Advantages of Arthroscopic Transosseous Suture Repair of the Rotator Cuff without the Use of Anchors
- 3.1k Downloads
Although arthroscopic anchor suturing is commonly used for rotator cuff repair and achieves good results, certain shortcomings remain, including difficulty with reoperation in cases of retear, anchor dislodgement, knot impingement, and financial cost. In 2005, we developed an anchorless technique for arthroscopic transosseous suture rotator cuff repair.
Description of Technique
After acromioplasty and adequate footprint decortication, three K-wires with perforated tips are inserted through the inferior margin of the greater tuberosity into the medial edge of the footprint using a customized aiming guide. After pulling the rotator cuff stump laterally with a grasper, three K-wires are threaded through the rotator cuff and skin. Thereafter, five Number 2 polyester sutures are passed through three bone tunnels using the perforated tips of the K-wires. The surgery is completed by inserting two pairs of mattress sutures and three bridging sutures.
We investigated the retear rate (based on MR images at least 1 year after the procedure), total score on the UCLA Shoulder Rating Scale, axillary nerve preservation, and issues concerning bone tunnels with this technique in 384 shoulders in 380 patients (174 women [175 shoulders] and 206 men [209 shoulders]). Minimum followup was 2 years (mean, 3.3 years; range, 2–7 years). Complete followup was achieved by 380 patients (384 of 475 [81%] of the procedures performed during the period in question). The remaining 91 patients (91 shoulders) do not have 1-year postsurgical MR images, 2-year UCLA evaluation or intraoperative tear measurement, or they have previous fracture, retear of the rotator cuff, preoperative cervical radiculopathy or axillary nerve palsy, or were lost to followup.
Retears occurred in 24 patients (24 shoulders) (6%). The mean overall UCLA score improved from a preoperative mean of 19.1 to a score of 32.7 at last followup (maximum possible score 35, higher scores being better). Postoperative EMG and clinical examination showed no axillary nerve palsies. Bone tunnel-related issues were encountered in only one shoulder.
Our technique has the following advantages: (1) reoperation is easy in patients with retears; (2) surgical materials used are inexpensive polyester sutures; and (3) no knots are tied onto the rotator cuff. This low-cost method achieves a low retear rate and few bone tunnel problems, the mean postoperative UCLA score being comparable to that obtained by using an arthroscopic anchor suture technique.
Level of Evidence
Level IV, therapeutic study. See Instructions for Authors for a complete description of levels of evidence.
KeywordsRotator Cuff Bone Tunnel Rotator Cuff Repair Great Tuberosity Axillary Nerve
Arthroscopic rotator cuff repair using anchors is a common surgical approach that achieves good results. Recently, there have been changes in anchor suture techniques, from single row to double row and then to suture bridge. However, these procedures have certain limitations: (1) when a retear occurs, reoperation is difficult because the anchors are attached to the greater tuberosity; (2) suture anchors are expensive; (3) dislodgement of the anchor sometimes occurs ; and (4) knot impingement occasionally occurs, this being attributable to knots placed on the rotator cuff .
In an attempt to overcome these limitations, we have developed a technique for arthroscopic transosseous suture repair of the rotator cuff without the use of anchors, which has been used at our hospital since April 2005. Because only sutures are used in this procedure, problems resulting from the above-mentioned anchor suture technique do not occur. The only surgical materials used are inexpensive polyester sutures. Furthermore, as knots are not tied onto the rotator cuff but rather to the lower margin of the greater tuberosity, this technique avoids knot impingement.
We describe the details of this procedure and report its outcomes in terms of (1) retear rate, (2) total score on the UCLA Shoulder Rating Scale , (3) axillary nerve injury, and (4) issues concerning bone tunnels, such as breakage by a suture.
