Increasing Muscle Extensibility: A Matter of Increasing Length or Modifying Sensation?

Viscoelastic Deformation

Many human studies^11,16,–,18 suggest that increases in muscle extensibility observed immediately after stretching are due to a lasting viscoelastic deformation. Skeletal muscles are considered to be viscoelastic. Like solid materials, they demonstrate elasticity by resuming their original length once tensile force is removed. Yet, like liquids, they also behave viscously because their response to tensile force is rate and time dependent.^1,14 An immediate increase in muscle length can occur due to the viscous behavior of muscles whenever they undergo stretch of sufficient magnitude and duration or frequency. This increased length is a viscoelastic deformation because its magnitude and duration are limited by muscles’ inherent elasticity.¹ Viscoelastic deformation has been tested in research using various stretching methods such as “static” (constant joint angle) stretches,^19–23 constant load,²⁴ contract/relax,²⁵ and repeated cyclic stretches.^23,26 Static stretching can be used to evaluate the property of viscoelastic stress relaxation. When stretch is applied to a muscle and the muscle is held in the stretched position for a period of time, as is the case with normal static stretching techniques, the muscle's resistance to stretch gradually declines (Fig. 3).^1,2,14,27 This decline in resistance to stretch is called viscoelastic stress relaxation and is expressed as a percentage of the initial resistance.^14,19,20 Constant load stretching, such as stretching that uses a fixed torque, can be used to evaluate the property of creep. Creep occurs when mechanical length gradually increases in response to a constant stretching force.^1,2,23

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Viscoelastic stress relaxation during static stretch. Peak torque occurs when muscle first reaches the final stretch position. Final torque is measured at the end of the static stretch holding phase. Viscoelastic stress relaxation (delta torque) is the decrease in torque and can be expressed as a percentage of peak torque: (peak torque − final torque) ÷ peak torque. Reprinted with permission of Wiley-Blackwell from: Magnusson SP. Passive properties of human skeletal muscle during stretch maneuvers: a review. Scand J Med Sci Sports. 1998;8:65–77.

The study most commonly used to support the theory that viscoelastic deformation is responsible for increases in human muscle extensibility is an animal study by Taylor et al.²³ The results of this study showed an immediate increase in MTU length induced by repeated cyclic and static stretches. The authors suggested that the observed length increases should be lasting due to the viscous properties of the MTU.²³ However, no further testing was performed to determine the duration and residual magnitude of these length increases.²³

In human studies, viscoelastic deformation and recovery have been tested on hamstring and ankle plantar-flexor muscles.^{20–22,24,28} The results refute viscoelastic deformation as a mechanism for lasting increases in muscle length and extensibility. These studies showed that the magnitude and duration of the length increases vary depending on the duration of the stretch and the type of stretching applied. All of these studies consistently showed viscoelastic deformation of human muscle to be transient in nature. With stretch application typical of that practiced in rehabilitation and sports, the biomechanical effect of viscoelastic deformation can be quite minimal and so short-lived that it may have no influence on subsequent stretches. In one hamstring muscle study, a static stretch of 45 seconds’ duration was found to have no significant effect on the next stretch performed 30 seconds later.²⁸ With 3 consecutive 45-second static stretches (30-second rest intervals between stretches), each stretch showed VESR of 20% during the static holding phase. However, the muscles had already recovered from the relaxation by the next stretch.²⁸ Similar results were demonstrated in a study of ankle plantar-flexor muscles.²¹ There was no change in stiffness of the ankle plantar-flexor muscles that underwent static stretches of: (1) 4 sets of 15 seconds’ duration and (2) 2 sets of 30 seconds’ duration (10-second rest intervals between stretches).²¹

