Differences in Shoulder Internal Rotation Strength Between Baseball Players With Ulnar Collateral Ligament Reconstruction and Healthy Controls

Abstract

Background:

Studies have indicated decreased shoulder internal rotation (IR) and external rotation (ER) strength in the throwing limb of baseball players after ulnar collateral ligament injury. There is limited evidence on the recovery of shoulder rotation strength after primary ulnar collateral ligament reconstruction (UCLR).

Hypothesis:

At the time of return to throwing, baseball players who underwent UCLR would demonstrate decreased IR and ER shoulder strength in the throwing arm as compared with healthy baseball players.

Study Design:

Cross-sectional study; Level of evidence, 3.

Methods:

Male competitive high school and collegiate baseball athletes participated in this study. Athletes who underwent UCLR were compared with healthy controls who were matched by age, height, weight, and position. Bilateral isometric shoulder ER and IR strength was measured using a handheld dynamometer for all participants at the time of initial evaluation (UCLR group) and throughout the course of a season (healthy group). Independent t tests were run to calculate mean differences in ER and IR shoulder strength between the groups, with significance set at P < .05.

Results:

A total of 86 baseball athletes participated in this study (43 UCLR group, 43 healthy group). At the time of return to throwing (mean ± SD, 194 ± 30 days postoperatively), the 2 groups demonstrated no significant differences in nonthrowing arm ER or IR strength (P = .143 and .994, respectively). No significant difference was found between groups for throwing arm ER strength (P = .921); however, the UCLR group demonstrated significantly less throwing arm IR strength than the healthy group (144.2 ± 27.8 vs 157.6 ± 27.1 N; P = .023).

Conclusion:

The results of this study demonstrate that throwing arm rotator cuff strength may not fully recover before the initiation of a return-to-throwing program after UCLR. These data provide a potential framework for clinicians to assist in the management and exercise prescription of the baseball athlete after UCLR and before medical release and the initiation of a return-to-throwing program.

The ulnar collateral ligament (UCL) is the primary restraint to valgus force at the elbow during the baseball throwing motion. ¹⁴ Throughout the last 3 decades, there has been a documented and steep rise in primary and revision UCL reconstruction (UCLR). ^7,9,25,45 Up to 25% of major league pitchers and 15% of minor league pitchers have reported a history of primary UCLR. ⁹ Revision rates of 15% to 37% for baseball pitchers ^25,45 and 3.4% to 8.2% for position players ⁷ have been reported after primary UCLR in professional baseball. While revision rates may vary among positions in baseball, it is well-established that decreased performance and career longevity are significantly diminished after surgery, making the focus of clinical outcomes after primary UCLR of paramount impact. ^3,26

A number of risk factors for excessive UCL stress and potential UCL injury or reinjury have been identified in the literature: glenohumeral range of motion measures, ^16,40 measures of throwing volume, ^1,34 kinetic chain deficits, ^15,18 and parameters of pitching performance. ^{2,4,13,29,31,39} Decreases in rotator cuff strength have also been linked to risk for elbow injury. ^6,17,38 Tyler et al ³⁸ noted a relationship between decreased supraspinatus strength and increased relative internal rotation (IR) strength and its association to elbow pain during overhand throwing. Preseason external rotation (ER) weakness has been associated with throwing-related, time loss injuries that required surgery in a cohort of professional baseball pitchers. ⁶ More specific to a UCL-injured cohort, baseball players with confirmed UCL injury have demonstrated decreased rotator cuff strength at the time of injury. ¹⁷ These studies identified a potential association between rotator cuff strength and elbow injury risk but did not address the relationship of rotator cuff strength after UCL injury and its relationship to secondary injury or return to prior level of competition.

The rotator cuff is one component of the kinetic chain that transfers the forces necessary to obtain high-velocity throws while controlling the stresses placed across the medial elbow, especially in the late cocking and acceleration phases. ⁸ Therefore, baseball players looking to return to preinjury throwing performance after UCLR must demonstrate adequate muscular force before the initiation of a return-to-throwing program. However, to date, no studies have examined rotator cuff strength at the time of return to throwing after UCLR.

The purpose of this study was to compare isometric IR and ER strength of the throwing and nonthrowing shoulders in male baseball players diagnosed with UCL tears with that of healthy age-matched baseball players at the time of return to throwing. It was hypothesized that baseball players who underwent UCLR would demonstrate significant decreases in IR and ER strength in the throwing arm as compared with that of healthy players.

