Sage Journals: Discover world-class research

Abstract

BACKGROUND:

In order to guide sports active subjects with musculoskeletal (MSK) shoulder dysfunctions it is essential to have precise insights of possible weak links in the kinetic chain of their shoulder girdle with access to reference data. Equally, it is uncertain whether so-called functional shoulder strength tests are complementary to an isokinetic shoulder strength test.

OBJECTIVE:

To establish reference values and to clarify the relationship between two upper extremity field tests – the Scapular Muscle Endurance test (SME) and the Closed Kinetic Chain Upper Extremity Stability Test (CKCUEST) and the isokinetic strength test of the shoulder internal (IR) and external rotators (ER).

METHODS:

Seventy-six male sport active men performed the CKCUEST, the SME, and underwent a bilateral isokinetic strength test to establish reference values of the isokinetic shoulder relative to body mass Peak Moment (PM/kg) and Power (Power/kg) for the IR and ER. A mixed-design ANOVA calculated possible side-to-side differences and one-way ANOVA was applied for eventually age group differences. Additionally, the relationship between the three tests was examined with a Spearman rank-order correlation coefficient.

RESULTS:

No significant correlations were indicated between the SME and the criterion isokinetic test and between the SME and CKCUEST.

CONCLUSIONS:

The main outcome parameters in this study warrant individual assessment as they measure different constructs, leading to a diverse profile of the shoulder.

Keywords

Scapular muscle endurance test upper extremity field test isokinetic test shoulder

1. Introduction

Shoulder pain (SP) is a matter of concern within the working [1, 2, 3] and general population [4, 5, 6]. This is shown by an increasing overall 1-year prevalence of chronic SP in the general population between the years 1994 and 2004 (46.7%–55.2%) [4]. Smaller percentages (2.28%; 15.7%) were observed in UK [6] and Norwegian [5] population indicating international differences and highly varying prevalence numbers. These percentages seem also in line with an active working male population showing a 1-year prevalence and incidence of acute SP around 16.2% [3] and 10.4% [1]. These percentages can be extended for the 1-year incidence of overall chronic SP in the French active working population [2].

Table 1
Anthropometric data of the subjects ( $N=$ 76) all displayed in (Median and (Interquartile range))

Age group	Age (y)	Height (m)	Weight (kg)	BMI (kg/m ${}^{2}$ )	Work index	Sport index	Leisure-time index	Total PA index
20–29 y ( $N=$ 25)	24 (22–28) ${}^{\text{a,b,c,d}}$	1.8 (1.77–1.84)	77 (73–82)	24 (23–25) ${}^{\text{a}}$	3.1 (2.8–3.6)	3.5 (3.3–3.9)	3 (2.5–3.5)	9.6 (8.8–10.4)
30–39 y ( $N=$ 18)	33 (31–37) ${}^{\text{a,b,c,d}}$	1.81 (1.75–1.84)	76 (73–87)	24 (23–26)	3.1 (2.6–3.4)	3.5 (2.8–4.1)	3.5 (3–3.8)	10.0 (9.2–11.0)
40–49 y ( $N=$ 20)	45 (41–47) ${}^{\text{a,b,c,d}}$	1.8 (1.75–1.83)	81 (73–87)	25 (24–27)	3.1 (2.7–3.6)	3.1 (2.8–4.0)	3 (2.5–3.5)	9.3 (8.9–10.3)
50–55 y ( $N=$ 13)	53 (51–54) ${}^{\text{a,b,c,d}}$	1.73 (1.7–1.83)	83 (75–89)	26 (25–28) ${}^{\text{a}}$	2.9 (2.6–3.4)	3.3 (2.8–3.8)	3.3 (2.4–3.4)	9.3 (7.6–10.6)

Abbreviations: BMI, body mass index; PA, physical activity. ${}^{\text{a,b,c,d}}$ the same letter indicates difference between these age groups ( $P<$ 0.05).

These numbers stress the importance of including future interventions to reduce the incidence of chronic SP at work within an active population [1, 2]. The interest of applying interventions to cope with chronic musculoskeletal (MSK) SP is increasing in order to avoid chronicity. Especially, chronic SP is mostly associated with increased disability at work or during leisure time, higher pain levels at 6 months, increased sick leave and increased medical consumption [5, 7].

For individualized interventions, several factors have to be taken into account since there is a positive association between SP at work and the following factors: age, high perceived physical exertion, work with arms above shoulder level [1], forceful effort and job demand [2]. From a prospective cohort study during 6 months within Dutch general practices 35%, 24% and 41% of the patients presented with respectively acute, sub-acute and chronic SP. Perceived causes of SP reported by the patient were mostly overuse and to a lesser extent trauma events [7]. Despite the detection of previously stated associated factors, the true mechanism of several shoulder pathology causing SP are not yet fully understood [1, 2, 5, 7, 8]. Moreover, for a subgroup of patients with SP with tear of supraspinatus, pain intensity is not a sensitive factor, rather, it is decreased external rotation (ER) PM/kg (Peak Moment) measured isokinetically [9]. Although patients with subacromial impingement show a decreased strength of the (IR) and ER on their affected side [10], the results vary depending on the isokinetic shoulder strength protocol [11]. A recent systematic review [12] has indicated that the supine isokinetic protocol has a relatively small measurement error [13] and is reliable [12].

In addition to shoulder strength testing, Physical Performance Tests (PPT) have become increasingly popular [14, 15]. They also have the advantage of being mainly multi-joint [31] and thus more functional than the isokinetic shoulder IR and ER test. The Closed Kinetic Chain Upper Extremity Stability Test (CKCUEST) seems to discriminate for subacromial impingement syndrome [16], but this evidence is still not confirmed as there are conflicting reports [17]. Limited but promising evidence exists also with regard to another functional PPT of the posterior kinetic chain, the Scapular Muscle Endurance test (SME), which may discriminate between chronic SP and asymptomatic subjects [18] and subacromial impingement syndrome [19]. Some reviews have proposed the CKCUEST as one of the more researched PPTs to measure functional performance of the shoulder [14, 20]. Another review discourages their use as single ‘best’ test to predict or screen for SP in an active population [21]. Hence there is a tendency to combine several PPTs [22, 23] since applied as a single PPT they still do not have sufficient measurement properties to guarantee their use in determining return to play or work [14]. Lastly, the relationship between the two previously mentioned PPTs and the golden standard isokinetic shoulder rotational strength [20, 24] is still uncertain except for the CKCUEST where two studies have already shown high to moderate correlations [25, 26] with the isokinetic shoulder IR and ER PM.

Thus the main objective of this study was to provide grouped age reference data for the isokinetic shoulder IR and ER strength and Power, but also for the 2 PPTs within sports active individuals and their inter correlations.

2. Methods

2.1 Participants

The sample of this study consisted of 76 men in active duty aged 20–55 years who were tested over a time interval of one year and a half starting from September 2018. Their anthropometric characteristics are outlined in Table 1. Participants were recruited from the Belgian Defense Forces after being informed about the goal and the inclusion and the exclusion criteria by e-mail, advertising poster, or verbal announcement. All tests conformed to The Code of Ethics of the World Medical Association (Declaration of Helsinki). After expressing their interest and understanding by signing the written informed consent, they were invited for an intake and eligibility screening. The ethical committee of University Hospital Center Brugmann approved this study (Reference CE 2018/93).

The following exclusion criteria were applied: 1. previous pain or having undergone surgery in the upper extremity or spine within a year before the onset of this study [27]; 2. failing to pass the annual side bridge military physical fitness test; or 3. having less than 60 ${}^{\circ}$ IR and 55 ${}^{\circ}$ ER in the shoulder. Also, having pain on one of the following clearing tests like the painful arc test, cross body adduction test, and 5-s of the CKCUEST led also to exclusion [16, 17]. The combat (C) subgroups included paratroopers or infantry soldiers; the combat service (CS) were mainly technicians and the combat service support (CSS) were mainly medical, administrative, and logistic corps.

