Acute effects of static, dynamic, and proprioceptive neuromuscular facilitation stretching on muscle power in women

The purpose of this study was to investigate the acute effects of 3 types of stretching-static, dynamic, and proprioceptive neuromuscular facilitation (PNF)-on peak muscle power output in women. Concentric knee extension power was measured isokinetically at 60 degrees x s(-1) and 180 degrees x s(-1) in 12 healthy and recreationally active women (mean age +/- SD, 24 +/- 3.3 years). Testing occurred before and after each of 3 different stretching protocols and a control condition in which no stretching was performed. During 4 separate laboratory visits, each subject performed 5 minutes of stationary cycling at 50 W before performing the control condition, static stretching protocol, dynamic stretching protocol, or PNF protocol. Three submaximal warm-up trials preceded 3 maximal knee extensions at each testing velocity. A 2-minute rest was allowed between testing at each velocity. The results of the statistical analysis indicated that none of the stretching protocols caused a decrease in knee extension power. Dynamic stretching produced percentage increases (8.9% at 60 degrees x s(-1) and 6.3% at 180 degrees x s(-1)) in peak knee extension power at both testing velocities that were greater than changes in power after static and PNF stretching. The findings suggest that dynamic stretching may increase acute muscular power to a greater degree than static and PNF stretching. These findings may have important implications for athletes who participate in events that rely on a high level of muscular power.     

A comparison of assisted and unassisted proprioceptive neuromuscular facilitation techniques and static stretching

A comparison of assisted and unassisted proprioceptive neuromuscular facilitation techniques and static stretching. J Strength Cond Res 26(5): 1238-1244, 2012-Proprioceptive neuromuscular facilitation (PNF) stretching often requires a partner. Straps are available allowing an individual to perform PNF stretching alone. It is not known if a strap provides similar improvements in the range of motion (ROM) as partner-assisted PNF or static stretching. The purpose of this study was to compare assisted and unassisted (with a strap) PNF stretching and static stretching. Hip joint ROM, reaction time (RT), and movement time (MT) were measured prestretching and poststretching. Thirteen recreationally active adults participated in this study. The participants were subjected to 5 different stretch interventions in a random order on separate days. Stretch conditions included unassisted PNF stretching using (a) isometric, (b) concentric, and (c) eccentric contractions with a stretch strap, (d) partner-assisted isometric PNF, and (e) static stretching. The RT, MT, dynamic, active, passive hip flexion angle, and angular velocity with dynamic hip flexion were measured before and after the intervention. The ROM improved (p < 0.05) 2.6, 2.7, and 5.4%, respectively, with dynamic, active static, and passive static ROM, but there was no significant difference between the stretching protocols. There was a main effect for time (p < 0.05) with all stretching conditions negatively impacting dynamic angular velocity (9.2%). Although there was no significant effect on RT, MT showed a negative main effect for time (p < 0.05) slowing 3.4%. In conclusion, it was found that all 3 forms of active stretching provided similar improvements in the ROM and poststretching performance decrements in MT and angular velocity. Thus, individuals can implement PNF stretching techniques with a partner or alone with a strap to improve ROM, but athletes should not use these techniques before important competitions or training because of the impairment of limb velocity and MT.     

The effects of proprioceptive neuromuscular facilitation and dynamic stretching techniques on vertical jump performance

The purpose of this study was to investigate the effects of 3 different warm-ups on vertical jump performance. The warm-ups included a 600-m jog, a 600-m jog followed by a dynamic stretching routine, and a 600-m jog followed by a proprioceptive neuromuscular facilitation (PNF) routine. A second purpose was to determine whether the effects of the warm-ups on vertical jump performance varied by gender. Sixty-eight men and women NCAA Division I athletes from North Dakota State University performed 3 vertical jumps on a Just Jump pad after each of the 3 warm-up routines. The subjects were split into 6 groups and rotated between 3 warm-up routines, completing 1 routine each day in a random order. The results of the 1-way repeated measures analysis of variance showed no significant differences in the combined (p = 0.927), men's (p = 0.798), or women's (p = 0.978) results. The results of this study showed that 3 different warm-ups did not have a significant affect on vertical jumping. The results also showed there were no gender differences between the 3 different warm-ups.     

