Suture anchors are used to attach soft tissues to bone. Numerous types of anchors are available, and studies have been done to evaluate their fixation strength.1,6,15 The strength of the repair depends on the soft tissue material properties, the stitch used, the suture material properties, the anchor design, and the material properties of the bone in which the anchor is placed.
Rotator cuff repairs using suture anchors can fail at the anchor-bone interface. Approximately 4% of failures of rotator cuff repairs that require revision surgery have been reported to be attributable to metal suture anchor pull-out from bone.4,5 However, the incidence of suture anchor pull-out in painful shoulders that previously had rotator cuff repair with bioabsorbable suture anchors has been reported to be much greater. In one series in which MRI findings in such patients were described, 30% had dislodgement of the suture anchor from the humeral head and retearing of the rotator cuff tendon.9
Poor bone quality in the region of the greater tuberosity of the humerus can increase the risk of suture anchor pull-out after rotator cuff repair. This may be attributable to osteoporosis, as patients who have rotator cuff tears develop are in the same age group as patients with osteoporosis. In addition, chronic disuse can lead to local osteopenia in the greater tuberosity.11,13 Chronic impingement can lead to the development of subchondral bone cysts that are located where an anchor would need to be placed, and in the revision situation, prior surgery and anchor placement may compromise bone quality in the region of the greater tuberosity.
Polymethylmethacrylate (PMMA) bone cement has been used to augment screw fixation in the spine3,7,16 and in trauma situations.12 In addition, PMMA alone has been used to anchor suture to bone without use of a traditional suture anchor.10 However, PMMA has not been studied as a means to augment suture anchor fixation. We hypothesized that suture anchor fixation can be augmented with PMMA cement, and that PMMA can be used to improve suture anchor fixation in a stripped anchor hole.
MATERIALS AND METHODS
Six matched pairs of proximal humeri were obtained and cleared of all soft tissue attachments. The age of the donors at death averaged 78 years (range, 60-91 years). Half were males and half were females. Dual energy xray absorptiometry (DEXA) scans were performed on each humeral head to investigate side-to-side variability in bone mineral density (QDR 4000 scanner, Hologic Inc, Waltham, MA). Each humeral shaft then was transected 10 cm distal to the inferior edge of the articular cartilage and potted in PMMA. The assignment of right or left side to each of the testing protocols, described below, was randomized.
Metal screw-like suture anchors (5-mm Fastin RC; Mitek, Norwood, MA), were used in this study. Two sites in each humerus were chosen for anchor insertion. They were located 1 cm and 3 cm posterior to the bicipital groove in the trough between the articular surface and the greater tuberosity. The right and left humeri were assigned randomly to one of two experimental groups. Both sites on each humerus received the same experimental treatment. These sites were spaced sufficiently far apart that there was no influence of one site on another.
Six humeri (one from each of the six matched pairs of humeri) were selected. In each humerus, two anchors were screwed into the bone 2 cm apart at the sites described above and per the manufacturer's instructions. These anchors were then individually cyclically loaded to failure (protocol to be described below). This was called the intact bone data group. These tests were done to establish a baseline for comparison of the other tests.
A volume of approximately 1.5 cm3 PMMA was mixed at room temperature and injected with a syringe into each stripped anchor hole. No additional attempt was made to pressurize the cement beyond the pressure that was applied as the cement was injected into the hole with the syringe. When the PMMA cured to a doughy consistency, a new suture anchor was inserted into each hole and the PMMA was allowed to fully harden. During insertion, less than ½ cc PMMA was extruded from the hole. The volume of this extruded cement was not measured specifically, however. The anchor eyelet was protected from cement by a sheath that is part of the standard anchor insertion instrumentation. The suture anchors in the cement-augmented stripped holes then were cyclically loaded per the same protocol with cycles to failure and maximum load recorded. This was the stripped anchor hole + cement data group. These tests were done to determine whether PMMA augmentation can restore or improve fixation of suture anchors in stripped anchor holes.
Using the six matched contralateral humeri that had not yet been tested, anchor sites were prepared by placing new suture anchors as described under Experiment 1. These suture anchors then were unscrewed without stripping the hole. These holes were injected with 1 cc PMMA and the suture anchors were replaced. The suture anchors were cyclically loaded per the same protocol with cycles to failure and maximum load recorded. This was the intact bone + cement data group. This test was done to determine whether cement augmentation can increase the strength of fixation of suture anchors in unstripped anchor holes.
