Acute effects of cryotherapy duration on upper extremity muscle strength, flexibility, endurance, and balance: a randomized trial

Abstract

Background

To compare the acute effects of different cryotherapy (CT) durations on shoulder muscle strength, flexibility, endurance, and upper extremity dynamic balance in healthy individuals.

Methods

Seventy-five healthy participants (51 females, 24 males; mean age 20.47 ± 1.32 years) were included in this prospective, single-blind, randomized trial. The participants were randomized into three groups (n = 25 per group) based on the duration of CT application: 10, 15, and 20 min. The primary outcome was muscle strength (hand-held dynamometer); secondary outcomes included upper body flexibility (Back Scratch Test), muscular endurance (Modified Push-up Test), and dynamic balance (Upper Quarter Y-Balance Test). All outcomes were assessed immediately pre- and post-CT application.

Results

No significant between-group differences were observed for any outcome following cryotherapy protocols, and effect sizes were generally small (partial η²=0.002–0.075, p > 0.05), indicating that the primary hypothesis was not supported. Exploratory within-group analyses demonstrated small reductions in external rotation strength after the 15-minute application (Δ=-1.02 kgF, d = 0.30) and in abduction strength after the 20-minute application (Δ=-0.99 kgF, d = 0.22), whereas no significant strength reductions were observed following the 10-minute application. Increases in endurance were observed across all groups (d = 0.28–0.43), and improvements in external rotation flexibility were noted following the 10- and 20-minute applications (d = 0.19–0.23). Dynamic balance was largely maintained, although an improvement in medial reach distance was observed in the 10-minute group (Δ = 3.45%, d = 0.46).

Conclusions

The 10-, 15-, and 20-minute cryotherapy protocols were associated with comparable acute responses in upper-extremity performance, with no between-group differences observed. Exploratory within-group analyses indicated favorable flexibility and endurance responses across durations, while small reductions in specific strength measures occurred only after the longer applications. Therefore, these findings remain exploratory, and no protocol demonstrated clear superiority.

Trial registration

ClinicalTrials.gov NCT05355584 (20/04/2022).

Introduction

Cryotherapy (CT) is the application of various cold sources to the body to induce physiological changes in human tissue [1]. CT encompasses various modalities, including simple ice packs, cold-water immersion, and whole-body cryotherapy [2]. These applications primarily aim to reduce tissue damage, muscle spasm, edema, and pain, and to facilitate a faster rehabilitation process after injury by decreasing intramuscular temperature and cellular metabolism [3–5]. As tissue temperature decreases, the firing rate of muscle spindles declines, leading to indirect physiological alterations within the muscle tissue [6]. Due to these alterations, motor performance—particularly muscle strength—is affected to varying degrees [7, 8]. Despite these variable effects on performance, CT remains widely applied because it effectively attenuates inflammatory responses, mitigates physiological stress from intense training, and accelerates overall recovery [9, 10].

Although recent evidence confirms that CT effectively promotes neuromuscular recovery [11], its acute impact on functional performance remains controversial, largely depending on the chosen modality and the duration of application [12–15]. Some studies report acute enhancements in muscle strength and endurance [16–18]. Conversely, others document immediate declines in muscle strength and athletic performance [19]. Similar contradictions exist for other parameters. For instance, CT may improve flexibility [20, 21] but shows no significant effect on balance [3, 22]. Many of these studies utilized protocols lasting 20 min or longer. However, returning to activity immediately after such prolonged exposure is known to induce significant performance deficits in athletes [19, 23].

Currently, there is no consensus in the literature on the practical CT duration, particularly for the complex biomechanics of the upper extremity. While cold water immersion for 11–15 min is reported to be more effective than passive recovery, applying an ice pack for more than 10 min has been shown to impair active muscle strength and activity [24]. Furthermore, prolonged durations (> 30 min) may pose a risk of nerve damage [25]. However, it remains unclear whether these general duration thresholds apply to the shoulder girdle. Among the limited studies in this area, Torres et al. investigated the acute effects of localized cryotherapy on shoulder strength and proprioception [26], while Ünver et al. explored shoulder joint functional balance [27]. While these studies provide insights into specific parameters from single, fixed-duration applications, the dose-response relationship across different exposure times within a comprehensive functional battery remains unclear.

To address this gap, this study aimed to compare the acute effects of 10-, 15-, and 20-minute cryotherapy durations on shoulder girdle muscle strength, flexibility, endurance, and dynamic balance in healthy young adults. We hypothesized that these acute functional responses would differ significantly across the three application durations. Unlike previous research, our study distinguishes itself by simultaneously comparing three distinct exposure times across a comprehensive set of performance metrics within a single protocol. By integrating these parameters, this study aims to provide clinicians with practical insights into selecting CT durations that support recovery while minimizing acute functional deficits.