Plastic Deformation of Connective Tissue

Another popular theory suggests that increases in human muscle extensibility observed immediately after stretching are due to “plastic,”^17,29–32 or “permanent”^17,30–36 deformation of connective tissue.³⁷ The classical model of plastic deformation would require a stretch intensity sufficient to pull connective tissue within the muscle past the elastic limit and into the plastic region of the torque/angle curve so that once the stretching force is removed, the muscle would not return to its original length but would remain permanently in a lengthened state (Fig. 4).^1,2 In 10 studies^17,29–37 that suggested plastic, permanent, or lasting deformation of connective tissue as a factor for increased muscle extensibility, none of the cited evidence was found to support this classic model of plastic deformation. The term “plastic deformation” often was considered only to be a synonym for deformation that is permanent in nature.^31,32

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

Model passive length/tension curve for biological tissue. The elastic region begins at the initial length and ends at the elastic limit. Increases in the length measurement due to applied tensile force are temporary. When tensile force is removed, the specimen will resume its original length. In the plastic region, with application of tensile force beyond the elastic limit, permanent deformation of the specimen will occur. In this case, once the tensile force is removed, the specimen will not return to its original length. Failure or rupture point is the last point on the curve. The length attained at the rupture point is the maximum length of the specimen. Note: Number values are absolute; curve is a theoretical illustration.

The evidence cited^{29–31,33–35,37} in support of this theory can be traced almost entirely to a study by Warren et al³⁸ performed on rat tail tendons and a review article by Sapega et al.³² Neither of these works recommended the classical model of plastic deformation, which requires high stretching loads, but instead suggested viscoelastic deformation: using lower stretching loads with prolonged stretch duration in order to facilitate “viscous flow” within the connective tissue.

Although model passive length/tension curves that include a plastic deformation phase may be applicable for some types of biological tissue, studies of muscle demonstrate a markedly different typical curve. A plastic deformation phase would be reflected on the passive length/tension curve by a decrease in its slope.² There was no evidence of a classic plastic deformation phase occurring in any of the cited studies.

Increased Sarcomeres in Series

Animal studies have demonstrated that the number of sarcomeres in series of a muscle can be changed by prolonged immobilization in extreme positions. That is, when muscles are immobilized in fully extended positions, there is an increase in the number of sarcomeres in series. Although often reported otherwise, these muscles demonstrated no overall change in muscle length because increases in the number of sarcomeres in series were offset by a concurrent decrease in sarcomere length.^39–41 When muscles are immobilized in shortened positions, there is a decrease in the number of sarcomeres in series and a concurrent decrease in muscle length.^39–41 Sarcomere number and muscle length in the shortened muscles have been found to increase to normal levels after recovery from immobilization.^39,40 These animal studies suggest that muscles adapt to new functional lengths by changing the number and length of sarcomeres in series in order to optimize force production at the new functional length.^39,41 Despite substantial differences between muscle immobilization and intermittent stretching, this research has been generalized to suggest that short-term (3- to 8-week) human stretching regimens cause similar increases in sarcomeres in series and a concurrent increase in length of the stretched muscles.^{7,11,12,17,31,42–45}

For obvious practical and ethical reasons, there are no human stretching studies that evaluated on a histological level whether the number of sarcomeres in series changes due to therapeutic intervention. Perhaps with development of imaging techniques, this will someday be a possibility.

Neuromuscular Relaxation

The rehabilitation literature often suggests that involuntary contraction of muscles due to a neuromuscular “stretch reflex” can limit muscle elongation during static stretching procedures.^{16,33,34,46–50} In order to increase muscle extensibility, it often has been proposed that slowly applied static stretch (used alone or in combination with therapeutic techniques associated with proprioceptive neuromuscular facilitation) stimulates neuromuscular reflexes that induce relaxation of muscles undergoing static stretch.^{16,17,34,36,46,47,49–53} Some authors,^34,53 furthermore, have suggested that neuromuscular reflexes adapt to repeated stretch over time, which enhances the stretched muscles’ ability to relax and results in increased muscle extensibility.

