Foam Rolling, Elbow Proprioception, Strength, and Functional Motor Performance

Abstract

Context

Foam rolling has recently been used frequently to increase flexibility. However, its effects on proprioception, strength, and motor performance are not well known. In addition, very few researchers have examined the effects of foam rolling on the upper extremity.

Objective

To investigate the effects of foam rolling on elbow proprioception, strength, and functional motor performance in healthy individuals.

Design

Randomized controlled clinical trial.

Setting

Exercise laboratory of School of Physical Therapy and Rehabilitation, Dokuz Eylul University.

Patients or Other Participants

Sixty healthy participants (mean age = 22.83 ± 4.07 years).

Intervention(s)

We randomly assigned participants to 2 groups: the foam-rolling group (FRG; 4 weeks of foam rolling for the biceps brachii muscle) or control group (CG; no foam rolling).

Main Outcome Measure(s)

We evaluated proprioception (joint position sense [JPS] and force matching), biceps brachii muscle strength, and functional motor performance (modified pull-up test [MPUT], closed kinetic chain upper extremity stability test [CKCUEST], and push-up test) at the baseline and at the end of the fourth and eighth weeks.

Results

The JPS at 45° of elbow flexion, muscle strength, CKCUEST, and push-up test results improved after foam rolling, and improvement was maintained at follow up (P < .017). Although the changes between groups for the results of proprioception and CKCUEST were similar among the 3 time points (P > .05), muscle strength improved from baseline to the second evaluation and from baseline to follow up (P < .001) in the FRG compared with the CG group (P = .004). The FRG group demonstrated better push-up test results than the CG group at the 3 time points (P = .040, P = .001, P < .001). Other data did not change (P values > .05).

Conclusions

Foam rolling was effective in improving elbow JPS at small flexion angles, biceps brachii strength, and some aspects of upper extremity functional motor performance. These effects were maintained at 4 weeks after application.

Key Points

• Four weeks of foam rolling to the biceps brachii improved its muscle strength, elbow joint position sense measured at a relatively small joint angle, and some aspects of upper extremity functional motor performance.

• The positive effects of 4 weeks of foam rolling lasted for 1 month.

• Foam rolling to the biceps brachii is not recommended to improve joint position sense and elbow force measured at larger angles.

Proprioception, which is a prerequisite for optimal muscular control, coordination, and stabilization,^1–4 is vital for a full movement repertoire⁵ that includes activities of daily living (ADLs; such as walking, reaching, and lifting objects)^2,6 as well as challenging athletic skills (such as jumping, shooting, and throwing).^1,3,4,7 Muscle strength, which is essential for general health and physical fitness, also facilitates participation in ADLs, physical activity, leisure activities, and sport performance.^8,9 Strength, proprioception, and neuromuscular control are combined in functional motor performance, which involves movements whose reflections are evident in ADLs and sports.^1,10 After an injury, proprioception decreases due to deafferentation caused by damage to mechanoreceptors, which disrupts the neuromuscular response required for joint stabilization in the regulation of function and, thus, decreases stability.¹ This can ultimately lead to further microtrauma and reinjury.^1,10 In this context, strength and neuromuscular control are critical to improve performance and to prevent^1,7 injury and reinjury in the upper extremity.¹¹ The recovery of decreased strength and neuromuscular control after injury is also a prerequisite for returning to sport.^1,11

Proprioception and neuromuscular control of the shoulder and elbow are necessary for accurate positioning of the hand and proper function of the upper extremity during ADLs and sport activities.^2–4,12 Although a growing number of researchers are addressing proprioception and neuromuscular control of the shoulder region, focusing on the elbow in this context is surprisingly still in its infancy but is of great importance. The elbow joint contributes to stabilization, the fulfillment of fine manipulative tasks, and achieving optimal performance in sports.^4,7 During multijoint activity, coordination among joints compensates for errors, thereby improving performance.¹² Erroneous deviations of the proximal limb segments can be offset by the distal joints, and this compensatory behavior can be organized based on proprioception.^3,7,9,12 Still, more errors seem to occur in elbow proprioception than in shoulder proprioception during sport activities such as overhead throws.¹² Also, deterioration of the proprioceptive input at the elbow resulted in greater disturbance of endpoint positioning of the arm movements than deterioration of the proprioceptive input at the shoulder.⁶ Further, sport activities often lead to injuries that impair elbow proprioception.⁹ Developing effective interventions to improve elbow proprioception and neuromuscular control appears to be crucial.^1,10