Efficacy of static stretching and proprioceptive neuromuscular facilitation stretch on hamstrings length after a single session

A number of studies have investigated the efficacy of several repetitions of proprioceptive neuromuscular facilitation stretching (PNF) and static stretching (SS). However, there is limited research comparing the effects of a single bout of these stretching maneuvers. The aim of this study was to compare the effectiveness of a single bout of a therapist-applied 30-second SS vs. a single bout of therapist-applied 6-second hamstring (agonist) contract PNF. Forty-five healthy subjects between the ages of 21 and 35 were randomly allocated to 1 of the 2 stretching groups or a control group, in which no stretching was received. The flexibility of the hamstring was determined by a range of passive knee extension, measured using a universal goniometer, with the subject in the supine position and the hip at 90° flexion, before and after intervention. A significant increase in knee extension was found for both intervention groups after a single stretch (SS group = 7.53°, p < 0.01 and PNF group = 11.80°, p < 0.01). Both interventions resulted in a significantly greater increase in knee extension when compared to the control group (p < 0.01). The PNF group demonstrated significantly greater gains in knee extension compared to the SS group (mean difference 4.27°, p < 0.01). It can be concluded that a therapist applied SS or PNF results in a significant increase in hamstring flexibility. A hamstring (agonist) contract PNF is more effective than an SS in a single stretching session. These findings are important to physiotherapists or trainers working in clinical and sporting environments. Where in the past therapists may have spent time conducting multiple repetitions of a PNF and an SS, a single bout of either technique may be considered just as effective. A key component of the study methodology was the exclusion of a warm-up period before stretching. Therefore, the findings of efficacy of a single PNF are of particular relevance in sporting environments and busy clinical settings where time may be limited.     

The effect of static, ballistic, and proprioceptive neuromuscular facilitation stretching on vertical jump performance

The purpose of this study was to compare the acute effects of different modes of stretching on vertical jump performance. Eighteen male university students (age, 24.3 +/- 3.2 years; height, 181.5 +/- 11.4 cm; body mass, 78.1 +/- 6.4 kg; mean +/- SD) completed 4 different conditions in a randomized order, on different days, interspersed by a minimum of 72 hours of rest. Each session consisted of a standard 5-minute cycle warm-up, accompanied by one of the subsequent conditions: (a) control, (b) 10-minute static stretching, (c) 10-minute ballistic stretching, or (d) 10-minute proprioceptive neuromuscular facilitation (PNF) stretching. The subjects performed 3 trials of static and countermovement jumps prior to stretching and poststretching at 5, 15, 30, 45, and 60 minutes. Vertical jump height decreased after static and PNF stretching (4.0% and 5.1%, p < 0.05) and there was a smaller decrease after ballistic stretching (2.7%, p > 0.05). However, jumping performance had fully recovered 15 minutes after all stretching conditions. In conclusion, vertical jump performance is diminished for 15 minutes if performed after static or PNF stretching, whereas ballistic stretching has little effect on jumping performance. Consequently, PNF or static stretching should not be performed immediately prior to an explosive athletic movement.     

Impact of prior exercise on hamstring flexibility: a comparison of proprioceptive neuromuscular facilitation and static stretching

Position stands from the American College of Sports Medicine and the Surgeon General site a need for strategies capable of enhancing the effectiveness of stretching on flexibility and joint range of motion. One strategy for enhancing flexibility that has received anecdotal support but lacks substantial experimental evidence is the impact of prior exercise. This study compared 5 minutes of static stretching and proprioceptive neuromuscular facilitation (PNF) on hamstring flexibility performed with and without exercise. Forty undergraduate student-athletes participated in a repeated measure, counterbalanced experimental design. Within-group comparisons indicated that PNF resulted in a significant (p < 0.05) increase in flexibility after 60 minutes of exercise when compared with baseline (9.6%) and without exercise (7.8%). No differences were observed with static stretching across time. In addition, no differences were observed between the groups at any time point. Results demonstrated that PNF performed after exercise enhanced acute hamstring flexibility, and implementing a PNF stretching routine following exercise may augment current stretching practices among athletes.     

Relationship between two proprioceptive measures and stiffness at the ankle.