2.2 Procedure

The same physiotherapist (first author), familiarized with all the previously stated tests, conducted all the tests. To prevent order biasing the test sequence was randomized. After each test, a 5-min break was allowed [28]. Before testing, all subjects were given a detailed explanation of the testing procedures. The following baseline measures were taken into account: weight, height, (BMI), age, and physical activity level indices assessed by the Baecke questionnaire [29]. The standard warm-up before the first test consisted of ten times the PNF (Proprioceptive Neuromuscular Facilitation) flexion and extension D1 and D2 Pattern for each side with a light elastic band [30]. After a demonstration, the subject was positioned in the start position for each of the following tests. Verbal and tactile feedback were given during the tests to ensure proper execution.

Field tests

The Scapular Muscle Endurance test (SME) was originally developed from a strengthening exercise described by Sahrmann [31]. The test is meant to assess bilateral scapular stability and endurance of the scapulothoracic muscles [32]. For this test, the subject stood in front of a wall, with the shoulders and elbows in a 90/90 ${}^{\circ}$ position without touching the wall. The subject held an adjustable digital dynamometer (weight $=$ 150 g, HDB 5K5N, KERN & SOHN GmbH, Balingen, Germany) keeping a 1 kgf resistance by externally rotating the shoulders (Fig. 1). The distance between hands was adjusted for the individual shoulder width. At the same time, a custom-made adjustable spacer (31–52 cm, weight $=$ 100 g) was held between the subject’s elbows also on shoulder width to maintain the test position and to avoid compensatory shoulder movements. The test was terminated when the subject was unable to maintain the 1 kgf resistance, dropped the adjustable spacer or failed to maintain 90/90 ${}^{\circ}$ position defined as touching the adjustable measuring rod (SECA 222, Hamburg, Germany) with the dynamometer [28]. The subject’s score was the time to failure (TTF) measured by a stopwatch.

Figure 1.

Scapular muscle endurance test.

The Closed Kinetic Chain Upper Extremity Stability Test (CKCUEST) is a functional tool to measure the upper extremity stability and Power [34, 35, 36]. The CKCUEST was administered as followed: the subject took a push-up position for the start of the test with each hand on one piece of tape and keeping the body straight. Both pieces of tape were placed at 91.44 cm distance between the medial borders of the hands. Hands were placed at shoulder height and feet were positioned at shoulder width. The test was performed with shoes to prevent slipping (Fig. 2). The test consisted of moving one hand to touch the other hand while stabilizing the whole body and alternating sides. This sequence was repeated as quickly as possible for 15-s. Each start and stop were indicated by a beep of the timer (GYMBOSS, St. Clair, USA), the same as used for the SME, to standardize and a 45-s rest followed each trial. The first trial was sub-maximal for familiarization. The three following trials were taken into account and were videotaped with a smartphone to ensure the accuracy of test scores. Proper execution of the test included completion of the test with a straight body, arms perpendicular to the ground, keeping the feet in start position, without touching the floor with other body parts than hands and feet [16]. Each improperly executed trial was repeated after 45-s.

Figure 2.

Closed kinetic chain upper extremity stability test.

The number of taps during the 15-s were recorded [35]. A tap counted when one hand touched the opposite hand. The mean value was calculated from the last three maximal trials and applied as the subject’s test score. Three types of scores were calculated: the raw score as the number of taps by the subject, the normalized score, the number of taps divided by the height, and a Power score by multiplying the average number of taps with 68% of the body weight in kilograms divided by 15 [37]. This last score represents the amount of work performed per time unit. The CKCUEST was found to be reliable in male college students (ICC 0.922) [35].

2.3 Isokinetic strength test

To have a more analytical or single-joint measure [38] of the shoulder’s, rotational strength a criterion standard isokinetic test is recommended [39]. We opted to test the participants in a supine position with the shoulder in 45 ${}^{\circ}$ of abduction (measured with a 30 cm goniometer) with 0 ${}^{\circ}$ forward flexion and the elbow in 90 ${}^{\circ}$ of flexion with neutral forearm rotation [12]. The elbow rested fixated in the rotation cuff pad, with the longitudinal axis of the humerus approximating the axis of the dynamometer and the hand gripping the input shaft (Fig. 3). Two straps were fixed diagonally over the trunk to promote isolated force from the shoulder muscles. The output was reported without gravitational correction to exclude measurement error because some subjects may find it difficult to relax during the weighting process [40]. The limits of the range of motion (ROM) were fixed at 60 ${}^{\circ}$ IR and 55 ${}^{\circ}$ ER [12]. All tests were executed using the Humac/Norm (by CSMI, Henley Healthcare, Sugarland, TX, USA).

Figure 3.

The isokinetic shoulder internal and external rotation test with the isokinetic dynamometer (CSMI, Humac/Norm 502140, Henley Healthcare, Sugarland, TX, USA).

All further settings of the dynamometer were adapted individually to create the above-mentioned test position. Through standardization, first the non-dominant limb was tested and then the dominant, defined as the preferred limb. During all tests, participants were encouraged orally, but visual feedback from the computer screen was not permitted. Three familiarization series of 6 repetitions at 120 ${}^{\circ}$ /s concentric-concentric were done at 50–75% of the maximal Power output with 30-s rest in between. Familiarization was completed with 60-s of rest, then the testing protocol started.

First, the subjects performed a maximal concentric-concentric IR-ER protocol at 120 ${}^{\circ}$ /s, four maximal repetitions were executed following 3 repetitions for familiarization. After a 60-s rest, 5 maximal repetitions were similarly executed in the concentric-concentric mode at 240 ${}^{\circ}$ /s. Lastly, 3 maximal repetitions were similarly executed in concentric-concentric mode at 60 ${}^{\circ}$ /s.

Table 2

Demographic characteristics of the subjects ( $N=$ 76) all displayed in frequencies and % of total

	20–29 y ( $N=$ 25)	30–39 y ( $N=$ 18)	40–49 y ( $N=$ 20)	50–55 y ( $N=$ 13)
Hand dominance ( $N=$ 76)
Left	4 (5%)	3 (4%)	2 (3%)	3 (3%)
Right	21 (28%)	15 (20%)	18 (24%)	10 (13%)
Profession ( $N=$ 76)
Combat	1 (1%)	2 (3%)	0 (0%)	0 (0%)
Combat service	0 (0%)	3 (4%)	2 (3%)	0 (0%)
Combat service support	24 (32%)	13 (17%)	18 (24%)	13 (17%)
Sports participation ( $N=$ 76)
All-round	14 (18%)	6 (8%)	7 (9%)	3 (4%)
Endurance	11 (14%)	12 (16%)	13 (17%)	10 (13%)

Following a 5-minute break, which allowed for a change of sides, the tests were repeated for the other limb using the same testing protocol as described above. The following outcomes were retained for further analysis: the bodyweight normalized peak moment (Nm/kgbw) and Power (Watt/kgbw) values for the IR, the ER, and the ER/IR ratio [12, 27].

2.4 Statistical analysis

All statistical tests were run on SPSS (IBM Corp, released 2019, IBM SPSS Statistics for Windows, Version 26.0, Armonk, NY). Outcome measures are displayed as mean $\pm$ standard deviation (SD) scores with 95% confidence interval for parametric and as median score with the interquartile scores for the non-parametric outcomes.

A descriptive analysis was performed on the anthropometric data and in those cases where the following parameters did not meet the assumption for normality: age, weight, BMI, preferred side, class, leisure-time index, and combined score, a Kruskal-Wallis ANOVA was performed. The demographic data were presented in frequencies.