Foam rolling is an affordable, easy, and time-efficient technique that has been commonly used as an exercise and massage tool in recent years.^13,14 Foam rolling increases flexibility and joint and fascial mobility.^15–18 However, reports from several research teams^15,17–19 were in part contradictory with respect to the effects of the technique on proprioception, strength or endurance, and performance. Authors investigating the effects of foam rolling on proprioception,¹⁵ strength or endurance,²⁰ or performance^17,20 described no effect or a decrease, whereas others found evidence to support improvements in proprioception,^15,19 strength or endurance,^16,18 or performance.¹⁸ Yet none of these studies focused on the upper extremity. The cylindrical structure of the material used in foam rolling, its texture, and the application of the technique with pressure from body weight may affect proprioception by stimulating the receptors.¹³ This may contribute to strength and functional stability or motor performance if the afferent pathways can be rearranged and the coactivation of force couples can be facilitated. Conversely, the application may not improve or may worsen strength due to its possible relaxation effect.^14,15

No investigators have examined the effects of foam rolling on functional motor performance of the elbow region, although the acute effects of the technique have generally been studied. High-quality and well-designed randomized controlled studies are warranted to assess the effects of the technique with a longer application period and appropriate follow up.¹⁴ Determining whether foam rolling is an effective intervention for improving elbow proprioception, strength, and upper extremity performance would be useful for guiding the health care team in implementing effective interventions to improve upper extremity performance and prevent musculoskeletal injuries.

Therefore, the purpose of our study was to evaluate the effects of foam rolling of the elbow at the end of the 4-week intervention and at 1-month follow up on proprioception, strength, and functional motor performance.

METHODS

Research Design

Between September 2017 and December 2018, 60 healthy university students participated in this randomized controlled study conducted at Dokuz Eylul University, School of Physical Therapy and Rehabilitation. The Ethics Committee of Dokuz Eylul University (Number:2017/17-42:3399-GOA) approved this study, and participants gave written informed consent. We randomly divided them into the foam-rolling group (FRG; n = 30) or control group (CG; n = 30; Figure 1) using a computerized random number generator (Random.org; Randomness and Integrity Services Ltd).

Figure 1.

Consolidated Standards of Reporting Trials (CONSORT) flow diagram of the study.

Participants

Volunteers were included if they were 18 years of age or older, did not have elbow pain in the 6 months before recruitment, and did not regularly perform upper extremity sports in the previous 6 months. Exclusion criteria were an open wound; acne or a similar skin problem that might prevent the application of the foam roller; experience with foam rolling or myofascial relaxation exercises; a history of or current upper extremity injury, upper extremity fracture, or surgery; or a systemic musculoskeletal disease, osteoporosis, diabetes mellitus with peripheral neuropathy, vertigo, cardiovascular disease, or pregnancy.

Given the lack of relevant studies, we assigned 30 individuals to each group to ensure parametric conditions instead of performing an a priori power analysis.

Procedures

Foam rolling and evaluations were performed on the dominant side. A rest period of 120 seconds was allowed between evaluations. Evaluations were done before administration, at the end of 4 weeks, and at the end of 8 weeks. Any learning effect was excluded by having participants perform a familiarization session for all tests.

Outcome Measures

Proprioception

Joint Position Sense (JPS).

The participant sat on the chair with the back supported and the hips flexed to approximately 90°. The forearm was placed on the arm of the chair, covered with a foam pad, in a semiprone position with the elbow fully extended at the starting position. The participant bent the elbow to the target angle (45°, 60°, and 75° of elbow flexion in random order [JPS_45°, JPS_60°, and JPS_75°, respectively]) as measured on a digital inclinometer (Baseline), and maintained this position for 5 seconds to learn this angle. Afterward, he or she returned to the initial position and rested for 5 seconds (Figure 2). The individual repeated this protocol 3 times with eyes open (EO) and eyes closed (EC). For the actual test, we applied the EC protocol 3 times.⁴ Our test-retest reliability in 10 patients was intraclass correlation coefficient (ICC) [3,k] = 0.78 for (JPS_45°), 0.94 (JPS_60°), and 0.78 (JPS_75°).

