Upper Extremity Musculoskeletal Characteristics and the Kerlan-Jobe Orthopaedic Clinic Questionnaire Score in Collegiate Baseball Athletes

For this descriptive cohort study, we recruited male baseball athletes from a National Collegiate Athletic Association Division I institution. The inclusion criteria were as follows: (1) male between the ages of 18 and 40 years, (2) currently rostered and participating as a varsity-level collegiate baseball athlete, and (3) cleared by medical personnel (certified AT or team physician) to participate in full physical activity. A total of 37 athletes were enrolled (Tables 1 and 2). All participants provided written informed consent, and the study was approved by the Duke Health Institutional Review Board.

All testing was carried out by a certified AT (M.S.F.), who is trained and proficient in UE posture, ROM, flexibility, and isometric-strength testing. Intertester reliability for the chosen methods has been established in a small sample of recreationally active individuals.¹⁶  All participants underwent 1 test session that occurred within 4 weeks of the start of the 2017 collegiate baseball season. The session lasted approximately 45 minutes and consisted of a UE injury history questionnaire, KJOC questionnaire, and assessments of posture, ROM, flexibility, and isometric strength.

During the test session, injury histories were collected from all athletes. They self-reported all shoulder-joint injuries sustained over their lifetimes. These injuries were confirmed by the certified AT responsible for their care if they were under the care of the certified AT at the time of injury; the injury information was collected as part of a preparticipation physical examination, or both.

All athletes completed the KJOC questionnaire during the test session. This questionnaire has been shown²  to be a valid, reliable, and responsive instrument for gathering patient-reported outcome measures in baseball athletes. The questionnaire comprises 10 questions about UE health and function. Athletes were instructed to place an x on a 10-cm line associated with each of the 10 questions. The position of the x indicated their current level of performance or function, or both. An x placed closer to the right indicated a higher rating for performance, function, or both.

We assessed forward head posture and bilateral forward shoulder posture.^17–19  A lateral photograph was taken of each athlete bilaterally. Before taking the lateral photograph, we provided the athletes with standardized instructions to look straight forward with their UEs at their sides as if they were “standing in a line.” Next, we placed marks on the humeral heads and the spinous process of the C7 vertebra, which are the necessary landmarks for assessing forward head and bilateral forward shoulder posture. After each photograph was uploaded to the computer, we used the ImageJ (version 1.51h; National Instititues of Health, Bethesda, MD) program to assess posture. To measure forward head posture using the angle tool, 1 line was drawn from the tragus to the spinous process of the C7 vertebra, and another line was drawn parallel to the floor from the spinous process of the C7 vertebra. To measure bilateral forward shoulder posture, 1 line was drawn from the spinous process of the C7 vertebra to the mark made on the humeral head, and another line was drawn parallel to the floor from the mark made on the humeral head. The angle formed at the intersection of the lines for both forward head posture and forward shoulder posture was measured and reported.

Upper extremity ROM and flexibility were measured using preestablished techniques^20–22  that have been shown to have good to excellent reliability.^16,20–22  The measures consisted of passive glenohumeral internal and external rotation, posterior shoulder tightness, and pectoralis minor length. All measures were collected for the dominant and nondominant UEs. The dominant UE was defined as the athlete's throwing arm. For each test of UE ROM and flexibility, we collected 3 measurements and averaged them for data analysis.

Passive glenohumeral internal and external rotation were used to isolate and measure the athlete's internal- and external-rotation ROM.²²  Each athlete lay supine on a treatment table with the shoulder positioned in 90° of abduction and the elbow flexed to 90°. The examiner stabilized the scapula during the assessment. Both assessments were based on the recommendations of Norkin and White.²²  All measurements were taken to the nearest 0.1°.