Previous research has investigated the role of proprioception and stiffness in the control of joint stability. However, to date, no research has been done on the relationship between proprioception and stiffness. Therefore, the purpose of this study was to determine the relationship between force sense, joint reposition sense, and stiffness at the ankle. A heterogeneous sample was obtained for this study; 20 of the 40 participants had a history of ankle sprains, and 13 of the 20 had been diagnosed by a physician (two mild ankle sprains, seven moderate sprains, four severe sprains). All subjects were asymptomatic and active at the time of the study. Active joint reposition sense was measured using a custom-built ankle goniometer, force sense was measured unilaterally and contralaterally with a load cell, and ankle muscle stiffness was measured via transient oscillation using a custom-built inversion-eversion cradle. We found no significant correlations between stiffness and joint reposition sense, with values of r ranging from 0.01 to 0.21. Significant correlations were found between stiffness and force sense. Specifically, contralateral force sense reproduction was significantly correlated to stiffness in the injured or "involved" ankle (r's ranging from 0.47 to 0.65; P< or =0.008). Whether the decreased ability to appropriately sense force (increased error) sends information to the central nervous system to increase muscle stiffness in response to an unexpected loss of stability, or whether these two phenomena function independently and both change concurrently as a result of injury to the system requires further investigation.     

Effects of ankle position during static stretching for the hamstrings on the decrease in passive stiffness

Static stretching is frequently performed to improve flexibility of the hamstrings, although the ankle position during hamstring stretching has not been fully investigated. We investigated the effects of ankle position during hamstring stretching on the decrease in passive stiffness. Fourteen healthy men performed static stretching for the hamstrings with the ankle dorsiflexed and plantar-flexed in a randomized order on different days. The hip was passively flexed to the maximum angle which could be tolerated without stretch pain with the knee fully extended; this was maintained for 5 min, with 1-min stretching performed in 5 sessions. Final angles and passive stiffness were measured before and after stretching. The final angle was defined as that formed by the tibia and horizontal plane when the knee was passively extended from hip and knee angles at 90° flexion to the maximum extension angle which could be tolerated without stretch pain. Passive stiffness was determined by the slope of torque–angle curve during the measurement of the final angle. The final angle significantly increased after stretching with the ankle dorsiflexed and plantar-flexed, whereas passive stiffness significantly decreased only after stretching with the ankle planter-flexed. The results suggest that passive stiffness decreases after stretching with the ankle planter-flexed but not after stretching with the ankle dorsiflexed, although the range of joint motion increases regardless of the ankle position during 5-min stretching for the hamstrings. These results indicate that static stretching should be performed with the ankle plantar-flexed when aiming to decrease passive stiffness of the hamstrings.     

Comparison of effects of static,proprioceptive neuromuscular facilitation and Mulligan stretching on hip flexion range of motion: a randomized controlled trial

The aim of this study was to compare the effects of static stretching, proprioceptive neuromuscular facilitation (PNF) stretching and Mulligan technique on hip flexion range of motion (ROM) in subjects with bilateral hamstring tightness. A total of 40 students (mean age: 21.5±1.3 years, mean body height: 172.8±8.2 cm, mean body mass index: 21.9±3.0 kg · m^-2) with bilateral hamstring tightness were enrolled in this randomized trial, of whom 26 completed the study. Subjects were divided into 4 groups performing (I) typical static stretching, (II) PNF stretching, (III) Mulligan traction straight leg raise (TSLR) technique, (IV) no intervention. Hip flexion ROM was measured using a digital goniometer with the passive straight leg raise test before and after 4 weeks by two physiotherapists blinded to the groups. 52 extremities of 26 subjects were analyzed. Hip flexion ROM increased in all three intervention groups (p<0.05) but not in the no-intervention group after 4 weeks. A statistically significant change in initial–final assessment differences of hip flexion ROM was found between groups (p<0.001) in favour of PNF stretching and Mulligan TSLR technique in comparison to typical static stretching (p=0.016 and p=0.02, respectively). No significant difference was found between Mulligan TSLR technique and PNF stretching (p=0.920). The initial–final assessment difference of hip flexion ROM was similar in typical static stretching and no intervention (p=0.491). A 4-week stretching intervention is beneficial for increasing hip flexion ROM in bilateral hamstring tightness. However, PNF stretching and Mulligan TSLR technique are superior to typical static stretching. These two interventions can be alternatively used for stretching in hamstring tightness.     