Figure 2.

Measurement of joint position sense.

Force Matching.

Force sense was assessed using force reproduction by limb matching that involved the use of a reference force, determined as a percentage of a maximal voluntary contraction (MVC), and attempting to replicate that force. The handheld dynamometer (HHD) is a small, portable, low-cost, and easy-to-use device that was valid and reliable in isolated strength testing of various muscles.²¹ Because of the limited studies of reliable force-reproduction measures in the elbow region, we used a force-matching test protocol that appeared to be compatible with the HHD.¹³ We measured force matching for elbow flexion. The participant sat in a chair with 90° of elbow and 90° of shoulder flexion, the forearm in supination, and the elbow resting on the table. First, we measured the MVC for the elbow flexors 3 times using an HHD (MicroFET3; Hoggan Scientific, LLC) and averaged the results. Next, we calculated 30% of the MVC and, after a 45-second rest, taught the participant the target contraction 3 times with EO and EC. Then we asked the individual to match the force with EC (Figure 3).²² Our test-retest reliability was ICC [3,k] = 0.83.

Figure 3.

Measurement of force matching.

We recorded the deviation (absolute error) from the target and averaged the 3 test results in all proprioception evaluations.

Muscle Strength

We measured biceps brachii muscle strength using the HHD. The participant lay in supine position with a neutral shoulder, elbow in 90° of flexion, and forearm supinated. We placed the HDD proximal to the styloid process and asked him or her to perform a gradually increasing MVC in 2 seconds and then maintain it for 5 seconds. The individual contracted the muscle as resistance was gradually applied via the handheld dynamometer until the examiner matched the participant's effort. We allowed a rest period of 1 minute between tests²³ and averaged the 3 test results. Our test-retest reliability was ICC [3,k] = 0.97.

Functional Motor Performance

Before the tests, the participant performed warmup from submaximal to maximal level. For a high-intensity effort, we orally encouraged each person throughout the evaluations.²⁴ A 1-minute rest was allowed between tests.

Closed Kinetic Chain Upper Extremity Stabilization Test.

For the closed kinetic chain upper extremity stabilization test (CKCUEST), we marked 2 lines (2 strips of athletic tape 3.8-cm wide) parallel to each other at a distance of 90 cm on the floor as measured with a standard cloth tape measure. The starting position for the test was 1 hand on each piece of tape while the body was in push-up position. The participant had to move both hands back and forth from each line as many times as possible in 15 seconds, men in the push-up position and women in the modified push-up position (kneeling; Figure 4). We allowed a 45-second rest period between repetitions and averaged the 3 test results. In addition to the average number of lines touched, the score and power were calculated using the following equation: score = average number of lines touched/height, power = ([68% weight · the average number of lines touched]/15).²⁵ Our test-retest reliability was ICC [3,k] = 0.91.

Figure 4.

Closed kinetic chain upper extremity stability test positioning: A, women. B, men.

Modified Pull-Up Test.

For the modified pull-up test (MPUT), we positioned the participant supine and adjusted the metal frame above his or her head just above the shoulder level. Men performed the pull-up test with support from their heels, and women were supported with a step just below the knees. To perform the test at full range, the participant started by holding the metal frame with the arms in full extension, pulling it parallel to the floor, and lowering the body with their elbows fully extended. During the test, we instructed the participant to limit the movement of the head and trunk and to maintain a smooth motion as much as possible (Figure 5). The individual performed the test with as many pulls as possible in 15 seconds during the 3 maximum tests, and we averaged the scores. A 45-second rest was permitted between repetitions.²⁴ Our test-retest reliability was ICC [3,k] = 0.93.

Figure 5.

Modified pull-up test positioning: A, women. B, men.

Push-Up Test.

