- Email: [email protected]
Cyclic testing of tibialis tendon allografts for anterior cruciate ligament reconstruction using suture-post versus spiked washer tibial fixation
Clinical Biomechanics 70 (2019) 8–15
Contents lists available at ScienceDirect
Clinical Biomechanics journal homepage: www.elsevier.com/locate/clinbiomech
Cyclic testing of tibialis tendon allografts for anterior cruciate ligament reconstruction using suture-post versus spiked washer tibial fixation
T
Keith L. Markolf , Dean Wang, Nirav B. Joshi, Edward Cheung, Frank A. Petrigliano, David R. McAllister ⁎
Department of Orthopaedic Surgery, David Geffen School of Medicine at UCLA, 10833 Le Conte Ave, 76-143 CHS, Los Angeles, CA 90095, USA
ARTICLE INFO
ABSTRACT
Keywords: Graft fixation Cyclic testing ACL reconstruction
Background: The purpose of this study was to directly compare spiked washer and suture-post tibial-sided fixation techniques used for anterior cruciate ligament reconstruction by measuring anterior tibial translation during cyclic tests. Methods: Fresh-frozen human knees were tested using a robotic system that applied 250 cycles of anteriorposterior tibial force (134 N) at 30° flexion, while recording tibial translation. Ten intact knees were tested to collect baseline data for native specimens. A single knee was selected to test ligament reconstructions using doubled tibialis tendon allografts. All grafts were fixed proximally using an EndoButton™, and the tibial end of the graft was fixed with either a spiked washer or with a suture post placed at two different locations (near and distant) relative to the tibial tunnel. Findings: Mean first cycle translation for intact knees was 4.8 (sd 1.8) mm; means after reconstruction were 2.6 (sd 0.9) mm (spiked washer), 10.1 (sd 1.9) mm (suture post near), and 10.4 (sd 1.5) mm (suture post distant). Corresponding means for translation increase over 250 cycles were 0.3 (sd 0.2) mm, 3.6 (sd 1.3) mm, 7.2 mm (sd 0.9) mm, and 8.0 (sd 1.3) mm. All mean increases (first cycle and cyclic) after ACL reconstruction were significantly greater than those for the intact knees, and all means with a suture post were significantly greater than those with a spiked washer. There were no significant differences between mean translations for near and distant suture post locations. Interpretation: Use of suture post fixation for anterior cruciate ligament reconstruction is questioned since increases in anterior tibial translation could lead to excessive post-operative knee laxity and possibly early clinical failure.
1. Introduction Rigid fixation is one of the most important factors influencing the mechanical performance of an anterior cruciate ligament (ACL) graft in the immediate post-operative period (Kurosaka et al., 1987). During the 12 week process of osseointegration (Rodeo et al., 1993), fixation must be secure enough to prevent graft slippage and/or local elongation at the fixation site. The tibial side fixation of a graft construct is generally the weak link because bone quality of the tibial metaphysis is inferior to that of the femur, and the tissue ends of a soft-tissue tendon graft are more difficult to secure (Brand Jr et al., 2000a; Brand Jr et al., 2000b). A suture-post (SP) is still commonly used to fix the tibial end of an ACL graft (Speziali et al., 2014). With this technique, suture at the free ends of a tendon graft are tied around a post beneath a flat washer. However, the suture stitches can tighten and compress into the graft
⁎
Corresponding author. E-mail address: [email protected] (K.L. Markolf).
https://doi.org/10.1016/j.clinbiomech.2019.07.025 Received 30 May 2018; Accepted 23 July 2019 0268-0033/ © 2019 Elsevier Ltd. All rights reserved.
tissue as they are subjected to repetitive loading. This could potentially result in increased laxity of the ACL-reconstructed knee. In addition, optimal placement of the SP relative to the tibial tunnel is unknown. In theory, fixing the SP at a location more distant from the tibial tunnel could result in increased knee laxity due to a longer and more compliant graft-suture construct, but this effect has not been investigated experimentally. Tibial fixation using a spiked washer (SW) should theoretically produce less laxity than a SP because the tissue is compressed directly onto cortical bone proximal to the sutures, thereby eliminating any effects of suture tightening within the graft tissue. There have been a number of prior studies comparing the strength and stiffness of various tibial fixation devices, with most performed using bovine tibias (Flanigan et al., 2012; Giurea et al., 1999; Micucci et al., 2010; Weimann et al., 2005) or porcine tibias (Efe et al., 2010; Herrera et al., 2010; Magen et al., 1999; Meuffels et al., 2010;
Clinical Biomechanics 70 (2019) 8–15
K.L. Markolf, et al.
Numazaki et al., 2002; Nyland et al., 2014; Prado et al., 2013). In these studies, tibial sided graft fixation was examined using uniaxial test machines to apply force directly in line with the graft tissue. However, a uniaxial test machine does not accurately represent in vivo knee loading conditions and the effects of local contact stresses at the graftbone interfaces within the bone tunnels. Therefore, it is desirable to evaluate tibial fixation devices by performing ACL reconstruction in human specimens and loading the knees in a manner that more closely replicates in vivo conditions. The objective of this study was to directly compare SW and SP tibial fixation by measuring anterior knee laxity during cyclic anteroposterior (AP) tibial loading of an ACL-reconstructed knee. We hypothesized that SP fixation would produce significantly greater anterior knee laxity than SW fixation, and that locating the SP more distant from the tibial tunnel exit (thereby increasing overall graft-suture construct length) would result in significantly increased anterior knee laxity. 2. Methods 2.1. Specimen preparation Fresh-frozen tibialis tendon allografts were used for this study. It has been shown previously that tibialis anterior and tibialis posterior tendons have comparable structural properties and display similar viscoelastic behavior (Pearsall 4th et al., 2003). Therefore, for the purposes of this study, tibialis anterior and posterior tendons were considered to be equivalent. A total of 15 anterior and 15 posterior tibialis tendon grafts were harvested from human donors (mean age 46 years, range 16–64). Within each fixation group, 5 anterior and 5 posterior tibialis tendons were tested in a randomized order. All grafts were cleared of adherent muscle fibers and surrounding soft tissues, wrapped in saline-soaked gauze, and stored frozen at −20 °C. On the day of testing, the graft was thawed to room temperature and kept moist with saline solution during specimen preparation, fixation, and biomechanical testing. The tendon was passed through the loop of a cortical fixation device (30 mm EndoButton™, Smith & Nephew, Andover, MA) to produce a double-stranded graft (Fig. 1). No.2 polyester suture (Ethibond™, Ethicon, Somerville, NJ) was sutured into each free tendon end of the construct using a running-locked suture technique. The cross-sectional area of each doubled graft was measured under a compressive pressure of 0.12 MPa applied for 2 min using an established technique (Grood et al., 1992). Ten fresh-frozen human cadaveric knee specimens were used to collect baseline ATT data for the native ACL (mean age 34 years, range 21–45). The tibia and femur were sectioned mid-shaft and scraped clean of soft tissue to within 10 cm of the joint line, and the bone ends were potted in cylindrical molds of polymethylmethacrylate (PMMA) for attachment to testing fixtures. The soft tissues surrounding the knee capsule were left intact to help prevent dehydration of ligamentous tissues within the knee. To reduce possible variability due to knee geometry and size, a single specimen was selected for use in testing all ACL reconstructions (male, 44 years of age). This allowed direct comparisons of tibial fixation methods within the same knee. The native ACL was excised and 11 mm diameter femoral and tibial tunnels were drilled within the footprints of the native ACL in accordance with current clinical practice using the anteromedial portal technique. An 11 mm femoral tunnel was drilled (inside-out) to a depth of 30 mm. An additional 4.5 mm tunnel (used for passage of the
Fig. 2. EndoButton™ femoral fixation at the lateral cortex of the femoral tunnel exit reinforced with PMMA acrylic.
EndoButton™) was drilled from the internal base of the femoral tunnel to the lateral cortex of the femur. An 11 mm tibial tunnel was drilled from the tibial cortex to the center of the ACL footprint using a drill guide. The EndoButton™ (with attached synthetic fiber loop and graft) was pulled through the femoral tunnel and 4.5 mm passage tunnel. When in place, the button contacted a small metal washer placed over the tunnel opening at the lateral femoral cortex (Fig. 2). The cortical bone surface at the exit of the femoral tunnel was reinforced with PMMA acrylic. These modifications were made to prevent the possibility of cumulative bone deformation at this site from repeated testing. Before insertion, each graft was preconditioned on a tensioning board (Graftmaster II; Smith & Nephew, Andover, MA) in accordance with clinical protocol for this device. The graft construct was fixed with an initial 89 N tensile force and held on the tension board for 20 min. This level of force has been used in previous studies (Figueroa et al., 2010; Höher et al., 2000; Howard et al., 1996). After preconditioning, the graft was immediately inserted into the knee and secured to the femur as described above. The tibial end of the graft was tensioned to 89 N at 30° of knee flexion during final fixation. This tensioning protocol was based on previously published randomized clinical trials (Kim et al., 2006; Yasuda et al., 1997). Tibial fixation of the graft was accomplished using either a SW (18 mm diameter with a 6.5 mm × 40 mm bicortical screw; Fig. 3A; Arthrex, Inc., Naples, FL) or a SP (18 mm flat washer with a 6.5 mm × 40 mm bicortical screw; Fig. 3B; Arthrex, Inc., Naples, FL). Both were secured with a nut against the opposite cortex. For SW grafts, the sutures from both ends of the graft were tied together and looped over the hook of a spring scale to maintain constant suture tension during final fixation. While under tension, the two graft ends were
Fig. 1. Tibialis tendon graft preparation. 9
Clinical Biomechanics 70 (2019) 8–15
K.L. Markolf, et al.
