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
Clinical Results of Hamstring Autografts in Anterior Cruciate Ligament Reconstruction: A Comparison of Femoral Knot/Press-Fit Fixation and Interference Screw Fixation
To compare the clinical outcomes of femoral knot/press-fit anterior cruciate ligament (ACL) reconstruction with conventional techniques using femoral interference screws.
Methods
Among patients who underwent arthroscopic ACL reconstruction with hamstring autografts, 73 were treated with either a femoral knot/press-fit technique (40 patients, group A) or femoral interference screw fixation (33 patients, group B). The clinical results of the 2 groups were retrospectively compared. The inclusion criteria were primary ACL reconstruction in active patients. The exclusion criteria were fractures, multiligamentous injuries, patients undergoing revision, or patients with contralateral ACL-deficient knees. In the femoral knot/press-fit technique, semitendinosus and gracilis tendons were prepared as 2 loops with knots. After passage through a bottleneck femoral tunnel, the grafts were fixed with a press-fit method (grafts' knots were stuck in the bottleneck of the femoral tunnel). A tie with Mersilene tape (Ethicon, Somerville, NJ) over a bone bridge for each tendon loop and an additional bioabsorbable interference screw were used for tibial fixation.
Results
The mean follow-up period was 38 months (range, 24 to 61 months). A significant improvement in knee function and symptoms was reported in most patients, as shown by improved Tegner scores, Lysholm knee scores, and International Knee Documentation Committee assessments (P < .01). The results of instrumented laxity testing, thigh muscle assessment, and radiologic assessment were clearly improved when compared with the preoperative status (P < .01). No statistically significant difference in outcomes could be observed between group A and group B (P = not significant).
Conclusions
In this nonrandomized study, femoral knot/press-fit ACL reconstruction did not appear to provide increased anterior instability compared with that of conventional femoral interference screw ACL reconstruction. Favorable outcomes with regard to knee stability and patient satisfaction were achieved in most of our ACL-reconstructed patients using femoral knot/press-fit fixation with hamstring tendon autograft.
Level of Evidence
Level IV, therapeutic case series.
The many different fixation devices using hamstring tendon in the femoral tunnel in anterior cruciate ligament (ACL) reconstruction remain a subject of debate. Current techniques include staples, interference screws, buttons, and cross-pins. All fixation methods should be stiff and strong enough to tolerate physiological forces until complete graft incorporation has occurred.
A prospective randomized study of anterior cruciate ligament reconstruction: A comparison of patellar tendon and quadruple-strand semitendinosus/gracilis tendons fixed with bioabsorbable interference screws.
Comparative outcome analysis of malpositioned and properly positioned fixation groups after hamstring autograft ACL reconstruction with femoral cross-pin fixation.
Anterior cruciate reconstruction using semitendinosus and gracilis tendons, bone patellar tendon, or quadriceps tendon-graft with press-fit fixation without hardware. A new and innovative procedure.
In this technique, fixation is close to the native ACL attachment site, potentially resulting in enhanced healing due to optimal ample graft and bone contact in the femoral tunnel. Biomechanical study also showed similar results when comparing this method with methods using the EndoButton (Smith & Nephew Endoscopy, Andover, MA).
However, a frequently encountered problem with this technique is insufficient length and thickness of the hamstring graft.
This study presented an arthroscopic-assisted ACL reconstruction using the femoral knot/press-fit technique for hamstring autografts. The purpose of this retrospective study was to compare clinical outcomes of patients undergoing femoral knot/press-fit ACL reconstruction with those of patients undergoing conventional femoral interference screw ACL reconstruction. Our hypothesis was that the results of the femoral knot/press-fit technique would be superior to those of the conventional femoral interference screw technique.
Methods
We considered 126 patients who had undergone arthroscopic ACL reconstruction using autologous semitendinosus and gracilis tendons between 1999 and 2010 at our clinics for this study. The inclusion criteria were primary ACL reconstruction in active patients. We excluded 17 cases that had fractures, had multiligamentous injuries, underwent revision, or had contralateral ACL-deficient knees. We also excluded 31 cases that underwent ACL reconstruction with periosteum-enveloping graft. There were 78 cases in total enrolled in our study.
In group A (femoral knot/press-fit fixation, 42 patients), the semitendinosus and gracilis tendons were prepared as 2 loops with knots. After passage through a bottleneck femoral tunnel, the grafts were fixed with a press-fit method (graft knots were stuck in the bottleneck of the femoral tunnel). A bioabsorbable interference screw (Bioscrew; Linvatec, Largo, FL) and a Mersilene tape (Ethicon, Somerville, NJ) tie placed over a bone bridge for each tendon loop were used for tibial fixation (Fig 1).
Fig 1In group A the grafts were fixed with a press-fit method (graft knots were stuck in the bottleneck) in the femoral tunnel. A bioabsorbable interference screw in the tibial tunnel and a Mersilene tape tie over a bone bridge for each tendon loop were used for tibial fixation.
