Proprioceptive deficits including impaired joint position sense (JPS) are considered to be contributing to impaired neuromuscular control, resulting in functional instability and repetitive injury.1 Furthermore, impaired JPS is associated with impaired muscle performance,2 and this combination is associated with functional deficits in people with musculoskeletal disorders.3 Therefore, it is important to identify effective interventions to enhance the JPS for preventing injuries and restoring functions in sports rehabilitation and musculoskeletal physiotherapy. Furthermore, sports performance can be associated with the magnitude of the JPS.4 Therefore, interventions to enhance the JPS would be concerns in the area of athletic training and sports science. Proprioceptive neuromuscular facilitation (PNF) can be a promising intervention, and several studies have investigated this subject in the past 15 years.5–7
PNF is an applied technique designed to promote the response of the neuromuscular mechanisms, such as mobility, muscular strength and endurance, joint stability, balance, and neuromuscular control,8 by stimulating the proprioceptors within the skin, joints, muscles, and tendons. There are several PNF techniques commonly mentioned in the literature, including the contract-relax method and the contract-relax-antagonist-contract method to improve mobility,9 as well as the repeated contraction method and the replication method to facilitate neuromuscular outputs. Several systematic reviews have discussed the effect of the contract-relax and contract-relax-antagonist-contract methods on the performance in power dominant tasks and endurance dominant tasks.10,11 Generally, the 2 methods tend to reduce performance in the power dominant tasks and endurance dominant tasks. However, to our knowledge, thus far, no systematic review has investigated the effect of PNF on the performance of JPS. Studies found that PNF enhanced activities of the superior parietal cortex12 and cerebellum,13 both of which contribute to the JPS.
It may be possible that PNF contributes to enhancement of the JPS, but the effect of PNF on the JPS is hypothesized to differ with different techniques, body regions, and physical problems. A meta-analysis involving several high-quality studies that involve similar techniques, outcome measures, and patient characteristics is required to comprehensively understand the effect of an intervention; however, it is unclear as to how many such studies are available. The present study aimed to examine existing peer-reviewed original studies that have investigated the effect of PNF techniques on the JPS in terms of the methodological quality, PNF techniques, outcomes, and participant characteristics. The findings of this study provide direction for future researches aimed at achieving a comprehensive understanding of the effect of PNF techniques on the JPS.
Methods
Design
This systematic review was undertaken in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines14 and updated method guidelines for Cochrane Musculoskeletal Group systematic reviews and meta-analyses.15 This study was registered in the international prospective register of systematic reviews (CRD42016041823).
Identification and Selection of Studies
A systematic search was performed using the refined key search terms (Appendix 1) on PubMed, EMBASE, MEDLINE, CINAHL, SocINDEX, Scopus, and the Cochrane Library from inception to January 2018. Cross-referencing was undertaken using the relevant reviews.
The following inclusion criteria were used: (1) assessment of the JPS; (2) peer-reviewed original studies with a randomized controlled trial or quasi-randomized controlled trial design; (3) participants with musculoskeletal disorders or healthy individuals (ie, neither animal studies nor those involving neurological problems); and (4) no cointervention with PNF, except for warm-up procedures. There was no language limitation. One author (H.T.) undertook database search. Screening of the title, abstract, and full text was performed by 2 authors (S.O. and Y.O.) independently. Full-text inspection was undertaken for the studies that either of the authors retained for full-text inspection during the screening process. There was no disagreement regarding the inclusion of the studies as per the inclusion criteria.
Assessment of Methodological Quality
We used the PEDro scale.16–18 In the PEDro scale, a high-quality study is defined as a study with a PEDro score ≥6.17 Reducing the cutoff score from 6 to 5 did not affect the overall outcome, and a cutoff score of 5 was used in some reviews.19–21 Therefore, we investigated if the number of high-quality studies changed using the 2 thresholds.
To further understand the methodological quality, the following 5 points were also assessed (Appendix 2): (1) approval from the ethics committee and informed consent (criterion A), (2) trial registration (criterion B), (3) inclusion of an adequate sample size (criterion C), (4) reliability of the outcome measures (criterion D), and (5) appropriate statistical assessment with effect size calculation (repeated-measures analysis of variance with effect size calculation for parametric analysis and Mann–Whitney U test using predifference to postdifference with effect size calculation; criterion E). These criteria have been used in other critical appraisal tools and CONSORT statement.22,23
The International Committee of Medical Journal Editors requested that all clinical trials be registered in the database before patient recruitment in 2005. Therefore, we evaluated if trial registration was undertaken for studies published after 2005.
The 2 authors (S.O. and Y.O.) assessed the methodological quality, and any disagreement between them was resolved by another author (H.T.). Agreement regarding the PEDro scale and the additional criteria between the 2 authors (S.O. and Y.O.) was examined with Cohen kappa and percentage agreement, where the following interpretations of the kappa value were used: <.4 = poor agreement, .41 to .60 = moderate agreement, .61 to .80 = good agreement, and .81 to 1.0 = very good agreement.24 All the authors were certified in the PEDro scale training program.
