Meniscal repair devices: a clinical and biomechanical literature review

Meniscal Repair Devices: A Clinical and Biomechanical Literature Review Eugene Farng, B.S.E., and Orrin Sherman, M.D.

Purpose: The development of new approaches to arthroscopic meniscal repair has spurred the concomitant publication of studies reviewing their use and biomechanical properties. The purpose of this article is to review both the devices and the literature surrounding their clinical and biomechanical properties. Type of Study: Literature review. Methods: Studies were initially gathered using a MEDLINE search, and additional information was found through cross references. We evaluate a series of studies comparing sutures, suture anchors, screws, staples, and a variety of other devices in terms of initial fixation strength, degradation profile, performance under cyclical loading, and clinical success. Results: In the traditional suture studies, vertical sutures are clearly superior to both horizontal sutures and knot-end techniques in terms of initial fixation strength and performance under cyclical loading. Unfortunately, multidevice studies have been less consistent and less conclusive. the Linvatec Biostinger, Smith & Nephew T-fix, and Bionx Meniscus Arrow have separately been shown to have superior initial fixation strength on par with suture techniques. After cyclical loading, horizontal sutures, vertical sutures, 16-mm Arrows, 13-mm Arrows, and the Smith & Nephew T-fix generally show higher fixation strengths. Only the Bionx Arrow, Linvatec Biostinger, and Clearfix Screw have been shown to retain their initial fixation strengths through four months of hydrolysis time. Conclusions: Data suggest that the biomechanical performance of some devices is nearly equivalent to current suture techniques. Ultimately, the combination of a simplified surgical technique, high clinical healing rates (75%-92%), and relatively minor complications makes these devices attractive for properly indicated meniscal tears. Key Words: Meniscal repair—Tensile strength— Cyclical loading—Absorption profile—Bioresorbable.

H

istorically, the discussion about meniscal repair has been characterized by the debate of meniscectomy versus meniscal repair. Some argued that meniscectomy was benign and that no significant differences were found in clinical results between partial and total meniscectomy.1 Others argued that meniscectomy led to long-term degenerative changes.2-5 With the conclusion that the menisci increases the surface area for femoral-tibial load-transmission,6-12 aids in the mechanics of joint lubrication,13 and acts as

From the Department of Orthopedic Surgery, New York University School of Medicine, New York, New York, U.S.A. Address correspondence and reprint requests to Orrin Sherman, M.D., Skirball Institute, Suite 8U, 530 First Ave, New York, NY 10016, U.S.A. E-mail: [email protected] © 2004 by the Arthroscopy Association of North America 0749-8063/04/2003-3492$30.00/0 doi:10.1016/j.arthro.2003.11.035

a secondary anterior-posterior stabilizer in anterior cruciate ligament (ACL)-deficient knees14,15 the focus of treatment has shifted toward preservation and repair of the meniscus whenever possible. With repair currently the preferred treatment, a number of different repair techniques have evolved. Repair techniques generally fall into 4 categories: open, arthroscopic inside-to-outside, arthroscopic outside-to-inside, and arthroscopic all-inside. Open repair techniques were used initially,16,17 but arthroscopic inside-to-outside approaches soon evolved18-20 to minimize the risks associated with open surgery and enable access to portions of the meniscus that are difficult to reach with an open approach. An outsideto-inside approach was later recommended21,22 to minimize the risk to posterior neurovascular structures. Eventually, an all-inside technique for posterior horn tears was developed23,24 that obviates posterior

Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 20, No 3 (March), 2004: pp 273-286

273

274

E. FARNG AND O. SHERMAN

FIGURE 1. Meniscal suturing techniques. (A, B, and C) Variations on vertical suturing techniques, (D) horizontal mattress stitch, and (E) knot-end stitch. (Copyright 1997 by SpringerVerlag. Reprinted with permission.45)

capsular exposure, further reducing neurovascular risk. Most recently, a number of all-inside arthroscopic meniscal repair devices have been introduced to the market, touting simpler surgical techniques, shorter surgical times, and reduced surgical risk. The purpose of this article is to review some of the more popular meniscal repair devices available to surgeons. We will review studies concerning the biomechanical strength compared with traditional suture techniques, considering both time and strength required for adequate apposition and meniscal healing. Finally, we examine some of the preliminary clinical data concerning these devices.

properly positioned. The attached 2-0 nonabsorbable monofilament suture can be tied according to the surgeon’s preference. Its has been used in simple repairs (such as vertical posterior horn tears, bucket handle tears, flap tears, and horizontal tears),26,27 as

SUTURE TECHNIQUES Irrespective of the surgical technique used to access the meniscus, generally 3 different suture techniques can be used. As shown in Fig 1, knot-end techniques22 use large, intra-articular knots to retain the central portion of the meniscus, while horizontal25 and vertical sutures loop through the central meniscus fragment. REPAIR DEVICES Aside from sutures, a large number of devices have been specifically designed for meniscus repair and are being marketed as such (Fig 2). The T-fix (Smith & Nephew Endoscopy, Andover, MA) suture anchor is an arthroscopically placed device that uses no posterior portals. It consists of a 17-gauge needle preloaded with a permanent 3-mm polyacetyl anchor that is deployed with a wire obturator when the needle tip is

FIGURE 2. (A) Meniscal repair devices: SDsorb Staple, Mitek Meniscal Repair System, Biomet Staple. (B) Meniscal repair devices: left to right, Mitek Meniscal Repair System, Clearfix Screw, Arthrex Dart, Bionx Meniscus Arrow, Linvatec Biostinger, Smith & Nephew T-fix, 2-0 Ethibond suture.

MENISCAL REPAIR DEVICES well as more complex repairs involving a fascial sheath and exogenous blood clot.28 Studies of endoscopically deployed anchors in cadaveric knees show little neurovascular risk to posterior structures, although tenodesis may be possible.29 Smith & Nephew has since introduced a follow-up device to the T-fix named the FasT-fix. This device consists of two 5-mm polymer suture anchors connected via a pretied, sliding, self-locking knot comprised of No. 0, nonabsorbable USP braided polyester suture. After the first anchor is inserted and selfdeployed, the second anchor is advanced and deployed a few millimeters away. A sliding, self-locking knot is then pushed down, maintaining the meniscus in a reduced position. In 1993, Albrecht-Olsen et al.30 described the development of a new bioresorbable arrow-like device made of self-reinforced polylactic acid (SR-PLLA). The Meniscus Arrow (Bionx, Blue Bell, PA) consists of a 10-, 13-, or 16-mm-long barbed stem to anchor the arrow into the peripheral meniscus and a T-shaped head to retain the central meniscus. Although the original design used a drill and pneumatic reciprocating instrument to insert the device, a gun-style crossbow is also available, improving the device’s ease-ofuse. Tests of crossbow-inserted Arrows have been shown to have equal or greater strength compared with traditionally inserted Arrows in pullout tests.31,32 Advantages of this device include lowered neurovascular risk as well as shorter surgical time.33 However, cases of loosening requiring removal,34 inflammatory foreign-body reaction,35 chondral injury,36,37 arrow fracture,38 aseptic synovitis,39 and cystic hematoma formation40 have all been reported. Two other similar tack-like devices are the Arthrex Meniscal Dart (Arthrex, Naples FL) and the Linvatec Biostinger (Linvatec, Largo, FL). The Arthrex Dart is composed of poly-L,D-lactic acid (PLDLA) with reverse barbs along a 10-, 12-, or 14-mm shaft, and the Linvatec BioStinger is a poly-L-lactic acid (PLLA) device with a barbed 10-, 13-, or 16-mm shaft. Both devices also have gun-style delivery mechanisms to simplify the surgical procedure. Because of the recent entry into the market, little or no literature devoted solely to these devices is available, and complications have not been widely published. Ethicon, a Johnson & Johnson subsidiary, produces 3 different meniscal repair devices. The Clearfix Meniscal Screw, acquired in 1999 when Johnson & Johnson bought Innovasive, is a needle-loaded, 2 ⫻ 10 mm, poly-L-lactide (PLLA) variable-pitched screw that uses its threads to compress the torn meniscus.

