Acute effects of static passive vs contract- relax short-duration stretching on isometric and isokinetic performance in knee muscles: A single-group,pilot study

Abstract

BACKGROUND:

Research suggests that the effect of short bouts of stretching on muscle strength and ROM gains depends on the total stretching volume.

OBJECTIVE:

We assessed the effects of short bouts of static passive (SSS) and contract-relax (SCR) stretching on range of motion (ROM) and muscular strength.

METHODS:

Twenty volunteers performed two stretch protocols in a randomized order on two separate days, using the SSS and SCR techniques. Maximal ROM was tested prior to (pre-S), immediately after (post-I) and 10 minutes after (post-10) the stretch protocol. Maximal isometric voluntary contraction (MVC), isokinetic concentric peak moment (F-PM) and angle of peak moment (F-PMa) in the knee flexors, and concentric flexion/extension PM ratio (F/E) were measured and evaluated before (pre) and after (post) the stretch protocol.

RESULTS:

Both stretching techniques showed significant time-effects (pre vs post-I and post-10) but no interaction effects (time $\times$ techniques). There were small but significant effects between intervention-paired comparisons for F-PM, F-PMa and F/E ratios after SSS, and for F-MVC and F-PMa measures after SCR stretch techniques.

CONCLUSIONS:

Short-stretch techniques have trivial and small effects on loss of strength, increase the ROM for at least 10 min and slightly decrease the concentric F/E ratio. Using SSS could create some risk of hamstring injuries.

1. Introduction

Stretching immediately prior to exercise has been claimed to decrease results in performance tests that require isolated force or power, and during the warm-up period, stretching has been shown to decrease several muscular performance variables [1, 2, 3].

However, the dose-response relationship between stretching and decreased muscle strength is still unclear. In addition, the procedures carried out during experimental protocols are not normally representative of stretching bouts during normal athletic warm-up procedures, as the former are usually more time-consuming [4].

The observed impairment may be related to a number of factors. Among these, stretch-induced strength loss is dependent on the specific muscular activation used for measuring strength loss, the muscle length at which strength is measured, the stretch intensity and stretching technique applied [5, 6, 7, 8], and the duration of the stretch [9].

Recent studies suggest that short-duration static stretching (SSS) prior to strength testing does not result in any significant reduction in muscular strength when the duration of the stretch is $⩽$ 45 s [10, 11, 12, 13]. Impairment of performance seems to be related to the total duration of the stretch and is not related to the subject’s age, gender or fitness level [14]. Static stretching lasting less than 30 seconds seems to have no significant effect on force production [15, 16, 17] and explosive muscular actions of very short duration ( $<$ 100 milliseconds) were not significantly influenced by active static stretching compared to actions using maximal muscle strength [18].

Other reports showed that static stretching ( $<$ 90 s total) with a stretch intensity less than the point of discomfort (POD) creates little or no impairment of muscular performance [6, 11, 19]. For example, with regard to effects on maximal muscular performance, a literature search yielded 149 findings from 106 articles in which only 44% of the findings indicated significant reductions in maximal strength, power or speed-dependent performance [11]. The effect sizes and percentage changes associated with various durations of static stretching on force and isokinetic power were 0.004 and $-$ 0.5 for durations of 0–30 s and 0.62 and $-$ 4.7 for durations of 30–90 s respectively [6]. Counter movement jump (CMJ) performance remained unchanged throughout a short stretching (15 s) protocol [19].

It appears that stretches of short duration do not provide sufficient stimulus to elicit performance impairments following a dynamic warm-up [20] or when followed by dynamic stretching [21]. However, SSS protocols seem to increase the ROM and they could be as effective as longer duration stretches [22].

Contract-relax stretching is a proprioceptive neuromuscular facilitation technique (PNF) that involves an isometric contraction in the target muscles while in a stretched position, and upon relaxation, the stretch is either maintained or increased to a greater range of motion. The contract-relax stretch technique appears to generate greater or similar overall strength losses compared with static stretching [23, 24], possibly due to higher muscle relaxation and subsequently increased muscle compliance [27]. However, self-administered PNF of short duration (5 s) does not appear to change the maximal voluntary contraction force [25].