Methods

Study design

This study was designed as a prospective, single-blind, randomized trial. The research was conducted at the Laboratories of the Department of Physiotherapy and Rehabilitation, Faculty of Health Sciences, Muğla Sıtkı Koçman University between April 2022 and May 2022. The study protocol was approved by the Muğla Sıtkı Koçman University Ethics Committee (Date: 15.12.2021, Decision No: 28) and was conducted in accordance with the principles of the Declaration of Helsinki. The study was registered at ClinicalTrials.gov (NCT05355584). Written informed consent was obtained from all participants prior to their inclusion in the study. The reporting of this study follows the CONSORT guidelines.

Participants

Volunteers studying at Muğla Sıtkı Koçman University Faculty of Health Sciences were recruited for this study following a faculty-wide announcement. A total of 105 volunteers were initially screened. Inclusion criteria comprised being a healthy volunteer aged 18–25 years. Exclusion criteria included having a history of shoulder injury or upper extremity surgery within the last two years, suffering from cervical or thoracic spine disorders or current upper extremity musculoskeletal pain, being a professional athlete, or having conditions affecting balance, proprioception, or vascularity (e.g., diabetes mellitus, atherosclerosis, Raynaud’s disease, cold intolerance). Following the eligibility assessment, 75 participants were enrolled in the study. To ensure the sample reflected the faculty’s gender distribution (33% male, 67% female), stratified sampling was used.

An a priori sample size calculation was conducted during the study planning phase using G*Power (F tests: ANCOVA, fixed effects, main effects, and interactions). With an assumed effect size of f = 0.40 (based on the muscle strength outcome [26]), α error probability of 0.05, power of 80%, three intervention groups, one covariate, and a numerator df of 2, the required total sample size was estimated to be 64 participants. To ensure an equal number of people in each group, the group size was set at 66, and to account for potential dropouts, the target sample size was set at 75.

Randomization and blinding

Following the baseline assessments, participants were randomly assigned to one of three experimental groups (10, 15, and 20 min) with a 1:1:1 allocation ratio. To ensure complete allocation concealment, an independent researcher generated the allocation sequence using a computer-based random number generator, stratified by sex to ensure an identical distribution of males and females across all three groups. A secondary researcher was responsible for participant enrollment, while the primary investigator assigned participants to their respective groups and administered the interventions.

Participants were blinded to the specific treatment duration and the hypothesis of the study. To maintain blinding, no clocks or timekeeping devices were present in the laboratory, and the start/end times were managed solely by the primary researcher. To ensure procedural consistency, all interventions and outcome assessments were conducted by a single licensed physiotherapist with extensive prior training in standardized testing protocols. Consequently, assessor blinding could not be implemented. To mitigate potential observer bias, all measurement protocols, equipment positioning, and verbal instructions were strictly standardized across all participants.

Procedures

Prior to data collection, participants’ demographic characteristics were recorded, followed by a sensory screening on the dominant shoulder to rule out sensory deficits using hot-cold discrimination tubes. Additionally, a cold sensitivity test was applied to the volar surface of the forearm for 2–3 min; participants who exhibited adverse reactions (e.g., pain, stiffness, dysesthesia, or skin discoloration) were excluded. All outcome measures were performed in the same standardized order immediately pre- and post-CT: muscle strength, flexibility, endurance, and balance. To minimize fatigue, a 1–2 min rest interval was provided between tests. Finally, following the completion of the study procedures, participants were monitored for 15 min, and their vital signs were assessed.

Intervention

All experimental procedures and cryotherapy interventions were conducted in a temperature-controlled laboratory (approximately 22–24 °C). The cryotherapy intervention was administered while the participant was seated in a chair with back support to ensure a relaxed posture. A standard gel cold pack (14 × 33 cm; Roll Brand) was used as the cryotherapy agent. Prior to application, to ensure a standardized initial cooling temperature, the cold packs were stored in a standard freezer at approximately − 18 °C for at least 24 h, and a fresh pack was used for each participant. The cold pack was centered over the acromion of the dominant shoulder (defined as the hand used for writing), covering the shoulder girdle muscles. To minimize the risk of skin burns (frostbite), a thin, damp cloth was placed between the cold pack and the skin, and the pack was secured with its integrated strap (Fig. 1).

Fig. 1.

Cryotherapy application to the shoulder girdle region

Participants were instructed to remain relaxed and minimize muscle contraction during the application to avoid influencing tissue temperature dynamics [28]. The researcher performed the cryotherapy application (10, 15, or 20 min) in accordance with each participant’s group allocation. Immediately after the intervention, the initial assessments were repeated. Throughout the application, the area was inspected intermittently for adverse skin reactions, and participants were queried regarding pain or discomfort. In the event of severe discomfort, the application was terminated immediately.

