- Email: [email protected]
In vitro analysis of rotator cuff repairs: A comparison of arthroscopically inserted tacks or anchors with open transosseous repairs
In Vitro Analysis of Rotator Cuff Repairs: A Comparison of Arthroscopically Inserted Tacks or Anchors With Open Transosseous Repairs Anikar Chhabra, M.D., Vipool K. Goradia, M.D., Eric I. Francke, M.D., Geoffrey S. Baer, M.D., Tim Monahan, M.D., Alex J. Kline, B.S., and Mark D. Miller, M.D.
Purpose: The purpose of this study was to employ a cyclic loading protocol to compare rotator cuff repair strengths of arthroscopically inserted cuff tacks and suture anchors with the traditional open transosseous suture repair. Type of Study: In vitro cadaveric analysis. Methods: Full-thickness 1 ⫻ 3-cm rotator cuff defects were created in 25 fresh-frozen cadaveric shoulders, and were randomized to 1 of 4 repair groups: (1) open repair with transosseous sutures, (2) arthroscopic repair with 2 singly loaded suture anchors, (3) arthroscopic repair with 2 doubly loaded suture anchors, and (4) arthroscopic repair with cuff tacks. All repairs were cyclically loaded from 10 to 180 N, and the numbers of cycles to 50% (5-mm gap) and 100% (10-mm gap) failure were recorded. Results: The number of cycles to 100% failure was significantly higher for the arthroscopic doubly loaded suture anchor repairs when compared with the (1) open transosseous suture repair (P ⫽ .009), (2) arthroscopic cuff tack repair (P ⫽ .003), and (3) arthroscopic singly loaded suture anchor repair (P ⫽ .02). Additionally, the number of cycles to 50% failure was significantly higher for all anchors versus open or tack repair (P ⫽ .03 for both). Conclusions: Immediate postoperative fixation of rotator cuff repairs with doubly loaded suture anchors was more stable than that provided by the open transosseous suture repairs, arthroscopic singly loaded suture anchors, or cuff tacks. However, additional evaluation is needed to examine the effects on the sustained strength of the repair throughout the healing process. Clinical Relevance: These in vitro results indicate that superior immediate postoperative fixation of rotator cuff repairs may be achieved with the doubly loaded suture anchors. However, additional evaluation is needed to examine the effects on the sustained strength of the repair throughout the healing process. Key Words: Rotator cuff tears—Absorbable anchors—Absorbable tacks—Cyclic loading—Cadaveric model.
T
raditionally, the gold standard for rotator cuff repairs has been the open transosseous suture repair. More recently, arthroscopic and arthroscopically assisted mini-open techniques of rotator cuff
From the Department of Orthopaedics, The University of Virginia (A.C., E.I.F., G.S.B., A.J.K., M.D.M.), Charlottesville; and Orthopaedic Research of Virginia (V.K.G., T.M.), Richmond, Virginia, U.S.A. Address correspondence and reprint requests to Mark D. Miller, M.D., Department of Orthopaedic Surgery, University of Virginia, McCue Center, Box 800753, Emmet St and Massie Rd, Charlottesville, VA 22903-0753, U.S.A. E-mail: mdm3p@ virginia.edu © 2005 by the Arthroscopy Association of North America 0749-8063/05/2103-3971$30.00/0 doi:10.1016/j.arthro.2004.11.018
repair have become more popular. Accompanying this increase in popularity, several new arthroscopic rotator cuff repair devices have been designed and marketed. Since the introduction of the suture anchor in 1985 by Goble et al.,1 numerous permutations of the device have been developed. Among the most widely used rotator cuff repair devices currently available are cuff tacks and suture anchors. Ideally, these devices should be easy to insert, provide strength of repair equivalent to the open transosseous repair, and maintain repair strength throughout the healing process.2 Many studies have looked at the initial ultimate tensile strength of rotator cuff repairs using suture anchors3-6 as well as anchor pull-out strength.3-8 More
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 21, No 3 (March), 2005: pp 323-327