275

Mitek Surgical Products, acquired by Ethicon in 1995, produces the other 2 devices. The Mitek Meniscal Repair System (Mitek, Westwood, MA) is a molded polymer device available in permanent polypropylene (Prolene) or absorbable polydioxanone (PDS) at lengths of 6 or 8 mm. It consists of a single curved shaft connecting 2 perpendicular cross-bars; the cross-bars serve to anchor the device into the tissue at either end. Finally, the newer Mitek RapidLoc (Mitek) consists of a PLA “backstop” soft-tissue anchor, a connecting suture (2/0 absorbable Panacryl or 2/0 permanent Ethibond [Ethicon, Somerville, NJ]), and a PLA “top hat” for meniscal apposition. After the soft-tissue anchor is deployed behind the tear, the top hat is advanced along with a pretied knot to maintain reduction of the tear. Because it is a relatively new device, no independent biomechanical studies of the RapidLoc have been published. Finally, staples represent the remaining approach to meniscal repair devices. The SDsorb Staple (Surgical Dynamics, Norwalk, CT) is an absorbable copolymer of a 72% polylactic acid and 18% polyglycolic acid. It consists of two 7-mm posts connected with a 4-mm braided cable, and they are delivered with a singlehanded disposable instrument loaded with 6 staples. Product literature claims 70% strength at 6 weeks with total absorption in 1 year. The Biomet meniscal staple (Biomet, Warsaw, IN) is a single rigid staple comprised of 82% PLLA and 12% polyglycolic acid (PGA). It can be inserted manually or with CO2 powered devices, and it is available in 11- or 13-mm lengths. BIOMECHANICAL STUDIES A number of studies have been published examining the biomechanical properties of various meniscal repair techniques. With some exceptions that are detailed below, these studies generally share a similar testing methodology. First, medial or lateral menisci are obtained from human, porcine, or bovine models. The menisci are usually isolated and excised, although 2 of the studies41,42 reviewed retain the meniscotibial attachments. A full-thickness vertical lesion is created in the peripheral third of the meniscus, a few millimeters away from the peripheral edge. The lesion is then repaired at a single site. The lesion is then completed anteriorly and posteriorly to ensure that the entire load is transmitted at the repair site and not through competent meniscal tissue. The 2 halves of the meniscus are then mounted on a testing machine, and the repair technique is loaded to failure under

276

E. FARNG AND O. SHERMAN

FIGURE 3. Suture-based meniscal repair techniques are loaded to failure in the meniscus. Kohn et al.43 use no simulated lesion. Kohn et al.43 and Post et al.41 pull directly on the suture. All other studies create a simulated, longitudinal, vertical tear in the peripheral third of the meniscus that is repaired with a single technique. The 2 meniscus halves are loaded separated, and the load-to-failure is recorded. Vertical sutures show 2⫻ to 4⫻ the failure strength when compared with knot-end techniques, and 1⫻ to 2.3⫻ the failure strength when compared with horizontal techniques. Although Seil et al.54 showed no difference between the failure strength of vertical or horizontal techniques, vertical sutures produce less tissue gapping under cyclical loading conditions.

tension. In general, only catastrophic failures are considered endpoints, and repairs are considered to fail when the device ruptures or pulls through the meniscal tissue. Only one study reported tissue gapping as a formal endpoint.55 These studies are summarized graphically in Figs 3 and 4.

loop, and a single vertical loop. Because the authors found that “the failure strengths were . . . unaffected by position in the meniscus or by intermeniscal differences,” menisci were reused and the same meniscus was loaded to failure several times with new sutures. Perhaps because of the higher average sample age (67 years), different suture material (3-0 Ethibond), faster rate of loading (50 mm/min), or the simulated lesion, ultimate failure strengths reported by Rimmer et al. were considerably lower than those reported by Kohn and Siebert. Still, the authors conclude that vertically oriented sutures are twice as strong as horizontally oriented sutures (29.3 ⫾ 8.4 N, P ⬍ .001), and that double sutures (63.2 ⫾ 14.5 N) are not stronger than single sutures (67.3 ⫾ 5.8 N). Therefore, the authors recommend single vertical loop sutures because they are more easily inserted. A 1997 study by Post et al.41 examines vertical mattress sutures, horizontal mattress sutures, and knot-end techniques using different suture materials in porcine menisci. Unlike in other experiments, the meniscus was left attached to the tibia. Furthermore, no knot was tied on the joint capsule; rather, the suture was directly loaded until suture rupture or tissue failure of the inner meniscal fragment. As expected, vertical sutures were significantly stronger than horizontal or knot-end techniques (P ⬍ .0001). Furthermore, as in previous studies,43,44 horizontal and knot-end techniques failed primarily by pulling out of the tis-

SUTURE STUDIES In 1989, Kohn and Siebert43 published one of the earliest studies on the biomechanical strength of meniscal repair techniques (Fig 3). With a loading rate of 5 mm/min in excised cadaveric menisci (17 to 41 years), vertical Vicryl sutures (Ethicon) showed a pullout strength of 105 ⫾ 4 N. This is significantly stronger than horizontal Vicryl sutures (89 ⫾ 4 N, P ⬍ .01), horizontal Ethibond sutures (44 ⫾ 18 N, P ⬍ .01), or knot-end techniques (24 ⫾ 9 N, P ⬍ .01). The authors propose that vertical sutures capture more circumferential collagenous fibers, providing higher failure strengths. Unlike many of the other studies reviewed below, no meniscal lesion was simulated, and the deforming force was loaded directly onto the suture. Furthermore, 3 different suture types were used, making direct comparisons with other studies difficult. In 1995, Rimmer et al.44 published another human cadaveric (mean, 67 years) study examining 3 types of arthroscopic meniscal suturing techniques using 3-0 Ethibond: a single horizontal loop, double vertical

FIGURE 4. A complete, longitudinal, vertical tear is created in the peripheral third of the meniscus and repaired with a single technique. The 2 meniscus halves are separated, and the load-to-failure is recorded. Albrecht-Olsen et al.33 show that the Bionx Meniscus Arrows have the same pullout strength as horizontal sutures, and Arnoczky and Lavagnino67 show that they have the same pullout strength as vertical sutures. Most other studies show only 4-070% of the initial fixation strength compared with horizontal suture techniques.

MENISCAL REPAIR DEVICES sue, while the vertical technique failed via suture failure. Consequently, the failure strength of horizontal sutures showed no difference with variation in suture material, whereas vertical sutures showed a statistically significant difference between the failure loads using different suture materials. That is, vertical mattress sutures using 1-PDS showed much higher load to failure (146 ⫾ 17.1 N) than vertical mattress sutures using 2-0 Ethibond (89.3 ⫾ 23.8 N) (P ⬍ .05). Finally, in 1997, Asik et al.45 published a study examining the strength of vertical, vertical mattress, vertical loop, horizontal mattress, and knot-end sutures using 1-Prolene in bovine menisci loaded at 5 mm/min. Although the absolute fixation strengths are higher than those found in cadaveric models, the conclusions are the same. Vertically oriented sutures (approximately 131 N) show significantly higher initial fixation strengths when compared with knot-end (64 ⫾ 5 N, P ⬍ .001) or horizontal techniques (98 ⫾ 5 N, P ⬍ .001). Again, horizontal and knot-end techniques all failed by tissue failure, whereas 9 of 12 vertical techniques failed by suture rupture. SINGLE DEVICE STUDIES After the introduction of the biodegradable Meniscus Arrow, a number of biomechanical studies were performed comparing the new device to more traditional suture techniques (Fig 4).46-49 As its inventors, Albrecht-Olsen et al.46 published the first biomechanical study on the all-inside Meniscus Arrow, comparing a single horizontal Maxon-0 suture (Syneture, Norwalk, CT) to the new Meniscus Arrow in excised bovine menisci at a loading rate of 5 mm/min. With failure strengths of 54.88 ⫾ 11.86 N for the suture and 54.25 ⫾ 7.0 N for the Arrow, the authors concluded that no statistical difference existed between the pullout strengths of horizontal sutures and Arrows in bovine menisci. Shortly thereafter, Dervin et al.47 published a similar study of the 13-mm Arrow using the same experimental setup as Rimmer et al.44 in cadaveric (mean age, 67 years) menisci. A 2-0 Ethibond vertical loop suture technique was used as a comparison, consistent with previous studies showing the superiority of vertical sutures.41,43-45 Repairs were loaded at 50 mm/ min. The authors noted no statistical differences between failure strength in medial and lateral menisci. The authors concluded that the Meniscus Arrow (29.6 ⫾ 10 N) has approximately half the failure strength of vertical sutures (58.3 ⫾ 6.2 N) (P ⬍ .001). These results correspond roughly with previous con-