Stretching and maximal muscle performance relationships have recently been questioned in narrative reviews [6, 12] and one systematic review [11]. This review revealed that while some studies have reported significant losses in lower limb muscle groups, others did not, and that overall, 50% of the findings indicated that no detrimental effects on strength were likely when the static stretch duration was 30–45 s, with the pooled estimate of the changes ( $-$ 4.2 $\pm$ 2.7%) lying well within the normal variability for maximum voluntary performance. However, it was suggested that none of the studies used an appropriate statistical tool for combining and analysing individual study findings in a quantitative manner. Thus, the precise magnitudes of short duration stretch-induced acute changes in muscular performance and improved range of motion, and whether they are as effective as longer duration stretches, are still unknown and their effects are not yet fully delimited.

The aim of this study was to analyse the acute effects of similar short-duration stretches and stretch tolerance dosages in two stretch techniques – short-duration static passive (SSS) and short-duration contract-relax (SCR) – in terms of their effect on maximal ROM and isometric and isokinetic muscular strength in knee flexor and extensor muscles.

2. Methods

2.1 Partcipants

Twenty healthy recreationally active male sports science students (age 21.92 $\pm$ 2.9 years, height 1.75 $\pm$ 0.08 m, weight 71.56 $\pm$ 12.13 kg) participated in this study. No participants had any history of recent musculoskeletal injuries or neuromuscular diseases specific to the lower limbs that required surgery, or a history of lower limb injury within the previous 12 months. All participants were informed of the benefits and risks of the investigation prior to signing an institutionally approved informed consent document to participate in the study. Informed consent was given in accordance with the Declaration of Helsinki II and approved by the Ethics Committee of Pablo de Olavide University, Seville, Spain.

Figure 1.

Timeline with the assessments at baseline, intervention and post-test within an experimental session.

2.2 Experimental procedure

All participants carried out both experimental stretch protocols (separated by 24 h) and their intra- and inter-stretch techniques were compared. Each participant completed a 5 min warm-up at 50–60 W (between 60–70 RPM) using a stationary cycle ergometer (Monark Ergomedic 828E), a pre-ROM assessment, pre-stretching strength tests, the stretch protocol, an immediate post-ROM assessment, post-stretching strength tests and a 10 min post-ROM assessment (Fig. 1).

Peak moment, angle at peak moment in maximal isometric contraction, concentric maximal muscular voluntary exertion and conventional concentric knee flexor-knee extensor ratios at higher peak moment angles in the knee flexor and extensor muscles were measured using an isokinetic device (Cybex Norm, Lumex, Ronkonkoma) for strength tests. The maximal passive hip flexion angle-ROM (stretch test) was evaluated using a seated straight leg raise test (SSLR), and was evaluated prior to (pre-S), immediately after (post-I) and 10 min after (post-10) each stretch protocol (Fig. 1). Stretch techniques and use of the dominant or non-dominant leg were randomized for each subject. The stretch protocol was performed using the SSS stretch technique and 24 h later performed using the SCR technique with the other leg and in the same position and vice versa.

2.3 Instrumentation and measurements

2.3.1 ROM assessments

To estimate the passive resistive moment in hip flexion ROM (musculotendinous passive stiffness; MTS) we used a seated straight leg raise test (Fig. 2A). After the warm-up and prior to the first strength test, each participant was placed in the rest position and the same investigator applied the corresponding stretch protocol. To avoid pelvic tilt when stretch was applied to a leg, the other leg was flexed and fixed, and the back of the head, upper back and the points of the ischial bones were in contact with the wall and the floor respectively when the movement was executed. The leg being tested was stabilized and restrained to avoid any knee flexion when slowly raising the leg. To ensure a similar intensity of stretch, each participant indicated the correct maximal stretch position (by saying “stop”) when they reached the threshold for their individual POD. For the correct stretch amplitude position (similar intensity) each subject was placed in the same region using the Visual Analogic Scale pain ruler (VAS) (Schlenker Enterprises Ltd, Lombard, USA), independently of the ROM achieved in the previous set.

Figure 2.

Seated straight leg raise test (SSLR) (A) and stretch techniques intervention protocol (B). Wide black arrows show the positions selected to stabilize and restrain knee joint movement and avoid knee flexion when slowly raising the leg in both the SSLR and stretch protocols. Wide white arrow indicates the leg’s direction against resistance offered by the examiner in the SCR stretch technique.

