life

Systematic Review
Proprioceptive Neuromuscular Facilitation-Based Physical
Therapy on the Improvement of Balance and Gait in Patients
with Chronic Stroke: A Systematic Review and Meta-Analysis
Phan The Nguyen 1,2,† , Li-Wei Chou 1,3,4,† and Yueh-Ling Hsieh 1, *

1 Department of Physical Therapy, Graduate Institute of Rehabilitation Science, China Medical University,
Taichung 406040, Taiwan; [email protected] (P.T.N.); [email protected] (L.-W.C.)
2 Department of Physical Therapy, Faculty of Nursing and Medical Technology, University of Medicine and
Pharmacy, Ho Chi Minh City 8428, Vietnam
3 Department of Physical Medicine and Rehabilitation, China Medical University Hospital,
Taichung 406040, Taiwan
4 Department of Rehabilitation, Asia University Hospital, Taichung 413505, Taiwan
* Correspondence: [email protected]
† These authors contributed equally to this work.

Abstract: The present study aims to determine the potential benefits of PNF on balance and gait
function in patients with chronic stroke by using a systematic review and meta-analysis. Systematic
review in the following databases: MEDLINE/PubMed, Physiotherapy Evidence Database (PEDro),
Cochrane Library and Google Scholar. Studies up to September 2020 are included. A systematic
database search was conducted for randomized control trials (RCTs) that investigated the effects
of PNF intervention in patients with chronic stroke using balance and gait parameters as outcome
Citation: Nguyen, P.T.; Chou, L.-W.;
measures. The primary outcomes of interest were Berg Balance Scale (BBS), Functional Reach Test
Hsieh, Y.-L. Proprioceptive (FRT), Timed Up and Go Test (TUG) and 10-Meter Walking Test (10MWT). Nineteen studies with
Neuromuscular Facilitation-Based 532 participants were included, of which twelve studies with 327 participants were included for
Physical Therapy on the meta-analysis. When the data were pooled, PNF made statistically significant improvements in
Improvement of Balance and Gait in balance with BBS, FRT and TUG (p < 0.05) or gait velocity with 10MWT (p < 0.001) when compared
Patients with Chronic Stroke: A to the control. This review indicates that PNF is a potential treatment strategy in chronic stroke
Systematic Review and rehabilitation on balance and gait speed. Further high-quality research is required for concluding a
Meta-Analysis. Life 2022, 12, 882. consensus of intervention and research on PNF.
https://doi.org/10.3390/life12060882

Academic Editors: Jessica Barlinn Keywords: stroke; postural balance; gait; proprioceptive neuromuscular facilitation; stroke rehabilitation
and Milan R. Vosko

Received: 14 May 2022

Accepted: 8 June 2022
1. Introduction
Published: 13 June 2022
Stroke survivors in the chronic stage, namely, more than six months after stroke onset,
Publisher’s Note: MDPI stays neutral
often have long-term residual and disabling deficits, especially on the impairment of motor
with regard to jurisdictional claims in
tasks [1–3]. Moreover, muscle coordination is often decreased after a stroke, leading to
published maps and institutional affil-
deficits for both gait and balance control [4]. Even for stroke patients with independent
iations.
motor functions, balance impairments and gait abnormalities are two of the most common
manifestations in the chronic stage [5]. These deficits severely impede the individuals’
abilities to participate in activities of daily living and negatively impact their quality of life.
Copyright: © 2022 by the authors.
In addition, they indicate an increased risk for falls and a greater likelihood of hospital or
Licensee MDPI, Basel, Switzerland. nursing-home admissions [5].
This article is an open access article Several motor functions can often be improved with time and through various reha-
distributed under the terms and bilitation strategies, including modality, manual and movement therapies. However, the
conditions of the Creative Commons asymmetric postural behavior of stroke survivors during standing and walking is often
Attribution (CC BY) license (https:// reinforced, maintained or only transitorily decreased in the chronic recovery stage [4,6].
creativecommons.org/licenses/by/ Asymmetric posture can also be due to impaired sensory inputs, including decreased
4.0/). perception of midline. Increased use of the unaffected side can be a result of this as well [7].

Life 2022, 12, 882. https://doi.org/10.3390/life12060882 https://www.mdpi.com/journal/life

Life 2022, 12, 882 2 of 18

In view of this, chronic stroke rehabilitation, including muscle re-education in both affected
and unaffected sides, should first emphasize the correction of the postural asymmetric
pattern by enhancing the balance control of particular motor tasks beneficial to gait.
There are many available modality and movement therapies of post-stroke rehabili-
tation that have positive effects on motor and gait functions in patients after stroke, e.g.,
cycling, treadmill walking and functional electrical stimulation [8,9]. However, they may
be expensive and provide a limited kind of movement. Proprioceptive neuromuscular fa-
cilitation (PNF) is a therapeutic approach that uses cutaneous, proprioceptive and auditory
input to produce functional improvement in motor output and can play a vital role in the
rehabilitation of many injuries. It is a specific manual technique controlled by physical
therapists to help improve a patient’s functional status by incorporating multiple planes
of movements, making the task more functional and effective in achieving patient goals.
As it exhibits effects on the improvement of pain, range of motion, muscle strength and
endurance, coordination and facilitation of proximal stability and functional progression,
it has been widely used for early rehabilitation of the acute or subacute phases for neu-
romuscular re-education to improve motor functions of patients with stroke [10,11]. This
method stimulates proprioceptive organs in muscles and tendons to improve muscular
functions, promotes the exploration of postural reflexes and prioritizes muscle contraction
for increasing strength, flexibility, balance and coordination [12–14]. Preliminary case
reports revealed that a PNF-based program has the potential to generate positive outcomes
on motor function in older adults with chronic stroke [12,15,16]. Two systematic reviews
with small samples (five and twelve studies, respectively) reported that PNF is an effective
treatment for improving gait-related outcome measures in patients with stroke [11,17].
However, despite an increase in the published literature on the effects of PNF, there is still
limited evidence from the meta-analysis of randomized-controlled trials (RCTs) to quantify
the efficacy of PNF-based approaches on the improvement of motor outcomes, especially
for balance and gait in patients with chronic stroke.
A concise and up-to-date overview of the effectiveness of PNF-based training on
balance and gait in patients with chronic stroke is currently lacking. This study is a
systematic and meta-analytical review of the available RCTs to examine the effects of
PNF on improvements of balance and gait functions in patients with chronic stroke only,
excluding those in acute and subacute phases after stroke.

2. Materials and Methods

A systematic review was conducted in accordance with the PRISMA guidelines [18]
and the review protocol was registered with OSF (URL: https://osf.io/p26rj (accessed on
27 October 2020)).

2.1. Search Strategies

A systematic search of the related literature published between 1960/01 and 2021/07
was performed in MEDLINE/PubMed, Physiotherapy Evidence Database (PEDro), Cochrane
Library, Google Scholar, CINHAL, Web of Science and China Knowledge Resource Inte-
grated (CNKI). Database-specific filters were used, as necessary, to complete searches in all
the specified databases. Search filters were utilized so that only human studies and RCTs
were included. Hand searches were completed using the reference lists of related articles.
No language or date restrictions were imposed on the search. The search strategies and
keywords are illustrated in Table 1.

2.2. Inclusion/Exclusion Criteria for Studies

The inclusion criteria were as follows: studies of adult patients (>18 years old) di-
agnosed with chronic stroke (more than 6 months after stroke onset), studies using PNF
alone or in combination with other therapies as intervention in experimental groups and
non-PNF-based intervention in control groups, studies that measured balance- and gait-
related changes, and studies that used a RCT design. The exclusion criteria were graduation
Life 2022, 12, 882 3 of 18

theses, books, conference proceedings, including posters and platforms, single case studies,
quasi-randomized clinical trials and qualitative studies. This systematic review protocol
followed the recommendations of the Preferred Reporting Items for Systematic Review and
Meta-Analysis (PRISMA) Protocols [18].

Table 1. Search strategies and keywords list.

