TOTAL ARC OF MOTION IN THE SIDELYING POSITION: EVIDENCE FOR A NEW METHOD TO ASSESS GLENOHUMERAL INTERNAL ROTATION DEFICIT IN OVERHEAD ATHLETES

Abstract

Background

Total arc of motion (TA) measured in a supine position has been utilized as a method to detect the presence of glenohumeral internal rotation deficit (GIRD) in overhead athletes. A component of supine TA is supine internal rotation (IR) range of motion (ROM), which has many variables including the amount and location of manual stabilization. A sidelying position for gathering IR ROM has recently been proposed and, when combined with supine external rotation (ER) ROM, constitutes a new method of quantifying TA. This new sidelying TA method, however, has no normative values for overhead athletes.

Purpose

The purposes of this study were to develop normative values for sidelying TA in overhead athletes, determine any ROM difference between supine and sidelying TA, and examine side‐to‐side differences within the two TA methods. A secondary purpose of the study was to examine for any effect of gender or level of competition on the two TA methods.

Study Design

Cross‐sectional study.

Methods

Passive supine IR ROM, supine ER ROM, and sidelying IR ROM were gathered on bilateral shoulders of 176 collegiate and recreational overhead athletes (122 male [21.4 ± 4.7 years, 71.7 ± 2.7 inches, 25.3 ± 2.7 BMI] and 54 female [21.4 ± 5.4 years, 67.6 ± 3.0 inches, 22.5 ± 2.37 BMI]).

Results

Sidelying TA mean for the dominant shoulder was 159.6 °±15.0 °; the non‐dominant shoulder was 163.3 °±15.3 °. Sidelying TA for both shoulders (p < 0.0001) was 14 ° less than supine TA. Both TA methods exhibited a 4 ° dominant‐shoulder deficit (p < 0.0001). For the dominant and non‐dominant shoulder, respectively, there was no gender (p = 0.38, 0.54) or level of competition (p = 0.23, 0.39) effect on sidelying TA.

Conclusion

In overhead athletes, sidelying TA is a viable alternative to supine TA when examining for the presence of GIRD. Gender and level of competition does not significantly affect sidelying TA, so the mean of 160 ° on the dominant shoulder and 163 ° on the non‐dominant shoulder can be used by clinicians.

Level of Evidence

Level 3

INTRODUCTION

Athletes who participate in overhead sports often develop a chronic adaptation in the available range of motion (ROM) of the dominant shoulder, resulting from repetitive exposure to these overhead activities.^1‐7 This adaptation typically involves a loss of dominant shoulder internal rotation (IR) ROM, commonly known as glenohumeral internal rotation deficit (GIRD). The presence of GIRD is often balanced with a concomitant increase in external rotation (ER) ROM of the same shoulder.^1‐5

Manske¹ has described two types of GIRD. First, anatomical GIRD is a necessary adaptation of the dominant shoulder, resulting from repetitive exposure to overhead activities in which there is a loss of IR ROM, along with an equivalent increase in ER ROM. The total arc of motion (TA), however, which is defined as the sum of IR ROM and ER ROM, remains relatively symmetrical when compared to the non‐dominant shoulder. This side‐to‐side TA symmetry has been observed in several studies of overhead athletes, ranging from youth to professional levels.^2‐7 The shift in dominant shoulder rotational motion with a TA that is equal to the non‐dominant shoulder is considered a normal osseous adaptation and has been attributed to an increase in humeral retroversion (HRV) seen in the dominant shoulder of overhead athletes.^6‐8 The increase in HRV allows greater shoulder ER ROM during the cocking phase in overhead sports, such as baseball, softball, volleyball, and tennis, assisting the shoulder in reaching its extremely high rotational velocities during the acceleration phase of these sports.^9‐10 Increased HRV has also been purported to be a protective mechanism to minimize anterior shoulder laxity in overhead athletes,^7,8,11,12 while also minimizing the risk of developing internal impingement.^12‐16

Conversely, pathologic GIRD is present when a greater amount of IR ROM loss than ER ROM gain on the dominant shoulder exists, resulting in a TA deficit on the dominant shoulder compared to the non‐dominant shoulder.¹ This loss of TA or greater loss of IR ROM on the dominant shoulder has been associated with higher risk of injury in the throwing shoulder and elbow of overhead athletes.^9,17‐21 In studies by Wilk et al, baseball pitchers who demonstrated a TA deficit of 5 ° or more on their throwing shoulder were 2.5 times more likely to sustain a shoulder injury²⁰ and 2.6 times more likely to sustain an elbow injury²¹ compared to pitchers without a TA deficit of their throwing shoulder. It has been suggested that overhead athletes who have a TA deficit of 5 ° or more on the dominant shoulder likely present with posterior shoulder soft tissue restrictions, namely posterior glenohumeral joint capsule thickening^13,22‐24 or increased posterior rotator cuff muscle stiffness,^1,25‐28 with these adaptations occurring secondary to chronic exposure to overhead sports.