The ROM angle was measured twice with an analogue inclinometer (Plurimeter-V, Dr Rippstein, Zurich, Switzerland), placed 15 cm from the kneecap, on the tibia. The plurimeter was set to 0 ${}^{\circ }$ when each subject was placed in the rest position and recorded the angle when the subject said “stop”. The mean of the two trials was used for all analyses. The ROM was also estimated using the vertical length from the malleolus peroneus to the floor, with a rigid rod perpendicular to the support surface graduated in millimetres giving a total range from 0 to 2100 mm, with a straight crossbar or spirit level (Lineal Flexometer, Stanley, France). We found a highly significant relationship between angle and vertical length in pre-S ( $r=$ 0.93; $p=$ 0.0005), post-I ( $r=$ 0.92; $p=$ 0.0005) and post-10 ( $r=$ 0.93; $p=$ 0.0005) measures.

2.3.2 Isokinetic and isometric test protocol

After the warm-up, maximal isometric voluntary contraction (MVC) and peak moment (PM) in isokinetic concentric (CON) mode were measured before each stretch protocol and 5 min after completing the stretching manoeuvre. The participants were seated on the dynamometer with the body stabilized by straps around the pelvis, chest, leg and ankle to avoid compensation. Because the strength tests were very simple in terms of neuromuscular coordination, the participants did not require a separate familiarization session. However, during the strength-test preparation time, they carried out one isometric attempt for 2 s while being shown their correspondence with the strength-isometric curve graph generated on a monitor, thus receiving not only verbal but visual feedback information. A 45 s rest was provided after this preliminary attempt, before beginning the measured strength-test. During the test the participants were instructed to keep their arms crossed in front of their chest. Participants were verbally encouraged to give maximal effort during the protocol and they were given visual and positive verbal feedback about the moment-time curve during individual contractions.

Isometric MVC flexion (F-MVC) was measured at 30 ${}^{\circ }$ of knee flexion and the participants performed two maximal repetitions for 3 s with a 10 s and 30 s rest between each repetition and each set respectively. After 1 min rest, the protocol then included 3 repetitions in concentric flexion-extension isokinetic mode at an angular speed of 60 ${}^{\circ }$ /s with a range of motion of 110 ${}^{\circ }$ flexion. The analysis included the concentric peak moment in flexion (F-PM) and the flexion/extension PM ratio was also calculated (F/E). The gravitational factor was calculated using a dynamometer and automatically compensated for during the measurements. The highest value achieved in all trials was used in the analyses.

2.3.3 Stretch technique dosage regime

The maximal passive hip flexion angle-ROM (stretch test) was evaluated using the SSLR test, and was evaluated prior to (pre-S), immediately after (post-I) and 10 min after (post-10) each stretch protocol. Stretch techniques and use of the dominant or non-dominant leg were randomized for each subject (Fig. 2B). The stretch protocol was performed using the SSS stretch technique and 24 h later performed using the SCR technique with the other leg and in the same position and vice versa. We used similar times under muscular tension, with two stretching techniques: SSS and SCR. The end of ROM was always increased using the same stretch tolerance dose to the point of discomfort (POD) with a VAS. Each stretching period lasted 3 min (1-min stretch-dose plus 2-min rest between sets) as measured with an MD1-003 Digital Stopwatch (Super A Scientific Co., Ltd, Zhejiang, China). A VAS using a 10 cm horizontal line with “no pain” and “worst pain possible” as the anchor points at each end, and a numbered scale from 0 to 10 on the reverse (hidden from the participant), with a clear PVC indicator slider, was used to assess participants’ stretch tolerance in left and right leg ROM tests [26]. The VAS score was collected immediately ( $<$ 10 s) following the stretch test.

2.3.4 Static passive stretch (SS) technique

From a seated straight leg raise position, the subject’s leg was elevated to a maximal passive hip flexion ROM with the knee straight, from rest to the individual’s POD, and held in this position for a specified period of time with a sustained force. In our protocol, the dose for the SSS was 4 sets $\times$ 15 s passive stretch/30 s rest, using assisted static stretch and passive recovery. Timing of the stretch commenced when the leg had reached the correct stretch position. Rest periods were provided at a neutral joint angle.

