REF-1642 · Giachetti Martin et al., 2026 · Frontiers in Sports and Active Living · Paper · PubMed

Across 15 crossover studies, carbon-plated running shoes made no significant difference to leg stiffness or to peak positive power at the knee, hip or toe joint against non-plated shoes; ankle power was lower, at the edge of significance (SMD −0.71, 95% CI −1.42 to 0.00, p = 0.05), and step frequency dipped non-significantly (SMD −0.16); GRADE certainty low to moderate

Main caveat: No injury outcome. Inference, since the abstract does not describe the protocols: joint power and leg stiffness are usually measured in short laboratory bouts…
fairly strong evidenceMeasured on recreational to highly trained athletesRead from the abstract only

Giachetti Martin and colleagues asked whether carbon-plated running shoes change running mechanics compared with shoes without a plate.

Following PRISMA and a registered protocol, they searched MEDLINE, Scopus, LILACS and Embase in July 2025 for crossover trials in healthy adults aged 18-70 who ran in both kinds of shoe, pooled standardised mean differences with random-effects models and rated certainty with GRADE.

The outcomes were step frequency, leg stiffness and peak positive joint power at the metatarsophalangeal (MTP, toe), ankle, knee and hip joints. Fifteen studies were included.

Leg stiffness and peak power at the knee, hip and MTP joint did not differ significantly between shoes; ankle power was lower in plated shoes, a borderline result (SMD −0.71, p = 0.05); step frequency fell slightly and not significantly (SMD −0.16, p = 0.06). Certainty ranged from low to moderate.

The authors conclude that plated shoes do not change joint power or leg stiffness consistently; the adaptations look subtle and may sit at the distal joints, with "no systematic redistribution of lower-limb mechanics".

Participants

Healthy adults aged 18-70 who ran in both plated and non-plated shoes (the inclusion criterion).
Not in the abstract: the number of runners, sex, training level, running speeds, shoe models.

Methods

Search: MEDLINE, Scopus, LILACS and Embase, July 2025; PRISMA; PROSPERO CRD420251058609.
Inclusion: crossover trials comparing carbon-plated with non-plated shoes in healthy adults aged 18-70.
Outcomes: step frequency (steps/min), leg stiffness (kN/m), peak positive joint power (W/kg) at the MTP, ankle, knee and hip.
Analysis: random-effects meta-analysis of standardised mean differences with 95% CIs.
Certainty: GRADE, low to moderate; which outcome received which rating is not in the abstract.
Not in the abstract: the risk-of-bias tool, heterogeneity, how many studies fed each outcome.

Key Results

Outcome, plated vs non-platedSMD (95% CI)p
Leg stiffness−0.12 (−0.46 to 0.23)0.44
Knee peak positive power0.21 (−0.10 to 0.52)0.12
Hip peak positive power−0.23 (−1.36 to 0.90)0.56
MTP peak positive power0.13 (−1.87 to 2.12)0.85
Ankle peak positive power−0.71 (−1.42 to 0.00)0.05 (borderline)
Step frequency−0.16 (−0.32 to 0.01)0.06
Our reading: the hip and MTP intervals span large effects in both directions, so "no significant difference" there means the data cannot tell, not that nothing changes; the knee and leg-stiffness intervals are narrower.

Practical Application

Plated shoes did not consistently shift work to the knee or hip. Pooled over 15 crossover studies, peak positive power at the knee and hip, and leg stiffness, did not differ from non-plated shoes; the one signal was less ankle power in plated shoes (borderline), with a slight, non-significant drop in step frequency.
This is not evidence of safety. Joint power is not tissue load or injury: the injury question for plated shoes is open (REF-724), and no shoe type has been shown to reduce running injuries (REF-716).
Inference, not tested here: if the shoe takes over part of the ankle's push-off work, a runner moving most runs into plated shoes may change how the calf and Achilles are loaded; bring them in gradually and treat new calf, Achilles or foot pain as a reason to step back.
Not settled by this design: effects at the hip and toe joint (intervals too wide), joint moments and tissue stress, fatigue late in a long run or race, and differences by sex, level or foot strike; the certainty rating per outcome is not in the abstract.

Limitations

No injury outcome. Inference, since the abstract does not describe the protocols: joint power and leg stiffness are usually measured in short laboratory bouts, so fatigue over a long run is probably not captured.
Peak positive joint power and leg stiffness only; joint moments, negative work and tissue loads are not pooled.
Wide intervals at the hip and MTP joints; certainty per outcome, heterogeneity and the number of studies per outcome are not in the abstract.
The abstract describes plated shoes as compliant foam plus a carbon plate, so, unless the non-plated shoes had the same foam (not stated in the abstract), the comparison does not isolate the plate from the foam.
Written from the abstract and the PubMed record; full text not read.
How this entry was checked and changed
Why this level

the abstract reports its GRADE certainty (low to moderate), so the risk of bias counts as read (rule 1); with part of the evidence at low certainty a review is A- at most; joint power and leg stiffness are measured mechanics, not markers

Content correction 2026-11-07
2026-11-07: added from the abstract and the PubMed record; level A- (meta-analysis of 15 crossover studies; GRADE certainty low to moderate reported in the abstract).