Double-row versus Single-row Cuff Repair*
Stay away from a man who’s all gurgle and no guts.
This book is intentionally oriented toward surgical techniques, emphasizing tips and tricks. In general, we believe it is more important to discuss techniques than controversies. However, a controversy has recently developed in the realm of arthroscopic rotator cuff repair to which we believe we should devote a chapter. That controversy revolves around single-row versus double-row arthroscopic rotator cuff repair. Two recently published clinical studies, one by Franceschi et al. (1) and one by Burks et al. (2) concluded that single-row techniques offered the same clinical and anatomic outcomes as double-row techniques. We believe these conclusions are erroneous.
First of all, one must recognize that contemporary state-of-the-art double-row repair differs from earlier-generation double-row techniques. The current suture-bridge technique (or transosseous-equivalent technique) (3,4) has bridging self-reinforcing sutures between the medial and the lateral rows. The first-generation double-row repair construct consisted of medial mattress sutures and lateral simple sutures without linkage between the two rows (5). The first-generation double-row constructs had superior biomechanical parameters compared to single-row constructs (6,7). However, they did not adequately resist rotational effects nor did they distribute forces across the entire rotator cuff insertional footprint.
Biomechanical testing has consistently demonstrated the superiority of a linked double-row construct (such as SutureBridge) to both unlinked double-row and single-row constructs. Currently, testing techniques that emphasize cyclic loading as well as resistance to shear, show that linked double-row constructs are the strongest.
We recently did a biomechanical study (8) on a linked double-row construct that showed that it possesses self-reinforcing properties similar to the Chinese finger trap. The self-reinforcing features of the linked suture bridge indicate that in vivo destructive forces can be neutralized and even harnessed to make the construct stronger under load.
It is ironic that we find ourselves in the awkward position of having to rationalize the use of a repair construct that is biomechanically superior to other constructs and that has been shown in several studies to improve tendon-to-bone healing (9–11). However, the aforementioned level I studies by Franceschi et al. (1) and by Burks et al. (2) have cast some doubt on the clinical superiority of linked double-row self-reinforcing constructs. Therefore, it is essential that we critically analyze these two studies to see if that doubt can be justified.
The study by Franceschi et al. (1) utilized a double-row repair technique with medial mattress sutures and lateral simple sutures, a construct devoid of suture linkage or bridging between the two rows. This first-generation double-row suture anchor repair technique, as previously discussed, embodies an inferior mechanical profile compared to the contemporary double-row linked bridging technique. The authors reported the use of an average of 1.9 suture anchors for single-row repairs and 2.3 suture anchors for double-row repairs. In general, one would expect that the double-row repairs would have had twice the number of suture anchors as the single-row repairs. However, in this study, the average number of suture anchors were nearly identical and it becomes difficult to differentiate these as truly differing repair techniques. As this is a level I study, the authors should be commended for conducting it in a randomized and controlled fashion. However, it suffers from a fatal flaw and essentially prevents one from deriving a valid conclusion based upon the results. Although the authors concede that a limitation to their study is that no formal power analysis was performed, reporting on 26 patients in each group provides a study that is clearly grossly underpowered when healing is the primary outcome variable (12). Thus, it becomes, even in the best of procedure comparisons (i.e., a true first-generation double-row compared to a single-row repair technique), impossible to conclude that differences do not exist between these two populations of patients. The study offers little to our knowledge or ability to compare a single-row to a double-row technique and does not speak in any way to the merit of the self-reinforcing linked bridging suture technique.