While there is agreement that GIRD, whether anatomical or pathologic, occurs in overhead athletes, the techniques used to assess this deficit are not universally agreed upon. The supine position with the arm abducted to 90 ° has traditionally been utilized to assess IR and ER ROM, as well as TA.^{6‐8,11,29‐33} Authors have noted, however, a lack of consistent stabilization of the scapula may be present when utilizing the supine position to assess the glenohumeral contribution to shoulder IR ROM.^3,5,9,34,35 In the supine position, a posteriorly‐directed manual stabilization force is typically applied to the anterior acromion and coracoid process, while simultaneously allowing humeral head rotation during IR ROM assessment. The amount of stabilization force provided is subjective in nature, and may vary between examiners. This lack of consistent stabilization has been implicated in the lower intra‐³⁵ and inter‐rater reliability^34,35 noted for the supine IR ROM position.

Attempts to evaluate supine IR ROM utilizing various methods of stabilization have been undertaken in recent years. Wilk et al³⁶ examined the reliability and IR ROM of three supine IR ROM methods that differed in the location and application of stabilization. It was determined that the method that included stabilization of the scapula to minimize its accessory motion of anterior tilting produced the greatest intra‐rater reliability of 0.62, but also the lowest inter‐rater reliability of 0.43. These low reliability values for the supine method with scapular stabilization lend further evidence to its potential drawback as a reliable clinical measure of IR ROM.

Moreover, numerous studies have investigated IR ROM differences in the supine position between the dominant and non‐dominant shoulders in overhead athletes. These results indicate that significant differences exist between sides, with the dominant shoulder having less IR ROM than the non‐dominant side.^{4,5,7,24,28,37‐41} Based upon this difference, previous authors suggest not using the non‐dominant shoulder for comparison when assessing IR ROM loss of the dominant shoulder.^42‐46

Researchers have recently examined the sidelying position as an alternative for measuring IR ROM.^27,35,41 In healthy subjects with non‐impaired shoulders, as well as those with shoulder pathology, Lunden and colleagues³⁵ found the full sidelying position to have greater intra‐ and inter‐rater reliability when assessing shoulder IR ROM, as compared to the supine position. In a small sample of Division I baseball and softball athletes, Kevern and associates²⁷ found that the sidelying position had the highest intra‐ and inter‐rater reliability for measuring IR ROM, as compared to a supine position both with and without scapular stabilization. It has been purported that the sidelying position allows the scapula to be stabilized by the individual's own body weight,^26,27,35 thus IR ROM measurements taken in this position are likely not influenced by variations in stabilization force provided by the examiner. Also, in contrast to the supine position, the sidelying position may limit the scapular accessory motion of anterior tilting that occurs during IR, leading to a more distinct capsular end‐feel³⁵ and likely isolating IR ROM to the glenohumeral joint.^27,35

While Kevern²⁷ defined sidelying TA as the sum of sidelying IR and sidelying ER, it may be more appropriate to consider a hybrid definition of sidelying TA. This hybrid sidelying TA is defined as the sum of sidelying IR and supine ER and may provide a more functional assessment of the IR and ER motion extremes that occur in overhead sports. Specifically, the assessment of IR ROM in the sidelying position places the shoulder in a combined position of flexion and horizontal adduction, positions that the shoulder assumes during the arm deceleration phase of throwing,^47,48 where a lack of IR ROM may be problematic. Dillman et al⁴⁷ noted that maximal IR ROM occurs at the end of the deceleration phase and is accompanied by horizontal adduction of the shoulder. The amount of horizontal adduction in this phase is more closely replicated in the sidelying IR position, as compared to the supine IR position. The assessment of ER ROM in the supine position, with the shoulder abducted to 90 º, mimics the cocking phase of many overhead sports where maximal ER ROM occurs.

Since GIRD is present in overhead athletes and research exists supporting the use of the sidelying position to assess IR ROM, a new method of quantifying total arc of motion, one that utilizes the sidelying position to assess IR ROM along with supine ER ROM, could be used to further differentiate between anatomical and pathologic GIRD in overhead athletes. Therefore, the primary purposes of this study of collegiate and recreational overhead athletes were to: 1) establish both dominant and non‐dominant shoulder normative values for sidelying TA; 2) compare sidelying TA to the traditional supine TA for both the dominant and non‐dominant shoulder; and 3) compare side‐to‐side differences for both supine and sidelying TA. A secondary purpose of this study was to determine if gender and level of competition affect the amount of sidelying TA. The hypotheses of this study were: 1) sidelying TA would be significantly smaller than supine TA for both shoulders; and 2) there would be no significant side‐to‐side difference for either supine or sidelying TA.