For the reference data, possible side differences (within subject variable) between age groups (between-subject variable) were analyzed by a mixed-design ANOVA for all isokinetic test parameters. The following parameters did not meet the assumption for normality: the Non-Dominant (ND) PM/kg and the ND PM/kg ratio at 120 ${}^{\circ}$ /s, but also the Dominant (D) and ND PM/kg ratio at 60 ${}^{\circ}$ /s. All parameters met the assumption for equal variance. The assumption for sphericity was not met as shown by the Mauchly’s test sphericity $\chi^{2}$ (629) $=$ 5510.577, $P<$ 0.001 thus the degrees of freedom were corrected using Greenhouse-Geisser estimates of sphericity ( $\varepsilon=$ 0.070). Post-hoc analyses were performed by a Tukey test. Furthermore, age groups differences were analyzed by a one-way ANOVA with additionally a Bonferroni post-hoc analyses. The latter tests were also used to examine age group differences between the SME and the CKCUEST. All parameters met the assumption for normality. Only the Power score of the CKCUEST didn’t meet the assumption homogeneity of variance (F (3, 65.952) $=$ 1.56, $P>$ 0.05). The statistical significance alpha level was set at 0.05. Cohen’s d Effect Size was calculated for significant findings.

To examine the correlation between the three test data a Spearman Rank-Order Correlation Coefficient was calculated on each dependent variable because most of these were not normally distributed based on the Shapiro-Wilk tests. This was also done between some isokinetic IR and ER parameters in order to support our conclusions. The following categories for the Spearman rank-order correlation coefficient were used: negligible ( $\leqslant$ 0.30), low ( $>$ 0.30 $\leqslant$ 0.50), moderate ( $>$ 0.50 $\leqslant$ 0.70), high ( $>$ 0.70 $\leqslant$ 9.00), and very high ( $>$ 0.90 $\leqslant$ 1.0) [41]. For all non-parametric test outcome measures, data are displayed in median with interquartile range (IQR) as the distance between the 25th and 75th percentiles [42].

A priori power analysis for the post-hoc mixed design ANOVA by the G-Power program (G*Power Version 3.1.9.4, Dusseldorf, Germany) concluded the following there was 81% chance of correctly rejecting the null hypothesis of no significant effect of the interaction with 4 age groups for a total of 48 subjects with 12 subjects per group.

3. Results

The descriptive anthropometric data are displayed in Table 1. Only the mean BMI of the youngest age group was significantly lower than those in the oldest age group at post hoc testing. The age groups were comparable for all other anthropometric characteristics. The demographic characteristics are presented in Table 2 with their percentage of appearance in the sample ( $N=$ 76). The sport participation measured by the Baecke et al. questionnaire [29] asked for the 2 main sports per subject. If the subject reported two kinds of endurance sports (i.e., running and cycling) they were counted as one and if 2 mixed sports (i.e., running and climbing) were reported this was counted as all-round sport.

Table 3 outlines the isokinetic reference values for the relative strength of the IR and ER measured at the 3 speeds. Table 4 displays similar results for the relative

Table 3
Reference values of isokinetic peak moment for internal and external rotation of the shoulder in Mean $\pm$ SD (95% Confidence interval) for the PM/kg (Nm/kg)

Angular speed	Age group	D	ND	% difference	D	ND	% difference	Ratio D	Ratio ND
		Internal rotators			External rotators
60 ${}^{\circ}$ /s	20–29 y ( $N=$ 25)	0.562 $\pm$ 0.092(0.520, 0.604)	0.566 $\pm$ 0.083(0.527, 0.606) ${}^{\text{a}}$	0.9 $\pm$ 7.2	0.395 $\pm$ 0.070(0.365, 0.422) ${}^{\text{a}}$	0.384 $\pm$ 0.064(0.356, 0.412)	2.5 $\pm$ 5.5	0.707 $\pm$ 0.090(0.670, 0.745)	0.684 $\pm$ 0.090(0.642, 0.726)
	30–39 y ( $N=$ 18)	0.537 $\pm$ 0.104(0.487, 0.586)	0.522 $\pm$ 0.108(0.475, 0.567)	2.8 $\pm$ 8.5	0.357 $\pm$ 0.071(0.323, 0.390)	0.348 $\pm$ 0.077(0.316, 0.381)	2.4 $\pm$ 6.5	0.668 $\pm$ 0.071(0.623, 0.712)	0.674 $\pm$ 0.119(0.625, 0.724)
	40–49 y ( $N=$ 20)	0.502 $\pm$ 0.128(0.455, 0.549)	0.480 $\pm$ 0.122(0.435, 0.524) ${}^{\text{a}}$	4.6 $\pm$ 8.1	0.352 $\pm$ 0.069(0.320, 0.384)	0.334 $\pm$ 0.070(0.303, 0.365)	5.1 $\pm$ 6.2	0.715 $\pm$ 0.121(0.672, 0.757)	0.714 $\pm$ 0.122(0.667, 0.761)
	50–55 y ( $N=$ 13)	0.477 $\pm$ 0.089(0.419, 0.535)	0.483 $\pm$ 0.079(0.428, 0.539)	1.3 $\pm$ 10	0.317 $\pm$ 0.079(0.278, 0.357) ${}^{\text{a}}$	0.331 $\pm$ 0.06.8(0.293, 0.370)	4.3 $\pm$ 7.6	0.649 $\pm$ 0.087(0.600, 0.701)	0.651 $\pm$ 0.082(0.593, 0.709)
120 ${}^{\circ}$ /s	20–29 y ( $N=$ 25)	0.522 $\pm$ 0.084(0.487, 0.557)	0.552 $\pm$ 0.074(0.521, 0.583) ${}^{\text{a,b}}$	5.6 $\pm$ 7.2	0.349 $\pm$ 0.069(0.323, 0.376)	0.372 $\pm$ 0.058(0.348, 0.396) ${}^{\text{a,b}}$	2.7 $\pm$ 5.5	0.679 $\pm$ 0.111(0.633, 0.725)	0.670 $\pm$ 0.074(0.632, 0.709)
	30–39 y ( $N=$ 18)	0.490 $\pm$ 0.094(0.443, 0.537)	0.498 $\pm$ 0.088(0.455, 0.542)	1.7 $\pm$ 8	0.318 $\pm$ 0.071(0.287, 0.350)	0.332 $\pm$ 0.076(0.294, 0.369)	2.1 $\pm$ 12.2	0.663 $\pm$ 0.088(0.619, 0.706)	0.662 $\pm$ 0.118(0.617, 0.707)
	40–49 y ( $N=$ 20)	0.469 $\pm$ 0.120(0.412, 0.525)	0.457 $\pm$ 0.112(0.405, 0.503) ${}^{\text{b}}$	2.5 $\pm$ 7.6	0.311 $\pm$ 0.056(0.282, 0.341)	0.313 $\pm$ 0.066(0.282, 0.343) ${}^{\text{b}}$	0.4 $\pm$ 6	0.681 $\pm$ 0.109(0.630, 0.731)	0.696 $\pm$ 0.113(0.653, 0.739)
	50–55 y ( $N=$ 13)	0.444 $\pm$ 0.090(0.390, 0.499)	0.461 $\pm$ 0.083(0.410, 0.511) ${}^{\text{a}}$	3.6 $\pm$ 9.4	0.292 $\pm$ 0.070(0.256, 0.329)	0.300 $\pm$ 0.068(0.259, 0.341) ${}^{\text{a}}$	2.5 $\pm$ 7.5	0.637 $\pm$ 0.071(0.594, 0.680)	0.617 $\pm$ 0.066(0.564, 0.670)
240 ${}^{\circ}$ /s	20–29 y ( $N=$ 25)	0.438 $\pm$ 0.090(0.400, 0.476)	0.469 $\pm$ 0.081(0.434, 0.504) ${}^{\text{a}}$	6.9 $\pm$ 6.3	0.287 $\pm$ 0.061(0.261, 0.313) ${}^{\text{a}}$	0.292 $\pm$ 0.057(0.266, 0.317)	1.7 $\pm$ 5.3	0.675 $\pm$ 0.119(0.631, 0.719)	0.629 $\pm$ 0.099(0.590, 0.668)
	30–39 y ( $N=$ 18)	0.425 $\pm$ 0.084(0.381, 0.470)	0.420 $\pm$ 0.076(0.379, 0.461)	1.2 $\pm$ 7.4	0.270 $\pm$ 0.065(0.240, 0.300)	0.265 $\pm$ 0.072(0.235, 0.295)	1.9 $\pm$ 6.2	0.638 $\pm$ 0.080(0.586, 0.689)	0.620 $\pm$ 0.104(0.574, 0.666)
	40–49 y ( $N=$ 20)	0.401 $\pm$ 0.107(0.360, 0.444)	0.400 $\pm$ 0.106(0.361, 0.440)	0.4 $\pm$ 7	0.268 $\pm$ 0.064(0.239, 0.297)	0.260 $\pm$ 0.065(0.229, 0.290)	3.2 $\pm$ 5.9	0.684 $\pm$ 0.117(0.630, 0.733)	0.660 $\pm$ 0.106(0.616, 0.704)
	50–55 y ( $N=$ 13)	0.374 $\pm$ 0.098(0.322, 0.427)	0.372 $\pm$ 0.086(0.323, 0.420) ${}^{\text{a}}$	0.7 $\pm$ 8.7	0.226 $\pm$ 0.072(0.190, 0.262) ${}^{\text{a}}$	0.234 $\pm$ 0.067(0.193, 0.274)	3.3 $\pm$ 7.3	0.595 $\pm$ 0.115(0.535, 0.656)	0.589 $\pm$ 0.075(0.535, 0.644)