In group B (femoral interference screw fixation, 36 patients), the graft was fixed with a bioabsorbable interference screw (Bioscrew) for the femoral side. The same fixation method as that used in group A was used for the tibial side.
Assignment of patients to group A or group B was not achieved by standard randomization. Rather, the assignment was made according to the timeline. The group B technique was performed from August 1999 to February 2003; the group A technique was performed from March 2003 to May 2010. All operative procedures were performed by the same orthopaedic surgeon. Before each procedure, the advantages, disadvantages, and technical difficulties associated with these arthroscopic procedures were explained in detail to all patients. All patients agreed to undergo these procedures. Approval for this study was obtained from the institutional review board at our institution.
The diagnosis of ACL injury depends primarily on physical examination using a manual anterior drawer test (ADT) and findings of ACL tear or rupture on magnetic resonance imaging. The main indication for ACL reconstruction is symptomatic ACL rupture with a positive ADT of at least 2+ (>1 cm). Among our 78 patients with ACL rupture, symptoms included declining sports activity or work activity levels (all patients), instability (65 patients), and pain with daily walking (21 patients).
For patients with a 3+ ADT (>1.5 cm) or associated injuries, ACL reconstruction was suggested immediately after recovery of range of motion (ROM). For those with a 2+ ADT (>1 cm), 3 months of nonoperative treatment focusing mainly on strengthening was recommended; only in those with unrelieved symptoms was ACL reconstruction suggested.
Surgical Techniques
Technique for Harvesting Gracilis and Semitendinosus Tendons
A horizontal incision of approximately 3 cm was made from 1.5 cm medial to the tibial tubercle extending distally. The sartorius fascia, encountered after dissection, was carried down through the subcutaneous tissue. The underlying semitendinosus and gracilis tendons were palpated and individually identified. The tendons were harvested with a tendon stripper with the knee flexed and traction applied on the grasping suture.
Graft Preparation
In group A the ends of each tendon were tied together with a simple knot. The knots were sutured with 4 to 5 stitches using No. 2 Ethibond (Ethicon). Two 5-mm strips of Mersilene tape were passed around the knotted and looped end of each graft. These tendons were maximally tightened under cyclic manual loading.
To distinguish between the loops, the Mersilene tape for the semitendinosus tendon was marked but the tape for the gracilis tendon remained unmarked. To measure the diameter, the loops were then pulled, one after the other, through a measuring template in 1.0-mm site increments. The loop length after knotting should equal the bottleneck thickness (approximately 5 to 6 mm) plus the intra-articular graft length (approximately 30 to 35 mm) plus the length of the tibial tunnel (approximately 30 to 35 mm). Ideally, the loop length after knotting totaled approximately 70 to 75 mm for the semitendinosus tendon and approximately 75 to 80 mm for the gracilis tendon. The intra-articular portions of the graft (between the femoral and tibial openings) were then marked (Fig 2).
Fig 2(A) The hamstring tendon was harvested. (B, C) The ends of each tendon were tied together in a simple knot. (D) The knot was sutured with 4 to 5 stitches using No. 2 Ethibond. (E) The tendon was maximally tightened under cyclic manual loading to minimize stress relaxation. (F) Two strips of Mersilene tape were passed around the knotted and looped end of each graft. The intra-articular portions of the graft were marked.
In group B 2 No. 2 Ethibond sutures were affixed to the end of the tendon using a running baseball whipstitch technique. Two 5-mm strips of Mersilene tape were passed around the looped end of each graft.
Femoral Tunnel Preparation
Under arthroscopy, the ACL disrupted stump was first debrided and the tibial footprint was identified. With a 7-mm-offset femoral drill guide, a 2.2-mm Kirschner wire was passed through the anteromedial portal and located at the 10- to 11-o'clock position in a right knee (1- to 2-o'clock position in a left knee). The goal of this procedure was to reproduce the anteromedial component of the ACL. The knee was then flexed to 135°. The K-wire was advanced to the skin level.
In group A the K-wire was overdrilled with a cannulated reamer from the anteromedial portal, matching the diameter of the 2 tendon parts of the loops (commonly 7 to 9 mm). The cannulated reamer was reamed to a depth of 10 to 15 mm.
A 6-/11-mm stepped impactor (barrel diameter, 11 mm; offset diameter, 6 mm) (Richard Wolf, Knittlingen, Germany) was inserted through the anteromedial portal. The 6-mm offset was stuck into the tunnel and held steadily. Under the guide of the prepositioned K-wire, the femoral tunnel was then created with a cannulated reamer. A 1.5-cm longitudinal incision was made at the center of the K-wire. Starting outside, the tunnel was over-reamed and stopped at the stepped impactor tip. The reamer size was increased to match the diameter of the knotted portion of the graft (commonly 13 to 14 mm). If the stepped impactor was unavailable, a cannulated handle could be used instead (2.8-mm-diameter handle of Corkscrew Suture Anchor; Arthrex, Naples, FL). A bottleneck-type tunnel was created, and the tunnel integrity was then confirmed under arthroscopic visualization (Fig 3, Video 1 [available at www.arthroscopyjournal.org]).