We used posterior shoulder tightness to assess the extensibility of the posterior shoulder capsule.²¹  The athlete lay supine on a treatment table with the shoulder flexed to 90° and the elbow flexed to 90°. The examiner horizontally adducted the shoulder by bringing the extremity across the athlete's chest until a normal end-feel was reached; during the test, the examiner ensured no movement occurred in the scapula. All measurements were taken to the nearest 0.1°.

Pectoralis minor length was used to assess the flexibility of the pectoralis minor muscle. The athlete lay supine on a treatment table with his arms resting comfortably on the treatment table beside his body and his hands folded across his abdomen. The examiner palpated and subsequently marked the posterior lateral tip of the acromion. Using a double square, the examiner measured from the top of the treatment table to the mark that was made on the posterior lateral tip of the acromion. All measurements were made level, perpendicular to the treatment table, and to the nearest millimeter.

Isometric strength of the UE was measured using previously established protocols^23,24  that have demonstrated good to excellent reliability.¹⁶  The middle trapezius, lower trapezius, rhomboid, glenohumeral internal and external rotation, pectoralis major (lower fibers), serratus anterior, supraspinatus, and upper trapezius were tested. A handheld dynamometer (HHD; Lafayette Instrument Co, Lafayette, IN) was used to perform all isometric-strength testing. We conducted all such trials as “make tests.” During a make test, the examiner holds the HHD steady as the individual exerts maximal force. Athletes were allowed 1 practice trial at 50% of their perceived maximal effort for each testing position. A total of 3 measured trials at 100% of the athlete's perceived maximal effort were performed by the dominant and nondominant UEs. Each trial lasted 5 seconds; a 60-second rest period was given between trials to prevent fatigue. Peak force was recorded to the nearest 0.1 kg, and the average of 3 trials each for the dominant and nondominant UEs was used for data analysis.

Middle trapezius isometric strength was measured with the athlete lying prone on a treatment table, the head positioned in neutral, and the elbow of the test limb fully extended.^23,24  The UE was placed in 90° of abduction and externally rotated so the thumb was pointing toward the ceiling. The examiner placed the HHD slightly proximal to the radial styloid and used the opposite hand to stabilize the scapula.

Lower trapezius isometric strength was measured with the athlete lying prone on a treatment table, the test limb positioned over the head in 135° of abduction, and the elbow fully extended.^23,24  The thumb of the test limb was pointed toward the ceiling. The examiner placed the HHD slightly proximal to the radial styloid process and used the other hand to stabilize the scapula.

Rhomboid strength was measured with the athlete lying prone on a treatment table.^23,24  The athlete's head rested on the forearm of the nontest limb, and the test limb was positioned in 90° of abduction with the elbow extended and the thumb pointing toward the floor. The examiner placed the HHD proximal to the ulnar styloid and used the opposite hand to stabilize the scapula. Athletes were instructed to push straight upward for all measurements, with the motion coming from the muscle being tested.

Glenohumeral internal- and external-rotation strength was measured with the athlete lying prone on a treatment table^23,24  and the test limb abducted to 90°. The athlete was instructed to rotate the elbow inward or outward. The examiner stood on the side of the test limb and placed the HHD on the volar or dorsal side of the forearm, slightly proximal to the wrist.

The lower fibers of the pectoralis major were tested with the athlete lying supine on a treatment table.¹⁶  The test limb was positioned with the shoulder abducted to 120°, the elbow fully extended, and the palm facing upward. The athlete was instructed to push up and inward toward the opposite hip while keeping the UE straight. The HHD was placed slightly proximal to the elbow.

The serratus anterior was tested with the athlete positioned short sitting on a plyometric box.^23,24  The test limb was placed in approximately 130° of flexion, the elbow was fully extended, and the palm was facing the ground. The HHD was placed just proximal to the wrist. The athlete was instructed to keep the UE straight and push straight toward the ceiling.

The supraspinatus was tested with the athlete positioned short sitting on a plyometric box.²⁴  The test limb was abducted to 90° and horizontally adducted to 40° anterior to the coronal plane with the elbow fully extended. The HHD was placed just proximal to the wrist. The athlete was instructed to push straight upward while maintaining the position of the thumb pointing straight toward the ceiling.