Abbreviations: Diff, difference; D, dominant side; ND, non-dominant side; PM, Peak Moment; ${}^{\text{a,b}}$ indicates between-groups differences ( $P<$ 0.05).

Table 4

Reference values of isokinetic angular speed for internal and external rotation of the shoulder in Mean $\pm$ SD (95% Confidence interval) for the Power/kg (Watts/kg)

Angular speed	Age group	D	ND	% difference	D	ND	% difference	Ratio D	Ratio ND
		Internal rotators			External rotators
60 ${}^{\circ}$ /s	20–29 y ( $N=$ 25)	0.464 $\pm$ 0.080(0.429, 0.499)	0.464 $\pm$ 0.071(0.430, 0.497) ${}^{\text{a}}$	0.1 $\pm$ 5.4	0.332 $\pm$ 0.069(0.307, 0.358) ${}^{\text{a}}$	0.328 $\pm$ 0.061(0.303, 0.354) ${}^{\text{a}}$	1.2 $\pm$ 5.1	0.713 $\pm$ 0.088(0.673, 0.754)	0.716 $\pm$ 0.091(0.671, 0.760)
	30–39 y ( $N=$ 18)	0.438 $\pm$ 0.076(0.396, 0.479)	0.417 $\pm$ 0.076(0.378, 0.456)	4.8 $\pm$ 6.3	0.296 $\pm$ 0.068(0.266, 0.327)	0.294 $\pm$ 0.072(0.264, 0.325)	0.6 $\pm$ 6.0	0.669 $\pm$ 0.073(0.621, 0.717)	0.702 $\pm$ 0.125(0.649, 0.754)
	40–49 y ( $N=$ 20)	0.419 $\pm$ 0.112(0.379, 0.458)	0.393 $\pm$ 0.107(0.355, 0.430) ${}^{\text{a}}$	6.4 $\pm$ 6.0	0.293 $\pm$ 0.052(0.264, 0.322)	0.283 $\pm$ 0.066(0.254, 0.312)	3.6 $\pm$ 5.7	0.723 $\pm$ 0.128(0.677, 0.768)	0.744 $\pm$ 0.133(0.694, 0.794)
	50–55 y ( $N=$ 13)	0.386 $\pm$ 0.076(0.337, 0.435)	0.389 $\pm$ 0.072(0.343, 0.435)	0.9 $\pm$ 7.4	0.272 $\pm$ 0.07(0.236, 0.307) ${}^{\text{a}}$	0.270 $\pm$ 0.060(0.234, 0.305) ${}^{\text{a}}$	0.7 $\pm$ 7.1	0.673 $\pm$ 0.114(0.617, 0.729)	0.663 $\pm$ 0.088(0.602, 0.725)
120 ${}^{\circ}$ /s	20–29 y ( $N=$ 25)	0.760 $\pm$ 0.139(0.703, 0.818)	0.815 $\pm$ 0.125(0.763, 0.866) ${}^{\text{a,b}}$	6.9 $\pm$ 10.4	0.527 $\pm$ 0.124(0.480, 0.574)	0.555 $\pm$ 0.100(0.512, 0.598) ${}^{\text{a,b}}$	5.1 $\pm$ 8.8	0.697 $\pm$ 0.130(0.667, 0.755)	0.682 $\pm$ 0.075(0.645, 0.718)
	30–39 y ( $N=$ 18)	0.724 $\pm$ 0.133(0.658, 0.790)	0.731 $\pm$ 0.132(0.665, 0.796)	0.9 $\pm$ 12.2	0.484 $\pm$ 0.121(0.429, 0.540)	0.486 $\pm$ 0.121(0.435, 0.536)	0.2 $\pm$ 10.4	0.666 $\pm$ 0.090(0.645, 0.748)	0.666 $\pm$ 0.105(0.623, 0.710)
	40–49 y ( $N=$ 20)	0.697 $\pm$ 0.183(0.612, 0.783)	0.685 $\pm$ 0.185(0.599, 0.772) ${}^{\text{b}}$	1.8 $\pm$ 11.6	0.467 $\pm$ 0.101(0.414, 0.519)	0.464 $\pm$ 0.104(0.416, 0.512) ${}^{\text{b}}$	0.6 $\pm$ 9.9	0.684 $\pm$ 0.104(0.674, 0.772)	0.695 $\pm$ 0.110(0.653, 0.736)
	50–55 y ( $N=$ 13)	0.637 $\pm$ 0.136(0.555, 0.720)	0.674 $\pm$ 0.126(0.598, 0.750) ${}^{\text{a}}$	5.6 $\pm$ 14.4	0.427 $\pm$ 0.126(0.362, 0.493)	0.441 $\pm$ 0.111(0.381, 0.500) ${}^{\text{a}}$	3.1 $\pm$ 12.2	0.645 $\pm$ 0.096(0.597, 0.718)	0.614 $\pm$ 0.070(0.563, 0.665)
240 ${}^{\circ}$ /s	20–29 y ( $N=$ 25)	0.938 $\pm$ 0.242(0.838, 1.04)	1.06 $\pm$ 0.232(0.966, 1.16) ${}^{\text{a}}$	12.4 $\pm$ 15.8 ${}^{*}$	0.648 $\pm$ 0.159(0.581, 0.715)	0.656 $\pm$ 0.158(0.586, 0.727) ${}^{\text{a}}$	1.3 $\pm$ 13.8	0.702 $\pm$ 0.140(0.650, 0.753)	0.625 $\pm$ 0.115(0.580, 0.670)
	30–39 y ( $N=$ 18)	0.953 $\pm$ 0.217(0.836, 1.07)	0.953 $\pm$ 0.208(0.842, 1.065)	0.1 $\pm$ 18.7	0.592 $\pm$ 0.182(0.513, 0.671)	0.595 $\pm$ 0.193(0.512, 0.678)	0.5 $\pm$ 16.3	0.614 $\pm$ 0.097(0.553, 0.674)	0.616 $\pm$ 0.121(0.563, 0.668)
	40–49 y ( $N=$ 20)	0.908 $\pm$ 0.294(0.797, 1.02)	0.893 $\pm$ 0.272(0.787, 1.00)	1.7 $\pm$ 17.7	0.589 $\pm$ 0.162(0.514, 0.664)	0.563 $\pm$ 0.175(0.484, 0.642)	4.5 $\pm$ 15.5	0.671 $\pm$ 0.132(0.613, 0.728)	0.643 $\pm$ 0.115(0.592, 0.693)
	50–55 y ( $N=$ 13)	0.786 $\pm$ 0.229(0.648, 0.924)	0.831 $\pm$ 0.232(0.699, 0.962) ${}^{\text{a}}$	5.5 $\pm$ 21.9	0.502 $\pm$ 0.178(0.409, 0.595)	0.489 $\pm$ 0.192(0.391, 0.587) ${}^{\text{a}}$	2.7 $\pm$ 19.2	0.623 $\pm$ 0.140(0.552, 0.694)	0.540 $\pm$ 0.089(0.478, 0.602)

