If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
To evaluate the biomechanical characteristics of recently introduced meniscal repair devices with a hand-tied, inside-out meniscal suture in a human meniscus model.
Methods
In detached adult human menisci, vertical longitudinal cuts were created 3 mm from the synovial-meniscal junction, simulating a bucket-handle meniscal tear. Each cut was repaired using a single device. Group 1 received a vertical mattress suture of No. 2-0 OrthoCord; group 2, TrueSpan device with PEEK (polyether ether ketone) anchors containing No. 2-0 OrthoCord suture; group 3, TrueSpan device with biodegradable poly-lactide–co-glycolide (PLGA) anchors containing No. 2-0 OrthoCord suture; group 4, Meniscal Cinch II device; group 5, AIR meniscal repair device; and group 6, FasT-Fix 360 device. All samples were preloaded at 5 N and cycled 200 times between 5 and 20 N. The specimens that survived cyclic loading were destructively tested at 12.5 mm/s. Endpoints included maximum load, displacement, stiffness, and failure mode.
Results
The mean failure loads were as follows: 95.8 N for OrthoCord suture, 87.1 N for TrueSpan with PEEK, 84.6 N for TrueSpan with PLGA, 48.6 N for Meniscal Cinch II, 72.3 N for AIR, and 68.1 N for FasT-Fix 360. Repairs performed with OrthoCord suture (P = .002) and both TrueSpan devices (P < .03) but not the FasT-Fix 360 device or AIR device were statistically significantly stronger than Meniscal Cinch II repairs. Mean cyclic displacement measured 1.1 mm for OrthoCord, 1.5 mm for TrueSpan with PEEK, 1.5 mm for TrueSpan with PLGA, 2.1 mm for Meniscal Cinch II, 1.1 mm for AIR, and 1.4 mm for FasT-Fix 360. The Meniscal Cinch II device showed more displacement than all other devices (P < .05). The FasT-Fix 360, AIR, and Meniscal Cinch II devices failed by anchor pullout from the peripheral meniscus. OrthoCord and both TrueSpan devices failed by suture pulling through the bucket-handle tissue.
Conclusions
OrthoCord suture is stronger than the AIR, FasT-Fix 360, and Meniscal Cinch II devices. The TrueSpan device with PEEK and TrueSpan device with PLGA are stronger than the Meniscal Cinch II device. The Meniscal Cinch II device failed during cyclic loading with greater cyclic displacement than the AIR device, FasT-Fix 360 device, OrthoCord, and TrueSpan device with PEEK. The Meniscal Cinch II, AIR, and FasT-Fix 360 devices failed by anchor pullout, whereas OrthoCord and both TrueSpan devices failed by suture pull-through.
Clinical Relevance
Some newly introduced all-inside meniscal repair devices show inferior failure strength compared with earlier versions that might adversely impact clinical outcomes.
Repairing torn meniscal tissue has gained increased emphasis based on recent publications that underscore the beneficial effects of successful meniscal healing.
Influence of medial meniscus bucket-handle repair in setting of anterior cruciate ligament reconstruction on tibiofemoral contact mechanics: A biomechanical study.
All-inside meniscal repair techniques predominate. The indications for meniscal repair have expanded beyond the classic vertical peripheral longitudinal tear to include meniscal root tears,
Repair augmentation of unstable, complete vertical meniscal tears with bone marrow venting procedure: A prospective, randomized, double-blind, parallel-group, placebo-controlled study.
These all-inside ultrahigh-molecular-weight polyethylene (UHMWPE) suture–containing devices avoid additional incisions about the knee, decrease the risk of neurovascular injury, and avoid other complications previously associated with traditional inside-out meniscal repair.
However, many of the prior devices have been significantly updated, and newer all-inside devices have been introduced. The tested devices may be newly introduced, have different-sized or -shaped anchors, or in the case of the TrueSpan device (DePuy Mitek, Raynham, MA), have a complete redesign of the knot-pulley mechanism. In addition, the TrueSpan device now comes with 2 different anchor materials (1 biodegradable).
The purpose of this study was to evaluate the biomechanical characteristics of recently introduced meniscal repair devices with a hand-tied, inside-out meniscal suture in a human meniscus model. The hypothesis was that these different UHMWPE suture–containing devices would show similar structural properties under cyclic loading to a UHMWPE-containing suture in the human meniscus model.
