Brian C. Lau, MD; Carolyn A. Hutyra, MMCi; and Dean C. Taylor, MD
Glenohumeral instability is one of the most common injuries in athletes.1,76 The National Collegiate Athletic Association Injury Surveillance System reported 4080 glenohumeral instability events, with an injury rate of 0.12 per 1000 exposures.1 Glenohumeral instability events in athletes are often traumatic and typically occur following a contact injury but may also occur with noncontact events. These instability events and potential reduction treatments are painful and can cause significant distress to athletes. In addition to significant discomfort, these injuries may affect immediate or long-term athletic performance. Poor self-reported outcomes have been demonstrated with disease-specific health-related quality of life and shoulder function with recurrent dislocations.2,3 However, the motivation to resume sports participation and its timing is a major aspect for athletes seeking care. Therefore, when making treatment decisions in athletes one must understand the timing of the season, risk factors, treatment options, and likely outcomes to optimize return-to-play and function.
Decision making in health care has undergone an evolution in recent decades. Historically, decision making in health care was based on experience, intuition, and largely on expert opinion. In the late 20th century, there was a push toward evidence-based and pattern-recognition models. This expanded our knowledge and allowed the growth of information, including risk factors for shoulder instability. However, it is important to note that all athletes are different, not only in their risk-factor profile but also in their preferences and goals. Of late, there has been a transition in medicine to one of personalization and customization of care. In cancer treatment, this involves genetic testing and treatments based on one’s genetic profile. How do we make this transition in glenohumeral instability? In athletes, customization of care may require consideration of peer or scholarship pressure to finish a season and continue participation in sports despite risk of recurrence of shoulder instability.
This combination of personalized medical evidence, preferences, and an engaging and accessible platform are required to create rich, personalized, and efficient decision-making models. In the case of shoulder instability, personalized medical evidence can provide individual-level outcome probabilities for operative and nonoperative treatment. Based on specific patient factors and high-quality evidence in the literature, decision aids can appropriately model patient outcomes such as individual patient chances of recurrent dislocation, chance of primary and revision stabilization, or chance of maintaining a stable shoulder at various time points.4 Meanwhile, patient preferences provide clinicians with a systematic measurement of utility for each attribute addressed while making these risk-benefit trade-offs.
The shoulder decision-making model validated by Mather et al5 and tested by Streufert and colleagues6 provided a 2-fold output. The model was designed for respondents to receive a summary of their preferences and treatment outcomes on completion of the tool, while physicians could receive the same information along with the patient’s ultimate choice of treatment (operative or nonoperative), patient activities and frequency of activities, patient concerns regarding factors such as pain and scarring, and patient desires to discuss long-term outcomes. This personalized output in a clinical setting can help clinicians differentiate between patients who are more likely to benefit from various operative or nonoperative treatments and help facilitate the conversation surrounding different management options.
Along with advances of decision-model tools, it is critical that physicians have up-to-date knowledge of the factors that influence treatment decisions to proper counsel their athletes. In this review of decision making in surgical treatment of athletes with shoulder instability, we will review the timing of the season, risk factors, treatments and outcomes, and personalized decision-making models.
TIMING OF THE SEASON
The timing of an athletic season in the consideration of management of shoulder instability is unique to athletes. The typical time range for return to sport for those who undergo nonoperative treatment is approximately 2 to 3 weeks.7–9 In comparison, the typical time range for return to sport following surgical management is typically 4 to 6 months.77,78 As such, in-season athletes typically pursue an initial course of rehabilitation, activity modification, and bracing as appropriate. A recurrent instability episode or failure of an athlete to perform sport-specific drills or at preinjury levels tends to be an indicator to pursue in-season operative management.79
Another factor in high school and collegiate athletes may be the timing in their high school or collegiate career. Freshman or sophomore athletes may benefit from earlier operative treatment to allow return to higher levels of performance for their junior and senior years. Conversely, junior and senior athletes may prefer nonsurgical management to allow them to play during collegiate or professional scouting periods or in the final stage of their playing careers.
Owens et al provided an algorithm for in-season decision making for athletes.10 Their recommendation was that in the event of an in-season acute shoulder instability event, athletes should be allowed to attempt to return to play as long as there is enough time left in the season to permit adequate rehabilitation. Their algorithm for in-season decisions is demonstrated in Figure 6-1.10
The primary focus of published literature for decision making in glenohumeral instability focuses on risk factors for recurrent instability to help stratify patients as high risk or low risk. Athletes are unique because by their nature they tend to have 2 of the major risk factors that have consistently been borne out in the literature to qualify as high-risk: young age and high-level of at-risk activities.3,11 The risks of recurrent instability events in this high-risk group have reports of up to 100%.12,13 Other risk factors to consider in athletes include sex, hand dominance, glenoid bone loss, and humeral bone loss. Understanding the role of these risk factors for recurrent instability will help during treatment decisions.