FARO Technologies Inc., Lake Mary, FL). The femur was clamped to a fixture mounted to a base plate with the knee flexed to 30°. Under forcemoment control, the robot applied 250 cycles of AP force to the tibia. Per Arnold et al. (2005) who measured tension in BTB grafts with an arthroscopically implantable force probe (AIFP) during 1500 flexionextension cycles, 70% of the total loss in graft tension occurred by cycle 100. To account for variation in graft tissue type and study methodology, we applied 250 cycles of AP tibial force. During each cycle, the robot applied a 134 N posterior tibial force followed by a 134 N anterior tibial force. Target force of 134 N has been used previously in robotic studies related to ACL reconstruction (Gabriel et al., 2004; Giffin et al., 2004; Kanamori et al., 2000; Loh et al., 2003; Sasaki et al., 2014; Zamarra et al., 2010). Applied tibial force and the corresponding AP tibial translation (ATT) were measured continuously during the test. Anterior knee laxity was defined as the ATT recorded during testing, and evaluated by examining the initial ATT (first cycle), cycle 250 ATT, and the increase in ATT from cycle 1 to cycle 250. Ten intact knees were first tested to document baseline increases in ATT with the native ACL. Then, one knee was selected for cyclic testing with all subsequent graft tissues. Three fixation groups (10 grafts each) were tested with this knee: SW, SP-near, and SP-distant. The testing order for all graft tissues was randomized. For the 10 intact knee specimens, the neutral position of the tibia relative to the femur was determined by finding the midpoint between the peak posterior and peak anterior tibial translations at 134 N of applied AP force. The mean neutral position was determined to be 4.8 mm anterior to the PCL endpoint. All ATT measurements were referenced to the intact AP neutral position in order to allow direct comparisons of ATT and increase in ATT between native knees and the two fixation constructs. 2.3. Statistics and data analysis Power analysis was performed based on variance seen in preliminary testing. A sample size of n = 6 was needed to detect a 1 mm difference in ATT with 80% power. To ensure statistical power was achieved, a sample size of n = 10 per fixation method was chosen. The variables analyzed were the first cycle ATT, cycle 250 ATT, and the increase in ATT from cycle 1 to cycle 250. Unpaired two-sample Student's t-tests were used for tibialis anterior versus tibialis posterior comparisons (cross-sectional area, first cycle ATT, and increase in ATT). A similar analysis was also used to compare first cycle ATT, cycle 250 ATT, and increase in ATT between near and distant post locations with the SP construct. A one-way ANOVA was used to compare means for first cycle ATT, cycle 250 ATT, and cyclic increases in ATT between the native ACL group and all fixation methods. Post hoc comparisons were made using Tukey's HSD procedure. The level of significance was p < 0.05.
Fig. 3. A) 18 mm diameter spiked washer with a 6.5 mm × 40 mm bicortical screw. B) 18 mm diameter flat washer with a 6.5 mm × 40 mm bicortical screw. For both fixation device, an opposing nut was used to maintain secure fixation.
positioned beneath the peripheral spikes of the washer and secured onto the PMMA re-enforced cortical surface with a bicortical screw and nut (Fig. 4A). The bicortical screw hole was 2 cm distal to the tibial tunnel, with the sutured portion of the graft tissue distal to the tissue compressed by the SW. For SP grafts, two post locations were evaluated: near (using the SW tibial hole -Fig. 4B) and distant (4 cm distal to the SW tibial hole Fig. 4C). For both groups, the suture ends had to be free so that they could be tied around the post. Therefore a separate suture loop, passing through both graft ends proximal to the sutured portion, was tensioned with the spring scale to maintain a constant 89 N tension in the graft tissue while the sutures were tied around the shank of the post using four square knots. The knotted portion of the sutures was then compressed down onto the PMMA re-enforced surface of the tibia using a screw and flat washer.
3. Results 3.1. Comparison of tibialis anterior and tibialis posterior tendon grafts There were no significant differences in mean cross-sectional area, mean first cycle ATT, or mean increase in ATT between tibialis anterior and posterior tendon grafts for any of the three testing groups (Table 1).
2.2. Specimen testing
3.2. Comparison of ATT between testing groups
A six DOF robotic manipulator (KR210; KUKA Robotics Corp., Clinton Township, MI) with a force-moment sensor (Omega Model Industrial Automation Load Cell, ATI Industrial Automation, Apex, NC) was used to perform all knee loading tests (Fig. 5). Prior to testing, the tibia was clamped in a fixture mounted to the force-moment sensor. Measurements were taken to ensure proper alignment of the anatomic joint coordinate system (Grood and Suntay, 1983) with the axes of the robot and load cell using a three dimensional digitizer (Faro Gage,
Mean first cycle ATT for the SW fixation group was significantly less than that for the intact ACL group (p < 0.01), while means for near and distant suture-post groups were significantly greater than the intact ACL group and SW group (p < 0.01), but not significantly different from each other (Table 2). Means for Cycle 250 ATT and increases in ATT were significantly greater (p < 0.01) than the intact ACL for all fixation groups. Means 10
Clinical Biomechanics 70 (2019) 8–15
K.L. Markolf, et al.
Fig. 4. Tibial fixation methods. A) Spiked washer fixation proximal to the suture. B) Suture-post fixation near to the tibial tunnel (at the spike washer location). C) Suture-post fixation distant to the tibial tunnel (40 mm distal to the spiked washer location). The bone around the fixation site was reinforced with PMMA acrylic.
for near and distant SP groups were significantly greater (p < 0.01) than the SW group, but not significantly different from each other (Table 2). A comparison of mean curves for intact and ACL reconstructed knees during cyclic testing (Fig. 6) illustrates the differences between the intact knee and both fixation constructs, both in terms of initial ATT and cyclic increases in ATT. It should be noted in Fig. 6 that during cyclic testing, the SP curve demonstrates a progressively longer region of low stiffness (slope of the force vs. ATT curve). During the final SP
test cycles, the tibia had displaced anteriorly approximately 10 mm before the SP construct began to resist the applied anterior tibial force. This indicates that the graft was essentially slack in this region, in marked contrast to the behavior of the native ACL and grafts fixed with a SW. At the conclusion of testing, clear evidence of suture tightening was observed for all SP grafts with deep suture grooves imprinted into the ends of the graft tissue (Fig. 7).
11
Clinical Biomechanics 70 (2019) 8–15
K.L. Markolf, et al.
Fig. 5. Six degree-of-freedom robot used to apply 250 cycles of 134 N anteroposterior tibial force while recording anterior tibial translation.
of this laxity is related to local deformations at the graft fixation interfaces on the tibia. This cyclic loading study compared ATTs with knees reconstructed using tibialis tendon grafts that were fixed at their tibial ends with a SW or SP. SP fixation produced significant increases in anterior knee laxity compared to SW fixation, regardless of the position of the post. First cycle ATT with the near SP was 4 times greater and the increase in ATT over 250 cycles was 2 times greater compared to a SW. After completion of 250 test cycles, the mean ATT with SW grafts was 1.1 mm greater than that for intact knees. In contrast, mean ATT's with both near and distant SP grafts were approximately 12 mm greater than the intact knee, a condition that would be unacceptable clinically. These finding are important, as increases in post-operative knee laxity has been associated with ACL graft rupture (Pinczewski et al., 2007). Clinical studies have demonstrated that patients undergoing ACL reconstruction using a hamstring autograft with SP fixation show comparable joint stability and subjective patient outcomes compared to those using a bone-patellar tendon-bone autograft or a hamstring autograft with interference screw fixation (Harilainen et al., 2006; Ma et al., 2004; Prodromos et al., 2005). However, only two prior studies have directly compared SW and SP fixation. While direct comparison with our study is difficult, the results of these studies support our findings. Magen et al. (1999) performed biomechanical testing of bovine extensor tendons fixed to porcine tibias using six devices. Among those tested were SP and a 20-mm metal SW. The SP construct exhibited lower stiffness (60 N/mm) and yield load (830 N) than the SW constructs (126 N/mm and 930 N). However, cycling was not performed. Höher et al. (2000) compared SP fixation to dual SW fixation (use of two washers) on isolated human cadaveric semitendinosus and gracilis tendons in a uniaxial test machine. Using a complex cyclic loading protocol, they evaluated the increase in graft elongation from the first cycle to the last cycle (500 cycles). Their SP constructs showed a 3.5 mm greater increase in graft elongation than SW constructs. Similarly, our near SP constructs had a 3.6 mm greater increase in ATT than the SW, and our distant SP constructs has a 4.4 mm greater increase in ATT than SW.