323
324
A. CHHABRA ET AL.
recently, Burkhart et al.9,10 published several articles advocating the use of cyclic testing to determine rotator cuff repair strength, because these tests provide a better simulation of the in vivo mechanism of rotator cuff repair failure. Using such a cyclic loading protocol, several studies have shown that bioabsorbable suture anchors and tack fixation are less prone to failure than transosseous suture fixation.11,12 In neither of these studies, however, were the repair devices inserted arthroscopically. To our knowledge, there are no published reports of studies that used arthroscopic insertion of rotator cuff repair devices into intact cadaver shoulders to determine repair strength. Arthroscopic knot tying is potentially more difficult than open knot tying and, as such, may affect the strength of repair. We believe that inserting the devices arthroscopically into intact cadaver shoulders best simulates how the devices will work in real clinical situations. The purpose of this study was to compare the rotator cuff repair strengths after cyclic loading of arthroscopically inserted bioabsorbable suture anchors with singly and doubly loaded suture anchors, arthroscopically inserted bioabsorbable cuff tacks, and open repairs with the traditional transosseous suture technique. We hypothesized that the doubly loaded suture anchors would show repair strength superior to the open repair, the singly loaded suture anchors, and the bioabsorbable cuff tacks. METHODS Twenty-eight fresh-frozen cadaver shoulders (mean age, 79 years) were obtained and stored at ⫺20°C. Each of the 28 specimens was then thawed at room temperature for 24 hours, and randomized to 1 of 4 treatment groups: (1) open technique repaired with transosseous sutures, (2) arthroscopic repair with 2 singly loaded 5-mm Twin-fix suture anchors (Smith & Nephew Endoscopy, Mansfield, MA), (3) arthroscopic repair with 2 doubly loaded 5-mm Twin-fix suture anchors, and (4) arthroscopic repair with Mitek cuff tacks (Mitek Surgical Products, Westwood, MA). A standard mini-open incision was made and a standard crescent-shaped full-thickness defect (30 mm anteriorto-posterior and 10 mm medial-to-lateral) was then created in the supraspinatus and infraspinatus tendons of each of the shoulders as was previously described by Burkhart et al.9,10 Three shoulders already had existing cuff tears that were extended to resemble the created rotator cuff tears. Four specimens were excluded from the study: 2 from group 1 (1 with a massive, irreparable rotator cuff tear, and 1 with a
degenerative tear with dimensions that were not consistent with our created defects), 1 from group 2, and 1 from group 3. In the excluded specimens in group 2 and group 3, very poor tissue quality was noted at the time the tears were created and the tissue was deemed inadequate for repair. The open repairs were performed by an orthopaedic sports fellowship–trained physician who performs mini-open repairs in a clinical setting, and the arthroscopic repairs were performed by an orthopaedic sports fellowship–trained physician who routinely uses arthroscopic repairs in his practice. Company representatives were present to ensure appropriate placement and use of the devices. For group 1, the repair was performed through the standard mini-open incision, and with groups 2, 3, and 4, a layered closure of the incision needed to create the initial cuff defect was performed prior to arthroscopic repair of the cuff. In group 1 (n ⫽ 5), each rotator cuff defect was repaired with two No. 2 Ethibond transosseous sutures (Ethicon, Somerville, NJ) using a Mason-Allen stitch technique.4 These sutures were placed through 3 drill holes that were spaced 10 mm apart and 30 mm distal to the tip of the greater tuberosity.13 For the arthroscopic repairs (groups 2, 3, and 4), standard posterior, anterior-superior, and lateral portals were used. A bursectomy was performed for visualization and soft tissue was removed from the greater tuberosity with a shaver. An accessory anterolateral portal was created for insertion of anchors and cuff tacks. In group 2 (n ⫽ 6), each defect was repaired with 2 Twin-fix singly loaded suture anchors using No. 2 braided polyester sutures. One of the 2 sutures from each anchor was removed. The sutures were placed at a 45° angle (the “deadman’s angle”) in the greater tuberosity. The sutures were placed approximately 1 cm medial to the leading edge of the cuff tear using standard arthroscopic suture passing instruments, creating a simple stitch. A lockable sliding knot backed with 3 reverse half-hitches on alternating posts was used for each suture.14 The technique for group 3 (n ⫽ 6) was identical to group 2 except that both sutures from each Twin-fix anchor was utilized for the repair (doubly loaded suture anchors). In group 4 (n ⫽ 7), each defect was repaired with two smooth bioabsorbable Mitek cuff tacks by mobilizing the leading edges of the tears to close the defect. The tacks were inserted (according to the manufacturer’s instrumentation and guidelines) through the tendon 5 mm from its free end and into a predrilled hole in the greater tuberosity. The specimens were dissected, leaving only the intact rotator cuff, humeral head, and proximal 15 cm
ROTATOR CUFF REPAIR: TACKS, ANCHORS, SUTURES TABLE 1.
325
Number of Cycles to 10-mm Gap Formation
Groups
No. of Specimens
5-mm Gap (mean cycles ⫾ SE) (range)
10-mm Gap (mean cycles ⫾ SE) (range)
Transosseous sutures Smith & Nephew singly loaded suture anchors Smith & Nephew doubly loaded suture anchors Mitek cuff tacks
5 6 6 7
3.2 ⫾ 3.8 (1-10) 24.2 ⫾ 37.8 (2-100) 30.7 ⫾ 39.1 (1-100) 3.4 ⫾ 4.8 (1-15)
25.0 ⫾ 25.5 (2-64) 31.5 ⫾ 36.2 (3-100) 83.5 ⫾ 40.4 (1-100) 8.63 ⫾ 9.9 (1-28)
of the humeral shaft. For each specimen, a 2.5-cm wide Nylon strap9,10 was attached to the proximal rotator cuff with a modified Krackow stitch using No. 5 Ethibond sutures. The specimens were kept moist with saline during preparation and testing. Each specimen was mounted on an MTS Mini Bionix testing machine (MTS Systems, Eden Prairie, MN). The humeral shaft was clamped to a platform and the Nylon strap was looped around a bar that was connected to the load cell. The repairs were cyclically loaded from 10 to 180 N with a 5-second cycle similar to that described by Burkhart et al.9,10 Gap formation at each repair site was measured using an extensometer. The number of cycles to 5-mm (50% failure) and 10-mm (100% failure) gap formation was recorded, along with the mechanism of failure.9,10 The data were then analyzed using a 1-way analysis of variance, followed by individual Student t tests with P ⬍ .05 being considered statistically significant. RESULTS Multiple failure modes were identified. In group 1, the transosseous sutures failed by the sutures pulling through the rotator cuff (n ⫽ 5). In group 2, the suture anchor loosened (n ⫽ 1), the suture broke at the anchor (n ⫽ 1), or the sutures pulled through the rotator cuff (n ⫽ 3). One specimen in this group did not fail through the number of cycles tested. For group 3, the only failure (n ⫽ 1) occurred when a suture anchor loosened. Five specimens in this group did not fail. For group 4, failures occurred when the tack heads broke off (n ⫽ 3) or when the tacks pulled out (n ⫽ 4). The average cycles to 100% failure (gap formation of 10-mm) were: 25 for the transosseous suture repair group, 31.5 for the singly loaded suture anchor group, 83.5 for the doubly loaded suture anchor group, and 8.6 for the cuff tack group (Table 1). Arthroscopically inserted doubly loaded suture anchors had significantly more cycles to 100% failure than (1) the transosseous suture repair (P ⫽ .009), (2)
the arthroscopic cuff tack repair (P ⫽ .003), and the (3) arthroscopically repaired singly loaded suture anchors (P ⫽ .02). However, no significant differences (P ⬎ .05) were found with the singly loaded suture anchor when compared with (1) the transosseous suture repair or (2) the arthroscopic cuff tack repair (Table 2). For 50% failure, no significant differences were present when comparing any of the 4 treatment groups. However, the number of cycles to 50% failure was significantly higher when combining both anchor groups versus (1) the transosseous open repair (P ⫽ .03) and (2) the arthroscopic tack repair (P ⫽ .03) (Table 2). DISCUSSION Rotator cuff tears have long been recognized as a disabling condition that usually can be effectively treated by surgical treatment, although recent debate has sought to elucidate the specific roles of traditional open techniques versus arthroscopic repairs. Advantages of arthroscopic repairs include complete mobilization and release of the rotator cuff, decreased surgical trauma to the deltoid, decreased postoperative
TABLE 2.