Suture anchors were pulled on along their axis of insertion (Fig 1). We thought that this represented a worst-case scenario and would be the best test of the anchoring capacity of the suture anchor. Mechanical testing of the suture anchor fixation was done according to a previously published protocol.15 The suture anchors were cyclically loaded with a preload of 4 N and with a load rate of 25 N/second. A 50-N maximum load was chosen for the first 10 cycles, and this was increased in 50-N increments after each 10 cycles with a maximum of 40 cycles performed. After 40 cycles (maximum load of 200 N), the suture anchor was loaded at a linear rate of 1 mm/second until failure. For each test, cycles to failure and the maximum load during the test were recorded. Twelve tests were done for each condition (intact bone, stripped anchor hole + cement, and intact bone + cement). All tests were done at room temperature and in air. The bone was kept moist during the testing.
Initial tests were done using a 0.51-mm diameter wire threaded through the eyelet of the suture anchor. In four initial tests, the wire broke before failure of the suture anchor. The results from these four tests and anchor sites were discarded from the final data set. To test the fixation of the anchor alone and remove wire breakage as a mode of failure, suture anchors then were threaded with 0.64-mm diameter stainless steel wire and were tested in the eight remaining fresh test sites. The wire was clamped to a materials testing machine (858 Mini Bionix, MTS, Eden Prairie, MN) with a distance from the bottom of the clamp to bone surface set at 4 cm. None of the tests with the 0.64-mm wire failed by wire breakage. Only the eight tests in each group done with the 0.64-mm diameter wire that ultimately failed in all cases by anchor pull-out are reported.
A repeated measures analysis of variance was chosen because we repeated measurements at the same site on the same bone, and the same site in the contralateral matched-pair bone. This study design and statistical analysis technique minimize concern regarding variability of bone density and trabecular orientation from site to site. Cycles to failure and maximum load were compared for the three groups. A Fisher protected least significant difference (PLSD) followup test was used to determine significance between each group with the significance level set at 0.05 (Statview v 5.0.1, SAS Institute, Cary, NC).
Bone mineral density as measured by DEXA scanning of the humeral head specimens averaged 0.44 g/cm2 (standard deviation = 0.03 g/cm2). Side-to-side difference in bone mineral density averaged 0.01 g/cm2, with a maximum difference of 0.04 g/cm2.
For unstripped anchors, the average number of cycles to failure with our testing protocol was 34% greater for anchors placed in PMMA-augmented holes compared with anchors inserted into bone without PMMA (38.1 versus 28.5 cycles to failure; p = 0.0005) (Table 1). The average maximum load carried by the suture anchor was 71% greater for anchors placed in PMMA-augmented holes compared with anchors inserted into intact bone (294 N versus 172 N; p = 0.0212) (Table 2).
For stripped anchor holes that are augmented with cement, the average number of cycles to failure with our testing protocol was 31% greater than the number of cycles to failure in the intact bone with an uncemented anchor (37.4 cycles versus 28.5 cycles; p = 0.0010). The average maximum load carried by the stripped, cemented suture anchor was 111% greater than the average maximum load carried by the anchors inserted into intact bone without PMMA (363 N versus 172 N; p = 0.0012).
No difference was seen in cycles to failure or maximum load between suture anchors cemented into holes drilled into intact bone or suture anchors cemented into stripped holes.
Suture anchors are used to attach soft tissues to bone, and pull-out failure can be a concern. Osteoporosis, disuse, fracture, poor bone preparation, cysts in the bone, or prior surgery can leave a bony bed of questionable quality. Having a way to augment suture anchor fixation to decrease the risk of suture anchor pull-out, or to improve fixation after an anchor strips out of bone, can be of value. With the ramped cyclic loading protocol used in this study, we found that cementing a suture anchor in place, regardless of whether it is in an intact or a stripped anchor hole, will increase cycles to failure by approximately 31% to 34% and maximum load by 71% to 111% compared with an uncemented anchor in intact bone.
When placing a suture anchor, the surgeon has made a decision regarding what anchoring position will lead to the most anatomic repair. If the suture anchor does not hold after it is placed, various options exist. Choosing another site can be problematic because it will lead to fixation at a different anatomic location, and this location, if it is near the original pull-out site, may have compromised bone from the nearby pull-out or from the generally poor condition of the bone that may have led the first anchor to not hold. Deeper placement of the suture anchor at the same site may be attractive from an anatomic standpoint, but does not guarantee improvement in bone quality and puts the suture at risk from rubbing against the edge of the hole.2 Transosseous suturing is an option, but usually requires a more extensive exposure, and still depends on the strength of the native bone at the bone bridge to achieve fixation. Although choosing another anchoring site, placing the anchor deeper at the same site, and electing to switch to transosseous suture fixation are options to be considered when a suture anchor does not hold after it is placed, cementing of the hole and replacement of the anchor in the cement will improve the strength of fixation, will not change the anatomic location of anchoring, will not introduce a risk of suture failure at the edge of the hole, and will not require a more extensive exposure to pass and tie transosseous sutures.