Table 1 Comparison of graft cross-sectional area and Anterior Tibial Translation (ATT) using tibialis anterior and tibialis posterior grafts (mean ± SD). Spiked washer 2
Cross-sectional area (mm ) Cycle 1 ATT (mm) Cycle 250 ATT (mm) ATT increase (mm) Near suture-post Cross-sectional area (mm2) Cycle 1 ATT (mm) Cycle 250 ATT (mm) ATT increase (mm) Distant suture-post Cross-sectional area (mm2) Cycle 1 ATT (mm) Cycle 250 ATT (mm) ATT increase (mm)
Tibialis anterior
Tibialis posterior
46.0 (10.7) 2.2 (1.1) 6.1 (2.3) 3.9 (1.8) Tibialis anterior 43.3 (8.1) 10.0 (1.2) 17.4 (1.8) 7.4 (1.1) Tibialis anterior 50.9 (7.4) 10.5 (1.5) 19.0 (2.0) 8.5 (1.3)
51.1 (8.0) 2.9 (0.5) 6.3 (0.8) 3.4 (0.7) Tibialis posterior 46.7 (7.4) 10.2 (2.6) 17.2 (2.7) 7.1 (0.8) Tibialis posterior 43.7 (4.5) 10.2 (1.7) 17.8 (2.0) 7.6 (1.2)
Table 2 Comparison of Anterior Tibial Translations (ATT) for intact knee, spiked washer, near suture-post, and distant suture-post testing groups (mean ± SD).
Intact knee Spiked washer Near suture-post Distant suture-post ⁎ † ‡
First Cycle ATT (mm)
Cycle 250 ATT (mm)
Increase in ATT (mm)
4.8 (1.8) 2.6 (0.9)⁎ 10.1 (1.9)†,‡ 10.4 (1.5)†,‡
5.1 (1.9) 6.2 (1.6)† 17.3 (2.1)†,‡ 18.4 (2.0)†,‡
0.3 3.6 7.2 8.0
(0.2) (1.3)† (0.9)†,‡ (1.3)†,‡
Significantly less than the native ACL (p < 0.01). Significantly greater than the native ACL (p < 0.01) Significantly greater than spiked washer (p < 0.01).
4. Discussion Rigid fixation of an ACL graft reconstruction is desirable to avoid unwanted post-operative anterior knee laxity. An important component 12
Clinical Biomechanics 70 (2019) 8–15
K.L. Markolf, et al.
Fig. 6. Tracing of 250 anteroposterior loading cycles of 134 N (applied tibial force vs. tibial translation). The curves for each condition represent the mean of all 10 specimens. Suture-post fixation (showing the near post location) has significantly greater first cycle ATT and increase in ATT over 250 loading cycles compared to both the native ACL and spiked washer fixation (p < 0.01). Peak posterior displacement was consistent for all tests.
with a SP increased by 2.0 mm after only 8 loading cycles. There is a major difference between the two fixation constructs that is important when interpreting our results. With SW fixation, the sutured portion of the graft was distal to the washer and could not affect ATT measurements. With SP fixation, the sutured portion of the graft was proximal to the tibial fixation site, thus any suture tightening around the graft tissue would manifest as additional ATT. We believe the low stiffness region of the SP test curve (Fig. 6) was caused by progressive suture tightening around the graft tissue as cyclic anterior tibial force was applied. The deep suture grooving within the SP grafts after suture removal was direct evidence of suture tightening (Fig. 7). We preconditioned all grafts on a tension board by applying 89 N of tensile force for 20 min. With SP grafts, this preconditioning protocol should have helped reduce the effects of suture tightening prior to graft implantation. However, our results showed that additional suture tightening with SP grafts occurred during cyclic testing. It is possible that some of the initial suture tightening produced by the tension board could have relaxed during the time the graft was removed from the tension board and fixed within the knee (typically < 5 min). It should also be noted that the tension board force level was a constant 89 N, while the cyclic tests were performed with 134 N anterior tibial force. Our findings related to the near or distant location of a SP relative to the tibial tunnel entrance deserve special mention. Clinically, the optimal position of the SP relative to the tibial tunnel is unknown. We hypothesized that moving SP fixation 4 cm distally would increase anterior laxity because this would result in a longer (and theoretically more compliant) graft construct. This was not found to be the case, as there were no significant differences in ATT and increase in ATT between near and distant SP locations. This finding suggests that any increase in anterior laxity produced by the additional 4 cm of graft tissue was insignificant compared to laxity increases produced by suture tightening. It should be noted that the distance of 4 cm was selected to simulate an extreme condition, and represented a far greater distal placement than a clinician would consider. Therefore, our results would indicate that clinicians using this method of graft fixation have considerable flexibility in near-distant placement for SP fixation in terms of its effect on anterior knee laxity. This study had several limitations. A single cadaveric knee was used for testing of all graft tissues to eliminate any potential variability due to anatomical factors. Laxity measurements using a different knee
Fig. 7. A double stranded tibialis graft with suture-post fixation after completion of cyclic testing and graft removal. Suture tightening at both ends of the graft tissue is apparent, as evidenced by deep grooves near the graft ends.
The majority of prior studies investigating graft fixation have loaded the graft in line with the axis of the tibial tunnel, which is non-physiologic. In vivo tibial loading occurs in the anterior direction, and the tensile force acting on the fixation construct is reduced due to the frictional losses as the graft bends around the edge of the tibial tunnel (Coleridge and Amis, 2004; Kousa et al., 2003; Magen et al., 1999; Novak et al., 1996). Our methodology more accurately simulated in vivo conditions than straight tensile testing. With our configuration, measured ATT also included any graft deformations at the tunnel edges that could manifest as increased knee laxity in the ACL-reconstructed knee. Grover et al. (2005) showed that initial tension of a soft-tissue ACL graft, inserted into a cadaveric knee, was not maintained after graft fixation. They found that initial graft tension in human cadaveric knees, reconstructed with double-looped bovine extensor tendons fixed with a SW, decreased by 50% after 3 treatments of 20 loading cycles of 100 N applied anterior tibial force (60 cycles total). This resulted in a mean anterior laxity increase of 2.0 mm. Comparatively, our mean ATT with a SW increased by 2.0 mm after 70 loading cycles while our mean ATT 13
Clinical Biomechanics 70 (2019) 8–15
K.L. Markolf, et al.
specimen could have varied slightly. However, we observed a low standard deviation among the 10 knee intact specimens used to perform baseline ATT measurements, and would not expect the relative laxity measurements between fixation configurations to vary substantially if a different knee had been used. With this approach, degradation of the knee specimen during testing was a concern, specifically local deformation of tibial cortical surfaces at the SW and SP fixation sites resulting from multiple graft insertions. To help minimize this problem, the femoral and tibial fixation sites were reinforced with PMMA acrylic prior to all testing, and a small washer was used beneath the EndoButton™. While these reinforcements are not done in the clinical setting, they were unlikely to significantly affect the findings of this study. Our findings do not take into account the healing process. Therefore, the results of this study are most applicable to the early postoperative period prior to graft incorporation. Lastly, although the testing of tibialis type (anterior versus posterior) and fixation method were randomized, any carryover effects from prior testing, such as tunnel widening or cortical bone/PMMA acrylic weakening, could have increased the variability of the measured ATT. Since the testing order was randomized, these effects would be equally spread throughout the course of the study, and any adverse effects due to degradation of the knee specimen should not have changed the relative differences between the groups.