Outcome measures

Muscle strength assessment

Isometric muscle strength of the dominant shoulder girdle was assessed as the primary outcome using a hand-held dynamometer (Model-01165, Lafayette, USA). This assessment method has demonstrated excellent intrarater reliability for the shoulder musculature (ICC > 0.85) [29], as well as strong concurrent validity when compared with gold-standard isokinetic dynamometry [30]. Measurements (kg) were obtained in a specific order: flexion, abduction, adduction, internal rotation (IR), external rotation (ER), and extension, respectively. Three trials were conducted for each muscle group, and the average was calculated. A 1-minute rest period was provided between assessments to minimize fatigue.

Testing positions followed standardized protocols [31]. Flexion and abduction were assessed in a seated position with the shoulder at 90° and elbow extended (Fig. 2A, B). Extension was measured in the prone position with pelvic stabilization at the posterior superior iliac spine (Fig. 2C). Adduction, IR, and ER were assessed in the supine position with pelvic stabilization at the anterior superior iliac spine. For rotators, the shoulder was abducted to 90° with 90° elbow flexion (Fig. 2D, E). For adduction, the dynamometer was positioned between the humerus and the trunk (Fig. 2F). In all tests, resistance was applied distally to the lever arm, and participants were instructed to avoid compensatory movements.

Fig. 2.

Assessment of shoulder muscle strength using a hand-held dynamometer. (A: Flexion, B: Abduction, C: Extension, D: Internal rotation, E: External rotation, F: Adduction)

Flexibility assessment

Shoulder functional flexibility was assessed in a standing position using the Back Scratch Test. This test has been established as a highly reliable and valid measure for evaluating composite upper-extremity range of motion (with reliability and validity coefficients ranging from 0.79 to 0.97) [32]. The dominant shoulder was evaluated in two distinct functional patterns: reaching over the shoulder (assessing external rotation) and reaching behind the back (assessing internal rotation). In both positions, the distance (cm) between the middle finger tips of the dominant and non-dominant hands was measured. A gap between fingers was recorded as a negative value (-), touching tips as zero (0), and an overlap as a positive value (+). The mean of three trials was used for analysis.

Muscle endurance assessment

Upper extremity muscular endurance was assessed using the Modified Push-up Test. This test is a widely recognized clinical measure for evaluating upper-body muscular endurance, demonstrating excellent reliability (reliability coefficients > 0.90) and validity [33]. Following a familiarization trial, participants adopted a prone position on the mat with both knees and elbows in flexion (chest close to the floor). From this starting position, participants were instructed to push their torso upward by fully extending their elbows, without altering the flexed knee position. The maximum number of correctly performed repetitions within 30 s was recorded [33].

A single trial was performed to minimize cumulative upper extremity fatigue. Repetitions were considered invalid and excluded if compensations occurred, such as altered knee position, hip hiking, excessive lumbar lordosis (swayback), excessive thoracic kyphosis, head retraction, or failure to achieve full elbow extension/flexion range of motion.

Upper extremity dynamic balance assessment

Upper extremity dynamic balance was assessed using the Upper Quarter Y-Balance Test (YBT-UQ). This test has demonstrated excellent reliability (test-retest ICCs ranging from 0.80 to 0.99) and validity as a functional measure of upper-extremity performance [34, 35]. The testing setup consisted of three tape measures fixed to the floor in medial, superolateral, and inferolateral directions, with angles of 135°, 135°, and 90° between them. To standardize foot stability, the test was performed without shoes. Participants adopted a push-up position with the dominant hand placed at the central junction, the shoulder in 90° flexion, the elbow extended, and the feet shoulder-width apart. Participants were instructed to reach as far as possible with the free hand in each direction and return to the starting position in a controlled manner (Fig. 3). Prior to the recorded assessments, participants were provided with three practice trials in each reach direction to familiarize themselves with the testing procedure. Three successful trials were recorded for each direction. A 1-minute rest period was provided between directions.

Fig. 3.

Upper Quarter Y-Balance Test assessment positions. (A: Medial, B: Superolateral, C: Inferolateral)

The absolute reach distance was determined by averaging three successful trials for each direction. Upper limb length was measured in a standing position (shoulder at 90° abduction, elbow extended, wrist neutral, fingers pointing forward) as the distance from the spinous process of the 7th cervical vertebra (C7) to the distal tip of the middle finger. To standardize results, the absolute reach distance was normalized to upper-limb length, yielding a relative score calculated as: Relative (normalized) Reach Distance (%) = (Absolute reach distance/limb length) × 100.

Statistical analysis

Data analysis was performed using IBM SPSS Statistics 25.0 (IBM Corporation, Armonk, NY, USA). The conformity of variables to normal distribution was examined using visual methods (histograms) and the Kolmogorov-Smirnov Test. Descriptive analyses were presented as mean ± standard deviation (SD) for normally distributed variables, median (minimum-maximum) for non-normally distributed variables, and frequency (percentage) for categorical variables.