277

clusions that Arrows have a failure strength approximately equal to horizontal sutures,46 which in turn have half the failure strength of vertical sutures.44 The authors also note that tissue gapping occurs at comparatively lower forces with the Arrows, which could impair the healing process. Unfortunately, this parameter was not formally measured. In 1999, Boenisch et al.48 published a Bionx-funded study comparing the pullout strength of 2-0 Ti-Cron vertical loop sutures (Syneture), 2-0 Ti-Cron horizontal loop sutures, and 10-, 13-, and 16-mm Arrows in excised bovine menisci. Repairs were rapidly loaded at 750 mm/min. Perhaps because of this significantly higher loading rate, no statistical difference was seen between the vertical (72.4 ⫾ 9.2 N) and horizontal (68.3 ⫾ 9.5 N) suture groups. In 9 of 10 trials for both vertical and horizontal groups, the mode of failure was suture failure. Unlike the original study of AlbrechtOlsen et al.,46 in this study, the 16-mm Arrows (52.7 ⫾ 11.2 N) were found to be significantly weaker than the suture techniques (P ⬍ .05). The length of Arrow did have an impact on pullout strength; as expected, 13-mm Arrows (39.4 ⫾ 10.3 N) and 10-mm Arrows (18.5 ⫾ 9.9 N) showed lower fixation strength (P ⬍ .05). This difference was attributed to the number of barbs available to grasp meniscal tissue, because subsequent tests on 10- and 13-mm Arrows with an equal number of anchoring barbs displayed similar pullout strengths (35.1 ⫾ 9.9 N and 33.1 ⫾ 8.8 N, respectively). Finally, angled insertion was shown to have significantly lower pull-out strength when compared with Arrows inserted parallel to the tibial plateau (P ⬍ .05). In 1999, Song and Lee49 performed another study comparing biodegradable Meniscus Arrows to various suture techniques in the lateral menisci of 35 Yorkshire pigs loaded at 10 mm/min. Vertical sutures (113.9 ⫾ 14.6 N) were shown to be stronger than horizontal sutures (75.1 ⫾ 18.4 N, P ⬍ .05) or Arrows (38.3 ⫾ 4.3 N, P ⬍ .05). In contrast with previous studies, single Arrows failed at significantly lower loads than horizontal sutures. In fact, menisci repaired with 2 Arrows (56.5 ⫾ 3.5 N) were shown to be equivalent to knot-end suture techniques (53.9 ⫾ 6.4 N). In 2001, Karaoglu et al50 published a study of the T-fix device recommending obliquely oriented sutures. In bovine menisci loaded at 5 mm/min, the repair strength of 3 T-fix anchors arranged in a triangle with 2 obliquely oriented sutures (71.0 ⫾ 5.9 N) was greater than that of 3 T-fix anchors arranged in a line with 2 horizontally oriented sutures (61.8 ⫾ 3.8

278

E. FARNG AND O. SHERMAN

N) (P ⬍ .01). Like vertical sutures, the obliquely oriented sutures are thought to capture a larger number of circumferential collagenous fibers, allowing stronger tissue fixation. An in vivo canine study of another meniscal repair device, a modified SDsorb staple, was published in 1996 by Koukoubis et al.42 This study focused on a smaller version of the SDsorb staple, because the standard version was found to be too large for the canine knee. A 1-cm bucket-handle lesion was artificially created in the posteromedial third of the meniscus and treated with staples, a 3-0 PDS horizontal mattress suture, or left alone. Subjects had complete freedom to bear weight, and were later euthanized at 6 weeks, 4 months, or 1 year. Mechanical and histologic examination followed, allowing the authors to conclude that the staple provided equal or better fixation in the short term (up to 4 months). Furthermore, no long-term (1 year) differences were seen between the mechanical strength of healed menisci. By 9 months, the staples were almost completely resorbed. Unfortunately, direct comparison with other studies is not possible for a number of reasons. First, no initial fixation data were obtained; the earliest test was performed at 6 weeks. Second, no other study uses a canine model. Finally, the meniscus was not excised from the tibia, and the meniscotibial ligaments were preserved. Because a number of specimens failed at this attachment, failure strengths may not accurately reflect the strength of the repair device. MULTIDEVICE STUDIES In 2000, Barber et al.32 published the first study examining multiple meniscal repair devices as well as standard suture techniques. Experiments were performed in porcine menisci, and repairs were loaded at 5 mm/min (Fig 5). Eleven different repair methods were tested: vertical simple suture, horizontal mattress suture, 2 vertical simple sutures exiting midsubstance, manually inserted 13-mm Meniscus Arrow, Crossbow-inserted 13-mm Meniscus Arrows, 13-mm Biostinger, 10-mm Clearfix screw, Sdsorb meniscal staple, horizontal mattress stitch using 2 T-fix devices, 8-mm Mitek meniscal repair system, and 10-mm Biomet staple. The authors used 2-0 Mersilene for all suture repairs. Statistically, 4 distinct levels of fixation were found (P ⬍ .05). First, the double vertical stitch failed at 113 N. Second, the single vertical stitch failed at 80 N. The third group consisted of the Biostinger (57 N), horizontal mattress stitch (56 N) and the T-fix device (50

FIGURE 5. A complete, longitudinal, vertical tear is created in the peripheral third of the meniscus and repaired with a single technique. The 2 meniscus halves are separated, and the load-to-failure is recorded. In porcine menisci, Barber and Herbert32 show that the T-fix and Linvatec Biostinger have fixation strengths equal to horizontal mattress sutures. In human menisci, Becker et al.52 also show that the T-fix has fixation strength equal to horizontal mattress sutures, but the Linvatec Biostinger is significantly weaker. In bovine menisci, Arnoczky and Lavagnino67 show that the Bionx Arrow provides fixation strengths equal to vertical sutures and was significantly stronger than all other devices.

N). Finally, the fourth group consisted of the remaining devices: the Meniscus Arrow (33 N), the Clearfix screw (32 N), the Sdsorb staple (31 N), the Mitek repair system (30 N), and the Biomet staple (27 N). No difference in failure strength was found between crossbow or manually inserted Arrows. Unlike Rimmer’s earlier study,44 single vertical stitches are significantly weaker than double vertical stitches. The Meniscus Arrow was found to be significantly weaker than the horizontal suture. The failure mechanism of all devices was noted, and interestingly, the Meniscus Arrow and Biostinger both failed exclusively by pullthrough. That is, the anchoring head pulled through the central meniscal fragment every time, suggesting that pulling directly on the implant would fail to characterize load-to-failure correctly. In early 2001, Walsh et al.51 reported the ultimate failure strength of 3-metric horizontal Ethibond sutures, 3-metric vertical Ethibond sutures, Bionx Meniscus Arrows, and SDI Meniscal Staples. Repairs were performed in bovine menisci and were loaded at 50 mm/min. All 4 types of fixation showed statistically significant differences in failure strength (P ⬍ .005). The vertical sutures failed at 73.9 ⫾ 6.6 N, the horizontal sutures failed at 63.2 ⫾ 9.8 N, the Arrows failed at 44.3 ⫾ 10.9 N, and the Staples failed at 17.8 ⫾ 4.1 N. The following year, Becker et al.52 performed a

