The Effects of Hamstring Stretching on Range of Motion: A Systematic Literature Review

Key Words: flexibility, hip, knee, PEDro

H

amstring stretching is popular among physical therapists, athletic trainers, and fitness/coaching professionals, who all have an interest in improving flexibility in both asymptomatic and symptomatic clientele. This popularity exists despite the lack of consensus regarding the effectiveness of using hamstring stretching to reduce injury risk, enhance athletic performance, or decrease postexercise muscle soreness.11,14,15,24,32,38,42 A large number of studies, including all of those

1

Executive Director, NH Musculoskeletal Institute, Manchester, NH. Assistant Professor, Franklin Pierce College, Division of Graduate and Professional Studies, Physical Therapy Department; Physical Therapist, Rehabilitation Services of Concord Hospital, Concord, NH. 3 Director, Apple Therapy Services, Manchester, NH. 4 Associate Professor, Bridgewater State College, School of Movement Arts, Health Promotion and Leisure Studies, Bridgewater, MA. Address correspondence to Laura C. Decoster, NH Musculoskeletal Institute, 35 Kosciuszko Street, Manchester, NH 03101. E-mail: [email protected] 2

Journal of Orthopaedic & Sports Physical Therapy

reviewed in this paper, have shown at least a temporary increase in range of motion associated with hamstring stretching. However, the numerous stretching methodologies proposed in the literature make it difficult for clinicians to identify the best clinical strategies for improving hamstring flexibility. This fact establishes a need to systematically summarize available evidence in an attempt to determine the most effective stretching activities for improving hamstring muscle flexibility. Methodological analysis of available evidence can help clinicians determine the relative effectiveness of interventions.6 The increasing popularity of such systematic reviews is directly related to the fact that they address specific clinical questions, use strict search techniques to identify pertinent articles, provide methodological scoring of the selected articles, and provide clinicians with quick access to vast amounts of literature within a specific content area.21 The purpose of this systematic review was to investigate the literature regarding the most effective positions, techniques, and durations of stretching to improve flexibility of the hamstring muscles in asymptomatic populations. 377

REVIEW

Study Design: Systematic literature review. Objective: Investigate the literature regarding the most effective positions, techniques, and durations of stretching to improve hamstring muscle flexibility. Background: Hamstring stretching is popular among physical therapists, athletic trainers, and fitness/coaching professionals; however, numerous stretching methodologies have been proposed in the literature. This fact establishes a need to systematically summarize available evidence in an attempt to determine the most effective stretching approach. Methods: A list of 28 pertinent manuscripts that included randomized and clinical trials was created according to specific inclusion/exclusion criteria. These manuscripts were critically reviewed for quality according to the Physiotherapy Evidence Database (PEDro) (10-point) scale and descriptive information about the stretching parameters employed in the research. Results: Cumulatively, 1338 healthy subjects were included in the reviewed studies. Methodological quality scores ranged from 2 to 8 (mean ± SD, 4.3 ± 1.6). Several methodological flaws were frequently recognized, including failure to conceal group allocation or perform blinded assessment. All studies reported improvements in range of motion after stretching. Conclusions: Overall, methodological quality was poor, with only 21.4% (6/28) of studies achieving a score between 6 and 8. Thus it was difficult to confidently identify 1 most effective hamstring stretching method. Instead, the evidence appears to indicate that hamstring stretching increases range of motion with a variety of stretching techniques, positions, and durations. J Orthop Sports Phys Ther 2005;35:377-387.

LITERATURE

Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at on January 15, 2017. For personal use only. No other uses without permission. Copyright © 2005 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

Laura C. Decoster, ATC 1 Joshua Cleland, DPT, OCS 2 Carolann Altieri, PT 3 Pamela Russell, PhD 4

METHODS

TABLE 1. Criteria used to determine study eligibility. Inclusion Criteria

Inclusion and Exclusion Criteria •

This systematic review included both single-group and nonrandomized designs for 2 reasons. First, in content areas where randomized controlled trials are lacking, the omission of other relevant clinical trials could significantly compromise the validity of the results of a systematic review.7 Further, it has been well documented1-4,8,9,14,22,23,27,30,37,39 that control groups not performing hamstring stretching do not demonstrate a significant increase in hamstring flexibility, so the use of a comparison group was not considered mandatory for inclusion in this review. A complete list of eligibility criteria is detailed in Table 1.

Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at on January 15, 2017. For personal use only. No other uses without permission. Copyright © 2005 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

Search Strategy Articles satisfying the eligibility criteria for this study were identified through an extensive search of Medline (1966-November, 2004), the Cumulative Index of Nursing and Allied Health (1982-November, 2004), SPORTDiscus (1949-November, 2004), and Embase (1988-November, 2004). Key words utilized for searches were hamstring stretching, lower extremity stretching, contract-relax stretching, ballistic stretching, static stretching, range of motion, flexibility, knee, hip, skeletal muscle, and proprioceptive neuromuscular facilitation (PNF). Abstracts identified through the above search methods were briefly reviewed by one of the authors to determine if the articles satisfied the inclusion and exclusion criteria. Articles with abstracts that appeared to satisfy the eligibility criteria were retrieved. Articles were also retrieved in circumstances where it could not be determined from the abstract if the article met the a priori eligibility requirements. Once the full articles were retrieved, they were independently reviewed by 2 authors to determine if they satisfied inclusion and exclusion criteria. In circumstances where the 2 reviewers disagreed regarding the eligibility of an article, a formal discussion took place until a consensus was reached. If a consensus could not be reached, a third author, designated as an arbitrator, was prepared to facilitate the decision regarding the eligibility of an article. At no time was the arbitrator called upon for this task, as a consensus regarding each selected and rejected article was always reached between the first 2 authors. If an abstract was identified but no full-length document was readily available, an attempt was made to contact 1 of the authors of the abstract to determine if a full manuscript had been published or was in press. If a published article existed, it was retrieved and evaluated for eligibility criteria in the systematic review. Abstracts were not included in this review. 378