Methods

Participants

This was a cohort study with institutional review board approval. Competitive high school and collegiate baseball players from across the United States were recruited to participate in this study. Male athletes who underwent UCLR were compared with healthy controls without a UCL tear matched by age, height, weight, and position. Assent and/or consent was collected before enrollment in the study.

Participants were considered for the study if they were a baseball athlete between the ages of 15 and 25 years. Inclusion criteria for the UCLR group were (1) an inability to throw as the result of an injury, (2) an inability to continue pitching or throwing in baseball at the level before the UCL injury, (3) clinical examination results positive for a UCL tear, (4) confirmation of a UCL tear via magnetic resonance imaging, and (5) intent to return to prior level of competition after UCLR and subsequent postoperative rehabilitation. Exclusion criteria were revision of primary UCLR, any full-thickness chondral defects >1 cm², and previous shoulder surgery for labral or rotator cuff pathology. Participants in the healthy group were included in this study if the baseball athlete (1) was between 15 and 25 years old, (2) had no history of UCLR or UCL repair, (3) had no history of shoulder surgery for labral or rotator cuff pathology in the past year, and (4) had no complaints of elbow or shoulder pain that limited or restricted recent participation. An investigator within our outpatient sports medicine clinic screened eligible participants and subsequently enrolled each individual who met the study criteria.

A total of 86 male competitive high school and collegiate baseball athletes participated in this study: 43 in the UCLR group and 43 in the healthy group. Figure 1 shows the recruitment process.

Figure 1.

Flowchart of study enrollment process. UCL, ulnar collateral ligament.

Surgical Procedure and Rehabilitation

The diagnosis of an elbow UCL tear was made by a fellowship-trained, board-certified orthopaedic surgeon (J.E.C.) and confirmed via magnetic resonance imaging. Participants who sustained a UCL tear were recruited during the initial evaluation by the participating physician (J.E.C.) and physical therapist (J.C.G.). All athletes in the UCLR group had surgery performed by the participating physician (J.E.C.) using the contralateral palmaris longus tendon graft; if the palmaris longus was absent in the contralateral extremity, the gracilis tendon graft was used. ²³ If the athlete’s contralateral palmaris longus tendon was available for a graft choice, a modified docked figure-of-8 procedure was performed ³⁰ ; if absent, the ipsilateral gracilis tendon graft was used via a docking method. ³³

Athletes in the UCLR group attended formal physical therapy with a standardized rehabilitation protocol (see Appendix Table A1) at the outpatient sports medicine clinic if they lived within a commutable distance. Those in the UCLR group who were unable to be seen at the outpatient sports medicine clinic participated in formal physical therapy in closer proximity to their homes with the same standardized rehabilitation protocol. Each patient attending formal physical therapy outside of the sports medicine clinic was evaluated by a board-certified, residency-trained sports physical therapist at each physician follow-up (every 4-6 weeks). During the follow-up visit, the participants were provided with written recommendations to improve their standardization of care based on input from the treating surgeon and objective clinical data as measured by the sports physical therapist. The participants of the healthy group were recruited from local high schools and colleges, and all were healthy at the time of data collection.

Rotator Cuff Strength Testing

Rotator cuff strength testing was performed in the outpatient sports medicine center for the UCLR group and at various high schools and universities for the healthy group. Strength testing was performed as previously described. ¹⁷

Using a handheld dynamometer (HHD; MicroFET 2 [Hoggan Scientific]), the primary examiner (J.C.G.) tested bilateral IR and ER isometric rotator cuff strength via a “make test” methodology. ³⁷ An HHD was used in place of an isokinetic testing device because it is readily available and time-efficient. Isometric rotator cuff strength testing using an HHD has been shown to be a reliable method to measure rotator cuff strength. ^20,32 To minimize variability with isometric rotator cuff strength testing, measurements were taken by the primary investigator (J.C.G.) to ensure consistency. Intrarater reliability standards were established in pilot testing for isometric rotator cuff strength and found to be good: ER, intraclass correlation coefficient (2,1) = 0.94 (SEM = 1.3); IR, intraclass correlation coefficient (2,1) = 0.93 (SEM = 2.1). Stabilization was not used, per previous literature ^5,32 citing high inter- and intraexaminer reliability with manual stabilization.