METHODS

All testing procedures were completed in a single session lasting approximately 30 minutes, except for the reliability portion of this study, which took place over two testing sessions on consecutive days. A cross‐sectional study design was employed in order to calculate mean values for supine and sidelying TA. Independent variables in the study included gender, level of competition in overhead sports, and upper extremity dominance, which was defined as the arm used by the participant to throw a ball. The dependent variables were supine and sidelying TA. This study was approved by the Institutional Review Board of St. Catherine University. Participants were recruited through a sample of convenience and were asked to review and sign a written consent and HIPAA authorization form prior to participating in the study.

Participants

Participants were included in this study if they were: 1) 18 years or age or older; 2) a current Division I or Division III collegiate athlete or a current recreational athlete participating in the overhead sports of baseball, softball, volleyball, or javelin throwing; and 3) symptom‐free in at least one shoulder. Participants were excluded from this study if they: 1) had a history of shoulder surgery, fracture, or dislocation; 2) had pain that limited ROM in both shoulders; 3) were currently participating in a medically‐supervised shoulder rehabilitation program; or 4) were unable to lie on their back or side. Participants were identified as overhead athletes based on their self‐report and also from information gathered on a demographics form that inquired about history of participation in overhead sports. In order to be enrolled in the study, recreational athletes had to have a history of competitive overhead sport participation prior to age 18 and also have participated in overhead sports at least twice per week for the previous five years prior to participation in the study. After inclusion and exclusion criteria were determined, 176 participants were enrolled in the study (Table 1), including 33 Division I, 81 Division III, and 62 recreational overhead athletes. Two athletes had a painful non‐dominant shoulder at the time of data collection, so only dominant shoulder data was collected and utilized for those participants. Therefore, this study included data from 176 dominant and 174 non‐dominant shoulders.

Table 1.

Characteristics of study participants (mean ± standard deviation)

	Age (years)	Age range (years)	Height (inches)	Weight (pounds)	BMI
Males (n=122)	21.4 ± 4.7	18‐47	71.7 ± 2.7	184.9 ± 23.1	25.3 ± 2.7
Females (n=54)	21.4 ± 5.4	18‐56	67.6 ± 3.0	145.9 ± 18.1	22.5 ± 2.3

BMI = Body mass index.

Raters

All ROM measurements were taken by five Doctor of Physical Therapy (DPT) student investigators from St. Catherine University or their faculty research advisor. In order to ensure consistency among raters, the student investigators were instructed in the correct technique and end‐feel assessment of shoulder IR ROM in supine and sidelying and shoulder ER ROM in supine. Instruction and training was provided by the faculty research advisor, a clinician with 25 years of physical therapy experience, who also had significant expertise in sidelying IR ROM assessment.

Instrumentation

Measurements of ROM were obtained using a bubble inclinometer (Johnson Level & Tool Manufacturing Company Inc. Mequon, WI) that was placed on the participant's distal forearm just proximal to the wrist, on the dorsal aspect for all IR ROM measurements (Figure 1) and on the ventral aspect for ER ROM measurements.

Figure 1.

Placement of bubble inclinometer for all IR ROM measurements.

Procedures

After providing written informed consent, each subject was asked to complete a brief questionnaire that included demographic information of age, gender, self‐reported height and weight, history of participation and current level of competition in overhead sports, as well as previous shoulder injuries or surgeries. Arm dominance was determined by asking the subject which arm they would use to throw a ball.

All ROM measurements were taken on a standard treatment table. Supine IR ROM was gathered with the subject in a supine position. Investigators passively abducted the humerus to 90 °, with the elbow flexed to 90 º and the forearm was maintained in a neutral rotation position. A small towel roll was placed under the distal humerus to limit any horizontal abduction of the humerus. Investigators used their thenar mass to apply a posteriorly‐directed stabilization force to the acromion and coracoid processes, thus limiting anterior tilting of the scapula; care was taken to avoid any contact over the humeral head. Investigators passively moved the shoulder into IR (Figure 2) and when a firm capsular end‐feel was noted or the scapula began to anteriorly tilt, the IR ROM measurement was obtained by placing the bubble inclinometer just proximal to the wrist joint on the dorsal forearm.

Figure 2.

Supine IR ROM position.