• • • •

Experimental (randomized controlled trials) and quasiexperimental (ie, pretest/posttest cohort) studies Intervention includes common/clinically used hamstring stretching Outcome measures include range of motion at the knee or hip Subjects aged 14 to 60 y Subjects healthy (no orthopedic or neurologic issues that would affect the ability to gain range of motion) Exclusion Criteria

• • • •

Non-English language studies Outcome not in (or not convertible to) degrees (ie, sit and reach) Instrumented stretching Abstract and unpublished data

Once all articles were obtained, their reference lists were manually searched to identify potentially eligible publications that were not identified during the database searches. These articles were then retrieved and underwent the same examination process as described above. After the final list of articles was agreed upon, data abstraction and methodological scoring of the articles ensued.

Data Extraction A list of variables of interest to practicing physical therapists and athletic trainers was developed a priori. Authors generated independent lists and then met and discussed the relevance of each item and determined a final list as a group. These variables included subject description, type of stretching (static, ballistic, dynamic, contract-relax), position (standing, sitting, supine), duration of stretch, number of repetitions, number of stretching sessions per week, total number of stretching sessions, length of follow-up, method of assessing hamstring length, range of motion change scores, use of prestretch modalities and/or warm-up, and statistical differences among stretching groups. A standardized form (Figure) was developed with all variables listed to facilitate accurate data abstraction and comparison.35 The aforementioned variables were abstracted from all articles independently by the first 3 authors. Once all data were abstracted, the 3 authors met and reviewed each item for all articles to assure unanimous agreement upon the variables abstracted from each article. Articles that were missing variables simply received a not available (NA) remark for that item. If the first 3 authors could not reach consensus, the fourth author was prepared to serve as an arbitrator to facilitate the final decision regarding the variable. This method of data abstraction has been advocated for use in systematic reviews by the Cochrane Collaboration.35 J Orthop Sports Phys Ther • Volume 35 • Number 6 • June 2005

PEDRo 1. Eligibility criteria specified (yes/no) ................................................ 2. Subjects randomly allocated to groups (yes/no) ............................ 3. Allocation was concealed (yes/no)................................................. 4. Groups similar at baseline (yes/no) ............................................... 5. Subjects were blinded to group (yes/no)........................................ 6. Therapists who administered therapy were blinded (yes/no) ........ 7. Assessors were blinded (yes/no) ................................................... 8. Minimum 85% follow-up (yes/no) ................................................... 9. Intent to treat analysis for at least one key variable (yes/no)......... 10. Results of statistical analysis between groups reported (yes/no) .. 11. Point measurements and variability reported (yes/no) .................. Total Score :

LITERATURE REVIEW

Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at on January 15, 2017. For personal use only. No other uses without permission. Copyright © 2005 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

Subjects n ............................................................................................................ Gender (male/female)........................................................................... Age (mean and standard deviation) ..................................................... Groups (description and number)......................................................... Stretching Parameters Duration per stretch (seconds) and protocol (days) ............................. Frequency (sessions per week)............................................................ Repetitions (number per session)......................................................... Position (standing, supine, seated) ...................................................... Supervised? (yes/no)............................................................................ Technique/s (contract-relax, static) ...................................................... Associated modalities (which and when).............................................. Warm-up (yes/no and type) .................................................................. Outcomes Measurement (eg, active or passive knee extension or hip flexion) .... Length of benefit (eg, minutes, hours, days) ........................................ Time between last stretch and measurement....................................... Miscellaneous Comments

Figure. Investigators used this form to record data about each study.

Assessment of Methodological Quality The Physiotherapy Evidence Database (PEDro) scale was used to evaluate the methodological quality of all selected articles. The PEDro scale is an 11-item scale designed to evaluate the methodological quality of clinical trials (Figure).19 All items in the list were to be scored as ‘‘yes,’’ ‘‘no,’’ or ‘‘don’t know.’’ Those items scoring a yes generated 1 point toward each article’s final score. Those items that received a no or don’t know received a zero score for that item. The total score for each article was calculated in accordance with published guidelines for the use of the PEDro scale. The first item was not included in the score, as it concerns the external validity of the study.19 Thus, only items 2 through 11 were used to score the methodological quality of the study as they concern a study’s internal validity. Potential scores ranged from 0 to 10, with 10 indicating the highest possible methodological quality. Although the PEDro scale is most commonly utilized for the scoring of randomized clinical trials, it was also used in this study to score the single-group pretest/posttest trials J Orthop Sports Phys Ther • Volume 35 • Number 6 • June 2005

to allow direct comparison of methodology between both study designs. Based on previous research, the PEDro scale’s interrater reliability is 0.55 (95% CI, 0.41-0.72).19 To ensure identical preparatory training, all of the assessors underwent training procedures in the PEDro scoring system, according to the Center for Evidence-Based Physiotherapy at the University of Sydney. PEDro scale scoring of all selected articles was independently performed by 3 authors. Once the articles had been independently scored, the 3 assessors met to discuss the determined scores for each article. Every item in the PEDro list was discussed and each assessor stated if the assessor gave the article a yes, no, or don’t know score. If consensus between all 3 assessors did not exist, the group immediately performed a detailed search for the criterion at issue. If consensus could not be reached after the group assessment, the fourth assessor was available to serve as an arbitrator to facilitate a final decision regarding the scoring of a particular item. In all cases, the first 3 authors were able to reach a 379

TABLE 2. Demographics by study. All reviewed studies presented in alphabetical order by first author. Number, gender, and age of subjects are detailed.

Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at on January 15, 2017. For personal use only. No other uses without permission. Copyright © 2005 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

Study Bandy and Irion1 Bandy et al2 Bandy et al3 Chan et al4 Cipriani et al5 DePino et al8 de Weijer et al9 Funk et al10 Halbertsma and Goeken12 Halbertsma et al13 Hartig and Henderson14 Hubley et al16 Moller et al20 Nelson et al22 Osternig et al23 Prentice26 Reid and McNair27 Roberts and Wilson28 Rowlands et al29 Spernoga et al30 Sullivan et al31 Taylor et al33 Wallin et al36 Webright et al37 Wiemann and Hahn39 Wiktorsson-Moller et al40 Willy et al41 Worrell et al43

n 57 58 93 40 23 (46 legs) 30 56 30 18 (dropped to 14) 16 298 30 8 69 30 46 43 24 43 (dropped to 37) 30 20 24 47 40 69 8 24 (dropped to 18) 19

Gender (Male, Female) 40, 17 41, 17 61, 32 24, 16 18, 5 30, 0 28, 28 30, 0 8, 10 10, 6 298, 0 Not stated 8, 0 69, 0 16, 14 Not stated 43, 0 19, 5 0, 43 30, 0 10, 10 12, 12 47, 0 22, 18 69, 0 8, 0 12, 6 10, 9

Age (y)* 26.11 (5.26) 26.21 (5.57) 26.24 (5.13) 20 (3) 22.8 (4.7) 19.8 (5.1) 18-42 18-22 26.5 24.6 (3.27) 20 14-60 Not stated 16.45 (.96) 24.6 (5.01) 18-34 15.8 (1) 20.5 (1.35) 20 (1.3) 18.8 (.063) 26.7 (2.42), 24.5 (1.37) 25.46 19-32 21.5 (2.8), 21.3 (3.6) 20-34 Not stated 21 25.7 (2.4), 26.7 (4.8)

* Data are present either as mean (± SD) or range.

consensus without the participation of the fourth author. Following this process, the total PEDro score for all articles was calculated. The mean score and measures of variability were then calculated for the entire group of articles that were included in the systematic review.

RESULTS The results of the search revealed 28 articles that satisfied the inclusion criteria. Cumulatively, 1338 subjects were included in the reviewed studies. All of the studies investigated the effects of hamstring stretching on healthy individuals only. Of the articles selected, 15 (54%) incorporated both male and female subjects,1-5,8,12,13,23,28,31,33,37,41,43 10 studies included only males,9,10,14,20,22,27,30,36,39,40 1 study included only females,29 and 2 other studies did not specify the gender of their subjects.16,26 The mean age of the subjects in the studies could not be determined for the overall group because numerous studies either did not provide ages20,40 or simply gave a range of ages without noting the mean.8,10,16,26,36,39 A listing of studies with data about the number, age, and gender distribution of subjects is presented in Table 2. Significant heterogeneity existed among the 380

studies relative to the stretching protocols investigated, and methodological quality varied widely.

Methodological Quality The overall range of methodological quality of the studies was 2 to 8, with a mean of 4.3 (SD, 1.6). Despite the relative heterogeneity of the studies selected for this review, a number of methodological criteria were consistently satisfied for the majority of studies. The most frequently satisfied criteria in the selected articles were the reporting of point measurements and variability (93%), followed by the reporting of results of statistical analyses among groups (86%), and the use of randomization (82%). Several methodological flaws were recognized among the studies, including failure to conceal group allocation, to blind subjects and therapists to groups, or to perform blinded assessment. The percentage of studies that satisfied each specific methodological scoring criterion can be found in Table 3. Table 4 lists all 28 articles selected for this review in order according to methodological quality, from highest to lowest. Further information in this table includes identification of outcome variable, protocol specifics (duration, frequency, repetitions, position), and data relative to the range of motion gains for the J Orthop Sports Phys Ther • Volume 35 • Number 6 • June 2005

TABLE 3. Number and the percentage of studies meeting each of the PEDro criteria. PEDro Criteria 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.

Eligibility criteria specified (yes/no) Subjects randomly allocated to groups (yes/no) Allocation was concealed (yes/no) Groups similar at baseline (yes/no) Subjects were blinded to group (yes/no) Therapists who administered therapy were blinded (yes/no) Assessors were blinded (yes/no) Minimum 85% follow-up (yes/no) Intent to treat analysis for at least 1 key variable (yes/no) Results of statistical analysis between groups reported (yes/no) Point measurements and variability reported (yes/no)

group experiencing the largest gain in each study. A brief description of the comparison groups used in all studies is listed as well.

Percent Meeting Criterion (%)

25 23 0 16 0 1 5 16 10 24 26

89% 82% 0% 57% 0% 4% 18% 57% 36% 86% 93%

ences were noted between groups. No significant differences (P = .08) were noted between pretest and posttest measurements for either the static or PNF group in the Worrell et al43 study, possibly as a result of low power.

groups by stretching position. The data for all groups from each study are compiled in this table, therefore the number of studies reported do not equal the total number of studies in this review. This table demonstrates that all stretching positions resulted in considerable range of motion gains, as compared to control groups (no stretching), which showed relatively minimal change (0.01°) in range of motion.