MENISCAL REPAIR DEVICES similar study using excised human menisci. Specimens were taken from patients (mean age, 57 years) receiving bicondylar knee arthroplasty because of varus osteoarthriosis. They examined 7 fixation techniques: horizontal mattress suture using 2-0 Ethibond, the Meniscus Arrow, Arthrex Dart, Linvatec BioStinger, Innovasive Meniscal Screw, Smith and Nephew T-fix, and the Mitek Meniscal Repair System. No statistical difference was seen between the 2 strongest techniques, the horizontal mattress suture (62.09 ⫾ 7.91 N) and T-fix techniques (51.35 ⫾ 16.31 N). The initial fixation strength of the Mitek fastener (32.67 ⫾ 2.97 N), was higher than the other remaining devices. The Bionx Arrow (24.70 ⫾ 6.39 N) and Linvatec Biostinger (25.44 ⫾ 10.20 N) had the same fixation strength, and the Clearfix screw (14.5 ⫾ 4.39 N) and the Arthrex dart were the weakest (9.52 ⫾ 2.33 N). Recently, in 2002, Rankin et al.53 published on the results of vertical sutures, horizontal sutures, Meniscus Arrows, and T-Fix repairs in bovine menisci loaded at 150 mm/min. Notably, repairs were performed using 3 sutures or devices for each repair, resulting in significantly higher fixation strengths. Vertical sutures (202 ⫾ 7 N) were stronger than horizontal sutures (170 ⫾ 12 N, P ⬍ .03), T-fix anchors (95.9 ⫾ 7.5 N, P ⬍ .0001), or Meniscus Arrows (99.4 ⫾ 7.5 N, P ⬍ .0001). Furthermore, analysis of tissue gapping using photoreflective markers and infrared cameras during loading showed that vertical sutures require significantly higher loads (109 ⫾ 7.3 N) to produce 1.0 mm of tissue gap when compared with horizontal sutures (35.5 ⫾ 3.5 N, P ⬍ .0001). Furthermore, the Arrow (19.1 ⫾ 3.5 N) and T-fix (17.4 ⫾ 2.1 N) device required even less force (P ⬍ .001). CYCLICAL LOADING Although initial fixation strengths are one measurement that can be used to evaluate the biomechanics of meniscal repair techniques, they fail to simulate physiologic repetitive loading conditions. To address this issue, Seil et al.54 evaluated both horizontal and vertical mattress sutures using absorbable 2-0 PDS and nonabsorbable 2-0 Ethibond in excised medial porcine menisci. Three different testing groups were loaded to failure at 50 mm/min: before cyclical loading, after 100 cycles between 5 and 20 N, and after 100 cycles between 5 and 40 N. Initial failure strengths (mean, approximately 60 N) showed no statistical difference between horizontal or vertical sutures or between su-

279

ture materials. Eight of sixty (13.3%) of the cyclically loaded sutures failed during cycling (7 at 40 N and 1 at 20 N). Furthermore, significant meniscal gapping was noted during the early portions of cyclical loading, with vertical sutures outperforming horizontal sutures in this regard (P ⬍ .001). Although the difference in ultimate failure strength after 100 cycles was not statistically significant in any of the groups, the failure of horizontal sutures to prevent tissue gapping in the early stages of loading suggests a failure of sufficient tear reduction. Thus, because vertical sutures show less tissue gapping, the authors conclude that they outperform horizontal sutures under cyclic loading. A similar study was performed by Bellemans et al.55 examining the ultimate failure strength and cyclical fatigue strength of vertical loop sutures, horizontal mattress sutures, the T-Fix device, Bionx Arrows (16, 13, and 10 mm), the Mitek Meniscal Repair System, SDI Staples, and the Arthrex Meniscal Dart. Fixation strength was examined using cadaveric lateral menisci (average age, 36 years). In the first half of the experiment, repairs were loaded at 50 mm/min until device, suture, or tissue failure. Based on the initial fixation strengths, the authors categorized the devices into 3 groups. The strongest group included the horizontal sutures (52.5 ⫾ 4.45 N), vertical sutures (46.3 ⫾ 8.05 N), T-fix, 16-mm Arrows (39.2 ⫾ 8.6 N), and 13-mm Arrows (32.8 ⫾ 8.8 N). The Mitek Meniscal Repair System displayed an intermediate fixation strength (18.1 ⫾ 3 N), and the 10-mm Arrow (18.8 ⫾ 2.65 N), SDSorb Staple (4.3 ⫾ 4.25 N), and Arthrex Dart (10.5 ⫾ 0.5 N) showed inferior fixation strength. In the second half of the experiment, cyclical loading was applied between 10 and 20 N at a frequency of 1 Hz and a speed of 50 mm/min. Performance under cyclical loading was considered to be the time or number of cycles until a gap of 3 mm was seen. Under cyclical loading, only 2 groups of devices were statistically distinguishable. The horizontal sutures (115 cycles), vertical sutures (185 cycles), T-fix (145 cycles), 16-mm Arrows (165 cycles), 13-mm Arrows (98 cycles), and Mitek Meniscal Repair System (105 cycles) maintained tissue apposition for a longer duration, and the 10-mm Arrow, SD Sorb Staple, and Arthrex Dart all failed within 15 cycles. As a follow-up to a previous study,52 Becker et al.56 examined the performance of multiple repair devices under cyclical conditions in human menisci. Repairs were loaded for 100 cycles between 5 and 15 N at 1 Hz, and then the repair was loaded to failure. At the start of cyclical loading, the Mitek Meniscal Repair

280

E. FARNG AND O. SHERMAN

System showed significantly higher levels of displacement (5.6 ⫾ 1 mm) compared with vertical sutures (1.8 ⫾ 0.6 mm), horizontal sutures (3.2 ⫾ 1.2 mm), Arrow (3.3 ⫾ 1.3 mm), Stinger (1.7 ⫾ 0.5 mm), or T-Fix (2 ⫾ 1.3 mm) techniques. Load to failure after cyclical loading was higher for the vertical sutures (61.1 ⫾ 3.4 N), horizontal sutures (57.4 ⫾ 6.8 N), and T-fix devices (48.7 ⫾ 7.2 N). The Arrow (32.7 ⫾ 5.8 N), Fastener (34.4 ⫾ 3.3 N), and Stinger (31.1 ⫾ 10.4 N) showed lower fixation strengths after cyclical loading (P ⬍ .05). Although vertical and horizontal sutures showed equivalent failure strengths after cyclical loading, vertical sutures showed superior stiffness and lower displacement during loading, leading the authors to recommend vertical techniques over horizontal ones. The Clearfix screw did not withstand the 15-N loading cycle. ABSORPTION PROFILE Another factor that can be expected to contribute to the in vitro biomechanical properties of these devices is their absorption profile. Because a number of these devices are made of PLA, polyglycolic acid (PGA), polydioxanone (PDS), or derivatives of these materials (SR-PLLA, PLDLA, PLA-co-PGA), their structural integrity and fixation strengths are expected to degrade with time. In vivo and in aqueous media, degradation of poly-alpha-hydroxy acids such as PLA and PGA proceeds via a combination of internal autocatalysis by carboxylic chain ends57 and random bulk hydrolysis of the ester bonds in the polymer chains.58,59 In the early stages, depolymerization results in a decrease of the molecular weight of the polymer, without any visible change in the implant or cellular response.58 Degradation eventually leads to fragmentation of the material to smaller parts, and when the molecular weight of the polymer drops below 5,000 D,58 a nonspecific foreign body reaction activates macrophages and polymorphonuclear leukocytes.59-61 Oligomers then undergo further hydrolytic degradation to endogenous substances or to compounds with well known metabolic pathways (lactic or glycolic acid). Factors intrinsic to the device such as molecular weight, sterilization technique, processing technique, crystallinity, thermal history, and geometry will affect the rate of degradation.62 Similarly, in vivo factors such as heat, movement, and nonenzymatic action can also influence the rate of degradation.63 In vivo, researchers have shown that the tensile strength of PLLA drops significantly between 6 and 12 weeks, with 3.2-mm braided fibers retaining only