Young age has been identified as one of the strongest risk factors of recurrent shoulder instability14–16 (Table 6-1). Some reports demonstrate up to a 100% redislocation rate in patients younger than 30 years.15 In a meta-analysis of a systematic review of 10 studies, Olds et al17 stratified patients into 2 groups: age 15 to 40 years and older than 40 years. In patients 40 years or younger, there was a 44% recurrence rate compared to 11% in those older than 40 years.17 This was equivalent to a 13.46× increased risk in younger patients. Moreover, the study found that those younger than 30 years had an even higher risk of recurrence at 50%.17 Interestingly, when the older group of patients was stratified into those between ages 41 and 60 years vs those older than 61 years, there was no difference in occurrence rate, 11% vs 10%, respectively.
Younger patients tend to be more active and engage in higher-risk activities. Therefore, it may be expected that adolescent patients would be at an even higher risk of recurrence. Robinson and Dobson18 studied patients age 15 to 20 years and identified a recurrence rate of 86.6% over 5 years. Zaremski et al19 performed a systematic review of 17 studies and stratified adolescents into separate groups: those younger than 14 years and those older than 14 years. They identified a recurrence rate of 42.3% in those older than 14 years, and interestingly found a lower rate of recurrence in those younger than 14 years at 25%.19 It is possible that changes in a young athlete’s physical strength and capabilities during puberty may lead to greater risks.
Lower recurrence rates in patients younger than 14 years may result from higher degrees of flexibility seen in younger patients or lower engagement in high-risk activities. As such, patients younger than 10 years seldom develop symptoms of shoulder instability.20
Recommendation: Primary surgical stabilization in contact or collision athletes in adolescence should be considered.
The physical and mental benefits of sports and competition are plentiful. These include improved cardiovascular health, physical strength, and mental mood, but also grit and perseverance. Sports also may place athletes at risk for injury. Shoulder instability may develop in athletes because of traumatic contact or collisions. The sports with the highest risks include American football, rugby, and soccer but may also include martial arts, boxing, ice hockey, and wrestling.11,21
Leroux and colleagues performed a systematic review of contact or collision athletes with anterior shoulder instability.21 This study evaluated outcomes following arthroscopic surgical management of shoulder instability and identified a 17.8% failure rate in contact or collision athletes.21 This is in comparison to the general population, which has an 8.5% and 8.0% recurrence rate following arthroscopic or open shoulder stabilization, respectively.22
|AGE GROUP, Y||RATE OF RECURRENT INSTABILITY, %|
Overhead activities in athletics or occupation have also been identified as a risk factor for recurrence.23 Sachs et al reported that those who worked with their arms above chest height had a 5.76× higher risk of developing recurrent instability.3
Recommendation: Contact, collision, and overhead athletes should consider operative stabilization following stabilization first-time shoulder dislocation.
Male sex has also been identified as a risk factor for recurrence.16,17,21,24 It is, however, difficult to discern whether sex itself is the cause for greater recurrence or participation in more at-risk activities. American football, for example, which has some of the highest risk of shoulder instability, is primarily played by male athletes. Moreover, even in sports played by both sexes such as lacrosse and ice hockey there are differences in rules and playing style that limit the amount of contact encountered by female athletes.
As such, there are conflicting reports in the literature. Hovelius et al performed a prospective multicenter clinical study with 25 years of follow that demonstrated no significant difference in recurrence of shoulder instability with respect to sex.24 However, a systematic review demonstrated that male athletes had a 46.84% compared to a 27.22% recurrence rate compared to female athletes.17 Interestingly, however, after age 40 years there were similar rates of recurrent shoulder instability following a single traumatic shoulder dislocation: 22% and 25% in men and women, respectively.17 One explanation for these findings is that younger individuals are more likely to participate in high-risk activities.
In the adolescent population, male athletes age 10 to 17 years had a 1.23× greater risk of recurrent shoulder instability compared to female athletes. The same study performed by Leroux and colleagues identified that male athletes age 15 to 17 years participating in contact or collision sports had the highest risk of recurrent dislocation.21
Recommendation: In patients with shoulder instability, sex should be considered along with other risk factors; however, sex should not be considered alone.