Maximal strength, number of repetitions, and total volume are differently affected by static-, ballistic-, and proprioceptive neuromuscular facilitation stretching

ABSTRACT: Barroso, R, Tricoli, V, dos Santos Gil, S, Ugrinowitsch, C, and Roschel, H. Maximal strength, number of repetitions, and total volume are differently affected by static-, ballistic-, and proprioceptive neuromuscular facilitation stretching. J Strength Cond Res 26(9): 2432-2437, 2012-Stretching exercises have been traditionally incorporated into warm-up routines before training sessions and sport events. However, the effects of stretching on maximal strength and strength endurance performance seem to depend on the type of stretching employed. The objective of this study was to compare the effects of static stretching (SS), ballistic stretching (BS), and proprioceptive neuromuscular facilitation (PNF) stretching on maximal strength, number of repetitions at a submaximal load, and total volume (i.e., number of repetitions × external load) in a multiple-set resistance training bout. Twelve strength-trained men (20.4 ± 4.5 years, 67.9 ± 6.3 kg, 173.3 ± 8.5 cm) volunteered to participate in this study. All of the subjects completed 8 experimental sessions. Four experimental sessions were designed to test maximal strength in the leg press (i.e., 1 repetition maximum [1RM]) after each stretching condition (SS, BS, PNF, or no-stretching [NS]). During the other 4 sessions, the number of repetitions performed at 80% 1RM was assessed after each stretching condition. All of the stretching protocols significantly improved the range of motion in the sit-and-reach test when compared with NS. Further, PNF induced greater changes in the sit-and-reach test than BS did (4.7 ± 1.6, 2.9 ± 1.5, and 1.9 ± 1.4 cm for PNF, SS, and BS, respectively). Leg press 1RM values were decreased only after the PNF condition (5.5%, p < 0.001). All the stretching protocols significantly reduced the number of repetitions (SS: 20.8%, p < 0.001; BS: 17.8%, p = 0.01; PNF: 22.7%, p < 0.001) and total volume (SS: 20.4%, p < 0.001; BS: 17.9%, p = 0.01; PNF: 22.4%, p < 0.001) when compared with NS. The results from this study suggest that, to avoid a decrease in both the number of repetitions and total volume, stretching exercises should not be performed before a resistance training session. Additionally, strength-trained individuals may experience reduced maximal dynamic strength after PNF stretching.     

Comparison of muscle activation levels during arm abduction in the plane of the scapula vs. proprioceptive neuromuscular facilitation upper extremity patterns

This study quantified activation of 8 muscles of the shoulder, trunk, and back during standing performance of (a) arm abduction in the plane of the scapula (scaption), (b) proprioceptive neuromuscular facilitation (PNF) diagonal 1 flexion (D1F), and (c) PNF diagonal 2 flexion (D2F) while lifting a dumbbell with the dominant hand. Twelve men (26.1 ± 4.4 years) and 13 women (24.5 ± 1.9 years) volunteered to participate. Electromyographic signals were collected with DE-3.1 double-differential surface electrodes at a sampling frequency of 1,000 Hz. Electromyographic signals were normalized to peak activity in the maximum voluntary isometric contraction (MVIC) trial and expressed as a percentage. One-way repeated-measures analysis of variance with Bonferroni corrections (α = 0.05) examined muscle activation patterns across the 3 conditions. For the middle trapezius, average activation was greater (p < 0.001) for D2F (70.5 ± 23.4% MVIC) than D1F (46.4 ± 19.6% MVIC). Lower trapezius average activation was greater (p < 0.001) for D2F (55.3 ± 23.8% MVIC) than D1F (40.1 ± 16% MVIC). The anterior deltoid showed greater activation (p = 0.009) for scaption (92.4 ± 26% MVIC) than D1F (74.4 ± 21.4% MVIC). The erector spinae showed greater activation for D2F (34.2 ± 12% MVIC; p < 0.001) and D1F (41.7 ± 21.4% MVIC; p < 0.001) than scaption (14.5 ± 12.3% MVIC). During D2F and scaption, all 6 muscles of the shoulder complex demonstrated very high activation levels (>60% MVIC) with the exception of the lower trapezius (55% MVIC). In contrast, erector spinae and external oblique muscles exhibited moderate activation (21-40% MVIC) during arm elevation. The 6 muscles of the shoulder complex displayed high to very high muscle activation at a level appropriate for strength training during all 3 exercise conditions.     

Mechanisms of calcium sequestration during facilitation at active zones of an amphibian neuromuscular junction

The calcium transients (Delta[Ca(2+)](i)) at active zones of amphibian (Bufo marinus) motor-nerve terminals that accompany impulses, visualized using a low-affinity calcium indicator injected into the terminal, are described and the pathways of subsequent sequestration of the residual calcium determined, allowing development of a quantitative model of the sequestering processes. Blocking the endoplasmic reticulum calcium pump with thapsigargin did not affect Delta[Ca(2+)](i) for a single impulse but increased its amplitude during short trains. Blocking the uptake of calcium by mitochondria with CCCP had little effect on Delta[Ca(2+)](i) of a single impulse but greatly increased its amplitude during short trains. This present compartmental model is compatible with our previous Monte Carlo diffusion model of Ca(2+) sequestration during facilitation [Bennett, M.R., Farnell, L., Gibson, W.G., 2004. The facilitated probability of quantal secretion within an array of calcium channels of an active zone at the amphibian neuromuscular junction. Biophys. J. 86(5), 2674-2690], with the single plasmalemma pump in that model now replaced by separate pumps for the plasmalemma and endoplasmic reticulum, as well as the introduction of a mitochondrial uniporter.     