Polymethymethacrylate likely augments anchor fixation by interdigitating with a greater surface of bone than the anchor alone (Fig 2). By interdigitating with more bone, more bone is involved in resisting suture anchor pull-out, thereby improving fixation. The degree to which PMMA will augment fixation and resist pull-out failure ultimately depends on multiple factors including the volume of cement injected, the viscosity of the cement injected, temperature, and pressurization. In addition, the bone must resist pull-out, and the material properties of the bone also influence the results. The degree to which these and related variables influence the effect of cement augmentation was not investigated and was considered outside the scope of this study.
Suture anchor fixation has been evaluated with a cyclical test using increasing load.14,15 We used a previously published loading protocol15 to be able to compare our results with those of other investigators. Without cement augmentation, we found that the maximum load applied to pull out the suture anchors in this study averaged 172 N. Other studies have found comparable and greater failure loads for various suture anchor types under similar ramped cyclic loading conditions.14,15 As a reference, the average uncemented suture anchor failure load we found in this study is greater than the failure load of simple and horizontal mattress sutures through rotator cuff tendon, but less than that of a Mason-Allen stitch or a so-called massive cuff stitch.8
Meyer et al,10 in describing a technique for using cement alone (without a suture anchor) to fix sutures in bone, recommended drilling a hole 3.5 mm in diameter and 10 mm deep, rinsing the hole, carefully placing a knotted suture at the base of the hole, injecting cement that was mixed at room temperature into the hole using a syringe, and then compacting the cement in the hole using a blunt wire. In our study, the bone holes were not of a uniform dimension. In intact bone, the holes generally were narrower as they were created by directly screwing the anchor into the bone. In stripped holes, the hole had been created by the pull-out of a suture anchor. These holes tended to be larger and less uniform in shape than the drilled holes in the study by Meyer et al. Cement was mixed and inserted in a similar fashion. We did not rinse the hole before placement of the suture anchor, but the anchor likely acted as the blunt wire did to compact the cement and cause it to interdigitate with the surrounding cancellous bone. The average pull-out forces that we report in this study are greater than reported in the study by Meyer et al,10 probably because we eliminated suture breakage as a mode of failure by threading the suture anchor with wire. In addition, the anchor eyelet was kept free of cement by a sheath that is a part of the standard insertion instrumentation. Keeping the eyelet free of cement also should be possible in vivo. This is a potential advantage over simply embedding the suture in PMMA, as a suture would be able to run through the eyelet of the anchor.
Forty cycles was the maximum number of cycles that could be attained in our experimental design. After 40 cycles, the suture anchors that still were embedded in bone were pulled out with a load to failure test. The absolute number of cycles the suture anchor will withstand before pulling out is dependent on the magnitude and direction of the applied cyclic load. The number of cycles to failure in our study can only be compared with other specimens that have been loaded with an identical loading regimen. Although testing of anchors by pulling along the axis of insertion often is considered the worst-case scenario, off-axis loading is more physiologically representative.
To what extent PMMA would increase load to failure in a simple pull-out test was not investigated as this is not a realistic failure mechanism. To what extent PMMA would augment cycles to failure in a traditional fatigue test with a single maximum load would depend on the chosen load. Based on the results of our study, one can be certain that in a fatigue test cycling from 0 to 50 N (the lowest cyclical load used in our study), the increase in cycles to failure with PMMA augmentation of suture anchor fixation would be greater than the 31% to 34% increase we found with our ramped cyclic loading protocol.
One difficulty that may arise with PMMA augmentation of suture anchor fixation involves administration of the cement into the suture anchor hole. While injecting PMMA into a suture anchor hole, care must be taken to keep the PMMA from spilling out of the hole, as loose pieces of PMMA may act as foreign bodies in the joint. For this reason, we think that this technique currently cannot be done arthroscopically, and we inject cement through a mini-open approach (Figure 3). We are studying the use of bioabsorbable cement to augment suture anchor fixation.