Baseline characteristics and outcomes were compared between groups using One-way ANOVA. The primary analyses were conducted using a repeated-measures analysis of variance (RM-ANOVA), with time (pre- and post-intervention) as the within-subject factor and cryotherapy duration (10, 15, and 20 min) as the between-subject factor. The primary hypothesis was evaluated based on the time × group interaction. When significant main or interaction effects were identified, post hoc pairwise comparisons were performed using the Bonferroni correction to control the family-wise type I error rate. Effect sizes for the primary analyses were expressed as partial eta-squared (partial η²). To describe pre- to post-intervention changes within each cryotherapy duration, exploratory within-group analyses were subsequently performed using paired-samples t-test. These within-group analyses were considered exploratory and hypothesis-generating and were interpreted cautiously; they were not used to infer superiority among cryotherapy durations. Cohen’s d was calculated to quantify the magnitude of within-group changes, and 95% confidence intervals (CIs) were reported where appropriate. Statistical significance was set at a two-sided p < 0.05.

Results

A total of 105 participants were initially recruited for the study. Following the exclusion of 30 individuals (6 did not meet the inclusion criteria, and 24 declined to participate), data from 75 participants (51 females and 24 males) were included in the final analysis (Fig. 4). The mean age of the participants was 20.47 ± 1.32 years (range: 18–25 years). Demographic characteristics are presented in Table 1. At baseline, there were no statistically significant differences between the groups regarding regular exercise habits, smoking status, muscle strength, flexibility, endurance, or balance (p > 0.05). No adverse effects were observed in any group.

Fig. 4.

CONSORT flow diagram of the study

Table 1.

Baseline characteristics of the participants

	10-minute (n = 25)	15-minute (n = 25)	20-minute (n = 25)
Age (years, median (min-max)) Gender (F/M) Height (cm, mean ± SD) Body weight (kg, median (min-max)) BMI (kg/m2, median (min-max)) Dominant Arm (R/L)	20 (19–22) 17/8 169.64 ± 9.94 58 (50–95) 21.48 (17.72–29.32) 22/3	20 (18–24) 17/8 167.20 ± 10.46 58 (40–92) 21.34 (17.71–29.70) 23/2	21 (18–24) 17/8 168.28 ± 9.74 60 (49–82) 21.26 (16.76–32.46) 25/0

BMI body mass index, F female, M male, R right, L left, SD standard deviation, n number of participants, p > 0.05

Primary outcome

Primary between-group analyses were performed using a repeated-measures ANOVA. No significant time × group interactions were observed for any shoulder muscle strength outcome, including flexion, extension, abduction, adduction, internal rotation, and external rotation (all p > 0.05), indicating that the acute changes in muscle strength were comparable across the 10-, 15-, and 20-minute cryotherapy protocols, and the primary hypothesis was not supported. The corresponding interaction effect sizes were small (partial η² = 0.002–0.038), and no cryotherapy duration demonstrated superiority over the others. Significant main effects of time were observed for shoulder abduction (p = 0.025) and external rotation strength (p = 0.024), indicating similar changes over time across all three groups (Table 2).

Table 2.

Primary inferential analyses (between-group) of muscle strength, flexibility, endurance, and balance

Outcome	Time p	Time x Group p	partial η²
Muscle Strength (Kg-F)
Flexion	0.211	0.923	0.002
Extension	0.410	0.867	0.071
Abduction	0.025	0.634	0.013
Adduction	0.692	0.350	0.029
İnternal rotation	0.516	0.675	0.011
External rotation	0.024	0.247	0.038
Flexibility (cm)
External rotation	0.000	0.060	0.075
Internal rotation	0.006	0.525	0.018
Endurance (repetitions)	0.000	0.751	0.008
Balance (normalized scor)
Medial (%)	0.236	0.076	0.069
Supero-lateral (%)	0.300	0.548	0.017
Infero-lateral (%)	0.786	0.498	0.019

Primary hypothesis testing was based on the Time × Group interaction, KgF=kilogram-force; cm = centimeter. Bold values indicate statistical significance (p < 0.05)

Exploratory within-group analyses demonstrated no significant changes in muscle strength following the 10-minute cryotherapy protocol. In the 15-minute group, external rotation strength decreased slightly from pre- to post-intervention (Δ = −1.02 kgF, 95% CI: −1.54 to − 0.50; p = 0.001; Cohen’s d = 0.30). Similarly, the 20-minute protocol was associated with a small reduction in shoulder abduction strength (Δ = −0.99 kgF, 95% CI: −1.93 to − 0.18; p = 0.019; Cohen’s d = 0.22). No other within-group changes reached statistical significance. As these analyses were exploratory, they should not be interpreted as evidence of the superiority of any cryotherapy duration (Table 3).