Main Term Keyword

“Stroke” OR “Cerebrovascular accident” OR “Cerebrovascular disease” OR “Cerebrovascular
#1 Population
disorders” OR “CVA” OR “Hemiplegic” OR “Hemiplegia”
“Proprioceptive Neuromuscular Facilitation” OR “PNF” OR “Proprioceptive” OR
#2 Intervention
“Neuromuscular Facilitation”
“Gait” OR “Walking” OR “Ambulation” OR “Lower extremity” OR “Endurance” OR “Balance” OR
#3 Outcomes
“Mobility” OR “Posture” OR “Postural control”
#4 Final Search #1 AND #2 AND #3

2.3. Screening Process

The initial identification and selection of studies in the electronic search was conducted
by two reviewers, who independently evaluated the titles and abstracts of all primary
studies. Then, the reviewers chose texts that were considered as eligible references and
evaluated whether they met the pre-established inclusion criteria. In case of a disagreement
between the two reviewers, the third reviewer would make the final decision.

2.4. Methodological Quality and Risk of Bias Assessment

The methodological quality of the studies was evaluated using the 11-item PEDro
Scale. Studies with PEDro scale scores of 9–10 were considered to be of excellent quality,
those with scores of 6–8 and 4–5 were of good and fair quality, respectively, and those
with scores below 4 were of poor quality [19]. The PEDro score demonstrated “fair” to
“excellent” inter-rater reliability (Intraclass Correlation Coefficient 0.53–0.91) for RCT of
physiotherapy interventions [20]. For studies not published on the PEDro database, but
that met the inclusion criteria, two reviewers would independently assess the quality of
these studies using the PEDro scale. The Cochrane Handbook for Systematic Reviews of
Interventions (Cochrane Collaboration) was used to evaluate the risk of bias in the included
trials (reported as low, high or unclear risks). The following criteria were used to assess the
risk of bias: the generation of randomization sequence, allocation concealment, blinding,
completeness of the data and reporting of outcomes. In case of a disagreement in the quality
assessment between the independent reviewers, a consensus was reached by a discussion
or consultation with a third reviewer of the research team.

2.5. Data Extraction and Analysis

Study characteristics, participant characteristics, interventions, duration of trial period,
period of outcome assessment and main results were extracted from the selected studies. A
meta-analysis of the study was performed using RevMan software, version 5.3 (The Nordic
Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark). Outcomes and
results were described. Meta-analysis was conducted on outcome measures used in at
least three RCT studies to assess the balance and gait. The pooled effect estimates were
computed with a random-effect model using the means, standard deviations (SDs) of the
post-intervention and number of participants [21]. Heterogeneity was assessed using I2
statistics. Sensitivity analysis was conducted to assess the influence of each study on the
combined results by excluding individual trials one at a time. Pooled-effect estimates were
obtained by comparing the change in the least square mean from baseline to endpoint
for each group, and were expressed as the mean difference (MD) with a 95% confidence
interval (CI) between groups. A p-value < 0.05 indicated a significant statistical difference.
Life 2022, 12, 882 4 of 18

3. Results
3.1. Study Selection
A total of 1253 potentially relevant studies were identified from the databases f the
initial search. Following deduplication, 518 articles underwent title and abstract screening.
In the end, 68 articles were included for full-text review, which further excluded 49 articles,
with the remaining 19 studies for qualitative synthesis [22–40]. Screening the evaluation
instruments used showed that the Berg balance scale (BBS), functional reach test (FRT),
timed up-and-go test (TUG) and 10-m walking test (10MWT) were used in at least three
separate studies, and their data in a total of 12 articles were combined for quantitative
Life 2022, 12, x FOR PEER REVIEW
review by meta-analysis. Figure 1 depicts the PRISMA flowchart. 5 of 19

Figure 1. Flowchart for study selection for systematic review and meta-analysis.
Figure 1. Flowchart for study selection for systematic review and meta-analysis.

3.2. Quality Appraisal and Risk of Bias

Table 2 shows the quality assessment of included studies according to the 19 items of
the PEDro scale. The PEDro scores of these 19 studies ranged from 5 to 7, with an average of
5.43 ± 0.62. As can be observed, 6 studies with scores of 6 or 7 are of good methodological
quality with low risk of bias [27,34,35,38–40]. PEDro scale items, namely, “Randomization”,
“Comparison at the baseline”, “Adequate follow-up”, “Comparisons between groups”
and “Measures of precision and variability”, were involved in the research designs of all
included studies. The most common reasons for lower scores with risk of bias were the lack
of “Allocation concealment” in 18 studies, “Blind participants” in the 19 included studies,
“Blind therapists” in 18 studies, “Blind evaluators” in 16 studies and “Intention-to-treat
analysis” in 17 studies (Table 2). The risk of bias in all studies did not achieve the low
 Life 2022, 12, 882 5 of 18

Life 2022, 12, x FOR PEER REVIEW 7 of 19

bias level and there were some concerns for all trials from the results evaluated using the
Cochrane Risk of Bias Assessment tool (Figure 2).

Figure 2. The risk of bias assessment summary using the Cochrane Risk of Bias Assessment tool.
Figure 2. The risk of bias assessment summary using the Cochrane Risk of Bias Assessment tool. The
The L sign indicates a low risk of bias, H indicates a high risk of bias and U sign indicates an unclear
L sign indicates a low risk of bias, H indicates a high risk of bias and U sign indicates an unclear risk
risk of bias [22–40].
of bias [22–40].

3.3. Participant Characteristics

The included studies had a total of 532 participants, with the sample size of each RCT
ranging from 10 to 72 participants. The mean age of participants of both genders ranged
from 34.3 to 83.6 years in the experimental groups and from 36.2 to 80.1 years in the control
groups. Types of stroke included hemorrhagic and ischemic strokes. Time since stroke
onset ranged from 7.6 to 69.4 months in the experimental groups and from 6.3 to 81.49
Life 2022, 12, 882 6 of 18

Table 2. Methodological quality of included studies on PEDro scale.

Scale Item

Measures of Precision and Variability

Comparisons between Groups
Comparison at the Baseline

Intention-to-Treat Analysis
Allocation Concealment

Adequate Follow-Up
Blinded Participants

Blinded Therapists

Blinded Evaluators
Randomization
Eligibility †

Total Score
Study/Author

Bang and Song, 2019 [22] X X X X X 5

Cheng et al., 2010 [23] X X X X X X 5
Hwangbo and Kim 2016 [24] X X X X X X 5
Joeng et al., 2012 [25] X X X X X 5
Kim and Kim 2018 [26] X X X X X X X 6
Kim and Kang 2018 [27] X X X X X 5
Kim and Kim, 2020 [28] X X X X X 5
Kim et al., 2015 [29] X X X X X X 5
Kim et al., 2020 [30] X X X X X 5
Kim et al., 2011 [31] X X X X X 5
Krukowska et al., 2016 [32] X X X X X X 5
Lee et al., 2012 [33] X X X X X X 5
Lim 2014 [34] X X X X X X X 6
Moon et al., 2010 [35] X X X X X X X 6
Park et al., 2016 [36] X X X X X X 5
Park 2017 [37] X X X X X X 5
Ribeiro et al., 2013 [38] X X X X X X X X 7
Seo et al., 2012 [39] X X X X X X X 6
Stephenson et al., 2014 [40] X X X X X X X 6
X: “yes”; †: does not contribute to total score.
Life 2022, 12, 882 7 of 18

3.3. Participant Characteristics

The included studies had a total of 532 participants, with the sample size of each
RCT ranging from 10 to 72 participants. The mean age of participants of both genders
ranged from 34.3 to 83.6 years in the experimental groups and from 36.2 to 80.1 years in the
control groups. Types of stroke included hemorrhagic and ischemic strokes. Time since
stroke onset ranged from 7.6 to 69.4 months in the experimental groups and from 6.3 to
81.49 months in the control groups.

3.4. Participant Inclusion Criteria

All studies selected patients with chronic stroke who had been diagnosed with stroke
at least six months earlier as a primary inclusion criteria. Of the nineteen studies ana-
lyzed, fifteen studies used stroke at the chronic stage for those who could understand and
follow the researcher0 s instructions according to Mini-Mental State Examination scores
(MMSE > 20 points) [22–28,30,31,34–36,39]. Two studies used the Brunnstrom scale [35,37]
and three studies used the Modified Ashworth Scale (MAS) [25,31,38] to describe the spastic
and involuntary muscle movement of lower limbs as inclusion criteria of a participant’s
motor severity. Twelve studies included participants who could walk with or without
assistance before treatment and ten studies selected participants who could follow simple
verbal instructions [22,26–31,34,35,37–40].