Brand Jr., J.C., Pienkowski, D., Steenlage, E., Hamilton, D., Johnson, D.L., Caborn, D.N., 2000a. Interference screw fixation strength of a quadrupled hamstring tendon graft is directly related to bone mineral density and insertion torque. Am. J. Sports Med. 28 (5), 705–710. Brand Jr., J., Weiler, A., Caborn, D.N., Brown Jr., C.H., Johnson, D.L., 2000b. Graft fixation in cruciate ligament reconstruction. Am. J. Sports Med. 28 (5), 761–774. Coleridge, S.D., Amis, A.A., 2004. A comparison of five tibial-fixation systems in hamstring-graft anterior cruciate ligament reconstruction. Knee Surg. Sports Traumatol. Arthrosc. 12 (5), 391–397. Efe, T., Bauer, J., Herdrich, S., et al., 2010. Comparison between bovine bone and titanium interference screws for implant fixation in ACL reconstruction: a biomechanical study. Arch. Orthop. Trauma Surg. 130 (8), 993–999. Figueroa, D., Calvo, R., Vaisman, A., Meleán, P., Figueroa, F., 2010. Effect of tendon tensioning: an in vitro study in porcine extensor tendons. Knee 17 (3), 245–248. Flanigan, D.C., Child, J.R., Litsky, A.S., 2012. Testing of double-stranded allografts used in ACL reconstruction. J Knee Surg 25 (5), 385–390. Gabriel, M.T., Wong, E.K., Woo, S.L., Yagi, M., Debski, R.E., 2004. Distribution of in situ forces in the anterior cruciate ligament in response to rotatory loads. J. Orthop. Res. 22 (1), 85–89. Giffin, J.R., Vogrin, T.M., Zantop, T., Woo, S.L., Harner, C.D., 2004. Effects of increasing tibial slope on the biomechanics of the knee. Am. J. Sports Med. 32 (2), 376–382. Giurea, M., Zorilla, P., Amis, A.A., Aichroth, P., 1999. Comparative pull-out and cyclicloading strength tests of anchorage of hamstring tendon grafts in anterior cruciate ligament reconstruction. Am. J. Sports Med. 27 (5), 621–625. Grood, E.S., Suntay, W.J., 1983. A joint coordinate system for the clinical description of three-dimensional motions: application to the knee. J. Biomech. Eng. 105 (2), 136–144. Grood, E.S., Walz-Hasselfeld, K.A., Holden, J.P., et al., 1992. The correlation between anterior-posterior translation and cross-sectional area of anterior cruciate ligament reconstructions. J. Orthop. Res. 10 (6), 878–885. Grover, D.M., Howell, S.M., Hull, M.L., 2005. Early tension loss in an anterior cruciate ligament graft. A cadaver study of four tibial fixation devices. J. Bone Joint Surg. Am. 87 (2), 381–390. Harilainen, A., Linko, E., Sandelin, J., 2006. Randomized prospective study of ACL reconstruction with interference screw fixation in patellar tendon autografts versus femoral metal plate suspension and tibial post fixation in hamstring tendon autografts: 5-year clinical and radiological follow-up results. Knee Surg. Sports Traumatol. Arthrosc. 14 (6), 517–528. Herrera, A., Martínez, F., Iglesias, D., Cegoñino, J., Ibarz, E., Gracia, L., 2010. Fixation strength of biocomposite wedge interference screw in ACL reconstruction: effect of screw length and tunnel/screw ratio. A controlled laboratory study. BMC Musculoskelet. Disord. 11, 139. Höher, J., Scheffler, S.U., Withrow, J.D., et al., 2000. Mechanical behavior of two hamstring graft constructs for reconstruction of the anterior cruciate ligament. J. Orthop. Res. 18 (3), 456–461. Howard ME, Cawley PW, Losse GM, Johnston RB 3rd. Bone-patellar tendon-bone grafts for anterior cruciate ligament reconstruction: the effects of graft pretensioning. Arthroscopy. 1996;12(3):287–292. Kanamori, A., Sakane, M., Zeminski, J., Rudy, T.W., Woo, S.L., 2000. In-situ force in the medial and lateral structures of intact and ACL-deficient knees. J. Orthop. Sci. 5 (6), 567–571. Kim, S.G., Kurosawa, H., Sakuraba, K., Ikeda, H., Takazawa, S., 2006. The effect of initial graft tension on postoperative clinical outcome in anterior cruciate ligament reconstruction with semitendinosus tendon. Arch. Orthop. Trauma Surg. 126 (4), 260–264. Kousa, P., Järvinen, T.L., Vihavainen, M., Kannus, P., Järvinen, M., 2003. The fixation strength of six hamstring tendon graft fixation devices in anterior cruciate ligament reconstruction. Part II: tibial site. Am. J. Sports Med. 31 (2), 182–188. Kurosaka, M., Yoshiya, S., Andrish, J.T., 1987. A biomechanical comparison of different surgical techniques of graft fixation in anterior cruciate ligament reconstruction. Am. J. Sports Med. 15 (3), 225–229. Loh, J.C., Fukuda, Y., Tsuda, E., Steadman, R.J., Fu, F.H., Woo, S.L., 2003. Knee stability and graft function following anterior cruciate ligament reconstruction: comparison between 11 o’clock and 10 o’clock femoral tunnel placement. 2002 Richard O’Connor Award paper. Arthroscopy 19 (3), 297–304. Ma, C.B., Francis, K., Towers, J., Irrgang, J., Fu, F.H., Harner, C.H., 2004. Hamstring anterior cruciate ligament reconstruction: a comparison of bioabsorbable interference screw and endobutton-post fixation. Arthroscopy 20 (2), 122–128. Magen, H.E., Howell, S.M., Hull, M.L., 1999. Structural properties of six tibial fixation methods for anterior cruciate ligament soft tissue grafts. Am. J. Sports Med. 27 (1), 35–43. Meuffels, D.E., Docter, P.T., van Dongen, R.A., Kleinrensink, G.J., Verhaar, J.A., Reijman, M., 2010. Stiffer fixation of the tibial double-tunnel anterior cruciate ligament complex versus the single tunnel: a biomechanical study. Arthroscopy 26 (9 Suppl), S35–S40. Micucci, C.J., Frank, D.A., Kompel, J., Muffly, M., Demeo, P.J., Altman, G.T., 2010. The effect of interference screw diameter on fixation of soft-tissue grafts in anterior cruciate ligament reconstruction. Arthroscopy 26 (8), 1105–1110. Novak, P.J., Wexler, G.M., Williams Jr., J.S., Bach Jr., B.R., Bush-Joseph, C.A., 1996. Comparison of screw post fixation and free bone block interference fixation for anterior cruciate ligament soft tissue grafts: biomechanical considerations. Arthroscopy 12 (4), 470–473. Numazaki, H., Tohyama, H., Nakano, H., Kikuchi, S., Yasuda, K., 2002. The effect of initial graft tension in anterior cruciate ligament reconstruction on the mechanical behaviors of the femur-graft-tibia complex during cyclic loading. Am. J. Sports Med. 30 (6), 800–805.
5. Conclusions Anterior knee laxity after cyclic loading was significantly greater with SP fixation compared to SW fixation regardless of post location. Suture tightening around the graft tissue was observed to be the major contributor to increased laxity with the SP construct. Increased anterior knee laxity in the early post-operative period could lead to knee instability and eventual clinical failure. Based on our findings, the efficacy of SP fixation for ACL reconstruction is questioned. Declaration of Competing Interest We wish to draw the attention of the Editor to the following facts which may be considered as potential conflicts of interest and to significant financial contributions to this work. [OR] We wish to confirm that there are no known conflicts of interest associated with this publication and there has been no significant financial support for this work that could have influenced its outcome. We confirm that the manuscript has been read and approved by all named authors and that there are no other persons who satisfied the criteria for authorship but are not listed. We further confirm that the order of authors listed in the manuscript has been approved by all of us. We confirm that we have given due consideration to the protection of intellectual property associated with this work and that there are no impediments to publication, including the timing of publication, with respect to intellectual property. In so doing we confirm that we have followed the regulations of our institutions concerning intellectual property. Acknowledgements This study was supported in part by the H H Lee Surgical Research Program and the Musculoskeletal Transplant Foundation (MTF Grant #20130216). Tissues for this study were provided by the Musculoskeletal Transplantation Foundation. References Arnold, M.P., Lie, D.T., Verdonschot, N., de Graaf, R., Amis, A.A., van Kampen, A., 2005. The remains of anterior cruciate ligament graft tension after cyclic knee motion. Am. J. Sports Med. 33 (4), 536–542.
14
Clinical Biomechanics 70 (2019) 8–15
K.L. Markolf, et al. Nyland, J., Lee, Y.H., McGinnis, M., Kibbe, S., Kocabey, Y., Caborn, D.N., 2014. ACL double bundle linked cortical-aperture tibial fixation: a technical note. Arch. Orthop. Trauma Surg. 134 (6), 835–842. Pearsall 4th, A.W., Hollis, J.M., Russell Jr., G.V., Scheer, Z., 2003. A biomechanical comparison of three lower extremity tendons for ligamentous reconstruction about the knee. Arthroscopy 19 (10), 1091–1096. Pinczewski, L.A., Lyman, J., Salmon, L.J., Russell, V.J., Roe, J., Linklater, J., 2007. A 10year comparison of anterior cruciate ligament reconstructions with hamstring tendon and patellar tendon autograft: a controlled, prospective trial. Am. J. Sports Med. 35 (4), 564–574. Prado, M., Martín-Castilla, B., Espejo-Reina, A., Serrano-Fernández, J.M., Pérez-Blanca, A., Ezquerro, F., 2013. Close-looped graft suturing improves mechanical properties of interference screw fixation in ACL reconstruction. Knee Surg. Sports Traumatol. Arthrosc. 21 (2), 476–484. Prodromos, C.C., Joyce, B.T., Shi, K., Keller, B.L., 2005. A meta-analysis of stability after anterior cruciate ligament reconstruction as a function of hamstring versus patellar tendon graft and fixation type. Arthroscopy 21 (10), 1202. Rodeo, S.A., Arnoczky, S.P., Torzilli, P.A., Hidaka, C., Warren, R.F., 1993. Tendon-healing
in a bone tunnel. A biomechanical and histological study in the dog. J. Bone Joint Surg. Am. 75 (12), 1795–1803. Sasaki, N., Farraro, K.F., Kim, K.E., Woo, S.L., 2014. Biomechanical evaluation of the quadriceps tendon autograft for anterior cruciate ligament reconstruction: a cadaveric study. Am. J. Sports Med. 42 (3), 723–730. Speziali, A., Delcogliano, M., Tei, M., et al., 2014. Fixation techniques for the anterior cruciate ligament reconstruction: early follow-up. A systematic review of level I and II therapeutic studies. Musculoskelet. Surg. 98 (3), 179–187. Weimann, A., Rodieck, M., Zantop, T., Hassenpflug, J., Petersen, W., 2005. Primary stability of hamstring graft fixation with biodegradable suspension versus interference screws. Arthroscopy 21 (3), 266–274. Yasuda, K., Tsujino, J., Tanabe, Y., Kaneda, K., 1997. Effects of initial graft tension on clinical outcome after anterior cruciate ligament reconstruction. Autogenous doubled hamstring tendons connected in series with polyester tapes. Am. J. Sports Med. 25 (1), 99–106. Zamarra, G., Fisher, M.B., Woo, S.L., Cerulli, G., 2010. Biomechanical evaluation of using one hamstrings tendon for ACL reconstruction: a human cadaveric study. Knee Surg. Sports Traumatol. Arthrosc. 18 (1), 11–19.
