Commentary & Perspective

Commentary & Perspective on
"The Outcome and Repair Integrity of Completely Arthroscopically Repaired Large and Massive Rotator Cuff Tears"
by Leesa M. Galatz, MD, et al.

Commentary & Perspective by
Thomas J. Gill, MD*,
Massachusetts General Hospital, Boston, Massachusetts

Arthroscopic Repair of Large and Massive Rotator Cuff Repairs: Reasons for Caution and Optimism

Arthroscopy and sports medicine have been at the forefront of the revolution in minimally invasive surgery. Early advances in arthroscopic techniques focused on the knee. As gains in arthroscopic knee surgery continued, the idea of translating these skills and techniques to the shoulder received considerable interest.

Arthroscopic Bankart repairs were among the first reconstructive efforts in the shoulder. Early results were not good, with some authors reporting a failure rate of more than 50%. Surgeons who were making use of tacks and transosseous sutures were not able to achieve the good and excellent results (a success rate of 90% or more) that Rowe and others reported with the use of open surgery. Early arthroscopic Bankart repair bore little resemblance to the more well-established open technique. The torn labrum was usually fixed "in situ," more medially on the glenoid neck than when performed in the open fashion. Techniques to shift the redundant capsule were not well developed, so instability often persisted. Once these shortcomings were identified and surgeons who were using arthroscopic techniques began to accomplish what is typically accomplished with open procedures (e.g., use of suture anchors placed on the articular surface of the glenoid rim, and shifting of the capsule), the outcomes of arthroscopic Bankart repairs began to approach those of open repairs.

With this in mind, the present paper by Galatz et al. can be evaluated in better context. Between 1997 and 2000, the authors performed 170 complete arthroscopic rotator cuff repairs. In this well-performed study, the authors describe a series of eighteen consecutive arthroscopic repairs of "large and massive" tears, tears measuring greater than 2 cm in diameter and thus involving "two tendons." This definition can be somewhat confusing and can make a comparison with other "large and massive repair" series difficult because a 2-cm tear is not "large." According to Cofield¹, large tears are defined as being 3 to 5 cm in length, and massive tears are longer than 5 cm. Tears that measure 2 cm in length will typically involve the infraspinatus muscle only minimally. Even with this definition, only eighteen of 170 tears fulfilled the criterion, which is less than one would expect at a tertiary referral center. The authors added that indications for surgery included a "failure of" nonoperative treatment of a chronic tear for a period of at least three months," although I would suggest that such a criterion is not only unnecessary but that such a delay may only serve to complicate the attempted repair.

All repairs were performed by a single, experienced shoulder surgeon. Evaluation was performed at twelve and twenty-four months after surgery with the use of the ASES (American Shoulder and Elbow Surgeons) shoulder score as well as with an ultrasound examination. The method of postoperative evaluation was rigorous and well performed. However, it would have been helpful if the authors had ordered an immediate postoperative ultrasound test to be performed at "time zero." By doing so, we could be sure that the later findings did in fact represent "recurrent tears" rather than incompletely repaired tendons at the time of surgery.

In the wake of the current enthusiasm and popular "preference" for arthroscopic surgical procedures, the results of this study are somewhat humbling. Although patients reported an overall improvement in function, pain, and strength, the results seemed to deteriorate as early as two years after the index procedure. The ASES scores fell from 84.6 to 79.9, only nine patients had scores that were >90, and six patients had scores that were less than or equal to 79. Average forward flexion fell from 152° to 142°. More worrisome was the fact that seventeen of eighteen patients had recurrent tears, most of which were equal in size to the original tear itself. While initial inspection of these results is discouraging, I believe that, on closer inspection, there is reason for optimism.