Men performed the test in the standard position (on their hands and feet), and women in the modified position (on their hands and knees). While the participant was in the prone position with the trunk straight and hands open at shoulder width, he or she performed push-ups. Each person began the test with the elbows fully extended. As the body descended to the ground, the individual bent the elbows until the humerus was parallel to the surface. Throughout the test, we instructed the participant to keep the head and trunk position straight. The participant performed a 15-second trial, rested for 45 seconds, and then pursued 3 maximal tests. The number of push-ups the participant completed in the 15-second bout was recorded, and the 3 results were averaged.²⁴ Our test-retest reliability was ICC [3,k] = 0.75.

Foam Rolling Protocol

The FRG performed the protocol for 1 session/d, 3 d/wk, for 4 weeks with the physiotherapist. We instructed participants not to do any other exercise or receive any other treatments until the end of the study, and we monitored their compliance during the weekly sessions.

During the application, the participant sat on the floor with the side of the trunk close to a rectangular step. We placed standard 15.3-cm foam-rolling material on the step, and the individual positioned the biceps brachii muscle with body weight on the foam-rolling material, the shoulder in 90° of abduction, and the elbow in full extension (Figure 6). We chose the biceps brachii muscle because it was more distinct and easily distinguishable than nearby muscles. Each person performed 2 sets (60 seconds each) of foam rolling in the form of 10 back-and-forth movements in 1 minute. We used a metronome to control the speed and allowed a 30-second rest between sets. To ensure adequate pressure during the application, we provided oral encouragement for the participant to place as much body weight as possible on the foam roller, pushing into discomfort but not pain. Before the application, we conducted a trial session for teaching purposes.²⁶

Figure 6.

Foam-rolling application.

The CG Protocol

We handed out an information brochure to the CG about proprioception, strength, function, and foam rolling. We asked them not to do regular sports during the study and to continue ADLs as usual, and we monitored their compliance monthly.

Statistical Analysis

We used SPSS (version 22.0; IBM Corp) for all data analyses. The Shapiro-Wilk test was calculated to determine if the continuous data were normally distributed. The Mann-Whitney U test was computed to compare height, body weight, body mass index, and age and the χ² test to compare the sex and dominant extremity distribution of the 2 groups. For the 3 evaluations, we conducted a Friedman analysis of variance to examine the changes in proprioception, strength, and performance data. Then, to identify the time interval in which a meaningful difference occurred based on the level of significance of the data, Wilcoxon analysis and Bonferroni correction were performed (P < .0167). The Mann-Whitney U test was used to compare changes among these 3 time intervals between groups. We set the significance level as P < .05. In the post hoc power analyses, we accepted ≥0.80 as sufficient in showing a difference.

RESULTS

Participants' mean age was 22.83 ± 4.07 years, and their mean body mass index was 22.44 ± 3.39 kg/cm². All participants attended all sessions, and none experienced any side effects.

Baseline participant characteristics (Table 1) and main outcome measurements (Tables 2–4) were similar among groups (P > .05) except for age (P = .01). Supplemental Tables 1–3 (available online at http://dx.doi.org/10.4085/1062-6050-0445.20.S1) present the effect sizes and post hoc power results.

Table 1.

Participants' Characteristics

Characteristic	Group		P Value
	Foam Rolling (n = 30)	Control (n = 30)
	Median (Minimum–Maximum)
Age, y	21.70 (18.00–27.00)	23.97 (18.00–47.00)	.019ab
	Mean ± SD
Height, cm	171.93 ± 8.33	168.63 ± 7.85	.071a
Weight, kg	68.93 ± 13.46	62.43 ± 11.06	.080a
Body mass index, kg/m2	23.16 ± 3.67	21.72 ± 2.98	.171a
	No.
Sex (men/women)	18/12	11/19	.071c
Dominant extremity (right/left)	(30/0)	(29/1)	.313c

^a Mann-Whitney U test.

^b P < .05.

^c χ² test.

Table 2.