3.5. Interventions Administered

Six studies described the effects of PNF intervention alone, compared to either non-
intervention [40] or other treatment strategies [22,25,32,33,38], such as treadmill training [22,38],
neurodevelopmental treatment (NDT) [32], general exercise [25] and weight-bearing ex-
ercise [33]. In these studies, the duration of PNF intervention was from 20 to 30 min per
session, with the rehabilitation program including 12 to 36 sessions in total.
There were 13 studies investigating the effects of PNF combined with non-PNF strategies
in comparison to electrical stimulation [30,35], conventional physical therapy [24,31,34,39],
treadmill training [27], NDT [29], constraint-induced movement therapy [23], eye move-
ment programs [28,36] or taping administered alone [26,37]. The duration of these PNF-
based combination therapeutic approaches ranged from 30 to 105 min per session, with
the rehabilitation program including a range of sessions from 1 to 40 sessions. One study
reported superior immediate effects of PNF combined with kinesio taping on gait function
compared with using PNF or taping alone [37]. All the characteristics and reviews of the
19 studies are described in Table 3.
Diverse PNF approaches were employed in the included studies, among which six
studies used the PNF pelvic pattern for gait training [23,25,35,38–40] and three studies
used PNF sprinter and skater patterns for balance and gait velocity training [29,33,34].
Significant improvements in the balance and gait were reported in one study using NDT
combined with PNF underwater (involved the sprinter and skate patterns) at a temperature
of 32–34 ◦ C and depth of 100 cm when compared to using NDT alone [29].

Table 3. Characteristics and outcome reviews of included studies.

Number of
RCT Study Participants Grouping and Total Sessions Outcome Significant
Author, Year (Mean Age Intervention (Times/Week) Measures Improvement *
in Years) (Time in Minutes)

BBS and in terms

14 EG: PNF (20) of gait speed,
Bang and Song EG: 7 (58.86 ± 6.49) CG: treadmill (20) 20 Balance: BBS cadence, step
2019 [22] CG: 7 # PNF neck pattern (5 for 4 weeks) Gait parameters
(57.71 ± 6.70) length, and
double-limb
Life 2022, 12, 882 8 of 18

Table 3. Cont.

Number of
RCT Study Participants Grouping and Total Sessions Outcome Significant
Author, Year (Mean Age Intervention (Times/Week) Measures Improvement *
(Time in Minutes)
in Years)
EG: PNF + CIMT
(45) Upper limb
CG: CIMT (30-60) function and
64 # PNF trunk and
Cheng et al., EG: 32 (52.3 ± 9.5) 40 fine motor STEF, 10MWT and
2010 [23] CG: 32 limbs, gait (5 for 8 weeks) evaluation: STEF MBI: EG.
(51.7 ± 10.3) patterns with Gait velocity:
resistance, stretch, 10MWT
rhythmic stability, ADL: MBI
dynamic reversal
EG: PNF (10) +
general therapeutic Trunk stability:
40 exercise (20) 30 FRT, activities of
Kim et al., EG: 20 (51.4 ± 5.7) FRT soleus and
2011 [31] CG: general (5 for 6 weeks) Muscle activity:
CG: 20 (53.5 ± 7.1) quadriceps: EG
therapeutic EMG
exercise (30)
# PNFstabilizing
reversal and
Rhythmic
stabilization (sitting,
standing)
EG: PNF (30) + BBS, dynamic
Hwangbo and 30 traditional 30 sitting,
EG: 15 (59.4 ± 9.1) rehabilitation (30) Trunk control: TIS
Kim, 2016 [24] CG: 15 (55.9 ± 9.8) (5 for 6 weeks) Balance: BBS coordination and
CG: traditional
rehabilitation (60) TIS: EG
# PNF neck pattern
(sitting)
EG: treadmill with
PNF lower-leg
taping (30)
CG: treadmill with Balance: TUG
Kim and Kang 27 30 Walking ability: TUG, 10MWT and
EG: 14 (51.4 ± 2.6) placebo lower-leg (5 for 6 weeks) 6MWT: EG
2018 [26] CG: 13 (51.5 ± 2.9) taping (30) 10MWT and
# PNF flexion–
6MWT
adduction–external
rotation pattern
EG: PNF (15) +
treadmill (15)
CG: treadmill
training (30) Balance: TUG
23 # PNF scapular and 30 Walking ability:
Kim and Kim EG: 12 (60.8 ± 3.1) TUG, 10MWT and
2018 [27] CG: 11 (60.6 ± 3.4) pelvic patterns with (5 for 6 weeks) 10MWT and 6MWT: EG
hold-relax, 6MWT
contract-relax, and
dynamic reversal
(sidelying)
EG: PNF (30) +
functional
electrical
stimulation (30)
CG: general
10 physical therapy
EG: 5 20
Kim and Kim (70.61 ± 13.08) (30) + functional Balance: BBS, TUG Balance: BBS, TUG
2020 [28] electrical (5 for 4 weeks) Gait parameters Gait Velocity
CG: 5
(71.00 ± 6.02) stimulation (30)
# PNF bilateral lower
extremity
asymmetric
flexion/extension
patterns
Life 2022, 12, 882 9 of 18

Table 3. Cont.

Number of
RCT Study Participants Grouping and Total Sessions Outcome Significant
Author, Year (Mean Age Intervention (Times/Week) Measures Improvement *
(Time in Minutes)
in Years)

EG: PNF Balance: BBS and

20 underwater (8) + 30 FRT BBS, FRT, 10MWT
Kim et al., EG: 10 (65.9 ± 6.2) Gait velocity:
2015 [29] CG: 10 (64.1 ± 3.6) NDT (30) (5 for 6 weeks) and TUG: EG
CG: NDT (30) 10MWT
Gait function: TUG
# PNF sprinter and
skate patterns
(standing,
underwater)
EG: eye movement
30
EG: 15 (56.8 ± 3.44) (15) + PNF (15) + Trunk control: TIS
Kim et al., conservative 24 TIS, COP, LOS,
CG: 15 (3 for 8 weeks) Balance: COP, LOS, BBS, FRT
2020 [30] (57.53±3.59) treatment (30) BBS, FRT
CG: conservative
treatment (60)
# PNF neck pattern
movement training
EG: PNF † Balance: force
72 platform (COP of
Krukowska et al., EG: 34 (52.7 ± 7.5) CG: Bobath-NDT † 36 Movement of COP:
2016 [32] # PNF diagonal and (6 for 6 weeks) field support and CG
CG: 39 (52.7 ± 6.3)
spiral patterns total path length)
EG1: PNF (30)
EG2: functional
weight bearing Weight bearing
27 exercise (30) and static balance
EG1: 9 (49.1 ± 9.0) CG: general capability: FICSIT-4, GBS:
12 EG1, EG2
Lee et al., 2012 [33] EG2: 10 exercise (30) (3 for 4 weeks) FICSIT-4, force FSST, TUG: EG1,
(51.7 ± 17.4) # PNF sprinter platform EG2
CG1: 9 (44.8 ± 8.6) pattern (side lying, Dynamic balance:
half standing, FSST and TUG
modified plantigrade
posture)
EG: PNF (15) +
conventional
physical therapy
(35)
22 CG: conventional Balance: FRT and
physical therapy 20 FRT, BBS and TUG:
Lim 2014 [34] EG: 11 (55.5 ± 5.4) (5 for 4 weeks) BBS
CG: 11 (56.4 ± 5.7) Gait function: TUG EG
(50)
# PNF pattern
sprinter and skater
(sitting and half
standing)
aPNFG: PNF (30)
cPNFG: PNF (15) + Upper limb
15 ES (15) functions: MFT
aPNFG: 5 CG: ES (30) Gait velocity:
Moon et al., (49.8 ± 2.9) # PNF scapula + 30 Upper limb
2010 [35] cPNFG: 5 (5 for 6 weeks) walking
function: aPNFG
(53.4 ± 2.5) upper limb- distance/minutes
CG: 5 (52.4 ± 7.1) combining-isotonic- Weight bearing:
dynamic reversal force platform
(side lying, sitting)
Balance: force
platform (Static and
dynamic balance:
EG: PNF (30 times) sway length and
+ EMP (20 times) † area with eye
CG: EMP (20 Static balance with
20 30 open/closed; eye closed: EG
Park et al.,
2016 [36]
EG: 10 (61.1 ± 8.2) times) † (5 for 6 weeks) Dynamic balance: Head alignment:
CG: 10 (60.2 ± 7.9) # PNF neck pattern limit of stability EG
with contract-relax with
(sitting) forward/backward
and left/right)
Head alignment:
GPS
Life 2022, 12, 882 10 of 18

Table 3. Cont.