As for the study by Burks et al. (2), the repair technique was not a bridging double-row construct nor a true double-row technique as it was initially described, but rather, a “triangle” repair with one medial anchor and two lateral anchors. Mazzocca et al. (14) have previously tested this “triangle” configuration and found its strength to be equivalent to single-row repairs. So, it is no surprise that this older generation nonbridging double-row repair construct was not superior to the single-row technique. Once again, these authors should be commended for conducting a level I randomized controlled trial. But, similar to the study by Francheschi et al. (1), the authors presented a grossly underpowered study (12) whereby valid comparisons of these techniques cannot be made. In essence, the authors predicted a 20% retear rate, yet only demonstrated a 10% retear rate in both groups and thus committed a type II error whereby 20 patients analyzed in each group cannot render a valid comparison of two surgical techniques. To summarize, while both of these studies are often quoted as substantive evidence that there are “no differences between single-row and double-row techniques,” they both suffer from an inability to make valid comparisons because they are grossly underpowered in addition to utilizing older-generation nonbridging repair constructs for their double-row repairs. Making the argument that differences do not exist between these repair techniques is no different than the historical argument that arthroscopic suture anchor repairs for instability should not be done because the earlier arthroscopic transglenoid instability repairs had very high failure rates.
Furthermore, multiple studies have shown that tendon-to-bone healing occurs in significantly greater numbers of patients who have had double-row cuff repairs than in those with single-row repairs (9–11). In the systematic review of the existing literature performed by Duquin et al. (11) that included more than 1,100 rotator cuff repairs, the authors demonstrated a statistically significant reduction in anatomic retear rates for true double-row repairs compared to single-row repairs for all tears >1 cm in length. While this study supports what some might believe to be a self-evident finding considering the biomechanical comparisons of these techniques, this review does not separate or discuss the results of the self-reinforcing bridging suture technique.
Recently, at the Closed Shoulder and Elbow Meeting (New York, NY, November, 2009), Gartsman et al. (14) presented a Level I prospective randomized comparison using ultrasound evaluation of single-row repairs of isolated supraspinatus tendon tears compared to the suture-bridge transosseous repair technique, a self-reinforcing bridging suture technique. The authors performed a power analysis and determined that for an 80% power, assuming a 12% difference between groups with p = 0.05, that 50 patients in each group would suffice. The authors used two anchors medially and two anchors laterally, with bridging sutures linking the two rows of anchors, and demonstrated that the single-row retear rate was 20% versus the transosseous equivalent suture-bridge technique that was 6%, a highly significant difference between the two groups. The limitations of this study include that it is not yet formally published and no functional outcomes were reported.
A consistent problem with existing literature is that aggregate scores (University of California at Los Angeles [UCLA], American Shoulder and Elbow Surgeons [ASES], Constant) and outcomes do not separate out strength as a primary outcome variable. Arguably, comparisons in strength following different repair techniques might yield additional findings not yet considered by current literature. We believe that the magnitude of improvement in external rotation strength and forward elevation strength after self-reinforcing suture-bridging double-row repair has been greater than that seen previously with single-row techniques. Most recently, Cole and associate authors have demonstrated a statistically significant improvement in forward elevation strength as used to, in part, calculate the Constant score when comparisons were made between double-row and single-row repair techniques (personal communication, November 2009).
Let’s take this line of reasoning a bit further. We have long known that arthroscopic debridement and decompression for the treatment of rotator cuff tears lead to significant pain relief and improvement in postoperative UCLA scores even in the face of little or no gain in strength. But is that level of functional improvement the benchmark that we want for evaluating our rotator cuff tears? Surely not!
In our opinion, the great advantage of cuff repair over cuff debridement is the ability to improve strength, yet strength is not properly (or at all) weighted in our current scoring systems. In light of this, we are currently evaluating the relative importance of strength return to patients in the outcome following rotator cuff repair and are developing a new outcome tool that adequately addresses strength by quantifying postoperative gains in strength. This is the only way that we can assess the clinical improvement that is directly attributable to tendon healing with improved transmission of muscle forces to the joint.
Figure 3.1 Schematic of a SpeedBridge (Arthrex, Inc., Naples, FL) rotator cuff repair. Two BioComposite SwiveLock C anchors preloaded with FiberTape are placed medially. The FiberTape sutures are passed through the rotator cuff 2 to 3 mm lateral to the musculotendinous junction. The FiberTape sutures from the medial anchors are crisscrossed and secured laterally with two BioComposite SwiveLock C suture anchors.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
