Functional Performance Tests Reveal Promising Results at 6 Months After Shoulder Stabilization Surgery

Abstract

Background:

Although athletes are mostly allowed to return to play 6 months after shoulder stabilization surgery, there are inadequate data about their functional status during this period.

Hypotheses:

Performance tests would reveal insufficiency in the functional capacity of shoulder 6 months after stabilization surgery.

Study Design:

Prospective cohort study.

Level of Evidence:

Level 3.

Methods:

A total of 32 male athletes with arthroscopic anterior capsulolabral repair (AACR) were included in the study. Shoulder internal and external rotator (IR-ER) strength was assessed using isokinetic dynamometer at 60°/s and 180°/s angular velocities preoperatively and 6 months postoperatively. Shoulder function was assessed with closed kinetic chain upper extremity stability (CKCUES) test, Y balance test-upper quarter (YBT-UQ), and unilateral seated shot-put test (USSPT) at 6 months postoperation. Western Ontario shoulder instability index (WOSI) and Tampa scale of kinesiophobia (TSK) were used for the self-assessment of the shoulder. Mixed-model ANOVA was used to analyze the changes in the IR-ER strength on both shoulders. Limb symmetry index (LSI) was calculated for the IR-ER strength, YBT-UQ, and USSPT scores.

Results:

Shoulder IR strength was higher at 6 months postoperatively compared with preoperatively. The LSI was 76.4% and 76.6% for ER strength, and 94.2% and 94% for IR strength at 60°/s and 180°/s angular velocities, respectively, at the postoperative 6 month timepoint. The mean CKCUES test score was 21.8 ± 2.6 touches and the LSI was 94.7% for the YBT-UQ and 102.5% for the USSPT. WOSI (P < 0.001) and TSK (P = 0.001) scores were significantly lower at 6 months postoperatively.

Conclusion:

Functional status of the patients with shoulder stabilization surgery improved considerably 6 months after surgery, yet they did not fully recover function.

Clinical Relevance:

Exercise programs focusing on shoulder ER strength and shoulder performance should be emphasized after stabilization surgery.

Surgical repair of the capsulolabral complex restores the stability of the glenohumeral joint and significantly decreases the risk of redislocation in patients with shoulder instability.^2,18,24,32 More than 80% of patients can return to play (RTP) at an average of 6.1 months after stabilization surgery.^1,9,12,14,19 Despite these successful results, redislocation rates of 4% to 19% after surgical repair have been reported.^2,20,28 In addition, between 9% and 35% of the patients could not RTP at preinjury levels and have had to decrease their activity.^1,9,14,19,26

Decision-making for RTP after shoulder stabilization surgery is based mostly on the time elapsed since surgery, shoulder range of motion (ROM), and whether patients can perform sport-specific activities without limitation. ³³ In recent years, due to the problems mentioned previously, studies have begun to evaluate shoulder function using standardized and more objective criteria to decide whether patients are functionally ready for RTP.^4,13,22,37 However, there is limited and conflicting knowledge about recovery of shoulder function at 6 months after stabilization surgery. Amako et al ⁴ reported that shoulder rotator cuff (RC) strength had recovered on the operated (OP) shoulder 6 months after surgery as a preoperative level was reached. In contrast, Frantz et al ¹³ reported decreased RC strength at 6 months after surgery compared with preoperatively. Unlike the others, Wilson et al ³⁷ compared the RC strength of the OP shoulder with the nonoperated (non-OP) shoulder. They found decreased RC strength on the OP shoulder compared with the non-OP shoulder at 6 months postoperatively and concluded that patients did not meet the strength criteria. Although muscle strength is the functional outcome most commonly assessed during decision-making for RTP, there is no consensus in the strength criteria among the studies. Amako et al ⁴ reported that RC strength was recovered based on the preoperative values, whereas Wilson et al ³⁷ reported inadequate RC strength based on the non-OP shoulder. Frantz et al ¹³ also aimed to reach the preoperative level, yet they assessed the RC strength with a manual muscle testing method, which has low reliability and limited diagnostic accuracy. ⁶

Another reason for the conflicting results between studies is the difference in study groups. The sport type and activity level of the patient are important factors affecting shoulder functional level, RTP decision, and reinjury risk after RTP. ¹⁴ However, previous studies were conducted on heterogenous groups from different populations and different types of sport. For better evaluation of shoulder function during decision-making for RTP, studies with more homogeneous groups, and assessment of athletes both pre and postoperatively, are needed.