15
Contents lists available at ScienceDirect
Clinical Biomechanics journal homepage: www.elsevier.com/locate/clinbiomech
Cyclic testing of tibialis tendon allografts for anterior cruciate ligament reconstruction using suture-post versus spiked washer tibial fixation
T
Keith L. Markolf , Dean Wang, Nirav B. Joshi, Edward Cheung, Frank A. Petrigliano, David R. McAllister ⁎
Department of Orthopaedic Surgery, David Geffen School of Medicine at UCLA, 10833 Le Conte Ave, 76-143 CHS, Los Angeles, CA 90095, USA
ARTICLE INFO
ABSTRACT
Keywords: Graft fixation Cyclic testing ACL reconstruction
Background: The purpose of this study was to directly compare spiked washer and suture-post tibial-sided fixation techniques used for anterior cruciate ligament reconstruction by measuring anterior tibial translation during cyclic tests. Methods: Fresh-frozen human knees were tested using a robotic system that applied 250 cycles of anteriorposterior tibial force (134 N) at 30° flexion, while recording tibial translation. Ten intact knees were tested to collect baseline data for native specimens. A single knee was selected to test ligament reconstructions using doubled tibialis tendon allografts. All grafts were fixed proximally using an EndoButton™, and the tibial end of the graft was fixed with either a spiked washer or with a suture post placed at two different locations (near and distant) relative to the tibial tunnel. Findings: Mean first cycle translation for intact knees was 4.8 (sd 1.8) mm; means after reconstruction were 2.6 (sd 0.9) mm (spiked washer), 10.1 (sd 1.9) mm (suture post near), and 10.4 (sd 1.5) mm (suture post distant). Corresponding means for translation increase over 250 cycles were 0.3 (sd 0.2) mm, 3.6 (sd 1.3) mm, 7.2 mm (sd 0.9) mm, and 8.0 (sd 1.3) mm. All mean increases (first cycle and cyclic) after ACL reconstruction were significantly greater than those for the intact knees, and all means with a suture post were significantly greater than those with a spiked washer. There were no significant differences between mean translations for near and distant suture post locations. Interpretation: Use of suture post fixation for anterior cruciate ligament reconstruction is questioned since increases in anterior tibial translation could lead to excessive post-operative knee laxity and possibly early clinical failure.
1. Introduction Rigid fixation is one of the most important factors influencing the mechanical performance of an anterior cruciate ligament (ACL) graft in the immediate post-operative period (Kurosaka et al., 1987). During the 12 week process of osseointegration (Rodeo et al., 1993), fixation must be secure enough to prevent graft slippage and/or local elongation at the fixation site. The tibial side fixation of a graft construct is generally the weak link because bone quality of the tibial metaphysis is inferior to that of the femur, and the tissue ends of a soft-tissue tendon graft are more difficult to secure (Brand Jr et al., 2000a; Brand Jr et al., 2000b). A suture-post (SP) is still commonly used to fix the tibial end of an ACL graft (Speziali et al., 2014). With this technique, suture at the free ends of a tendon graft are tied around a post beneath a flat washer. However, the suture stitches can tighten and compress into the graft
⁎
Corresponding author. E-mail address: [email protected] (K.L. Markolf).
https://doi.org/10.1016/j.clinbiomech.2019.07.025 Received 30 May 2018; Accepted 23 July 2019 0268-0033/ © 2019 Elsevier Ltd. All rights reserved.
tissue as they are subjected to repetitive loading. This could potentially result in increased laxity of the ACL-reconstructed knee. In addition, optimal placement of the SP relative to the tibial tunnel is unknown. In theory, fixing the SP at a location more distant from the tibial tunnel could result in increased knee laxity due to a longer and more compliant graft-suture construct, but this effect has not been investigated experimentally. Tibial fixation using a spiked washer (SW) should theoretically produce less laxity than a SP because the tissue is compressed directly onto cortical bone proximal to the sutures, thereby eliminating any effects of suture tightening within the graft tissue. There have been a number of prior studies comparing the strength and stiffness of various tibial fixation devices, with most performed using bovine tibias (Flanigan et al., 2012; Giurea et al., 1999; Micucci et al., 2010; Weimann et al., 2005) or porcine tibias (Efe et al., 2010; Herrera et al., 2010; Magen et al., 1999; Meuffels et al., 2010;
Clinical Biomechanics 70 (2019) 8–15
K.L. Markolf, et al.
Numazaki et al., 2002; Nyland et al., 2014; Prado et al., 2013). In these studies, tibial sided graft fixation was examined using uniaxial test machines to apply force directly in line with the graft tissue. However, a uniaxial test machine does not accurately represent in vivo knee loading conditions and the effects of local contact stresses at the graftbone interfaces within the bone tunnels. Therefore, it is desirable to evaluate tibial fixation devices by performing ACL reconstruction in human specimens and loading the knees in a manner that more closely replicates in vivo conditions. The objective of this study was to directly compare SW and SP tibial fixation by measuring anterior knee laxity during cyclic anteroposterior (AP) tibial loading of an ACL-reconstructed knee. We hypothesized that SP fixation would produce significantly greater anterior knee laxity than SW fixation, and that locating the SP more distant from the tibial tunnel exit (thereby increasing overall graft-suture construct length) would result in significantly increased anterior knee laxity. 2. Methods 2.1. Specimen preparation Fresh-frozen tibialis tendon allografts were used for this study. It has been shown previously that tibialis anterior and tibialis posterior tendons have comparable structural properties and display similar viscoelastic behavior (Pearsall 4th et al., 2003). Therefore, for the purposes of this study, tibialis anterior and posterior tendons were considered to be equivalent. A total of 15 anterior and 15 posterior tibialis tendon grafts were harvested from human donors (mean age 46 years, range 16–64). Within each fixation group, 5 anterior and 5 posterior tibialis tendons were tested in a randomized order. All grafts were cleared of adherent muscle fibers and surrounding soft tissues, wrapped in saline-soaked gauze, and stored frozen at −20 °C. On the day of testing, the graft was thawed to room temperature and kept moist with saline solution during specimen preparation, fixation, and biomechanical testing. The tendon was passed through the loop of a cortical fixation device (30 mm EndoButton™, Smith & Nephew, Andover, MA) to produce a double-stranded graft (Fig. 1). No.2 polyester suture (Ethibond™, Ethicon, Somerville, NJ) was sutured into each free tendon end of the construct using a running-locked suture technique. The cross-sectional area of each doubled graft was measured under a compressive pressure of 0.12 MPa applied for 2 min using an established technique (Grood et al., 1992). Ten fresh-frozen human cadaveric knee specimens were used to collect baseline ATT data for the native ACL (mean age 34 years, range 21–45). The tibia and femur were sectioned mid-shaft and scraped clean of soft tissue to within 10 cm of the joint line, and the bone ends were potted in cylindrical molds of polymethylmethacrylate (PMMA) for attachment to testing fixtures. The soft tissues surrounding the knee capsule were left intact to help prevent dehydration of ligamentous tissues within the knee. To reduce possible variability due to knee geometry and size, a single specimen was selected for use in testing all ACL reconstructions (male, 44 years of age). This allowed direct comparisons of tibial fixation methods within the same knee. The native ACL was excised and 11 mm diameter femoral and tibial tunnels were drilled within the footprints of the native ACL in accordance with current clinical practice using the anteromedial portal technique. An 11 mm femoral tunnel was drilled (inside-out) to a depth of 30 mm. An additional 4.5 mm tunnel (used for passage of the
Fig. 2. EndoButton™ femoral fixation at the lateral cortex of the femoral tunnel exit reinforced with PMMA acrylic.
EndoButton™) was drilled from the internal base of the femoral tunnel to the lateral cortex of the femur. An 11 mm tibial tunnel was drilled from the tibial cortex to the center of the ACL footprint using a drill guide. The EndoButton™ (with attached synthetic fiber loop and graft) was pulled through the femoral tunnel and 4.5 mm passage tunnel. When in place, the button contacted a small metal washer placed over the tunnel opening at the lateral femoral cortex (Fig. 2). The cortical bone surface at the exit of the femoral tunnel was reinforced with PMMA acrylic. These modifications were made to prevent the possibility of cumulative bone deformation at this site from repeated testing. Before insertion, each graft was preconditioned on a tensioning board (Graftmaster II; Smith & Nephew, Andover, MA) in accordance with clinical protocol for this device. The graft construct was fixed with an initial 89 N tensile force and held on the tension board for 20 min. This level of force has been used in previous studies (Figueroa et al., 2010; Höher et al., 2000; Howard et al., 1996). After preconditioning, the graft was immediately inserted into the knee and secured to the femur as described above. The tibial end of the graft was tensioned to 89 N at 30° of knee flexion during final fixation. This tensioning protocol was based on previously published randomized clinical trials (Kim et al., 2006; Yasuda et al., 1997). Tibial fixation of the graft was accomplished using either a SW (18 mm diameter with a 6.5 mm × 40 mm bicortical screw; Fig. 3A; Arthrex, Inc., Naples, FL) or a SP (18 mm flat washer with a 6.5 mm × 40 mm bicortical screw; Fig. 3B; Arthrex, Inc., Naples, FL). Both were secured with a nut against the opposite cortex. For SW grafts, the sutures from both ends of the graft were tied together and looped over the hook of a spring scale to maintain constant suture tension during final fixation. While under tension, the two graft ends were
Fig. 1. Tibialis tendon graft preparation. 9
Clinical Biomechanics 70 (2019) 8–15
K.L. Markolf, et al.