Abbreviations: Diff, difference; D, dominant side; ND, non-dominant side; ${}^{*}$ indicates significant side difference; ${}^{\text{a,b}}$ indicates between-groups differences ( $P<$ 0.05).

Table 5

Reference values of the CKCUEST and the SME (min:sec) in Mean $\pm$ SD (95% CI), minimum and maximum

	Age group	Mean $\pm$ SD (95% CI)	Min	Max
SME (min:sec)	20–29 y ( $N=$ 25)	02:33 $\pm$ 00:59 (02:09, 02:58)	01:15	05:28
	30–39 y ( $N=$ 18)	02:51 $\pm$ 01:20 (02:11, 03:30)	01:05	05:30
	40–49 y ( $N=$ 20)	02:28 $\pm$ 00:58 (02:01, 02:55)	01:04	04:51
	50–55 y ( $N=$ 13)	02:22 $\pm$ 01:04 (01:43, 03:00)	00:57	04:16
Raw CKCUEST (taps)	20–29 y ( $N=$ 25)	27 $\pm$ 3.8 (25.2, 28.4) ${}^{\text{a}}$	19	35
	30–39 y ( $N=$ 18)	27 $\pm$ 3.4 (25.3, 28.6) ${}^{\text{b}}$	22	34
	40–49 y ( $N=$ 20)	25 $\pm$ 3.9 (23.5, 27.1)	17	33
	50–55 y ( $N=$ 13)	23 $\pm$ 3.5 (21.2, 25.5) ${}^{\text{a,b}}$	16	29
Normalized CKCUEST (taps/height)	20–29 y ( $N=$ 25)	15 $\pm$ 2.1 (14, 15.7)	12.8	19.2
	30–39 y ( $N=$ 18)	15 $\pm$ 1.7 (14.1, 15.8)	9.3	18.8
	40–49 y ( $N=$ 20)	14 $\pm$ 2.2 (13.2, 15.2)	9.6	18.5
	50–55 y ( $N=$ 13)	13 $\pm$ 2.1 (12, 14.6)	12.8	17
Power CKCUEST (taps * weight 68%/15)	20–29 y ( $N=$ 25)	94 $\pm$ 16.2 (87.5, 100.9)	68.9	136
	30–39 y ( $N=$ 18)	96 $\pm$ 12.9 (89.8, 102.7)	74.9	115.2
	40–49 y ( $N=$ 20)	94 $\pm$ 18.2 (85.5, 102.6)	59	123.7
	50–55 y ( $N=$ 13)	86 $\pm$ 9.2 (80, 91.1)	68.1	102.5

Abbreviations: SME, Scapular Muscle Endurance test, CKCUEST, Closed Kinetic Chain Upper Extremity Stability Test; ${}^{\text{a,b}}$ indicates between-groups differences ( $P<$ 0.05).

Power values. The mixed-design ANOVA showed some significant side differences over the different age groups $F$ (35, 1656) $=$ 237.747, $P<$ 0.001. These side differences were significant between the age groups, $F$ (3, 72) $=$ 4.365, $P=$ 0.007. There was no significant interaction $F$ (105, 2520) $=$ 1.353, $P=$ 0.011 between the side differences and the age groups. Nevertheless, post hoc analyses showed only one significant mean side-to-side difference at 240 ${}^{\circ}$ /s: IR strength for the youngest age group (12.4%, $P=$ 0.002, $d=$ 0.519). Moreover, for every speed and muscle group some significant differences were found between age groups, but not for the ER/IR ratios.

Table 5 outlines the functional shoulder test reference values for the SME and the three CKCUEST scores. Post hoc analyses showed only for the raw CKCUEST scores a significantly higher score in the two youngest groups versus the oldest group ( $P<$ 0.05), compared to the SME test and the other CKCUEST scores where no significant differences appeared between the age groups. Table 6 outlines several significant positive and negative values of the correlation coefficients ( $r_{s}$ ) between de CKCUEST and the criterion isokinetic test. No significant correlation was found between the SME and the latter or between the SME and the CKCUEST ( $-0.176-0.027=r_{s}$ ). For this reason no additional tables were created.

The values of the correlation coefficient between the IR and ER power and the PM at 60 ${}^{\circ}$ /s, at 120 ${}^{\circ}$ /s and at 240 ${}^{\circ}$ /sec were very high: 0.89–0.97 and statically significant ( $P<$ 0.01).

4. Discussion

The primary goal of this study was to establish reference grouped age values for the CKCUEST, the SME, and the isokinetic shoulder strength and Power for the rotators in sports-active men and to examine the possible relationship between these tests. The main observations were: (1) when using these reference values age should be taken into account with the exception for the SME test; (2) the findings of these tests are not interchangeable.

4.1 Reference values for the field and isokinetic tests

In the current study, the CKCUEST findings were comparable to these of elite track and field athletes [43]. Although 90% of the present healthy cohort had a medical, administrative or logistic job, the CKCUEST is commonly used in patient and general populations [16, 17]. Comparing the youngest age group of the present study to another but smaller group of overhead active and sedentary males, the mean score of 27 $\pm$ 3.8 taps of the present study is similar to their male athlete’s (27.13 $\pm$ 3.15 taps) and their sedentary population (25.30 $\pm$ 3.68 taps) [16] but stresses the influence of sport participation in the CKCUEST raw score. The same is not true for the mean CKCUEST Power score as the score in the present study (94 $\pm$ 16.2 taps) was slightly higher in comparison with their active (92.78 $\pm$ 13.92 taps) and almost the same as their sedentary male population (93.37 $\pm$ 16.10 taps) [16]. The same mean score was also higher than in a small age-matched pa-

Table 6
Correlation coefficient ( $r_{s}$ ) between the CKCUEST and criterion isokinetic test for the normalized strength and Power based on the Spearman rank-order correlation coefficient