This study aimed to investigate pre- and postoperative shoulder function (strength and functional outcomes) in athletes from different types of sport who have had shoulder stabilization surgery. We hypothesized that, after shoulder stabilization surgery, athletes would demonstrate decreased functional outcomes at 6 months after surgery. The second hypothesis was that there would be differences in functional status of athletes from different types of sport.

Methods

Study Design and Ethical Approval

This was a prospective cohort study investigating the functional status of the patients with arthroscopic anterior capsulolabral repair (AACR) preoperatively and 6 months postoperatively. The study was conducted between April 2019 and July 2021 at Hacettepe University, Faculty of Physical Therapy and Rehabilitation and approved by the ethical committee of the Hacettepe University. It was carried out according to the Declaration of Helsinki. All subjects signed an informed consent form.

Patients

A total of 32 male athletes diagnosed with anterior shoulder instability and planned for stabilization surgery were included in the study. The inclusion criteria were being between 18 and 45 years old, having no previous shoulder surgery, having no injury on the contralateral side over the last year, being either a professional or a recreational athlete, and being able to perform preoperative assessments without limitation. The dominant shoulder was operated on in 17 patients and the nondominant shoulder in 15 patients. The mean time from the initial dislocation to surgery was 11.8 ± 15.3 months. The demographics of the patients were recorded before the assessments (Table 1).

Table 1.

Characteristics of participants

Demographics
Age, years ± SD	24.5 ± 5.6
Height, cm ± SD	177.2 ± 7.2
Weight, kg ± SD Time to surgery, mo	77.4 ± 12.3 11.8 ± 15.3
Number of Participants in Each Sport Group
Type 4	11
Type 3	10
Type 2	11
Type 1	0
ROM on Operated Shoulder
	Preoperative	Postoperative
Flexion, deg	176.6 ± 9.3	179 ± 2.5
Abduction, deg	177.5 ± 7.8	180 ± 3
External rotation, deg	102 ± 11.5	100 ± 7.4
Internal rotation, deg	64 ± 10.8	69 ± 7.2

ROM, range of motion. deg: degree, mo: month.

Patients were classified according to the potential risk of shoulder injury based on the type of the sport they performed, as described by Allain. ³ Type 4 sports included overhead hitting movement with the addition of sudden stops such as basketball, handball, volleyball, goalkeeping, rugby, judo, karate, or wrestling. Type 3 sports included overhead sports with hitting movements such as climbing, weightlifting, shot putting, crawl or butterfly swimming, pole vaulting, figure skating, canoeing, pitching, golf, and field hockey. Type 2 sports included high-impact sports such as martial arts except for judo and karate, cycling, motocross, soccer, rugby with exception of hooking, water skiing, downhill skiing, parachuting, and equestrianism. Type 1 sports included nonimpact sports such as rowing, fencing, breaststroke, diving, gymnastics (without apparatus), skiing, shooting, and sailing.

Procedures

Surgical Procedure

A similar surgical procedure was conducted in all patients by 3 different surgeons having experience in shoulder surgery for >15 years. During the operation, patients were placed in a bench-chair position under general anesthesia. An anteroinferior portal was opened for imaging of the shoulder joint. None of the patients had >20% glenoid rim defect or off-track Hill-Sachs lesions. Four patients had superior labrum anterior-posterior (SLAP) lesions. Later, other portals (posterior and anterosuperior) were created. The capsulolabral complex was repaired using 2 suture anchors in all cases (Osteoraptor 2.9 mm, Smith & Nephew). The SLAP lesion was repaired using 1 anchor.