Patients are asked to use a soft brace that holds the arm in 45°flexion and 45° internal rotation for 4 weeks. Passive elevation is started 2 weeks after surgery by a physiotherapist. When the physiotherapist starts passive elevation, the patient is asked to be careful not to actively elevate the affected arm. Active elevation is permitted at 4 weeks and the patients are permitted to drive 6 weeks after surgery. In shoulders with tears greater than 3 cm, the duration of postoperative fixation and start times of passive elevation, active elevation, and driving are each extended by 2 weeks.
Patients and Methods
Retear rate according to tear size
< 1 cm (n = 96)
1–3 cm (n = 219)
3–4.7 cm (n = 69)
Overall (n = 384)
Number of shoulders with retears
Retear rate (%)
Scores on the UCLA Shoulder Rating Scale (n = 384 shoulders)
Time of evaluation
UCLA score (points)
No damage to the axillary nerve was detected by postoperative EMG in any patient. The distance between the top of the greater tuberosity and the insertion point of the K-wire on postoperative MR images (Fig. 12) averaged 17.7 mm (range, 6–34 mm).
A bone tunnel-related issue was encountered in only one shoulder during the procedure. This issue was breakage of the bone tunnel by the suture having been passed through it. No other damages of the bone tunnel were observed on postoperative radiographs, CT scans, and MR images.
Currently, arthroscopic rotator cuff repair is the most widely used technique for managing rotator cuff tears. Therefore, improvement of arthroscopic cuff repair is a major topic. Because anchors are used in almost all arthroscopic rotator cuff repairs, various anchor-related shortcomings remain with this technique, such as difficulty with revision surgery because of the presence of anchors in the greater tuberosity, anchor dislodgement, and knot impingement. We have developed a more economical technique in which the only extra cost involved is that of the suture materials, and problems related to anchors are eliminated. In this study, we describe the details of this procedure and report its outcomes in terms of (1) retear rate, (2) total score on the UCLA Shoulder Rating Scale , (3) axillary nerve preservation, and (4) issues concerning bone tunnels.
In our series, the overall postoperative retear rate was 6% using this technique. This rate is low compared with rates reported for suture-anchor-based methods [2, 10, 14, 20]. In this context, it should be noted that there are no global tears (medial-to-lateral diameter of torn rotator cuff greater than 5 cm) in our series because we use an open multiple muscle transfer technique for such tears. Because retear rates are proportional to tear size in other studies [10, 20] and were in our series, exclusion of patients with such large tears could explain our low retear rate. Therefore, it is meaningless to simply compare the overall retear rates in our series with those reported by others whose series had a different distribution of tear size. Nevertheless, the retear rate in our series was low when compared with those reported by others for small and medium tears [10, 20]. This relatively low retear rate is a potential advantage of our procedure, in which the rotator cuff is pulled peripherally by mattress sutures and compressed tightly to the footprint by bridging sutures.
Two years after our procedure, the mean UCLA total shoulder score was 32.7 points. This is comparable to other series that have reported this end point at a minimum of 2 years after suture-anchor repairs [5, 7, 10, 11].
Protection of the axillary nerve is critically important in this procedure, in which K-wires are inserted through the inferior margin of the greater tuberosity. Postoperative MR images showed that the mean distance between the top of the greater tuberosity and the insertion point of the K-wires was 17.7 mm. Given that the mean distance between the upper margin of the greater tuberosity and the axillary nerve is variously reported as 35 mm  and 45.6 mm  (Fig. 12), the axillary nerve should be relatively safe from injury from K-wires in this procedure. We confirmed the absence of axillary nerve palsy by clinical findings and postoperative EMGs.
Problems with bone tunnels are a theoretical and perhaps an actual concern with transosseous techniques. Sixteen years ago, a cadaver study showed that transosseous rotator cuff suturing was inferior to suture-anchor repairs as a result of cyclic loading . However, the transosseous suturing used in that study was simple suturing of the rotator cuff using a short bone tunnel in an area of weak bone cortex on the lateral margin of the footprint. We believe that the approach studied by Burkhart et al.  is substantially different from our technique, which combined a long bone tunnel extending from the inferior margin of the greater tuberosity to the medial edge of the footprint with two sets of mattress and three bridging sutures. Therefore, we believe that the conclusions of the cadaver study  do not apply to our technique. In our study, breakage of the bone tunnel was identified during the procedure in only one patient; this patient had severe osteoporosis that was apparent on plain radiographs of the lumbar spine.
