Return to Play After Arthroscopic Stabilization for Posterior Shoulder Instability-A Systematic Review.

PURPOSE: To ascertain the rate and timing of return to play (RTP) and the availability of specific criteria for safe RTP after arthroscopic posterior shoulder stabilization.
METHODS: Medline, EMBASE, and the Cochrane Library were searched according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines to find studies on arthroscopic posterior shoulder stabilization. Studies were included if they reported RTP data or rehabilitation protocols and excluded if concomitant procedures influenced the rehabilitation protocol. Rate and timing of RTP, along with rehabilitation protocols, were assessed.
RESULTS: This review found 25 studies, including 895 cases, meeting the study's inclusion criteria. The majority of patients were male (82.7%), with an age range of 14 to 66 years and a follow-up range of 4 to 148.8 months. The overall RTP rate ranged from 62.7% to 100.0%, and 50.0% to 100.0% returned to the same level of play. Among collision athletes, the overall rate of RTP was 80.0% to 100.0%, with 69.2%-100.0% returning to the same level of play. In overhead athletes, the overall rate of RTP was 85.2% to 100.0%, with 55.6% to 100.0% returning to the same level of play. Four studies (128 patients) specifically addressed the timing of RTP, and the range to RTP was 4.3 to 8.6 months. Specific RTP criteria were reported in a majority of studies (60%), with the most reported item being restoration of strength (44%).
CONCLUSION: There is a high rate of return to sport after arthroscopic posterior shoulder stabilization, ranging from 4.3 to 8.6 months after surgery. Return to preinjury level is higher for collision athletes compared with overhead athletes. However, there is inadequate reporting of RTP criteria in the current literature, with no clear timeline for when it is safe to return to sport. LEVEL OF EVIDENCE: IV, systematic review of level II to IV studies.

Posterior shoulder instability (PSI) represents an estimated 2% to 10% of all shoulder instability cases.1, 2, 3, 4, 5 PSI often presents insidiously, as the most common symptom is shoulder pain and not an acute instability event. Repetitive microtrauma to the shoulder, atraumatic causes including baseline ligamentous laxity, and acute traumatic events have been proposed and suspected in PSI.,6, 7, 8 PSI is commonly seen in athletes who perform high-demand, dynamic posterior-loading activities, such as weightlifters and American football linemen, as well as athletes in sports that reward increased shoulder range of motion, including swimming and gymnastics. It is occasionally seen in the young military population.,,8, 9, 10, 11, 12 PSI may be treated conservatively or operatively, with management dictated by patient factors (age, compliance, comorbidities), evidence of any bony pathology involving the glenohumeral joint, and mechanism of instability.,

Burkhead and Rockwood found nonoperative management to be less successful in patients with a history of a traumatic event, as 16% of patients with a traumatic event had clinical success as opposed to 70% to 89% of patients without a traumatic event. Surgery is commonly considered when patients continue to experience instability symptoms and fail nonoperative treatment.,, However, although surgical correction for PSI has provided improved clinical outcomes, the rates of return to play (RTP) for PSI are unclear.,,16, 17, 18, 19, 20 Additionally, there is no consensus for guidelines or criteria on how to allow patients to return to play safely after posterior shoulder stabilization.

The purpose of this systematic review is to ascertain the rate and timing of RTP and the availability of specific criteria for safe RTP after arthroscopic posterior shoulder stabilization. The hypothesis is that there is a significant rate of RTP after arthroscopic posterior stabilization but substantial differences in RTP protocols.

Methods

Search Strategy and Study Selection

To collate the existing evidence related to rehabilitation protocol and RTP after posterior shoulder stabilization, a systematic review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Eligibility criteria for this review consisted of clinical studies of patients undergoing arthroscopic posterior shoulder stabilization in which rehabilitation protocols or RTP data were reported. Further study characteristics required for eligibility included publication in a peer-reviewed journal and availability of the full text of the study. Studies were deemed ineligible for this review if they included concomitant procedures that explicitly influenced the rehabilitation protocol or patients with multidirectional laxity. Review articles, case reports, surgical technique articles, biomechanical studies, and non–English language articles were not included.