Proprioception Results Between Groups^a

Outcome Measure	Group	Measurement Timepoint			Change Between Timepoints			Change Over Time Between Groups, Pb	Binary Comparison of Measurement Times, Pc
		First	Second	Third	Second − First	Third − Second	Third − First
Joint position sense, °
45°	FRG	4.86 ± 3.24	3.04 ± 1.87	3.82 ± 2.01	−1.82 ± 3.72	0.77 ± 2.67	−1.04 ± 3.05	.010f	2 > 1 (.01)bg
									1 = 3 (.09)
									2 = 3 (.14)
	CG	4.93 ± 3.06	4.81 ± 2.76	4.59 ± 2.87	−0.12 ± 3.49	−0.21 ± 2.80	−0.33 ± 3.27	.723	NA
60°	FRG	4.07 ± 1.93	4.56 ± 2.61	3.59 ± 1.51	0.48 ± 3.17	−0.96 ± 2.82	−0.47 ± 2.40	.387	NA
	CG	4.86 ± 3.58	4.19 ± 2.74	4.43 ± 2.59	−0.66 ± 3.46	0.23 ± 3.49	−0.42 ± 4.25	.712	NA
75°d	FRG	3.00 (0.66–10.33)	3.49 (0.66–7.66)	3.00 (0.66–8.00)	0.00 (−7.00 to 5.33)	−0.83 (−4.67 to 4.33)	0.00 (−8.33 to 5.67)	.607	NA
	CG	3.66 (0.66–11.33)	3.00 (1.33–8.66)	2.49 (0.66–6.33)	−0.33 (−5.67 to 5.33)	−0.33 (−5.67 to 3.67)	−0.83 (−6.67 to 5.33)	.387	NA
Pe
45°					.069	.108	.332	NA	NA
60°					.290	.234	.482	NA	NA
75°					.544	.906	.450	NA	NA
Force match, kgd
Biceps brachii	FRG	0.30 (0.06–0.70)	0.03 (0.33–1.13)	0.30 (0.06–0.76)	0.000 (−0.50 to 0.67)	−0.11 (−0.70 to 0.57)	−0.01 (−0.57 to 0.57)	.655	NA
	CG	0.31 (0.06–0.96)	0.316 (0.10–0.76)	0.28 (0.10–0.83)	−0.03 (−0.60 to 0.60)	−0.08 (−0.43 to 0.43)	−0.01 (−0.50 to 0.30)	.519	NA
Pe					0.701	0.796	0.819	NA	NA

Abbreviations: CG, control group; FRG, foam rolling group; NA, not applicable.

^a First = baseline, second = after 4 wk, third = after 8 wk. Values are presented as mean ± SD unless otherwise indicated.

^b Friedman analysis of variance.

^c Wilcoxon signed-rank test.

^d Values are presented as median (minimum–maximum).

^eMann-Whitney U test.

^fP < .05.

^g P < .0167.

Table 4.

Functional Motor Performance Results Between Groups^a

Functional Motor Performanceb	Group	Measurement Timepoint			Change Between Timepoints			Change Over Time Between Groups, Pc	Binary Comparison of Measurement Times (Pd)
		First	Second	Third	Second − First	Third − Second	Third − First
MPUT, No. pull-ups	FRG	8.78 ± 2.36	9.18 ± 2.43	9.61 ± 2.66	0.42 ± 1.53	0.42 ± 1.48	0.82 ± 1.57	.077	NA
	CG	8.46 ± 2.05	8.31 ± 2.22	8.65 ± 2.69	−0.15 ± 1.99	0.34 ± 1.64	0.18 ± 2.09	.805	NA
Push-ups, No.e	FRG	12.00 (7.00–17.00)	13.16 (7.00–22.00)	13.50 (7.66–21.00)	1.66 (−3.33–10.00)	−1.00 (−3.00–4.00)	−0.666 (−2.67–8.00)	.005h	2 > 1 (.003)cg
									3 > 1 (.001)cg
									2 = 3 (.084)
	CG	11.00 (6.33–17.33)	12.16 (7.00–16.66)	10.99 (5.33–16.33)	0.16 (−7.67 to 4.67)	0.66 (−4.33 to 3.00)	1.83 (−5.33 to 2.33)	.057	NA
Pf
MPUT					.354	.652	.187	NA	NA
Push up					.040g	.001g	<.001h	NA	NA
Closed kinetic chain upper extremity stability testg
Average	FRG	19.52 ± 4.59	21.82 ± 5.06	21.77 ± 4.82	2.29 ± 3.08	−0.44 ± 2.35	2.25 ± 3.46	<.001h	2 > 1 (.001)cg
									3>1 (.001)cg
									2 = 3 (.72)
Score		0.28 ± 0.07	0.32 ± 0.08	0.32 ± 0.07	0.03 ± 0.04	−0.001 ± 0.035	0.03 ± 0.05	.001h	2 > 1 (.000)cg
									3 > 1 (.001)cg
									2 = 3 (.74)
Power		59.69 ± 12.83	66.57 ± 12.99	66.45 ± 12.43	6.88 ± 9.80	−0.12 ± 7.70	6.75 ± 11.64	.001h	2 > 1 (.001)cg
									3 > 1 (.002)cg
									2 = 3 (.79)
Average	CG	20.69 ± 3.25	21.65 ± 4.11	22.22 ± 4.59	0.95 ± 2.84	0.56 ± 2.00	1.52 ± 3.46	.550	NA
Score		0.31 ± 0.05	0.32 ± 0.06	0.33 ± 0.07	0.01 ± 0.04	0.009 ± 0.030	0.02 ± 0.05	.055	NA
Power		59.37 ± 15.17	61.82 ± 15.84	63.32 ± 16.66	2.45 ± 8.38	1.49 ± 5.59	3.95 ± 9.66	.055	NA
Ph
Average					.124	.464	.348	NA	NA
Score					.137	.424	.344	NA	NA
Power					.110	.684	.322	NA	NA