Efficacy of two different stretch training programs (passive vs. proprioceptive neuromuscular facilitation) on shoulder and hip range of motion in older people

The aim of this study was to determine the influence of 2 methods of stretch training (passive and proprioceptive neuromuscular facilitation [PNF]) on range of motion (ROM) in older people between the age of 60 and 70 years over a period of 13 weeks. Fifty-four participants (39 women and 15 men) were divided into 3 groups: passive (n = 17; 66.5 ± 6.5 years), PNF (n = 17; age, 64.7 ± 4.0 years old), and control (n = 17; age, 66.4 ± 4.5 years). The subjects trained 2 times per week on nonconsecutive days for 13 weeks. Each training session included 2 flexibility exercises focused on the shoulder and hip joints. The PNF group performed 6 seconds of passive stretching, 3 seconds of muscular contractions, and 2 seconds of relaxation. The passive group performed 10 seconds of stretching and 5 seconds of relaxation. This sequence was repeated 3 times by each group. The control group did not perform any stretching. In the PNF group, there was an increase in hip ROM (p < 0.001) between pretest and posttest in the passive group and an improvement (p < 0.001) was observed between pretest and posttest, whereas in the control group, there was a significant decrease (p < 0.01) in hip ROM between pretest and posttest. In shoulder ROM, there was an increase (p < 0.001) between pretest and posttest in the passive group and an improvement (p < 0.001) was observed between pretest and posttest in the PNF group. There were no changes in shoulder ROM between pretest and posttest in the control group. The analysis of variance showed significant differences in hip and shoulder ROM between passive and control groups and PNF and control groups, but no significant differences were found between passive and PNF. The main finding was that the ability of physically active older people to increase ROM in response to stretching techniques is similar for both passive and PNF techniques.     

Joint stiffness of the ankle and the knee in running

The spring-mass model is a valid fundament to understand global dynamics of fast legged locomotion under gravity. The underlying concept of elasticity, implying leg stiffness as a crucial parameter, is also found on lower motor control levels, i.e. in muscle-reflex and muscle-tendon systems. Therefore, it seems reasonable that global leg stiffness emerges from local elasticity established by appropriate joint torques. A recently published model of an elastically operating, segmented leg predicts that proper adjustment of joint elasticities to the leg geometry and initial conditions of ground contact provides internal leg stability. Another recent study suggests that in turn the leg segmentation and the initial conditions may be a consequence of metabolic and bone stress constraints. In this study, the theoretical predictions were verified experimentally with respect to initial conditions and elastic joint characteristics in human running. Kinematics and kinetics were measured and the joint torques were estimated by inverse dynamics. Stiffnesses and elastic nonlinearities describing the resulting joint characteristics were extracted from parameter fits. Our results clearly support the theoretical predictions: the knee joint is always stiffer and more extended than the ankle joint. Moreover, the knee torque characteristic on the average shows the higher nonlinearity. According to literature, the leg geometry is a consequence of metabolic and material stress limitations. Adapted to this given geometry, the initial joint angle conditions in fast locomotion are a compromise between metabolic and control effort minimisation. Based on this adaptation, an appropriate joint stiffness ratio between ankle and knee passively safeguards the internal leg stability. The identified joint nonlinearities contribute to the linearisation of the leg spring.     