Table 3.

Exploratory within-group analyses

Outcome	Group	Pre Mean ± SD	Post Mean ± SD	Δ [95% CI]	p (within group)	Cohen’s d
Muscle Strength (Kg-F)
Flexion	10-minute	18.60 ± 4.38	18.08 ± 4.25	-0.52 [-1.73–0.69]	0.386	0.12
	15-minute	18.74 ± 5.33	18.24 ± 4.92	-0.50 [-1.86–0.84]	0.446	0.10
	20-minute	17.66 ± 4.51	17.43 ± 4.66	-0.23 [-1.19–0.73]	0.627	0.05
Extension	10-minute	17.76 ± 4.66	18.13 ± 4.68	0.37 [-0.53–1.55]	0.527	0.08
	15-minute	17.56 ± 4.98	17.58 ± 4.37	0.02 [-1.17–1.21]	0.969	0.00
	20-minute	16.87 ± 4.12	17.20 ± 3.91	0.33 [-0.32–0.98]	0.301	0.08
Abduction	10-minute	17.05 ± 4.46	16.49 ± 4.01	-0.56 [-1.59–0.47]	0.274	0.13
	15-minute	16.84 ± 4.72	16.49 ± 4.19	-0.35 [-1.46–0.75]	0.514	0.08
	20-minute	16.48 ± 4.67	15.49 ± 4.30	-0.99 [-1.93– (-0.18)]	0.019	0.22
Adduction	10-minute	21.90 ± 5.63	21.63 ± 5.58	-0.27 [-1.45–1.50]	0.754	0.05
	15-minute	20.41 ± 5.03	21.39 ± 4.96	0.98 [-0.20–2.16]	0.102	0.20
	20-minute	21.21 ± 4.63	20.98 ± 4.69	-0.23 [-1.52–0.99]	0.698	0.05
İnternal rotation	10-minute	16.30 ± 3.77	16.12 ± 4.01	-0.18 [-1.21–0.85]	0.721	0.05
	15-minute	16.04 ± 3.52	16.16 ± 3.88	0.12 [-0.73–0.97]	0.772	0.03
	20-minute	16.59 ± 4.71	16.10 ± 4.19	-0.49 [-1.60–0.62]	0.350	0.11
External rotation	10-minute	14.76 ± 3.52	14.68 ± 3.55	-0.08 [-1.06–0.90]	0.371	0.12
	15-minute	15.30 ± 3.94	14.28 ± 3.87	-1.02 [-1.54–(-0.50)]	0.001	0.30
	20-minute	14.83 ± 4.65	14.35 ± 4.49	-0.48 [-1.35–0.39]	0.266	0.24
Flexibility (cm)
External rotation	10-minute	4.25 ± 4.94	5.17 ± 4.94	0.92 [0.06–1.78]	0.036	0.19
	15-minute	7.19 ± 3.52	7.34 ± 3.79	0.15 [-0.40–0.70]	0.586	0.04
	20-minute	4.39 ± 5.44	5.71 ± 6.15	1.32 [0.61–2.03]	0.001	0.23
Internal rotation	10-minute	1.46 ± 5.81	2.17 ± 6.14	0.71 [-0.04–1.45]	0.062	0.12
	15-minute	3.55 ± 5.74	3.85 ± 5.68	0.30 [-0.59–1.20]	0.486	0.05
	20-minute	-0.17 ± 6.98	0.75 ± 7.07	0.92 [0.14–1.73]	0.023	0.13
Endurance (repetitions)
	10-minute	19.68 ± 7.45	21.76 ± 7.57	2.08 [1.47–2.69]	0.000	0.28
	15-minute	16.56 ± 5.51	19.08 ± 6.17	2.52 [1.59–3.45]	0.000	0.43
	20-minute	18.68 ± 7.60	21.04 ± 7.73	2.36 [1.38–3.34]	0.000	0.31
Y-Balance Test (normalized scor)
Medial (%)	10-minute	103.06 ± 6.61	106.51 ± 8.41	3.45 [0.58–5.07]	0.016	0.46
	15-minute	106.67 ± 8.22	106.40 ± 8.66	-0.27 [-2.22–1.72]	0.792	0.03
	20-minute	101.69 ± 8.92	101.46 ± 11.61	-0.23 [-2.85–2.25]	0.810	0.02
Supero-lateral (%)	10-minute	55.67 ± 10.31	57.24 ± 12.11	1.57 [-0.59–3.15]	0.170	0.12
	15-minute	54.22 ± 10.49	55.17 ± 12.22	0.95 [-1.05–2.51]	0.405	0.07
	20-minute	54.53 ± 11.80	54.37 ± 12.36	0.16 [-2.63–2.16]	0.842	0.02
Infero-lateral (%)	10-minute	70.19 ± 10.10	70.29 ± 10.93	0.10 [-2.82–2.89]	0.955	0.00
	15-minute	70.68 ± 12.82	71.60 ± 14.12	0.92 [-1.28–2.64]	0.483	0.06
	20-minute	71.84 ± 11.94	70.31 ± 14.42	-1.53 [-3.68–1.14]	0.289	0.10