16% of the initial tensile strength.64 Furthermore, in vivo degradation rates are higher than in vitro.64 Similarly, 90% of the initial bending strength of PDLLA cubs and rods is retained through 6 weeks of in vivo degradation, but drops to 60% by 12 weeks and 6% by 16 weeks.65 Therefore, when evaluating the use of these bioabsorbable meniscal repair devices, surgeons must consider both the duration of fixation that the repair will provide, and the time required for adequate healing. In 1997, Barrett et al.66 published a study examining the differences between absorbable 2-0 PDS and nonabsorbable 2-0 Ethibond on meniscal repair using a horizontal inside-out technique. In that study, 82 meniscal repairs were evaluated at an average of 24.1 months with clinical examination, range of motion testing, McMurray examination, detectable swelling, locking, joint-line tenderness, and a 15-question subjective visual analog evaluation by the patient. No second-look arthroscopy was performed. With no failures requiring subsequent surgery, permanent sutures showed a statistically significant (P ⫽ .02) difference when compared with absorbable sutures, which had an 18% repeat surgery rate. Furthermore, mean values of all 15 visual analog questions were worse for absorbable sutures, and Tegner scores of absorbable sutures (3.04) are lower than permanent sutures (4.25) (P ⫽ .0001). Thus, the authors conclude that permanent sutures are recommended over absorbable sutures as PDS, which appear to absorb too quickly to allow sufficient time of meniscal apposition for successful healing. Although inferring the loss of fixation strength from known material properties may be sufficient for simple sutures, this process is not applicable to complex devices, because their morphology can be expected to contribute significantly to implant degradation patterns and loss of fixation strength. Therefore, Arnoczky and Lavagnino67 examined the fixation strength of these absorbable devices as a function of in vitro hydrolysis time in bovine menisci. The repair strength of 2-0 PDS vertical suture, the Bionx Meniscus Arrow, Linvatec Biostinger, Innovasive Clearfix screw, Surgical Dynamics staple, and Mitek Meniscal repair system were studied at 0, 6, 12, and 24 weeks. Hydrolysis was simulated by incubating the repaired menisci at a pH of 7.4 and temperature of 37°C in phosphate-buffered saline containing 10% antibioticantimycotic solution as well as protease inhibitors. Control groups indicated that no significant decrease in fixation strength occurred because of tissue degradation over the 24 week study, and Arrows inserted

MENISCAL REPAIR DEVICES into 24-week incubated menisci showed the same fixation strength as Arrows inserted into fresh menisci. Initial fixation strength of the Bionx Meniscus Arrow (57.7 ⫾ 13.8 N) was equal to the 2-0 PDS vertical suture (51.7 ⫾ 2.7 N) and significantly higher than all other devices (P ⬍ .05). After 6 weeks of hydrolysis, none of the devices showed significant decreases in fixation strength. At 12 weeks, the PDS based devices (the Mitek Meniscal Repair System and the 2-0 PDS suture) showed significant weakening. The Mitek meniscus repair system lost all fixation, and the 2-0 PDS suture retained only approximately 35% of its initial fixation strength (P ⬍ .05). By 24 weeks, the Surgical Dynamics staple (a PLA and PGA copolymer), Mitek meniscus repair system, and 2-0 PDS suture failed to provide any repair strength at all. None of the other devices (Arrow, Biostinger, Screw) showed any loss of strength over the 24-week period. REQUIRED HEALING TIME Because many meniscal repair devices are absorbable, the length of time required for adequate meniscal healing must also be considered. A biomechanical study by Roeddecker et al.68 in rabbits examined the tearing energy of repaired versus normal menisci as a function of time. Simulated lesions were created in rabbit menisci and assigned to 1 of 3 treatment groups: no treatment, repair with 4-0 Vicryl suture, or repair with fibrin sealant. Subjects were allowed to heal for 6 or 12 weeks and then euthanized. The energy required to tear the Vicryl-repaired meniscal scar was 26.3% and 23.3%, respectively, when compared with the healthy meniscus. Similarly, at 6 and 12 weeks, the energy required to tear the fibrin glue-repaired scar was 42.50% and 42.51% when compared with the healthy meniscus. Thus, even after 12 weeks of healing, the scar still shows significantly lower strength compared with a noninjured meniscus. In 1991, Morgan et al.69 published a study of 74 arthroscopic meniscal repairs evaluated by secondlook arthroscopy, noting that about 4 months were required for visual evidence of meniscal healing to appear. The authors concluded that all asymptomatic repairs were fully or partially healed; clinical evaluation gave no false-negative results. REQUIRED STRENGTH Another consideration of interest is the fixation strength required to maintain the meniscus in a re-

281

duced position during ambulation and rehabilitation. Unfortunately, this information has not been accurately characterized. Biomechanical studies in the past have suggested that joint forces in the knee can be as high as 4 times body weight during level walking,70 and other studies claim that the menisci transmit up to 55%6 or even 99%12 of the load in the knee. Therefore, a 75-kg individual could be expected to transmit compressive loads of 1,250 to 3,000 N, and loads of this magnitude are clearly beyond the limits of individual repair techniques. Fortunately, this compressive load is transmitted across the meniscus, and only a fraction of this weight is borne by the radially directed repair device. At the other extreme, significantly lower forces in repair devices were shown by Kirsch71 in 5 freshfrozen human knees under non–weight-bearing conditions. A 3-cm peripheral longitudinal tear was created in the posterior horn of medial and lateral menisci and repaired with a specially prepared T-fix and load cell assembly. A simulated quadriceps force produced non–weight bearing, active knee movement through 100°, showing less than 10 N forces in the suture. CLINICAL OUTCOMES Clinical evaluations of these repair devices have also been reported; however, only the earlier devices have a large enough patient base and follow-up information for evaluation. In early 1997, Escalas et al.72 reported that the 6-month follow-up evaluation of 20 patients with T-fix repaired medial menisci reported showed that 90% of patients returned to preinjury activity levels. Later that year, Barrett et al.73 reported on the results of 21 meniscal repairs using the T-fix suture anchor with a minimum of 1-year follow-up evaluation. Evaluations for joint line tenderness, effusion, and McMurray testing concluded that 81% of the repairs were successful. Avascular zone 2 repairs showed significantly lower healing rates than the peripheral zone 0 or 1 repairs. With a 92% success rate on vertical or bucket-handle tears of zone 0 or 1 repairs, the authors conclude that the amount of apposition provided by the T-fix device is clinically sufficient. Similarly, early clinical studies of the Bionx Arrow also show encouraging results. Albrecht-Olsen et al.33 performed a prospectively randomized study of 68 patients with arthroscopic evaluation comparing the healing rate using Arrows versus the inside-out horizontal technique using Maxon-0 sutures. Unlike previous studies72-74 meniscal healing was evaluated with

282

E. FARNG AND O. SHERMAN

second look arthroscopy at 3 to 4 months. Only tears in the red-red or red-white regions were repaired. Healing rates were 91% for the Arrow group and 75% for the suture group. Only 5 failures were clinically detected; the other 6 were not discovered until repeat arthroscopy, suggesting that only half of all failed repairs present with significant clinical symptoms. The authors found that surgical time was cut in half using the Meniscus Arrow, and that “there was no statistical differences in healing (total and partial) between Arrows and sutures in the subgroups with isolated lesions, lesions in ACL-reconstructed knees, and lesions in ACL-insufficient knees.” H¨urel et al.74 reviewed 26 repairs (24 medial, 2 lateral) in 25 (16 men, 9 women) patients at an average follow-up time of 16.7 (range, 12-22) months. An average of 2.8 Arrows per tear were used. Using the modified Marshall rating system, 88% of patients were clinically rated as good or excellent. Complications included infection, hemarthrosis, and Arrowinduced irritation. The authors conclude that despite its increased cost, Arrows are an attractive alternative to traditional techniques because of the shorter surgical time, easier technique, and decreased risk of neurovascular injury. In 2001, Venkatachalam et al.75 reported on the results of 92 meniscus repairs using sutures, Meniscus Arrows, and T-Fix suture anchors. At an average follow-up time of 21 months, the authors report that repairs using sutures alone had higher clinical success rates (78.6%) than repairs using Arrows (56.5%) or T-fix devices (57.1%). The overall clinical success rate for the population was 66.1%. The overall complication rate of 11.3% included two isolated broken Arrows as well as two Arrows causing synovitis and articular cartilage damage. Unfortunately, as the authors note, surgical technique was not controlled or randomized, and selection bias may have influenced these results. Furthermore, the authors do not discuss the statistical significance of the results, making it difficult to conclude that suture techniques have a higher success rate than Meniscus Arrow or T-Fix suture anchors. Most recently, in the first half of 2002, 3 different studies reported similar findings concerning 2-year follow-up results of meniscal repair using the Bionx Meniscus Arrows. In January, Jones et al.76 reported only 2 clinical failures requiring a second operation in 39 repaired menisci (5%) at an average follow-up time of 29.7 months (minimum, 24 months). No failures were reported in subgroups requiring concominant ACL reconstruction; both failures occurred in patients

