Abstract
The cross-stitch peripheral suture has good strength, but the large amount of exposed suture on the tendon surface has restricted its clinical usage. We report a method of embedded cross-stitch that incorporates cross-stitches into peripheral sutures and reduces the amount of exposed suture on the tendon surface. Thirty-three fresh pig flexor tendons were divided equally into three groups and repaired with cross-stitch, embedded cross-stitch, or modified Halsted sutures. The tendons were tested in an Instron tensile machine to assess the mechanical performance of these repairs. With an identical number of strands across the repair site, the gap formation and ultimate forces of the embedded cross-stitch method were statistically greater than those of the cross-stitch and modified Halsted methods. The embedded cross-stitch method also had significantly greater stiffness and energy to failure than the cross-stitch method. The embedded cross-stitch method, with little suture exposure on the tendon and sufficient strength, presents an alternative to the current cross-stitch peripheral repair.
INTRODUCTION
Peripheral sutures for repair of flexor tendon injuries in the hand include simple running, locking running, cross-stitch, horizontal mattress intrafiber, and Halsted continuous horizontal mattress sutures (Lin et al., 1989; Mashadi and Amis, 1989; Silfverskiöld and Anderson, 1993; Wade et al., 1989). Of these methods, running peripheral sutures have been popularly used in clinical practice and, although cross-stitch sutures have sufficient strength, only Silfverskiöld and May (1994), Sirotakova and Elliot (1999) and Taras et al. (1996) have reported their use in clinical practice. Massive exposure of the sutures on the surface of the repaired tendon presents an obstacle to their wider use. A simple running peripheral suture with multiple core sutures provide sufficient strength for a flexor tendon to withstand the tension which occurs in aggressive active rehabilitation programs (Sandow and McMahon, 1996; Savage and Risitano, 1992; Tang 1994; Tang and Shi, 1992; Tang et al., 1994). Nevertheless, stronger peripheral sutures with a simpler core suture may also provide the tendon with sufficient strength, and preferable to multiple core sutures when the anteroposterior diameter of the lacerated tendon is small.
We report a method that incorporates cross-stitches into peripheral sutures and remarkably decreases suture exposure on the tendon surface. We tested the gap formation force, ultimate strength, stiffness, and energy to failure of the tendons repaired with this “embedded cross-stitch” method, and compared its mechanical performance with that of the cross-stitch and a modified Halsted methods.
MATERIALS AND METHODS
Thirty-three fresh flexor tendons were obtained from adult pig trotters. Their structure is similar to that of human tendons and they have been used in previous experimental studies (Cao et al., 2002; Savage, 1985; Wang et al., 2003; Zatiti et al., 1998). After skin incision, the flexor digitorum profundus tendons were exposed through a sheath incision and were completely transected with a surgical blade at the level of the metatarsophalangeal joint. The tendon laceration corresponded to an injury around the metacarpophalangeal joint level in zone 2.
Operative techniques
The 33 flexor tendons were randomly divided into three groups of 11 tendons each and were repaired with either cross-stitch, embedded cross-stitch, or modified Halsted sutures. As illustrated in Fig 1, the cross-stitch peripheral suture was according to Silfverskiöld and Anderson’s description (Silfverskiöld and Anderson, 1993). The embedded cross-stitch suture involved embedment of cross-stitch suture underneath the epitenon of the repaired tendon. Only 2–3 mm of each strand of the embedded cross-stitch suture which ran across the tendon repair site and horizontal strands that grasped the tendon were exposed on the tendon surface. The modified Halsted suture was based on the description of Wade et al. (1989), but the length of exposed horizontal strands was reduced to minimize suture exposure. 5-0 monofilament nylon (Ethilon, Ethicon Inc., Somerville, NJ, USA) was used for all the repairs, and an identical number of peripheral strands (12) were used for each tendon. The length of the tendon with the peripheral suture was uniformly 12 mm, with sutures running about 6 mm into each stump.
Biomechanical testing
The tendons were subjected to a load-to-failure test in an Instron tensile machine (Model 4411, Instron Corp., Canton, MA, USA). The load cell of the tensile machine was 500 N. The tendons were mounted securely in upper and lower clamps of the machine, with a uniform initial distance between the two clamps of 5 cm. Prior tests ensured no slipping or tearing at the clamp–tendon junction when the tendons were loaded until complete failure. With a preload of 1.0 N, the overhead crossbar connected to the upper clamp was advanced at a constant speed of 25 mm/min. The preload and the rate of the tendon pull were to simulate loading conditions of a mobilizing tendon during active finger flexion (Cao et al., 2002; Tan et al., 2003; Tang et al., 2001a; Tang and Xie, 2001; Wada et al., 2001, 2002; Xie et al., 2002). A micrometer digital caliper was attached close to the tested tendon segment and the force that produced 2 mm of separation of tendon stumps was recorded as the 2 mm gap formation force. The tendon was pulled until complete pullout or rupture of all stitches. During the loading process, the load and displacements of the tendons were measured with Instron Series IX software and were stored in the connecting computer.