For sidelying IR ROM measurements, participants were positioned lying on the side to be measured and the humerus was passively flexed to 90 º to place it in a position of horizontal adduction. The elbow was placed in 90 º of flexion and the forearm was maintained in a neutral rotation position. In order to standardize the sidelying position, investigators visually aligned the bilateral acromion processes in a vertical plane (Figure 3), using methods described by Lunden et al.³⁵ The scapula was then protracted slightly by the investigator pulling gently on the distal humerus, stopping when the vertical alignment of the acromion processes first began to be altered. The pelvis not contacting the table was aligned in a neutral position (neither protracted or retracted) relative to the opposite side pelvis. Investigators passively moved the shoulder into IR (Figure 3) and when a firm capsular end‐feel was noted that stopped shoulder motion, the IR ROM measurement was obtained by placing the bubble inclinometer just proximal to the wrist joint on the dorsal forearm. For the sidelying position there was no manual stabilization force applied by the investigators to the scapula.

Figure 3.

Sidelying IR ROM position.

For supine ER ROM the humerus was positioned similar to the supine IR ROM position, although no scapular stabilization was provided over the anterior shoulder. The shoulder was passively moved into ER and upon achieving a capsular end‐feel the measurement was obtained by placing the bubble inclinometer just proximal to the wrist joint on the ventral aspect of the forearm.

Prior to data collection, the order of the two IR ROM positions was randomized to minimize any possible order effect on the IR ROM values. Measurements in these two IR ROM positions were performed first, followed lastly by ER ROM. Range of motion was assessed first on the dominant shoulder and then on the non‐dominant shoulder, using the same order of positions for both shoulders. Two measurements were obtained for each ROM position and the mean was used for subsequent data analysis. The ROM values were not blinded to the investigator, as they recorded the values onto a data form for each subject. Supine TA was operationally defined in this study as the sum of supine IR ROM and supine ER ROM, while sidelying TA was defined as the sum of sidelying IR ROM and supine ER ROM.

In order to pool data from all investigators for analysis, intra‐rater and inter‐rater reliability were determined using 18 subjects. This took place prior to any data collection for the main part of the study. Measurements of supine IR ROM, sidelying IR ROM, and supine ER ROM were collected as previously described on two consecutive days, with two measurements taken each day for all three ROM positions. Both the order of raters and the three measurement positions were randomized for each subject. All measurements for the reliability portion of this study were read and recorded by a student physical therapist who was not a part of the study, therefore allowing the raters to be blinded to the values of the measurements. The mean of these measurements was used for subsequent reliability determination.

Statistical Analysis

Using the 18 subjects recruited for the reliability portion of this study, intraclass correlation coefficients (ICCs) were calculated to quantify intra‐rater reliability (ICC_1,1) for each of the six raters for the supine IR, sidelying IR, and supine ER ROM positions. Intraclass correlation coefficient values were also used to calculate inter‐rater reliability (ICC_2,k) using the mean of each rater's two measurements taken on each subject in the three ROM positions. The faculty research advisor was used as the reference standard for all inter‐rater reliability calculations. Intraclass correlation coefficient values were classified as: excellent (0.90‐0.99); good (0.80‐0.89); fair (0.70‐0.79); and poor (<0.69).⁴⁹ For both intra‐rater and inter‐rater reliability, 95% confidence intervals (CI) were also calculated.

The demographic variables of age, height, weight, and body mass index (BMI) were gathered for each subject and summary descriptive statistics (mean, standard deviation, range) were determined for each of these variables for the entire sample of subjects.

Means and standard deviations were determined for supine and sidelying TA on both the dominant and non‐dominant shoulder. Within both supine and sidelying TA, paired t‐tests were performed to compare the means of the dominant and non‐dominant shoulder. Within both the dominant and non‐dominant shoulders, paired t‐tests were also utilized to compare the means of supine and sidelying TA. In addition, independent t‐tests were used to examine the means of each gender within the supine and sidelying TA on both the dominant and non‐dominant shoulder.

Repeated measures analyses of variance (ANOVAs) were calculated to examine the differences in means of supine and sidelying TA for both the dominant and non‐dominant shoulder across the three levels of competition (Division I, Division III, recreational athletes). For ANOVAs that were significant, a post‐hoc analysis was performed using paired t‐tests, with Bonferroni multiple comparisons correction applied to these t‐tests.

The standard error of measurement (SEM) was also calculated for supine and sidelying TA on the dominant and non‐dominant shoulder. The level of significance was set a priori at p ≤ 0.05 for all statistical analyses performed in this investigation. Number Cruncher Statistical Software (NCSS) was used for all statistical analyses.

RESULTS

The results for intra‐rater and inter‐rater reliability, including 95% confidence intervals, appear in Tables 2 and 3, respectively. Intra‐rater ICC values across the six raters for the three ROM positions that comprise supine and sidelying TA ranged from fair to excellent.⁴⁹ The sidelying IR ROM position had the greatest intra‐rater reliability with a range of values from 0.87‐0.97 across the raters. Inter‐rater ICC values for the three ROM positions ranged from good to excellent,⁴⁹ with the sidelying IR position having the highest inter‐rater reliability value of 0.91.