Effects of stretching on maximal anaerobic power: the roles of active and passive warm-ups

The purpose of the study was to provide practical suggestions on the effect of stretching on the maximal anaerobic power preceded by active or passive warm-up. To this aim, 15 relatively fit male subjects (age 23 +/- 0.2 years, height 177 +/- 2 cm, body mass 74 +/- 2 kg; [mean +/- SE]) randomly performed a series of squat jumps (SJ) and countermovement jumps (CMJ). Jumps were preceded alternatively by: i) passive stretching of lower limbs muscles; ii) active warm-up (AWU); iii) passive warm up (PWU); and iv) the joining of stretching with either active warm-up (AWU+S) or passive warm-up (PWU+S). In control conditions (C) only jumps were required. For the 2 jumps the flight time (Ft), the peak force (Pf), and the maximal power (Wpmax) were calculated. It resulted that Ft, Pf, and Wmax values were significantly higher: i) after AWU than after PWU and PWU+S in CMJ; and ii) in AWU as compared to those of other protocols of SJ. Stretching did not negatively affect the maximal anaerobic power, per se, but seems to inhibit the effect of AWU. The results suggested that AWU seemed to increase vertical jump performance when compared to PWU, presumably due to an increase in metabolic activity as a consequence of AWU, which did not occur in PWU, despite the same skin temperature. Passive stretching alone seemed not to negatively influence vertical jump performance, whereas, if added after AWU, could reduce the power output.     

The effect of proprioceptive training on neuromuscular function in patients with patellar pain

Eight patients with patellar pain underwent knee proprioceptive training. The maximal knee extension torque associated with the Vastus Lateralis EMG signal increased (p 0.001 and 0.039). Although muscle balance was not improved, all the patients improved their clinical scores.     

Structure and function of the ankle dorsiflexor muscles in young and moderately active men and women.

The aim was to investigate determinants of ankle dorsiflexor muscle (DF) strength and size in moderately active young men and women (n = 30; age 20-31 yr). Concentric (Con) and eccentric (Ecc) strength were measured isokinetically. Magnetic resonance imaging was used to determine the muscle cross-sectional area (CSA). Multiple biopsies were obtained from the tibialis anterior muscle to determine total numbers, areas (Area I and II) and proportions (Prop I and II) of type I and II fibers, respectively, and relative contents of myosin heavy chain (MHC) isoforms MHC1, MHC2a, and MHC2x. Women had lower Con and Ecc strength (24 and 27%; P < 0.01), smaller CSA (19%; P < 0.001), lower Ecc DF specific strength (strength/CSA) (10%; P < 0.01), and smaller Area I and Area II (21 and 31%; P < 0.01) than men. Prop I, MHC1, estimated total number of fibers, and Con DF specific strength were similar for both sexes. Con DF strength was up to 72% determined by CSA and Prop I, and Ecc DF strength was up to 81% determined by CSA, Prop I, and sex; variables other than CSA explained at most 9%. Body weight and fiber areas explained >50% of the variation in CSA. In conclusion, CSA was the predominant determinant of DF strength, CSA was to a great extent determined by the body weight and the sizes of muscle fibers, and sex differences in Ecc specific strength require further study.     

Effects of passive Bi-axial ankle stretching while walking on uneven terrains in older adults with chronic stroke

Many people with stroke experience foot drop while walking. Further, walking on uneven surfaces is a common fall risk for these people that hinder with their daily life activities. In addition, a few years after a stroke, lower-limb exercises become less focused, especially the ankle joint movement. The objective of this study is to determine the gait performance of older adults with chronic stroke on an uneven surface in relation to ankle mobility after a four-week bi-axial ankle range of motion (ROM) exercise session. Fifteen older adults with chronic post-stroke hemiparesis (N = 15; mean age = 65 years) participated in a total of 12 bi-axial ankle ROM exercises that consisted of three 30-min training sessions per week for four weeks. Basic clinical tests and gait performance in even and uneven surfaces were evaluated before and after training. Participants with chronic post-stroke hemiparesis showed significantly improved ankle functions, decreased ankle stiffness (from 0.140 ± 0.059 to 0.128 ± 0.067 N·m/°; p = 0.025), and increased paretic ankle passive ROMs (dorsiflexion(DF)/plantarflexion(PF): from 27.3 ± 14.7° to 50.6 ± 10.3°, p < 0.001; inversion(INV)/eversion(EV): 21.7 ± 9.7° to 28.6 ± 9.9°; p = 0.033) after training. They exhibited significant improvements in the walking performance over an uneven surface, step kinematics (walking speed 0.257 ± 0.17 to 0.320 ± 0.178 m/s; p = 0.017; step length: 0.214 ± 0.109 to 0.243 ± 0.108 m; p = 0.009), and clinical balance and mobility (Berg balance scale: 47.2 ± 4.7 to 50.1 ± 3.9, p = 0.0001; timed-up and go test: 23.9 ± 10.3 to 20.2 ± 7.0 s, p = 0.0156). This study is the first research to investigate the walking performance on uneven surfaces in the elderly with chronic stroke in relation to the ankle biomechanical property changes.