Experimental evidence does not support any of these assertions.^13,14,54,55 Stretch reflexes have been shown to activate during very rapid and short stretches of muscles that are in a mid-range position, producing a muscle contraction of short duration.⁵⁴ However, most studies of subjects who were asymptomatic and whose muscles were subjected to a long, slow, passive stretch into end-range positions did not demonstrate significant activation of stretched muscles.^14,54,56,57 Even studies that simulated ballistic (cyclic and high-velocity) stretching demonstrated no evidence of significant stretch reflex activation of muscles both in human²⁶ and animal²³ models. In a study that evaluated the effects of a single “contract-relax” stretch²⁵ and in short-term (3 and 6 weeks’ duration) stretching studies,^45,58 no significant electromyographic activity of the stretched muscles was found and no shift of passive torque/angle curves was observed. The increase in end-range joint angles, therefore, could not be attributed to neuromuscular relaxation.^14,25,45,58

Single Stretching Session

A modification in sensation that occurs immediately after a single stretching session was first reported in 1996 in a study by Magnusson et al²⁵ (using a passive knee extension test involving a Kin-Com dynamometer* with a modified thigh pad) that investigated the effects of a 10-second static hold versus a single contract/relax stretch on human hamstring muscles. Halbertsma et al,⁵⁹ using an instrumented passive straight-leg raise method, tested subjects just prior to and immediately following a 10-minute hamstring muscle stretching session. In both of these studies, there was no shift of passive torque/angle curves (Fig. 1). but an increase in applied torque and increased end-range joint angles were observed (Fig. 2). Subsequent studies showed that sensory perception in response to stretching of human hamstring muscles is acutely modified by assuming a stooped versus an upright trunk position⁶⁰ and is similarly modified by a single 90-second static stretch and 10 repeated cyclic (“ballistic”) stretches.²⁶

Short-Term (3- to 8-Week) Stretching Programs

In a study investigating the biomechanical effects of a 4-week hamstring muscle stretching program, Halbertsma and Göeken¹⁵ (using an instrumented passive straight-leg raise test) found that sensations of pain onset and pain or stretch tolerance occurred at increased torques, resulting in increases in hamstring muscle extensibility (pain onset: mean increase=10°, 95% confidence interval=5°–14°; pain or stretch tolerance: mean increase=5°, 95% confidence interval=1°–9°). Concurrently, no shift of passive torque/angle curves was observed. Magnusson et al⁵⁸ reported similar results in a 3-week hamstring muscle stretching study that induced a 17-degree mean increase in end-range joint angles using an endpoint of pain onset. These results—no shift in passive torque/angle curves accompanied by increases in end-range joint angles—have been supported repeatedly in studies involving hamstring muscles and using various stretching and testing methods.^{17,42,43,45,61,62} Modification of subjects’ sensory response to stretch after short-term stretching programs also has been demonstrated in the rectus femoris muscle⁶³ and in ankle plantar-flexor muscles.^44,64 Studies involving subjects with spinal cord injuries showed no evidence of a shift in torque/angle curves after 4-week programs of sustained 30-minute daily stretching of hamstring⁶⁵ and ankle plantar-flexor⁶⁶ muscles, further supporting the notion that short-term stretching does not alter torque/angle relationships.

Long-Term (>8 Weeks) and Chronic Stretching Programs

The effect of longer-term stretching programs (>8 weeks) and rigorous chronic stretching regimens on passive torque/angle curves has not yet been evaluated.¹⁴

Conflicting Terminology

Throughout the rehabilitation literature regarding the effects of stretching, confusion arises due to inconsistent use of terminology among studies. Some of the above-cited studies confirmed that increases in muscle extensibility occurred after stretching, whereas others claimed that muscle extensibility did not increase. On the surface, it appears that these studies had conflicting results, but the difference merely resides in the definition of muscle extensibility. The studies reporting increases in muscle extensibility used a sensory endpoint, which indicates that the selected sensation had onset later during stretch application, allowing increases in end-range joint angles.^15,59 The studies reporting no increases in muscle extensibility used an endpoint of standardized torque, which gives some evidence that there was no shift of the torque/angle curves or change in muscle stiffness.^61,62,65 Taken together, the findings of all of these studies support the sensory theory to explain increases in end-range joint angles.