Effect of Stretching Technique on Hamstring Flexibility Gains Data were further analyzed relative to technique utilized (static, PNF, ballistic) for all groups (not solely the group with the best outcome) and compared in Table 6. The table reveals that the groups performing static stretching achieved greater mean range of motion gains than groups performing proprioceptive neuromuscular facilitation techniques. Of the 28 studies included in this systematic review, only 4 studies2,26,39,43 directly compared the effects of different stretching techniques (Table 7). Bandy et al2 demonstrated that static stretching produced more than double the range of motion gains compared to dynamic stretching. Prentice et al26 demonstrated that both static and slow-reversal-hold stretching produced significant increases in range of motion compared to baseline. In addition, there was a significant difference between groups, with the slow-reversal-hold group experiencing a 12° gain, while the static group gained only 9°. Wiemann et al39 demonstrated that both static and ballistic stretching resulted in significant range of motion gains over baseline; however, no significant differJ Orthop Sports Phys Ther • Volume 35 • Number 6 • June 2005

Four studies1,3,5,28 directly compared the duration of stretching between groups (Table 8). Bandy and Irion1 determined that stretching durations of 30 and 60 seconds resulted in significantly greater range of motion improvements than a 15-second stretch. Because there was no statistically significant improvement when stretching time was increased from 30 to 60 seconds, the authors concluded that 30 seconds was ideal. Bandy et al3 reported that stretching 1 time daily for 30 seconds was just as effective as stretching once per day for 60 seconds or 3 times per day for 30 or 60 seconds. In the study conducted by Cipriani et al,5 stretching for 10 seconds 6 times was shown to be equally as effective as stretching 2 times for 30 seconds over 3- and 6-week periods. Roberts and Wilson28 demonstrated that a 15-second stretch repeated 3 times was more effective than a 5-second stretch repeated 9 times.

Effect of Prestretching Warm-up on Hamstring Flexibility Gains Finally, only 2 studies were designed to determine the effectiveness of prestretching activity or modality use (Table 9). DeWeijer et al8 compared stretching with and without warm-up, to warm-up alone, and to a control group. While both stretching groups achieved significant hamstring flexibility gains, neither of the nonstretching groups made gains. Taylor et al33 compared stretching alone to stretching preceded by either moist heat or cold application. All 3 groups demonstrated comparable range of motion improvements with no statistically significant difference among conditions. 381

REVIEW

Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at on January 15, 2017. For personal use only. No other uses without permission. Copyright © 2005 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

Effect of Stretching Duration on Hamstring Flexibility Table 5 shows mean range of motion gains for all Gains

LITERATURE

Effect of Stretching Position on Hamstring Flexibility Gains

Number Meeting Criterion (n)

382

J Orthop Sports Phys Ther • Volume 35 • Number 6 • June 2005

5 5

5

Cipriani et al5 Sullivan et al31

Bandy et al3

10.0 8.5 33.6

6.0 12.04 9.0 8.9 7.0 5.3 5.2

5

4

4 4

4

4 4

3 2 2

2 2

2 2

Roberts and Wilson28 Reid and McNair27 Wallin et al36 Rowlands et al29

Willy et al41

30 Spernoga et al 8 DePino et al

Moller et26 al20 Prentice Wiktorsson-Moller et al40 Halbertsma et al13 Hartig and Henderson14 Osternig et al23 Halbertsma and Goeken12 KE SLR

SLR KE

SLR SLR SLR

KE KE

KE

SLR SLR

KE

KE

KE KE

KE

SLR KE

KE SLR

KE KE

KE KE KE SLR

KE

Outcome

Supine Seated

Standing Standing

Standing Supine and Seated Standing with APT Supine Standing with APT Standing Supine Not stated

Standing

Seated Standing with APT Not stated

Standing Standing with APT Standing

Seated Not Stated

Seated Standing

Standing Standing Seated Not clear

Supine

Position

PNF PNF

Static Static

PNF PNF PNF

PNF Static

Static

PNF PNF

Static

‘‘Static active’’

Static PNF

Static

Static Static/PNF

Static Static

Static Static

Static Static Active Ballistic

Static

Technique

× × × ×

30 s 30 s 30 s 15-s reps

5/wk, 6 wk

7/wk, 6 wk 4/wk, 2 wk

1/wk, 2 wk Single session

1/wk, 3 wk 3/wk, 6 wk

5/wk, 6 wk 5/wk, 6 wk 2/d, 6 wk Single session

Single session

Protocol

× 19-s cycles 3/wk, 4 wk × 20-s cycles 2/wk, 6 wk 2 × 30 s 5/wk, 6 wk

5/wk, 6 wk

3/wk, 5 wk

1 × 80 s 1 × 10 min

10 × 30 s 5 × 30 s

Single session 2/d, 4 wk

Single session 4/d, 7 wk

5 × 16-s cycles Single session 3 × 20-s cycles 3/wk, 10 wk 6 × 16-s cycles Single session

5 × 26-s cycles Single session 4 × 30 s Single session

5 3 × 2

3 × 30 s

3 × 15 s

5 × 30 s 3/wk, 8 wk 4 × 20-s cycles 5/wk, 3 wk

1 × 30 s

6 × 10 s 1 × 30 s

3 × 30 s 15 min

1 × 60 s 1 × 30 s

1 1 1 3

3 × 30 s

Duration

Active No

No No

Active Not stated Active

Active Active

Active

No Active

No

Active

Active No

No

No Active

No No

Heat No

No No Active Active

Active

Warm-up

Static, control Control

Control Control

No comparison Static No comparison

Control Control

Ballistic str Various protocols, control Control

Control

9 × 5 s, control

3 × 30 s, 3 × 60 s, 1 × 60 s 4 wks, control Static × 15 s

15 s, 60 s, control Dynamic, control 15 s, control 15 s static, WU, control Ice + str, str Eccentric strength/ control Heat Cycling with and without str 2 × 30 s Posterior pelvic tilt