Table 1

Test-retest relative and absolute reliability in strength and stretch test variables

		Pre-S			Post-I			Post-10
Stretch technique	Variable	ICC	95% CI	CV	ICC	95% CI	CV	ICC	95% CI	CV
Static passive	Angle	0.992	0.981–0.997	4.4%	0.987	0.969–0.995	5.3%	0.995	0.988–0.998	3.3%
Contract-relax		0.985	0.964–0.994	5.8%	0.994	0.986–0.997	4.9%	0.995	0.988–0.998	4.0%
		Pre			Post
		ICC	95% CI	CV	ICC	95% CI	CV
Static passive	F-MVC	0.994	0.987–0.997	3.8%	0.985	0.967–0.993	4.9%
	E-MVC	0.98	0.954–0.991	5.3%	0.983	0.961–0.992	4.0%
Contract-relax	F-MVC	0.983	0.962–0.992	6.4%	0.998	0.973–0.995	4.1%
	E-MVC	0.982	0.960–0.992	5.5%	0.993	0.985–0.997	3.5%

Test-retest absolute reliability was measured by the standard error of measurement (SEM), which was expressed in relative terms through the coefficient of variation (CV), whereas relative reliability was assessed by the intraclass correlation coefficients (ICC) at 95% Confidence Interval (CI) calculated with the one-way random effects model. F-MVC and E-MVC: Maximal isometric voluntary contraction in Knee Flexors and Knee Extensors respectively.

2.3.5 Contract-relax proprioceptive neuromuscular facilitation technique

From a seated straight leg raise position, the subject’s leg was elevated to a maximal passive hip flexion ROM with the knee straight, from rest to the individual POD. In this position, the participant performed 4 sets of maximal isometric contractions for 3 s, lowering the leg against resistance offered by the examiner, relaxing for 2 s between attempts. Each set, using the SCR, involved 3 contract-relax cycles, and a rest period of 30 seconds was provided between sets (4 sets $\times$ 3 reps of (3 s-maximal isometric exertion and 2 s rest)/30 s). After each each individual contraction, a new ROM end-point was established using a contract-relax process of passively lengthening the muscle to the point of limitation and then performing the next isometric contraction.

2.4 Statistical analyses

Values are reported as mean $\pm$ SD. The normality of distribution of the measures was examined with the Shapiro-Wilk test and the homogeneity of variance across stretch techniques (SSS vs SCR) was verified using the Levene’s test. Test-retest absolute reliability was measured by the standard error of measurement (SEM) which was expressed in relative terms by the coefficient of variation (CV), whereas relative reliability was assessed by the intraclass correlation coefficients (ICC, 95% CI) calculated with the one-way random effects model for average measures. The SEM was calculated as the root mean square of total mean square intra-subject. Pearson’s test for correlation, a Wilcoxon paired samples test and one-way repeated measures ANOVA 2 $\times$ 3 (technique x time course) with Bonferroni corrections were used to compare dependent variables for each stretch technique and time point. The data were assessed for clinical significance using an approach based on the magnitudes of change. Effect size (Cohen’s d) was calculated in order to estimate the magnitude of the stretch technique effects on the selected ROM and strength variables using the following criteria: $>$ 0.2 (small), $>$ 0.5 (moderate) and $>$ 0.8 (large) for recreationally trained individuals [27].

The chances that the true values for each condition were beneficial/better, unclear or detrimental/worse for performance were calculated. The chances of beneficial/better or detrimental/worse effects were assessed qualitatively as follows: $<$ 1% almost certainly not; 1–5% very unlikely; 5–25% unlikely; 25–75% possible; 75–95% likely; 95–99% very likely; and $>$ 99% almost certain. If the chances of having beneficial/better or detrimental/worse effects were $>$ 5%, the true difference was assessed as unclear [28, 29]. Inferential statistics based on interpretation of magnitudes of effects were calculated using a custom-built spreadsheet for the analysis of controlled trials [30]. The significance level was set to a value of $p⩽$ 0.05. Statistical analyses were conducted using SPSS Statistics software version 18 (SPSS Inc., Chicago, IL).

3. Results

In this study, the acute effects of short stretches on isometric and isokinetic muscular strength were investigated in 20 participants using SSS and SCR stretch techniques. Table 1 shows the test-retest reliability of the scores. The ICC and CV values show an acceptable reliability in all measures.