Data Analyses
Data were extracted by the 2 authors (S.O. and H.T.) independently. Any disagreement was resolved by another author (Y.O.).
Effect size (η2) was calculated, where η2 values of <.003, .01 to .039, .06 to .11 and >.14 represent no effect, small effect, intermediate effect, and large effect size, respectively.25 A positive value in the effect size indicated increased JPS after PNF compared with that with other approaches (control groups), including the wait-and-see method. Data of the control group were extracted from the wait-and-see method when it was reported; otherwise, data of the control groups were extracted from other approaches compared with the PNF group. A negative effect size value indicated decreased JPS after PNF compared with that with the control groups. Effect size (η2) for the interaction effect and intervention effect was calculated. When the effect size (η2) for the interaction effect and/or the intervention effect was not reported, the η2 value between the PNF group and the control group was calculated using the within-group change scores. Furthermore, when the within-group change scores were not reported and when the baseline scores were assumed to be comparable, the η2 value between the PNF group and the control group was calculated using the postintervention scores.19 The corresponding author of the literature was contacted twice via e-mail within a 1-month interval in case of insufficient information for data extraction.
Results
Study Selection
Figure 1 presents the flow of study selection. Nine studies were assessed for the methodological quality. Lazarou et al5 provided us with unpublished data of effect sizes.
—Flow of the study selection. JPS indicates joint position sense; RCT, randomized controlled trial.
Citation: Journal of Sport Rehabilitation 29, 4; 10.1123/jsr.2018-0498
Characteristics of the Studies
Table 1 summarizes the 9 studies. Two studies were published in Japanese, while the others were in English.
Summary of the 9 Studies Analyzed in This Review
| Study/corresponding author responded or not (including not contactable) | Design | Participants | Interventions relevant to this review | Measures | Effect size of intervention in comparison with a control condition (methods of effect size calculation) |
|---|---|---|---|---|---|
| Lazarou et al5 Responded | RCT | n = 20 • PNF (n = 10) • Control (n = 10) Eligibility • Ankle sprain • No participation in any form of supervised training after the injury • Exclusion criteria: chronic ankle instability, grade 3 ankle sprain, medial or interosseous ligament ankle sprain, concurrent fracture, history of nerve injury or ankle surgery to lower limbs, and further ankle injury after the sprain Age • PNF: mean (SD) = 22 (4) y • Control: mean (SD) = 22 (2) y Gender • PNF: 7 women and 3 men • Control: 7 women and 3 men | Body part: ankle PNF • RC using rhythmic stabilization first for the diagonal ankle movements without movements of the hip and knee (10-s isometric contraction of agonistic and antagonistic muscles, 2-min resting, and combination of resisted active concentric contraction for 5 s, resisted isometric contraction for 5 s, and resisted eccentric contraction for 5 s of agonistic muscles, with no rest between contractions) • 5–15 repetitions/set with 30-s rest between sets • 20 min of PNF training/session with a total training time of 50–60 min/session • 10 sessions for 6 wk Control • Balance training for 50–60 min/session (wobble board, firm surface, and soft-surface activities with eyes open) • 10 sessions for 6 wk | Outcome measures relevant to this review • Active repositioning acuity • Target position: 10° of dorsiflexion, 15° of plantar flexion, 30° of plantar flexion • Storage time: 15 s • Follow-up: 6-wk postintervention (follow-up 1) and 8 wk after the follow-up 1 (follow-up 2) Measurement side: injured side Measurement posture: sitting Measurement method: • Absolute error • Biodex System II Pro isotonic dynamometer (Loredan Biomedical, Inc, Sacramento, Calif) • Internal goniometer of the Biodex Repetitions for the measure: 3 | η2 = .0025 (between-group difference in prechange to postchange scores, follow-up 1, 10° of dorsiflexion) η2 = .0004 (between-group difference in prechange to postchange scores, follow-up 1, 15° of plantar flexion) η2 = .0529 (between-group difference in prechange to postchange scores, follow-up 1, 30° of plantar flexion) η2 = .1156 (between-group difference in prechange to postchange scores, follow-up 2, 10° of dorsiflexion) η2 = .0081 (between-group difference in prechange to postchange scores, follow-up 2, 15° of plantar flexion) η2 = .0016 (between-group difference in prechange to postchange scores, follow-up 2, 30° of plantar flexion) |