WU, str, control

Comparison Groups

Not stated Not stated

Immediate Not stated

90 min Not stated Immediate

0-32 min 1 min

Not stated

Not stated 1d

Not stated

Not stated

1d Not stated

2d

Not stated Same day

Not stated 15 min

Immediate 2d

0, .25, 1, 4, 24 h 2d 2d Not stated Immediate

Time Elapsed Before Poststretching Measurement

Abbreviations: APT, anterior pelvic tilt; KE, knee extension; PNF, proprioceptive neuromuscular facilitation; Reps, repetitions; ROM, range of motion; SLR, straight leg raise; str, stretch; WU, warm-up. * No preintervention measurement taken. † Stretching specifics not reported: ‘‘The stretching exercises used were taken from Anderson’s book on stretching.’’

7.8 7.0

12.3

8.5

5 5

11.2 9.5

11.5

28.0 13.0

Chan et al 43 Worrell et al

4

5 5

Funk et al10 Hubley et al16 N/A* 9

5.7 12.0

6 5

14.0 12.5 11.4 10.2 8.4

8

ROM Gain (deg)

6 6 6 6

1

PEDro Score

Bandy and Irion Bandy et al2 Webright et al37 Wiemann and Hahn39 Taylor et al33 Nelson et al22

de Weijer et al9

Study

TABLE 4. All 28 studies in order (highest to lowest) of methodological quality; protocol and gain data relate to stretching group with best outcome

Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at on January 15, 2017. For personal use only. No other uses without permission. Copyright © 2005 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

DISCUSSION To our knowledge, this is the first attempt to systematically summarize hamstring-stretching studies to identify the most effective positions, techniques, and duration of stretching for improving hamstring flexibility. We did identify other systematic reviews of hamstring-stretching research, but they were concerned with stretching to prevent injuries.34,38 Although we must acknowledge the low methodological quality of the majority of the 28 studies included in this review, the evidence does appear to indicate that hamstring stretching improves flexibility for at least the time elapsed between the last stretching session and measurement in each study (Table 2). Clearly, however, the quality of research must influence one’s clinical decisions.

As healthcare professions continue the paradigm shift toward evidence-based practice, it has become increasingly important to determine the most efficient and effective treatment techniques to achieve desired outcomes. Studies with better methodological quality would provide greater evidence as to which stretching techniques might be most advantageous to utilize in a clinical setting. Therefore, it should be recognized that studies demonstrating the greatest range of motion gains do not necessarily provide the greatest evidence in support of a particular stretching

Effect of Stretching Position on Hamstring Flexibility Gains Stretching position (standing, seated, supine) did not make a significant difference in the magnitude of

REVIEW

TABLE 5. This table breaks the studies down by outcome measurement used and shows the number of studies to use each stretching position. The range of range-of-motion gains and the mean for the highest-gain outcome groups is shown in the last 2 columns. Outcome Measurement (Position)

Studies (n)

Range of Mean Gains (deg)

Mean of Best Group Gains (deg)*

10 4 3 4 1 1 11

7.0-12.5 5.7-11.2 5.3-14.0 6.0-28.0 5.2 12.04 –2.9-3.0

10.60 9.00 9.00 14.30 5.20 12.04 0.01

Knee extension/standing Knee extension/seated Knee extension/supine Straight leg raise/standing Straight leg raise/seated Straight leg raise/supine Control * Mean of groups with highest gain in each study.

TABLE 6. This table breaks the studies down by stretching technique outcome measurement used and shows the number of studies to use each technique. The range of range-of-motion gains and the mean for the highest-gain outcome groups is shown in the last 2 columns. Stretching Technique (Outcome Measurement) Static (knee extension) Static (straight leg raise) PNF (knee extension) PNF (straight leg raise)

Studies (n)

Range of Mean Gains (deg)

12 3 2 5

5.7-12.5 8.9-28.0 5.3-7.8 5.2-33.6

Mean of Best Group Gains (deg)* 10.6 18.5 6.6 14.2

Abbreviation: PNF, proprioceptive neuromuscular facilitation. * Mean of groups with highest gain in each study.

J Orthop Sports Phys Ther • Volume 35 • Number 6 • June 2005

LITERATURE

Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at on January 15, 2017. For personal use only. No other uses without permission. Copyright © 2005 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

Methodological Quality

technique if they exhibited a low PEDro score. The average methodological score for all 28 articles was 4.3 out of a possible 10 points. It has been suggested that a 70% methodological score is essential to be considered a valid clinical trial.18 Just 1 of the 28 reviewed studies met this criterion.8 The failure of most of the studies to meet methodological quality criteria significantly impacts the validity of evidence that can be abstracted. Perhaps one of the most critical deficits noted was the lack of blinding of the data collection technician. It is understandable that in studies investigating the effects of a particular stretching technique, it would be difficult or impossible to blind the subjects and the instructing clinicians. However, the lack of blinding of the individuals assessing range of motion gains should be considered a significant methodological flaw. Without adequate blinding, the researcher’s expectation can result in the incorporation of bias into the outcomes.25 Surprisingly, only 51-4,8 of the 28 articles (18%) utilized blinded assessors. One additional study27 stated that the principal investigator was blind to subject allocation, but failed to state what that investigator’s role entailed so it was unclear whether he acted as an assessor.