Conflicting Interpretations

Although the results of many of the supporting studies were similar, not all of these studies attributed the findings to a change in sensory perception.^43–45,64 Some studies^43–45 suggest instead that because there is increased applied torque, a longer torque/angle curve, and increased end-range joint angles, the stretching program has induced structural changes within the muscle, such as increased sarcomeres in series and a mechanical increase in muscle length. If an increase in the number of serial sarcomeres is accompanied by an increase in length of the muscle, there should be an observable right shift in the entire passive torque/angle curve similar to the shift in passive length/tension curves shown in the animal studies (Fig. 1).^39,40,67 Without a concurrent right shift in passive torque/angle curves, there is no evidence of an increase in muscle length. The theory of structural adaptation occurring after a short-term stretching program also does not explain similar increases seen immediately after a single stretching session that occur without a regimented stretching program.^14,25,59,60

Conflicting Results

Although there is growing evidence to support the theory that increases in muscle extensibility observed after stretching are due to modified sensation only, there are a few conflicting reports. In a study of ankle plantar-flexor muscles, Guissard and Duchateau⁶⁸ observed a right shift of torque/angle curves that occurred over a 6-week training period. This result may have been due to the vigorous design of the stretching program, which was performed 5 days per week and took 20 minutes to complete. This stretching dosage for a single muscle group on a single limb is well in excess of the 15 to 150 seconds^{29,44,64,69,70} of daily stretch typically used in sports and research but may be applicable in rehabilitation settings. Thirty days after the stretching program ended, increases in extensibility and muscle length were partially maintained. More research is needed to determine: (1) whether increases in muscle length are an appropriate and desirable outcome of treatment and (2) the most efficient therapeutic intervention and dosage to induce and maintain length increases.

Muscle Extensibility

Muscle extensibility is a critical dimension of muscle length. Tests of extensibility (traditionally called “muscle length tests”) were developed with the idea that there is a “normal” or ideal range of muscle extensibility that promotes optimal kinematics, resulting in efficient motion, enhancing the ability to adapt to imposed stresses, and potentially decreasing the risk of injury.^71,72 It is suggested that when a particular muscle or muscle group demonstrates insufficient extensibility (appearing to be “short”), motion between joint surfaces that the muscle crosses may be limited, resulting in restricted joint motion. When the muscle or muscle group demonstrates excessive extensibility (appearing to be “long”), motion between the joint surfaces also may be excessive, resulting in excessive joint motion.

Whether insufficient or excessive, a deviation from optimal extensibility is thought to precipitate unusual wear patterns on capsular structures and articular surfaces of involved joints. It is suggested that deviations from optimal extensibility contribute to muscle imbalances, faulty posture, and dysfunctional movement.^71,73 Although guidelines for what constitutes insufficient, optimal, and excessive extensibility measurements are based on the science of kinematics, their clinical validity has rarely been studied.

Although kinematic analysis is concerned with the motion that occurs at the joint and can identify the clinical extensibility measurements that are theoretically optimal, it is not concerned with analyzing the forces causing the motion.^1,72 Except in cases where bony approximation is the limiting factor, this type of analysis does not clearly define what should constitute the endpoint of stretch application. Perhaps this is the reason that stretch endpoints often are poorly defined and inconsistent among texts and research studies and in clinical practice.

An endpoint of the examiner's perception of “(firm) resistance” is suggested in some texts^72,74 and often is used in research, although many studies did not measure the amount of applied torque required to reach this point.^{35,49,53,69,70} The validity and reliability of this endpoint are highly questionable because without quantitative measurement, there is no way to be certain that torque is being applied consistently.⁷⁵ There also is evidence that the amount of torque applied by trained therapists can vary markedly—as much as 40-fold for a single subject.⁷⁶ Even if torque were standardized, how would the most clinically relevant torque be determined?