Abbreviations: CG, control group; CKCUEST,; FRG, foam rolling group; MPUT, modified pull-up test; NA, not applicable.

^a First = baseline, second = after 4 wk, third = after 8 wk. Values are presented as mean ± SD unless otherwise indicated.

^bAverage No. of lines touched, score (inch⁻¹) and power (kg).

^cFriedman analysis of variance.

^dWilcoxon signed-rank test.

^eValues are presented as median (minimum–maximum).

^fMann-Whitney U test.

^gP < .0167.

^hP < .05.

Proprioception

Joint Position Sense

In the FRG, JPS_45° was more accurate after 4 weeks of intervention than at the first assessment (P = .01). The values of the CG did not change (P = .72). No differences were present between groups after 4 (P_{JPS_45°} = 0.06) or 8 (P_{JPS_45°} = 0.33) weeks. Other JPS data did not change over time in either group (P_{JPS_60°} = 0.38 for FRG, P_{JPS_60°} = 0.71 for CG, P_{JPS_75°} = 0.60 for FRG, P_{JPS_75°} = 0.38 for CG; Table 2).

Force Matching

The force-matching data did not vary over time in either group (P = .65 for FRG, P = .51 for CG; Table 2).

Muscle Strength

In the second (P_2nd-1st = .002) and third (P_3rd-1st = .001) assessments, muscle strength increased in the FRG compared with the first assessment. The CG's values did not differ (P = .85). The changes between the first and second (P_2nd-1st = .004) assessments and between the first and third assessments (P_3th-1st = .001) within each group were different. However, the changes between the second and the third assessment results were not different between groups (P = .98; Table 3).

Table 3.

Muscle Strength Results Between Groups^a

Muscle Strength	Group	Measurement Timepoint			Change Between Timepoints			Change Over Time Between Groups, Pb	Binary Comparison of Measurement Times (Pc)
		First	Second	Third	Second − First	Third − Second	Third − First
Biceps brachii, kg	Foam rolling	19.95 ± 4.30	21.33 ± 5.42	21.21 ± 4.70	1.29 ± 2.15	−0.02 ± 2.17	1.26 ± 1.91	<.001f	2 > 1 (.002)be
									3 > 1 (.001)be
									2 = 3 (.98)
	Control	18.58 ± 4.47	18.27 ± 3.54	18.29 ± 3.97	−0.31 ± 1.99	0.02 ± 1.56	−0.27 ± 1.42	0.851	NA
Pd					.004e	.994	.001e	NA	NA

Abbreviation: NA, not applicable.

^a First =baseline, second = after 4 wk, third = after 8 wk.

^b Friedman analysis of variance.

^c Wilcoxon signed-rank test.

^d Mann-Whitney U test.

^e P < .0167.

^f P < .05.