383

TABLE 7. This table details the studies that directly compared stretching techniques and includes a brief description of groups and stretching protocols used by each, the range of motion gained by each group, and the authors’ conclusions.

Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at on January 15, 2017. For personal use only. No other uses without permission. Copyright © 2005 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

Author Year (Outcome PEDro Measurement) Score Bandy et al2 (KE)*

6

Prentice26 (SLR)*

2

Wiemann and Hahn39 (SLR)†

6

Worrell et al43 (KE)

5

Group 1 (N, Gain)

Group 2 (N, Gain)

1 × 30 s static, 6 × 5 s dystanding, namic knee 5/wk, 6 wk extension, (19, 11.42°)* 5/wk, 6 wk, supine (19, 4.3°)* 3 × 10 s static, 3 × 10 s, PNF, supine, supine, 3/wk, passive, 10 wk (23, 3/wk, 10 wk 12°) (23, 8.9°) 15 s of reps of 3 × 30 s × 3 3 ballistic different motions, static single sesstretches, sion (16, single ses8.4°) sion (14, 7.8°) 4 × 15-20 s, 4 × 20 s PNF, static, standing (19, standing, 9.5°) 5/wk, 3 wk (19, 8°)

Group 3 (N, Gain)

Group 4 (N, Gain)

Group 5 (N, Gain)

Author Conclusions

Control (no stretching) (20, 0.7°)

NA

NA

Group 1 gained more than twice as much range compared to group 2; Dynamic stretching must be questioned Both improved range but the PNF method was superior

Control (no stretching) (46, 3.4°)

NA

NA

12 reps, resistance training, single session (12, 2°)

15 m stationary cycling, single session (12, 4°)

Control (rest) (16, 1.9°)

Programs to increase flexibility should include both stretching and exercise

NA

NA

NA

No significant difference between techniques; high variability

Abbreviations: PNF, proprioceptive neuromuscular facilitation; reps, repetitions; SLR, straight leg raise. * Statistical significance found between groups and compared to baseline for stretching groups. † Statistical significance found between baseline and groups 1, 2, and 4.

range-of-motion improvements, although there is evidence suggesting that pelvic position during standing stretches is important.31 All stretching positions resulted in considerable range-of-motion improvements when compared to control groups (Table 5). The study29 that demonstrated the greatest overall range of motion improvements, significantly exceeding all of the other studies, with the exception of Cipriani5 (33.6° and 28°, respectively), who incorporated both sitting and standing hamstring stretches into a single protocol. Perhaps the combination of stretching positions helped maximize range of motion gains.

Effect of Stretching Technique and Duration on Hamstring Flexibility Gains The variety of stretching techniques (static, PNF, ballistic) and limited direct comparisons in the reviewed articles make it difficult to determine the most effective technique. In the 4 studies that compared the relative effects of stretching techniques,2,26,39,43 all techniques resulted in statistically significant increases in range of motion; however, no technique was consistently found to be superior. This suggests that all techniques may deliver range of motion improvements. 384

Stretching duration was the focus of 4 reviewed studies.1,3,5,28 It appears that a single 30-second stretching bout may be the most effective practice and that periods greater than 30 seconds are no more effective.1,3 However, studies have shown that stretching for shorter durations with increased repetitions can result in similar range of motion gains.5,28 Also, on the topic of stretching duration, some authors suggest that the total stretching time is more important than the number of repetitions.3,5,28 The length of a stretching protocol is also of interest to clinicians. Those studies investigating the effectiveness of stretching techniques over a longer period (weeks as opposed to a single session) achieved a greater increase in range of motion; however, those studies employing only a single session of stretching also made statistically significant range of motion gains.

Effect of Prestretching Warm-up on Hamstring Flexibility Gains Although only 2 studies set out to determine the effectiveness of prestretching activities, they were high-scoring studies. DeWeijer et al8 (PEDro score, 8) determined that physical activity prior to stretching did not improve the effectiveness of the stretch. Taylor et al33 (PEDro score, 6) concluded that J Orthop Sports Phys Ther • Volume 35 • Number 6 • June 2005

TABLE 8. This table details the studies that directly compared stretching duration and includes a brief description of groups and stretching protocols used by each, the range of motion gained by each group and the authors’ conclusions. Group 1 (N, Gain)

Group 2 (N, Gain)

Group 3 (N, Gain)

Group 4 (N, Gain)

Group 5 (N, Gain)

1 × 15 s static, standing, 5/wk, 6 wk (14, 3.8°)

1 x 30 s static, standing, 5/wk, 6 wk (14, 12.5°)

1 x 60 s static, standing, 5/wk, 6 wk (14, 10.9°)

Control NA (no stretching) (15, 0.27°)

Bandy et al3 (KE) 5

3 x 60 s static, standing, 5/wk, 6 wk (18, 10.5°)

3 x 30 s static, standing, 5/wk, 6 wk (19, 10.1°)

1 x 60 s static, standing, 5/wk, 6 wk (18, 10.5°)

1 x 30 s static, standing, 5/wk, 6 wk (18, 11.5°)

Cipriani et al5 (SLR)

5

6 x 10 s static, 2 x 30 s static, standing, standing, 7/wk, 3/6 wk 7/wk, 3/6 wk (23, 17.4/28°) (23, 14.5/ 24.2°)

NA

NA

Roberts and Wilson28 (KE)

5

9 x 5 s ‘‘static 3 x 15 s ‘‘static NA active’’, 3/wk, active,’’ 3/wk, 5 wk (8, 4.6°) 5 wk (8, 7.8°)

NA

30 sec and 60 sec significantly better than pretest and 15 sec, but not significantly different from each other; 30 sec is ideal Control 1 × 30 seconds (no stretching) is effective; no (20, 0.60°) significant difference with increased duration or daily frequency NA Both protocols had significant gains at both 3 and 6 weeks, with no difference between the protocols; the key was total daily stretch time NA 15-s stretch gained significantly more active knee extension than 5-s stretch

Abbreviations: KE, knee extension; SLR, straight leg raise.

prestretching application of modalities, either heat or cold, did not change the outcome of stretching activities.