Methods
This study used 60 fresh-frozen adult human meniscus specimens obtained from 30 knees (both medial and lateral menisci were used). Specimens were excluded if any tearing, degenerative changes, or calcification were present. The mean donor age was 55 ± 18 years (range, 19-100 years). The donors were evenly divided into male and female donors. Pair matching of the meniscus to the repair device was not performed. No damaged menisci or degenerative meniscal tears were observed in the harvested specimens.
The specimens were removed from the freezer and allowed to thaw overnight. After harvesting of the meniscus from the tibia, a vertical longitudinal cut 3 mm from the periphery of each meniscus was created with a knife by experienced fellowship-trained orthopedic surgeons (F.A.B., M.S.H., and W.A.). This cut was not initially extended into the anterior and posterior meniscal horns to allow better control of the test meniscus during repair device insertion. A single suture or meniscal repair device was placed in each meniscus to approximate the 2 meniscal edges. All meniscal repair sutures or devices were placed in the central portion of the meniscus, and both medial and lateral menisci were used. Once the repair construct was placed and the repair complete, the remaining anterior and posterior tissue bridges were divided, completely separating the meniscus into 2 segments connected only by the single repair.
Testing was conducted on an E1000 ElectroPuls materials testing machine (Instron, Norwood, MA). Each segment of the repaired meniscus was held with 2 custom metal clamps that were in turn attached to the materials testing machine, and cyclic loading was performed at a displacement rate of 5 mm/min, with the distraction stress always parallel to the axis of the repair device being tested (Fig 1). This allowed the consistent application of force to the repair system without plastic deformation.
Fig 1Testing was conducted on a materials testing machine with each materials segment held with 2 metal clamps. (Copyright F. Alan Barber, M.D.)
After the initial preload of 5 N for 60 seconds, a cyclic loading force between 5 and 20 N was applied at 1 Hz for 200 cycles. Specimens that survived cycling were then destructively tested (loaded to failure) at a rate of 12.5 mm/min. Load and displacement were sampled continuously at 10 Hz. The displacement was measured by the travel of the actuator of the materials testing machine.
A total of 10 samples for each repair device were prepared by the 3 orthopedic surgeons (F.A.B., M.S.H., and W.A.). The menisci were not randomly distributed among groups. The sampling rate for force and position data was 50/s. These data were downloaded into a spreadsheet (Excel [Microsoft Office 365], version 1902; Microsoft, Redmond, WA) and analyzed using statistical software (Social Science Statistics, www.socscistatistics.com).
The endpoints for this testing were as follows: (1) maximum failure load seen at destructive testing after the successful completion of 200 cycles; (2) cyclic displacement, calculated as the difference in the average trough of the first 10 cycles and that of the final 10 cycles (between cycles 190 and 200); (3) stiffness, determined from the maximal endpoints of the linear region of the load-displacement plot; and (4) mode of failure, determined by visual inspection of the failure result by an individual who did not participate in the device insertion (D.B.S.).
The repair devices tested included the following: Group 1 received a single vertical suture of No. 2-0 OrthoCord (DePuy Mitek) (55% polydioxanone and 45% UHMWPE) placed 3 mm inside the meniscal cut, extending from the superior surface to the inferior surface of the meniscus, and hand tied on the peripheral meniscal capsule using 6 square knots (control group) (Fig 2). Group 2 underwent repair using the TrueSpan device with PEEK (polyether ether ketone) anchors placed with a vertical mattress stitch on the superior meniscal surface, 3 mm inside the meniscal cut and angled to orient one arm toward the superior peripheral capsule and the second toward the inferior meniscal capsule (Fig 3). The TrueSpan device with PEEK contains a strand of No. 2-0 OrthoCord that is doubled between the 2 different-sized PEEK anchors (6 mm × 1.4 mm and 4.6 mm × 1.7 mm) and inserted by a gun with a needle. A sliding locking knot is made, located outside the second PEEK anchor to be inserted, creating a repair with 2 sutures crossing the meniscal surface with no exposed knot.
Fig 2A single vertical suture of No. 2-0 OrthoCord is placed 3 mm inside the meniscal cut and hand tied on the capsular side of the medial meniscus to approximate the 2 meniscal fragments. (Copyright F. Alan Barber, M.D.)