| Bjorklund et al26 Not responded | RCT (cross-over with washout period of a minimum of 2 d [mean 7.2 d]) | n = 18 • PNF (n = 18) • Control (n = 18) Eligibility • Healthy individuals (no current musculoskeletal problem in the right shoulder) • Dominant side of right Age • Men: mean (SD) = 24 (3) y • Women: mean (SD) = 21 (2) y Gender: 9 men and 9 women | Body part: shoulder PNF 1 • CR stretching for shoulder horizontal abduction (isometric contraction in the direction of the horizontal adduction for 5 s, 2- to 3-s resting, and then static stretching in the direction of horizontal abduction for 20 s × 3 sets) PNF 2 • CR stretching for shoulder horizontal adduction (isometric contraction in the direction of the horizontal abduction for 5 s, 2- to 3-s resting, and then static stretching in the direction of horizontal adduction for 20 s × 3 sets) Control • 5-min resting in sitting | Outcome measures relevant to this review • Active repositioning acuity • Target position: 15°and 30° horizontal angle between the arm and a sagittal plane • Storage time: 5 s Measurement side: right side Measurement posture: sitting Measurement method: • Variable error • Absolute error • Electromagnetic tracking system Repetitions for the measure: 6 | η2 = −.030 (between-group difference in prechange to postchange scores, variable error, PNF 1 vs Control, 15° horizontal angle between the arms) η2 = −.011 (between-group difference in prechange to postchange scores, variable error, PNF 1 vs Control, 30° horizontal angle between the arms) η2 = −.059 (between-group difference in prechange to postchange scores, variable error, PNF 2 vs Control, 15° horizontal angle between the arms) η2 = .018 (between-group difference in prechange to postchange scores, variable error, PNF 2 vs Control, 30° horizontal angle between the arms) η2 = −.009 (between-group difference in prechange to postchange scores, absolute error, PNF 1 vs Control, 15° horizontal angle between the arms) η2 = .007 (between-group difference in prechange to postchange scores, absolute error, PNF 1 vs Control, 30° horizontal angle between the arms) η2 = −.009 (between-group difference in prechange to postchange scores, absolute error, PNF 2 vs Control, 15° horizontal angle between the arms) η2 = −.011 (between-group difference in prechange to postchange scores, absolute error, PNF 2 vs Control, 30° horizontal angle between the arms) |
| Padua et al7 Not responded | RCT | n = 57 • PNF (n = 13) • Control (n = 15) • Open kinetic chain exercise (n = 13) • Close kinetic chain exercise (n = 13) Eligibility • Healthy individuals • No history of injury to the upper-extremity within the preceding 12 mo • No history of glenohumeral subluxation or dislocation • No current involvement in formal shoulder exercise program Age • PNF: mean (SD) = 21.9 (1.93) y • Control: mean (SD) = 20.7 (1.6) y Gender: unknown | Body part: shoulder Warm-up for 2 min with an Airdyne bicycle using only the arms (both groups) PNF • RC of shoulder patterns (D1 and D2) • 3 sets × 10 repetitions (first with 50% maximum voluntary effort, second with 75% maximum voluntary effort, and third with 100% maximum voluntary effort) • 1-min rest between sets • 90-s rest between exercise patterns • 3 times/wk for 5 wk • D1: shoulder extension–abduction–internal rotation • D1: shoulder flexion–adduction–external rotation • D2: shoulder extension–adduction–internal rotation • D2: shoulder flexion–abduction–external rotation Control • Act as usual • No participation in any form of upper-extremity exercise training • 5 wk | Outcome measures relevant to this review • Active repositioning acuity • Target position: 30°internal rotation, 30° external rotation, 75° external rotation • Storage time: 10 s Measurement side: dominant side Measurement posture • Supine • 90° shoulder abduction and elbow in 90° flexion Measurement method • Absolute error • LIDO Multi-Joint II isokinetic dynamometer (Loredan Biomedical, Inc, Sacramento, CA, USA) • Dynamometer speed of 300°/s (no resistance during rotation) • Electrogoniometer of the isokinetic dynamometer Repetitions for the measure: 3 | η2 = .035 (between-group difference in postintervention score, 30° internal rotation) η2 = .063 (between-group difference in postintervention score, 30° external rotation) η2 = .035 (between-group difference in postintervention score, 75° external rotation) |
| Streepey et al27 Not responded | Quasi-RCT | n = 18 • PNF (n = 9) • Control (n = 9) Eligibility: Healthy individuals (free from current injury and no history of serious injury requiring surgery to the knee of their dominant side) Age: 18–30 y Gender: 10 women and 8 men | Body part: knee PNF • CR stretching of the quadriceps and hamstring (stretching for 30 s with mild discomfort, isometric contraction for 10 s, and then 30-s resting × 3 sets with a 30-s break) • Hip extension with knee flexion for quadriceps stretching • Hip flexion with knee extension for hamstring stretching Control • Sham passive movement (passive movement of the hip into flexion with knee extension and passive movement of the hip into extension with knee flexion without either mild discomfort or isometric contraction) | Outcome measure relevant to this review • Kinesthesia of the knee from a 135°angle between the thigh and lower leg Measurement side: dominant side Measurement posture: sitting Measurement method • Absolute error • Knee flexion/extension was undertaken at a velocity of 0.4°/s Repetitions for the measure: 5 knee extension and 5 knee flexion in random order | η2 = .22 (interaction effect) η2 = .012 (between-group difference in postintervention score) |