Within-group analyses were exploratory and descriptive in nature. These analyses were not used to infer superiority among cryotherapy durations. Inferential comparisons between durations were based on the Time × Group interaction presented in Table 2. SD standard deviation, CT cryotherapy, CI confidence interval, Δ mean difference; repetitions number of repetitions completed in 30 s. Bold values indicate statistical significance (p < 0.05)

Secondary outcomes

No significant time × group interactions were observed for shoulder flexibility, muscle endurance, or Upper Quarter Y-Balance Test performance (all p > 0.05), with generally small interaction effect sizes (partial η² = 0.008–0.075). Although significant main effects of time were identified for external rotation flexibility (p < 0.001), internal rotation flexibility (p = 0.006), and muscular endurance (p < 0.001), these changes occurred similarly across the three intervention groups (Table 2).

Exploratory within-group analyses showed modest improvements in flexibility and endurance across the cryotherapy protocols. In the 10-minute group, external rotation flexibility (Δ = 0.92 cm, 95% CI: 0.06–1.78; p = 0.036; d = 0.19), muscular endurance (Δ = 2.08 repetitions, 95% CI: 1.47–2.69; p < 0.001; d = 0.28), and medial reach distance during the Upper Quarter Y-Balance Test (Δ = 3.45%, 95% CI: 0.58–5.07; p = 0.016; d = 0.46) increased. In the 15-minute group, muscular endurance increased (Δ = 2.52 repetitions, 95% CI: 1.59–3.45; p < 0.001; d = 0.43), whereas no significant changes were observed in flexibility or dynamic balance. In the 20-minute group, external rotation flexibility (Δ = 1.32 cm, 95% CI: 0.61–2.03; p = 0.001; d = 0.23), internal rotation flexibility (Δ = 0.92 cm, 95% CI: 0.14–1.73; p = 0.023; d = 0.13), and muscular endurance (Δ = 2.36 repetitions, 95% CI: 1.38–3.34; p < 0.001; d = 0.31) improved, whereas no significant within-group changes were observed for dynamic balance (Table 3).

Discussion

The primary finding of this randomized trial was that no significant between-group differences were observed for any outcome. Furthermore, the associated effect sizes were generally small (partial η² = 0.002–0.075), indicating broadly comparable responses across the 10-, 15-, and 20-minute cryotherapy protocols. Accordingly, our primary hypothesis of duration-dependent superiority was not supported, as no cryotherapy duration demonstrated superiority over the others. However, exploratory within-group analyses suggested distinct response patterns. While all durations were associated with improvements in selected flexibility, endurance, and balance measures following cryotherapy protocols, within-group data analysis indicated that longer exposures (15 and 20 min) were associated with significant decreases in specific muscle strengths. Importantly, given the lack of significant between-group differences, these within-group trends remain strictly exploratory, and no single duration can be definitively deemed superior for strength preservation.

This study presents several key strengths. To the best of our knowledge, this is the first randomized trial to compare the acute effects of different cryotherapy (CT) durations on shoulder girdle muscle strength, flexibility, endurance, and upper extremity balance in healthy individuals. Literature on specific CT durations is scarce; previous studies have predominantly focused on the lower extremity, while research involving the upper extremity has been restricted to a narrow range of parameters [24, 26]. Furthermore, the underlying mechanisms of CT remain not fully understood, with contradictory reports on its efficacy [15]. Therefore, this study addresses important gaps in the literature by directly comparing three CT durations across multiple objective outcomes—including muscle strength, flexibility, endurance, and dynamic balance. This comprehensive evaluation provides a valuable baseline reference for designing future controlled trials.

Our findings provide insight into the duration-dependent responses to CT. Existing evidence indicates that CT can aid recovery without reducing performance [36], limit post-exercise increases in rotator cuff cross-sectional area [37], and provide phase-specific benefits: pre- and per-cooling may enhance active performance, whereas post-cooling mainly supports recovery [38]. Consistent with these broader trends, although no duration demonstrated superiority, our exploratory within-group analyses showed favorable patterns in endurance, flexibility, and balance across all three cryotherapy durations. However, longer cold exposure was associated with significant within-group reductions in specific maximal isometric strength measures. Specifically, significant reductions in external rotation strength were observed after the 15-minute application (d = 0.30), whereas similar decreases in abduction strength were noted after the 20-minute application (d = 0.22). In contrast, no significant strength reductions were detected following the 10-minute application.