with isolated meniscal tears. In that study, 12 patients (31.6%) reported transient soft-tissue irritation or tenderness that generally resolved in 12 months. Two of these patients required removal of subcutaneous Arrow fragments, and two others with recurrent symptoms required repeat surgery and partial meniscectomy. In March 2002, Petsche et al.77 reported similar results, with 2 failures in 29 patients (6.8%) at an average follow-up time of 24 months (minimum, 12 months). Five of the patients (17.2%) reported local, transient skin irritation that resolved within 7 months. No surgery was required to remove Arrow fragments, and no complaints of parasthesias or neurologic symptoms were reported. Intermediate-term results (average, 2.3 years) of 32 patients undergoing meniscus repair using Meniscus Arrows and concurrent ACL reconstruction were reported by Gill and Diduch78 in 2002. Clinical criteria showed a success rate of 90.6%, with all failures occurring in the red-white zone. Rehabilitation protocol was unchanged from isolated ACL reconstruction protocols, with early weight bearing as tolerated and immediate range of motion exercises. Other than the failed repairs, the authors do not note any Arrowrelated complications. Also in 2002, Laprell et al79 reported on the results of 37 repaired menisci using the Mitek Meniscal Repair System. With a minimum follow-up time of 12 months, the authors found a clinical success rate of 86%. All failures occurred in the middle third of the meniscus (red-white zone). Seventeen patients underwent repeat arthroscopy at 6 to 8 weeks for an ACL reconstruction using bone–patellar tendon– bone autograft, and in all 17 cases, repaired menisci were still reduced and stable to probing. Complications included subcutaneous device migration as well as grade II chondromalacia of femoral condyle in an area corresponding to repair device. DISCUSSION Based on pilot studies, Post41 concluded that the load to failure does not vary based on the portion of the meniscus used, whether the meniscus was taken from the medial or lateral compartment, or if the meniscus is reused. Still, despite attempts to use similar experimental methodology, a large number of variations exist between studies, making direct interstudy comparisons very difficult. First, several different meniscus models have been used, including cadaveric, porcine, bovine, and canine. With the

MENISCAL REPAIR DEVICES characterization of the material properties of human80,81 and animal82,83 menisci, interspecies differences clearly exist. Statistically significant differences were found among human, bovine, monkey, canine, sheep, and porcine menisci when attempting to quantify biomechanical parameters (aggregate modulus and permeability) for use in a linear biphasic model.84 Sheep, the animal model whose biomechanical parameters most closely matched human menisci, were not used for any of the biomechanical studies reviewed here. This may be because of anatomic constraints, because sheep menisci are about half the size of human menisci. Another confounding factor includes the retention of the meniscotibial ligament. Although most of the menisci tested are excised and isolated, a few studies retain the meniscotibial ligaments and anchor the tibia to the testing apparatus. Third, variations in the geometry of the experimental setup, that is, the distance between the lesion, repair entry-site, and joint capsule, can significantly alter the amount of anchoring tissue available and therefore the failure strength of devices. Finally, the rate of loading varies between studies. This is significant because the meniscus is a viscoelastic material that displays variable load-deflection curves at different rates of loading.82 Whether or not this significantly alters the load-tofailure of the repair devices is unclear. When evaluating suture techniques, researchers generally accept that vertical sutures are superior to both horizontal and knot-end techniques (Fig 3). Nearly every suture study presented shows a higher initial fixation strength for vertical sutures when compared with horizontal or knot-end techniques.32,41,43-45,49 Only 2 studies show nonsignificant differences in suture strength.48,54 The absolute fixation strength for vertical sutures ranges from approximately 60 N54 to 146 N,41 and horizontal sutures range from 29 N44 to 98.5 N.45 Initial fixation strengths for knot-end techniques range from 24 N43 to 70 N.41 With significant variations in experimental design, however, interstudy comparisons of absolute strengths are misleading at best. Relative differences in strength also vary, with vertical sutures ranging from approximately 154 to approximately 2.344 times as strong as horizontal sutures. Vertical sutures have also generally shown superiority under cyclical loading conditions.54,56 Unfortunately, the results are less consistent for other meniscal repair devices. The Bionx Meniscus Arrow has received the most attention, with 10 studies reviewed here (Fig 4). Generally, with initial pullout strengths for 13-mm Arrows ranging from 29.6 N47 to 57.7 N,46 the Meniscus Arrow has less initial fixation

283

strength than suture techniques. Seven of the 10 studies show that 13-mm Arrows have from 40% to 70% of the initial fixation strength of horizontal sutures,32,48,49,51,52,55,56 but another claims 50% the initial fixation strength of vertical sutures.47 However, some studies suggest that Arrows are equivalent to suture techniques, showing that Arrows have 99% of the strength compared with horizontal sutures46 or that they are actually equal to vertical sutures.67 Interestingly, these studies were performed in bovine menisci. The other bovine study48 showed that 13-mm Arrows have lower initial fixation strengths than both horizontal and vertical suture techniques. Finally, 4 multidevice studies were reviewed (Fig 5), using 3 different meniscus models (cadaveric, porcine, and bovine). As the current gold standard, suture techniques are the baseline for comparison of these devices. Barber and Herbert32 concluded that vertical sutures were superior to all other techniques, while the Biostinger and T-fix were equal to horizontal suture techniques. All other devices (Bionx Arrow, Clearfix screw, SDsorb staple, Mitek Meniscal Repair System, and Biomet staple) provided inferior strength. Becker et al.52 also concluded that the T-fix provided fixation equal to horizontal sutures but also concluded that the Mitek Meniscal Repair System was stronger than the equally rated Biostinger and Arrow. The screw and the dart both provided inferior strengths. Arnoczky and Lavagnino67 concluded that Arrows provided fixation equal to the vertical sutures. The Mitek meniscal repair system, Innovative Biostinger, and Clearfix screw were found to have slightly lower fixation strengths, and the SDsorb staple was found to be the weakest. In cadaveric menisci, Bellemans et al.55 concluded that the vertical loop suture, horizontal mattress suture, T-Fix device, 16-mm Arrow, and 13-mm Arrow all provided equivalent fixation strength as well as superior performance under cyclical conditions. The Mitek Meniscal Repair System provided an intermediate fixation strength, but its performance under cyclical conditions was on par with the other devices. The 10-mm Arrow, SDSorb Staple, and Arthrex Dart provided inferior fixation strength as well as inferior performance under cyclical loading. In summary, 3 studies of the T-fix32,52 show that it has equivalent pullout strength to horizontal sutures. Of the nonsuture– based devices, the Arrow,67 Biostinger,32 and Mitek Meniscal Repair System52 have been shown to be arguably equivalent to suture techniques. In vitro device hydrolysis was a significant factor for the PDS-based Mitek Meniscal Repair Sytem,