| Minshull et al28 Not responded | RCT | n = 18 • PNF (n = 9) • Control (n = 9) Eligibility • Healthy individuals • Recreational sports participants (3 times/wk) without systematic flexibility or strength training Age • PNF: mean (SD) = 20.3 (2.2) y • Control: mean (SD) = 20.7 (2.3) y Gender: 18 men | Body part: knee PNF • CRAC of the hamstring (isometric contraction of hip extension and knee flexion for 10 s, 5-s resting, and 10-s active assisted hip flexion with knee extension × 3 sets with a 60-s break) • 2 times/wk for 8 wk Control • Passive stretching of the hamstring (10-s static stretching, 5-s resting, and 10-s static stretching × 3 sets with a 60-s break) • 2 times/wk for 8 wk | Outcome measures relevant to this review • Active repositioning acuity • Target position: 160° angle between the thigh and lower leg • Storage time: unknown Measurement side: preferred side Measurement posture: prone Measurement method • Percentage relative to the target angle of knee flexion [(observed performance score − target performance score)/(target performance score) × 100%]a • Instrument with dynamometer • Repetitions for the measure: 5 | η2 = −.0004 (between-group difference in postintervention score) |
| Sookhee and Hohee29 Not responded | RCT | n = 54 • PNF (n = 26) • Control (n = 28) Eligibility • Patients with nonspecific low back pain • Age between 19 and 65 y • Exclusion criteria: structural differences in lower-extremity, pathological findings of tumors, a history of surgery for severe trauma, a rheumatoid disease, a neurological problem, recent treatment to relive pain or medication, treatment in a large hospital within the past 3 mo, or an exercise habit of > 3 d/wk Age • PNF: mean (SD) = 58.0 (32.1) y • Control: mean (SD) = 60.8 (30.2) y Gender: unknown | Body part: lower-extremity Warm-up with hyperthermia treatment for 20 min (both groups) PNF • RC (lower-extremity flexion–adduction–external rotation with knee flexion, sprinter, lifting)a • 10-s exercise and 10-s rest • 3 sets × 4 times/wk for 6 wk Control • Stabilization exercise (supine, bridge, quadruped, standing position) • 10-s exercise and 10-s rest • 3 sets × 4 times/wk for 6 wk | Outcome measures relevant to this review • Active repositioning acuity • Target position: 30° trunk flexion • Storage time: a few seconds • Follow-up: 3-wk postintervention (follow-up 1) and 6-wk postintervention (follow-up 2) Measurement side: not applicable Measurement posture: standing Measurement method • Angle at Th12 spinal processa • Digital goniometer Repetitions for the measure: 3 | η2 = −.006 (between-group difference in postintervention score, follow-up 1) η2 = .0009 (between-group difference in postintervention score, follow-up 2) |
| Cho et al6 Not responded | RCT | n = 40 • PNF (n = 19) • Control (n = 21) Eligibility: healthy individuals (free from orthopedic abnormalities and no history of knee injury) Age • PNF: mean (SD) = 20.2 (0.9) y • Control: mean (SD) = 20.3 (1.3) y Gender • PNF: 9 women and 10 men • Control: 10 women and 11 men | Body part: knee PNF • CR of the hamstring at 90° knee flexion (isometric contraction of knee flexion for 7 s and then 5-s resting × 3 sets) • CR of the quadriceps at 90° knee flexion (isometric contraction of knee extension for 7 s and then 5-s resting × 3 sets) • 10 s between the 2 different CR interventions Control • Without interventions | Outcome measures relevant to this review • Active repositioning acuity • Target position: 60°, 90°, 120°, and 150° angle between the thigh and lower leg • Storage time: 5 s Measurement side: unknown Measurement posture: prone Measurement method • Absolute error • Isokinetic muscular strength equipment (CSMI) Repetitions for the measure: unknown | η2 = .28 (between-group difference in postintervention score, a target of 60°) η2 = .20 (between-group difference in postintervention score, a target of 90°) η2 = .14 (between-group difference in postintervention score, a target of 120°) η2 = .19 (between-group difference in postintervention score, a target of 150°) |
| Furuya et al30 Responded | RCT | n = 30 (from 19 patients) • PNF (n = 14) • Control (n = 16) Eligibility • Patients undergoing knee arthroplasty (3 wk after the surgery) • Exclusion criteria: diseases of central nervous system, history of medical diagnosis in the back or hip, unable to flex the knee >90°, unable to extend the knee >45°, exclusion from a postsurgical protocol (walk training after the next day of the surgery and discharging at 3 wk after the surgery) Age: mean (SD) = 72.4 (5.2) y Gender: 16 women and 3 men | Body part: knee PNF • Replication of hip flexion–adduction–external rotation pattern with minimizing hip movements in sitting (45° knee extension from 90° flexion × 5 sets) Control • Active knee extension (45° knee extension from 90° flexion × 5 sets) | Outcome measures relevant to this review • Active repositioning acuity • Target position: 135° angle between the thigh and lower leg • Storage time: 5 s Measurement side: surgical side Measurement posture: sitting Measurement method • Absolute error • Analysis of static image from a side view using Image J Repetitions for the measure: 5 | η2 = .08 (interaction effect) η2 = .02 (main effect of intervention) |