The upper trapezius was tested with the athlete positioned short sitting on a plyometric box.^23,24  His hands were relaxed in his lap, and his face was turned just slightly away from the test limb. The examiner placed the HHD through a therapy strap and held it over the acromion process. The athlete was instructed to depress the shoulder before test initiation and to raise his shoulders toward his ears or shrug his shoulders upward to initiate the test.

To score the KJOC questionnaire, each 10-cm line associated with the 10 questions was measured. The measurement was taken from the farthest point on the left side of the line to the x that was marked as a measure of the current level of performance, function, or both.²  All measures were recorded to the nearest 0.1 cm. The measurements for all 10 questions were summed to represent an overall score for the UE level of performance, function, or both out of a total of 100. The measurements for questions 1 through 5 were summed to represent an overall score for UE pain. The measurements from questions 6 through 10 were summed to represent an overall score for UE function.

All UE isometric-strength measurements were normalized to body weight (%BW). Normalized UE isometric strength was used to calculate strength for the scapular muscle group, strength for the glenohumeral muscle group, and overall total UE strength for both the dominant and nondominant limbs.

We assessed all variables for normality. All variables assessed as normal were analyzed with an independent-samples t test to determine if differences existed between pitchers and position players and between athletes who had or had not sustained an injury in the year before testing. For all variables that were not normally distributed, we calculated a Mann-Whitney U test for analyses between pitchers and position players and between athletes with or without injury in the year before testing. The α level was set a priori at .05. We used SPSS (version 24; IBM Corp, Armonk, NY) and Stata 14 (StataCorp LLC, College Station, TX) for all statistical analyses.

A stepwise multiple regression model was fit using Stata 14 to determine if the predictor variables assessed in this study could predict the KJOC scores. The predictor variables were GIRD, total arc of motion (passive glenohumeral internal and external rotation), posterior shoulder tightness, total strength of the scapular muscles, total strength of the glenohumeral muscles, and total strength of the entire shoulder complex. The response variables were the KJOC total score, KJOC total score for questions 1 through 5, and KJOC total score for questions 6 through 10. We used the Pearson product moment correlation for normal variables and the Spearman rank correlation for nonnormal variables to determine if any relationships existed between the predictor variables and response variables chosen for this study. We calculated β coefficients to assess the relative predictive power of each predictor variable. We set the α level at .10 to determine which of the predictor variables would be included in the final regression model.

Means and standard deviations (SDs) for all baseball athletes and for baseball athletes stratified by position and UE injury history are presented in Tables 3 through 6 and Supplemental Tables 1 through 4 (available online at http://dx.doi.org/10.4085/1062-6050-81-18.S1). These data provide a reference dataset describing posture and UE ROM, flexibility, and strength in Division I collegiate baseball athletes. Pairwise correlations between the response variables and the predictor variables selected for this study are presented in Table 7. We observed no correlations between the predictor and response variables.

Multiple linear regression models that retained predictor variables are presented in Tables 8 and 9. Based on the multiple linear regression model, 1 predictor variable was maintained in the final equation for the response variable KJOC total score: history of UE injury in the year before testing. This multiple linear regression model accounted for 7.80% of the variance in KJOC total score (P = .07). Based on the multiple linear regression model, 2 predictor variables were maintained in the final equation for the response variable KJOC total score for questions 1 through 5: history of UE injury in the year before testing and posterior shoulder tightness. This multiple linear regression model accounted for 14.40% of the variance in KJOC total score for questions 1 through 5 (P = .047). The multiple linear regression model retained no predictor variables for KJOC total score for questions 6 through 10.