Table 2.

Intra‐rater reliability range (95% CI) of the 6 raters for the three ROM positions that comprise supine and sidelying TA using ICC (model [1,1])

	Supine IR	Sidelying IR	Supine ER
Intra‐rater reliability	0.78‐0.92	0.87‐0.97	0.79‐0.95
	(0.70‐0.96)*	(0.79‐0.99)*	(0.75‐0.99)*

IR=Internal rotation; ER = External rotation

CI = confidence interval, * significant at p < 0.001

Table 3.

Inter‐rater reliability (95% CI) of the three ROM positions that comprise supine and sidelying TA using ICC (model [2,k]).

	Supine IR	Sidelying IR	Supine ER
Inter‐rater reliability	0.84 (.74‐.91)*	0.91 (.83‐.96)*	0.84 (.80‐.90)*

IR=Internal rotation; ER=External rotation

CI = confidence interval, * significant at p < 0.001

Mean, standard deviation, and standard error of measurement for supine and sidelying TA of the dominant and non‐dominant shoulders are noted in Table 4. The amount of supine TA of the dominant shoulder was 174.0 º (±17.1 °), while the non‐dominant shoulder was 177.8 º (±17.3 °). The amount of sidelying TA of the dominant shoulder was 159.6 º (±15.0 °), while the non‐dominant shoulder was 163.3 º (±15.3 °). As depicted in Table 4, paired t‐tests determined that the 3.8 º and 3.7 º loss in motion on the dominant shoulder for supine and sidelying TA, respectively, was statistically significant (p < 0.0001). When comparing the difference between the supine and sidelying TA values, results of paired t‐tests revealed a significant difference of 14.4 º on the dominant shoulder and 14.5 º on the non‐dominant shoulder, with sidelying TA having less motion than supine TA for each shoulder (Table 5).

Table 4.

Supine and sidelying total arc of motion (mean ± standard deviation).

	Dominant shoulder (° ± SD)	Non‐dominant shoulder (° ±SD)	Mean Difference (°)	p	Dominant shoulder SEM (°)	Non‐Dominant shoulder SEM (°)
Supine TA	174.0 ± 17.1	177.8 ± 17.3	3.8	<0.0001*	4.2	4.2
Sidelying TA	159.6 ± 15.0	163.3 ± 15.3	3.7	<0.0001*	3.7	3.8

^*significant at p < 0.05; SEM = Standard Error of Measure; TA=Total arc of motion

Table 5.

Difference in total arc of motion positions by shoulder (mean ± standard deviation).

	Dominant shoulder (° ± SD)	Non‐dominant shoulder (° ± SD)
Supine TA	174.0 ± 17.1	177.8 ± 17.3
Sidelying TA	159.6 ± 15.0	163.3 ± 15.3
TA Difference	14.4° *	14.5 ° †

^*significant at p < 0.0001;

^†significant at p < 0.0001; TA = Total arc of motion

Means and standard deviations for supine and sidelying TA of the dominant and non‐dominant shoulders by gender are noted in Tables 6 and 7, respectively. When examining for possible gender differences on the dominant shoulder, independent t‐tests revealed a significantly smaller supine TA of 7.9 ° for males compared to females (171.4 ° vs. 179.3 °, p = 0.004). Males also had a smaller sidelying TA of 2.2 ° on the dominant shoulder compared to females (158.7 ° vs. 160.9 °), however the difference was not significant (p = 0.38). For the non‐dominant shoulder, independent t‐tests revealed a significantly smaller supine TA of 6.0 ° for males compared to females (173.5 ° vs. 179.5 °, p = 0.01). Males also had a smaller sidelying TA of 1.5 ° compared to females (162.4 ° vs. 163.9 °), but again, this difference was not significant (p = 0.54).

Table 6.

Dominant shoulder gender differences for supine and sidelying total arc of motion (mean ± standard deviation).

	Male (° ± SD)	Female (° ± SD)	Difference (°)	p
Supine TA	171.4 ± 17.4	179.3 ± 14.6	7.9	0.004*
Sidelying TA	158.7 ± 15.9	160.9 ± 12.9	2.2	0.38

^*significant at p < 0.05; TA = Total arc of motion

Table 7.

Non‐dominant shoulder gender differences for supine and sidelying total arc of motion (mean ± standard deviation).