| Ito et al31 Responded | RCT | n = 18 • PNF (n = 9) • Control (n = 9) Eligibility: healthy individuals (no history of knee injury) Age: mean (SD) = 35.3 (7.5) y Gender: 9 women and 9 men | Body part: knee PNF • Replication of hip flexion–adduction–external rotation pattern with minimizing hip movements in sitting (45° knee extension from 90° flexion × 5 sets) Control • Active knee extension (45° knee extension from 90° flexion × 5 sets) | Outcome measures relevant to this review • Active repositioning acuity • Target position: 135° angle between the thigh and lower leg • Storage time: 5 s Measurement side: dominant side Measurement posture: sitting Measurement method • Absolute error • Analysis of static image from a side view using Image J Repetitions for the measure: 5 repetitions | η2 = .28 (between-group difference in prechange to postchange scores) |
Abbreviations: CR, contract-relax; CRAC, contract-relax-antagonist-contract; PNF, proprioceptive neuromuscular facilitation; RC, repeated contraction; RCT, randomized controlled trial.
aDifficult for clear understanding from the original paper.
Table 2 presents the PEDro scores and additional criteria A to E. For the PEDro scale, the reviewers scored 99 items and agreed on 96 items (97.0% agreement). The kappa value with 95% confidence intervals was .94 (.87–1.00) (P < .001), and the overall interreviewer agreement of the methodological quality was good. With respect to the additional criteria A to E scale, the reviewers scored 45 items and agreed on all the items (100% agreement). The kappa value with 95% confidence intervals was 1.0 (1.0) (P < .001), and the overall interreviewer agreement of the methodological quality was good. More than 80% of studies failed to fulfill criterion 3 (concealed allocation), criterion 5 (subject blinding), criterion 6 (therapist blinding), criterion 9 (intention-to-treat analysis), criterion B (trial registration), criterion C (inclusion of an adequate sample size), and criterion E (appropriate statistical assessment with effect size calculation). Regardless of the PEDro cutoff score (5 or 6), 6 out of 9 studies were not considered to be of high quality. Furthermore, none of the studies satisfied all the additional criteria (criteria A–E). All the studies except the one by Padua et al7 were published after 2005; however, only one study5 registered the trial.
Methodological Quality
| PEDro criteria | Additional criteria | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Study | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | A | B | C | D | E | PEDro total from criteria 2 to 11 (/10) | Total (/16) |
| Lazarou et al5 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 9 | 13 |
| Bjorklund et al26 | 1 | 1 | 0 | 1 | 0 | 0 | 0 | 1 | 0 | 1 | 1 | 1 | 0 | 0 | 1 | 0 | 5 | 8 |
| Padua et al7 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 1 | 0 | 1 | 1 | 1 | 0 | 0 | 1 | 0 | 5 | 7 |
| Streepey et al27 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 1 | 0 | 1 | 1 | 1 | 0 | 0 | 1 | 0 | 4 | 6 |
| Minshull et al28 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 1 | 0 | 0 | 0 | 0 | 4 | 5 |
| Sookhee and Hohee29 | 1 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 4 | 5 |
| Cho et al6 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 3 | 5 |
| Furuya et al30 | 0 | 1 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 4 | 4 |
| Ito et al31 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 3 |
Note: A greater total score generally indicates a better methodological quality (0 = not satisfied and 1 = satisfied).
Reporting criteria: 1, eligibility criteria; 2, random allocation; 3, concealed allocation; 4, baseline comparability; 5, subject blinding; 6, therapist blinding; 7, assessor blinding; 8, follow-up; 9, intention-to-treat; 10, between-group statistical comparisons; 11, measures of variability; A, approval from the ethics committee and informed consent; B, trial registration; C, inclusion of an adequate sample size; D, reliability of the outcome measures; E, appropriate statistical assessment with effect size calculation.
Discussion
The current study systematically reviewed the studies that investigated the effect of PNF techniques on the JPS in terms of methodological quality, PNF techniques, outcomes, and participant characteristics. This current review provides us suggestions for enhancing the methodological quality by understanding the common errors in existing studies and gaps in the literature.