Number of
RCT Study Participants Grouping and Total Sessions Outcome Significant
Author, Year (Mean Age Intervention (Times/Week) Measures Improvement *
(Time in Minutes)
in Years)
20
aPNFG: 7 aPNFG: PNF (30) Cadence, speed,
(57.3 ± 9.4) cPNFG: PNF (30) +
1 Gait parameters
Park 2017 [37] cPNFG: 7 kinesio taping and stride length:
(51.7 ± 6.5) CG: kinesio taping cPNFG
CG: 6 (64.8 ± 15.2)
#PNF sprinter and
skater patterns
(sitting, standing)
EG: PNF (30) Motor recovery
CG: treadmill with Ankle dorsiflexion
20 12 and basic mobility: during swing
Ribeiro et al., EG: 9 (58.3 ± 8.9) partial body
2013 [38] (4 for 3 weeks) STREAM
CG: 11(56.5 ± 8.3) weight support ADL: Motor FIM phase: EG
(30) Gait parameters
# PNF scapular and
pelvic patterns
(sidelying, sitting
and standing with
stretching and
maximum
resistance)
EG: PNF-based
walking exercise
(30) + general
physical therapy Gait function: All parameters of
40 temporal, spatial
EG: 20 (64.1 ± 3.2) 20
Seo et al., 2012 [39] (30) (5 for 4 weeks) gait performance
CG: 20 (65.8 ± 6.0) CG: general parameters and
FAP and FAP: EG
physical therapy
(30) + walking
exercise (30)
# PNF gait training

EG: PNF gait

training (30)
CG1: treadmill
18 with body weight Gait velocity and Gait velocity and
Stephenson et al., EG: 6 (63.3 ± 12.4) support (20) 12
CG1: 6 (55.0 ± 9.4) cadence: 10MWT cadence: EG > CG1
2014 [40] CG2: no (3 for 4 weeks) Gait disability: WGS total score:
CG2: 6 interventions
(63.8 ± 12.2) # PNF pelvic and WGS EG
lower extremity
patterns and gait
training
The 10-m walking test (10MWT); 6-minute walk test (6MWT); activity of daily living (ADL); Berg balance scale
(BBS); center of pressure (COP); constraint-induced movement therapy (CIMT); Dynamic gait index (DGI);
eye movements program (EMP); Figure 8 walking test (F8W); four Square step test (FSST); frailty and injuries
cooperative studies of intervention techniques (FICSIT-4); functional ambulation performance (FAP); functional
independence measure (FIM); functional reach test (FRT); global postural system (GPS); good balance system
(GBS); modified Barthel index (MBI); PNF-alone group (aPNFG); PNF-combined group (cPNFG); simple test for
evaluating hand function (STEF); the stroke rehabilitation assessment of movement (STREAM); timed up-and-go
test (TUG); trunk impairment scale (TIS). # PNF techniques in italic * Indicates significant differences compared to
other groups. † Indicates no time data provided by authors.

3.6. Outcome Measures

Outcome measures related to motor function, including balance and gait, were used in
all the included studies. Identical measurements of BBS, TUG, FRT and 10WMT on balance
and gait function were performed in twelve studies [22–31,33,34] on which meta-analysis
was conducted for this review study (Table 3).

3.6.1. Balance
Outcome measures, namely, the frailty and injuries cooperative studies of interven-
tion techniques (FICSIT-4, tests of static balance), four square step test [33], measure-
Life 2022, 12, 882 11 of 18

ment of center of pressure (COP) and velocity moment from force platform [28,32,36];
BBS [22,24,28–30,34]; FRT [28,29,31,34] and TUG [26,27,29,30,33,34] were adopted by the
included studies to assess balance function. The findings in these studies revealed signifi-
cant differences in the balance function before and after PNF intervention [22,24,26–34,36]
and between patients receiving PNF and controls [22,24,26–30,34].

Berg Balance Scale

The BBS, which has proven reliability and validity, was used to examine the ability
of patients to balance before and after the treatment. The BBS consists of 14 items, each
of which has a minimum score of 0 and a maximum score of 4; the maximum score is
56. A higher score of BBS indicates good steady state and proactive balance [41]. There
were six studies showing significant improvements in BBS scores in patients with chronic
stroke receiving PNF-based intervention, compared to those undergoing non-PNF inter-
ventions [22,24,28–30,34]. Meta-analysis results show significant differences in BBS for
Life 2022, 12, x FOR PEER REVIEW participants in the PNF group (n = 63) compared to the control group (n = 63) (MD 12 of 19
= 2.90,
95% CI: 1.97~3.84, p < 0.001) with low statistical heterogeneity (p = 0.25, I2 = 25%, Figure 3).

Figure 3. Forest plot of effect

Figureof3.PNF intervention
Forest on BBS
plot of effect scores.
of PNF Abbreviations:
intervention on BBSIV: inverse
scores. variance; CI:IV:
Abbreviations: confidence interval;
inverse variance;
SD: standard deviation.CI: Refs. [22,24,28–30,34]
confidence interval; SD: standard deviation. Refs. [22,24,28–30,34].

Functional
Functional Reach
Reach Test
Test
The
The FRT
FRT measures
measures the the distance
distance (in
(in centimeters)
centimeters) between
between thethe start
start and
and end
end positions
positions
while standing independently, raising an arm 90 ◦ from the torso and reaching out with-
while standing independently, raising an arm 90° from the torso and reaching out without
out losing
losing balance
balance (i.e.,(i.e.,
takingtaking a step).
a step). A far A far distance
distance of the of
FRT the FRT indicates
indicates good proac-
good proactive bal-
tive balance [42]. Of the 19 studies included, 4 showed significant improvements
ance [42]. Of the 19 studies included, 4 showed significant improvements in the maximal in the
maximal
horizontalhorizontal
distances distances
for FRT infor FRT in
patients patients
with chronic with chronic
stroke after stroke after intervention
PNF-based PNF-based
intervention compared to those before PNF treatment [31] and in controls
compared to those before PNF treatment [31] and in controls after non-PNF interventions after non-PNF
interventions [28,29,34].
[28,29,34]. The The meta-analyses
meta-analyses showed significant
showed significant changes inchanges in FRT performance
FRT performance between
between participants in the PNF group (n = 56) and controls (n = 56) (MD = 2.49 cm, 95% CI:
participants in the PNF group (n = 56) and controls (n =2 56) (MD = 2.49 cm, 95% CI: 0.55–
0.55–4.43, p = 0.01) with higher heterogeneity (p = 0.005,2I = 77%, Figure 4A). The observed
4.43, p = 0.01) with higher heterogeneity (p = 0.005, I = 77%, Figure 4A). The observed
heterogeneity was attributed to the magnitude of the study of Kim et al. [29]. After ex-
heterogeneity was attributed to the magnitude of the study of Kim et al. [29]. After ex-
cluding the study of Kim et al. [29], the overall pooled effect was enhanced (MD = 3.40 cm,
cluding the study of Kim et al. [29], the overall pooled effect was enhanced (MD = 3.40 cm,
95% CI: 2.30–4.50, p < 0.05) with low heterogeneity (p = 0.61, I22 = 0%, Figure 4B).
95% CI: 2.30–4.50, p < 0.05) with low heterogeneity (p = 0.61, I = 0%, Figure 4B).
Timed up and Go Test
The TUG test is a performance-based measure of balance and functional mobility. The
time taken to sit on an armchair, stand up at the starting signal, walk 3 m and return to
the sitting position is measured. The scores of ten seconds or less indicate normal mobility
and balance, 11–20 s is within the normal limits for frail, elderly and disabled patients [42].
Of the 19 included studies, 6 studies showed significant differences in TUG in patients
with chronic stroke after PNF-based intervention compared to those before treatment [33]
and controls with non-PNF interventions [26,27,29,30,34]. Meta-analysis results revealed
significant differences between participants in the PNF group (n = 61) and controls (n = 59)
 ance [42]. Of the 19 studies included, 4 showed significant improvements in the maximal
horizontal distances for FRT in patients with chronic stroke after PNF-based intervention
compared to those before PNF treatment [31] and in controls after non-PNF interventions
[28,29,34]. The meta-analyses showed significant changes in FRT performance between
Life 2022, 12, 882 participants in the PNF group (n = 56) and controls (n = 56) (MD = 2.49 cm, 95% CI:12 0.55–
of 18
4.43, p = 0.01) with higher heterogeneity (p = 0.005, I2 = 77%, Figure 4A). The observed
heterogeneity was attributed to the magnitude of the study of Kim et al. [29]. After ex-
cluding
(MD = −the study
2.25 of CI:
s, 95% Kim−et al. [29],
3.16~ thepoverall
−1.35, < 0.001)pooled effect
with low was enhanced
heterogeneity (p =(MD
0.35,= I3.40 cm,
2 = 10%,
95% CI:5).
Figure 2.30–4.50, p < 0.05) with low heterogeneity (p = 0.61, I = 0%, Figure 4B).
2