FARO Technologies Inc., Lake Mary, FL). The femur was clamped to a fixture mounted to a base plate with the knee flexed to 30°. Under forcemoment control, the robot applied 250 cycles of AP force to the tibia. Per Arnold et al. (2005) who measured tension in BTB grafts with an arthroscopically implantable force probe (AIFP) during 1500 flexionextension cycles, 70% of the total loss in graft tension occurred by cycle 100. To account for variation in graft tissue type and study methodology, we applied 250 cycles of AP tibial force. During each cycle, the robot applied a 134 N posterior tibial force followed by a 134 N anterior tibial force. Target force of 134 N has been used previously in robotic studies related to ACL reconstruction (Gabriel et al., 2004; Giffin et al., 2004; Kanamori et al., 2000; Loh et al., 2003; Sasaki et al., 2014; Zamarra et al., 2010). Applied tibial force and the corresponding AP tibial translation (ATT) were measured continuously during the test. Anterior knee laxity was defined as the ATT recorded during testing, and evaluated by examining the initial ATT (first cycle), cycle 250 ATT, and the increase in ATT from cycle 1 to cycle 250. Ten intact knees were first tested to document baseline increases in ATT with the native ACL. Then, one knee was selected for cyclic testing with all subsequent graft tissues. Three fixation groups (10 grafts each) were tested with this knee: SW, SP-near, and SP-distant. The testing order for all graft tissues was randomized. For the 10 intact knee specimens, the neutral position of the tibia relative to the femur was determined by finding the midpoint between the peak posterior and peak anterior tibial translations at 134 N of applied AP force. The mean neutral position was determined to be 4.8 mm anterior to the PCL endpoint. All ATT measurements were referenced to the intact AP neutral position in order to allow direct comparisons of ATT and increase in ATT between native knees and the two fixation constructs. 2.3. Statistics and data analysis Power analysis was performed based on variance seen in preliminary testing. A sample size of n = 6 was needed to detect a 1 mm difference in ATT with 80% power. To ensure statistical power was achieved, a sample size of n = 10 per fixation method was chosen. The variables analyzed were the first cycle ATT, cycle 250 ATT, and the increase in ATT from cycle 1 to cycle 250. Unpaired two-sample Student's t-tests were used for tibialis anterior versus tibialis posterior comparisons (cross-sectional area, first cycle ATT, and increase in ATT). A similar analysis was also used to compare first cycle ATT, cycle 250 ATT, and increase in ATT between near and distant post locations with the SP construct. A one-way ANOVA was used to compare means for first cycle ATT, cycle 250 ATT, and cyclic increases in ATT between the native ACL group and all fixation methods. Post hoc comparisons were made using Tukey's HSD procedure. The level of significance was p < 0.05.
Fig. 3. A) 18 mm diameter spiked washer with a 6.5 mm × 40 mm bicortical screw. B) 18 mm diameter flat washer with a 6.5 mm × 40 mm bicortical screw. For both fixation device, an opposing nut was used to maintain secure fixation.
positioned beneath the peripheral spikes of the washer and secured onto the PMMA re-enforced cortical surface with a bicortical screw and nut (Fig. 4A). The bicortical screw hole was 2 cm distal to the tibial tunnel, with the sutured portion of the graft tissue distal to the tissue compressed by the SW. For SP grafts, two post locations were evaluated: near (using the SW tibial hole -Fig. 4B) and distant (4 cm distal to the SW tibial hole Fig. 4C). For both groups, the suture ends had to be free so that they could be tied around the post. Therefore a separate suture loop, passing through both graft ends proximal to the sutured portion, was tensioned with the spring scale to maintain a constant 89 N tension in the graft tissue while the sutures were tied around the shank of the post using four square knots. The knotted portion of the sutures was then compressed down onto the PMMA re-enforced surface of the tibia using a screw and flat washer.
3. Results 3.1. Comparison of tibialis anterior and tibialis posterior tendon grafts There were no significant differences in mean cross-sectional area, mean first cycle ATT, or mean increase in ATT between tibialis anterior and posterior tendon grafts for any of the three testing groups (Table 1).
2.2. Specimen testing
3.2. Comparison of ATT between testing groups
A six DOF robotic manipulator (KR210; KUKA Robotics Corp., Clinton Township, MI) with a force-moment sensor (Omega Model Industrial Automation Load Cell, ATI Industrial Automation, Apex, NC) was used to perform all knee loading tests (Fig. 5). Prior to testing, the tibia was clamped in a fixture mounted to the force-moment sensor. Measurements were taken to ensure proper alignment of the anatomic joint coordinate system (Grood and Suntay, 1983) with the axes of the robot and load cell using a three dimensional digitizer (Faro Gage,
Mean first cycle ATT for the SW fixation group was significantly less than that for the intact ACL group (p < 0.01), while means for near and distant suture-post groups were significantly greater than the intact ACL group and SW group (p < 0.01), but not significantly different from each other (Table 2). Means for Cycle 250 ATT and increases in ATT were significantly greater (p < 0.01) than the intact ACL for all fixation groups. Means 10
Clinical Biomechanics 70 (2019) 8–15
K.L. Markolf, et al.
Fig. 4. Tibial fixation methods. A) Spiked washer fixation proximal to the suture. B) Suture-post fixation near to the tibial tunnel (at the spike washer location). C) Suture-post fixation distant to the tibial tunnel (40 mm distal to the spiked washer location). The bone around the fixation site was reinforced with PMMA acrylic.
for near and distant SP groups were significantly greater (p < 0.01) than the SW group, but not significantly different from each other (Table 2). A comparison of mean curves for intact and ACL reconstructed knees during cyclic testing (Fig. 6) illustrates the differences between the intact knee and both fixation constructs, both in terms of initial ATT and cyclic increases in ATT. It should be noted in Fig. 6 that during cyclic testing, the SP curve demonstrates a progressively longer region of low stiffness (slope of the force vs. ATT curve). During the final SP
test cycles, the tibia had displaced anteriorly approximately 10 mm before the SP construct began to resist the applied anterior tibial force. This indicates that the graft was essentially slack in this region, in marked contrast to the behavior of the native ACL and grafts fixed with a SW. At the conclusion of testing, clear evidence of suture tightening was observed for all SP grafts with deep suture grooves imprinted into the ends of the graft tissue (Fig. 7).
11
Clinical Biomechanics 70 (2019) 8–15
K.L. Markolf, et al.
Fig. 5. Six degree-of-freedom robot used to apply 250 cycles of 134 N anteroposterior tibial force while recording anterior tibial translation.
of this laxity is related to local deformations at the graft fixation interfaces on the tibia. This cyclic loading study compared ATTs with knees reconstructed using tibialis tendon grafts that were fixed at their tibial ends with a SW or SP. SP fixation produced significant increases in anterior knee laxity compared to SW fixation, regardless of the position of the post. First cycle ATT with the near SP was 4 times greater and the increase in ATT over 250 cycles was 2 times greater compared to a SW. After completion of 250 test cycles, the mean ATT with SW grafts was 1.1 mm greater than that for intact knees. In contrast, mean ATT's with both near and distant SP grafts were approximately 12 mm greater than the intact knee, a condition that would be unacceptable clinically. These finding are important, as increases in post-operative knee laxity has been associated with ACL graft rupture (Pinczewski et al., 2007). Clinical studies have demonstrated that patients undergoing ACL reconstruction using a hamstring autograft with SP fixation show comparable joint stability and subjective patient outcomes compared to those using a bone-patellar tendon-bone autograft or a hamstring autograft with interference screw fixation (Harilainen et al., 2006; Ma et al., 2004; Prodromos et al., 2005). However, only two prior studies have directly compared SW and SP fixation. While direct comparison with our study is difficult, the results of these studies support our findings. Magen et al. (1999) performed biomechanical testing of bovine extensor tendons fixed to porcine tibias using six devices. Among those tested were SP and a 20-mm metal SW. The SP construct exhibited lower stiffness (60 N/mm) and yield load (830 N) than the SW constructs (126 N/mm and 930 N). However, cycling was not performed. Höher et al. (2000) compared SP fixation to dual SW fixation (use of two washers) on isolated human cadaveric semitendinosus and gracilis tendons in a uniaxial test machine. Using a complex cyclic loading protocol, they evaluated the increase in graft elongation from the first cycle to the last cycle (500 cycles). Their SP constructs showed a 3.5 mm greater increase in graft elongation than SW constructs. Similarly, our near SP constructs had a 3.6 mm greater increase in ATT than the SW, and our distant SP constructs has a 4.4 mm greater increase in ATT than SW.