Functional Motor Performance

Closed Kinetic Chain Upper Extremity Stabilization Test

In the FRG, after both 4 (P_{average_2nd-1st} = .001, P_{score_2nd-1st} < .001, P_{power_2nd-1st} = .001) and 8 (P_{average_3rd-1st} = .001, P_{score_3rd-1st} = .001, P_{power_3rd-1st} = .002) weeks of application, the CKCUEST results improved compared with the preintervention values (P_average < .001_, P_score = .001, P_power = .001). When we compared the changes in the groups over time, no difference between measurement times was evident (P_{average_2nd-1st} = .12_, P_{average_3rd-2nd} = .46, P_{average_3rd-1st} = .34; P_{score_2nd-1st} = .13_, P_{score_3rd-2nd} = .42, P_{score_3rd-1st} = .34; P_{power_2nd-1st} = .11_, P_{power_3rd-2nd} = .68, P_{power_3rd-1st} = .32). The CG's values did not change (P_average = .55_, P_score = .055, P_power = .055).

Modified Pull-Up Test

The MPUT data did not vary over time in either group (P = .07 for FRG, P = .80 for CG).

Push-Up Test

In the FRG, the push-up results improved after 4 (P_2nd-1st = .003) and 8 (P_3rd-1st = .001) weeks of intervention compared with preintervention, but the CG values did not change (P = .057). When we compared the changes in the groups over time, the FRG results improved compared with the CG results at all time intervals (P_2nd-1st = .04_, P_3rd-2nd = .001, P_3rd-1st < .001; Table 4).

DISCUSSION

The JPS at 45° of elbow flexion, muscle strength, and CKCUEST and push-up test results improved after 4 weeks of foam rolling, and these improvements were maintained for 1 month. Compared with the CG, muscle strength in the FRG improved from baseline to the second evaluation and from baseline to follow up. The FRG was superior to the CG in the improved push-up test results at the 3 time points.

Possible mechanical, neurophysiological, and psychological mechanisms for the beneficial effects of foam rolling include improved proprioceptive feedback,^13,15 muscle firing rate and fiber recruitment,^15,18 circulation,^13,14,27 mobility,¹⁷ autonomic nervous system (ANS) activation,²⁸ and perceptions of wellbeing.²⁹

Foam rolling improved JPS_45°. At the end of week 8, the results deteriorated slightly compared with the values at 4 weeks and improved compared with the baseline values, but these changes were not statistically significant. Therefore, we can say that the values obtained after application were maintained for 1 more month. However, this result should be interpreted with caution because of its similarity with the baseline value. Although JPS did not change in the CG and the improvement in JPS_45° in the FRG was greater than in the CG, the failure to show a difference between groups may have been attributable to our insufficient statistical power (post hoc power = 0.21–0.54). The improvements in JPS_45° may have been due to the following mechanisms. During rolling, pressure-related elongation in the muscle may activate the muscle spindle and Golgi tendon organs (GTOs) and the mechanoreceptors in the fascia, so that the muscle supplies more proprioceptive feedback to the central nervous system.^13,15 The relatively normal JPS values⁴ and the more accurate baseline values for JPS_60° and JPS_75° than JPS_45° could be the reasons for the lack of improvement in these JPS values.

Foam rolling did not change force matching. Force sense arises from the sense of tension peripherally (afferent feedback from the muscle) and the sense of effort centrally. The main factor in predicting target force appears to be the sense of effort.³⁰ Foam rolling does not seem to have a significant central effect. David et al¹⁵ also found improvement in knee JPS but not in force sense after foam rolling. They suggested that most of the time spent during foam rolling application is at the muscle belly, which stimulates the muscle spindles more, and less time is spent on the tendons, which stimulates the GTOs less. Therefore, they associated their findings with the fact that force-sense testing acts on the GTO, while the JPS test acts on the GTO and muscle spindles.¹⁵ This could also explain our results.