Life 2022, 12, x FOR PEER REVIEW 13 of 19

to the sitting position is measured. The scores of ten seconds or less indicate normal mo-
bility and balance, 11–20 s is within the normal limits for frail, elderly and disabled pa-
tients [42]. Of the 19 included studies, 6 studies showed significant differences in TUG in
patients with chronic stroke after PNF-based intervention compared to those before treat-
ment [33] and controls with non-PNF interventions [26,27,29,30,34]. Meta-analysis results
Figure 4. (A) Forest plot of effect
Figure of PNF
4. (A) Forestintervention on FRT.
plot of effect Refs
of PNF [28,29,31,34];
intervention on(B) sensitivity
FRT. analysis of(B)
Refs [28,29,31,34]; effect of PNF
sensitivity
revealed significant differences between participants in the PNF group (n = 61) and con-
intervention on FRT. analysis
Abbreviations:
of effect IV: inverse
of PNF variance;onCI:
intervention FRT.confidence interval;
Abbreviations: SD: standard
IV: inverse deviation.
variance; Refs.
CI: confidence
[29,30,31,34] trols (n = 59) (MD = −2.25 sec, 95% CI: −3.16~−1.35, p < 0.001) with low heterogeneity (p =
interval; SD: standard deviation. Refs. [29–31,34].
0.35, I2 = 10%, Figure 5).
Timed Up and Go Test
The TUG test is a performance-based measure of balance and functional mobility.
The time taken to sit on an armchair, stand up at the starting signal, walk 3 m and return

Figure
Figure 5.
5. Forest
Forest plot
plot of
of effect
effect of
of PNF
PNF intervention
intervention on
on TUG
TUG test.
test. Abbreviations:
Abbreviations: IV:
IV: inverse
inverse variance;
variance;
CI: confidence interval; SD: standard deviation. Refs [26,28,29,33,34].
CI: confidence interval; SD: standard deviation. Refs [26,28,29,33,34].

3.6.2. Gait
Outcome measures,
measures, namely,
namely,10MWT
10MWT[23,25–27,29,40],
[23,25–27,29,40],TUGTUG [29,34], 6-minute
[29,34], 6-minute walking
walk-
test test
ing [26,27], walking
[26,27], distance
walking per
distance perminute
minute[35],
[35],kinematic
kinematicgait
gait parameters [22,30,37–39],
[22,30,37–39],
functional ambulation performance
performance [39], Wisconsin
Wisconsin gait
gait scale [40], dynamic gait index and
Figure 8 walking test [25] were adopted by the includedincluded studies
studies toto assess
assess gait
gait function.
function.
The findings
findingsin inthese
thesestudies
studies reveal significant differences in gait function before
reveal significant differences in gait function before and afterand
after PNF intervention [22,23,25–30,34,35,37–40] and between patients receiving
PNF intervention [22,23,25–30,34,35,37–40] and between patients receiving PNF and con- PNF and
controls [22,23,25–27,29,30,34,39].
trols [22,23,25–27,29,30,34,39].

10-Meter Walking
10-Meter Walking Test
Test
The 10MWT is aaperformance
The 10MWT is performancemeasure
measure used
used to to assess
assess gaitgait velocity
velocity in meters
in meters per
per sec-
second over a 10 m distance. It can be employed to determine functional mobility and
ond over a 10 m distance. It can be employed to determine functional mobility and gait.
gait. Good walking-speed performance required for the subject to walk 10 m on a course
Good walking-speed performance required for the subject to walk 10 m on a course indi-
indicates good functional mobility in individuals with chronic stroke, also as a practical
cates good functional mobility in individuals with chronic stroke, also as a practical and
and informative functional sixth “vital sign” for all patients [43]. Of the 19 included
informative functional sixth “vital sign” for all patients [43]. Of the 19 included studies, 5
studies, 5 showed significant differences in 10MWT performance before and after PNF
showed significant differences in 10MWT performance before and after PNF intervention
and between patients with chronic stroke treated with PNF and non-PNF interventions
[23,25–27,29]. Meta-analysis results also revealed significant differences in 10MWT per-
formance between patients with chronic stroke in the PNF group (n = 78) and controls (n
= 78) (MD = −2.15 sec, 95% CI: −2.87~−1.43, p < 0.001) with low heterogeneity (p = 0.08, I2 =
 trols [22,23,25–27,29,30,34,39].

10-Meter Walking Test

The 10MWT is a performance measure used to assess gait velocity in meters per sec-
ond over a 10 m distance. It can be employed to determine functional mobility and gait.
Life 2022, 12, 882
Good walking-speed performance required for the subject to walk 10 m on a course indi-13 of 18
cates good functional mobility in individuals with chronic stroke, also as a practical and
informative functional sixth “vital sign” for all patients [43]. Of the 19 included studies, 5
showed significant
intervention differences
and between in 10MWT
patients withperformance before
chronic stroke and after
treated withPNF intervention
PNF and non-PNF
and between patients with chronic stroke treated with PNF and non-PNF interventions
interventions [23,25–27,29]. Meta-analysis results also revealed significant differences in
[23,25–27,29]. Meta-analysis results also revealed significant differences in 10MWT per-
10MWT performance between patients with chronic stroke in the PNF group (n = 78) and
formance between patients with chronic stroke in the PNF group (n = 78) and controls (n
controls (n = 78) (MD = −2.15 s, 95% CI: −2.87~−1.43, p < 0.001) with low heterogeneity
= 78) (MD 2 == 0%,
−2.15 sec, 95% CI: −2.87~−1.43, p < 0.001) with low heterogeneity (p = 0.08, I2 =
(p =0%,
0.08,
Figure 6). Figure 6).
I

Figure 6. Forest plot of effect of PNF intervention on 10MWT. Abbreviations: IV: inverse variance; CI:
confidence interval; SD: standard deviation. Refs. [23,25–27,29].