T Test P Values
Comparison Groups Open v singly loaded anchors Open v doubly loaded anchors Open v tacks Singly loaded anchors v doubly loaded anchors Singly loaded anchors v tacks Doubly loaded anchors v tacks Anchors combined v open Anchors combined v tacks
5-mm Gap Formation
10-mm Gap Formation
.117 .073 .472
.368 *.009 .116
.388 .118 .074 *.029 *.030
*.021 .093 *.003 .147 *.004
NOTE. Statistically significant values are indicated with an asterisk. “Anchors combined” represents the singly loaded and doubly loaded suture groups combined.
326
A. CHHABRA ET AL.
pain, earlier time to rehabilitation, and decreased postoperative stiffness.15,16 Disadvantages of a complete arthroscopic repair include technical difficulty and uncertainty of bone-tendon fixation. Recently, Severud et al.17 reported that all arthroscopic rotator cuff repairs with suture anchors had shoulder scores similar to those of mini-open repairs at final followup, but no patients had adhesive capsulitis in the arthroscopic group compared with 4 in the mini-open group. Additionally, at 6 and 12 weeks, the arthroscopic group had significantly better motion. There were no anchor-related complications.17 To our knowledge, although there are data comparing the pull-out strength of several suture anchors,18 there are no biomechanical studies that compare arthroscopically inserted single and double suture anchors with each other, with arthroscopically inserted cuff tacks, and with the traditional open transosseous suture repair. We believe that inserting these devices arthroscopically and then testing them with a cyclic loading protocol best simulates true rotator cuff repair conditions. The modes of failure seen in this study are consistent with previous studies looking at failure mechanisms of rotator cuff repairs. All failures in the open transosseous suture repair occurred by the sutures pulling out of the rotator cuff repair site. The majority of failures of the singly loaded suture anchors occurred by the suture pulling through the repair, whereas the only doubly loaded suture anchor that failed occurred because the anchor bone fixation failed. Multiple modes of failure were seen in the tack group, including breakage of the head of the tack and pulling of the tack through the repair. These modes of failure are consistent with those previously reported in the literature.11 The modes of failure in this study are informative, but no conclusions can be drawn because the quality of rotator cuff tissue, or bone density and quality were not quantified or examined in this study. This study found statistically more cycles to 100% failure with the arthroscopically inserted doubly loaded Smith & Nephew suture anchor fixation (83.5 cycles) when compared with all other groups, including the open transosseous suture repair (25.0 cycles). Burkhart et al.9,10 reported 285 cycles to 100% failure with 3 arthroscopically inserted Mitek-RC anchors compared with 188 cycles to failure with 3 simple transosseous sutures. In the current study, the average numbers of cycles to 100% failure were less than those reported in the Burkhart studies.9,10 We theorize that these differences are a result of 2 main factors. First, we used 2 sutures or 2 anchors instead of 3
anchors to repair the rotator cuff defects that were larger in size than those of Burkhart et al.9,10 Second, our cadavers were older (mean age of 79 years). In the study by Burkhart et al.,10 10-mm failure (100% failure) in specimens more than 45-years-old with Mitek anchor fixation occurred at 91 cycles,10 a value similar to those in the current study. Goradia et al.11 had similar cycles to failure, but demonstrated that Acufex (Smith & Nephew) bioabsorbable tacks and Mitek Super suture anchors were comparable to each other and both were superior to an open transosseous suture repair.11 The modified Mason-Allen stitch used in our transosseous suture group has been shown to provide better fixation under cyclic loading conditions when compared with numerous other techniques.5 Burkhart’s failures in this group occurred when the sutures pulled through bone.10 Use of the modified Mason-Allen stitch may explain why our failure modes were different from those reported by Burkhart et al.9,10 Burkhart’s Mitek anchor groups failed primarily through the tendons, just as our singly loaded suture anchor group did. Because the suture anchor group failed primarily through the tendon, the modified Mason-Allen stitch should be considered with suturebased anchors. In a study by Rossouw et al.,12 a different method of cyclic loading was used than that described in the current study for rotator cuff defects repaired with Mitek GII anchors and No. 2 braided polyester sutures. Their failure mode was primarily suture breakage at the knots or at the anchors and not failure through the tendon. The enhanced tendon fixation may have occurred because of the locking-type stitch they used. The ideal type of stitch to be used with suture anchors has yet to be determined, and was not within the scope of this investigation. As shown by our 50% failure (5-mm gap formation) data, there were significantly more cycles to failure when comparing all anchors (singly and doubly loaded) with open repairs and to the tacks. Because these modes of failure usually occurred within the first few cycles, this demonstrates less initial displacement with the inserted anchors. These data are not consistent with a study in sheep by Lewis et al.,19 which showed that the bone tunnel suture technique produced a stronger initial attachment than the suture anchors. Several potential pitfalls with our study should be pointed out. First, our study did not address the quality of the rotator cuff tissue in the cadaver specimens, or the osteoporotic nature of the bone. Also, we acknowledge that it may be difficult to mechanically compare
ROTATOR CUFF REPAIR: TACKS, ANCHORS, SUTURES the “created defects” used in biomechanical studies with true rotator cuff tears seen in a patient. These factors may have relevance when applying our data to clinical situations. Second, as with any cadaveric cyclic loading study, in vivo remodeling of the bone tendon interface cannot be accounted for. The body’s response to the repair and other clinical factors will play a role in determining a successful clinical outcome. Additionally, the minimum initial stability at the interface of the repair site needed to achieve a successful repair is unknown, and may vary with clinical factors such as patient age and quality of rotator cuff tissue.