Screening, eligibility, and inclusion were determined by 2 independent reviewers (J.W.F. and E.T.H.). The Medline, EMBASE, and Cochrane Library databases were searched from their inception to December 20, 2019, using the following algorithm: posterior AND shoulder instability AND (arthrosc∗). The titles and abstracts of returned results were screened according to the described eligibility criteria (Fig 1). Potentially relevant studies were identified, and the associated full text was reviewed. The reference lists of all relevant studies were screened for any articles not identified through the database search. Studies were included based on agreement of both independent reviewers. Any instances of disagreement were resolved through consultation with the senior author.

Fig 1

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow chart.

Assessment of Level and Quality of Evidence

The level of evidence (LOE) of the included studies was evaluated based on the criteria from the Oxford Centre for Evidence-Based Medicine. The quality of studies was assessed based on the criteria from the Modified Coleman Methodology Score as performed by Ramponi et al. The methodological quality of the RTP guidelines described in each study was assessed based on the criteria outlined by Zaman et al. Instances of scoring discrepancy were resolved through consultation with the senior author to reach a consensus.

Data Extraction

Two reviewers independently extracted data in duplicate from the included studies using a predesigned data collection form. Screening was performed to remove duplicate patients between studies. Demographic variables included total number of patients, sex ratio, patient age, and follow-up time. Variables related to RTP were recorded, including percentage of patients returning to play, ability to return to the preoperative level of play, patient-reported timing of return to athletic activity, and reasons for not returning to sport.

Statistical Analysis

Statistical analysis was performed using SPSS (release 2013; IBM SPSS Statistics for Macintosh, version 22.0. Armonk, NY). Descriptive statistics were determined for all categorical and continuous variables. Categorical variables were reported as frequencies with percentages, and continuous variables were reported as a weighted mean with an estimated standard deviation. The quality of RTP for each study was determined according to the quality of the RTP guidelines outlined by Zaman et al. For all analyses, P values <.05 were considered to be statistically significant.

Results

Initially, 1303 studies were identified. After removal of duplicates and non-English studies, 794 studies were further analyzed. After application of inclusion and exclusion criteria, 25 studies reporting on 996 shoulders were included in the final analysis (Fig 1).

Study Characteristics

Overall, 25 clinical studies (LOE II, 3; LOE III, 6; LOE IV, 16) reported RTP rates, including 996 shoulders. There were 740 males and 155 females, with ages ranging from 14.8 to 66 years, who were followed up for 4 to 148.8 months (Table 1).

Table 1

Study Characteristics

Reference	LOE	QOE	Patients	Age (y)	Male sex (%)	Follow-up (mo)
Andrieu et al. 201725	III	63	101	28.8	74.2	49.7
Arner et al. 201526	IV	79	56	17.9 (14.8 to 25.5)	100	44.7 (24 to 98)
Bahk et al. 201027	IV	62	29	26.3 (18.3 to 43.4)	97	66 (24 to 148.8)
Badge et al. 200912	IV	87	11	24.8 (15 to 36)	100	32 (17 to 54)
Bradley et al. 200618	II	92	100	23.3 (15 to 61)	77	27.7 (12 to 77)
Bradley et al. 201328	II	72	200	24.3 (15 to 65)	79	36.7 (12 to 115)
Castagna et al. 200729	IV	63	9	25.2	77.8	34.2 (28 to 39)
Eckenrode et al. 200930	IV	65	5	20.2 (18 to 22)	100	NR
Garret et al. 201731	III	49	25	30 (16 to 45)	64	NR
Hines et al. 201837	III	52	32	30.8 (20 to 47)	93.8	53.7 (25 to 82)
Katthagen et al. 201733	III	69	38	27.6 (13 to 66)	92.1	49.2 (24 to 93.6)
Kercher et al. 201935	IV	76	32	20.5 (16 to 41)	100	41.58 (24 to 92)
Kim et al. 200336	IV	72	15	21 (17 to 25)	93	39 (31 to 47)
Kraeutler et al. 201834	III	74	22	26.6 (17 to 45)	95	43.2 (26.4 to 88.8)
Lacheta et al. 201932	IV	72	7	23.5 (17 to 43)	100	96 (36 to 120)
Lenart et al. 201219	IV	70	19	21.4 (15 to 33)	81	36 (12 to 67)
Mair et al. 19989	IV	73	9	18.8 (16 to 21)	100	30 (24 to 42)
McClincy et al. 201538	III	77	48	17.8	71	37 (12 to 97)
McClincy et al. 202039	IV	80	68	17.2 (14 to 19)	NR	45
Papendick and Savoie 199520	IV	62	41	23 (15 to 42)	NR	10 (4 to 41)
Radkowski et al. 200840	II	90	98	22.9	76.5	27
Robins et al. 201741	IV	58	42	NR	NR	39.6
Wanich et al. 201242	IV	68	12	20.3 (16 to 33)	100	33.6 (18 to 64)
Wolf and Eakin 199843	IV	60	5	26 (14 to 54)	79	33 (24 to 45)
Wooten et al. 201544	IV	74	22	17.3	86.30	63 (24 to 115)