4. Discussion
The current study exclusively focused on the chronic stroke population. To our
knowledge, this study is the first systematic review and meta-analysis examining the effects
of PNF-based physical therapy on the improvement of balance and gait function in patients
with chronic stroke. A previous systematic review and meta-analysis concerning four
studies using trunk PNF patterns demonstrated positive effects of PNF on trunk control
and balance in both the acute and subacute stages of stroke [44]. Another systematic review
with five included studies suggested that PNF improved gait parameters in patients with
stroke [11]. The current meta-analysis provides evidence supporting the beneficial effects
of the PNF-based physical therapy approach on the improvement of balance and gait
velocities with many specific PNF patterns and techniques by assessing 10MWT, BBS, FRT
and TUG in patients with chronic stroke. While our findings on balance and gait functions
are comparable with the previous results from pooled patients with stroke, mainly at the
acute and subacute stages [11,44], the current review demonstrated the positive effects of
PNF intervention in strengthening the impaired balance and gait in patients with stroke,
specifically at the chronic stage.
This present meta-analysis with a statistical evidence for BBS, FRT and TUG mea-
surements shows that the potential PNF patterns and techniques adopted in the included
studies are appropriate for improving static and dynamic balance abilities during postural
changes and mobility in patients with stroke, specifically at the chronic stage. Among
those studies using BBS, FRT and TUG measurements, diversified PNF patterns were used,
including PNF sprinter and skater exercise [25,29,34], neck pattern [22,24,28], scapular
and pelvic patterns in a side-lying position [27] and both leg patterns [30]. The positive
outcomes obtained suggest that PNF could facilitate core muscle control, which in turn
improves balance through coordination movement and enhances balance ability by stimu-
lating a proprioceptive sense of muscles and tendons [45]. The patterns of PNF exercises
have a spiral, diagonal direction, which further emphasizes the functional training on
trunk stability aiming to enhance balance in a lateral direction [46,47]. The lateral balance
of trunk control, which was more affected by stroke than balance in the anteroposterior
direction, seems to be a primary target for rehabilitation [48]. BBS and TUG can provide
the clinical validity of balance capacity measures, including the performance of lateral,
static and dynamic balance control [48–50]. In line with the previous results, BBS, FRT
and TUG in patients with chronic stroke were found to improve after PNF. Moreover, the
beneficial effects of PNF therapy on gait function in patients with chronic stroke were
also observed, especially in a walking speed of 10MWT as revealed in the meta-analysis
results [23,25–27,29,40]. Taken together, the findings suggest that PNF intervention in-
creases lateral, static and dynamic balance to promote functional balance and the mobility
Life 2022, 12, 882 14 of 18

of patients with stroke at the chronic stage. The previous studies have shown that balance
may be a predictor of gait performance in patients with chronic hemiparetic stroke [51,52],
implying a strong correlation between balance and gait parameters. There were four studies
in this review presenting the significant effect of PNF on the improvement in balance and
gait speed for patients with chronic stroke [26,27,29,34].
Previous studies have shown that the pelvic pattern of PNF helps to improve control
of the pelvis, which is crucial for maintaining trunk control, gait and balance through
the stimulation of muscle and joint proprioception [53]. Specific core-stability training
for patients with stroke would improve not only trunk function, but also balance and
mobility. Moreover, it would lead to greater improvement compared to a conventional
comprehensive rehabilitation program [54]. Among those studies on gait function, the
pelvic pattern of PNF was commonly used in gait training programs [23,25,35,38–40],
which aim to increase core stability for promoting ambulation in patients with stroke.
Several included studies also demonstrated that PNF pattern exercise using sprinter and
skater also contributed to enhance the balance and gait functions in patients with chronic
stroke [29,33,34,37]. Regarding the duration of the PNF treatment program, a 30-min PNF
intervention at least for 12 sessions in most of the included studies has been shown to
improve balance and gait abilities in patients with chronic stroke [24,25,33,37,39]. In view of
these findings, long-term PNF intervention aiming to promote trunk control and lower-limb
strength was recommended for increasing balance ability and walking speed in patients
with chronic stroke. PNF can still benefit patients by enhancing their balance and gait
abilities at more than 6 months after stroke onset. Hence, the inclusion of PNF in a routine
treatment regime of chronic stroke individual can be supported.
Consistent with the previous reviews, the emphasis of this work was on balance and
gait functions as major deficits of chronic stroke hampering functional recovery in neurore-
habilitation [55–57]. Comparative studies with alternative PNF treatments were analyzed
in this review, which would shed light on the significant differences of PNF interventions
on motor impairments by conducting kinematic parameters, subjective reports or objective
measures and activity limitations examined during the chronic stages of recovery. These
findings need to be further integrated into current practice recommendations.
Evidence for stroke rehabilitation relating to walking ability, postural control, muscle
strength and functional recovery is becoming increasingly available in the form of high-
quality RCTs that can inform clinical guidelines as well as high-level government strategies
with respect to stroke [58]. Individualized, patient-centered, evidence-based physical
treatments, with consideration of all the available treatment components, should be selected
by using a mix of components from different approaches. PNF is one of the effective
physical interventions to decrease muscle spasticity and improve lower-limb function
and gait speed in post-stroke survivors, as well as cycling, treadmill exercise, functional
electrical stimulation and deep dry needling [8,9,59].

4.1. Limitations of Included Studies

Of the 19 included studies, 6 were of good quality while 10 were of fair quality ac-
cording to the PEDro scale. Insufficient good-quality studies may influence the precision,
applicability and confidence of our results and recommendations. Hence, more quality
studies should be included in future research. Moreover, the included studies have the
following limitations. First, most studies had a small sample size. Second, the RCTs
involved experimental groups receiving PNF with different intervention periods, interven-
tion times, environments and more diverse tasks, which had variations in intervention
implementation, making it difficult to interpret the isolated effects of PNF in the treatment
of this population. Third, biased results arose from ignoring minimal clinically important
differences in each assessment in each trial. It may not seem as though the difference meets
the threshold for minimal clinically important differences across the assessments; however,
there were statistically significant differences between the experimental and control groups.
Fourth, follow-up observation reports were lacking. In view of these limitations, the current
Life 2022, 12, 882 15 of 18

results should be interpreted with caution and cannot be generalized to all patients with
chronic stroke.

4.2. Strengths and Weaknesses of This Review

The current study is an updated systematic review and meta-analysis on the literature
published from 1960 to 2021 addressing PNF for people with stroke at a chronic stage. The
included studies were not restricted to RCTs published in English, thus offering a more
comprehensive understanding on the effect of PNF intervention in the published literature
related to balance and gait in patients with chronic stroke.
However, this review has significant limitations. Out of 19 analyzed RCTs, 15 were
conducted in Korea and 4 studies in the United States, China, Brazil and Poland, respec-
tively, which may limit generalizability. The included studies covered only RCTs; not
including the published clinical trials may be a significant miss and underestimates the
effects of PNF on patients with chronic stroke. The small number of RCTs reflects the
difficulties in conducting such investigations on patients with stroke at the chronic stage,
which may be related to the PNF technique, the study design itself or patient-related or
ethical issues. Publication bias may also be another reason for the small sample. Second,
different methodologies for evaluating balance and gait functions were used, making it
impossible to conduct a meta-analysis of all the selected studies. This reflects the variation
in assessments of balance and gait functions affected by PNF interventions in chronic stroke.
We attempted to maintain an acceptable level of rigor and quality to recommendations
through a similar process of clinical and scientific inquiry to examine the current literature.

4.3. Implications for Clinical Practice and Research

At more than 6 months after stroke onset, PNF can still benefit patients by enhancing
their balance and gait abilities. In particular, PNF focusing on trunk control in patients
with stroke can improve their balance function and ability to walk. Using PNF pelvic, neck,
sprinter and skater patterns for trunk control in various positions with isotonic, dynamic
reversal, stretching, reversal stabilization or resisted techniques can provide strategy im-
plementation for clinical practice on balance and gait functions at the chronic stage for
patients with stroke. It is worth noting that the effect of PNF on patients with chronic stroke
is not based on short-term effects, but requires long-term continuous treatment to obtain
more persistent benefits. The findings of this review indicate that there were statistically
significant improvements to the balance and gait speed for patients with chronic stroke.
However, a statistically significant difference does not necessarily mean that it ensures a
clinically meaningful difference as well. Further high-quality research with the best avail-
able evidence is required for concluding a consensus of clinical intervention and research
on PNF.

5. Conclusions
The most significant recovery of movement is generally considered to occur within the
first six months following a stroke, with spontaneous recovery slowing down after that time.
That does not mean that patients with chronic stroke should be prevented from addressing
intensive therapy for motor recovery after six months. The results from this systematic
review with meta-analysis suggest that PNF-based physical therapy has statistical effects
on the improvement of balance and gait speed in individuals at least 6 months after a
stroke. Although positive statistical effects were found in this study, more advancing
rehabilitation studies with the sciences of neuroplasticity are needed for concluding a
consensus of the clinical and research significances of data regarding the relative efficacy of
PNF, including the technique components, dosage parameters and practice conditions in
future meta-analyses.