In ROM assessment, a correlation in VAS scores was found between pre-S vs post-I ( $r=$ 0.65; $p<$ 0.001), pre-s vs post-10 ( $r=$ 0.63; $p<$ 0.001) and post-I vs post-10 ( $r=$ 0.74; $p<$ 0.001). There were no statistically significant differences in ROM values and VAS scores between legs for each subject or between stretch techniques at each point.

There was a high correlation between angle and vertical length for estimated ROM in the SSLR test (range between 0.86 and 0.97; $p<$ 0.01). We found no statistically significant differences in ROM values and VAS-scores between legs (dominant and non-dominant) or between stretch techniques.

Table 2

Differences in both static passive and contract-relax short-duration stretch techniques before and after each stretch and strength protocol

					Changes for post-I Vs pre-S			Changes for post-10 Vs pre-S
		Pre-S	Post-I	Post-10	% change	ES	% changes better/	% change	ES	% changes better/
Stretch technique	Variables	mean $\pm$ (SD)	mean $\pm$ (SD)	mean $\pm$ (SD)	mean $\pm$ (SD)	(95% CI)	trivial/worse effects	mean $\pm$ SD	(95% CI)	trivial/worse effects
Static passive	Angle ( ${}^{\circ }$ )	38.79 $\pm$ 13.33	46.58 $\pm$ 14.51 ${}^{***}$	44.63 $\pm$ 13.71 ${}^{***}$	23.2 $\pm$ 19.5	0.56 (0.33–0.79)	100/0/0 almost certain positive	17.7 $\pm$ 15.6	0.42 (0.24–0.60)	99/1/0 very likely positive
Contract-relax	Angle ( ${}^{\circ }$ )	42.27 $\pm$ 15.87	49.02 $\pm$ 21.93 ${}^{*}$	47.38 $\pm$ 21.99 ${}^{*}$	13.7 $\pm$ 13.6	0.31 (0.08–0.53)	83/17/0 likely positive	9.5 $\pm$ 13.6	0.41 (0.18–0.63)	96/4/0 very likely positive
					Changes pre-post
		Pre-Stretch	Post-Strecth		% change	ES	% changes better/
		mean $\pm$ (SD)	mean $\pm$ (SD)		mean $\pm$ (SD)	(95% CI)	trivial/worse effects
Static passive	F-PM (N.m)	114.84 $\pm$ 31.59	110.72 $\pm$ 30.35 ${}^{*}$		$-$ 3.0 $\pm$ 6.4	$-$ 0.13 ( $-$ 0.23–0.02)	0/92/8 likely trivial
	F-PMFa ( ${}^{\circ }$ )	27.40 $\pm$ 5.85	24.68 $\pm$ 5.38 ${}^{*}$		$-$ 8.0 $\pm$ 20.5	$-$ 0.45 ( $-$ 0.86–0.04)	0/11/89 likely negative
	F/E ratio	0.61 $\pm$ 0.07	0.59 $\pm$ 0.09 ${}^{*}$		$-$ 4.2 $\pm$ 7.3	$-$ 0.32 ( $-$ 0.55–0.08)	0/16/84 likely negative
Contract-relax	F-MVC (N.m)	138.90 $\pm$ 47.55	126.42 $\pm$ 35.24 ${}^{*}$		$-$ 6.8 $\pm$ 11.8	$-$ 0.25 ( $-$ 0.49–0.02)	0/32/68 likely negative
	F-PMa ( ${}^{\circ }$ )	28.75 $\pm$ 7.86	25.79 $\pm$ 7.84 ${}^{*}$		$-$ 9.2 $\pm$ 15.7	$-$ 0.36 ( $-$ 0.66–0.06)	0/13/87 likely negative

Intra-group significant differences with respect to pre-test: ${}^{*}p<$ 0.05; ${}^{**}p<$ 0.01; ${}^{***}p<$ 0.001; ES: Effect Size. Cohen’s d (Standardized differences) CI: 95% Confidence Interval; F-PM: Peak concentric moment production in knee flexors. F-PMa: Angle in higher peak concentric moment production in knee flexors. F-MVC: Maximal isometric voluntary contraction in knee flexors. No interaction effects (time x techniques) were found.