Our study does have several limitations. Because we did not have a cohort for comparison, we compared our UCLA shoulder scores and retear rates with data from other published studies. The second limitation is that we did not perform EMG on all shoulders. However, we examined all shoulders clinically monthly for the first postoperative year and again at a minimum of 2 years. We identified no axillary nerve palsies. In addition, we assessed postoperative tendon integrity by MRI at a minimum of 1 year (mean, 13 months), whereas we performed clinical assessments 2 years after surgery. We believe this limitation is at least partially offset by our consistent approach to assessment of tendon integrity, which involved postoperative MRI in all patients. Although 19% of patients who underwent this procedure were lost to followup, we believe the followup rate of 81% was sufficient to accurately evaluate the procedure. Because we performed this technique for all complete rotator cuff tears except for global tears (> 5 cm in medial-to-lateral diameter), there was no selection bias. There also is a limitation related to the surgical technique. Our method cannot be used in shoulders where the stump of the torn rotator cuff does not emerge across the top of the humeral head under traction. By expanding decortication of the footprint to the inner side, large rotator cuff tears also can be repaired. However, as shoulders with tears larger than 4 cm had a retear rate of 23%, we do not recommend this technique in such cases.
With our approach, we found a low retear rate, excellent shoulder scores, no nerve injuries, and minimal problems related to bone tunnels. It also allowed us to avoid the use of suture anchors, which can dislodge and add cost to the procedure. In rotator cuff repairs, transosseous sutures may generate a greater bonding force than anchor sutures . Because suturing restricts rotator cuff movement , transosseous suturing is advantageous in rotator cuff repair. In our procedure, mattress sutures are required to draw the cuff stump peripherally and apply adequate initial fixing power to the footprint. Because the K-wires are threaded diagonally through the rotator cuff, strong traction is exerted by the sutures on the rotator cuff.
The presence of anchors in the greater tuberosity makes reoperation difficult. These disadvantages are especially obvious when numerous metal anchors have been used. While probably preferable, absorbable anchors still have the disadvantage of requiring drill holes in the bone. Although revision arthroscopic rotator cuff repair using anchors has been reported [9, 12, 13, 18], the authors’ only reference to previously placed anchors is as follows: “they are removed only when prominent or crowding of the greater tuberosity is seen” . In many cases in which anchor suturing is used, several anchors are inserted into the greater tuberosity, making it difficult to place additional anchors or create bone tunnels for transosseous suture repair. The situation worsens when additional anchors are inserted and second retears occur. Because our anchorless technique uses only polyester sutures, it offers the added advantage of facilitating revision surgery. To obtain satisfactory initial fixing power without applying excessive tension to the rotator cuff, the use of absorbable mattress sutures should be considered. Thus, a comparison study of nonabsorbable and absorbable sutures is required. It is our intention to pursue this issue.
We thank Ryo Murata MD and Shuhei Ogino MD, two of the three surgeons cited in the text who performed this procedure, and Sayoko Kuroda for English proofreading.
Supplementary material 1 (MPG 48700 kb)
Supplementary material 2 (MPG 42624 kb)
- 7.Esch JC, Tamborlane JW. Arthroscopic rotator cuff repair. In: Johnson DH, Pedowitz RA, eds. Practical Orthopaedic Sports Medicine & Arthroscopy. Philadelphia, PA: Lippincott Williams & Wilkins; 2007:285–293.Google Scholar
- 16.Nagasawa R. [Knot impingement after rotator cuff repair][in Japanese]. Katakansetsu (The Shoulder Joint). 2010;34:815–819.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.