For isometric testing, the participant sat upright facing the examiner at the end of a treatment table with the arm positioned at the side in 0° of abduction, 90° of elbow flexion, and neutral rotation of the shoulder. A seated position with neutral abduction and rotation was chosen because of the previously reported inter- and intraexaminer reliability for this position. ³² The primary examiner then placed the HHD on the dorsal or volar side of the distal radioulnar joint to assess isometric ER or IR strength, respectively. The participant was verbally cued to sit tall with scapular retraction. The primary examiner instructed the participant to rotate his arm outward (ER) or inward (IR) with maximum effort for up to 5 seconds while maintaining the testing arm at the side with the same amount of elbow flexion. Minimal tactile cueing was applied to the lateral aspect of the participant’s’ elbow to maintain 0° of shoulder abduction during the test, to avoid compensation in the frontal plane (Figure 2). The primary investigator utilized a 10-point visual analog scale to monitor reports of pain during testing; any participant’s reported pain >2 points resulted in the exclusion of his results. The mean measurement of 2 bilateral trials was documented for IR and ER. If there was wide discrepancy in measurement during one of the trials (>3.0 N), an additional trial was recorded for consistency.

Figure 2.

Measurement for isometric rotator cuff (A) internal and (B) external rotation strength testing. The tester is pictured to the side of the participant to allow for visualization of dynamometer placement. The participant was instructed to rotate his arm inward (internal rotation) or outward (external rotation) with maximum effort while the investigator used minimal tactile cueing at the participant’s elbow to decrease frontal plane compensation.

For the UCLR group, medical release for return to throwing was based on clinical examination by the participating physician (J.E.C.) and a series of objective measures and patient-reported outcomes: humeral torsion, passive glenohumeral range of motion, and rotator cuff strength, as well as the Lower Quarter Y-Balance Test and the Kerlan-Jobe Orthopaedic Clinic Questionnaire. Rotator cuff strength testing was assessed at time of return to throwing (mean ± SD, 194 ± 30 days) by the primary investigator (J.C.G.) at the outpatient sports medicine center. All participants in the healthy group underwent strength testing before their fall or spring baseball season using the same methods as the UCLR group.

Statistical Analysis

The mean IR and ER strength of the throwing and nonthrowing shoulders was compared between the UCLR and healthy groups. Independent t tests and chi-square analysis, when appropriate, were used to calculate mean differences between groups for all demographic data and for shoulder ER and IR strength. The significance level was set at P < .05.

Post hoc power analysis was performed using G*Power 3.1, with power set at 0.80 (1 – β) and α = .05 (2-tailed) and with a calculated effect size of 0.49 for differences in throwing arm IR between groups. This revealed a power of 0.61 for our sample size.

Results

Of the 43 participants in the UCLR group, 23 were seen at our sports medicine clinic, while 20 were seen at formal physical therapy clinics in closer proximity to the participants’ hometowns. Of the 43 participants in the healthy group, 29 were measured preseason, 11 postseason, and 3 midseason. No participant had to withdraw from the study because of pain during testing. Descriptive characteristics are presented in Table 1. There were no significant differences in age, height, limb dominance, or weight between the UCLR and healthy groups.

Table 1

Participant Demographics ^a

	UCLR Group (n = 43)	Healthy Group (n = 43)	P
Age, y	18.1 ± 1.8	19.7 ± 1.6	.70
Height, cm	186.3 ± 6.1	186.0 ± 6.9	.50
Weight, kg	87.3 ± 9.4	87.0 ± 10.4	.69
Dominant limb			.09
Right	38	32
Left	5	11
Playing experience, y	13.5 ± 1.9	14.9 ± 1.9	.89
Position			.11
Pitcher	36	28
Catcher	1	1
Infielder	1	8
Outfielder	5	5
Utility	0	1

^a Values are reported as mean ± SD or No. of participants. UCLR, ulnar collateral ligament reconstruction.

The mean values for IR and ER rotator cuff strength of the throwing and nonthrowing arms are presented in Table 2. There was no significant difference between groups with regard to nonthrowing arm shoulder ER and IR strength. There was also no significant difference between groups for throwing arm shoulder ER strength. However, the UCLR group demonstrated significantly less throwing arm shoulder IR strength as compared with the healthy group (144.2 ± 27.8 vs 157.6 ± 27.1 N; P = .023, d = 0.49).

Table 2

Rotator Cuff Strength at the Time of Return to Throwing After UCLR vs Healthy Controls ^a

	Rotator Cuff Strength, N
	UCLR Group (n = 43)	Healthy Group (n = 43)	P
Nonthrowing arm
ER	116.8 ± 20.2	110.1 ± 18.1	.143
IR	149.9 ± 24.2	149.9 ± 26.4	.994
Throwing arm
ER	117.9 ± 25.8	116.5 ± 19.0	.921
IR	144.2 ± 27.8	157.6 ± 27.1	.023

^a Values are reported as mean ± SD. Bold P value indicates statistically significant difference between groups (P < .05). ER, external rotation; IR, internal rotation; UCLR, ulnar collateral ligament reconstruction.

