Background The cross-education effect after unilateral resistance training (CE-URT) describes force production gains in a contralateral non-trained homologous muscle. While CE-URT is established in healthy populations, its role in clinical populations has not been systematically assessed. The CE-URT magnitude may be influenced by training parameters (i.e. contraction type, intensity, anatomical region and training–testing contraction specificity), but their contribution remains unclear in experimental models of immobilisation in healthy immobilised participants (HIP) and clinical populations, particularly in patients with musculoskeletal conditions (PMC) and hemiparesis after stroke (PHAS), where CE-URT has been primarily applied. Objective We aimed to assess the CE-URT on force production capacity in HIP, PMC and PHAS, to examine the modulatory role of key URT prescription parameters. Methods PubMed, Scopus, CINAHL and Web of Science were searched for randomised and non-randomised controlled trials (≥ 3 weeks) assessing CE-URT compared to control conditions (immobilisation, conventional rehabilitation or within-subject control periods) on contralateral force production capacity. Risk of bias (Downs & Black) and certainty of evidence (Grading of Recommendations Assessment, Development and Evaluation [GRADE]) were assessed. A three-level random-effects meta-analysis with subgroup analyses was conducted according to population (HIP/PMC/PHAS), contraction type (eccentric/concentric/isometric/concentric-eccentric-coupled), training intensity (maximal/submaximal), training–testing contraction specificity (specific/non-specific) and anatomical region (upper/lower limb). Effect sizes (ESs) were expressed as standardised mean differences with 95% confidence intervals (95% CIs). Results Twenty-four studies (low to very-low risk of bias) were included. Compared with control conditions, CE-URT induced a moderate effect on maximal strength (ES [95% CI] = 0.654 [0.393, 0.915], P < 0.001; Q(52) = 163.14, P < 0.001; moderate-GRADE) and a small effect on rapid force production (ES [95% CI] = 0.367 [0.115, 0.619], P = 0.007; Q(17) = 7.59, P = 0.97; low-GRADE). Large effects were observed in PHAS (ES [95% CI] = 1.263 [0.439, 2.087], P = 0.008; Q(9) = 48.28, P < 0.001; moderate-GRADE) and HIP models (ES [95% CI] = 0.845 [0.533, 1.158], P < 0.001; Q(20) = 18.63, P = 0.546; high-GRADE). Large effects were found in isometric (ES [95% CI] = 1.080 [0.286, 1.873], P = 0.014; Q(8) = 41.22, P < 0.001; moderate-GRADE), testing-training contraction specific (ES [95% CI] = 1.072 [0.449, 1.695], P = 0.003; Q(10) = 42.04, P < 0.001; moderate-GRADE), upper-limb (ES [95% CI] = 0.857 [0.529, 1.185], P < 0.001; Q(32) = 76.35, P < 0.001) and eccentric protocols (ES [95% CI] = 0.831 [0.491, 1.171], P < 0.001; Q(13) = 10.62, P = 0.64; high-GRADE). Moderate effects were observed after maximal (ES [95% CI] = 0.759 [0.256, 1.262], P = 0.005; Q(17) = 67.54, P < 0.001; very low-GRADE) or submaximal intensity protocols (ES [95% CI] = 0.587 [0.273, 0.902], P = 0.001; Q(34) = 91.05, P < 0.001; low-GRADE), and training–testing contraction non-specific protocols (ES [95% CI] = 0.542 [0.280, 0.805], P < 0.001, Q(41) = 97.667, P = 0.008; low-GRADE). No significant effect was observed in PMC (P = 0.150), concentric (P = 0.411) or lower-limb protocols (P = 0.093). Conclusions The CE-URT was shown in experimental models of disuse in HIP and clinical conditions with PHAS. The CE-URT appears promising for recovery of force production capacity in clinical scenarios. However, further research is needed to optimise intervention protocols and increase certainty of the evidence, particularly in PMC.

