Open versus Arthroscopic Repair for Shoulder Instability: What’s Best?

Published on 11/03/2015 by admin

Last modified 11/03/2015

Print this page

This article have been viewed 1015 times

Chapter 99 Open versus Arthroscopic Repair for Shoulder Instability: What’s Best?

ROBERT. LITCHFIELD, MD, FRCSC, FINTAN. SHANNON, MD

Shoulder dislocations are a common sports injury, especially in contact sports such as hockey,^1,² rugby,^3,⁴ and all forms of football,^5,⁶ but also in sports associated with falls, such as skiing,^7,⁸ and in sports in which shoulder movements are integral to the game, such as tennis, badminton, swimming, and baseball.^3,^{9, 10} Most of these dislocations involve anterior dislocation,^3,^7,^8,¹¹^–¹⁵ during which the glenohumeral joint actually is disrupted. In some, the dislocation reduces itself spontaneously.¹⁶ Other patients require manual reduction, often requiring anesthesia. Immediate conservative management after reduction may consist of a sling, more to alleviate pain and worry than to protect the joint, and avoidance of sporting activities for an extended time, possibly associated with analgesics, physiotherapy, or both; but the research on the acute management of shoulder dislocations is quite inconclusive,¹⁷ and there is extreme variability in clinical practice.¹⁸^–²⁰

Irrespective of how shoulder dislocations are managed immediately after injury,²¹ the risk for redislocating a shoulder is high without surgical repair, in some series approaching 90%, especially in younger athletes.^2,²¹ Conservative management, including the use of a sling in the short term and physiotherapy, is of uncertain benefit²¹ and may actually increase this risk by delaying more definitive treatment.²² Consequently, surgery commonly is considered, especially for those who are young and those wanting to return to athletic activities.

Currently, the surgical options can be classified broadly into open surgical repair and arthroscopic repair; however, ongoing debate rages regarding which is most effective for the treatment of shoulder dislocations, which has led to two meta-analyses and a further published review of the literature.²³^–²⁵ Certainly, early investigators seemed to identify either no difference between the approaches or some superiority of open surgery, but this conclusion largely was based on a single outcome: joint stability. Despite these early results, the arthroscopic repair of shoulder instability seems to have gathered momentum since the late 1990s.^24,²⁵ This chapter is a review of the relatively limited comparative evidence, examining not only joint stability, but also several other outcomes of clinical interest. Note that these comparisons all are relatively recent, the first publication having been in 1996.²⁶

SURGICAL OPTIONS

As stated earlier, the surgical options can be divided broadly into those performed via open surgery and those performed arthroscopically, but there is a variety of techniques that can be utilized within each approach. This especially appears to be true with arthroscopic shoulder repair, possibly because this approach is newer, and hence more in its evolutionary phase.^24,²⁵ Some of the initial arthroscopic approaches included staple capsulorrhaphy, transglenoid suturing, bioabsorbable tack fixation, suture anchor fixation with capsular plication, and the combined use of intra-articular and extra-articular Suretac sutures.²⁶^–⁴¹ More recently, most surgeons utilize a suture anchor system with or without knot tying. This variety, in itself, makes it difficult to compare between studies, and between the arthroscopic and open surgery approaches.

OUTCOMES OF INTEREST

Because there has been variability in the surgical approaches used, there also has been considerable variability in the outcomes measured. Virtually all investigators who have assessed open or arthroscopic repair alone or relative to each other have utilized recurrent instability as an outcome, and often as the primary outcome. However, how recurrent instability is defined has differed, some including a positive apprehension sign as an indicator of instability. No other single outcome has spanned all studies, but several investigators have adopted the use of validated multidimensional instruments, especially the shoulder-specific (joint-specific) evaluation instruments called the Rowe Rating Scale,^26,^29,^31,^33,³⁵^–³⁸ ^,40 which initially was called the Rating Sheet for Bankart Repair, and the Constant Score.^28,^{31, 33, 35, 37, 38} Other shoulder-specific summation scores that have been utilized are the modified Rowe Rating Scale,²⁸ the American Shoulder and Elbow Surgeons (ASES) Standardized Shoulder Assessment Form,^39,⁴¹ and the University of California Los Angeles (UCLA) Shoulder Rating Scale.^26,³⁶ Kirkley and colleagues’⁴³ Western Ontario Shoulder Instability (WOSI) scale, another instrument used in one of the comparative studies,⁴² offers the advantage of being a disease-specific, rather than just a joint-specific, instrument, in that it specifically evaluates shoulder function in patients with shoulder instability. Like the Rowe and Constant scales, each of these instruments generates a summation score, often expressed as a percentage, which is derived from evaluations of individual parameters, such as pain, function, range of motion (ROM), strength, and stability, though which parameters are included, and how they are weighted varies from scale to scale. Some investigators also have compared the open and arthroscopic approaches with respect to the change in summation score from pre-operative baseline to final evaluation.