Data for age and follow-up are mean (range).

LOE, level of evidence; NR, not reported; QOE, quality of evidence.

Rate and Time of RTP

Rates of return to play were reported in 25 studies. The overall rate of RTP was 62.7% to 100%; in 18 studies, 50.0% to 100% returned to the same level of play. Four studies (128 patients) reported time of return to play, which was 4.3 to 8.6 months (Table 2).

Table 2

Rate and Time of Return to Play

Reference	Return to Play (%)	Return to Play at Same or Higher Level (%)	Time (mo)
Andrieu et al. 201725	62.7	NR	7.7
Arner et al. 201526	92.9	78.6	NR
Bahk et al. 201027	100	100	4.3
Badge et al. 200912	84.6	65.4	NR
Bradley et al. 200618	89.0	67.0	NR
Bradley et al. 201328	90.0	63.5	NR
Castagna et al. 200729	100	100	NR
Eckenrode et al. 200930	80.0	80.0	NR
Garret et al. 201731	71.4	NR	NR
Hines et al. 201837	87.5	NR	NR
Katthagen et al. 201733	78.6	78.6	NR
Kercher et al. 201935	93.8	62.5	NR
Kim et al. 200336	100	NR	NR
Kraeutler et al. 201834	68.2	50.0	8.6
Lacheta et al. 201932	100	100	NR
Lenart et al. 201219	100	100	NR
Mair et al. 19989	100	100	NR
McClincy et al. 201538	85.4	60.4	NR
McClincy et al. 202039	89.0	76.8	NR
Papendick and Savoie 199520	100	NR	NR
Radkowski et al. 200840	88.8	73.5	NR
Robins et al. 201741	91.9	NR	NR
Wanich et al. 201242	91.7	91.7	5.9
Wolf and Eakin 199843	100	NR	NR
Wooten et al. 201544	80.0	68.0	NR

Abbreviation: NR, not reported.

Rate of RTP Among Collision and Overhead Sports

Among collision athletes, the overall rate of return to play was 80% to 100%, with 69.2% to 100% returning to the same level of play. In overhead athletes, the overall rate of return to play was 85.2% to 100%, with 55.6% to 100% returning to the same level of play (Table 3).

Table 3

Rate of Return to Play in Collision and Overhead Athletes

Reference	Return to Play (%)	Return to Play at Same or Higher Level (%)	Time (mo)
Collision athletes
Arner et al. 201526	92.9	78.6	NR
Badge et al. 200912	100	100	4.3
Bradley et al. 200618	86.3	74.5	NR
Bradley et al. 201328	90.6	69.2	NR
Castagna et al. 200729	100	100	NR
Eckenrode et al. 200930	80.0	80.0	NR
Kim et al. 200336	100	NR	NR
Lacheta et al. 201932	100	100	NR
Lenart et al. 201219	100	100	NR
Mair et al. 19989	100	100	NR
Robins et al. 201741	91.9	NR	NR
Wolf and Eakin 199843	100	NR	NR
Overhead athletes
Kercher et al. 201935	93.8	62.5	NR
Kim et al. 200336	100	NR	NR
Lenart et al. 201219	100	100	NR
McClincy et al. 201538	85.4	60.4	NR
Papendick and Savoie 199520	100	NR	NR
Radkowski et al. 200840	85.2	55.6	NR
Wanich et al. 201242	91.7	91.7	5.9
Wolf and Eakin 199843	100	NR	NR

Abbreviation: NR, not reported.