Regardless of the PEDro cutoff score, 6 out of 9 studies were not considered to be of high quality. Furthermore, none of the studies satisfied all the additional criteria. Even in the study by Lazarou et al,5 wherein the methodological quality was considered high as per the PEDro score, additional criteria C (inclusion of an adequate sample size) and D (reliability of the outcome measures) were not satisfied. These faults in the additional criteria can influence conclusions of the study5; thus, the results should be cautiously interpreted. No study satisfied criterion C (inclusion of an adequate sample size) and criterion E (appropriate statistical assessment with effect size calculation). Underpowered results and inappropriate statistical assessment or the lack of effect size calculation influenced the study conclusions. Furthermore, it was found that most studies in the current review did not register the trial and did not follow global publication standards. Therefore, we recommend that we should assume the lack of convincing information about the effect of PNF on the JPS. In order to fully understand the effect of PNF on the JPS, further well-designed studies are required.
Considerations for Future Studies
We calculated the effect sizes for the effect of PNF on the JPS in each study. A large effect size was observed in 2 studies6,31 although the methodological quality was poor in each study. These studies suggest promising conditions that can be used in future trials. First , regarding a target joint, the 2 studies with a large effect size6,31 used the knee. Thus, the knee can be a promising target in future trials.
Regarding the PNF technique, Cho et al6 used contract-relax and Ito et al31 used replication. Only the postintervention scores were available from the Cho et al’s study6; thus, effect size was calculated using the postintervention scores. In the study by Ito et al,31 only within-group changes in the scores were available; thus, the effect size was calculated based on the within-group change scores. Statistically, an effect size calculated using the within-group change scores is considered more accurate than that calculated using the postintervention scores. Therefore, a promising intervention in a future trial may be a replication, and this would not be surprising considering findings in previous studies that investigated brain area facilitated by PNF.12,13 The replication includes rotational elements of joint movements, and all previous studies that demonstrated enhanced activities of the superior parietal cortex and cerebellum12,13 used PNF techniques with rotational elements. Therefore, rotational elements of movements may be a key factor for PNF techniques to enhance the JPS, and future studies are required.
Regarding participants, the 2 studies with a large effect size6,31 recruited healthy individuals. When we explored the possibilities of enhancing the performance in activities that require fine motor control with PNF, a study involving healthy individuals may be considered. By contrast, although Furuya et al30 and Ito et al31 used replication targeting the knee, Furuya et al30 did not detect an effect size as large as that in Ito et al’s study.31 This difference may have been associated with the difference in the recruited participants because Furuya et al30 recruited patients undergoing knee arthroplasty, while Ito et al31 recruited healthy individuals. When we explored effective interventions to improve impaired JPS, a trial with participants with more preserved mechanoreceptors than that in patients undergoing knee arthroplasty would be demanded. It has been known that patients with anterior cruciate ligament injury have impaired JPS,32–35 and thus, it may be important to undertake a future trial including patients with anterior cruciate ligament injuries to investigate effect of replication technique on the JPS.
Study Limitations
The current review has 2 limitations. First, the ideal effect size calculation should have been performed with the interaction effect and intervention effect. However, such a statistical analysis was reported by Furuya et al30 only. It was impossible to obtain additional data for the statistical analyses, and effect size calculations were undertaken using the within-group change scores or the postintervention scores in most studies.6,7,26–29,31 It can be assumed that the calculated effect sizes in the current study can be different from the genuine effect sizes. The use of the effect sizes is recommended only for future research designs. Second, there were 2 studies with unclear information regarding the interventions and/or outcomes,28,29 where it was impossible to clarify the information from the corresponding authors. We could not reperform these studies, and these results should be cautiously interpreted.
Conclusions
We found that there were no multiple studies with high methodological quality and similar PNF techniques, outcomes, and participant characteristics. More high-quality studies are required to achieve a comprehensive understanding regarding the effect of PNF on the JPS.
References
- 1.↑
Lephart SM, Pincivero DM, Giraldo JL, Fu FH. The role of proprioception in the management and rehabilitation of athletic injuries. Am J Sports Med. 1997;25:130–137. PubMed ID: 9006708 doi:10.1177/036354659702500126
- 2.↑
Mota N, Ribeiro F. Association between shoulder proprioception and muscle strength in water polo players. Isokinet Exerc Sci. 2012;20:17–21. doi:10.3233/IES-2011-0435
- 3.↑
van der Esch M, Steultjens M, Harlaar J, Knol D, Lems W, Dekker J. Joint proprioception, muscle strength, and functional ability in patients with osteoarthritis of the knee. Arthritis Rheum. 2007;57:787–793. PubMed ID: 17530678 doi:10.1002/art.22779
- 4.↑
Kaya D, Callaghan MJ, Donmez G, Doral MN. Shoulder joint position sense is negatively correlated with free-throw percentage in professional basketball players. Isokinet Exerc Sci. 2012;20:189–196. doi:10.3233/IES-2012-0458
- 5.↑
Lazarou L, Kofotolis N, Malliou P, Kellis E. Effects of two proprioceptive training programs on joint position sense, strength, activation and recurrent injuries after ankle sprains. Isokinet Exerc Sci. 2017;25:289–300. doi:10.3233/IES-171146
- 6.↑
Cho SI, Lee DY, Hong JH, Yu JH, Kim JS. Effect of hold and relax technique on knee joint position awareness in normal adults. Indian J Sci Technol. 2015;8(19).