The results relating to the percentage decrease in passive resistance to stretch (estimated as ROM changes) and performance variables before and after stretch treatment are shown in Table 2. Intra-group differences were observed. The ROM showed statistically significant differences and increased from pre-S to post-I and post-10 in both SSS and SCR stretch techniques. Statistically significant differences were not found in the changes between post-I and post-10 in any SSS ( $p<$ 0.07) or SCR ( $p<$ 0.07) techniques. With regard to inter-group differences, the magnitude of the changes was statistically higher in SSS only in post-I. However, in the Time $\times$ Technique interaction statistically significant differences were not found, and no superiority in ROM gain can be assumed for any of the techniques. After 10 min there were no statistically significant differences in the magnitude of change between stretch techniques.

Statistically significant differences in pre-post stretch techniques were obtained in the following cases: F-PM, F-PMa and concentric F/E ratio ( $p<$ 0.05; small ES in all variables) after SSS; and F-MVC and F-PMa ( $p<$ 0.05; small ES in both cases) after applying the SCR stretch technique.

4. Discussion and conclusions

The aim of the present study was to examine the acute effects of two different short-duration stretching protocols on the isometric and isokinetic strength of knee flexor and extensor muscles.

In the present study, the final ROM increased by 23.2 $\pm$ 19.5% (moderate effect size) and by 17.7 $\pm$ 15.6% (small to moderate effect size) immediately and 10 min after SSS protocol stretching, respectively. These results, even with shorter stretches, are supported by studies that reported gains in stretching following passive resistance to stretch [7, 31] and reported that these gains remained for 5–30 min post-stretching [9, 32]. After SCR protocol stretching, final ROM increased by 13.7 $\pm$ 13.6% (small effect size) immediately after and by 9.5 $\pm$ 13.6% (small to moderate effect size) 10 min after the stretch protocol, which is in line with previous work [24, 31]. These positive changes (moderate and small ES in SSS and SCR stretch techniques respectively) suggest that short-duration static stretch and contract-relax stretch protocols enhance ROM immediately and for at least 10 min after the conclusion of the stretch protocol. These changes may return to baseline levels within 10 to 20 minutes, in line with results achieved with higher stretch-time doses [9, 32], but this was not measured. Statistically significant Time $\times$ Technique interaction differences were not found and the increases in ROM were not significantly different between SSS and SCR techniques. These results do not match those of other researchers, who found that increases in ROM were significantly greater immediately after PNF than after static stretch techniques [24, 31].

Although previous work suggests that the time-course for the decrease in musculo-tendinous stiffness may be dose-dependent for more practical stretching durations of 2 to 8 min [9], in this study, using only 3 minutes of stretching intervention for each technique, the relative magnitude of the decrease in passive resistance to stretch showed no differences between stretch techniques.

Stretch-induced strength loss in F-PM was found to be significant, with a decrease of 3.0 $\pm$ 6.4% after the SSS stretch protocol, but considering the trivial effect sizes observed, this difference is unlikely to be relevant or meaningful. For F-MVC, the Stretch-induced strength loss showed a significant decrease of 6.8 $\pm$ 11.8% after SCR stretch protocols but there was a small ES for negative changes. This is not in agreement with the observed negative effects of stretching on strength performance in previous studies with similar stretch durations [1, 33]. We found that strength loss was greater after SCR stretching, but the small ES shows that this difference is unlikely to be relevant.

Our findings contrast with studies that used static or contract-relax stretch techniques with similar or more time under tension and found no changes in musculo-tendinous stiffness, and no compromised maximal muscle performance [11, 15, 17]. Possibly the magnitude of change in ROM pre and post an SSS protocol that promotes a stretch-induced and likely trivial concentric strength loss could be due to a minor modification in the material behaviour of the aponeurosis-tendon system, by stress relaxation and/or plastic deformation, creating a more compliant muscle tendon unit or/and central nervous system inhibitory mechanism [34]. The stretch-induced isometric strength loss generated by a short SCR protocol could perhaps be explained more by fatigue due to specific isometric muscular activation. However, the magnitudes of relative changes in F-MVC and F-PM were relatively small in terms of effect size (likely negative and trivial, respectively). In addition, the isometric and isokinetic moment in the SSS protocol and the concentric isokinetic moment in the SCR stretch protocol were unaltered. In this study the acute effect of short stretching (15 s) and intervention (3 min total stretching protocol) on SSS and SCR stretching techniques appears to be a decrease in strength variables. However, the likely trivial effects of the change suggest similarity in the results of this study with others, which found no changes in maximal muscle strength, with similar or more time under tension in both stretch techniques.