RTP Criteria

Overall return-to-play criteria were reported in the majority of studies (60%), with the most commonly report item being restoration of shoulder strength (44%). There was a small discrepancy in reported time of return, with 9 studies (36%) reporting return at either 4 or 6 months, 6 months being the most commonly used time point (24%). Other criteria including range of motion (ROM) (40%), absence of pain (20%), completion of sport-specific rehabilitation (16%), and proprioception (4%) were also reported. The mean RTP quality of evidence was 1.8 (range 0 to 4) (Table 4).

Table 4

Return-to-Play Criteria

Overall	15 (60)
Strength	11 (44)
Range of motion	10 (40)
Time	9 (36)
Pain	5 (20)
Sport-specific rehabilitation	4 (16)
Proprioception	1 (5)

Data are n (%).

Discussion

The most important finding from this systematic review of the literature was a high rate of return to sport after arthroscopic posterior shoulder stabilization, with the majority of patients returning to the same level of play. Both collision and overhead athletes returned to play at high rates, but one third of overhead athletes were unable to return to their preinjury status. However, there is inadequate reporting of RTP criteria in the current literature. Additionally, it remains unclear in the literature when it is safe to return to play, with appreciable variations in when athletes could return. Time, strength, and ROM were the most commonly reported criteria for RTP, but a few studies used sports-specific rehabilitation protocols.

Return to play is a key and desirable outcome after orthopaedic procedures. Healthy, active patients place great importance on returning to sports and returning to the same (or higher) level as preinjury. This systematic review demonstrated that RTP was achievable at a high rate after posterior shoulder stabilization.,,,25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44 Bradley et al. evaluated 200 shoulders, diagnosed with unidirectional recurrent posterior shoulder instability, and noted that 90% returned to play. Both McClincy et al. and Radkowski et al. compared surgical outcomes and RTP rates in throwing and nonthrowing athletes. McClincy et al. reported that 86% of the athletes were able to return to play, with no significant difference between the throwing and nonthrowing athletes. Radkowski et al. reported that throwing and nonthrowing athletes returned to play at 85% and 91%, respectively. These reported findings indicate that athletes of varying sports-specific motions return to play at high levels after posterior shoulder stabilization. However, it is still unknown whether these players were able to sustain their activity level, or what the impact of returning to play had on the durability of their posterior shoulder stabilization. It is worth mentioning that the reported rates compare well to other commonly performed sports surgeries such as anterior shoulder stabilization, anterior cruciate ligament reconstruction, and medial patellofemoral ligament surgery.,45, 46, 47

Although several studies reported high rates of RTP, the number of athletes who return to the same or higher level varies. Studies by Bradley et al. and Bahk et al. reported similar RTP rates. However, return to the same or higher level rates was ∼20% lower, suggesting excellent results after surgery but room for improvement, and future research with regard to returning with the same ability before the athlete’s injury., A study in 2009 documented the outcomes and RTP rates for 5 collegiate Division I wrestlers after posterior shoulder stabilization. Four of the 5 athletes returned to their preinjury status (the fifth graduated from college), including 2 achieving All-American honors. Although they make up a small sample size, these wrestlers were able to recover fully and perform at an elite level. Lacheta et al. followed athletes after arthroscopic posterior bony Bankart bridge repair technique and found all 9 athletes returned to play and achieved their previous athletic level. The type of sport had no significance on the outcomes or the ability to return to the same or higher level of performance among this cohort.