Normalized β coefficients for the multiple linear regression model are presented in Table 10. The β coefficients showed that, for every 1-SD increase in UE injury history, the predicted decrease of the KJOC total score would be 0.297 SD. For the total of questions 1 through 5 of the KJOC score, the greatest effect was observed with posterior shoulder tightness. For every 1-SD increase in posterior shoulder tightness, a decrease of 0.348 SD in the KJOC total score for questions 1 through 5 would be expected.

Perceived UE pain and function, as measured by the KJOC questionnaire, has not been linked to UE musculoskeletal characteristics, including posture, ROM, flexibility, and strength. Ultimately, this may allow for better discrimination among baseball athletes who do not have any UE pain, are playing with UE pain, or are not playing due to UE pain. Therefore, the primary purpose of our study was to describe the musculoskeletal characteristics of the UE—posture, ROM, flexibility, and isometric strength—in a population of baseball athletes and to further analyze potential differences between player position and injury history. The secondary purpose of our study was to determine if these musculoskeletal characteristics had predictive capability for the KJOC scores of baseball athletes.

Our results demonstrated differences in the musculoskeletal characteristics of the UE when the baseball population was stratified by position and 1-year injury history. We also observed a relationship between KJOC total score for questions 1 through 5 and a history of UE musculoskeletal injury and posterior shoulder tightness. The hypothesis that predictive capabilities would be present between posture, UE ROM, flexibility, and isometric strength and the KJOC questionnaire score was partially supported, and both UE musculoskeletal injury history in the past year and posterior shoulder tightness were partially predictive of the KJOC score.

Comparisons of UE musculoskeletal characteristics between pitchers and position players illustrated group differences. Based on averages, pitchers demonstrated less forward shoulder posture in the dominant shoulder than position players. A forward shoulder posture may be attributed to imbalances in the strength of the scapular-stabilizer muscles, tightness, or both in the anterior shoulder musculature.^17–19  Comparisons of ROM and flexibility between pitchers and position players revealed differences in external-rotation ROM and, subsequently, total arc of motion. Pitchers displayed greater external-rotation ROM and total arc of motion than position players. Greater external-rotation ROM has been noted in pitchers and may be due to bony and muscular adaptations that have been attributed to the pitching motion.¹⁶  Generally, pitchers demonstrated less UE strength than position players. These differences were evident in the strength of the middle and lower trapezius, rhomboid, and serratus anterior and the calculated total strength for the scapular muscles. Less strength in pitchers may be explained by a lack of focus on pure strengthening exercises to preserve the shoulder for pitching activities. A certain level of strength must be maintained to achieve optimal performance, but the fatigue caused by repetitive overhead throwing may affect pitchers' ability to perform effective strength training.^4,7  In addition to strength, athletes must maintain the muscular endurance necessary to overcome the fatigue associated with the throwing motion.

The KJOC questionnaire was developed to assess the functional status of the UE, specifically in athletes who participate in overhead sports.^2,25  Traditional methods of examining pain, including the Disabilities of the Arm, Shoulder and Hand questionnaire, do not satisfy the need to assess sport-specific characteristics,^2,25,26  including subtle changes in endurance, velocity, power, and ball control.²⁵  Researchers^3,25,26  have reported that the KJOC score can distinguish among professional baseball athletes who do not have pain, are playing with pain, or are not playing due to pain. A KJOC score threshold of 81.3 is more sensitive and accurate than the threshold for other questionnaires used to assess the UE.²⁶  A KJOC score greater than 81.3 indicates with 95.1% accuracy that a patient has returned to play.²⁶  The total KJOC score for all baseball athletes tested in our study was 93.17, indicating that these athletes scored above the return-to-play standard as assessed by the KJOC questionnaire. The pitchers in our study scored just slightly lower than the position players (KJOC scores = 91.28 and 95.33, respectively). We found greater differences between athletes who did or did not have a history of UE musculoskeletal injury in the year before testing (KJOC scores = 85.30 and 94.74, respectively). The baseball athletes who had a history of UE musculoskeletal injury in the year before testing scored just slightly higher than the KJOC score used as a return-to-play criterion. The direct relationship between a 1-year history of UE musculoskeletal injury in baseball athletes and the resultant KJOC score may indicate a need to address pain and self-perceived performance in athletes who are rehabilitating after UE musculoskeletal injury. Obtaining baseline KJOC scores before a UE musculoskeletal injury event may allow for more individualized clinical care and return-to-play decisions. Baseline KJOC scores can be compared with all subsequent KJOC scores to determine an individual KJOC score threshold for return to play after UE musculoskeletal injury.