Postoperative Rehabilitation

A standardized postoperative rehabilitation program was applied to all patients by a single physical therapist with 8 years of experience in shoulder rehabilitation. Patients used a shoulder sling, which was positioned in internal rotation (IR), for 4 weeks in the postoperative period. Passive-controlled motion was started 3 weeks postoperatively and active exercises were initiated after the sling was removed. RC strengthening and scapular control exercises were started after the sling was removed and increased progressively. Patients started controlled stretching exercise in postoperative week 6. Full active ROM was aimed for by the end of postoperative week 12. Functional rehabilitation was initiated after postoperative week 12. ¹⁶

Outcome Measures

Shoulder ROM was assessed preoperatively and 6 months postoperatively using a universal goniometer. Shoulder flexion, abduction, IR, and external rotation (ER) were assessed while patients were in the supine position. IR and ER were evaluated at 90° of shoulder abduction and 90° of elbow flexion position. Two physical therapists conducted the ROM measurements. One physical therapist passively moved the patient’s arm, while the other measured the range of the movement with the goniometer. ³⁶

Shoulder rotational muscle strength was assessed using an isokinetic dynamometer preoperatively and 6 months postoperatively (IsoMed 2000, D&R GmbH). Patients were seated on the isokinetic device, while their upper extremity was positioned in 45° of shoulder abduction, 90° of elbow flexion, and neutral forearm position.^4,5 Shoulder testing range was determined as 30° of IR to 45° of ER from the neutral position (Figure 1). ⁵ Before the tests, 5 submaximal repetitions were carried out for familiarization. During the testing procedure, patients performed maximal IR and ER against the device at 60°/s and 180°/s angular velocities. Patients exerted 10 repetitions at the 180°/s and 6 repetitions at the 60°/s and rested for 1 min between each set. Verbal encouragement was given during the tests. First the non-OP side and then the OP side was tested. The body weight-normalized peak torques for each test were recorded for the analyses. ⁴

Figure 1.

Isokinetic test position.

The closed kinetic chain upper extremity stability (CKCUES) test was used to assess shoulder stability and anaerobic performance 6 months postoperatively.^31,34 The test was performed in the traditional push-up position, with hands at 36 inches apart. ³¹ During the testing procedure, patients were instructed to reach out with 1 hand to the opposite hand, touch it and return to the starting position while maintaining the correct posture (Figure 2). After a familiarization trial, 3 rounds of 15-second testing trials with 45-second resting periods were performed. The number of touches was measured in each trial and the average of the 3 trials was recorded. ⁷

Figure 2.

Closed kinetic chain upper extremity stability test.

The Y balance test-upper quarter (YBT-UQ) was used to assess the stability of the upper extremity and the upper body 6 months postoperatively. During the testing procedure, patients adopted the push-up position with the tested hand on the stance platform and the thumb behind the red line. In this position, patients were asked to push the indicator with their free hand as far as possible in the medial, superolateral, and inferolateral directions and return to the initial position (Figure 3). Patients performed 1 practice trial and 3 testing trials for each direction with each side with 45 seconds of resting intervals. First the non-OP side and then the OP side was tested. ⁷ The average of 3 trials was recorded for each direction. The composite score of 3 directions was used for the analyses. ⁷

Figure 3.

Y balance test-upper quarter.

Unilateral seated shot-put test (USSPT) was used to assess the upper extremity anaerobic power 6 months postoperatively. ⁸ Patients were seated on a standard armchair without armrests while the lower legs were placed on another chair in front of them. They held a 3-kg medicine ball at shoulder height and were asked to push the ball as far as possible (not to throw). During the test, patients were instructed to keep their back in the chair without any trunk movement while the nonthrowing arm was placed on the opposite shoulder (Figure 4). Participants performed 2 practice trials (with 75% and 100% efforts, respectively) and followed by 3 testing trials with 45 seconds of resting periods. The testing order was first the non-OP side then the OP side. The distance patients pushed the medicine ball was recorded and the average of 3 trials was used for the analyses. Verbal encouragement was given during the test. ²¹

Figure 4.

Unilateral seated shot-put test.

The Western Ontario shoulder instability index (WOSI) was used for self-assessment of shoulder function preoperatively and 6 months postoperatively. WOSI evaluates shoulder function with 21 questions about daily life, sports, work, and quality of life. ¹⁷ The total score can range between 0 and 2100 points, in which 0 represents no deficits in shoulder function and 2100 represents the worst. We presented patients’ scores as the percentage of the deficit (0%, no deficits; 100%, the worst) which is more useful clinically. ²⁵

Statistical Analyses

SPSS Version 21 (IBM Corporation) was used for the statistical analyses. Mean and standard deviations were calculated for the quantitative variables. A 2 × 2 mixed model analysis of variance (ANOVA) with 2 within factors (time by shoulder) was used to analyze the differences between the pre and postoperative RC strength on the OP and non-OP shoulders in all patients. When there were significant interactions/main effects, pairwise comparisons were conducted to analyze side-to-side and pre and postoperative differences.