Isokinetic linear speed	PM/kg (Nm/kg)		Power/kg (watts/kg)		PM/kg (Nm/kg)		Power/kg (watts/kg)		PM/kg (Nm/kg)		Power/kg (watts/kg)
	Raw CKCUEST				Normalized CKCUEST				Power CKCUEST
	D	ND	D	ND	D	ND	D	ND	D	ND	D	ND
60 ${}^{\circ}$ /s IR	0.519 ${}^{**}$	0.425 ${}^{**}$	0.484 ${}^{**}$	0.455 ${}^{**}$	0.526 ${}^{**}$	0.439 ${}^{**}$	0.500 ${}^{**}$	0.464 ${}^{**}$	0.238 ${}^{*}$	0.234 ${}^{*}$	0.195	0.214
60 ${}^{\circ}$ /s ER	0.332 ${}^{**}$	0.344 ${}^{**}$	0.290 ${}^{*}$	0.322 ${}^{**}$	0.321 ${}^{**}$	0.353 ${}^{**}$	0.282 ${}^{*}$	0.326 ${}^{**}$	0.178	0.188	0.154	0.161
60 ${}^{\circ}$ /s ER/IR	$-$ 0.270 ${}^{*}$	$-$ 0.091	$-$ 0.257 ${}^{*}$	$-$ 0.132	$-$ 0.299 ${}^{**}$	$-$ 0.105	$-$ 0.293 ${}^{*}$	$-$ 0.155	$-$ 0.185	$-$ 0.046	$-$ 0.106	$-$ 0.055
120 ${}^{\circ}$ /s IR	0.471 ${}^{**}$	0.453 ${}^{**}$	0.502 ${}^{**}$	0.456 ${}^{**}$	0.467 ${}^{**}$	0.475 ${}^{**}$	0.501 ${}^{**}$	0.466 ${}^{**}$	0.244 ${}^{*}$	0.236 ${}^{*}$	0.280 ${}^{*}$	0.244 ${}^{*}$
120 ${}^{\circ}$ /s ER	0.251 ${}^{*}$	0.296 ${}^{**}$	0.310 ${}^{**}$	0.331 ${}^{**}$	0.256 ${}^{*}$	0.323 ${}^{**}$	0.297 ${}^{**}$	0.356 ${}^{**}$	0.145	0.134	0.206	0.194
120 ${}^{\circ}$ /s ER/IR	$-$ 0.253 ${}^{*}$	$-$ 0.173	$-$ 0.239 ${}^{*}$	$-$ 0.169	$-$ 0.257 ${}^{*}$	$-$ 0.171	$-$ 0.265 ${}^{*}$	$-$ 0.171	$-$ 0.164	$-$ 0.149	$-$ 0.114	$-$ 0.103
240 ${}^{\circ}$ /s IR	0.495 ${}^{**}$	0.461 ${}^{**}$	0.467 ${}^{**}$	0.441 ${}^{**}$	0.500 ${}^{**}$	0.466 ${}^{**}$	0.451 ${}^{**}$	0.449 ${}^{**}$	0.271 ${}^{*}$	0.254 ${}^{*}$	0.309 ${}^{**}$	0.278 ${}^{*}$
240 ${}^{\circ}$ /s ER	0.288 ${}^{*}$	0.255 ${}^{*}$	0.327 ${}^{**}$	0.306 ${}^{**}$	0.285 ${}^{*}$	0.273 ${}^{*}$	0.321 ${}^{**}$	0.317 ${}^{**}$	0.148	0.152	0.215	0.223
240 ${}^{\circ}$ /s ER/IR	$-$ 0.238 ${}^{*}$	$-$ 0.178	$-$ 0.167	$-$ 0.121	$-$ 0.249 ${}^{*}$	$-$ 0.181	$-$ 0.159	$-$ 0.127	$-$ 0.185	$-$ 0.083	$-$ 0.165	$-$ 0.067

Abbreviations: IR, Internal Rotators, ER, External Rotators, D, dominant side; ND, non-dominant side; PM, peak moment. ${}^{*}$ Correlation is significant at the 0.05 level (2-tailed). ${}^{**}$ Correlation is significant at the 0.01 level (2-tailed).

tient and general population (22; 22 taps) [17] also indicating that higher physical activity levels give higher raw CKCUEST scores. Moreover, the age effect apparent between the youngest age groups and the oldest within the present cohort is supported by another study among overhead athletes [25] indicating higher raw CKCUEST scores can be expected in younger cohorts.

Considering the SME test, we did not find an age effect on the test scores. Since this test is not frequently applied the only comparison could be made to two studies with smaller asymptomatic cohorts [28, 42]. The mean score of 153 $\pm$ 59-s in the study is higher than the median score of 115.0-s (89.5–135.5) in a young healthy population (mean age 22; min-max: 19–24 years) [42] and much higher in comparison with another healthy population (mean age 20.42 $\pm$ 1.04 years) with a mean score of 37.19 $\pm$ 23.28-s [28]. This can possibly be explained by the difference in age group definition. Moreover, we used an additional item for the SME test (i.e., the measuring rod as extra objectification), an adjustable elbow support and the pull-dynamometer. For these reasons correct statement of the properties of the measuring items is recommended to standardize the further application of the SME test in athletes.

Isokinetic strength and Power of IR and ER are mostly used in MSK screenings for the shoulder so several reference values do exist to compare with [44]. Unfortunately, for the supine lying protocol tested with the arm positioned at 45 ${}^{\circ}$ abduction and 0 ${}^{\circ}$ flexion ( $<$ 15 subjects) reference data is limited to one study [12]. Two studies conducted the supine lying protocol with the arm at 90 ${}^{\circ}$ of abduction [40, 45]. We preferred the testing protocol of Forthomme et al. at 45 ${}^{\circ}$ of abduction to avoid pain provocation and its solid reproducibility [12]. We only retained the bodyweight normalized data because this method made it easier to compare with different populations in other studies.

One meaningful advantage of this method is its bilateral design and the detection of asymmetries. The only significant side asymmetry (12.4%) in favor of the ND side was detected for the IR Power at 240 ${}^{\circ}$ /s in the youngest age group. In variance with this finding in a study of US military personnel (Marines), there was an asymmetry in IR strength of $>$ 20% at 60 ${}^{\circ}$ /s, measured in sitting, which led to a significant increase in the odds of reporting shoulder injury in comparison with lower asymmetries ( $<$ 20%). However, when sex was taken into account only the female Marines ( $N=$ 91) kept showing this tendency [46]. This provides some evidence to take the specific nature of the screened population into account, as well as the critical effect of the test protocol, as the above finding could be affected by the test position.

The study [12], that inspired us using the specific supine protocol, tested a rather small sample ( $N=$ 12; 24.4 $\pm$ 1.7 y) and reported only absolute values, leaving only the ER/IR ratios for comparison. Our mean shoulder ER/IR ratios at 60 ${}^{\circ}$ /s and 240 ${}^{\circ}$ /s of the youngest group (D: 70.7 $\pm$ 9.0%; ND: 68.4 $\pm$ 9.0; D: 67.5 $\pm$ 11.9%; ND: 62.9 $\pm$ 9.9) were higher in comparison with those derived from the above study (60 ${}^{\circ}$ /s D: 68 $\pm$ 4%, 240 ${}^{\circ}$ /s D: 63 $\pm$ 7%) [12]. Another study consisting of 438 male military recruits matching the youngest age group of the present study applied the supine protocol in 90 ${}^{\circ}$ of abduction and found a ER/IR ratio of 79.6 $\pm$ 18.1% for the ND shoulder at 60 ${}^{\circ}$ /s, which is 10% higher than in the present study [40]. The same was true for a male age-matched elite handball team ( $N=$ 108) who also used the same isokinetic protocol and found ER/IR ratios at 60 ${}^{\circ}$ /s and 240 ${}^{\circ}$ /s (D: 73 $\pm$ 12%, ND: 81 $\pm$ 13%; D: 78 $\pm$ 14%, ND: 82 $\pm$ 17%) [45]. This was also found among amateur handball players (male $N=$ 34 and female $N=$ 32) for the ER/IR ratios at 60 ${}^{\circ}$ /s and 240 ${}^{\circ}$ /s (D: 68 $\pm$ 13%, D: 70 $\pm$ 13%) [47]. All these different ER/IR ratios demonstrate how important it is to have reference data per specific population and test protocol.