	Male (° ± SD)	Female (° ± SD)	Difference (°)	p
Supine TA	173.5 ± 13.8	179.5 ± 13.5	6.0	0.01*
Sidelying TA	162.4 ± 12.8	163.9 ± 12.6	1.5	0.54

^*significant at p < 0.05; TA = Total arc of motion

Results of repeated measures ANOVAs for supine and sidelying TA on the dominant shoulder across the three levels of competition are depicted in Figure 4. Supine TA mean values ranged from 170.5 °‐174.8 °, while sidelying TA values were smaller and ranged from 155.4 °‐159.1 °. Recreational athletes demonstrated the smallest mean for both TA positions on the dominant shoulder. There was no significant difference, however, across the levels of competition for either the supine TA (F_2,175 = 0.77; p = 0.47) or sidelying TA (F_2,175 = 1.5; p = 0.23) for the dominant shoulder. Figure 5 depicts the results of repeated measures ANOVAs for supine and sidelying TA on the non‐dominant shoulder across the three levels of competition. Similar to the dominant shoulder, there was no significant difference across the levels of competition for either supine TA (F_2,175 = 2.47; p = 0.09) or sidelying TA (F_2,175 = 0.95; p = 0.39) for the non‐dominant shoulder. Supine TA mean values ranged from 172.0 °‐180.3 °, while sidelying TA values ranged from 160.5 °‐164.5 °. Recreational athletes also had the smallest mean value for both TA positions on the non‐dominant shoulder.

Figure 4.

Dominant shoulder total arc of motion (TA) by level of competition (mean ± standard deviation).

Figure 5.

Non‐dominant shoulder total arc of motion (TA) by level of competition (mean ± standard deviation).

DISCUSSION

A primary aim of this study was to establish normative values for sidelying TA in overhead athletes using a novel technique to examine for GIRD. Results from the 176 participants revealed that the normative sidelying TA was 160 ° and 163 ° for the dominant and non‐dominant shoulder, respectively. These values were lower than those obtained for supine TA, which were 174 ° for the dominant and 178 ° for the non‐dominant shoulder. These supine TA values are similar to those reported by previous authors who utilized this same position, with typical values ranging between 167 ° to 200 °.^{2,6,7,20,21,25,29,31‐33,50} Supine TA values in the present study are likely on the lower end of this range due to the care taken by the examiners to stabilize the scapula, thus limiting the amount of anterior tilting and effectively isolating IR ROM to the glenohumeral joint.

The result for the dominant shoulder sidelying TA of 160 ° is in relative agreement with Kevern,²⁷ who utilized two raters to examine the reliability of three methods to determine TA on the dominant shoulder of 38 Division I collegiate baseball and softball players. These raters determined that the sidelying TA of the dominant shoulder was approximately 160 ° and 147 °, respectively. While Kevern²⁷ obtained sidelying IR ROM using the same position as the current study, they gathered ER ROM in the sidelying position. In contrast, the current study utilized a supine position to gather ER ROM, which was added to the sidelying IR ROM value for determination of sidelying TA. It appears that supine ER ROM may be greater than when measuring it in a sidelying position. Using the supine position to measure ER ROM places the shoulder in the same functional position where maximal ER ROM is achieved during the cocking phase that occurs in overhead sports such as baseball, softball, tennis serving, and volleyball spiking. It is the opinion of the current authors that gathering the two motions that comprised sidelying TA in this study, sidelying IR ROM and supine ER ROM, may provide a viable alternative when assessing the functional total arc of motion in overhead athletes.

To the authors’ knowledge, the effect of gender on TA in overhead athletes has yet to be examined. Results of this study indicate that there was a gender effect for supine TA, with females having 8 ° more motion on their dominant shoulder than their male counterparts. Further examination of the component motions of supine TA revealed that females had a non‐significant 2 ° loss of supine ER ROM (females = 118.1 °; males = 119.9 °) and a significant 10 ° increase in supine IR ROM (females = 61.2 °; males = 51.6 °). It appears that the difference in supine IR ROM between genders was primarily responsible for the significant gender difference in supine TA. This finding agrees with previous researchers who have noted increased supine IR ROM in females compared to males.^51‐53 In contrast to supine TA, there was no difference between genders for sidelying TA of the dominant shoulder, including no difference in the component motions of sidelying IR ROM and supine ER ROM. Our findings concur with those of Carcia and colleagues⁴¹ who found no gender difference for sidelying IR ROM when examining both the dominant and non‐dominant shoulder in a population of asymptomatic, collegiate, non‐athletes. Due to no gender difference being observed for sidelying TA, rehabilitation professionals can have confidence in using the overall normative value of 160 ° for the dominant shoulder and 163 ° for the non‐dominant shoulder of collegiate and recreational overhead athletes. These normative values, however, are not applicable to professional or youth athletes, as they were not represented in this sample of participants.