The importance of subject sensation as an endpoint has largely been overlooked. To date, endpoints of subject sensation are widely used in research, but basic texts describing muscle extensibility assessment have not clearly and unequivocally made this recommendation. Is passive muscle stiffness necessary to stop joint motion, or is it possible that just the subject's sensory perception of stiffness or perception of moderate stretch can be a limiting factor? Studies evaluating the biomechanical effects of stretching reveal that in controlled clinical settings under the condition of slowly applied passive stretch, it is subject sensation—not the degree of stiffness—that limits joint motion. Researchers have been able to apply passive torque up to the sensory endpoint of pain^25,58 or stretch tolerance^15,59 without being limited by stiffness. It seems reasonable that subject sensation could both alter and reflect the way the tested muscle is routinely used in function. Further research is needed to determine whether subject sensation is a significant factor in limiting joint motion during functional motion and whether muscle extensibility measurements are truly a reflection of the way muscles are used during function.

Although subject sensation is the most frequently used endpoint in human stretching research, there is little consensus regarding which sensation is most relevant clinically. A wide range of sensations has been used in research, from the subject's perception of a “pull,”⁴⁷ to varying degrees of:

• “resistance”^16,37

• “stretch”^{15,17,19,20,22,42,44,45,59,63}

• “discomfort”^{7,29,33,69–71}

• “tightness”^{25,26,28,58,69,70,77}

• “stiffness”⁷⁸

• “pain”^{7,15,18,25,26,58,60,64,79}

A change in endpoints from detection of the “first sensation of pain” to pain or stretch tolerance can result in a change in end-range joint angles that varies markedly among subjects.¹⁵ This has been demonstrated in the hamstring musculature of otherwise “normal” subjects assessed with “short” hamstring muscles to range anywhere from no change at all to a 20-degree increase in end-range joint angle values.¹⁵ Further research is needed to assess which sensation is most clinically relevant.

Subject sensation is—at the very least—an important endpoint of the torque/angle curve and may give information regarding how the muscle is routinely used during functional activity. However, extensibility measurements alone are only one dimension of muscle length and may not accurately reflect the actual length of the muscle. This has been demonstrated repeatedly in studies that included evaluation of torque⁸⁰ and cross-sectional area.^81,82 Torque and cross-sectional area measurements provide critical information that allows a more precise muscle length assessment.

Torque

When the dimension of tension is included in muscle length evaluation, a passive torque/angle curve can be constructed. This curve shows the relationship between individual extensibility measurements and the torque required to attain each measurement (the torque/angle relationship). Using this curve, important biomechanical properties such as stiffness, compliance, energy, and hysteresis can be assessed.² This information allows evaluation of an individual's muscle for comparison before and after an intervention, thus showing the effect of the intervention on the tested muscle's biomechanical properties. Use of this type of testing led to the development of the sensory theory.^15,25,58,59

Torque/angle curves, however, may not fully reflect actual muscle length. Torque measurements quantify a muscle's resistance to passive stretch, and this resistance is partly determined by thickness of the muscle. Other factors being equal, a thicker muscle demonstrates increased stiffness at a given joint angle, which causes the muscle to appear shorter on a torque/angle curve. A thinner muscle, other factors being equal, demonstrates decreased stiffness at a given joint angle, causing the muscle to appear longer on a torque/angle curve. In order to evaluate the contribution of muscle thickness to passive resistance, measurement of cross-sectional area is required.

Cross-Sectional Area

Measurement of cross-sectional area is, by itself, valuable. Changes in cross-sectional area indicate an intervention effect of muscle hypertrophy (when increased) and atrophy (when decreased).

When assessment of muscle cross-sectional area is combined with torque and joint angle, the biomechanical properties of muscles of different thicknesses can be compared.^1,81,82 Measurements of stress (tension/cross-sectional area), as well as normalized stiffness, compliance, energy, and hysteresis values, can be derived. These normalized values allow researchers to determine to what degree muscle cross-sectional area contributes to observed passive resistance and biomechanical properties.