Several studies reported data on overhead and collision subcohorts, determining whether the physical motion of the shoulder or the type of sport had an impact on RTP rates and return to preinjury status.,,,,,,, Our systematic review revealed that overhead and collision athletes returned to play at similar rates, but collision athletes were more likely than overhead athletes to return to the same or higher level. A study in 2015 assessed 56 American football players, concluding that 93% of the players successfully returned to sport. However, only 79% were capable of returning to their preinjury status. Another study in 2015 compared nonthrowing athletes to overhead throwing athletes, documenting similar RTP rates. Although overhead throwing athletes returned to play at a rate of 85%, only 60% were able to return to the same or higher level. The discrepancy in return to preinjury status between collision and overhead athletes suggests the need for validated RTP criteria.

Most of the selected studies reported general criteria for RTP, with most reporting time, ROM, and strength as important components. The majority of the studies allowed RTP at 6 months. Developing a verified RTP guideline may potentially help decrease the rate of recurrence of instability. Several studies reported the rate of recurrence within each respective cohort after posterior shoulder stabilization. Bahk et al. and Hines et al. reported 3.4% and 6% recurrence rates, respectively. Bradley et al. prospectively reviewed 200 patients, and 14 of the 200 required arthroscopic revision. A systematic review conducted in 2018 defined RTP criteria after operative stabilization for traumatic anterior shoulder instability in hopes of reducing the recurrence rate. Much like the Hurley et al. systematic review that reported RTP rates after anterior shoulder stabilization, creating a validated checklist for a safe RTP after posterior shoulder stabilization would help improve surgical outcomes, decrease recurrence rates, and help close the gap between RTP and rates of return to the same or higher level.

Several factors have been postulated and correlated with athletes who are unable to return to play, such as poor healing, timing of life events (graduation from high school or college), and loss of interest in the injury-causing sport. Bahk et al. evaluated clinical outcomes and attempted to identify predictors of success after arthroscopic posterior Bankart reconstruction, concluding that 96.6% of the cohort reported success, with 84.6% returning to sports. Of the 8 athletes who did not return to sports, 75% documented that it was because of their shoulder. A 2015 study evaluated American football players after arthroscopic stabilization of PSI, with a majority returning to sport. However, the study team reported that some players returned at a different level or not at all because of injuries other than PSI or because they chose to partake in a different sport. Military personal were evaluated for the presence of posterior glenoid bone loss after arthroscopic isolated stabilization of the posterior labrum, investigating its impact on return to duty, complications, and surgical outcomes. Patients were separated according to mean posterior glenoid bone loss: <13.5%, considered minimal loss, and >13.5%, considered subcritical. Patients with >13.5% bone loss were less likely to return to full duty compared with those with minimal bone loss, 14.3% versus 8%, respectively. Baseball players who underwent arthroscopic posterior labral repair were assessed for surgical outcomes and described a large RTP rate. However, pitchers were less likely than position players to return to preinjury levels (41% versus 86%). This suggests that it may be harder to return to certain positions in baseball and potentially other major sports after PSI.

Limitations

Systematic reviews are vulnerable to limitations, including possible biases in the selected studies. Specific variables restrict the conclusions made from this study, including the retrospective composition, low level of evidence, and irregular reporting of RTP information. Because of limitations in the included studies’ reporting, we were unable to analyze whether demographic factors or dominance of shoulder were potential risk factors for inability to RTP and return to the same or higher level. Additionally, we were unable to determine whether concomitant pathologies hindered an athlete’s ability to return to play or return to their preinjury status. A few studies reported a mean follow-up <12 months, questioning the potential effect of a return at 4 months on overall outcomes and a return to athletic activities. Lastly, the long-term effects of loss of ROM and overconstraint could not be assessed given the short- and medium-term nature of the reviewed studies.

Conclusion

There is a high rate of return to sport after arthroscopic posterior shoulder stabilization, ranging from 4.3 to 8.6 months after surgery. Return to preinjury level is higher for collision athletes compared with overhead athletes. However, there is inadequate reporting on RTP criteria in the current literature, with no clear timeline for when it is safe to return to sport.