Discussion

In the present study, no differences were found for nonthrowing arm isometric ER or IR strength or for throwing arm isometric ER strength. Baseball players who underwent UCLR demonstrated decreased isometric IR strength in the throwing arm when compared with healthy controls matched by age, height, weight, and position. The results support our hypothesis regarding differences in IR strength of the throwing arm, although no differences in ER strength of the throwing arm were found.

Several studies have investigated normative and baseline data in regard to rotator cuff strength in Little League, ¹⁹ high school, ^22,38 and professional baseball athletes. ⁶ Conflicting evidence exists regarding rotator cuff strength in these populations. ^6,19,22,38 In a study of 165 uninjured high school students, Hurd et al ²² found that shoulder ER strength on the throwing limb was less than that of the nonthrowing limb, whereas IR strength was higher on the throwing limb when compared with the nonthrowing limb. Similarly, when comparing isokinetic values for concentric and eccentric rotator cuff strength in high school baseball athletes, Mulligan et al ²⁸ found that eccentric and concentric IR strength was higher on the throwing limb than the nonthrowing limb and that eccentric and concentric ER strength was lower on the throwing limb when compared with the nonthrowing limb. One study ³⁶ cited greater shoulder ER and IR strength on the throwing limb versus the nonthrowing limb at the end of a competitive season, suggesting that time of year may play a role in the interpretation of strength results for healthy baseball athletes. When we consider other factors that may influence strength measurements, more recent evidence suggests that there is no difference in rotator cuff strength of the throwing limb by influence of position (pitcher vs position player). ¹¹

Previous work ¹⁷ on rotator cuff strength in baseball players diagnosed with a UCL injury revealed significant decreases in throwing arm shoulder IR and ER strength at the time of injury when compared with healthy controls matched by age, height, and weight. These findings, in combination with the results of the current study, suggest that athletes who sustain a UCL injury may recover ER strength throughout a course of recovery and rehabilitation but may not fully recover IR strength as compared with healthy controls. Interestingly, the healthy group in our study had weaker IR strength (157.6 ± 27.1 N) than the healthy group in the Garrison et al ¹⁷ study (174.9 ± 20.7 N), despite utilizing the same methodology and measuring device and including participants who were similar in age, height, and weight. These factors suggest that the deficit in IR strength after UCLR may be larger than that in the present data set. The authors from the previous study investigating rotator cuff strength in a UCL-injured cohort ¹⁷ did note that pain may have had an influence on the deficits at the time of injury; however, a visual analog scale was used to monitor for pain during rotator cuff strength testing, and they did not report any withdrawal attributed to pain during testing.

Several studies ^21,24,32 have examined normative HHD isometric rotator cuff strength data, including ER:IR strength ratios. The use of an ER:IR strength ratio may prove beneficial to quantify normal strength while correcting for differences in the chosen units of strength measured. Two studies ^21,24 analyzed ER:IR strength ratios utilizing a fixed HHD.

In a study examining rotator cuff strength in a group of 20- to 29-year-old men (n = 24), Hughes et al ²¹ found the ER:IR strength ratio to be 0.66 per a stabilized HHD. However, the testing positions in this study for ER and IR strength differed from those in the current study. In another study of rotator cuff strength using a stabilized HHD, Kolber et al ²⁴ reported ER:IR ratios ranging from 0.76 to 0.78; yet, their results should be used with caution when applied to the current data, as their study comprised healthy male and female participants. More recently, Riemann et al ³² analyzed the rotator cuff strength and ratios of 90 healthy men (mean age, 23.3 years) with current or past upper extremity sport participation. They utilized 3 testing procedures, including a seated-at-neutral measure of IR and ER strength similar to the procedure used in the present study. In the seated-at-neutral position, the previous authors found an ER:IR strength ratio of 0.87 on the dominant limb. In the present study, the ER:IR ratio of the dominant limb was 0.81 on average, indicating that the overall strength of the rotator cuff in our UCLR group may be decreased at the time of return to throwing as compared with a healthy population.