Generalized laxity, as measured by Beighton score, plays an important role in the incidence of shoulder instability.25–27 Patients that have greater laxity may have a higher likelihood of instability events, which may manifest as subluxations or true dislocations. As such, Owens et al identified hyperlaxity as a risk factor for recurrent instability.25 Patients with hyperlaxity demonstrate a 2.68× likelihood of experiencing a recurrent shoulder instability compared to those without hyperlaxity. However, these patients are also more likely to cope better with these events because components of hyperlaxity may be hereditary and present since early childhood. These patients may also demonstrate components of multidirectional instability. Patients with multidirectional instability have experience poor outcomes and return to play following surgical management.28 It is important to evaluate for generalized laxity and multidirectional instability because these patients may best benefit from a longer course of physical therapy.28
Recommendation: Generalized laxity and multidirectional instability should be evaluated in athletes. Patients with atraumatic multidirectional instability may benefit from an initial course of physical therapy.
The rate of dislocations in relation to hand dominance has been reported with mixed findings. Lim et al identified no statistical difference in incidence of shoulder instability but a trend toward the nondominant hand, 52.9% compared to 47.1% in the dominant hand.29 (Lim 2018) Conversely, Longo et al30 found the opposite, with 70% of events occurring in the dominant hand vs nondominant hand. However, there is a paucity of literature on the effect of hand dominance on the natural history of recurrence and outcomes in relation to non-operative or surgical management.
Hand dominance, however, may play a role in decision making in shoulder instability following a first-time dislocation to the throwing shoulder. The dominant shoulder in the throwing athletes may be more troublesome with microinstability of the shoulder, which may occur following nonoperative treatment. However, surgical treatment has a risk of loss of external rotation, which may affect an athlete’s throwing mechanics. Throwing athletes must consider whether the risks of recurrent instability or microinstability would be tolerable in their throwing shoulder in context with the risk of loss of external rotation with surgical management.
In these scenarios, the work of Kavaja31 and Plath32 may offer a treatment option. These studies suggest that delaying surgery and waiting to see whether the patient develops instability symptoms after a first-time traumatic shoulder dislocation does not lead to a less-favorable prognosis of instability, quality of life, or glenohumeral joint osteoarthritis.31,32 In throwing athletes, this would allow them to determine whether microinstability develops or whether they experience any adverse effects to their throwing mechanics. If symptoms develop, then surgical management may be warranted.
Recommendation: Throwing athletes may benefit from a trial of nonoperative management following a first-time shoulder dislocation. If symptoms prevent return to the same level of activity, then a thorough discussion should occur regarding risk of loss of range of motion following surgical management.
Glenoid Bone Loss
Part of the pathophysiology of anterior-inferior shoulder dislocations is impaction of the humeral head along the anterior-inferior aspect of the glenoid face.79 This may lead to a bony Bankart, which equates to a bony fracture. After a single first-time dislocation, the occurrence of glenoid rim fracture was found in 22%.33 Interestingly, several studies found that the presence of a large bony Bankart may actually afford a protective effect against recurrent instability (odds ratio, 0.51).34–36 This may result from greater reservation on the part of the physician and the athlete to rush rehabilitation and/or the result of early surgical fixation of the fracture fragment. However, more commonly bone loss is the result of repetitive microinstability or recurrent dislocations, which lead to attritional bone loss.
Recent focus has been on the degree of attritional bone loss and its effect on recurrence and surgical outcomes. The widest portion of the anterior-to-posterior dimension of the glenoid is on average 25 mm. Therefore, attrition of 6 to 8 mm may represent 24% to 32% bone loss. This degree of bone loss can actually change the appearance of the glenoid as described by Burkhart et al.37 This led to the concept of the inverted pear with the glenoid wider superiorly than inferiorly, which disrupted the arc of motion, particularly with abduction and external rotation leading to increased risk of recurrence.
Similarly, Piaseck et al found that osseous defects between 9% and 15% of glenoid width (< 3-4 mm) were usually trivial, whereas defects greater than 20%j to 30% (> 6-10 mm) were significant.38 The authors described that these changes were even more important in patients with high athletic demands. This led to the traditional notion that bone loss greater than 20% to 25% was considered critical bone loss leading to high clinical recurrence and surgical failure.39,40
In 2017, Shin et al identified that even smaller amounts of bone loss may lead to poor surgical outcomes.41 In a study of 169 patients, the study identified that the critical bone loss was 17.3%. Patients with greater than 17.3% bone loss required revision surgery because of recurrence in 42.3% of patients compared to 3.7% in patients with bone loss less than 17.3% of the glenoid.41
The degree of bone loss, particularly in athletes, is critical in treatment decision making. The greater the bone loss, the greater likelihood of recurrence, which should portend to a greater likelihood of surgery. Moreover, significantly higher percentage of glenoid bone loss may be better served with open vs bone block procedures.