Fig 3The TrueSpan device with PEEK has a strand of No. 2-0 OrthoCord that is doubled between 2 different-sized PEEK anchors (6 mm × 1.4 mm and 4.6 mm × 1.7 mm). A sliding locking knot is provided, located outside the second PEEK anchor, creating a repair with 2 sutures crossing the meniscus, without a knot on the meniscal surface. (Copyright F. Alan Barber, M.D.)
Group 3 underwent repair using the TrueSpan device with poly-lactide–co-glycolide (PLGA) anchors similarly placed with a vertical mattress stitch on the superior meniscal surface, 3 mm inside the meniscal cut in the same fashion as in group 2. The TrueSpan device with PLGA also contains a single No. 2-0 OrthoCord with the same configuration and delivery mechanism.
Group 4 underwent repair using the Meniscal Cinch II device (Arthrex, Naples FL) placed, as with the other devices, in a vertical mattress configuration on the superior surface of the meniscus, 3 mm inside the meniscal cut. This device has a coreless No. 2-0 FiberWire suture (Arthrex) containing a pre-tied sliding locking knot attached to 2 PEEK implants measuring 1 mm × 5 mm (Fig 4). The insertion device has a curved needle.
Fig 4The Meniscal Cinch II has a coreless No. 2-0 FiberWire suture with a pre-tied sliding locking knot attached to two PEEK implants measuring 1 mm × 5 mm. (Copyright F. Alan Barber, M.D.)
Group 5 underwent repair using the AIR meniscal repair system (Stryker, Kalamazoo, MI) placed 3 mm inside the meniscal cut, creating a vertical mattress configuration on the superior surface of the meniscus (Fig 5). The AIR device has 2 PEEK anchors (1 mm × 1 mm × 5 mm) containing a pre-tied sliding locking knot connected by a No. 2-0 braided UHMWPE suture inserted using a flexible 17-gauge needle.
Fig 5The AIR meniscal repair system has 2 PEEK anchors (1 mm × 1 mm × 5 mm) connected by a No. 2-0 braided ultrahigh-molecular-weight polyethylene suture with a sliding locking knot inserted using a flexible 17-gauge needle. (Copyright F. Alan Barber, M.D.)
Finally, group 6 underwent repair using the FasT-Fix 360 device (Smith & Nephew Endoscopy, Andover, MA) placed using a needle-insertion device in a vertical mattress configuration on the superior surface of the meniscus, 3 mm inside the meniscal cut. This device has 2 arrow-shaped PEEK anchors (the first measures approximately 1 mm × 5 mm and the second measures approximately 5 mm × 1.5 mm × 0.7 mm) connected by No. 2-0 braided UHMWPE suture with a pre-tied sliding locking knot (Fig 6).
Fig 6The FasT-Fix 360 device has 2 arrow-shaped PEEK anchors (approximately 1 mm × 5 mm and 5 mm × 1.5 mm) connected by No. 2-0 braided ultrahigh-molecular-weight polyethylene suture with a pre-tied sliding locking knot. (Copyright F. Alan Barber, M.D.)
Data were recorded and calculated using Excel (version 1902). Analysis-of-variance testing (Social Science Statistics, www.socscistatistics.com) was performed, and if the data indicated statistical significance, Duncan multiple range tests were performed. Statistical significance was defined as P < .05.
Results
Although each test group started with 10 specimens, the TrueSpan device with PLGA, Meniscal Cinch II, and FasT-Fix 360 were unsuccessfully inserted in 1 specimen each, making subsequent testing impossible. This left only 9 specimens tested in those 3 groups.
The mean load-to-failure data (in newtons) of the tested meniscal repairs (devices and suture) are shown in Table 1. The OrthoCord suture repair was statistically significantly stronger than the Meniscal Cinch II repair (P = .002; 95% confidence interval [CI], 76.7-114.9 N), even when samples that failed during cyclic loading (3 samples) were excluded (P = .03), as were the FasT-Fix 360 device (P = .001; 95% CI, 79.7-111.9 N) and AIR device (P = .002; 95% CI, 85.9-105.7 N). The OrthoCord repair did not exhibit a statistically significant difference compared with either TrueSpan repair. Both TrueSpan repairs (with PEEK and with PLGA anchors) were stronger than the Meniscal Cinch II repair (P = .03 and P = .02, respectively). The 95% CIs were 63.4 to 110.8 N and 64.0 to 105.2 N, respectively. No other statistically significant differences were found in maximum load-to-failure strength.