284

E. FARNG AND O. SHERMAN

which degraded in 12 weeks. This was faster than PDS suture.67 Combined with the clinical conclusion that the PDS suture absorbs too quickly for adequate meniscal healing,66 it is possible that the PDS Mitek Meniscal Repair System does not maintain tissue apposition for a sufficient duration. Still, the lone clinical report on results using the Mitek system is encouraging, with an 86% clinical success rate. The only other device to show any loss of strength during 24 weeks of in vitro hydrolysis are the PDS suture and SDstaple. We should note that device degradation occurs more quickly in vivo than in vitro.64 How much biomechanical strength is required of devices after meniscal repair is not clear. Sutures may only be subjected to 10-N of tension through non– weight-bearing active motion, but, conversely, the menisci may also be responsible for transmitting as much as 3,000-N loads during level walking. Further complicating the issue is the fact that the experimental design of these biomechanical studies fails to accurately reflect in vivo loading conditions. Compressive and shear forces between the tibia and femur produce circumferential hoop stresses in the menisci.6 These forces are not simulated when testing repairs to failure with radially directed tensile loads. Clinically, short-term follow-up studies on these repair devices have generally shown a clinical success rate of 86%79 to 95%,76 although one study reports a success rate of approximately 57%.75 This compares well with open techniques (84%25 to 88%17) and arthroscopic suture techniques (98.6%).22 We must note, however, that the true rate of anatomic failure as shown on second-look arthroscopy may be as high as twice the rate of clinical failure.33 The long-term implications of an anatomically failed but asymptomatic repair have not been established, and the rate of repeat tear and repeat surgery in this subpopulation is unknown. CONCLUSIONS Failure rates are affected by initial fixation strength, cyclical stresses, and device hydrolysis. From a biomechanical standpoint, vertical sutures provide superior initial fixation strengths when compared with horizontal sutures, which in turn are equivalent or superior to most of the available devices. Under cyclical conditions, the vertical sutures are again superior to horizontal sutures in terms of displacement and tissue gapping.54-56 The T-Fix, Meniscus Arrow, and Mitek Meniscal Repair System have also been shown to be comparable to suture techniques under cyclical

conditions. If an absorbable device is used, it should retain its strength for at least 4 months, and the Bionx Arrow, Linvatec Biostinger, and Clearfix Screw have been shown to fit this criterion. With up to 2 years of follow-up evaluation, reports of clinical success rates around 90%72-77 suggest that Arrow techniques provide sufficient tissue apposition for clinical success in the short to intermediate term. Only vertical sutures, horizontal sutures, and the Bionx Arrow have sufficient biomechanical and clinical data to support use. T-Fix laboratory testing suggests that is has sufficient biomechanical performance, but only a few published clinical series support its use thus far. The initial fixation strength of the Mitek Meniscal Repair System appears adequate, but hydrolysis data suggest that the PDS version degrades too quickly. One published report on the Mitek Meniscal Repair System exists supports its use. At this time, no clinical studies to support the use of the other devices have been published. These devices have considerably simplified surgical technique and decreased surgical risk and are therefore an attractive alternative to sutures. However, healing rates for meniscal repair have not changed with the introduction of these new devices, and poorly indicated tears still do not carry a good prognosis for healing. Surgeons may be tempted to treat more complex meniscal tears with these new, simple devices, but healing rates for complex or avascular tears remain poor. REFERENCES 1. Tapper E, Hoover N. Late results after meniscectomy. J Bone Joint Surg Am 1969;51:517-526. 2. Fairbank TJ. Knee joint changes after meniscectomy. J Bone Joint Surg 1948;30:664-670. 3. Appel H. Late results after meniscectomy in the knee joint: A clinical and roentologcal follow-up investigation. Acta Orthop Scand Suppl 1970;133:100-111. 4. Jackson JP. Degenerative changes in the knee after meniscectomy. BMJ 1968;2:525-527. 5. Johnson RJ, Kettelkamp DB, Clark W, Leaverton P. Factors affecting late results after meniscectomy. J Bone Joint Surg Am 1974;56:719-729. 6. Krause WR, Clemson MS, Pope MH, et al. Mechanical changes in the knee after meniscectomy. J Bone Joint Surg Am 1976;58:599-604. 7. Ahmed A, Burke D. In vitro measurement of static pressure distribution in synovial joints: I. Tibial surface of the knee. J Biomech Eng 1983;105:216-225. 8. Baratz M, Fu F, Mengato R. Meniscal tears: The effect of meniscectomy and of repair on intraarticular contact areas and stress in the human knee. Am J Sports Med 1986;14:270-275. 9. Kurosawa H, Fukubayashi T, Nakajima H. Load bearing mode of the knee joint: Physical behavior of the knee joint with or without menisci. Clin Orthop 1979;149:283-290. 10. Radin E, Lamotte F, Maquet P. Role of the menisci in the

MENISCAL REPAIR DEVICES

11. 12.

13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.

31.

32. 33. 34. 35. 36.

distribution of stress in the knee. Clin Orthop 1984;185:290294. Walker P, Erkman M. The role of the menisci in force transmission across the knee. Clin Orthop 1975;109:104-192. Seedhom B, Hargreaves D. Transmission of the load in the knee joint with special reference to the role of the menisci: II. Experimental results, discussion, and conclusions. Eng in Med 1979;8:220-228. MacConaill MA. The movements of bones and joints: The synovial fluid and its assistants. J Bone Joint Surg Br 1950; 32:244-252. Levy IM, Torzilli PA, Warren RF. The effect of medial meniscectomy on anterior-posterior motion of the knee. J Bone Joint Surg Am 1982;64:883-888. Shoemaker SC, Markolf KL. The role of the meniscus in the anterior-posterior stability of the loaded anterior cruciate– deficient knee. J Bone Joint Surg Am 1986;68:71-79. Cassidy R, Shaffer A. Repair of peripheral meniscal tears. Am J Sports Med 1981;9:209-214. Dehaven K, Black K, Griffiths H. Open meniscus repair, technique and two to nine year results. Am J Sports Med 1989;17:788-795. Henning CE. Arthroscopic repair of meniscis tears. Orthopedics 1983;6:1130-1132. Henning CE, Lynch MA, Clark JR. Vascularity for healing of meniscus repairs. Arthroscopy 1987;3:13-18. Schulte KR, Fu FH. Meniscal repair using the inside-to-outside technique. Clin Sports Med 1996;15:455-467. Warren R. Arthroscopic meniscal repair. Arthroscopy 1985;1: 170-172. Morgan CD, Casscells SW. Arthroscopic meniscus repair: A safe approach to the posterior horns. Arthroscopy 1986;2:3-12. Morgan CD. The “all inside” meniscus repair. Arthroscopy 1991;7:120-125. Morgan CD. “All inside” arthroscopic meniscus repair. Tech Orthop 1993;8:105-112. Hamberg P, Gillquist J, Lysholm J. Suture of new and old peripheral meniscus tears. J Bone Joint Surg Am 1983;65:193197. Barrett GR, Richardson K, Koening V. T-Fix endoscopic meniscal repair: Technique and approach to different types of tears. Arthroscopy 1995;11:245-251. Barrett GR, Treacy SH, Ruff CG. The T-fix technique for endoscopy meniscus repair. Am J Knee Surg 1996;9:151-156. Barrett GR, Treacy SH. Use of the T-Fix suture anchor in fascial sheath reconstruction of complex meniscal tears. Arthroscopy 1996;12:251-255. Coen MJ, Caborn DNM, Urban W, et al. An anatomic evaluation of T-Fix suture device placement for arthroscopy allinside meniscal repair. Arthroscopy 1999;15:275-280. Albrecht-Olsen P, Kristensen G, Tormala P. Meniscus buckethandle fixation with an absorbable Biofix tack: Development of a new technique. Knee Surg Sports Traumatol Arthrosc 1993;1:104-106. Tormala P, Karhi O, Koho P, Tamminmaki M. A novel inserter instrument (Crossbow) for installation of self-reinforced bioabsorbable arrows into meniscus tissue. Knee Surg Sports Traumatol Arthrosc 2000;6:370-372. Barber FA, Herbert MA. Meniscal repair devices. Arthroscopy 2000;16:613-618. Albrecht-Olsen P, Kristensen G, Burgaard P, et al. The arrow versus horizontal suture in arthroscopy meniscus repair. Knee Surg Sports Traumatol Arthrosc 1999;7:268-273. Hutchinson MR, Ash SA. Failure of a biodegradable meniscal arrow. Am J Sports Med 1999;27:101-103. Menche DS, Phillips GI, Pitman MI, Steiner GC. Inflammatory foreign-body reaction to an arthroscopic bioabsorbable meniscal arrow repair. Arthroscopy 1999;15:770-772. Ross G, Grabill J, McDevitt E. Chondral injury after meniscal

37. 38. 39. 40. 41. 42. 43. 44. 45. 46.

47. 48.

49. 50.

51. 52.

53. 54. 55. 56. 57. 58.

