.10), while younger patients showed 0.40 (95% CI 0.10–0.71). The difference between the two age groups was statistically significant (p = 0.038) — the older patients gained roughly double.
• Chronic stroke patients benefited more. Trials with longer time since stroke showed SMD 0.76 (95% CI 0.51–1.01), versus 0.44 (95% CI 0.10–0.77) for shorter-duration patients.
• Severe strokes responded too. The severe-impairment subgroup improved at SMD 0.69 (95% CI 0.45–0.93), slightly more than the mild subgroup at 0.54 (95% CI 0.18–0.91).
• Shorter programs packed the punch. Shorter interventions showed SMD 0.68 (95% CI 0.45–0.90) versus 0.53 (95% CI 0.09–0.96) for longer ones.

Sensitivity analysis showed the pooled result was stable — no single trial carried the finding. Egger’s test found no sign of publication bias (p = 0.454).

The table tennis trials inside

Two of the ten trials used table tennis games specifically — and one of them was the largest study in the entire analysis. Anwar and colleagues (2022) randomly assigned 74 stroke survivors in Pakistan to Nintendo Wii training or conventional therapy, with five 30-minute sessions per week for four weeks (PMID 35699989). Kong and colleagues (2016) enrolled 40 patients in 60-minute table tennis game sessions, tracking FMA-UE along with the Action Research Arm Test and the Stroke Impact Scale.

The authors note that these two table tennis trials, together with one golf study, accounted for a larger share of the between-study variability — evidence that sample and program differences, not a uniform recipe, drove the pooled result.

Why swinging works

The review’s authors point to a chain of mechanisms. Racket-style tasks demand rapid reactions, precise ball striking, and continuous coordination between both sides of the body — the exact circuitry stroke damages. Fast swinging movements also recruit type II muscle fibers, the fast-twitch units that slow, repetitive clinic exercises under-stimulate. And on the motivation side, games provide scores, instant feedback, and just enough fun that patients actually do the repetitions. Adherence is rehabilitation’s oldest problem; a rally you want to win solves some of it.

This lines up with earlier table tennis rehabilitation science. A 2024 EEG study of seated table tennis play in stroke patients found significant increases in beta-wave activity across frontal and parietal brain regions after just eight sessions — markers of the motor-planning and attention circuits recovery depends on (Seo & Kim 2024).

The honest caveats

Only ten trials went into the pool, and the variation between them was high (I² = 82.2%, p < 0.001) — different games, doses, stroke severities, and timeframes. The subgroup findings are exploratory, not confirmatory. Some trials lacked blinding, which can inflate benefit estimates in rehabilitation research. And the pooled effect covers swing-based games as a class; it cannot isolate table tennis games alone, even though they form the largest single-trial contribution. The authors themselves call for standardized protocols before firm dosing guidance is possible.

What it means

For stroke survivors and rehabilitation clinics, the practical takeaway is modest hardware, real returns: motion-sensing game consoles — much of it Wii-era equipment now available second-hand for pocket change — turned racket-swing drills into measurable arm recovery in trial after trial, on top of whatever conventional therapy delivered. The strongest signal was in exactly the patients rehab struggles with most: older survivors, months or years past their stroke, whose recovery had slowed. No prescription, no clinic queue — just a paddle, a sensor, and a ball that never stops coming back.

Sources

1. Cai W, Zheng Y, Wu J, Hu G, Takeshita D. “The effectiveness of virtual reality/motion-sensing games incorporating swing-based elements on upper limb function after stroke: a systematic review and meta-analysis.” BMC Neurology 2026;26:365. PMID 42104238.
2. Anwar N, Karimi H, Ahmad A, Gilani SA, Khalid K. “Virtual Reality Training Using Nintendo Wii Games for Patients With Stroke: Randomized Controlled Trial.” JMIR Serious Games 2022;10(2):e29830. PMID 35699989.
3. Seo S, Kim Y. Sitting table tennis program for stroke rehabilitation. Journal of Physical Medicine Rehabilitation & Disabilities 2024;10:090.