Limitations One possible limitation of this systematic review is the fact that the authors who selected and assessed the articles were not blinded to author names, journal, and institutional affiliation, which has been identified as a source of possible bias.6 However, studies investigating the disclosure of author affiliation could not correlate unblinded assessment and reviewer bias.7,17 Another limitation is the exclusive use of articles written in English, making it possible to omit methodologically sound studies written in other languages. Also, the heterogeneous nature of the studies included did not allow for meta-analysis of J Orthop Sports Phys Ther • Volume 35 • Number 6 • June 2005

the results or calculation of effect sizes for comparison. Finally, it should also be noted that this systematic review only investigated the effects of hamstring stretching on asymptomatic individuals. Consequently, the results of the studies reported here may not apply directly to patient populations.

CONCLUSION The methodological weakness of the reviewed studies makes it difficult to confidently identify 1 most effective hamstring stretching method. Instead, the evidence appears to indicate that hamstring stretching increases range of motion with a variety of stretching techniques, positions, and durations. Future systematic reviews and studies should focus on determining the most efficient practice in symptomatic populations. 385

REVIEW

Bandy and Irion1 6 (KE)

Author Conclusions

LITERATURE

Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at on January 15, 2017. For personal use only. No other uses without permission. Copyright © 2005 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

Author Year (Outcome PEDro Measurement) Score

Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at on January 15, 2017. For personal use only. No other uses without permission. Copyright © 2005 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

TABLE 9. This table details the studies that directly compared prestretching warm-up and/or modalities and includes a brief description of groups and stretching protocols used by each, the range of motion gained by each group, and the authors’ conclusions. Author Year (Outcome Measurement)

PEDro Score

de Weijer et al9 (KE)

8

10 min stair climbing + 3 × 30 s static, supine (14, 14°)

10 min stair climb- 3 × 30 s static, ing (14, 1.2°) supine (14, 13.1°)

Control (14, -0.3)

Taylor et al33 (KE)

6

20 min moist heat, 1 × 60 s static, seated (7, 5.7°)

20 min cold, 1 × 60 s static, seated (9, 3.62°)

N/A

Group 1 (N, Gain)

Group 2 (N, Gain)

REFERENCES 1. Bandy WD, Irion JM. The effect of time on static stretch on the flexibility of the hamstring muscles. Phys Ther. 1994;74:845-850; discussion 850-842. 2. Bandy WD, Irion JM, Briggler M. The effect of static stretch and dynamic range of motion training on the flexibility of the hamstring muscles. J Orthop Sports Phys Ther. 1998;27:295-300. 3. Bandy WD, Irion JM, Briggler M. The effect of time and frequency of static stretching on flexibility of the hamstring muscles. Phys Ther. 1997;77:1090-1096. 4. Chan SP, Hong Y, Robinson PD. Flexibility and passive resistance of the hamstrings of young adults using two different static stretching protocols. Scand J Med Sci Sports. 2001;11:81-86. 5. Cipriani D, Abel B, Pirrwitz D. A comparison of two stretching protocols on hip range of motion: implications for total daily stretch duration. J Strength Cond Res. 2003;17:274-278. 6. Clarke M, Oxman A. Cochrane Reviewers’ Handbook 4.1. Available at: http://www.cochrane.org. The Cochrane Library; 2001. 7. De Bie R. Methodology of systematic reviews: an introduction. Phys Ther Rev. 1996;1:47-51. 8. DePino GM, Webright WG, Arnold BL. The efficacy of stretching for prevention of exercise-related injury: a systematic review of the literature. J Athl Train. 2000;35:56-59. 9. de Weijer VC, Gorniak GC, Shamus E. The effect of static stretch and warm-up exercise on hamstring length over the course of 24 hours. J Orthop Sports Phys Ther. 2003;33:727-733. 10. Funk D, Swank AM, Adams KJ, Treolo D. Efficacy of moist heat pack application over static stretching on hamstring flexibility. J Strength Cond Res. 2001;15:123126. 386

Group 3 (N, Gain)

1 × 60 s static, seated, (8, 4.1°)

Group 4 (N, Gain)

Author Conclusions Addition of warm-up exercise did not significantly increase range gain (note: this study also investigated length of benefit; range gains started to decline after 15 min) All yielded range improvement, but no significant difference identified between groups: adequate stretch can occur without thermal modality