285

repair with bioabsorbable arrows. Arthroscopy 2000;16:754756. Anderson K, Marx RG, Hannafin J, Warren RF. Chondral injury following meniscal repair with a biodegradable implant. Arthroscopy 2000;16:749-753. Calder SJ, Myers PT. Broken arrow: A complication of meniscal repair. Arthroscopy 1999;15:651-652. Song EK, Lee KB, Yoon TR. Aseptic synovitis after meniscal repair using the biodegradable Meniscus Arrow. Arthroscopy 2001;17:77-80. Hechtman KS, Uribe WJ. Cystic hematoma formation following use of a biodegradable arrow for meniscal repair. Arthroscopy 1999;15:207-210. Post WR, Akers SR, Kish V. Load to failure of common meniscal repair techniques: Effects of suture technique and suture material. Arthroscopy 1997;13:731-736. Koukoubis TD, Glisson RR, Faegin JA, et al. Meniscal fixation with an absorbable staple. Knee Surg Sports Traumatol Arthrosc 1997;5:22-30. Kohn D, Siebert W. Meniscus suture techniques: A comparative biomechanical cadaver study. Arthroscopy 1989;5:324327. Rimmer MG, Nawana S, Keene GCR, Pearcy MJ. Failure strengths of different meniscal suture techniques. Arthroscopy 1995;11:146-150. Asik M, Sener N, Akpmar S, et al. Strength of different meniscus suturing techniques. Knee Surg Sports Traumatol Arthrosc 1997;5:80-83. Albretcht-Olsen P, Lind T, Kristensen G, Falkenberg B. Failure strength of a new Meniscus Arrow repair technique: Biomechanical comparison with horizontal suture. Arthroscopy 1997;13:183-187. Dervin GF, Downing KJW, Keene GCF, McBride DG. Failure strengths of suture versus biodegradable arrow for meniscal repair. Arthroscopy 1997;13:296-300. Boenisch UW, Faber KJ, Ciarelli M, et al. Pull-out strength and stiffness of meniscal repair using absorbable arrows or ti-cron vertical and horizontal loop sutures. Am J Sports Med 1999;27:626-631. Song EK, Lee KB. Biomechanical test comparing the load to failure of the biodegradable Meniscus Arrow versus meniscal suture. Arthroscopy 1999;15:726-732. Karaoglu S, Duygulu F, Inan M, Baktir A. Improving the biomechanical properties of the T-fix using oblique direction: in vitro study on bovine menisci. Knee Surg Sports Traumatol Arthrosc 2002;10:198-201. Walsh SP, Evans SL, O’Doherty DM, Barlow IW. Failure strengths of suture vs biodegradable arrow and staple for meniscal repair. Knee 2001;8:151-156. Becker R, Schroder M, Starke C, et al. Biomechanical investigations of different meniscal repair implants in comparison with horizontal sutures on human meniscus. Arthroscopy 2001;17:439-444. Rankin CC, Linter DM, Noble PC, et al. A biomechanical analysis of meniscal repair techniques. Am J Sports Med 2002;30:492-497. Seil R, Rupp S, Kohn DM. Cyclic testing of meniscal sutures. Arthroscopy 2000;16:505-510. Bellemans J, Vandenneucker H, Labey L, Van Audekercke R. Fixation strength of meniscal repair devices. Knee 2002;9:1114. Becker R, Starke C, Heymann M, Nebelung W. Biomechanical properties under cyclic loading of seven meniscus repair techniques. Clin Orthop Rel Res 2002;400:236-245. Vert M, Mauduit J, Li S. Biodegradation of PLA/GA polymers: Increasing complexity. Biomaterials 1994;15:12091213. Blasier RD, Bucholz R, Cole W, et al. Bioresorbable implants:

286

59. 60.

61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72.

E. FARNG AND O. SHERMAN

Applications in Orthopaedic Surgery. Instruct Course Lect 1997;46:531-546. Hollinger JO, Battistone GC. Biodegradable bone repair materials: Synthetic polymers and ceramics. Clin Orthop 1986; 207:290-305. Lam KH, Schakenrad JM, Esselbrugge H, et al. The effect of phagocytosis of poly(L-lactic acid) fragments on cellular morphology and viability. J Biomed Mater Res 1993;27:15691577. Tabata Y, Ikada Y. Macrophage phagocytosis of biodegradable microspheres composed of L-lactic acid/glycolic acid homo- and copolymers. J Biomed Mater Res 1988;22:837-858. Bostman OM. Absorbable implants for the fixation of fractures. J Bone Joint Surg Am 1991;73:148-153. Weiler A, Hoffmann RF, Stahelin AC, et al. Biodegradable implants in sports medicine: The biological base. Arthroscopy 2000;16:305-321. Laitinen O, Tormala P, Taurio R, et al. Mechanical properties of biodegradable ligament augmentation device of poly(Llactide) in vitro and in vivo. Biomaterials 1992;13:1012-1016. Heidemann W, Jeschkeit S, Ruffieux K, et al. Degradation of poly(D,L)lactide implants with or without addition of calciumphosphates in vivo. Biomaterials 2001;22:2371-2381. Barrett GR, Richardson K, Ruff CG, Jones A. The effect of suture type on meniscus repair. Am J Knee Surg 1997;10:2-9. Arnoczky SP, Lavagnino M. Tensile fixation strengths of absorbable meniscal repair devices as a function of hydrolysis time. Am J Sports Med 2001;29:118-123. Roeddecker K, Muennich U, Nagelschmidt M. Meniscal healing: A biomechanical study. J Surg Res 1994;56:20-27. Morgan CD, Wojtys EM, Casscells CD, Casscells SW. Arthroscopic meniscal repair evaluated by second-look arthroscopy. Am J Sports Med 1991;19:632-638. Morrison JB. Function of the knee joint in various activities. Biomed Eng 1969;4:573-580. Kirsch L, Kohn D, Glowik A. Forces in medial and lateral meniscus sutures during knee extension: An in vitro study. J Biomechanics 1999;31:104 (suppl 1). Escalas F, Quadras J, Caceres E, Benaddi J. T-Fix anchor

73.

74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84.

sutures for arthroscopic meniscal repair. Knee Surg Sports Traumatol Arthrosc 1997;5:72-76. Barrett GR, Treacy SH, Ruff CH. Preliminary results of the T-Fix endoscopic meniscus technique in an anterior cruciate ligament reconstruction population. Arthroscopy 1997;13:218223. H¨urel C, Mertens F, Verdonk R. Biofix resorbable Meniscus Arrow for meniscal ruptures: Results of a 1-year follow-up. Knee Surg Sports Traumatol Arthrosc 2000;8:46-52. Venkatachalam S, Godsiff SP, Harding ML. Review of the clinical results of arthroscopic meniscal repair. Knee 2001;8: 129-133. Jones HP, Lemos MJ, Wilk RM, et al. Two year follow-up of meniscal repair using a bioabsorbable arrow. Arthroscopy 2002;18:64-69. Petsche TS, Selesnick H, Rochman A. Arthroscopic meniscus repair with bioabsorbable arrows. Arthroscopy 2002;18:246253. Gill SS, Diduch DR. Outcomes after meniscal repair using the Meniscus Arrow in knees undergoing concurrent anterior cruciate ligament reconstruction. Arthroscopy 2002;18:569-577. Laprell H, Stein V, Peterson W. Arthroscopic all-inside meniscus repair using a new refixation device: A prospective study. Arthroscopy 2002;18:387-393. Mathur PD, McDonald JR, Ghormley RK. A study of the tensile strength of the menisci of the knee. J Bone Joint Surg Am 1949;31:650-654. Bullough PG, Munuera L, Murphy J, Weinstein AM. The strength of the menisci of the knee as it relates to their fine structure. J Bone Joint Surg Br 1970;52:564-570. Uezaki N, Kobayashi A, Matsushige K. The viscoelastic properties of the human semilunar cartilage. J Biomech 1978;12: 65-73. Proctor CS, Schmidt MB, Whipple RR, et al. Material properties of the normal medial bovine meniscus. J Orthop Res 1989;771-782. Joshi MD, Suh JK, Marui T, Woo SL. Interspecies variation of compressive biomechanical properties of the meniscus. J Biomed Mater Res 1995;29:823-828.