Biceps brachii strength increased after 4 weeks of foam rolling, and this improvement was maintained for 1 more month. Increased neural stimulation via foam rolling may enhance the firing rate and patterning of muscle fiber recruitment.^15,18 The technique might have caused elongation of the shortened sarcomeres with ischemic compression and a greater contribution to muscle contraction.³¹ Moreover, reactive hyperemia due to pressure may lead to increased oxygen uptake and decreased production of nociceptive and inflammatory substances, which may result in less damage to muscle fibers and greater power production.^13,32

The pressure applied during foam rolling may cause an increase in blood supply and tissue biochemical changes similar to those of other massage techniques. These changes include increased circulating neutrophil levels³³; much smaller increases in postexercise plasma creatine kinase³³; activation of sensors for the transcription of cytochrome c oxidase subunit VIIb (COX7B) and NADH dehydrogenase 1 (ND1), indicating the generation of new mitochondria³⁴; and fewer active immune cytokines, reflecting less cellular stress and inflammation.³⁴ With the formation of new mitochondrial cells, the muscle tissue can be better oxygenated and muscle strength may increase.¹⁸

Another possible mechanism is ANS activation with stimulation of the interstitial type III and IV receptors that respond to light touch and Ruffini terminations in the fascia that respond to deep continuous pressure. Stimulation of these receptors can reduce sympathetic tone, increase γ motor-neuron activity, and promote the relaxation of intrafascial smooth muscle cells.²⁸ The ANS can also change vasodilatation and fascial viscosity. Optimal relaxation of muscle and fascia might have had a positive effect on the muscle length-tension relationship.^13,14,35

Strength gains were maintained at 1 month after the application. Stimulation of the proprioceptors in the myofascia may have resulted in neural and myofascial adaptations. Fascia is thought to have a memory due to the mechanoreceptor and nociceptor structures it contains. Neural inputs to the brain and myofascial memory changes can occur from applications that use contact and pressure, such as foam rolling. In addition, collagen in the fascia is deposited along the direction of stretching at the molecular and macroscopic levels. Mechanical loads affect collagen alignment and deposition. Furthermore, the extracellular matrix can also be effective in this memory. Muscle appears to have memory due to central motor learning ability and deoxyribonucleic acid-containing nuclei within the muscle. These all point to a myofascial memory and, moreover, myofascial awareness.³⁶ In future studies, the duration of this effect should be investigated with a longer follow-up.

Inconsistent immediate outcomes were noted in examinations of the effects of foam rolling on performance.^18,29,37,38 We found that the CKCUEST and push-up test results improved after 4 weeks of foam rolling and that this improvement was maintained for 1 more month. As discussed earlier, neural and myofascial adaptations and myofascial memory or awareness may explain maintenance of the positive results.

Foam rolling can influence performance by producing a warmup effect that can be attributed to circulatory changes and fascial relaxation.¹⁴ Changes in tissue perfusion, increased plasma nitric oxide levels, decreased arterial stiffness, and improved vascular endothelial function²⁷ may lead to changes in the afferent muscle fibers and ANS activation. As performance mainly depends on muscle strength and neuromuscular control,¹⁴ improved strength and proprioception may account for better performance based on dynamic stabilization. Foam rolling did not improve or deteriorate the MPUT results. Through the increased dynamic neuromuscular stabilization with intense mechanoreceptor stimulation via foam rolling,^1,5 our performance tests were conducted in a weight-bearing position, and requiring more joint stability may have better reflected the effects of the application.²³ Lastly, enhanced performance may have been due to increased mobility^17,37 or a psychological environment conducive to better performance by reduced fatigue perception via stimulation of parasympathetic activity.^14,17,29

Limitations

The first limitation of this study was the lack of blinding of the examiner who obtained the measurements. Therefore, we took strict precautions to avoid bias. This examiner was not allowed to read the results on the measurement device throughout the tests. A trained assistant, blinded to the group assignments, read and recorded the results. To prevent bias in the functional performance tests, the examiner did not see the first evaluation results while recording the subsequent evaluation results. The other author, who was unaware of the group assignment of the participants, performed the data analyses. Another limitation could be the lack of standardization of the pressure during foam rolling.

Although we provided oral encouragement to the participants to achieve adequate pressure on the foam roller, we recommend ensuring standardization of the pressure with objective methods during future research.

CONCLUSIONS

Foam rolling to the biceps brachii was a safe method for improving its muscle strength, elbow JPS measured at 45°, and some aspects of upper extremity functional motor performance. After the 4-week training, improvements continued for 1 more month. Foam rolling had no effect on JPS or force matching measured at larger angles. In future studies, the long-term effects of the technique and related mechanisms and its effectiveness in the athletic population and on various elbow conditions should be investigated.