Load was then plotted against displacement to obtain a load–displacement curve. The maximal load of the sutures was the peak force shown on the load–displacement curve. Stiffness of the repaired tendon was calculated as the slope of the linear portion of the load–displacement curve, and the energy to failure was the area under the load–displacement curve up to the point of peak load (Fig 2).
Data analysis
The gap formation force, ultimate strength, stiffness, and energy to failure were analysed statistically by one-way analysis of variance (ANOVA). Power analysis was done to ensure a suitable sample size of this study. When a significant difference was found between the three groups, Student’s t tests were used to determine the statistical difference between pairs of data. Values are reported as mean ± SD.
RESULTS
Gap formation force
The results of the tests of the three methods are shown in Table 1. The 2 mm gap formation force of the embedded cross-stitch method was 49.2 ± 5.6 N, which was statistically greater than that of the cross-stitch and modified Halsted methods (P <0.001). The 2 mm gap formation force of the modified Halsted method was also statistically greater than that of the cross-stitch method (P <0.05).
Ultimate strength
The ultimate strength of the embedded cross-stitch method was 68.3 ± 6.3 N, which was significantly greater than that of the cross-stitch and modified Halsted method (P <0.001). The ultimate strength of the modified Halsted method was not statistically different from that of the cross-stitch method.
Stiffness
The tendons repaired with the embedded cross-stitch method displayed significantly greater stiffness than those with the cross-stitch method (P <0.001). The stiffness of the tendons repaired with the modified Halsted method was not statistically different from that of those repaired with the cross-stitch method, but was significantly less than that of the tendons with embedded cross-stitch method (P <0.001).
Energy to failure
The tendons repaired with the embedded cross-stitch method required significantly greater energy to produce repair failure than those repaired with the cross-stitch method (P <0.001) or the modified Halsted (P <0.01). Energy to failure was not statistically different between the tendons with cross-stitch sutures and those with the modified Halsted sutures.
DISCUSSION
The development of this new suture method was inspired by our recent study in which embedment of the majority of peripheral suture strands within the tendon greatly decreased exposure of suture on the tendon surface and increased the strength of repairs (Wang and Tang, 2002). Our cross-stitch suture embedded much of the suture beneath the epitenon and thus reduced suture exposure on the tendon surface. It also significantly increased the gap formation force, ultimate strength, and energy to failure of the tendon repairs though used an identical number of strands across the repair sites. Stiffness reflects a tendon’s capacity to withstand elongation under tensile load (Moneim et al., 2002; Tang et al., 2001b; Wang and Tang, 2002; Zobitz et al., 2001) and the embedded cross-stitch produced greater stiffness than the current cross-stitch suture. Embedment of cross-stitch strands within the tendon also increased the area of tendon–suture interface, which may in turn resist tendon deformities.
Recent studies have showed that cross-stitch sutures exposed on the tendon surface are associated with significantly increased friction and work of digital flexion (Angeles et al., 2002; Momose et al., 2001), probably due to the excessive exposure of the sutures. Although the strength of cross-stitch repairs is widely recognized (Elliot 2002; Kim et al., 1996; Kubota et al., 1996; Silfverskiöld and May, 1994; Tang et al., 2001c; Tran et al., 2002), its clinical application has been very limited by its extensive suture exposure. Bulkiness of tendon stumps and the exposed tendon sutures with the current cross-stitch method may present another problem. Since the suture–tendon junction of the cross-stitch method is actually far away from the tendon laceration level, the repair does little to smooth or immobilize the repair site. In contrast, the embedded cross-stitch method decreases the distance between the tendon–suture junction and the laceration and may result in a smoother repair. During the tendon repair, we placed the cross-stitch suture at a depth of about 1 mm, which probably reduces the bulkiness of the repair; deeper placement of the cross-stitch may increase the bulkiness of the tendon.
The embedded cross-stitch and modified Halsted methods differ in their configuration of suture strands beneath the epitenon of the repaired tendon. The length of suture strand exposure on the tendon surface is almost identical with both methods having minimal suture exposure. The increase in strength of the embedded cross-stitch over that of the modified Halsted resulted from the existence of locking components, rather than the grasping components of the modified Halsted method. The studies of Hotokezaka and Manske (1997) showed that locking sutures provide greater strength than grasping sutures. We believe that the strength of the modified Halsted method is sufficient for a peripheral suture, but the embedded cross-stitch method further increases the strength of the repairs.
Our embedded cross-stitch method has some of the merits of a simple peripheral suture and some of those of cross-stitch locking sutures. Multiple core sutures with a simple running peripheral suture are considered the primary option for digital flexor tendon repairs (Gill et al., 1999; Savage and Risitano, 1989; Tang et al., 1994; Thurman et al., 1998; Veitch et al., 2000; Wang et al., 2003; Xie et al., 2002). Nevertheless, the embedded cross-stitch with a simple core suture provides an alternative for tendons that are too small to accommodate multiple core sutures or require a combination of core sutures with stronger peripheral sutures to provide strength.