Table 1 Comparison of graft cross-sectional area and Anterior Tibial Translation (ATT) using tibialis anterior and tibialis posterior grafts (mean ± SD). Spiked washer 2
Cross-sectional area (mm ) Cycle 1 ATT (mm) Cycle 250 ATT (mm) ATT increase (mm) Near suture-post Cross-sectional area (mm2) Cycle 1 ATT (mm) Cycle 250 ATT (mm) ATT increase (mm) Distant suture-post Cross-sectional area (mm2) Cycle 1 ATT (mm) Cycle 250 ATT (mm) ATT increase (mm)
Tibialis anterior
Tibialis posterior
46.0 (10.7) 2.2 (1.1) 6.1 (2.3) 3.9 (1.8) Tibialis anterior 43.3 (8.1) 10.0 (1.2) 17.4 (1.8) 7.4 (1.1) Tibialis anterior 50.9 (7.4) 10.5 (1.5) 19.0 (2.0) 8.5 (1.3)
51.1 (8.0) 2.9 (0.5) 6.3 (0.8) 3.4 (0.7) Tibialis posterior 46.7 (7.4) 10.2 (2.6) 17.2 (2.7) 7.1 (0.8) Tibialis posterior 43.7 (4.5) 10.2 (1.7) 17.8 (2.0) 7.6 (1.2)
Table 2 Comparison of Anterior Tibial Translations (ATT) for intact knee, spiked washer, near suture-post, and distant suture-post testing groups (mean ± SD).
Intact knee Spiked washer Near suture-post Distant suture-post ⁎ † ‡
First Cycle ATT (mm)
Cycle 250 ATT (mm)
Increase in ATT (mm)
4.8 (1.8) 2.6 (0.9)⁎ 10.1 (1.9)†,‡ 10.4 (1.5)†,‡
5.1 (1.9) 6.2 (1.6)† 17.3 (2.1)†,‡ 18.4 (2.0)†,‡
0.3 3.6 7.2 8.0
(0.2) (1.3)† (0.9)†,‡ (1.3)†,‡
Significantly less than the native ACL (p < 0.01). Significantly greater than the native ACL (p < 0.01) Significantly greater than spiked washer (p < 0.01).
4. Discussion Rigid fixation of an ACL graft reconstruction is desirable to avoid unwanted post-operative anterior knee laxity. An important component 12
Clinical Biomechanics 70 (2019) 8–15
K.L. Markolf, et al.
Fig. 6. Tracing of 250 anteroposterior loading cycles of 134 N (applied tibial force vs. tibial translation). The curves for each condition represent the mean of all 10 specimens. Suture-post fixation (showing the near post location) has significantly greater first cycle ATT and increase in ATT over 250 loading cycles compared to both the native ACL and spiked washer fixation (p < 0.01). Peak posterior displacement was consistent for all tests.
with a SP increased by 2.0 mm after only 8 loading cycles. There is a major difference between the two fixation constructs that is important when interpreting our results. With SW fixation, the sutured portion of the graft was distal to the washer and could not affect ATT measurements. With SP fixation, the sutured portion of the graft was proximal to the tibial fixation site, thus any suture tightening around the graft tissue would manifest as additional ATT. We believe the low stiffness region of the SP test curve (Fig. 6) was caused by progressive suture tightening around the graft tissue as cyclic anterior tibial force was applied. The deep suture grooving within the SP grafts after suture removal was direct evidence of suture tightening (Fig. 7). We preconditioned all grafts on a tension board by applying 89 N of tensile force for 20 min. With SP grafts, this preconditioning protocol should have helped reduce the effects of suture tightening prior to graft implantation. However, our results showed that additional suture tightening with SP grafts occurred during cyclic testing. It is possible that some of the initial suture tightening produced by the tension board could have relaxed during the time the graft was removed from the tension board and fixed within the knee (typically < 5 min). It should also be noted that the tension board force level was a constant 89 N, while the cyclic tests were performed with 134 N anterior tibial force. Our findings related to the near or distant location of a SP relative to the tibial tunnel entrance deserve special mention. Clinically, the optimal position of the SP relative to the tibial tunnel is unknown. We hypothesized that moving SP fixation 4 cm distally would increase anterior laxity because this would result in a longer (and theoretically more compliant) graft construct. This was not found to be the case, as there were no significant differences in ATT and increase in ATT between near and distant SP locations. This finding suggests that any increase in anterior laxity produced by the additional 4 cm of graft tissue was insignificant compared to laxity increases produced by suture tightening. It should be noted that the distance of 4 cm was selected to simulate an extreme condition, and represented a far greater distal placement than a clinician would consider. Therefore, our results would indicate that clinicians using this method of graft fixation have considerable flexibility in near-distant placement for SP fixation in terms of its effect on anterior knee laxity. This study had several limitations. A single cadaveric knee was used for testing of all graft tissues to eliminate any potential variability due to anatomical factors. Laxity measurements using a different knee
Fig. 7. A double stranded tibialis graft with suture-post fixation after completion of cyclic testing and graft removal. Suture tightening at both ends of the graft tissue is apparent, as evidenced by deep grooves near the graft ends.
The majority of prior studies investigating graft fixation have loaded the graft in line with the axis of the tibial tunnel, which is non-physiologic. In vivo tibial loading occurs in the anterior direction, and the tensile force acting on the fixation construct is reduced due to the frictional losses as the graft bends around the edge of the tibial tunnel (Coleridge and Amis, 2004; Kousa et al., 2003; Magen et al., 1999; Novak et al., 1996). Our methodology more accurately simulated in vivo conditions than straight tensile testing. With our configuration, measured ATT also included any graft deformations at the tunnel edges that could manifest as increased knee laxity in the ACL-reconstructed knee. Grover et al. (2005) showed that initial tension of a soft-tissue ACL graft, inserted into a cadaveric knee, was not maintained after graft fixation. They found that initial graft tension in human cadaveric knees, reconstructed with double-looped bovine extensor tendons fixed with a SW, decreased by 50% after 3 treatments of 20 loading cycles of 100 N applied anterior tibial force (60 cycles total). This resulted in a mean anterior laxity increase of 2.0 mm. Comparatively, our mean ATT with a SW increased by 2.0 mm after 70 loading cycles while our mean ATT 13
Clinical Biomechanics 70 (2019) 8–15
K.L. Markolf, et al.
specimen could have varied slightly. However, we observed a low standard deviation among the 10 knee intact specimens used to perform baseline ATT measurements, and would not expect the relative laxity measurements between fixation configurations to vary substantially if a different knee had been used. With this approach, degradation of the knee specimen during testing was a concern, specifically local deformation of tibial cortical surfaces at the SW and SP fixation sites resulting from multiple graft insertions. To help minimize this problem, the femoral and tibial fixation sites were reinforced with PMMA acrylic prior to all testing, and a small washer was used beneath the EndoButton™. While these reinforcements are not done in the clinical setting, they were unlikely to significantly affect the findings of this study. Our findings do not take into account the healing process. Therefore, the results of this study are most applicable to the early postoperative period prior to graft incorporation. Lastly, although the testing of tibialis type (anterior versus posterior) and fixation method were randomized, any carryover effects from prior testing, such as tunnel widening or cortical bone/PMMA acrylic weakening, could have increased the variability of the measured ATT. Since the testing order was randomized, these effects would be equally spread throughout the course of the study, and any adverse effects due to degradation of the knee specimen should not have changed the relative differences between the groups.