- 7.↑
Padua DA, Guskiewicz KM, Prentice WE, Schneider RE, Shields EW. The effect of select shoulder exercises on strength, active angle reproduction, single-arm balance, and functional performance. J Sport Rehabil. 2004;13:75–95. doi:10.1123/jsr.13.1.75
- 8.↑
Sharman MJ, Cresswell AG, Riek S. Proprioceptive neuromuscular facilitation stretching: mechanisms and clinical implications. Sports Med. 2006;36:929–939. PubMed ID: 17052131 doi:10.2165/00007256-200636110-00002
- 9.↑
Hindle KB, Whitcomb TJ, Briggs WO, Hong J. Proprioceptive neuromuscular facilitation (PNF): its mechanisms and effects on range of motion and muscular function. J Hum Kinet. 2012;31:105–113. PubMed ID: 23487249 doi:10.2478/v10078-012-0011-y
- 10.↑
Peck E, Chomko G, Gaz DV, Farrell AM. The effects of stretching on performance. Curr Sports Med Rep. 2014;13:179–185. PubMed ID: 24819010 doi:10.1249/JSR.0000000000000052
- 11.↑
Behm DG, Blazevich AJ, Kay AD, McHugh M. Acute effects of muscle stretching on physical performance, range of motion, and injury incidence in healthy active individuals: a systematic review. Appl Physiol Nutr Metab. 2016;41:1–11. PubMed ID: 26642915 doi:10.1139/apnm-2015-0235
- 12.↑
Lial L, Moreira R, Correia L, et al. Proprioceptive neuromuscular facilitation increases alpha absolute power in the dorsolateral prefrontal cortex and superior parietal cortex. Somatosens Mot Res. 2017;34:204–212. PubMed ID: 29096587 doi:10.1080/08990220.2017.1392298
- 13.↑
Shuratani T, Nitta O, Matsuda M, Tada Y, Senoo A, Yanagisawa K. The effects of a static contraction of pelvic anterior elevation on the brain activities induced by a fMRI in the normal volunteers. J Jpn Acad Health Sci. 2011;14:205–212.
- 14.↑
Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Int J Surg. 2010;8:336–341. PubMed ID: 20171303 doi:10.1016/j.ijsu.2010.02.007
- 15.↑
Ghogomu EA, Maxwell LJ, Buchbinder R, et al. Updated method guidelines for Cochrane Musculoskeletal Group systematic reviews and metaanalyses. J Rheumatol. 2014;41(2):194–205. PubMed ID: 24293581 doi:10.3899/jrheum.121306
- 16.↑
de Morton NA. The PEDro scale is a valid measure of the methodological quality of clinical trials: a demographic study. Aust J Physiother. 2009;55:129–133. PubMed ID: 19463084 doi:10.1016/S0004-9514(09)70043-1
- 17.↑
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. PubMed ID: 12882612
- 18.↑
Sherrington C, Herbert RD, Maher CG, Moseley AM. PEDro. A database of randomized trials and systematic reviews in physiotherapy. Man Ther. 2000;5:223–226. PubMed ID: 11052901 doi:10.1054/math.2000.0372
- 19.↑
Takasaki H, May S. Mechanical diagnosis and therapy has similar effects on pain and disability as ‘wait and see’ and other approaches in people with neck pain: a systematic review. J Physiother. 2014;60:78–84. PubMed ID: 24952834 doi:10.1016/j.jphys.2014.05.006
- 20.