These findings are consistent with previous studies reporting transient decreases in muscle strength and proprioceptive performance following prolonged shoulder cooling [26, 39]. Specifically, previous research demonstrates that a 15-minute application can be associated with reductions in shoulder strength [26], and a 20-minute protocol has been reported to induce transient neuromuscular changes in the deltoid and rotator cuff, prompting caution when immediate maximal force is required [39]. These duration-dependent strength losses are commonly attributed to deeper tissue cooling, which may increase intramuscular viscosity and slow nerve conduction velocity. Indeed, each 1 °C drop in muscle temperature is associated with an approximately 2% decrease in maximum isometric force [7]. Although some studies suggest that specific cryotherapy protocols may improve muscle strength, these effects appear to be limited to intermittent, very short exposures, such as 3-minute applications [18].

In the context of the present study, while the lack of within-group reduction following the 10-minute protocol might hypothetically suggest a smaller degree of tissue cooling and neuromuscular alteration compared with longer applications, this observation must be interpreted with strict caution, given the absence of significant between-group differences. Furthermore, tissue cooling may be inherently influenced by individual factors such as subcutaneous fat thickness. Because direct skin and intramuscular temperatures were not measured in this study, the exact thermal dose cannot be determined from application time alone; therefore, these physiological interpretations should be viewed as strictly exploratory.

Similarly, the effects of CT protocols on flexibility remain equivocal. While some studies report positive effects of CT on flexibility [40, 41], others note increased stiffness and reduced flexibility following a 20-minute application [42], or minimal impact on muscle tone following a 10-minute application [43]. However, because these studies primarily examined the lower extremity, our upper-extremity findings provide an important extension of the evidence base. In the current study, while no significant between-group differences were observed for flexibility measures (partial η² = 0.018–0.075), exploratory within-group analyses indicated acute increases in shoulder flexibility across all groups. Specifically, these within-group increases were significant in external rotation flexibility after the 10-minute (d = 0.19) and 20-minute (d = 0.23) applications, and in internal rotation flexibility after the 20-minute application (d = 0.13). It is important to note that the absence of significant between-group differences may be partially related to the measurement tool; while practical, the Back Scratch Test is a composite measure that lacks joint-specific goniometric precision, which may limit its sensitivity to detect subtle, duration-dependent changes in glenohumeral range of motion. These relatively small effect sizes suggest modest acute changes that may reflect transient neuromuscular adaptations and local hypoalgesia, reducing stretch perception and increasing pressure pain threshold, thereby facilitating range of motion [21]. However, without a non-cooling control group, these flexibility gains cannot be definitively attributed to these cryotherapy protocols, and alternative explanations, such as warm-up or familiarization effects from the preceding maximal strength assessments, may also have contributed.

Regarding muscle endurance, previous lower-extremity studies utilizing application durations from 2.5 to 30 min generally report positive outcomes [44, 45]. Because CT responses are highly muscle-specific [17], evaluating the shoulder complex specifically provides valuable clinical insights. In the present study, although no significant between-group differences were detected, exploratory within-group analyses indicated acute increases in shoulder endurance across all three groups, with small-to-moderate effect sizes (d = 0.28–0.43). Consequently, as a hypothesis-generating observation, 10-, 15-, and 20-minute CT protocols may be associated with acute support of upper-extremity endurance. This aligns with existing literature recommending specific CT protocols for restoring muscle endurance and function during rehabilitation [46]. Nevertheless, because all participants received active CT and no non-cooling control condition was included, it cannot be determined whether these improvements were attributable solely to the CT protocols. Repeated testing, familiarization, motivational factors, or natural variability may also have contributed to the observed changes. Furthermore, using a single trial for the Modified Push-up Test—a deliberate choice to avoid fatigue before the balance assessments—may have reduced the stability and sensitivity of the endurance measurement, which could also have contributed to the absence of significant between-group differences.

The literature presents conflicting findings regarding the effects of CT protocols on balance. While adverse effects on the mediolateral component were observed following a 15-minute application [47], other studies found no detriment to lower limb balance after 20 min [48]. However, the predominant focus on the lower extremity leaves the specific influence of application duration on upper-limb balance ambiguous. Addressing the upper extremity, previous evidence demonstrates that a 20-minute application does not impair dynamic balance; conversely, it has been associated with enhanced medial and superolateral scores, supporting its use prior to performance [27]. Similarly, our findings indicate that dynamic balance was not adversely affected by 15- or 20-minute applications. Furthermore, although no significant between-group differences were observed, exploratory within-group analyses indicated acute improvements in medial directional scores following the 10-minute protocol (d = 0.46). Our findings, therefore, suggest that the applied shoulder cryotherapy protocols do not adversely affect upper-extremity dynamic balance in the short term. However, because the Upper Quarter Y-Balance Test involves motor coordination and was repeated after the intervention, and because no non-cooling control group was included, potential learning, practice, and familiarization effects cannot be excluded. Therefore, these within-group changes should be interpreted primarily as descriptive observations rather than definitive effects of the cryotherapy interventions.