CONCLUSIONS In summary, the results of this experiment support our hypothesis that doubly loaded suture anchors provide stronger initial fixation under submaximal cyclic loading for repair of 1 ⫻ 3-cm crescent-shaped rotator cuff tears when compared with singly loaded suture anchors, cuff tacks, and open transosseous suture repairs. Although previous studies have looked at the repair strengths achieved with various fixation devices, this is the first study to our knowledge that looked specifically at true arthroscopic insertion of these devices, thereby accounting for potential difficulties in arthroscopic knot tying that could feasibly lead to decreased repair strength. And although our data suggest that initial repair strengths are superior with doubly loaded suture anchors, additional evaluation is needed to examine the effects on the sustained strength of the repair throughout the healing process. REFERENCES 1. Goble EM, Somers WK, Clark R, Olsen RE. The development of suture anchors for use in soft tissue fixation to bone. Am J Sports Med 1994;22:236-239.
327
2. Burkhart SS. Biomechanics of rotator cuff repair: Converting the ritual to a science. Instr Course Lect 1998;47:43-50. 3. Craft DV, Moseley JB, Cawley PW, Noble PC. Fixation strength of rotator cuff repairs with suture anchors and the transosseous suture technique. J Shoulder Elbow Surg 1996;5:32-40. 4. Gerber C, Schneeberger AG, Beck M, Schlegel U. Mechanical strength of repairs of the rotator cuff. J Bone Joint Surg Br 1994;76:371-380. 5. Hecker AT, Shea M, Hayhurst JO, Myers ER, Meeks LW, Hayes WC. Pull-out strength of suture anchors for rotator cuff and Bankart lesion repairs. Am J Sports Med 1993;21:874-879. 6. Reed SC, Glossop N, Ogilvie-Harris DJ. Full-thickness rotator cuff tears. A biomechanical comparison of suture versus bone anchor techniques. Am J Sports Med 1996;24:46-48. 7. Barber FA, Herbert MA, Click JN. Suture anchor strength revisited. Arthroscopy 1996;12:32-38. 8. Barber FA, Herbert MA, Click JN. Internal fixation strength of suture anchors—Update 1997. Arthroscopy 1997;13:355-362. 9. Burkhart SS, Johnson TC, Wirth MA, Athanasiou KA. Cyclic loading of transosseous rotator cuff repairs: Tension overload as a possible cause of failure. Arthroscopy 1997;13:172-176. 10. Burkhart SS, Diaz Pagan JL, Wirth MA, Athanasiou KA. Cyclic loading of anchor-based rotator cuff repairs: Confirmation of the tension overload phenomenon and comparison of suture anchor fixation with transosseous fixation. Arthroscopy 1997;13:720-724. 11. Goradia VK, Mullen DJ, Boucher HR, Parks BG, O’Donnell, JB. Cyclic loading of rotator cuff repairs: A comparison of bioabsorbable tacks with metal suture anchors and transosseous sutures. Arthroscopy 2001;4:360-364. 12. Rossouw DJ, McElroy BJ, Amis AA, Emery RJ. A biomechanical evaluation of suture anchors in repair of the rotator cuff. J Bone Joint Surg Br 1997;79:458-461. 13. Caldwell GL, Warner JP, Miller MD, Boardman D, Towers J, Debski R. Strength of fixation with transosseous sutures in rotator cuff repair. J Bone Joint Surg Am 1997;79:1064-1068. 14. Wiley WB, Goradia VK. The Tuckahoe knot: A secure locking slip knot. Arthroscopy 2004;20:556-559. 15. Lo IKY, Burkhart SS. Current concepts in arthroscopic rotator cuff repair. Am J Sports Med 2003;2:308-324. 16. Yamaguchi K, Levine WN, Marra G, Galatz LM, Klepps S, Flatow E. Transitioning to arthroscopic rotator cuff repair: The pros and cons. J Bone Joint Surg Am 2003;85:144-154. 17. Severud EL, Ruotolo C, Abbott DD, Nottage WM. All-arthroscopic versus mini-open rotator cuff repair: A long-term retrospective outcome comparison. Arthroscopy 2003;3:234-238. 18. Schneeberger AG, Von Roll A, Kalberer F, Jacob HAC, Gerber C. Mechanical strength of arthroscopic rotator cuff repair techniques. J Bone Joint Surg Am 2002;84:2152-2160. 19. Lewis CW, Schlegal TF, Hawkins RJ, James SP, Turner AS. Comparison of tunnel suture and suture anchor methods as a function of time in a sheep model. Biomed Sci Instrum 1999; 35:403-408.