Another summation score that has been used is the 36-Item Short Form Health Survey (SF-36),²⁷ an instrument with several subscales to evaluate self-perceived patient well-being. The SF-36 is not specific to any joint or to any 1 medical or surgical condition; as such, it allows for comparison between conditions, which can be useful if one wishes to determine the relative worth of treatment of 1 condition versus another.

More individual outcomes of interest include the apprehension sign; the requirement for further surgical procedures on the involved shoulder; joint ROM, especially external rotation when the shoulder is abducted and forward flexion; loss of joint ROM, relative to presumed premorbid baseline or “normal” ROM; pain severity, for example, as rated on a visual analogue scale; the rate and time to return to athletic activities; the rate and time to return to prior athletic activities; the rate and time to return to contact or so-called collision athletic activities; and radiographic evidence of cystic change or drill holes. No article has reported on all these outcomes, or even the majority, again making interstudy comparisons difficult.

REVIEW OF OUTCOMES

Instability

As shown in Table 99-1, the earliest studies comparing arthroscopic repair of shoulder instability with the long-established open approach demonstrated superiority of the latter^26,^{29, 40} (Level III). For example, among 12 patients undergoing open repair and 15 arthroscopy, Geiger and coworkers²⁹ identified a 44% rate of instability among those who had undergone arthroscopy but no instances of instability in those who had had open surgery (P < 0.05). Similarly, in the studies by Guanche and researchers²⁶ (Level III) and Steinbeck and Jerosch⁴⁰ (Level II), the rates of instability were 33% and 17% in those who had had arthroscopy versus 8% and 6% in the open surgery group, respectively. Though clinically appreciable, these latter 2 differences did not achieve statistical significance. Nonetheless, these 3 studies were among the 6 included in the meta-analysis published by Freedman and colleagues²⁴ in 2004 and the 11 studies included in the meta-analysis performed by Mohtadi and coworkers²⁵ in 2005, both of which conclude that open is superior to arthroscopic repair, in terms of instability rates (Level III).

TABLE 99-1 Published Studies Comparing Open versus Arthroscopic Repair of Shoulder Instability

Later studies have been much less consistent. A few studies have uncovered some advantage of open surgery, such as Rhee and colleagues ³⁷ (25% arthroscopy vs. 13% open; P < 0.05), Hubbell and coworkers³⁰ (60% vs. 0%; P < 0.05), Cole and coauthors²⁷ (24% vs. 16%; P value not significant [NS]), Karlsson and researchers³³ (15% vs. 10%; P value NS), and Sperber and investigators³⁸ (23% vs. 13%; P value NS) (Level III). In contrast, several other studies revealed no difference at all between the 2 surgical approaches,^28,^{31, 35, 36, 39, 42} and 1 study actually demonstrated a clinically impressive, albeit not statistically significant, difference in favor of arthroscopy (6% arthroscopy vs. 24% open; P value NS) (Level III). In some instances, no cases of postrepair instability were identified in either group.^28,^{35, 39}

As stated earlier, 1 problem interpreting these data is that instability has not been consistently defined. For example, Kartus and colleagues ³⁵ limited the designation of instability to shoulder dislocation, which clearly might have reduced the numbers this group considered unstable versus other investigative teams that included a positive apprehension sign. Nonetheless, what is apparent is that it is less clear now that open repair provides a lower risk for instability than it may have been 1 decade ago, possibly because of improved arthroscopic techniques and equipment, and increased experience among surgeons. It also is apparent, and quite understandable, that the results of arthroscopy are quite operator dependent.