This study is also the first to examine the possible influence of level of competition on TA. There was no level of completion difference between Division I, Division III, or recreational overhead athletes for either supine or sidelying TA. This finding was noted for both the dominant and non‐dominant shoulder. As in the gender discussion above, these findings lend further evidence that the sidelying TA normative value of 160 ° for the dominant shoulder and 163 ° for the non‐dominant shoulder for can be used across Division I, Division III, and recreational overhead athletes. Again, caution is warranted when applying these norms to youth or professional athletes, as they were not examined in this study. Since this is the first study to examine sidelying TA using sidelying IR ROM and supine ER ROM, additional studies are warranted to further establish normative values for this new TA method.

The participants in this investigation had no difficulty assuming the sidelying position and did not complain of pain as the shoulder was internally rotated to its end‐feel. This observation agrees with Lunden et al³⁵, who found that less than 1% of their subjects, including both asymptomatic and those with various shoulder pathologies, were unable to assume a sidelying positon for IR ROM assessment. A possible reason for participants in the current study being able to assume a sidelying position and undergo pain‐free IR ROM is that the scapula was allowed to protract prior to undertaking IR ROM assessment. In effect, the positioning of the scapula into protraction allowed it to assume its scapular plane orientation. This positioning of the scapula is similar to a proposed sleeper stretch modification in which the trunk is allowed to rotate 20 °‐30 ° posterior to the coronal plane of the body, thus placing the humerus in a scapular plane orientation and minimizing the risk of producing subacromial impingement during IR.⁵⁴ In addition, during overhead sports the scapula protracts on the rib cage during the arm deceleration phase, during which time the shoulder is undergoing rapid internal rotation and horizontal adduction relative to the thorax.^47‐48 The current authors feel it is important to assess sidelying IR ROM with the scapula in some protraction, simulating its functional position during the deceleration phase of overhead sports.

A significant 14 ° difference was noted on both the dominant and non‐dominant shoulder when comparing supine and sidelying TA, with sidelying TA having less motion. The investigation by Lunden³⁵ is the only previous study that examined sidelying TA using supine ER ROM. Results of their study revealed a similar difference between supine and sidelying TA, 12 ° and 18 °, respectively, for the two raters. Since supine ER ROM was used in calculating both supine and sidelying TA values and was therefore a constant, the 14 ° difference in TA in the current study can be directly attributed to the difference in IR ROM between the two positions, with sidelying IR having less motion that supine IR (dominant shoulder IR ROM: supine = 54.5 °; sidelying = 40.0 °; non‐dominant shoulder IR ROM: supine = 67.7 °; sidelying = 53.2 °). The dominant shoulder sidelying IR ROM value of 40 ° is the same as that reported by Lunden³⁵ for their asymptomatic participants in the sidelying position, but is smaller than the 48 ° noted by Carcia,⁴¹ whose participants did not have a history of overhead athletic activity. This finding indicates that repetitive exposure to overhead sports may lead to a loss of IR ROM on the dominant shoulder. Interestingly, the current study's non‐dominant shoulder IR ROM value of 53 ° is the same as that reported by Carcia⁴¹ for their non‐dominant shoulder, revealing that the non‐dominant shoulder is likely to have similar IR ROM regardless of participation in overhead sports.

The smaller amount of IR ROM in sidelying is likely due to the tension placed upon the posterior shoulder soft tissues structures, such as the glenohumeral joint capsule and posterior shoulder musculature. When IR ROM was assessed in a supine position, the shoulder was elevated to 90 °. In sidelying, the shoulder remained elevated to 90 ° but instead assumed a position of flexion/horizontal adduction. A position of horizontal adduction has been demonstrated to increase the strain on the posterior shoulder soft tissue structures when the shoulder undergoes IR,⁵⁵ leading to a reduction in the available IR ROM in this horizontal adduction position.^27,35 In addition, it is the experience of the investigators in the current study that the sidelying position yields a dramatically more distinct, capsular end‐feel when assessing IR ROM, as compared to the supine position. This observation may be the result of the increased posterior shoulder soft tissue strain during IR in the sidelying position. Lastly, it has been suggested that the sidelying position limits the amount of scapular anterior tilting that typically accompanies shoulder IR ROM, thus isolating IR ROM in the sidelying position to that coming from the glenohumeral joint.^26,27,35 In addition, the factors of increased posterior shoulder soft tissue strain and limited anterior tilting of the scapula in the sidelying position may account for the increased reliability of sidelying IR ROM that was noted in this study, as well as by others.^27,35

Sidelying TA may be a viable alternative to assess GIRD in overhead athletes, as opposed to the traditional supine method. The primary osseous factor in GIRD is humeral retroversion and due to its bony nature, the impact of HRV on IR ROM likely does not depend on the position in which this motion is gathered. In contrast, the posterior shoulder soft tissue components of GIRD have significantly greater strain in a sidelying position compared to supine when the shoulder is undergoing IR.⁵⁵ Therefore, sidelying TA may be a viable method to ascertain the soft tissue component of GIRD in overhead athletes.