Purpose: The purpose of this study was to employ a cyclic loading protocol to compare rotator cuff repair strengths of arthroscopically inserted cuff tacks and suture anchors with the traditional open transosseous suture repair. Type of Study: In vitro cadaveric analysis. Methods: Full-thickness 1 ⫻ 3-cm rotator cuff defects were created in 25 fresh-frozen cadaveric shoulders, and were randomized to 1 of 4 repair groups: (1) open repair with transosseous sutures, (2) arthroscopic repair with 2 singly loaded suture anchors, (3) arthroscopic repair with 2 doubly loaded suture anchors, and (4) arthroscopic repair with cuff tacks. All repairs were cyclically loaded from 10 to 180 N, and the numbers of cycles to 50% (5-mm gap) and 100% (10-mm gap) failure were recorded. Results: The number of cycles to 100% failure was significantly higher for the arthroscopic doubly loaded suture anchor repairs when compared with the (1) open transosseous suture repair (P ⫽ .009), (2) arthroscopic cuff tack repair (P ⫽ .003), and (3) arthroscopic singly loaded suture anchor repair (P ⫽ .02). Additionally, the number of cycles to 50% failure was significantly higher for all anchors versus open or tack repair (P ⫽ .03 for both). Conclusions: Immediate postoperative fixation of rotator cuff repairs with doubly loaded suture anchors was more stable than that provided by the open transosseous suture repairs, arthroscopic singly loaded suture anchors, or cuff tacks. However, additional evaluation is needed to examine the effects on the sustained strength of the repair throughout the healing process. Clinical Relevance: These in vitro results indicate that superior immediate postoperative fixation of rotator cuff repairs may be achieved with the doubly loaded suture anchors. However, additional evaluation is needed to examine the effects on the sustained strength of the repair throughout the healing process. Key Words: Rotator cuff tears—Absorbable anchors—Absorbable tacks—Cyclic loading—Cadaveric model.
T
raditionally, the gold standard for rotator cuff repairs has been the open transosseous suture repair. More recently, arthroscopic and arthroscopically assisted mini-open techniques of rotator cuff
From the Department of Orthopaedics, The University of Virginia (A.C., E.I.F., G.S.B., A.J.K., M.D.M.), Charlottesville; and Orthopaedic Research of Virginia (V.K.G., T.M.), Richmond, Virginia, U.S.A. Address correspondence and reprint requests to Mark D. Miller, M.D., Department of Orthopaedic Surgery, University of Virginia, McCue Center, Box 800753, Emmet St and Massie Rd, Charlottesville, VA 22903-0753, U.S.A. E-mail: mdm3p@ virginia.edu © 2005 by the Arthroscopy Association of North America 0749-8063/05/2103-3971$30.00/0 doi:10.1016/j.arthro.2004.11.018
repair have become more popular. Accompanying this increase in popularity, several new arthroscopic rotator cuff repair devices have been designed and marketed. Since the introduction of the suture anchor in 1985 by Goble et al.,1 numerous permutations of the device have been developed. Among the most widely used rotator cuff repair devices currently available are cuff tacks and suture anchors. Ideally, these devices should be easy to insert, provide strength of repair equivalent to the open transosseous repair, and maintain repair strength throughout the healing process.2 Many studies have looked at the initial ultimate tensile strength of rotator cuff repairs using suture anchors3-6 as well as anchor pull-out strength.3-8 More
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 21, No 3 (March), 2005: pp 323-327
323
324
A. CHHABRA ET AL.
recently, Burkhart et al.9,10 published several articles advocating the use of cyclic testing to determine rotator cuff repair strength, because these tests provide a better simulation of the in vivo mechanism of rotator cuff repair failure. Using such a cyclic loading protocol, several studies have shown that bioabsorbable suture anchors and tack fixation are less prone to failure than transosseous suture fixation.11,12 In neither of these studies, however, were the repair devices inserted arthroscopically. To our knowledge, there are no published reports of studies that used arthroscopic insertion of rotator cuff repair devices into intact cadaver shoulders to determine repair strength. Arthroscopic knot tying is potentially more difficult than open knot tying and, as such, may affect the strength of repair. We believe that inserting the devices arthroscopically into intact cadaver shoulders best simulates how the devices will work in real clinical situations. The purpose of this study was to compare the rotator cuff repair strengths after cyclic loading of arthroscopically inserted bioabsorbable suture anchors with singly and doubly loaded suture anchors, arthroscopically inserted bioabsorbable cuff tacks, and open repairs with the traditional transosseous suture technique. We hypothesized that the doubly loaded suture anchors would show repair strength superior to the open repair, the singly loaded suture anchors, and the bioabsorbable cuff tacks. METHODS Twenty-eight fresh-frozen cadaver shoulders (mean age, 79 years) were obtained and stored at ⫺20°C. Each of the 28 specimens was then thawed at room temperature for 24 hours, and randomized to 1 of 4 treatment groups: (1) open technique repaired with transosseous sutures, (2) arthroscopic repair with 2 singly loaded 5-mm Twin-fix suture anchors (Smith & Nephew Endoscopy, Mansfield, MA), (3) arthroscopic repair with 2 doubly loaded 5-mm Twin-fix suture anchors, and (4) arthroscopic repair with Mitek cuff tacks (Mitek Surgical Products, Westwood, MA). A standard mini-open incision was made and a standard crescent-shaped full-thickness defect (30 mm anteriorto-posterior and 10 mm medial-to-lateral) was then created in the supraspinatus and infraspinatus tendons of each of the shoulders as was previously described by Burkhart et al.9,10 Three shoulders already had existing cuff tears that were extended to resemble the created rotator cuff tears. Four specimens were excluded from the study: 2 from group 1 (1 with a massive, irreparable rotator cuff tear, and 1 with a