Similar to the early reports on arthroscopic Bankart repairs, some of the failures in this study can be attributed to the described surgical technique. The authors report that they used a simple stitch configuration and an absorbable anchor. Multiple investigators have demonstrated that a simple stitch is a particularly weak configuration for a rotator cuff repair. Simple stitches can be easily pulled through the tendon. Gerber et al.² performed an elegant study of the optimal characteristics for rotator cuff repair, and they reported that a Mason-Allen tendon-grasping technique through bone tunnels with the use of a number-3 nonabsorbable braided suture and cortical reinforcement was optimal. While such a technique is not readily accessible when suture anchors and a closed technique are used, I would suggest that, at the least, a horizontal mattress suture be used for an arthroscopic repair.

Postoperatively, patients were started immediately on a passive range-of-motion program for six weeks, followed by an active and active-assisted program. Three patients were noted to have "less structurally satisfactory" repairs, and so the coracoacromial ligament was not resected. It would be interesting to know if and when the authors would consider opening an arthroscopic repair if they were not completely satisfied with the integrity of the repair.

The authors conclude that arthroscopic repair of the rotator cuff may not yield as strong a repair as traditional open techniques, especially with poor-quality tissue. There is little doubt that the technique used in this report, a simple stitch used in a single row, is not as strong as several commonly used open techniques, such as a medial row of anchors with horizontal mattress sutures in conjunction with a lateral transosseous suture line utilizing Mason-Allen suture technique. With re-tear rates approaching 50% in some open studies³, a double-row repair that does not use simple suture techniques appears to be optimal, whether performed open or arthroscopically.

In trying to decide between an open or arthroscopic repair, there are several factors that deserve consideration. Commonly stated advantages of arthroscopic repairs include less postoperative pain, a decreased risk of deltoid dehiscence, possible accelerated recovery and rehabilitation, and improved cosmesis. I believe that the difference in postoperative pain between the two techniques is overstated. I have personally noticed little difference between arthroscopic and open approaches with regard to postoperative pain. Furthermore, I do not believe that patients who have had arthroscopic repairs can undergo rehabilitation "faster." The technique by which a repair is performed has nothing to do with the biology of tendon healing to bone. If anything, an argument can be made to accelerate the rehabilitation regimen in a patient who has had an open repair because of the strength of the repair at time zero. Regardless of the technique used, the overriding principle when repairing massive tears should be to preserve the deltoid. While this may be a clear advantage to the arthroscopic approach, a careful deltoid-splitting technique rather than a detaching technique can minimize the risk of deltoid injury with open surgery.

To put this study in context, I believe that it is comparable with the early reports of arthroscopic Bankart repairs. It serves as a baseline to which future studies will be compared. The authors should be commended for performing such a careful and detailed investigation. I applaud them for their candor in reporting that seventeen of the eighteen repairs had retorn. Clearly, ultrasound confirmation of the integrity of a repair is a critical element in the assessment of repair outcomes, since subjective assessments alone appear to overstate the results. The question remains as to what degree functional outcomes following rotator cuff surgery are dependent on the integrity of the repair.

The ultimate message of this study is that we should continue to strive to replicate our open techniques when performing arthroscopic rotator cuff repairs. This includes using horizontal mattress or Mason-Allen suture techniques and perhaps double row repairs. By doing so, it is more likely that arthroscopic and open results will become equivalent. Patients in whom a repair fails will take little solace from the fact that the surgical intervention was performed arthroscopically rather than in an open fashion.

*The author did not receive grants or outside funding in support of his research or preparation of this manuscript. He did not receive payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organization with which the author is affiliated or associated.

References

1. DeOrio JK, Cofield RH. Results of a second attempt at surgical repair of a failed initial rotator-cuff repair. J Bone Joint Surg Am. 1984;66:563-7.
2. Gerber C, Schneeberger AG, Perren SM, Nyffeler RW. Experimental rotator cuff repair. A preliminary study. J Bone Joint Surg Am. 1999;81:1281-90.
3. Harryman DT 2nd, Mack LA, Wang KY, Jackins SE, Richardson ML, Matsen FA 3rd. Repairs of the rotator cuff. Correlation of functional results with integrity of the cuff. J Bone Joint Surg Am. 1991;73:982-9.

Copyright © 2004 by the The Journal of Bone and Joint Surgery, Inc.