The most common pain in the UE occurs in the glenohumeral joint and has been reported by 32% of pitchers in 7% of pitching appearances.⁴  Glenohumeral joint pain has been associated with changes in UE ROM and flexibility, UE strength, limb fatigue, and decreased self-perceived performance.⁴  We demonstrated that a 1-year history of UE musculoskeletal injury and posterior shoulder tightness were related to the KJOC total score for questions 1 through 5. The relationship between UE musculoskeletal injury and the KJOC score indicated that, after UE musculoskeletal injury, changes in pain and function of the UE were likely to occur. This may be due to the UE musculoskeletal characteristics not meeting optimal functional levels after UE musculoskeletal injury, translating to pain, limb fatigue, and decreased self-perceived performance. Athletes with a 1-year history of UE musculoskeletal injury scored consistently lower on the KJOC questionnaire than athletes who did not report such a history. This highlights the continued need for injury rehabilitation and prevention programs after return-to-play criteria have been met and athletes have returned to full participation in baseball activity. The relationship between posterior shoulder tightness and the KJOC score emphasizes the importance of continued participation in ROM and flexibility-training programs throughout a baseball athlete's career, as well as after UE musculoskeletal injury. Interestingly, strength was not related to the KJOC score, which may suggest that unilateral strength was not necessarily related to changes in the KJOC score. In baseball athletes, determining whether differences exist between the dominant and nondominant UEs and examining these differences as they relate to the KJOC score may be important.

We acknowledge that our study had inherent limitations. Changes in the musculoskeletal characteristics of the UE, pain, function, and self-perceived performance may be directly related to the type of UE injury sustained. For this study, all injuries were categorized as general UE musculoskeletal injuries, regardless of the type or location. This general categorization may have affected the predictive capability of a 1-year history of UE musculoskeletal injury on the KJOC score. The outcomes may change when UE musculoskeletal injuries are stratified by location or injury severity; researchers should examine this possibility, which would require a larger sample size and a greater number of UE musculoskeletal injuries than were available in our study. We also did not assess the strength of the individual UE muscles. Instead, we summed the strength of the individual UE muscles to calculate the strength of the scapular and glenohumeral muscle groups and overall UE strength. Addressing the relationship between each UE muscle's strength and the KJOC score may be necessary to determine how different injury types affect the KJOC score. Lastly, the statistical analysis included multiple t tests, which we acknowledge may have increased the risk of a type 1 error.

Upper extremity musculoskeletal pain and injury are common in baseball athletes^1,3,4,7,26  and may negatively affect player safety and performance.^3,4,7,26  Our results demonstrated that a 1-year history of UE musculoskeletal injury and posterior shoulder tightness were predictive of the KJOC total score for questions 1 through 5 in baseball athletes. These results highlight the need for a continued focus on self-perceived pain, function, and performance after UE musculoskeletal injury. Therefore, when baseball athletes return to sport after UE injury, clinicians need to screen for continuing pain and diminished self-perceived performance. Upper extremity injury-prevention and rehabilitation programs should be aimed at caring for each athlete's physical and mental wellbeing to provide the most effective treatment.

Supplemental Tables. Posture and strength of pitchers versus position players; Posture and strength of injured versus uninjured players.

Found at DOI: http://dx.doi.org/10.4085/1062-6050-81-18.S1