degenerative tear with dimensions that were not consistent with our created defects), 1 from group 2, and 1 from group 3. In the excluded specimens in group 2 and group 3, very poor tissue quality was noted at the time the tears were created and the tissue was deemed inadequate for repair. The open repairs were performed by an orthopaedic sports fellowship–trained physician who performs mini-open repairs in a clinical setting, and the arthroscopic repairs were performed by an orthopaedic sports fellowship–trained physician who routinely uses arthroscopic repairs in his practice. Company representatives were present to ensure appropriate placement and use of the devices. For group 1, the repair was performed through the standard mini-open incision, and with groups 2, 3, and 4, a layered closure of the incision needed to create the initial cuff defect was performed prior to arthroscopic repair of the cuff. In group 1 (n ⫽ 5), each rotator cuff defect was repaired with two No. 2 Ethibond transosseous sutures (Ethicon, Somerville, NJ) using a Mason-Allen stitch technique.4 These sutures were placed through 3 drill holes that were spaced 10 mm apart and 30 mm distal to the tip of the greater tuberosity.13 For the arthroscopic repairs (groups 2, 3, and 4), standard posterior, anterior-superior, and lateral portals were used. A bursectomy was performed for visualization and soft tissue was removed from the greater tuberosity with a shaver. An accessory anterolateral portal was created for insertion of anchors and cuff tacks. In group 2 (n ⫽ 6), each defect was repaired with 2 Twin-fix singly loaded suture anchors using No. 2 braided polyester sutures. One of the 2 sutures from each anchor was removed. The sutures were placed at a 45° angle (the “deadman’s angle”) in the greater tuberosity. The sutures were placed approximately 1 cm medial to the leading edge of the cuff tear using standard arthroscopic suture passing instruments, creating a simple stitch. A lockable sliding knot backed with 3 reverse half-hitches on alternating posts was used for each suture.14 The technique for group 3 (n ⫽ 6) was identical to group 2 except that both sutures from each Twin-fix anchor was utilized for the repair (doubly loaded suture anchors). In group 4 (n ⫽ 7), each defect was repaired with two smooth bioabsorbable Mitek cuff tacks by mobilizing the leading edges of the tears to close the defect. The tacks were inserted (according to the manufacturer’s instrumentation and guidelines) through the tendon 5 mm from its free end and into a predrilled hole in the greater tuberosity. The specimens were dissected, leaving only the intact rotator cuff, humeral head, and proximal 15 cm
ROTATOR CUFF REPAIR: TACKS, ANCHORS, SUTURES TABLE 1.
325
Number of Cycles to 10-mm Gap Formation
Groups
No. of Specimens
5-mm Gap (mean cycles ⫾ SE) (range)
10-mm Gap (mean cycles ⫾ SE) (range)
Transosseous sutures Smith & Nephew singly loaded suture anchors Smith & Nephew doubly loaded suture anchors Mitek cuff tacks
5 6 6 7
3.2 ⫾ 3.8 (1-10) 24.2 ⫾ 37.8 (2-100) 30.7 ⫾ 39.1 (1-100) 3.4 ⫾ 4.8 (1-15)
25.0 ⫾ 25.5 (2-64) 31.5 ⫾ 36.2 (3-100) 83.5 ⫾ 40.4 (1-100) 8.63 ⫾ 9.9 (1-28)
of the humeral shaft. For each specimen, a 2.5-cm wide Nylon strap9,10 was attached to the proximal rotator cuff with a modified Krackow stitch using No. 5 Ethibond sutures. The specimens were kept moist with saline during preparation and testing. Each specimen was mounted on an MTS Mini Bionix testing machine (MTS Systems, Eden Prairie, MN). The humeral shaft was clamped to a platform and the Nylon strap was looped around a bar that was connected to the load cell. The repairs were cyclically loaded from 10 to 180 N with a 5-second cycle similar to that described by Burkhart et al.9,10 Gap formation at each repair site was measured using an extensometer. The number of cycles to 5-mm (50% failure) and 10-mm (100% failure) gap formation was recorded, along with the mechanism of failure.9,10 The data were then analyzed using a 1-way analysis of variance, followed by individual Student t tests with P ⬍ .05 being considered statistically significant. RESULTS Multiple failure modes were identified. In group 1, the transosseous sutures failed by the sutures pulling through the rotator cuff (n ⫽ 5). In group 2, the suture anchor loosened (n ⫽ 1), the suture broke at the anchor (n ⫽ 1), or the sutures pulled through the rotator cuff (n ⫽ 3). One specimen in this group did not fail through the number of cycles tested. For group 3, the only failure (n ⫽ 1) occurred when a suture anchor loosened. Five specimens in this group did not fail. For group 4, failures occurred when the tack heads broke off (n ⫽ 3) or when the tacks pulled out (n ⫽ 4). The average cycles to 100% failure (gap formation of 10-mm) were: 25 for the transosseous suture repair group, 31.5 for the singly loaded suture anchor group, 83.5 for the doubly loaded suture anchor group, and 8.6 for the cuff tack group (Table 1). Arthroscopically inserted doubly loaded suture anchors had significantly more cycles to 100% failure than (1) the transosseous suture repair (P ⫽ .009), (2)
the arthroscopic cuff tack repair (P ⫽ .003), and the (3) arthroscopically repaired singly loaded suture anchors (P ⫽ .02). However, no significant differences (P ⬎ .05) were found with the singly loaded suture anchor when compared with (1) the transosseous suture repair or (2) the arthroscopic cuff tack repair (Table 2). For 50% failure, no significant differences were present when comparing any of the 4 treatment groups. However, the number of cycles to 50% failure was significantly higher when combining both anchor groups versus (1) the transosseous open repair (P ⫽ .03) and (2) the arthroscopic tack repair (P ⫽ .03) (Table 2). DISCUSSION Rotator cuff tears have long been recognized as a disabling condition that usually can be effectively treated by surgical treatment, although recent debate has sought to elucidate the specific roles of traditional open techniques versus arthroscopic repairs. Advantages of arthroscopic repairs include complete mobilization and release of the rotator cuff, decreased surgical trauma to the deltoid, decreased postoperative
TABLE 2.