Pre and postoperative RC strength on the OP and non-OP shoulder in the different sports groups were analyzed using 3 × 2 × 2 mixed model ANOVA. Patients’ sport type was determined as the between-subject factor, while time and shoulder were determined as within-subject factors.

Pre and postoperative WOSI scores were analyzed in all patients and in 3 groups separately using the Wilcoxon signed-rank test. Limb symmetry index (LSI) was used to express the OP shoulder function compared with the non-OP shoulder at 6 months postoperatively (score of OP shoulder/score of non-OP shoulder × 100). The LSI was calculated for RC strength, YBT-UQ, and USSPT. The acceptable criterion for the LSI was determined as >90%. ²³

Results

Considering All Patients

There was a significant time-by-shoulder interaction for the shoulder ER strength at both 60°/s (F_1,31 = 15.04, P = 0.01) and 180°/s (F_1,31 = 15.1, P < 0.001) angular velocities. Pairwise comparisons revealed similar RC strength at the postoperative 6 months compared with the preoperatively at 60°/s (P = 0.65) and 180°/s (P = 0.02). However, the ER strength on the OP shoulder was lower compared with the nonoperative shoulder at the postoperative 6 months at both 60°/s (P < 0.001) and 180°/s (P < 0.001). The LSI was 76.4% and 76.6% at 60°/s and 180°/s angular velocities, respectively. No significant shoulder by time interaction was observed for the shoulder IR strength at 60°/s (F_1,31 = 0.75, P = 0.39) and 180°/s (F_1,31 = 3.3, P = 0.07). Yet, the main effect of time was significant (P < 0.05). The IR strength was significantly higher 6 months postoperatively compared with preoperatively on both shoulders. The LSI at the 60°/s and 180°/s angular velocities were 94.2% and 94%, respectively.

Considering Different Sports Groups

There was no difference in the IR and ER strength changes on the OP and non-OP shoulder between the sports groups (P > 0.05). However, the main effect of time was significant (P < 0.05). Higher IR strength on the OP shoulder was observed in type 2 and 4 sports groups at 6 months postoperatively compared with preoperatively (P < 0.05) while pre and postoperative IR strength were similar in the type 3 sport group (P > 0.05) (Table 2). Shoulder ER strength was similar at the postoperative 6 months compared with the preoperatively at 60°/s and 180°/s angular velocities in all 3 groups (Table 3).

Table 2.

Shoulder internal rotator strength measurements ^a

Internal Rotator Strength
Patient Groups		Preop 60°/ sec	Postop 60°/ sec	Preop 180°/ sec	Postop 180°/ sec	LSI
Type 2	OP	0.57 ± 0.05	0.64 ± 0.06*	0.52 ± 0.07	0.61 ± 0.06*	95.3%
	Non-OP	0.62 ± 0.08	0.67 ± 0.03	0.58 ± 0.08	0.64 ± 0.02*
Type 3	OP	0.55 ± 0.2	0.58 ± 0.15	0.49 ± 0.19	0.56 ± 0.15	90.3%
	Non-OP	0.62 ± 0.11	0.65 ± 0.09	0.61 ± 0.13	0.62 ± 0.08
Type 4	OP	0.61 ± 0.17	0.72 ± 0.18*	0.53 ± 0.16	0.70 ± 0.19*	93.3%
	Non-OP	0.69 ± 0.2	0.75 ± 0.15	0.63 ± 0.16	0.75 ± 0.14*

LSI, limb symmetry index at the postoperative 6 months; non-OP, nonoperated shoulder; OP, operated shoulder; postop, postoperatively; preop, preoperatively.

^aMeasurements were represented after peak torque/body weight normalization.

^*Significantly higher strength compared with preoperatively.

Table 3.