Cross-Education Effect on Force Production Capacity After Unilateral Resistance Training in Clinical Populations: A Systematic Review with Meta-analysis

期刊论文
发表日期：

2026年7月6日

查看全文

Omar Valdés,
Emeric Chalchat,
Carlos Rehbein,
Oscar Núñez,
Sebastián García-Vicencio,
Vincent Martin,
Rodrigo Ramirez-Campillo,
Luis Peñailillo

摘要

Background

The cross-education effect after unilateral resistance training (CE-URT) describes force production gains in a contralateral non-trained homologous muscle. While CE-URT is established in healthy populations, its role in clinical populations has not been systematically assessed. The CE-URT magnitude may be influenced by training parameters (i.e. contraction type, intensity, anatomical region and training–testing contraction specificity), but their contribution remains unclear in experimental models of immobilisation in healthy immobilised participants (HIP) and clinical populations, particularly in patients with musculoskeletal conditions (PMC) and hemiparesis after stroke (PHAS), where CE-URT has been primarily applied.

Objective

We aimed to assess the CE-URT on force production capacity in HIP, PMC and PHAS, to examine the modulatory role of key URT prescription parameters.

Methods

PubMed, Scopus, CINAHL and Web of Science were searched for randomised and non-randomised controlled trials (≥ 3 weeks) assessing CE-URT compared to control conditions (immobilisation, conventional rehabilitation or within-subject control periods) on contralateral force production capacity. Risk of bias (Downs & Black) and certainty of evidence (Grading of Recommendations Assessment, Development and Evaluation [GRADE]) were assessed. A three-level random-effects meta-analysis with subgroup analyses was conducted according to population (HIP/PMC/PHAS), contraction type (eccentric/concentric/isometric/concentric-eccentric-coupled), training intensity (maximal/submaximal), training–testing contraction specificity (specific/non-specific) and anatomical region (upper/lower limb). Effect sizes (ESs) were expressed as standardised mean differences with 95% confidence intervals (95% CIs).

Results

Twenty-four studies (low to very-low risk of bias) were included. Compared with control conditions, CE-URT induced a moderate effect on maximal strength (ES [95% CI] = 0.654 [0.393, 0.915], P < 0.001; Q(52) = 163.14, P < 0.001; moderate-GRADE) and a small effect on rapid force production (ES [95% CI] = 0.367 [0.115, 0.619], P = 0.007; Q(17) = 7.59, P = 0.97; low-GRADE). Large effects were observed in PHAS (ES [95% CI] = 1.263 [0.439, 2.087], P = 0.008; Q(9) = 48.28, P < 0.001; moderate-GRADE) and HIP models (ES [95% CI] = 0.845 [0.533, 1.158], P < 0.001; Q(20) = 18.63, P = 0.546; high-GRADE). Large effects were found in isometric (ES [95% CI] = 1.080 [0.286, 1.873], P = 0.014; Q(8) = 41.22, P < 0.001; moderate-GRADE), testing-training contraction specific (ES [95% CI] = 1.072 [0.449, 1.695], P = 0.003; Q(10) = 42.04, P < 0.001; moderate-GRADE), upper-limb (ES [95% CI] = 0.857 [0.529, 1.185], P < 0.001; Q(32) = 76.35, P < 0.001) and eccentric protocols (ES [95% CI] = 0.831 [0.491, 1.171], P < 0.001; Q(13) = 10.62, P = 0.64; high-GRADE). Moderate effects were observed after maximal (ES [95% CI] = 0.759 [0.256, 1.262], P = 0.005; Q(17) = 67.54, P < 0.001; very low-GRADE) or submaximal intensity protocols (ES [95% CI] = 0.587 [0.273, 0.902], P = 0.001; Q(34) = 91.05, P < 0.001; low-GRADE), and training–testing contraction non-specific protocols (ES [95% CI] = 0.542 [0.280, 0.805], P < 0.001, Q(41) = 97.667, P = 0.008; low-GRADE). No significant effect was observed in PMC (P = 0.150), concentric (P = 0.411) or lower-limb protocols (P = 0.093).

Conclusions

The CE-URT was shown in experimental models of disuse in HIP and clinical conditions with PHAS. The CE-URT appears promising for recovery of force production capacity in clinical scenarios. However, further research is needed to optimise intervention protocols and increase certainty of the evidence, particularly in PMC.