Table 1Mean Load to Failure of Tested Meniscal Devices
Device
n
Force, N
Mean
SD
Maximum
Minimum
OrthoCord
10
95.8
16.4
124.9
66.5
TrueSpan with PEEK
10
87.1
33.2
144.4
44.5
TrueSpan with PLGA
9
84.6
15.9
113.9
64.3
Meniscal Cinch II
9
48.6
38.1
113.2
5.1
AIR
10
72.3
15.1
89.5
44.5
FasT-Fix 360
9
68.1
13.7
88.6
51.5
NOTE. The mean load to failure of those Meniscal Cinch II devices that completed cyclic loading successfully was 66.9 N.
PLGA, poly-lactide–co-glycolide; SD, standard deviation.
The mean stiffness data (in newtons per millimeter) of the tested meniscal repairs (devices and suture) are shown in Table 2. Repairs performed with the AIR device (P = .003; 95% CI, 7.8-13.3 N/mm), OrthoCord (P = .04; 95% CI, 6.595-14.5 N/mm), and TrueSpan device with PLGA (P = .02; 95% CI, 12.96-18.3 N/mm) all showed greater stiffness than the Meniscal Cinch II repair. The AIR device showed significantly greater stiffness than the FasT-Fix 360 device (P = .009; 95% CI, 14.3-18.5 N/mm).
Table 2Mean Stiffness of Tested Meniscal Repairs and Mean Cyclic Displacement Over 200 Cycles
Device
Stiffness, N/mm
Cyclic Displacement, mm
Mean
SD
Mean
SD
OrthoCord
15.7
4.9
1.1
0.3
TrueSpan with PEEK
14.0
2.7
1.5
0.4
TrueSpan with PLGA
15.6
3.6
1.5
0.5
Meniscal Cinch II
10.5
3.8
2.1
0.6
AIR
16.4
2.6
1.1
0.4
FasT-Fix 360
12.1
3.7
1.4
0.2
PLGA, poly-lactide–co-glycolide; SD, standard deviation.
The mean cyclic displacement data (in millimeters) of the tested meniscal repairs (devices and suture) are presented in Table 2. The AIR device showed statistically significantly less displacement over 200 cycles than the Meniscal Cinch II device (P = .0005; 95% CI, 1.7-2.5 mm), TrueSpan device with PEEK (P = .03; 95% CI, 1.2-1.8 mm), and TrueSpan device with PLGA (P = .03; 95% CI, 1.2-1.87 mm). In addition to the AIR device, the FasT-Fix 360 device (P = .004; 95% CI, 1.7-2.4 mm), OrthoCord suture (P = .0004; 95% CI, 1.7-2.4 mm), and TrueSpan device with PEEK (P = .03; 95% CI, 1.1-1.8 mm) all showed statistically significantly less displacement over 200 cycles than the Meniscal Cinch II device. The OrthoCord suture showed less displacement than the TrueSpan device with PEEK (P = .04; 95% CI, 1.2-1.7 mm) and TrueSpan device with PLGA (P = .04; 95% CI, 1.2-1.8 mm).
Different modes of failure were observed (Table 3). The principal failure mode for the FasT-Fix 360, AIR, and Meniscal Cinch II devices was with the anchor pulling out of the peripheral meniscal rim. The principal mode of failure for the OrthoCord suture, TrueSpan device with PEEK, and TrueSpan device with PLGA was with the repair suture pulling through the bucket-handle tissue.
Cyclic loading in a human meniscus model showed that a hand-tied OrthoCord suture repair provided the highest mean load-to-failure strength (95.8 N). This was significantly stronger than the AIR, FasT-Fix 360, and Meniscal Cinch II devices. The TrueSpan devices with either PEEK or PLGA anchors were stronger than the Meniscal Cinch II device. Furthermore, the Meniscal Cinch II device was the only device tested with specimens that failed to complete cyclic testing (n = 3).
