Stretching is component of health and performance governed largely by conventional wisdom with most people naive to what effects stretching has on performance, risk of injury, or even what stretching does from a mechanical or physiologic perspective. This blog summarizes the evidence from systematic reviews on a large number of research studies about the acute effects of stretching.
Acute Effects of Stretching
A recent systematic review by Behm, Blazevich, Kay, and Mchugh examined the acute (immediate) effects of static stretching (SS), dynamic stretching (DS), and proprioceptive neuromuscular facilitation (PNF) stretching on muscle performance, injury risk, range of motion (ROM) and the physiologic mechanisms underlying these changes.
First, let me quickly explain the difference in SS, DS, and PNF stretching. SS are stretches that are held at a certain muscle length usually to achieve some pre-determined degree of subjective “stretch” sensation. DS involves repeated movements to elongate a muscle through a pre-determined range of motion. PNF stretching usually involves one of two methods often done with the help of a partner. The first is contract-relax (CR.) Contract-relax is the use of a static stretch followed by an isometric contraction of the agonist muscle group (muscle being stretched) followed with relaxation into further stretch. The second method of PNF stretching is termed contract-relax, agonist-contract (CRAC.) This is the same as CR except when relaxing out of the isometric contraction the individual then attempts to contract the antagonist muscle while stretching the agonist further.
This review showed that acute SS does seem to slightly impair subsequent muscle performance variables with a dose dependent relationship. When stretches were held for less than 60 seconds a average weighted decrease between studies of 1.1% was noted in performance variables such as sprint velocity, jump height, and strength during knee extensor maximal volitional contraction. When static stretching was held greater than 60 seconds performance decreased 4.6%. It also appears that strength deceases are greater when a muscle is tested at shorter muscle lengths as compared to longer muscles lengths (-10.2% vs +2.2%.)
When looking at DS this review showed a weighted performance effect of improving performance 1.3% when averaged between studies. In general there were fewer studies that showed a negative effect of DS on performance as compared to those of SS. DS does not seem to have a clear dose dependent relationship in duration as SS does. Unfortunately there is large variation in the studies examined in terms of the amplitude of the dynamic stretches and few studies examined the frequency of the dynamic stretching movements so it is unclear of how the range of motion or the speed or frequency of movement might affect subsequent performance.
This review examined the effects of PNF contract-relax techniques and showed an average decrease between studies of 4.4% in performance despite most studies showing nonsignificant changes. The review hypothesized that PNF stretching may follow similar patterns of a dose dependent response as SS with stretches less than 60 seconds resulting in less performance loss as compared to stretches held for greater than 60 seconds. There were 9 studies that actually compared the effects SS vs PNF and showed PNF had greater performance decreases (-6.4%) as compared to SS (-2.3%.)
The mechanisms underlying these strength losses are not completely evident but there is some evidence giving insight into this. One hypothesis is that changes in tendon stiffness may cause a muscle to function at shorter and weaker lengths. The review cites counter evidence to this with a study showing that the gastrocnemius produces less force following stretching despite being at the same length; this would suggest potential decreases in central (efferent) drive to muscles.
Another hypothesis provided is that mechanical stretching imposes stress into a muscle-tendon unit with the decreased blood flow, which occurs during stretching, possibly leading to increased metabolic end products. Animal studies have shown these end product accumulation in response to acute stretch, though this has not been examined in humans. It is also postulated that stretching may cause an impaired transmission of the action potential across the sarcolemma (muscle cell membrane.) And while there is evidence for a decrease in EMG in response to stretching, there is also evidence showing no changes in EMG; furthermore it is unclear how EMG would influence the transmission of electrical activity at the level of the sacrolemma. There is some evidence that there is a change in the efferent activity of motoneurons as a result in reduced facilitation from muscle spindles and would help explain why strength losses are greater at short muscle lengths.
The authors of this review did a good job of placing these performance decreases in perspective. Firstly, the performance measures were assessed on average 3-5 minutes after stretching; that is a pretty quick turnaround and likely not matched to what is done in sport. Secondly, the relative decreases in performance were quite small and may not even manifest themselves in changes in performance in sport or in more complex tasks. Thirdly, these changes may not persist if dynamic activity is performed subsequent to the stretching and prior to performance testing. However, if a concern still exists for decreased muscle performance from stretching interventions then one should consider incorporating dynamic stretching as there appears to be smaller deceases and perhaps slight improvements in performance subsequent to dynamic stretching.
The review also examined the effects of pre-activity stretching on injury risk. Twelve studies (SS or PNF, none used DS) were incorporated into the analysis with eight showing some effectiveness and four showing no effect; no studies showed any increase in injury risk. The reduction in injury risk is most evident when assessed in the context of sprint running-type sports as compared to endurance sports with a average risk reduction of 54% to acute muscle injury (as a side note, keep in mind this is a relative-risk reduction and not an absolute risk reduction. For example if hamstring strains occur at a rate of 0.27/1000 training hours then the subsequent risk would be 0.15/1000 training hours.) All-injury risk reduction is less evident as pre-activity stretching does not seem as helpful in preventing overuse injury.
Perhaps the most commonly cited reason for stretching is to improve flexibility or to “decrease stiffness.” This relates to ROM improvements from stretching. The current body of literature does not support the notion of one method of stretching as being superior to another for improving ROM. Improvements in ROM are noted to be impermanent from acute stretching with increase ROM lasting anywhere from 5 to 120 minutes depending on the study.
This improvement in ROM from stretching is most attributed predominantly from an increase in tolerance to stretch, not necessarily a change in muscle stiffness; though evidence does exist towards both explanations. It is important to understand what is meant by “stiffness” as it is a scientific term relating to a stress-strain curve. Stiffness is the amount of force (stress) required to deform, or strain, a material. It is represented by the slope of a stress-stain curve; a simplified example I have constructed below.
So the response to stretching is not to decrease stiffness which would manifest by shifting that curve to the right. Rather that curve remains the same, the individual is just able to tolerate going further along that curve into more strain (stretch.)
To summarize, while there does appear to be decreases in muscle performance in response to pre-activity stretching these values are very small and likely would not be too impactful in sport or performance in more complex tasks. Furthermore, these performance changes average about 1% when performed for less than 60 seconds, which, while significant in a research setting, would almost assuredly go unnoticed in normal performance situations. If you are concerned with the loss of strength or performance then dynamic stretching seems to be the way to go with potential, though very small, performance improvements noted. In terms of reducing injury risk, stretching does not seem to affect injury risk in endurance sports or for overuse injuries but does seem beneficial for reducing the risk of muscle injuries in higher velocity, repetitive movement-based activities. Range of motion is reliably increased with each type of stretching (SS, DS, PNF) with no clear superior method evident in the literature. The method of this improvement in range of motion is most likely predominantly from a perceptual change entitling more of an increase in tolerance to stretch.
So if you like to stretch before you train, continue to do so. If not, and you are performing sprint-like or high velocity movements then perhaps consider the use of either short duration static stretching followed by dynamic activity or by dynamic stretching. In my opinion, the most beneficial approach to pre-exercise activity is to perform a thorough warm-up with the goal of increase body temperature using movements that replicate those movements which will occur during training.
Stay tuned for another blog on the chronic effects of stretching such as with programs of stretching taking place over many weeks.
- Behm DG, Blazevich AJ, Kay AD, Mchugh M. Acute effects of muscle stretching on physical performance, range of motion, and injury incidence in healthy active individuals: a systematic review. Appl Physiol Nutr Metab. 2016;41(1):1-11.