T Test P Values
Comparison Groups Open v singly loaded anchors Open v doubly loaded anchors Open v tacks Singly loaded anchors v doubly loaded anchors Singly loaded anchors v tacks Doubly loaded anchors v tacks Anchors combined v open Anchors combined v tacks
5-mm Gap Formation
10-mm Gap Formation
.117 .073 .472
.368 *.009 .116
.388 .118 .074 *.029 *.030
*.021 .093 *.003 .147 *.004
NOTE. Statistically significant values are indicated with an asterisk. “Anchors combined” represents the singly loaded and doubly loaded suture groups combined.
326
A. CHHABRA ET AL.
pain, earlier time to rehabilitation, and decreased postoperative stiffness.15,16 Disadvantages of a complete arthroscopic repair include technical difficulty and uncertainty of bone-tendon fixation. Recently, Severud et al.17 reported that all arthroscopic rotator cuff repairs with suture anchors had shoulder scores similar to those of mini-open repairs at final followup, but no patients had adhesive capsulitis in the arthroscopic group compared with 4 in the mini-open group. Additionally, at 6 and 12 weeks, the arthroscopic group had significantly better motion. There were no anchor-related complications.17 To our knowledge, although there are data comparing the pull-out strength of several suture anchors,18 there are no biomechanical studies that compare arthroscopically inserted single and double suture anchors with each other, with arthroscopically inserted cuff tacks, and with the traditional open transosseous suture repair. We believe that inserting these devices arthroscopically and then testing them with a cyclic loading protocol best simulates true rotator cuff repair conditions. The modes of failure seen in this study are consistent with previous studies looking at failure mechanisms of rotator cuff repairs. All failures in the open transosseous suture repair occurred by the sutures pulling out of the rotator cuff repair site. The majority of failures of the singly loaded suture anchors occurred by the suture pulling through the repair, whereas the only doubly loaded suture anchor that failed occurred because the anchor bone fixation failed. Multiple modes of failure were seen in the tack group, including breakage of the head of the tack and pulling of the tack through the repair. These modes of failure are consistent with those previously reported in the literature.11 The modes of failure in this study are informative, but no conclusions can be drawn because the quality of rotator cuff tissue, or bone density and quality were not quantified or examined in this study. This study found statistically more cycles to 100% failure with the arthroscopically inserted doubly loaded Smith & Nephew suture anchor fixation (83.5 cycles) when compared with all other groups, including the open transosseous suture repair (25.0 cycles). Burkhart et al.9,10 reported 285 cycles to 100% failure with 3 arthroscopically inserted Mitek-RC anchors compared with 188 cycles to failure with 3 simple transosseous sutures. In the current study, the average numbers of cycles to 100% failure were less than those reported in the Burkhart studies.9,10 We theorize that these differences are a result of 2 main factors. First, we used 2 sutures or 2 anchors instead of 3
anchors to repair the rotator cuff defects that were larger in size than those of Burkhart et al.9,10 Second, our cadavers were older (mean age of 79 years). In the study by Burkhart et al.,10 10-mm failure (100% failure) in specimens more than 45-years-old with Mitek anchor fixation occurred at 91 cycles,10 a value similar to those in the current study. Goradia et al.11 had similar cycles to failure, but demonstrated that Acufex (Smith & Nephew) bioabsorbable tacks and Mitek Super suture anchors were comparable to each other and both were superior to an open transosseous suture repair.11 The modified Mason-Allen stitch used in our transosseous suture group has been shown to provide better fixation under cyclic loading conditions when compared with numerous other techniques.5 Burkhart’s failures in this group occurred when the sutures pulled through bone.10 Use of the modified Mason-Allen stitch may explain why our failure modes were different from those reported by Burkhart et al.9,10 Burkhart’s Mitek anchor groups failed primarily through the tendons, just as our singly loaded suture anchor group did. Because the suture anchor group failed primarily through the tendon, the modified Mason-Allen stitch should be considered with suturebased anchors. In a study by Rossouw et al.,12 a different method of cyclic loading was used than that described in the current study for rotator cuff defects repaired with Mitek GII anchors and No. 2 braided polyester sutures. Their failure mode was primarily suture breakage at the knots or at the anchors and not failure through the tendon. The enhanced tendon fixation may have occurred because of the locking-type stitch they used. The ideal type of stitch to be used with suture anchors has yet to be determined, and was not within the scope of this investigation. As shown by our 50% failure (5-mm gap formation) data, there were significantly more cycles to failure when comparing all anchors (singly and doubly loaded) with open repairs and to the tacks. Because these modes of failure usually occurred within the first few cycles, this demonstrates less initial displacement with the inserted anchors. These data are not consistent with a study in sheep by Lewis et al.,19 which showed that the bone tunnel suture technique produced a stronger initial attachment than the suture anchors. Several potential pitfalls with our study should be pointed out. First, our study did not address the quality of the rotator cuff tissue in the cadaver specimens, or the osteoporotic nature of the bone. Also, we acknowledge that it may be difficult to mechanically compare
ROTATOR CUFF REPAIR: TACKS, ANCHORS, SUTURES the “created defects” used in biomechanical studies with true rotator cuff tears seen in a patient. These factors may have relevance when applying our data to clinical situations. Second, as with any cadaveric cyclic loading study, in vivo remodeling of the bone tendon interface cannot be accounted for. The body’s response to the repair and other clinical factors will play a role in determining a successful clinical outcome. Additionally, the minimum initial stability at the interface of the repair site needed to achieve a successful repair is unknown, and may vary with clinical factors such as patient age and quality of rotator cuff tissue.