Boyles R, Toy P, Mellon J Jr, Hayes M, Hammer B. Effectiveness of manual physical therapy in the treatment of cervical radiculopathy: a systematic review. J Man Manip Ther. 2011;19:135–142. PubMed ID: 22851876 doi:10.1179/2042618611Y.0000000011
- 21.↑
Coppola SM, Collins SM. Is physical therapy more beneficial than unsupervised home exercise in treatment of post surgical knee disorders? A systematic review. Knee. 2009;16:171–175. PubMed ID: 18851916 doi:10.1016/j.knee.2008.09.001
- 22.↑
Takasaki H, Lim ECW, Soon B. The effect of shoulder muscle fatigue on active repositioning acuity and scapulothoracic resting alignment: a systematic review with meta-analysis. Phys Ther Sport. 2016;20:61–78. PubMed ID: 27080109 doi:10.1016/j.ptsp.2016.01.001
- 23.↑
Takasaki H, Miki T. The impact of continuous use of lumbosacral orthoses on trunk motor performance: a systematic review with meta-analysis. Spine J. 2017;17:889–900. PubMed ID: 28323240 doi:10.1016/j.spinee.2017.03.003
- 25.↑
Cohen J. Statistical Power Analysis for the Behavioural Sciences. 2nd ed. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.; 1988.
- 26.↑
Bjorklund M, Djupsjobacka M, Crenshaw AG. Acute muscle stretching and shoulder position sense. J Athl Train. 2006;41:270–274. PubMed ID: 17043694
- 27.↑
Streepey JW, Mock MJ, Riskowski JL, Vanwye WR, Vitvitskiy BM, Mikesky AE. Effects of quadriceps and hamstrings proprioceptive neuromuscular facilitation stretching on knee movement sensation. J Strength Cond Res. 2010;24:1037–1042. PubMed ID: 20300021 doi:10.1519/JSC.0b013e3181d09e87
- 28.↑
Minshull C, Eston R, Bailey A, Rees D, Gleeson N. The differential effects of PNF versus passive stretch conditioning on neuromuscular performance. Eur J Sport Sci. 2014;14:233–241. PubMed ID: 23688197 doi:10.1080/17461391.2013.799716
- 29.↑
Sookhee B, Hohee S. The effects of proprioceptive neuromuscular facilitation and stabilizing exercise on trunk repositioning errors. J Phys Ther Sci. 2012;24:1017–1020. doi:10.1589/jpts.24.1017
- 30.↑
Furuya H, Ito T, Sumiya K, Tomoshige S, Tanaka T. The effect of PNF techniques with replication for the joint position sense in patients undergoing knee arthroplasty. PNF Res. 2009;10:28–32.
- 31.↑
Ito T, Saito T, Sato H, Shinoda M. The influence of replication on knee joint kinesthetic sense. PNF Res. 2007;7:39–43.
- 32.↑
Relph N, Herrington L. Knee joint position sense ability in elite athletes who have returned to international level play following ACL reconstruction: a cross-sectional study. Knee. 2016;23:1029–1034. PubMed ID: 27712856 doi:10.1016/j.knee.2016.09.005
- 33.
Friden T, Roberts D, Ageberg E, Walden M, Zatterstrom R. Review of knee proprioception and the relation to extremity function after an anterior cruciate ligament rupture. J Orthop Sports Phys Ther. 2001;31:567–576. PubMed ID: 11665744 doi:10.2519/jospt.2001.31.10.567
- 34.
Relph N, Herrington L, Tyson S. The effects of ACL injury on knee proprioception: a meta-analysis. Physiotherapy. 2014;100:187–195. PubMed ID: 24690442 doi:10.1016/j.physio.2013.11.002
- 35.↑
Kim HJ, Lee JH, Lee DH. Proprioception in patients with anterior cruciate ligament tears: a meta-analysis comparing injured and uninjured limbs. Am J Sports Med. 2017;45:2916–2922. PubMed ID: 28060536 doi:10.1177/0363546516682231
Appendix 1. Search Strategy
(stretching OR “Proprioceptive Neuromuscular Facilitation”) AND (sensorimotor OR proprioception OR “joint position sense”)
Appendix 2. Additional Criteria for the Assessment of the Methodological Quality
| [A. Approval from the ethics committee and informed consent] | |
| Have the authors documented the ethical approval for the research and gained informed consent from the participants? | |
| Yes | |
| No or Unable to determine | |
| [B. Trial registration] | |
| Have the authors registered the clinical trial at a clinical trial registry and documented the ID numbers? | |
| Yes | |
| No or Unable to determine | |
| [C. Incusion of an adequate sample size] | |
| Have the authors justified their sample size through a power calculation or post-hoc analysis (and recruited sufficient numbers)? | |
| Yes | |
| No or Unable to determine | |
| [D. Reliability of the outcome measures] | |
| Have the authors documented the evidence of the reliability of the outcome measures relevant to this review? For studies that have provided references of other works regarding reliability or have demonstrated the accuracy of the outcome measures, this question should be answered with yes. | |
| Yes | |
| No or Unable to determine | |
| [E. Appropriate statistical assessment with effect size calculation] | |
| Did the authors use appropriate statistical analyses for evaluating the results as per their aim, and have they documented the effect size? When a repeated-measures ANOVA with effect size for parametric analysis and Mann–Whitney U test using pre–post difference with effect size is used, this question should be answered with yes. | |
| Yes | |
| No or Unable to determine | |