It is likely that given the magnitude of the changes, after 10 minutes the stretch-induced isometric strength losses are overcome, as shown in studies of the gastrocnemius medialis muscle [32] and counter-movement height performance [35]. Previous research found that 4 $\times$ 30 s each with 30 s recovery at 50% POD [36] or moderate-intensity stretching [37] produced no significant changes in strength.

The stretch protocols generated an alteration in angle-moment relationships (length-tension relationships) with a decrease of 8.8 $\pm$ 20.5% and 9.2 $\pm$ 15.7% in F-PMa after SSS and SCR stretch protocols, respectively, indicating a small but significant change in ES.

If the F-PMa in both SSS and SCR protocols occurred earlier, a possible explanation could be that the stretching results in a slight rightward shift in the length-tension curve, suggesting that an acute lengthening of the muscle with stretching did occur, so that strength decreased at short muscle lengths and increased at long muscle lengths. This muscle length-dependent effect is consistent with previous work on static stretching [23, 26] and contract-relax stretching [23]. Thus, despite a small stretch-induced strength loss with both stretch protocols there seemed to be an increased extensibility of the musculotendinous unit, which may contribute to reducing the susceptibility to strain injury [38, 39].

However, our results show that the relative change in F/E ratio was significantly reduced by 4.2 $\pm$ 7.3% (small ES) with the SSS protocol and remained unaltered with the SCR stretch protocol. The SSS has a major effect on the moment-angle relationship during the total ROM, explained by the F-PM reduction produced at a lower angle (24.68 $\pm$ 5.38 to 27.40 $\pm$ 5.85; $p<$ 0.05). These statistically significant differences found in pre-post F-PM values after SSS could be the reason for the likely negative relative effects in the concentric F/E ratio, of less than 0.6. These adaptations could be risk factors for hamstring injuries [40, 41] and SCR stretching may be better.

These findings suggest that even static stretching and smaller time-stretch interventions may affect the conventional or mixed F/E ratios, as found in previous reports using greater time-doses of static stretch [19, 42]. However, the F-PM reduction is produced at a lower angle in SSS, which decreases the F/E ratio and could be a risk factor for hamstring injuries.

The present study had several methodological limitations. First, there was no control group. Given the small (but significant) changes between both protocols, there was no way to determine whether the changes occurred only because of the stretching protocols. Second, the decrease in musculo-tendinous stiffness was not measured, and was instead estimated indirectly from the decreased peak force and/or increased ROM. A decrease in viscoelastic properties promoted by stretching reduces musculo-tendinous stiffness and hence muscles’ capacity to produce force. It would have been interesting to re-measure the ROM at 15, 20, 25 and 30 minutes post-stretch to outline a precise time-course of the intervention-induced effects. Third, in our study, strength tests were only conducted twice (pre-post) and we did not measure the time to total restoration of baseline values. For the ROM, it would have been very useful to determine the evolution of latency in decrease-retain-increase effects. Fourth, ROM, stretch technique and performance should have been assessed for both legs in each subject, randomizing the dominant leg and the technique to obtain complete information for each subject. Fifth, the testing should have included sprint, squat and/or jump performance tests, to allow a comprehensive analysis of the specific influence of each stretch-dose and technique on realistic athletic performance.

In conclusion, the results of the present study show that both SSS and SCR stretching techniques, with short stretch-time doses, produced an immediate increase in ROM that was retained for at least 10 min. There was also a significant but likely trivial and likely negative loss of strength performance in F-PM and F-MVC with SSS and SCR stretch techniques, respectively. Both techniques slightly but significantly modified the angle-torque relationships but only SSS generated a likely relative effect on conventional isokinetic flexion/extension ratios. A decrease in the performance of isokinetic tests should not necessarily be extrapolated to indicate a decrease in athletic performance, especially when the magnitudes of changes are trivial or likely negative, as shown by a small effect size.