The greatest mean cyclic displacement was observed with the Meniscal Cinch II device. In contrast, the AIR device, FasT-Fix 360 device, OrthoCord suture, and TrueSpan device with PEEK showed statistically significantly less displacement than the Meniscal Cinch II device. OrthoCord suture repair showed less displacement than both TrueSpan repairs (P = .04).
The principal failure mode for the Meniscal Cinch II, AIR, and FasT-Fix 360 devices was anchor pullout from the meniscus (peripheral rim). In contrast, the principal failure mode for OrthoCord suture and both TrueSpan devices was suture pulling through the meniscal tissue (bucket-handle fragment).
The mechanical properties of self-adjusting UHMWPE suture repair devices were previously reported using a porcine meniscus model.
reported that OrthoCord suture repair (222 N) was stronger after 500 cycles than repairs performed with older versions of the FasT-Fix device (110 N), FiberWire (117 N), and UHMWPE suture (132 N); moreover, the latter showed significantly more elongation over 500 cycles (P < .05). The porcine model was reported to show comparable results to a young adult human meniscus.
showed that a vertical orientation of an all-inside device had less displacement, greater stiffness, and greater strength after cyclic loading than a horizontal orientation (P < .05). Performing our study in human cadaveric menisci allows for a comparison to the historical human biological data; the performance of repair devices in animal models may not replicate the human clinical condition.
This study evaluated newer versions of older all-inside devices. A comparison of data from a prior study
to those in our study proves interesting. Using the 200-cycle human meniscus model data indicates that in the previous study, the control suture repair using OrthoCord suture showed failure loads of about 88 N, which is less than the failure load in our study (95.6 N). The Cinch (Arthrex), the previous version of the meniscal-cinch device, showed failure loads of about 71 N; this does not compare favorably with the newer Meniscal Cinch II, which failed at about 49 N. The OmniSpan (DePuy Mitek), an older version of the TrueSpan device, showed failure loads of about 87 N, which is the same as the failure load of the current TrueSpan (87 N). In the prior study, the FasT-Fix 360 device failed at 60 N, whereas in this study, the mean failure load was 68 N. There was not prior iteration of the AIR device.
The hypothesis that these newer UHMWPE suture–containing devices would show similar structural properties under cyclic loading to UHMWPE-containing suture in the human meniscus model was not supported by this study. In fact, significant differences in ultimate failure loads, cyclic displacement, and modes of failure were observed, which may have clinical implications.
In the event that the meniscal repair fails, failure by the repair device’s anchors pulling out of the peripheral meniscal tissue is problematic. This creates the potential for loose anchors in the joint, which may result in chondral abrasion, an increased tendency for deployment failure (during insertion), and intra-articular migration of the anchors. This could be due to size or shape differences of the associated anchors. Specifically, the Meniscal Cinch II and AIR devices have 2 smaller anchors (5 mm × 1 mm). The FasT-Fix 360 device has 2 arrow-shaped PEEK anchors (1 mm × 5 mm and 5 mm × 1.5 mm). In contrast, the TrueSpan device has 2 larger anchors (6 mm × 1.4 mm and 4.6 mm × 1.7 mm). These larger anchors were less prone to pullout but could theoretically cause greater meniscal disruption during the initial insertion, as well as subsequently, should pullout occur. However, the amount of meniscal damage during insertion is probably more dependent on the needle size than the anchor size. It is well established that a vertical suture orientation provides greater repair strength than horizontal repair sutures by more effectively capturing the circumferentially oriented collagen bundles.
During the postoperative period, meniscal repairs are subjected to both compressive and shear loads. Distraction loads are not significant at a meniscal repair site.
The role of the meniscal repair suture or all-inside repair device is to resist the shear stresses on the meniscal segments while healing and rehabilitation progress.
This study shows that the TrueSpan device provides repair strength comparable to an inside-out repair performed with a UHMWPE-containing suture (OrthoCord). However, the Meniscal Cinch II repair device showed significantly lower mean failure loads, completed fewer load cycles, and had significantly greater displacement than the other tested devices. These data do not provide a reason for these results. It may be that load to failure is a more relevant mechanical parameter than cyclic displacement. In fact, it is more likely that failure of a meniscal repair is the result of repetitive cyclic loading. In this respect, both TrueSpan devices showed significantly greater cyclic displacement than the OrthoCord suture.