In conclusion, the primary finding of this study is that no cryotherapy duration demonstrated statistical superiority over the others for upper-extremity performance; thus, our initial hypotheses regarding between-group superiority were not statistically supported. Nevertheless, exploratory within-group analyses revealed that the 10-, 15-, and 20-minute CT protocols were all associated with acute, descriptive improvements in flexibility and endurance. While the 10-minute application was the only protocol not followed by significant within-group reductions in maximal isometric strength, the absence of a non-cooling control group precludes definitive clinical recommendations. Therefore, the suggestion that a 10-minute protocol might be preferable when preserving muscle force remains a tentative, hypothesis-generating observation rather than an evidence-based prescription, warranting confirmation in future controlled trials.

Limitations and future directions

This study has several limitations. First, the sample consisted of healthy, predominantly female university students from a single institution, which limits the generalizability of these results to professional athletes, older adults, or patients with musculoskeletal injuries. Additionally, the study was not powered to detect sex-specific responses, as each group included only eight male participants; therefore, robust sex-stratified analyses could not be performed. Second, the study focused exclusively on the acute effects of specific cryotherapy durations (10, 15, and 20 min). Third, the absence of a non-cryotherapy control group makes it difficult to isolate the true intervention effects from potential learning, familiarization, or repeated-testing effects. Specifically, because only three practice trials were provided for the Y-Balance Test—fewer than the 5–6 trials often recommended to reach a performance plateau—we acknowledge that the observed post-test improvements in balance may partly reflect task familiarity. In addition, the fixed testing sequence (strength, flexibility, endurance, balance) with short rest intervals may have introduced sequencing bias, as fatigue from the endurance test could have influenced subsequent dynamic balance scores.

Fourth, using a single unblinded assessor for both interventions and outcome assessments introduces a potential risk of observer bias. Furthermore, a formal assessment of intra-rater reliability for the hand-held dynamometry was not conducted for this sample. Fifth, defining the cryotherapy “dose” solely by application duration, without measuring skinfold thickness or actual tissue temperature, limits mechanistic interpretations, given that subcutaneous fat acts as a thermal insulator. Finally, regarding the measurement tools, the Back Scratch Test is a composite measure lacking goniometric precision. While employing a single trial for the Modified Push-up Test was a deliberate methodological choice to prevent excessive muscular fatigue from confounding the subsequent dynamic balance assessments, we acknowledge that a single trial may not fully capture the inherent variability of performance-based endurance tests.

Future studies should employ longitudinal designs and non-treatment control groups to examine whether these functional and strength-related responses are modifiable over time. Investigating both the acute and long-term effects of cryotherapy across specific athletic and rehabilitation contexts, while addressing these methodological gaps, will further clarify its optimal application.

Conclusion

This study investigated the acute effects of three cryotherapy durations (10, 15, and 20 min) on upper-extremity performance and found no significant between-group differences. Therefore, the primary hypothesis of duration-dependent superiority was not supported. While exploratory within-group analyses indicated that all durations tended to show similar descriptive improvements in flexibility and endurance, the 10-minute protocol was the only duration not associated with significant within-group reductions in maximal isometric muscle strength. These findings suggest that application time may be associated with differences in the balance between functional gains and strength preservation. However, given the absence of a non-cooling control group, these acute within-group responses do not establish the definitive superiority of a single duration. Instead, they provide descriptive and exploratory insights that may inform future research, serving as hypothesis-generating observations rather than definitive evidence for individualized cryotherapy prescription in training and rehabilitation.

Abbreviations

CT

Cryotherapy

IR

Internal rotation

ER

External rotation

YBT-UQ

Upper Quarter Y-Balance Test

RM-ANOVA

Repeated-measures analysis of variance

ICC

Intraclass Correlation Coefficient

Authors’ contributions

FB.Y: Conceptualization, Methodology, Writing- original Draft, Data curation, Analysis, Investigation, Visualization. B.CY: Conceptualization, Methodology, Analysis, Supervision, Writing- review & editing. All authors have read and agreed to the published version of the manuscript.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

Data available on request from the authors.

Declarations

Ethical approval and consent to participate

The study was approved by The Muğla Sıtkı Koçman University Clinical Research Ethics Committee (Decision Date: 15.12.2021, Decision No: 28). The study was registered at ClinicalTrials.gov (NCT05355584). The study was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants included in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.