CONCLUSIONS In summary, the results of this experiment support our hypothesis that doubly loaded suture anchors provide stronger initial fixation under submaximal cyclic loading for repair of 1 ⫻ 3-cm crescent-shaped rotator cuff tears when compared with singly loaded suture anchors, cuff tacks, and open transosseous suture repairs. Although previous studies have looked at the repair strengths achieved with various fixation devices, this is the first study to our knowledge that looked specifically at true arthroscopic insertion of these devices, thereby accounting for potential difficulties in arthroscopic knot tying that could feasibly lead to decreased repair strength. And although our data suggest that initial repair strengths are superior with doubly loaded suture anchors, additional evaluation is needed to examine the effects on the sustained strength of the repair throughout the healing process. REFERENCES 1. Goble EM, Somers WK, Clark R, Olsen RE. The development of suture anchors for use in soft tissue fixation to bone. Am J Sports Med 1994;22:236-239.
327
2. Burkhart SS. Biomechanics of rotator cuff repair: Converting the ritual to a science. Instr Course Lect 1998;47:43-50. 3. Craft DV, Moseley JB, Cawley PW, Noble PC. Fixation strength of rotator cuff repairs with suture anchors and the transosseous suture technique. J Shoulder Elbow Surg 1996;5:32-40. 4. Gerber C, Schneeberger AG, Beck M, Schlegel U. Mechanical strength of repairs of the rotator cuff. J Bone Joint Surg Br 1994;76:371-380. 5. Hecker AT, Shea M, Hayhurst JO, Myers ER, Meeks LW, Hayes WC. Pull-out strength of suture anchors for rotator cuff and Bankart lesion repairs. Am J Sports Med 1993;21:874-879. 6. Reed SC, Glossop N, Ogilvie-Harris DJ. Full-thickness rotator cuff tears. A biomechanical comparison of suture versus bone anchor techniques. Am J Sports Med 1996;24:46-48. 7. Barber FA, Herbert MA, Click JN. Suture anchor strength revisited. Arthroscopy 1996;12:32-38. 8. Barber FA, Herbert MA, Click JN. Internal fixation strength of suture anchors—Update 1997. Arthroscopy 1997;13:355-362. 9. Burkhart SS, Johnson TC, Wirth MA, Athanasiou KA. Cyclic loading of transosseous rotator cuff repairs: Tension overload as a possible cause of failure. Arthroscopy 1997;13:172-176. 10. Burkhart SS, Diaz Pagan JL, Wirth MA, Athanasiou KA. Cyclic loading of anchor-based rotator cuff repairs: Confirmation of the tension overload phenomenon and comparison of suture anchor fixation with transosseous fixation. Arthroscopy 1997;13:720-724. 11. Goradia VK, Mullen DJ, Boucher HR, Parks BG, O’Donnell, JB. Cyclic loading of rotator cuff repairs: A comparison of bioabsorbable tacks with metal suture anchors and transosseous sutures. Arthroscopy 2001;4:360-364. 12. Rossouw DJ, McElroy BJ, Amis AA, Emery RJ. A biomechanical evaluation of suture anchors in repair of the rotator cuff. J Bone Joint Surg Br 1997;79:458-461. 13. Caldwell GL, Warner JP, Miller MD, Boardman D, Towers J, Debski R. Strength of fixation with transosseous sutures in rotator cuff repair. J Bone Joint Surg Am 1997;79:1064-1068. 14. Wiley WB, Goradia VK. The Tuckahoe knot: A secure locking slip knot. Arthroscopy 2004;20:556-559. 15. Lo IKY, Burkhart SS. Current concepts in arthroscopic rotator cuff repair. Am J Sports Med 2003;2:308-324. 16. Yamaguchi K, Levine WN, Marra G, Galatz LM, Klepps S, Flatow E. Transitioning to arthroscopic rotator cuff repair: The pros and cons. J Bone Joint Surg Am 2003;85:144-154. 17. Severud EL, Ruotolo C, Abbott DD, Nottage WM. All-arthroscopic versus mini-open rotator cuff repair: A long-term retrospective outcome comparison. Arthroscopy 2003;3:234-238. 18. Schneeberger AG, Von Roll A, Kalberer F, Jacob HAC, Gerber C. Mechanical strength of arthroscopic rotator cuff repair techniques. J Bone Joint Surg Am 2002;84:2152-2160. 19. Lewis CW, Schlegal TF, Hawkins RJ, James SP, Turner AS. Comparison of tunnel suture and suture anchor methods as a function of time in a sheep model. Biomed Sci Instrum 1999; 35:403-408.