‘Pushing’ versus ‘holding’ isometric muscle actions; what we know and where to go: A scoping and systematic review with meta-analyses

Background

Pushing/pulling isometric muscle actions (PIMA) are commonly used to assess strength, fatigability, and neuromechanical function, whereas holding isometric muscle actions (HIMA), applied in rehabilitation and performance settings, remain less clearly defined and comparatively understudied. Evidence suggests that PIMA and HIMA may elicit distinct neural and cardiovascular responses, yet inconsistent operational definitions complicate interpretation and application. This review synthesized research directly comparing PIMA and HIMA to clarify their physiological profiles, identify research gaps and explore practical relevance.

Methods

The protocol was pre-registered with PROSPERO (CRD42024530386). Databases were searched for peer-reviewed studies comparing PIMA and HIMA. Study quality and risk-of-bias were evaluated, and meta-analyses and meta-regressions were performed on time-to-task-failure (TTF), ratings of perceived exertion (RPE), heart rate (HR), and mean arterial pressure (MAP).

Results

Fifty-four studies (publication year 2012.9±6.9; 1995-2024) were identified (N=919 participants; ∼29.8±10.7 years). Thirty-five studies reported performance measures, 45 examined neural outputs, and 14 assessed cardiovascular or metabolic responses. Meta-analysis revealed longer TTF for PIMA vs. HIMA at the same absolute intensity (n=407; g =−0.74, p <0.001), except for two studies on axial muscles ( g =1.78–3.59, p <0.001). Individual-study patterns suggest diminishing TTF differences at higher intensities; however, since two other studies found clear differences, this may reflect methodological heterogeneity rather than a true intensity effect. No significant differences were identified for HR, MAP, or RPE at relative time points, except for higher RPE at 50%TTF during PIMA. Qualitatively, PIMA was associated with higher peak torques and discharge rates, whereas HIMA was associated with higher burst rates, glucose uptake, and force fluctuation increases.

Conclusions

These mechanistic distinctions may hold practical relevance as PIMA may be beneficial for prolonged activation and agonist neuromuscular adaptations. In contrast, HIMA could provide diagnostic value, injury-prevention potential, and time-efficient muscular, neural, and cardiovascular adaptations in rehabilitation. Methods varied widely across studies, making additional meta-analyses impossible. Randomized controlled trials are required to confirm the use of PIMA vs HIMA in clinical or performance contexts.

Key Points

• Pushing/pulling and holding isometric actions produce distinct neuromuscular and physiological responses and should not be considered equivalent.

• Pushing/pulling actions generally support longer force maintenance, with task- and muscle-specific exceptions.

• Neural, mechanical, and metabolic characteristics differ, with pushing/pulling emphasizing antagonist force output while holding is more neuromuscular complex and metabolically taxing.

• These differences suggest divergent applications, with pushing/pulling suited to performance goals and holding suited to rehabilitation and specific diagnostic contexts.