Brand Jr., J.C., Pienkowski, D., Steenlage, E., Hamilton, D., Johnson, D.L., Caborn, D.N., 2000a. Interference screw fixation strength of a quadrupled hamstring tendon graft is directly related to bone mineral density and insertion torque. Am. J. Sports Med. 28 (5), 705–710. Brand Jr., J., Weiler, A., Caborn, D.N., Brown Jr., C.H., Johnson, D.L., 2000b. Graft fixation in cruciate ligament reconstruction. Am. J. Sports Med. 28 (5), 761–774. Coleridge, S.D., Amis, A.A., 2004. A comparison of five tibial-fixation systems in hamstring-graft anterior cruciate ligament reconstruction. Knee Surg. Sports Traumatol. Arthrosc. 12 (5), 391–397. Efe, T., Bauer, J., Herdrich, S., et al., 2010. Comparison between bovine bone and titanium interference screws for implant fixation in ACL reconstruction: a biomechanical study. Arch. Orthop. Trauma Surg. 130 (8), 993–999. Figueroa, D., Calvo, R., Vaisman, A., Meleán, P., Figueroa, F., 2010. Effect of tendon tensioning: an in vitro study in porcine extensor tendons. Knee 17 (3), 245–248. Flanigan, D.C., Child, J.R., Litsky, A.S., 2012. Testing of double-stranded allografts used in ACL reconstruction. J Knee Surg 25 (5), 385–390. Gabriel, M.T., Wong, E.K., Woo, S.L., Yagi, M., Debski, R.E., 2004. Distribution of in situ forces in the anterior cruciate ligament in response to rotatory loads. J. Orthop. Res. 22 (1), 85–89. Giffin, J.R., Vogrin, T.M., Zantop, T., Woo, S.L., Harner, C.D., 2004. Effects of increasing tibial slope on the biomechanics of the knee. Am. J. Sports Med. 32 (2), 376–382. Giurea, M., Zorilla, P., Amis, A.A., Aichroth, P., 1999. Comparative pull-out and cyclicloading strength tests of anchorage of hamstring tendon grafts in anterior cruciate ligament reconstruction. Am. J. Sports Med. 27 (5), 621–625. Grood, E.S., Suntay, W.J., 1983. A joint coordinate system for the clinical description of three-dimensional motions: application to the knee. J. Biomech. Eng. 105 (2), 136–144. Grood, E.S., Walz-Hasselfeld, K.A., Holden, J.P., et al., 1992. The correlation between anterior-posterior translation and cross-sectional area of anterior cruciate ligament reconstructions. J. Orthop. Res. 10 (6), 878–885. Grover, D.M., Howell, S.M., Hull, M.L., 2005. Early tension loss in an anterior cruciate ligament graft. A cadaver study of four tibial fixation devices. J. Bone Joint Surg. Am. 87 (2), 381–390. Harilainen, A., Linko, E., Sandelin, J., 2006. Randomized prospective study of ACL reconstruction with interference screw fixation in patellar tendon autografts versus femoral metal plate suspension and tibial post fixation in hamstring tendon autografts: 5-year clinical and radiological follow-up results. Knee Surg. Sports Traumatol. Arthrosc. 14 (6), 517–528. Herrera, A., Martínez, F., Iglesias, D., Cegoñino, J., Ibarz, E., Gracia, L., 2010. Fixation strength of biocomposite wedge interference screw in ACL reconstruction: effect of screw length and tunnel/screw ratio. A controlled laboratory study. BMC Musculoskelet. Disord. 11, 139. Höher, J., Scheffler, S.U., Withrow, J.D., et al., 2000. Mechanical behavior of two hamstring graft constructs for reconstruction of the anterior cruciate ligament. J. Orthop. Res. 18 (3), 456–461. Howard ME, Cawley PW, Losse GM, Johnston RB 3rd. Bone-patellar tendon-bone grafts for anterior cruciate ligament reconstruction: the effects of graft pretensioning. Arthroscopy. 1996;12(3):287–292. Kanamori, A., Sakane, M., Zeminski, J., Rudy, T.W., Woo, S.L., 2000. In-situ force in the medial and lateral structures of intact and ACL-deficient knees. J. Orthop. Sci. 5 (6), 567–571. Kim, S.G., Kurosawa, H., Sakuraba, K., Ikeda, H., Takazawa, S., 2006. The effect of initial graft tension on postoperative clinical outcome in anterior cruciate ligament reconstruction with semitendinosus tendon. Arch. Orthop. Trauma Surg. 126 (4), 260–264. Kousa, P., Järvinen, T.L., Vihavainen, M., Kannus, P., Järvinen, M., 2003. The fixation strength of six hamstring tendon graft fixation devices in anterior cruciate ligament reconstruction. Part II: tibial site. Am. J. Sports Med. 31 (2), 182–188. Kurosaka, M., Yoshiya, S., Andrish, J.T., 1987. A biomechanical comparison of different surgical techniques of graft fixation in anterior cruciate ligament reconstruction. Am. J. Sports Med. 15 (3), 225–229. Loh, J.C., Fukuda, Y., Tsuda, E., Steadman, R.J., Fu, F.H., Woo, S.L., 2003. Knee stability and graft function following anterior cruciate ligament reconstruction: comparison between 11 o’clock and 10 o’clock femoral tunnel placement. 2002 Richard O’Connor Award paper. Arthroscopy 19 (3), 297–304. Ma, C.B., Francis, K., Towers, J., Irrgang, J., Fu, F.H., Harner, C.H., 2004. Hamstring anterior cruciate ligament reconstruction: a comparison of bioabsorbable interference screw and endobutton-post fixation. Arthroscopy 20 (2), 122–128. Magen, H.E., Howell, S.M., Hull, M.L., 1999. Structural properties of six tibial fixation methods for anterior cruciate ligament soft tissue grafts. Am. J. Sports Med. 27 (1), 35–43. Meuffels, D.E., Docter, P.T., van Dongen, R.A., Kleinrensink, G.J., Verhaar, J.A., Reijman, M., 2010. Stiffer fixation of the tibial double-tunnel anterior cruciate ligament complex versus the single tunnel: a biomechanical study. Arthroscopy 26 (9 Suppl), S35–S40. Micucci, C.J., Frank, D.A., Kompel, J., Muffly, M., Demeo, P.J., Altman, G.T., 2010. The effect of interference screw diameter on fixation of soft-tissue grafts in anterior cruciate ligament reconstruction. Arthroscopy 26 (8), 1105–1110. Novak, P.J., Wexler, G.M., Williams Jr., J.S., Bach Jr., B.R., Bush-Joseph, C.A., 1996. Comparison of screw post fixation and free bone block interference fixation for anterior cruciate ligament soft tissue grafts: biomechanical considerations. Arthroscopy 12 (4), 470–473. Numazaki, H., Tohyama, H., Nakano, H., Kikuchi, S., Yasuda, K., 2002. The effect of initial graft tension in anterior cruciate ligament reconstruction on the mechanical behaviors of the femur-graft-tibia complex during cyclic loading. Am. J. Sports Med. 30 (6), 800–805.
5. Conclusions Anterior knee laxity after cyclic loading was significantly greater with SP fixation compared to SW fixation regardless of post location. Suture tightening around the graft tissue was observed to be the major contributor to increased laxity with the SP construct. Increased anterior knee laxity in the early post-operative period could lead to knee instability and eventual clinical failure. Based on our findings, the efficacy of SP fixation for ACL reconstruction is questioned. Declaration of Competing Interest We wish to draw the attention of the Editor to the following facts which may be considered as potential conflicts of interest and to significant financial contributions to this work. [OR] We wish to confirm that there are no known conflicts of interest associated with this publication and there has been no significant financial support for this work that could have influenced its outcome. We confirm that the manuscript has been read and approved by all named authors and that there are no other persons who satisfied the criteria for authorship but are not listed. We further confirm that the order of authors listed in the manuscript has been approved by all of us. We confirm that we have given due consideration to the protection of intellectual property associated with this work and that there are no impediments to publication, including the timing of publication, with respect to intellectual property. In so doing we confirm that we have followed the regulations of our institutions concerning intellectual property. Acknowledgements This study was supported in part by the H H Lee Surgical Research Program and the Musculoskeletal Transplant Foundation (MTF Grant #20130216). Tissues for this study were provided by the Musculoskeletal Transplantation Foundation. References Arnold, M.P., Lie, D.T., Verdonschot, N., de Graaf, R., Amis, A.A., van Kampen, A., 2005. The remains of anterior cruciate ligament graft tension after cyclic knee motion. Am. J. Sports Med. 33 (4), 536–542.
14
Clinical Biomechanics 70 (2019) 8–15
K.L. Markolf, et al. Nyland, J., Lee, Y.H., McGinnis, M., Kibbe, S., Kocabey, Y., Caborn, D.N., 2014. ACL double bundle linked cortical-aperture tibial fixation: a technical note. Arch. Orthop. Trauma Surg. 134 (6), 835–842. Pearsall 4th, A.W., Hollis, J.M., Russell Jr., G.V., Scheer, Z., 2003. A biomechanical comparison of three lower extremity tendons for ligamentous reconstruction about the knee. Arthroscopy 19 (10), 1091–1096. Pinczewski, L.A., Lyman, J., Salmon, L.J., Russell, V.J., Roe, J., Linklater, J., 2007. A 10year comparison of anterior cruciate ligament reconstructions with hamstring tendon and patellar tendon autograft: a controlled, prospective trial. Am. J. Sports Med. 35 (4), 564–574. Prado, M., Martín-Castilla, B., Espejo-Reina, A., Serrano-Fernández, J.M., Pérez-Blanca, A., Ezquerro, F., 2013. Close-looped graft suturing improves mechanical properties of interference screw fixation in ACL reconstruction. Knee Surg. Sports Traumatol. Arthrosc. 21 (2), 476–484. Prodromos, C.C., Joyce, B.T., Shi, K., Keller, B.L., 2005. A meta-analysis of stability after anterior cruciate ligament reconstruction as a function of hamstring versus patellar tendon graft and fixation type. Arthroscopy 21 (10), 1202. Rodeo, S.A., Arnoczky, S.P., Torzilli, P.A., Hidaka, C., Warren, R.F., 1993. Tendon-healing
in a bone tunnel. A biomechanical and histological study in the dog. J. Bone Joint Surg. Am. 75 (12), 1795–1803. Sasaki, N., Farraro, K.F., Kim, K.E., Woo, S.L., 2014. Biomechanical evaluation of the quadriceps tendon autograft for anterior cruciate ligament reconstruction: a cadaveric study. Am. J. Sports Med. 42 (3), 723–730. Speziali, A., Delcogliano, M., Tei, M., et al., 2014. Fixation techniques for the anterior cruciate ligament reconstruction: early follow-up. A systematic review of level I and II therapeutic studies. Musculoskelet. Surg. 98 (3), 179–187. Weimann, A., Rodieck, M., Zantop, T., Hassenpflug, J., Petersen, W., 2005. Primary stability of hamstring graft fixation with biodegradable suspension versus interference screws. Arthroscopy 21 (3), 266–274. Yasuda, K., Tsujino, J., Tanabe, Y., Kaneda, K., 1997. Effects of initial graft tension on clinical outcome after anterior cruciate ligament reconstruction. Autogenous doubled hamstring tendons connected in series with polyester tapes. Am. J. Sports Med. 25 (1), 99–106. Zamarra, G., Fisher, M.B., Woo, S.L., Cerulli, G., 2010. Biomechanical evaluation of using one hamstrings tendon for ACL reconstruction: a human cadaveric study. Knee Surg. Sports Traumatol. Arthrosc. 18 (1), 11–19.
15