Shoulder external rotator strength measurements ^a

External Rotator Strength
Patient Groups		Preop 60°/ sec	Postop 60°/ sec	Preop 180°/ sec	Postop 180°/ sec	LSI
Type 2	OP	0.27 ± 0.04	0.27 ± 0.04	0.22 ± 0.06	0.23 ± 0.03	79.3%
	Non-OP	0.24 ± 0.06	0.33 ± 0.04*	0.22 ± 0.04	0.29 ± 0.05*
Type 3	OP	0.21 ± 0.07	0.22 ± 0.04	0.18 ± 0.08	0.18 ± 0.05	64.2%
	Non-OP	0.28 ± 0.04	0.32 ± 0.08	0.23 ± 0.06	0.28 ± 0.08
Type 4	OP	0.29 ± 0.1	0.24 ± 0.12	0.28 ± 0.07	0.27 ± 0.08	79.4%
	Non-OP	0.31 ± 0.1	0.25 ± 0.08	0.37 ± 0.07	0.34 ± 0.1*

LSI, limb symmetry index at the postoperative 6 months; non-OP, nonoperated shoulder; OP, operated shoulder; postop, postoperatively; preop, preoperatively.

^aMeasurements were presented after peak torque/body weight normalization.

^*Significantly higher strength compared with preoperatively.

The LSI for the YBT-UQ test was 94.7% and 102.5% for the USSPT. The mean CKCUES test score was 21.8 ± 2.6 touches for all patients. Functional scores for each sport group are presented separately in Table 4.

Table 4.

Functional scores of patients 6 months postoperatively

		YBT-UQ (cm)		USSPT (m)
Patient Groups	CKCUES	OP	Non-OP	OP	Non-OP
Type 2	22.2 ± 4.1	68.6 ± 8.2	71.4 ± 9.3	4 ± 0.4	3.7 ± 0.3
Type 3	21.4 ± 2.2	70 ± 5.2	74.7 ± 6	4 ± 0.5	3.9 ± 0.2
Type 4	21.7 ± 1.7	68.8 ± 6	72.4 ± 5.2	4.3 ± 0.6	4.2 ± 0.6

CKCUES, closed kinetic chain upper extremity stability test; non-OP, nonoperated shoulder; OP, operated shoulder; USSPT, unilateral seated shot-put test; YBT-UQ, Y balance test- upper quarter. cm: centimeter, m: meter.

Shoulder flexion (P = 0.22), abduction (P = 0.18), ER (P = 0.27), and IR (P = 0.14) at the 6 month postoperative timepoint were similar to the preoperative level in all patients. WOSI scores were significantly lower (P < 0.001) at the postoperative 6 months compared with preoperatively. Patient mean disability was 57% preoperatively and 17.3% at 6 months postoperatively based on the WOSI score.

Discussion

The results of the study did not support our first hypothesis in that patients with AACR demonstrated promising results in terms of shoulder muscle strength and functional outcomes at the postoperative 6-month timepoint. The USSPT and YBT-UQ test scores and shoulder IR strength were within the acceptable range. However, shoulder ER strength and CKCUES test score of the patients were below the desired level. The results also did not support our second hypothesis since the patients from different sports types showed similar functional outcomes at the postoperative 6 months. Considering the self-functional assessment, patients still perceived some discomfort and disability after 6 months from the surgery.

Recovery of RC strength is one of the main goals of postoperative rehabilitation in patients with shoulder stabilization surgery. In clinical practice, postoperative RC strength is aimed to reach preoperative levels to allow patients to RTP. However, both the current and previous studies found that patients with shoulder instability also display decreased RC strength on the injured shoulder compared with the uninjured one in the preoperative period.^10,11,29 This shows that even if the RC strength reaches preoperative levels after stabilization surgery, it would still be insufficient. Amoko et al ⁴ reported decreased RC strength on the OP shoulder compared with the non-OP shoulder, although the patients reached the pre-operative level at the postoperative 6 months. In the current study, postoperative RC strength was also found to reach the preoperative level, yet it was still inadequate compared with the non-OP shoulder. Consequently, comparing the RC strength of the OP shoulder with that of the non-OP shoulder may provide more accurate data during the decision-making for RTP. If pre- and postoperative shoulder strength are compared, a higher level of shoulder strength should be expected in the postoperative period.