Author Contributions: Conceptualization, L.-W.C. and P.T.N.; methodology, P.T.N.; software, P.T.N.;
validation, Y.-L.H.; formal analysis, Y.-L.H.; data curation, Y.-L.H.; writing—original draft preparation,
Life 2022, 12, 882 16 of 18

P.T.N. and Y.-L.H.; writing—review and editing, Y.-L.H. All authors have read and agreed to the
published version of the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Adamson, J.; Beswick, A.; Ebrahim, S. Is stroke the most common cause of disability? J. Stroke Cerebrovasc. Dis. Off. J. Natl. Stroke
Assoc. 2004, 13, 171–177. [CrossRef] [PubMed]
2. Bushnell, C.; Bettger, J.P.; Cockroft, K.M.; Cramer, S.C.; Edelen, M.O.; Hanley, D.; Katzan, I.L.; Mattke, S.; Nilsen, D.M.; Piquado,
T.; et al. Chronic stroke outcome measures for motor function intervention trials: Expert panel recommendations. Circ. Cardiovasc.
Qual. Outcomes 2015, 8 (Suppl. 3), S163–S169. [CrossRef]
3. Kollen, B.; van de Port, I.; Lindeman, E.; Twisk, J.; Kwakkel, G. Predicting improvement in gait after stroke: A longitudinal
prospective study. Stroke 2005, 36, 2676–2680. [CrossRef] [PubMed]
4. Allen, J.L.; Kesar, T.M.; Ting, L.H. Motor module generalization across balance and walking is impaired after stroke. J. Neurophysiol.
2019, 122, 277–289. [CrossRef] [PubMed]
5. Briggs, R.; O’Neill, D. Chronic stroke disease. Br. J. Hosp. Med. 2016, 77, C66–C69. [CrossRef]
6. Beyaert, C.; Vasa, R.; Frykberg, G.E. Gait post-stroke: Pathophysiology and rehabilitation strategies. Neurophysiol. Clin. 2015, 45,
335–355. [CrossRef]
7. Jamal, K.; Leplaideur, S.; Rousseau, C.; Chochina, L.; Moulinet-Raillon, A.; Bonan, I. Disturbances of spatial reference frame and
postural asymmetry after a chronic stroke. Exp. Brain Res. 2018, 236, 2377–2385. [CrossRef]
8. Hakakzadeh, A.; Shariat, A.; Honarpishe, R.; Moradi, V.; Ghannadi, S.; Sangelaji, B.; Nakhostin Ansari, N.; Hasson, S.; Ingle, L.
Concurrent impact of bilateral multiple joint functional electrical stimulation and treadmill walking on gait and spasticity in
post-stroke survivors: A pilot study. Physiother. Theory Pract. 2019, 37, 1368–1376. [CrossRef]
9. Shariat, A.; Nakhostin Ansari, N.; Honarpishe, R.; Moradi, V.; Hakakzadeh, A.; Cleland, J.A.; Kordi, R. Effect of cycling and
functional electrical stimulation with linear and interval patterns of timing on gait parameters in patients after stroke: A
randomized clinical trial. Disabil. Rehabil. 2021, 43, 1890–1896. [CrossRef]
10. Guiu Tula, F.X.; Cabanas Valdes, R.; Sitja Rabert, M.; Urrutia, G.; Gomara Toldra, N. The Efficacy of the proprioceptive
neuromuscular facilitation (PNF) approach in stroke rehabilitation to improve basic activities of daily living and quality of life: A
systematic review and meta-analysis protocol. BMJ Open 2017, 7, e016739. [CrossRef]
11. Gunning, E.; Uszynski, M.K. Effectiveness of the Proprioceptive Neuromuscular Facilitation method on gait parameters in
patients with stroke: A systematic review. Arch. Phys. Med. Rehabil. 2019, 100, 980–986. [CrossRef] [PubMed]
12. Cayco, C.S.; Gorgon, E.J.R.; Lazaro, R.T. Effects of proprioceptive neuromuscular facilitation on balance, strength, and mobility of
an older adult with chronic stroke: A case report. J. Bodyw. Mov. Ther. 2017, 21, 767–774. [CrossRef]
13. Gabriel, D.A.; Kamen, G.; Frost, G. Neural adaptations to resistive exercise: Mechanisms and recommendations for training
practices. Sports Med. 2006, 36, 133–149. [CrossRef] [PubMed]
14. Shimura, K.; Kasai, T. Effects of proprioceptive neuromuscular facilitation on the initiation of voluntary movement and motor
evoked potentials in upper limb muscles. Hum. Mov. Sci. 2002, 21, 101–113. [CrossRef]
15. Cayco, C.S.; Gorgon, E.J.R.; Lazaro, R.T. Proprioceptive neuromuscular facilitation to improve motor outcomes in older adults
with chronic stroke. Neurosciences 2019, 24, 53–60. [CrossRef] [PubMed]
16. Ribeiro, T.S.; de Sousa e Silva, E.M.; Sousa Silva, W.H.; de Alencar Caldas, V.V.; Silva, D.L.; Costa Cavalcanti, F.A.; Lindquist, A.R.
Effects of a training program based on the proprioceptive neuromuscular facilitation method on post-stroke motor recovery: A
preliminary study. J. Bodyw. Mov. Ther. 2014, 18, 526–532. [CrossRef]
17. Jeun, Y.J. Meta-analysis of the Effect of Proprioceptive Neuromuscular Facilitation Training on Gait Ability in Patients with Storke.
J. Korea Soc. Comput. Inf. 2021, 26, 165–172.
18. Moher, D.; Shamseer, L.; Clarke, M.; Ghersi, D.; Liberati, A.; Petticrew, M.; Shekelle, P.; Stewart, L.A.; Group, P.-P. Preferred
reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst. Rev. 2015, 4, 1. [CrossRef]
19. Foley, N.C.; Teasell, R.W.; Bhogal, S.K.; Speechley, M.R. Stroke Rehabilitation Evidence-Based Review: Methodology. Top. Stroke
Rehabil. 2003, 10, 1–7. [CrossRef]
20. Foley, N.C.; Bhogal, S.K.; Teasell, R.W.; Bureau, Y.; Speechley, M.R. Estimates of quality and reliability with the physiotherapy
evidence-based database scale to assess the methodology of randomized controlled trials of pharmacological and nonpharmaco-
logical interventions. Phys. Ther. 2006, 86, 817–824. [CrossRef]
21. Borenstein, M.; Hedges, L.V.; Higgins, J.P.; Rothstein, H.R. A basic introduction to fixed-effect and random-effects models for
meta-analysis. Res. Synth. Methods 2010, 1, 97–111. [CrossRef] [PubMed]
Life 2022, 12, 882 17 of 18