When compared to the non‐dominant shoulder, the dominant shoulder had a 4 º deficit for both supine and sidelying TA. While the finding of this sidelying TA deficit is novel, the dominant‐shoulder deficit for supine TA is similar to those found by previous authors.^6,7 The supine TA deficit in the current study fell within the 5 ° threshold suggested by Manske et al¹ for an indication of anatomical GIRD, which is likely mediated by a difference in HRV between the dominant and non‐dominant shoulder. Approximately 40% of our participants, however, exceeded this 5 ° dominant–shoulder deficit threshold for both supine and sidelying TA, and thus may be at risk for pathologic GIRD.¹ While Wilk et al^20,21 has demonstrated an association between a supine TA deficit of 5 ° or more on the dominant side and increased risk of shoulder and elbow injuries in overhead athletes, similar studies need to be undertaken utilizing sidelying TA to determine the dominant‐shoulder deficit that increases the risk of upper extremity injuries in overhead athletes.

The SEM for sidelying TA was relatively low at 3.7 ° and 3.8 ° for the dominant and non‐dominant shoulder, respectively. This dominant‐shoulder finding is similar to that found by Kevern²⁷ and indicates that a true side‐to‐side difference was present for sidelying TA, as the difference between sidelying TA means of the dominant and non‐dominant shoulder was 3.7 ° (Table 4). In addition, supine TA had a slightly higher SEM of 4.2 ° for both shoulders. Sidelying TA, therefore, had slightly less error associated with IR ROM measurements taken in the sidelying versus supine position, as the supine ER position was common to both methods of TA determination. The lower SEM for sidelying TA is also likely linked to the higher intra‐ and inter‐rater reliability values found for the sidelying IR position. Also, the SEM for supine TA was slightly greater than the dominant‐shoulder deficit and, while statistically significant, this deficit may not indicate clinical significance using supine TA. In clinical practice it is important to use measurement positions that have small SEM and are therefore sufficiently accurate to evaluate for motion deficits in the shoulders of overhead athletes. Total arc of motion deficits of as little as 5 °, for example, have been linked to upper extremity injures in overhead athletes.^20,21

This investigation has some limitations. First, the mean values for supine and sidelying TA reported in this study were gathered on asymptomatic adults who were collegiate or recreational overhead athletes. The finding of this study, therefore, cannot be generalized to populations of varying ages, level of competition, or to those who are experiencing current shoulder pathologies. Specifically, no data was gathered on youth or professional‐level overhead athletes and the authors suggest additional supine and sidelying TA data be gathered for these groups. Second, multiple raters, including current DPT students, were utilized to collect data. While the authors acknowledge this potential limitation, care was taken to establish both intra‐ and inter‐rater reliability for all raters prior to any supine or sidelying TA data collection. The reliability results have confirmed those of previous authors who have investigated the sidelying position.^27,35 Third, variations in the amount of scapular tilting in the resting position were not controlled for in this study and, therefore, they may have had an impact on the amount of IR ROM obtained in the supine and sidelying positions. Future studies should examine the effect of variations in the resting or static position of the scapula on IR ROM, using three‐dimensional motion analysis equipment. Lastly, the authors are aware that athletes with current shoulder pain or those who may be recovering from a surgical procedure may not be able to assume the sidelying position or have pain‐free IR ROM during its assessment in this position.

CONCLUSION

This investigation examined the differences between traditional supine TA and a newly proposed sidelying TA, one that is comprised of sidelying IR ROM and supine ER ROM. A normative value of 160 ° and 163 ° for sidelying TA was established for the dominant and non‐dominant shoulder, respectively, regardless of gender or level of overhead athletic competition. There was also a difference between the two TA methods, with sidelying TA having 14 ° less motion than supine TA for both the dominant and non‐dominant shoulder.

In addition, a dominant‐shoulder deficit of 4 ° was noted for both supine and sidelying TA. When comparing this 4 ° deficit to the calculated SEM for sidelying TA, it was revealed that this deficit indicated a meaningful clinical difference and suggests further investigation be performed for sideling TA to determine the amount of TA deficit that places the dominant shoulder at greater risk for upper extremity injury. Youth and professional athletes should also be examined in future studies utilizing sidelying TA. Results of this investigation indicate that sidelying TA may be a viable alternative to examine for GIRD in the overhead athlete.