Previously, only a small number of studies have investigated RC strength after stabilization surgery.^4,37 Both Wilson et al ³⁷ and Amako et al ⁴ reported decreased shoulder ER strength on the OP shoulder compared with the non-OP shoulder at 6 months postoperatively. Park et al ²² evaluated RC strength recovery after shoulder stabilization surgery in professional baseball players. They decided on RTP based on RC strength recovery and delayed RTP until RC strength reached the desired level. It took 8.4 months for RC strength to recover and for their patients to be allowed to RTP. Despite taking a longer time than the average, all but 1 patient returned to play at preinjury level. ²² This signifies the importance of RC strength recovery. In the current study, we also found decreased ER strength on the OP shoulder compared with the non-OP shoulder at the postoperative 6 month timepoint. In contrast, the IR strength of the OP shoulder was at the desired level (similar to that of the non-OP shoulder and higher than the preoperative level). In such cases, where the RC strength was improved but not fully recovered, whether patients are allowed to RTP may depend on their sport type.

Recovery of IR and ER strength was similar among the different sports groups. Regardless of sport type, IR strength recovered 6 months after surgery while ER strength did not. For the lower shoulder injury risk groups (type 2 and perhaps type 3) decreased ER strength can possibly be neglected, but for type 4 sport, which includes overhead and collision sports, IR-ER strength recovery is more important. Increased demanding activities during type 4 sports increases the risk of redislocation. Therefore, having adequate IR and ER strength is necessary to avoid secondary injuries.^15,27 Therefore, RTP can be delayed until having adequate shoulder ER strength, especially in patients performing type 4 sports.

Functional outcomes of the shoulder joint are among other parameters used to determine whether patients are ready to RTP after surgery. ³⁵ Considering the YBT-UQ, similar results should be expected on both sides in the RTP phase.^7,34 With an acceptable LSI score (94.7%) at 6 months postoperatively, our patients demonstrated an acceptable shoulder stability level on the OP side. For the USSPT, patients should have either similar or higher scores on the OP shoulder (depending on the shoulder dominance) compared with the non-OP shoulder. ⁸ Since more than half of our patients were operated on the dominant shoulder, we expected 100% or higher LSI score from the USSPT. Having an average of 102.5% LSI score, patients in the current study also met the expectations in terms of shoulder anaerobic power. In contrast to the YBT-UQ and USSPT, our patients had a score of 21.8 touches in the CKCUES test, which was lower compared with literature reports. Previous studies reported 27.9 and 24.7 touches in overhead and recreational athletes, respectively.^7,31 In contrast, Wilson et al ³⁷ reported that a score of 21 touches in the CKCUES test would be sufficient for RTP after stabilization surgery. The CKCUES test is affected greatly by sport type, competition level, and anthropometric measurements. ³⁰ In addition, population differences could be another factor for diversity among CKCUES test scores. Therefore, it was hard to conclude whether the CKCUES test score of the patients was within the acceptable range. Overall, in parallel with the RC strength recovery, the functional scores of our patients improved considerably and achieved the desired level.

Significant improvements in the WOSI score at 6 months postoperatively indicate that patient perception concerning their shoulder had advanced and they felt more comfortable. Previous studies also revealed increased functional status after shoulder stabilization surgery based on the patient-reported outcome measures assessment.^9,14,19 However, both current and previous studies found that patients still had a small level of discomfort and anxiety.^9,14,19 This could be one of the reasons for having lower scores in some of the functional tests. Therefore, in parallel with the functional status, specific methods addressing the education of the brain should also be included in the rehabilitation program.

The main limitation of the study was the small number of participants and short duration of follow-up. However, this was a prospective cohort study that investigated shoulder function both pre- and postoperatively. Most of the patients did not come for the preoperative assessments, and therefore were not enrolled in the study. Second, although all 3 surgeons had >15 years experience in shoulder surgery and performed similar surgical procedures, there could still be small differences among them. Finally, we could not evaluate the full recovery of the shoulder function since we did not have a chance to assess the preinjury performance of the athletes.

The strength of the present study was assessing shoulder strength and functional status both pre- and postoperatively on both OP and non-OP shoulders.

Conclusion

An acceptable LSI score compared with the non-OP shoulder and a higher functional level compared with preoperatively should be the goal in patients with shoulder stabilization surgery. At 6 months postoperatively, shoulder muscle strength and functional status demonstrated favorable results but did not fully recover. Shoulder IR strength, YBT-UQ, and USSPT scores were at the desired level, whereas shoulder ER strength and CKCUES test score were not. Patients performing low-risk sports can be allowed to RTP yet it should be delayed for those performing type 4 sports.