Recommendation: Glenoid bone loss should be measured in all cases. Glenoid bone loss greater than 17.3% will benefit from primary stabilization in contact and collision athletes. In athletes with larger than 25% bone loss, primary bone procedure may be considered.
Humeral Bone Loss
In conjunction with glenoid bone loss from anterior-inferior dislocation of the glenohumeral joint is the development of an impaction injury in the posterior-superior aspect of the humeral head. The depression in the humeral head may engage with the glenoid and result in a recurrent instability episode. The incidence of humeral head impression has been shown to range from 70% to 100% for patients with first-time shoulder dislocations.33,42,43 Studies have shown that the presence of a humeral bone impaction or bone loss result in a 1.55× greater likelihood of a recurrent instability episode.27,36 Sekiya and colleagues sought to answer the degree of humeral bone loss that becomes clinically significant.44 In their study, biomechanical testing demonstrated humeral head lesions greater than 25% of the articular surface significantly increased the risk of recurrent instability.44 The authors’ recommendation, therefore, was to treat humeral head lesions greater than 25%.44
Although there are no comparative studies to provide a confident recommendation, treatment of isolated humeral bone loss may be targeted with a remplissage procedure, which is a capsulotenodesis of the infraspinatous tendon into the humeral defect. Lin et al performed a systematic review of remplissage procedures into defects measuring 20% to 40% of humeral volume and identified a wide range of values of return to sport ranging from 56.9% to 100% and return to previous level of sport of 41.7% to 100%.45 (Lin 2018) The authors’ recommendations were that patients with subcritical glenoid bone loss with a humeral head lesion greater than 25% should undergo a labral repair with a remplissage.45 In athletes, one must also consider that the remplissage procedure is nonanatomic and may lead to loss of motion. Depending on the sport, this loss of motion may be unacceptable. Other considerations include bone grafting of the lesion and interaction with any glenoid bony loss-glenoid track. Given the wide range of return to sport, it is difficult to determine when best to recommend a remplissage procedure, but it should be considered.
Recommendation: In athletes with humeral bone loss greater than 25%, primary stabilization with or without remplissage may be considered. One must consider possible consequences of a nonanatomic procedure and loss of motion, which may affect performance in athletes.
Combined Bone Loss—Glenoid Track
Combined bone loss is a frequently encountered scenario because glenoid bone loss incidence is 20% to 70% and humeral head bone loss ranges from 70% to 100% following instability episodes.18 Gowd et al performed a systematic review and found that established critical glenoid bone loss values were not equally relevant in the setting of bipolar bone loss.46 In all treatment options, nonoperative and surgical, there were higher rates of recurrence in the setting of bipolar bone loss.46
Arciero et al47 evaluated the combined biomechanical effects of bipolar lesions using 3-dimensional printing from computed tomography scans of 142 patients with varying degrees of glenoid and humeral head lesions. They defined humeral head lesions of 25% loss of humeral volume as small and lesions with loss of 50% humeral volume as medium. The study found that medium-sized humeral head impaction became clinically significant with as little as 2 mm (8%) of glenoid bone loss.47 This suggests that any humeral head lesion of greater than 50% volume would become clinically significant with any degree of glenoid bone loss.
In 2007, Yamamoto and colleagues introduced the concept of glenoid track to account for the interaction of glenoid and humeral bone loss.48 The glenoid track represents the contact area of glenoid on humeral head as the arm is brought through physiologic range of motion. The glenoid track migrates from inferomedial to superolateral on the humeral head and is defined as the region from the medial margin of the rotator cuff footprint to a point 84% of the width of the native glenoid medially onto the articular surface of the humerus.48 Di Giacomo furthered the concept by identifying humeral head lesions contained within this contact area as on-track and those that had contact outside or medial to the glenoid track as off-track.49 He found that off-track lesions were more likely to engage and result in recurrent dislocation.49
Lau et al evaluated glenoid track in adolescent and adult patients and found that the greater incidence of recurrent shoulder instability in adolescents may be due in part to a higher incidence of off-track lesions compared to adults.50 (Lau 2017) Moreover, patients with 2 or more shoulder dislocations also demonstrated greater incidence of off-track lesions.50
Recommendation: Evaluation of bipolar bone loss and glenoid track should be considered in all athletes. Patients with glenoid off-track lesions may benefit from shoulder stabilization.
TREATMENTS AND OUTCOMES
Several studies have investigated different treatments such as immobilization in a sling, immobilization in external rotation, and surgical treatment.2,9,51–56 These studies suggest there are several possible treatment options that may offer satisfactory outcomes. The major concern for athletes is the rate of return to sport and the level of performance upon return. See Table 6-2 for an overview of recurrence rates and return to play rates following nonoperative, arthroscopic repair, open repair, or coracoid transfer.