It is important to note that although IR strength may continue to improve after medical release and return to throwing, a structured return-to-throwing program alone may not improve shoulder IR strength. In a study ⁴⁴ of 9 NCAA Division II collegiate baseball pitchers (National Collegiate Athletic Association), isokinetic IR peak torque did not change over the course of a competitive baseball season. Furthermore, IR strength was observed to decrease over the course of a season in a cohort of high school pitchers. ²⁷ As such, it can be hypothesized from these data that neither baseball pitching nor a return to throwing progression alone will improve IR strength. This observation could be attributed to the fact that IR eccentric workload was shown to decrease after a pitching load of 60 pitches. ¹⁰ The internal rotators of the shoulder are highly active during throwing; however, the amount of muscle endurance, the type of muscle contraction required, and the overall load of the internal rotators to complete a competitive season (or a return-to-throwing program) may not be enough stimuli to see clinically and statistically meaningful strength changes. While these studies ^10,27,44 may help inform how IR strength changes with throwing, they included healthy collegiate and high school baseball athletes; it is unclear how these data would translate if the same testing protocols were applied to younger postoperative UCLR athletes. Future studies are warranted to investigate changes in IR strength throughout the course of a return-to-throwing program and/or a competitive season after UCLR. Additional research is needed to identify the long-term implications of deficits in throwing arm IR strength.

The combination of previous rotator cuff strength literature in relation to elbow injury in baseball players and findings from the current study suggests that the implementation of an isolated IR strengthening exercise program throughout rehabilitation after UCLR may be beneficial. Current rehabilitation protocols typically recommend strengthening of the rotator cuff for the internal and external rotators, either directly or through co-contraction, but with an observed emphasis on posterior rotator cuff strengthening (ER strengthening). ^12,43 The strength of the posterior rotator cuff and external rotators is crucial for controlling the shoulder during baseball throwing and pitching and, in turn, mitigating forces across the medial elbow. ^13,35 Decreased throwing arm shoulder ER strength has also been associated with throwing-related pain in adolescent pitchers, ³⁶ emphasizing the importance of ER strength. The data in the present study suggest that continued isolated strengthening of the internal rotators should be implemented and encouraged throughout a rehabilitation protocol to properly restore IR strength before return to throwing. With isolated rotator cuff strengthening, clinicians should continue to follow the most current recommendations for rehabilitation after UCLR that address the entire kinetic chain. ^41,42

Limitations

This study is not without limitations. The participants in the UCLR group were measured at the time of return to throwing. We do not know if IR strength would have resolved after a return-to-throwing program at the time of full, unrestricted return to play or return to performance. Additionally, these athletes were not measured before UCLR. It is possible that athletes in the UCLR group were weaker at baseline versus the healthy controls, and the potential strength deficit may have played a contributing role in the development of UCL-related pain. Future studies should consider examining longitudinal strength to determine if decreasing shoulder strength is a risk factor for the development of UCL injury.

Although a majority of the UCLR group was seen at the outpatient sports medicine center, some of the participants were unable to be seen at the same clinic. The sports physical therapists at our outpatient sports medicine center made efforts to standardize the care of these athletes via written and verbal communication with the physical therapists overseeing the postoperative rehabilitation of the injured athletes at other locations; however, the lack of ability to directly oversee and provide postoperative rehabilitation for every participant may have influenced these results.

The participants in the healthy group were recruited from local high schools and universities, with measurements performed before the fall or spring baseball season. There is currently limited evidence available to say how time of year or season may affect rotator cuff strength, but this could have influenced the results of the strength measures for the healthy group. In the present study, ER and IR strength were evaluated isometrically with the participant in a seated position and the testing arm held to the side with the elbow flexed.

Although this does not represent a functional throwing position, it does allow for better standardization and reliability among measures. Additionally, the interpretation of rotator cuff strength tested in this position should be used with caution. Given the clinical nature of this study, rotator cuff strength assessments are subject to the involvement of secondary stabilizing muscles. Specifically, the deficits observed in IR strength may be attributed to rotator cuff weakness and potential contribution from the pectoralis major and latissimus dorsi muscles. To reduce the influence of these factors, patient cuing and training of the assessors were performed, as demonstrated by the acceptable level of measurement reliability.

Finally, post hoc analysis for the current study revealed 0.61 power with the current sample size. To reach appropriate statistical power (0.80), we would have needed to recruit 48 more participants (24 per group). This could change the results of significance testing with the addition of more participants.

Conclusion

Baseball players who underwent UCLR demonstrated decreased shoulder IR rotator cuff strength at the time of return to throwing when compared with healthy controls. These data provide a potential framework for clinicians in the treatment of baseball athletes after UCLR to ensure that rotator cuff strength has been adequately restored. Knowledge of these strength deficits may assist the clinician in the management and exercise prescription of the baseball athlete after UCLR and before medical release and the initiation of a return-to-throwing program.