11. Gleim GW, McHugh MP. Flexibility and its effects on sports injury and performance. Sports Med. 1997;24:289-299. 12. Halbertsma JP, Goeken LN. Stretching exercises: effect on passive extensibility and stiffness in short hamstrings of healthy subjects. Arch Phys Med Rehabil. 1994;75:976-981. 13. Halbertsma JP, van Bolhuis AI, Goeken LN. Sport stretching: effect on passive muscle stiffness of short hamstrings. Arch Phys Med Rehabil. 1996;77:688-692. 14. Hartig DE, Henderson JM. Increasing hamstring flexibility decreases lower extremity overuse injuries in military basic trainees. Am J Sports Med. 1999;27:173-176. 15. Herbert RD, Gabriel M. Effects of stretching before and after exercising on muscle soreness and risk of injury: systematic review. BMJ. 2002;325:468. 16. Hubley CL, Kozey JW, Stanish WD. The effects of static stretching exercises and stationary cycling on range of motion at the hip joint. J Orthop Sports Phys Ther. 1984;6:104-109. 17. Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials. 1996;17:1-12. 18. Labelle H, Guibert R, Joncas J, Newman N, Fallaha M, Rivard CH. Lack of scientific evidence for the treatment of lateral epicondylitis of the elbow. An attempted meta-analysis. J Bone Joint Surg Br. 1992;74:646-651. 19. Maher CG, Sherrington C, Herbert RD, Moseley AM, Elkins M. Reliability of the PEDro scale for rating quality of randomized controlled trials. Phys Ther. 2003;83:713-721. 20. Moller M, Ekstrand J, Oberg B, Gillquist J. Duration of stretching effect on range of motion in lower extremities. Arch Phys Med Rehabil. 1985;66:171-173. 21. Montori VM, Swiontkowski MF, Cook DJ. Methodologic issues in systematic reviews and meta-analyses. Clin Orthop Relat Res. 2003;43-54. J Orthop Sports Phys Ther • Volume 35 • Number 6 • June 2005

Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at on January 15, 2017. For personal use only. No other uses without permission. Copyright © 2005 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

35.

36. 37.

38. 39. 40.

41.

42. 43.

387

REVIEW

J Orthop Sports Phys Ther • Volume 35 • Number 6 • June 2005

34.

static stretch on hamstring muscle length. J Orthop Sports Phys Ther. 1995;21:283-286. Thacker SB, Gilchrist J, Stroup DF, Kimsey CD, Jr. The impact of stretching on sports injury risk: a systematic review of the literature. Med Sci Sports Exerc. 2004;36:371-378. van Tulder M, Furlan A, Bombardier C, Bouter L. Updated method guidelines for systematic reviews in the cochrane collaboration back review group. Spine. 2003;28:1290-1299. Wallin D, Ekblom B, Grahn R, Nordenborg T. Improvement of muscle flexibility. A comparison between two techniques. Am J Sports Med. 1985;13:263-268. Webright WG, Randolph BJ, Perrin DH. Comparison of nonballistic active knee extension in neural slump position and static stretch techniques on hamstring flexibility. J Orthop Sports Phys Ther. 1997;26:7-13. Weldon SM, Hill RH. The efficacy of stretching for prevention of exercise-related injury: a systematic review of the literature. Man Ther. 2003;8:141-150. Wiemann K, Hahn K. Influences of strength, stretching and circulatory exercises on flexibility parameters of the human hamstrings. Int J Sports Med. 1997;18:340-346. Wiktorsson-Moller M, Oberg B, Ekstrand J, Gillquist J. Effects of warming up, massage, and stretching on range of motion and muscle strength in the lower extremity. Am J Sports Med. 1983;11:249-252. Willy RW, Kyle BA, Moore SA, Chleboun GS. Effect of cessation and resumption of static hamstring muscle stretching on joint range of motion. J Orthop Sports Phys Ther. 2001;31:138-144. Worrell TW, Perrin DH. Hamstring muscle injury: the influence of strength, flexibility, warm-up and fatigue. J Orthop Sports Phys Ther. 1992;20:154-159. Worrell TW, Smith TL, Winegardner J. Effect of hamstring stretching on hamstring muscle performance. J Orthop Sports Phys Ther. 1994;20:154-159.

LITERATURE

22. Nelson RT, Bandy WD. Eccentric training and static stretching improve hamstring flexibility of high school males. J Athl Train. 2004;39:254-258. 23. Osternig LR, Robertson RN, Troxel RK, Hansen P. Differential responses to proprioceptive neuromuscular facilitation (PNF) stretch techniques. Med Sci Sports Exerc. 1990;22:106-111. 24. Pope RP, Herbert RD, Kirwan JD, Graham BJ. A randomized trial of preexercise stretching for prevention of lower-limb injury. Med Sci Sports Exerc. 2000;32:271-277. 25. Portney L, Watkins M. Foundations of Clinical Research: Applications to Practice. 2nd ed. Upper Saddle River, NJ: Prentice Hall Health; 2000. 26. Prentice WE. A comparison of static stretching and PNF stretching for improving hip joint flexibility. Athl Train. 1983;18:56-59. 27. Reid DA, McNair PJ. Passive force, angle, and stiffness changes after stretching of hamstring muscles. Med Sci Sports Exerc. 2004;36:1944-1948. 28. Roberts JM, Wilson K. Effect of stretching duration on active and passive range of motion in the lower extremity. Br J Sports Med. 1999;33:259-263. 29. Rowlands AV, Marginson VF, Lee J. Chronic flexibility gains: effect of isometric contraction duration during proprioceptive neuromuscular facilitation stretching techniques. Res Q Exerc Sport. 2003;74:47-51. 30. Spernoga SG, Uhl TL, Arnold BL, Gansneder BM. Duration of maintained hamstring flexibility after a one-time, modified hold-relax stretching protocol. J Athl Train. 2001;36:44-48. 31. Sullivan MK, Dejulia JJ, Worrell TW. Effect of pelvic position and stretching method on hamstring muscle flexibility. Med Sci Sports Exerc. 1992;24:1383-1389. 32. Taimela S, Kujala UM, Osterman K. Intrinsic risk factors and athletic injuries. Sports Med. 1990;9:205-215. 33. Taylor BF, Waring CA, Brashear TA. The effects of therapeutic application of heat or cold followed by