The inside-out meniscal suture repair is an established and effective standard for addressing appropriate meniscal tears. No difference has been shown between the use of absorbable sutures and use of nonabsorbable sutures, and the use of absorbable materials in all-inside meniscal repair devices should have no impact on clinical healing.
Furthermore, recent evidence has failed to show a difference in the risk of meniscal repair failure between men and women at mid-term follow-up although men report higher activity levels.
This study has several limitations. This was a time-zero bench test performed at room temperature and in a nonaqueous environment. The differences observed may not translate into clinical differences. Meniscal explants were used instead of intact knees. Some of the donors of human cadaveric meniscal specimens were older than a clinically appropriate population, and degenerative meniscal segments could alter the biomechanical performance. The meniscus of a 19-year-old patient will most definitely not be the same in quality or quantity as that of a 100-year-old patient. This is particularly critical when evaluating the method of failure because older menisci would arguably fail by pullout first, before suture failure. This test does not replicate the mechanism of meniscal repair failure, applying distractive forces instead of compressive, axially applied rotational loads or shear stresses. These data represent only some aspects of repair device performance and do not necessarily correlate with clinical healing. The repair devices are designed for arthroscopic use but were inserted using a non-arthroscopic, “open” technique. The different groups of repairs underwent different techniques including the fact that the sutures in the control group may align differently than those using anchors. The number of specimens available is limited by material availability, and no a priori power analysis was performed. The risk of selection bias due to confounding variables exists. Finally, these data cannot be compared directly with those of biomechanical studies using other biological substrates such as porcine or bovine menisci.
Conclusions
OrthoCord suture is stronger than the AIR, FasT-Fix 360, and Meniscal Cinch II devices. The TrueSpan device with PEEK and TrueSpan device with PLGA are stronger than the Meniscal Cinch II device. The meniscal Cinch II device failed during cyclic loading with greater cyclic displacement than the AIR device, FasT-Fix 360 device, OrthoCord, and TrueSpan device with PEEK. The Meniscal Cinch II, AIR, and FasT-Fix 360 devices failed by anchor pullout, whereas OrthoCord and both TrueSpan devices failed by suture pull-through.
Acknowledgment
The authors acknowledge Peter Wronski at the RIH Orthopaedic Foundation for his help with testing and data analysis.
Influence of medial meniscus bucket-handle repair in setting of anterior cruciate ligament reconstruction on tibiofemoral contact mechanics: A biomechanical study.
Repair augmentation of unstable, complete vertical meniscal tears with bone marrow venting procedure: A prospective, randomized, double-blind, parallel-group, placebo-controlled study.
The authors report the following potential conflicts of interest or sources of funding: This research was funded by a grant from DePuy Synthes Mitek Sports Medicine. F.A.B. receives grant support and support for travel to meetings from DePuy Mitek. In addition, F.A.B. is a consultant for DePuy Mitek and receives grant support, payment for lectures, and royalties from DePuy Mitek, outside the submitted work. D.B.S. is an employee of Johnson & Johnson, which developed the test suture used in this study. M.S.H. receives support for travel to meetings from DePuy Mitek. W.A. receives support for travel to meetings from DePuy Mitek. Full ICMJE author disclosure forms are available for this article online, as supplementary material.
M.S.H. received support for travel to meetings from DePuy Mitek. W.A. received support for travel to meetings from DePuy Mitek.
Meniscus repairs for vertical, peripheral tears can be troublesome due to poor tissue quality and/or vascularity that can lead to re-rupture and subsequent removal. The gold standard, inside-out repair technique, has been challenged by all-inside devices for the benefit of improved efficiency and less morbidity but for the sake of expense and potential structural inferiority. Successful meniscus repair requires multiple components, only one of which is deciding the repair construct of choice. I feel the most important aspect will always be the indication based on tear configuration while respecting biology, because all fixation will eventually fail if the meniscus does not ultimately heal.
30313-3&id=gr1.jpg){kind=link}
30313-3&id=gr2.jpg){kind=link}
30313-3&id=gr3.jpg){kind=link}
30313-3&id=gr4.jpg){kind=link}
30313-3&id=gr5.jpg){kind=link}
30313-3&id=gr6.jpg){kind=link}