4.2 Relationship between field- and isokinetic testing

We observed rather moderate to low Spearman correlation coefficients between IR and ER strength at 60 ${}^{\circ}$ /s and the CKCUEST scores ( $r=$ 0.33–0.52). Two other studies which have examined this relationship (different dynamometer and test position) indicated very high Pearson correlation coefficients ( $r=$ 0.90–0.93) [48] within a mixed group Korean adults ( $N=$ 20 F $N=$ 20 M). Moderate Pearson correlation coefficients ( $r=$ 0.55–0.57) were present within again a mixed group of US active adults ( $N=$ 11 females; $N=$ 19 males) [26]. Lee et al. reported the mean values for the raw score of the CKCUEST (13.31 $\pm$ 4.78 taps within males) [48] which were much lower than the age-matched scores of 27 $\pm$ 3.8 taps in the present study, This could indicate a difference in activity level between both populations but also gender mixing and thus could partly explain the difference in size of correlations.

However these mixed correlations possibly reflect the difference in construct between both tests, which could be explained by the difference in muscle performance between the tests. During the CKCUEST, the muscles are providing isometric stabilization of the joint with glenohumeral movement while during isokinetic testing they are specifically executing a movement in the glenohumeral joint. Due to the lack of other studies examining the construct validity of the CKCUEST and the SME, the present finding provides a preliminary evidence that these three tests are not measuring the same construct in the present cohort. This supports the use of a combination of single-joint and multi-joint tests with upper limb evaluations [38].

Finally, in this study we have chosen the bodyweight normalized units for both strength and Power in view of their wide use. The addition of the latter parameter was motivated by the aspiration to derive additional information (beyond the PM). However, the very high correlations between these two outcome measures render Power a largely redundant parameter while leaving the PM (or PM/kg) as the quintessential isokinetic outcome parameter.

In terms of this study’s limitations, more participants could allow stronger comparisons between other studies and different populations (age groups, physical activity level subgroups and gender). In addition, previous studies already demonstrated significant age effects on the CKCUEST and hence future reference data studies should take this element into account, especially as in the present study we have under-represented the 50–55 age group ( $N=$ 13). Moreover, during conduction of the three tests, some excluded subjects reported unease such as neck pain during the SME ( $N=$ 1), painful wrists during CKCUEST ( $N=$ 1) and some subjects ( $N=$ 5) reported pain during the isokinetic test at the beginning of the IR or the last degrees of ER. Furthermore, the fixed hand distance of 91.44 cm (36 inch) for the CKCUEST could have been more challenging for some participants. Finally, this study applied a concentric isokinetic protocol in supine with specific speeds based on Forthomme et al. [12]. Running an eccentric isokinetic test within a sports active patient population could elicit higher pain levels for this reason only concentric tests were executed.

5. Conclusion

Low to moderate Spearman correlation coefficients between the isokinetic and functional tests is an indication that including both tests in shoulder girdle screening could have an additional value.

In view of the analysis, we recommend that the PM is used as the representative parameter while using the range of the concentric ER/IR ratios: 58.9%–71.5%, as a general reference for similar cohorts when tested using the same protocol.

Author contributions

CONCEPTION: T. Van Haele, L. Moerman, N. Duvigneaud and D. Van Tiggelen.

PERFORMANCE OF WORK: T. Van Haele, L. Moerman, N. Duvigneaud and D. Van Tiggelen.

INTERPRETATION OR ANALYSIS OF DATA: T. Van Haele, L. Moerman, N. Duvigneaud and D. Van Tiggelen.

PREPARATION OF THE MANUSCRIPT: T. Van Haele, L. Moerman, N. Duvigneaud and D. Van Tiggelen.

REVISION FOR IMPORTANT INTELLECTUAL CONTENT: L. Moerman, N. Duvigneaud and D. Van Tiggelen.

SUPERVISION: L. Moerman, N. Duvigneaud and D. Van Tiggelen.

Ethical considerations

All tests conformed to The Code of Ethics of the World Medical Association (Declaration of Helsinki). The ethical committee of University Hospital Center Brugmann approved this study (Reference CE 2018/ 93). All participants signed a written informed consent, prior to participation.

Funding

There was no financial support that could have influenced this study.

Footnotes

Acknowledgments

A word of thanks is directed to the Director of the Military Hospital Queen Astrid for providing the infrastructure and for supporting this work. In addition, the authors want to thank all participants for their willingness to participate in this study.

Conflict of interest

The authors declare no conflict of interest.

References

Bodin

Sérazin

Descatha

Leclerc

Goldberg

, et al. Effects of individual and work-related factors on incidence of shoulder pain in a large working population. J Occup Health. 2012; 54(4): 278-88.

Herin

Vezina

Thaon

Soulat

Paris

, group E. Predictors of chronic shoulder pain after 5 years in a working population. J Pain. 2012; 153(11): 2253-9.

Roquelaure

Leclerc

Touranchet

Sauteron

Melchior

, et al. Epidemiologic surveillance of upper-extremity musculoskeletal disorders in the working population. Arthritis Rheumatol. 2006; 55(5): 765-78.

Engebretsen

Grotle

Natvig

. Patterns of shoulder pain during a 14-year follow-up: Results from a longitudinal population study in Norway. J Shoulder Elb. Surg. 2015; 7(1): 49-59.

Svebak

Hagen

Zwart

J-A

. One-year prevalence of chronic musculoskeletal pain in a large adult norwegian county population: Relations with age and gender – the HUNT study. J Musculoskelet Pain. 2010; 14(1): 21-8.

Linsell

Dawson

Zondervan

Rose

Randall

Fitzpatrick

, et al. Prevalence and incidence of adults consulting for shoulder conditions in UK primary care; patterns of diagnosis and referral. Rheumatology (Oxford). 2006; 45(2): 215-21.

Reilingh

Kuijpers

Tanja-Harfterkamp

van der Windt

. Course and prognosis of shoulder symptoms in general practice. Rheumatology (Oxford). 2008; 47(5): 724-30.

Overbeek

Geurkink

de Groot

Klop

Nagels

Nelissen

, et al. Shoulder movement complexity in the aging shoulder: A cross-sectional analysis and reliability assessment. J Orthop Res. 2020.

Miller

Higgins

Dong

Collins

Bean

Seitz

, et al. Association of strength measurement with rotator cuff tear in patients with shoulder pain: The rotator cuff outcomes workgroup study. Am J Phys Med Rehabil. 2016; 95(1): 47-56.

10.

Dulgeroglu

Kirbiyik

Ersoz

Ozel

. Evaluation of shoulder rotational strength in patients with subacromial impingement syndrome using a computerized isokinetic dynamometer. J Musculoskelet Pain. 2013; 21(1): 23-30.

11.

Land

Gordon

Watt

. Isokinetic clinical assessment of rotator cuff strength in subacromial shoulder impingement. Musculoskelet Sci Pract. 2017; 27: 32-9.

12.

Forthomme

Dvir

Crielaard

Croisier

. Isokinetic assessment of the shoulder rotators: A study of optimal test position. Clin Physiol Funct Imaging. 2011; 31(3): 227-32.

13.

Sørensen

Oestergaard

Tulder

Petersen

. Measurement Properties of Isokinetic Dynamometry for Assessment of Shoulder Muscle Strength: A Systematic Review. Arch Phys Med Rehabil. 2020; 102.

14.

Tarara

Fogaca

Taylor

Hegedus

. Clinician-friendly physical performance tests in athletes part 3: A systematic review of measurement properties and correlations to injury for tests in the upper extremity. Br J Sports Med. 2016; 50(9): 545-51.

15.

Borms

Maenhout

Cools

. Incorporation of the kinetic chain into shoulder-elevation exercises: Does it affect scapular muscle activity? J Athl Train. 2020; 55(4): 343-9.

16.

Tucci

Martins

Sposito Gde

Camarini

de Oliveira

. Closed kinetic chain upper extremity stability test (CKCUES test): A reliability study in persons with and without shoulder impingement syndrome. BMC Musculoskelet Disord. 2014; 15: 1.

17.

Sciascia

Uhl

. Reliability of strength and performance testing measures and their ability to differentiate persons with and without shoulder symptoms. Int J Sports Phys Ther. 2015; 10(5): 655-66.