When the cemented suture anchors pulled out from bone, they typically removed more bone with them than the uncemented suture anchors. Pull-out of cemented anchors typically left a hole measuring from 6 to 9 mm in diameter, which equaled approximately the diameter of bone that had been infiltrated with cement. In one case (Test Site 3-intact bone + cement) pull-out of the suture anchor resulted in a fracture of the greater tuberosity that hinged on periosteum and cortical bone laterally. This fracture occurred in the specimen with the greatest pull-out force (623 N). In the clinical setting, this likely would have necessitated greater exposure and transosseous suture fixation for repair.
The fact that more bone is disrupted after cement augmentation is indicative of the involvement of more bone in resisting pull-out failure, and the greater force and energy needed to cause failure. Although the holes left behind by cemented anchor failure generally were larger than the noncemented anchor failure holes, none of the holes left by cemented anchor failure disrupted a large enough part of the bone to make recementation or traditional revision open repair with transosseous sutures impossible.
In the longer term, success depends on the healing of the tendon back to bone. The suture anchor will only play a role in the initial fixation of the tendon while the biologic healing process takes place. Although one may ask whether cement intrusion into bone could limit tendon healing to bone, we think that this is unlikely, as the cement inter-digitates only 1 to 2 mm into the cancellous bone of the humeral head around the suture anchor, and this interdigitation is deep to the tendon-bone contact surface (Fig 2). Although we have not specifically studied this issue, we expect that there would still be abundant surface area of tendon-bone contact outside this potential zone of influence for biologic healing to take place.
We investigated the augmentation of suture anchor fixation with PMMA bone cement. Regardless of whether the suture anchor hole is stripped, cementing the suture anchor in place with PMMA results in more cycles to failure and greater maximum load compared with non-PMMA-augmented suture anchors in intact bone. Polymethylmethacrylate can be used to augment fixation of a suture anchor or to recover and improve fixation in a stripped anchor hole, reducing the risk of anchor pull-out failure.
We thank Mitek for providing the suture anchors used in this study.
1. Barber FA, Herbert MA, Richards DP. Sutures and suture anchors: update 2003. Arthroscopy
2. Bynum CK, Lee S, Mahar A, Tasto J, Pedowitz R. Failure mode of suture anchors as a function of insertion depth. Am J Sports Med
3. Cook SD, Salkeld SL, Stanley T, Faciane A, Miller SD. Biomechanical study of pedicle screw fixation in severely osteoporotic bone. Spine J
4. Cummins CA, Murrell GA. Mode of failure for rotator cuff repair with suture anchors identified at revision surgery. J Shoulder Elbow Surg
5. Djurasovic M, Marra G, Arroyo JS, Pollock RG, Flatow EL, Bigliani LU. Revision rotator cuff repair: factors influencing results. J Bone Joint Surg Am
6. 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
7. Jang JS, Lee SH, Rhee CH, Lee SH. Polymethylmethacrylate-augmented screw fixation for stabilization in metastatic spinal tumors: technical note. J Neurosurg
. 2002;96 (1 suppl): 131-134.
8. Ma CB, McGillivray JD, Clabeaux J, Lee S, Otis JC. Biomechanical evaluation of arthroscopic rotator cuff stitches. J Bone Joint Surg Am
9. Magee T, Shapiro M, Hewell G, Williams D. Complications of rotator cuff surgery in which bioabsorbable anchors are used. AJR Am J Roentgenol
10. Meyer DC, Jacob HA, Pistoia W, von Roll A, Gerber C. The use of acrylic bone cement for suture anchoring. Clin Orthop Relat Res
11. Meyer DC, Fucentese SF, Koller B, Gerber C. Association of osteopenia of the humeral head with full-thickness rotator cuff tears. J Shoulder Elbow Surg
12. Motzkin NE, Chao EY, An KN, Wikenheiser MA, Lewallen DG. Pull-out strength of screws from polymethylmethacrylate cement. J Bone Joint Surg Br
13. Neer CS2nd, Craig EV, Fukuda H. Cuff-tear arthropathy. J Bone Joint Surg Am
14. Schneeberger AG, von Roll A, Kalberer F, Jacob HAC, Gerber C. Mechanical strength of arthroscopic rotator cuff techniques: an in vitro study. J Bone Joint Surg Am
15. Tingart MJ, Apreleva M, Zurakowski D, Warner JJ. Pullout strength of suture anchors used in rotator cuff repair. J Bone Joint Surg Am
16. Wittenberg RH, Lee KS, Shea M, White AA3rd, Hayes WC. Effect of screw diameter, insertion technique, and bone cement augmentation of pedicular screw fixation strength. Clin Orthop Relat Res