22. Bang, D.H.; Song, M.S. The effect of neck pattern of PNF on balance and walking ability in patients with chronic stroke. PNF Mov.
2019, 17, 47–56.
23. Cheng, K.; Yin, L.C.; Tsai, F.; Hao, S.B. Effect of improved PNF technique on lower limb function of patients with cerebral
apoplexy hemiplegia. China Modern Doctor. 2010, 48, 162–163. (In Chinese)
24. Hwangbo, P.N.; Kyoung, D.K. Effects of proprioceptive neuromuscular facilitation neck pattern exercise on the ability to control
the trunk and maintain balance in chronic stroke patients. J. Phys. Ther. Sci. 2016, 28, 850–853. [CrossRef]
25. Jeong, W.S.; Park, S.K.; Park, J.H.; Lee, H.G.; Kim, K.Y. Effect of PNF combination patterns on muscle activity of the lower
extremities and gait ability in stroke patients. J. Korea Cont. Assoc. 2012, 12, 318–328. (In Korean) [CrossRef]
26. Kim, B.R.; Kang, T.W. The effects of proprioceptive neuromuscular facilitation lower-leg taping and treadmill training on mobility
in patients with stroke. Int. J. Rehabil. Res. Int. Z. Fur Rehabil. Rev. Int. De Rech. De Readapt. 2018, 41, 343–348. [CrossRef]
27. Kim, C.H.; Kim, Y.N. Effects of proprioceptive neuromuscular facilitation and treadmill training on the balance and walking
ability of stroke patients. J. Korea Phys. Ther. 2018, 30, 79–83. [CrossRef]
28. Kim, D.H.; Kim, K.H.; Lee, S.M. Effects of eye movement with PNF neck movement on trunk stability and standing position
balance ability of chronic stroke patients. Phys. Med. 2020, 30, 206–213. [CrossRef]
29. Kim, K.; Lee, D.K.; Jung, S.I. Effect of coordination movement using the PNF pattern underwater on the balance and gait of stroke
patients. J. Phys. Ther. Sci. 2015, 27, 3699–3701. [CrossRef]
30. Kim, S.M.; Kim, Y.M. Effect of both lower extremities proprioceptive neuromuscular facilitation training with functional electrical
stimulation on the balance and gait of stroke patient: A randomized controlled trial. Korean Soc. Phys. Med. 2020, 15, 123–132.
[CrossRef]
31. Kim, Y.H.; Kim, E.J.; Gong, W.T. The effects of trunk stability exercise using PNF on the functional reach test and muscle activities
of stroke patients. J. Phys. Ther. Sci. 2011, 23, 699–702. [CrossRef]
32. Krukowska, J.; Bugajski, M.; Sienkiewicz, M.; Czernicki, J. The influence of NDT-Bobath and PNF methods on the field support
and total path length measure foot pressure (COP) in patients after stroke. Neurol. Neurochir. Pol. 2016, 50, 449–454. [CrossRef]
[PubMed]
33. Lee, J.S.; Nam, K.W.; Kim, K.Y.; Yoon, J.W.; Park, J.H. Effect of weight bearing exercise on weight bearing and balance for patients
with chronic stroke. J. Korean Soc. Phys. Ther. 2012, 24, 253–261. (In Korean)
34. Lim, C.G. The effects of proprioceptive neuromuscular facilitation (PNF) pattern exercise using the sprinter and the skater on
balance and gait function in the stroke patients. J. Korean Soc. Phys. Ther. 2014, 26, 249–256. (In Korean)
35. Moon, S.H.; Hong, W.S.; Kim, S.S.; An, H.J.; Song, Y.H.; Kim, Y.K.; Choi, J.H.; Kim, B.K.; Kim, S.H.; Choi, W.S. The impact of
functional electrical stimulus and proprioceptive neuromuscular facilitation to scapula adductor on upper limb functions and
gait of the patients with stroke. J. Int. Acad. Phys. Ther. Res. 2010, 1, 143–148.
36. Park, S.E.; Min, K.O.; Lee, S.B.; Choi, W.S.; Kim, S.H. Effect of eye movements and proprioceptive neuromuscular facilitation on
balance and head alignment in stroke patients with neglect syndrome. J. Phys. Ther. Sci. 2016, 28, 596–601. [CrossRef]
37. Park, S.J. The immediate effects of proprioceptive neuromuscular facilitation with taping on gait parameters in patients with
chronic stroke. J. Phys. Ther. Sci. 2017, 29, 2018–2021. [CrossRef]
38. Ribeiro, T.; Britto, H.; Oliveira, D.; Silva, E.; Galvao, E.; Lindquist, A. Effects of treadmill training with partial body weight
support and the proprioceptive neuromuscular facilitation method on hemiparetic gait: A randomized controlled study. Eur. J.
Phys. Rehabil. Med. 2013, 49, 451–461.
39. Seo, K.C.; Lee, J.H.; Lee, S.Y. Impact of PNF-based walking exercise on a ramp on gait performance of stroke patients. J. Phys.
Ther. Sci. 2012, 24, 1243–1246. [CrossRef]
40. Stephenson, J.; Maitland, M.; Beckstead, J.; Anemaet, W.K. Locomotor training on a treadmill compared with pnf training in
adults with chronic stroke. Technol. Innov. 2014, 15, 325–332. [CrossRef]
41. Downs, S.; Marquez, J.; Chiarelli, P. The Berg Balance Scale has high intra- and inter-rater reliability but absolute reliability varies
across the scale: A systematic review. J. Physiother. 2013, 59, 93–99. [CrossRef]
42. Portnoy, S.; Reif, S.; Mendelboim, T.; Rand, D. Postural control of individuals with chronic stroke compared to healthy participants:
Timed-Up-and-Go, Functional Reach Test and center of pressure movement. Eur. J. Phys. Rehabil. Med. 2017, 53, 685–693.
[CrossRef] [PubMed]
43. In, T.S.; Jung, J.H.; Jung, K.S.; Cho, H.Y. Effect of Sit-to-Stand Training Combined with Taping on Spasticity, Strength, Gait Speed
and Quality of Life in Patients with Stroke: A Randomized Controlled Trial. Life 2021, 11, 511. [CrossRef] [PubMed]
44. Shinde, K.; Ganvir, S. Effectiveness of trunk proprioceptive neuromuscular facilitation techniques after stroke: A meta-analysis.
Natl. J. Med. Allied. Sci. 2014, 3, 29–34.
45. Wang, T.J.; Belza, B.; Elaine Thompson, F.; Whitney, J.D.; Bennett, K. Effects of aquatic exercise on flexibility, strength and aerobic
fitness in adults with osteoarthritis of the hip or knee. J. Adv. Nurs. 2007, 57, 141–152. [CrossRef]
46. Rode, G.; Tiliket, C.; Boisson, D. Predominance of postural imbalance in left hemiparetic patients. Scand J. Rehabil. Med. 1997, 29,
11–16.
47. Espi-Lopez, G.V.; Lopez-Martinez, S.; Ingles, M.; Serra-Ano, P.; Aguilar-Rodriguez, M. Effect of manual therapy versus proprio-
ceptive neuromuscular facilitation in dynamic balance, mobility and flexibility in field hockey players. A randomized controlled
trial. Phys. Ther. Sport. 2018, 32, 173–179. [CrossRef]
Life 2022, 12, 882 18 of 18

48. van Nes, I.J.; Nienhuis, B.; Latour, H.; Geurts, A.C. Posturographic assessment of sitting balance recovery in the subacute phase
of stroke. Gait Posture 2008, 28, 507–512.
49. Manaf, H.; Justine, M.; Omar, M. Functional balance and motor impairment correlations with gait parameters during timed up
and go test across three attentional loading conditions in stroke survivors. Stroke Res. Treat. 2014, 2014, 439304. [CrossRef]
50. Kim, Y.N.; Lee, D.K. Comparison between Aquatic and Ground Environments of Rhythmic Initiation for Postural Control. J. Phys.
Ther. Sci. 2012, 24, 1269–1271. [CrossRef]
51. Dettmann, M.A.; Linder, M.T.; Sepic, S.B. Relationships among walking performance, postural stability, and functional assessments
of the hemiplegic patient. Am. J. Phys. Med. 1987, 66, 77–90. [PubMed]
52. Nardone, A.; Godi, M.; Grasso, M.; Guglielmetti, S.; Schieppati, M. Stabilometry is a predictor of gait performance in chronic
hemiparetic stroke patients. Gait Posture 2009, 30, 5–10. [CrossRef] [PubMed]
53. Wang, R.Y. Effect of proprioceptive neuromuscular facilitation on the gait of patients with hemiplegia of long and short duration.
Phys. Ther. 1994, 74, 1108–1115. [CrossRef] [PubMed]
54. Haruyama, K.; Kawakami, M.; Otsuka, T. Effect of core stability training on trunk function, standing balance, and mobility in
stroke patients. Neurorehabilit. Neural. Repair. 2017, 31, 240–249. [CrossRef]
55. Geurts, A.C.; de Haart, M.; van Nes, I.J.; Duysens, J. A review of standing balance recovery from stroke. Gait Posture 2005, 22,
267–281. [CrossRef]
56. Mayo, N.E.; Wood-Dauphinee, S.; Ahmed, S.; Gordon, C.; Higgins, J.; McEwen, S.; Salbach, N. Disablement following stroke.
Disabil. Rehabil. 1999, 21, 258–268. [CrossRef]
57. Smith, M.T.; Baer, G.D. Achievement of simple mobility milestones after stroke. Arch. Phys. Med. Rehabil. 1999, 80, 442–447.
[CrossRef]
58. Platz, T. Evidence-Based Guidelines and Clinical Pathways in Stroke Rehabilitation-An International Perspective. Front. Neurol.
2019, 10, 200. [CrossRef]
59. Ghannadi, S.; Shariat, A.; Ansari, N.N.; Tavakol, Z.; Honarpishe, R.; Dommerholt, J.; Noormohammadpour, P.; Ingle, L. The Effect
of Dry Needling on Lower Limb Dysfunction in Poststroke Survivors. J. Stroke Cerebrovasc. Dis. Off. J. Natl. Stroke Assoc. 2020,
29, 104814. [CrossRef]