18.

Eraslan

Gelecek

Genc

. Effect of scapular muscle endurance on chronic shoulder pain in textile workers. J Back Musculoskelet Rehabil. 2013; 26(1): 25-31.

19.

Atta

Ata

Aneis

Diab

. Correlation between scapular muscle endurance and core muscle endurance in subject with chronic shoulder pain. J Adv Multidiscip Res. 2018; 5(10): 4157-4161.

20.

Lau

. 2017. Clinical assessment tests evaluating risk factors for shoulder injury in the kinetic chain of overhead athletes. Bachelor’s thesis. School of Physiotherapy. Amsterdam.

21.

Hegedus

Vidt

Tarara

. The best combination of physical performance and self-report measures to capture function in three patient groups. Phys Ther Rev. 2013; 19(3): 196-203.

22.

Terry

Thelen

Crowell

Goss

. The musculoskeletal readiness screening tool-athlete concern for injury & prior injury associated with future injury. Int J Sports Phys Ther. 2018; 13(4): 595-604.

23.

Hearn

Rhon

Goss

Thelen

. Evaluation of a novel field expedient musculoskeletal readiness screening tool in an army basic training population. Mil Med. 2017; 182(7): e1862-e8.

24.

Land

Gordon

. Clinical assessment of factors associated with subacromial shoulder impingement: A systematic review. Phys Ther Rev. 2017; 21(3-6): 192-206.

25.

Borms

Cools

. Upper-extremity functional performance tests: Reference values for overhead athletes. Int J Sports Med. 2018; 39(6): 433-441..

26.

Popchak

Poploski

Patterson-Lynch

Nigolian

Lin

. Reliability and validity of a return to sports testing battery for the shoulder. Phys Ther Sport. 2021; 48: 1-11.

27.

Borms

Maenhout

Cools

. Upper quadrant field tests and isokinetic upper limb strength in overhead athletes. J Athl Train. 2016; 51(10): 789-96.

28.

Hazar Kanik

Pala

Gunaydin

Sozlu

Alkan

Basar

, et al. Relationship between scapular muscle and core endurance in healthy subjects. J Back Musculoskelet Rehabil. 2017; 30(4): 811-7.

29.

Baecke

Burema

Frijters

. A short questionnaire for the measurement of habitual physical activity in epidemiological studies. Am J Clin Nutr. Am J Clin Nutr. 1982; 36(5): 936-42.

30.

Cools

Vanderstukken

Vereecken

Duprez

Heyman

Goethals

, et al. Eccentric and isometric shoulder rotator cuff strength testing using a hand-held dynamometer: reference values for overhead athletes. Knee Surg Sports Traumatol Arthrosc. 2016; 24(12): 3838-47.

31.

Sahrmann

Azevedo

Dillen

. Diagnosis and treatment of movement system impairment syndromes. Braz J Phys Ther. 2017; 21(6): 391-399.

32.

Andias

Neto

Silva

. The effects of pain neuroscience education and exercise on pain, muscle endurance, catastrophizing and anxiety in adolescents with chronic idiopathic neck pain: A school-based pilot, randomized and controlled study. Physiother Theory Pract. 2018; 34(9): 682-91.

33.

Edmondston

Wallumrød

MacLéid

Kvamme

Joebges

Brabham

. Reliability of isometric muscle endurance tests in subjects with postural neck pain. J Manipulative Physiol Ther. 2008; 31(5): 348-54.

34.

Ellenbecker

Manske

Davies

. Closed kinetic chain testing techniques of the upper extremities. Int J Sports Med. 2000; 9: 219-29.

35.

Goldbeck

Davies

. Test-retest reliability of the closed kinetic chain upper extremity stability test: A clinical field test. J Sport Rehabil. 2000; 9: 35-45.

36.

Taylor

Wright

Smoliga

DePew

Hegedus

. Upper-extremity physical-performance tests in college athletes. J Sport Rehabil. 2016; 25(2): 146-54.

37.

Tucci

Felicio

McQuade

Bevilaqua-Grossi

Camarini

Oliveira

. Biomechanical analysis of the closed kinetic chain upper-extremity stability test. J Sport Rehabil. 2017; 26(1): 42-50.

38.

Mayer

Schlumberger

Cingel

Henrotin

Laube

Schmidtbleicher

. Training and testing in open versus closed kinetic chain. Isokinet Exerc Sci. 2003; 11: 181-7.

39.

Gremeaux

Croisier

Forthomme

. Isokinetic intervention in microtraumatic shoulder instability: An update. J Sports Med Phys Fitness. 2012; 52(4): 413-23.

40.

Lategan

. Isokinetic norms for ankle, knee, shoulder and forearm muscles in young South African men. Isokinet Exerc Sci. 2011; 19(1): 23-32.

41.

Hinkle

Wiersma

Jurs

. Applied Statistics for the Behavioral Sciences. 6

{}^{\text{th}}

ed. Boston, Mass: Houghton Mifflin; 2003.

42.

Ozyurek

SBD

Genc

. Determining normative values for scapular muscle endurance in college-aged individuals. Med Sport. 2018; 71: 505-13.

43.

ZieliDski

Król-ZieljDska

Kusy

. Changes in physical activity of elite track and field athletes in selected age categories. Stud Phys Cult Tour. 2006; 13: 185-187.

44.

Sørensen

Oestergaard

van Tulder

Petersen

. Measurement properties of isokinetic dynamometry for assessment of shoulder muscle strength: A systematic review. J Arch Phys Med Rehabil. 2021 Mar; 102(3): 510-520.

45.

Forthomme

Croisier

Delvaux

Kaux

Crielaard

Gleizes-Cervera

. Preseason strength assessment of the rotator muscles and shoulder injury in handball players. J Athl Train. 2018; 53(2): 174-80.

46.

Eagle

Connaboy

Nindl

Allison

. Significantly increased odds of reporting previous shoulder injuries in female marines based on larger magnitude shoulder rotator bilateral strength differences. Orthop J Sports Med. 2018; 6(2).

47.

Forthomme

Wieczorek

Frisch

Crielaard

Croisier

. Shoulder pain among high-level volleyball players and preseason features. Med Sci Sports Exerc. 2013; 45(10): 1852-60.

48.

Lee

Kim

. Reliability and validity of the closed kinetic chain upper extremity stability test. J Phys Ther Sci. 2015; 27(4): 1071-3.

49.

Dvir,

. Relevant, less relevant and irrelevant isokinetic strength test parameters: Some critical comments. Mov Sport Sci. Mot, 2014; 85(3): 15-21.

Upper extremity field tests and isokinetic shoulder internal and external strength and power in sports active males

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1. Introduction

Table 1 Anthropometric data of the subjects ( N = 76) all displayed in (Median and (Interquartile range))

2.1 Participants

2.2 Procedure

Field tests

3. Results

Table 3 Reference values of isokinetic peak moment for internal and external rotation of the shoulder in Mean ± SD (95% Confidence interval) for the PM/kg (Nm/kg)

4.1 Reference values for the field and isokinetic tests

Table 6 Correlation coefficient ( r s ) between the CKCUEST and criterion isokinetic test for the normalized strength and Power based on the Spearman rank-order correlation coefficient

5. Conclusion

Author contributions

Ethical considerations

Funding

Footnotes

Acknowledgments

Conflict of interest

References

Table 1
Anthropometric data of the subjects ( $N=$ 76) all displayed in (Median and (Interquartile range))

Table 3
Reference values of isokinetic peak moment for internal and external rotation of the shoulder in Mean $\pm$ SD (95% Confidence interval) for the PM/kg (Nm/kg)

Table 6
Correlation coefficient ( $r_{s}$ ) between the CKCUEST and criterion isokinetic test for the normalized strength and Power based on the Spearman rank-order correlation coefficient