Fig 3(A, B) Under arthroscopic visualization, the K-wire was overdrilled with a cannulated reamer from the anteromedial portal, matching the diameter of the 2 tendon parts of the loops (commonly 8 to 9 mm). A stepped impactor was inserted through the anteromedial portal. (C) The 6-mm offset was stuck into the tunnel and held steadily. Under the guide of the prepositioned K-wire, the femoral tunnel was then created by the cannulated reamer. Starting outside, the tunnel was over-reamed and stopped at the stepped impactor tip. (D) A bottleneck-type tunnel was created, and the tunnel integrity was then confirmed under arthroscopic visualization.
In group B the femoral tunnel was prepared by the traditional method and was reamed with a 9- to 11-mm reamer.
Tibial Tunnel Preparation
The tibial remnant of the ligament was preserved when possible and only partially removed to improve visualization. We placed the tibial tunnel in the posterior one-half and medial-lateral center of the native ACL tibial footprint. With the drill angle guide adjusted to 50°, the tibial guide pin was placed. The tibial tunnel was then reamed under direct visualization with an 8- or 9-mm cannulated reamer, depending on the diameter of the 2 tendon parts of the loops.
Graft Passage and Femoral-Side Fixation
The Mersilene tapes on the 2 grafts were secured to the prepositioned No. 5 Ethibond and drawn through the lateral aspect of the knee through the intercondylar notch and out the entrance of the tibial tunnel.
In group A the gracilis loop with the thin knot followed the semitendinosus loop. The 2 loops were pulled firmly. A sudden jerk indicated that the semitendinosus knot had reached the step in the tunnel (i.e., the bottleneck). To anchor the knots, maximum traction was applied to both loops.
In group B the graft was also drawn from the posterior aspect of the knee through the intercondylar notch and into the tibial tunnel. The graft was fixed with a 7 to 9 × 20–mm Bioscrew in the femoral tunnel.
Fixation of Tibial Site
The tendon bundle of the graft was fixed within the tibial tunnel (close to the joint line) with a 7 to 9 × 20–mm Bioscrew by an outside-in technique. A 4.5-mm drill hole was made 1 cm distal to the tibial end of the tunnel. A bone bridge was created by tunneling under the bone with a right-angle clamp. A No. 2 Ethibond suture was passed from the tibial tunnel opening by this right-angle clamp. The Mersilene tapes were separated and secured to the No. 2 Ethibond. The Mersilene tape strips were pulled under the bone bridge by this No. 2 Ethibond.
In all patients, maximum traction was exerted on the Mersilene tapes, and the semitendinosus tape ends were first tied with 4 knots. Stability was confirmed manually, and the tapes of the second loop (gracilis) were then tied, using the same procedure.
Knee stability was assessed and the graft was rechecked for tension under arthroscopy. Layered closure of each incision was performed. A sterile dressing was then applied, and the knee was placed in a knee brace and locked in full extension.
Postoperative Rehabilitation
The operative knee was protected by a knee brace in the fully extended position for the first week, and partial weight bearing was allowed, as tolerated. Quadriceps isometric exercises, a straight leg–raising exercise, and a passive ROM exercise were begun as early as possible. The brace was adjusted to a flexion angle of 30° after 1 week and increased by 30° every week, unlocked at 6 weeks, and discarded after 8 weeks. The ROM should quickly recover to complete flexion and extension.
Aggressive quadriceps and hamstring muscle strengthening exercises were initiated. Patients usually resumed normal daily activities at approximately 3 months after surgery and typically returned to sports activity after 6 months. Full recovery to the preinjury sports level can generally be achieved between 9 and 12 months after surgery.
Follow-Up Assessment
Outcome assessments were standardized and conducted by research assistants to effectively audit the results of the procedures. Clinical evaluations of knee function and stability were performed preoperatively and at last follow-up after more than 2 years postoperatively. All patients were followed up using the same strategy: During the first 2 years after surgery, we asked patients to return to our clinic regularly. After 2 years, we asked patients to return to our clinic, which depended on the patients' willingness. If the patient was not willing to return to our clinic, we performed our final outcome analyses. (We asked the patient, “Would you like to come back to our clinic next year?” If patent answered no, we performed our final outcome analyses.) After 5 years, we did not ask the patients to come back to our clinic except when they had a problem. All assessments were rated according to the guidelines of the International Knee Documentation Committee (IKDC).
The Lysholm knee score was used to evaluate subjective symptoms, including the presence of a limp, use of a support, evidence of joint locking, knee instability, joint pain, swelling, and impaired squatting and stair-climbing ability both preoperatively and postoperatively.
Manual knee laxity tests (Lachman test, ADT, and pivot-shift test) were performed and assessed for both legs. For the measurement of anterior tibiofemoral translation, the ADT and pivot-shift test were performed on the ACL-reconstructed knee and contralateral knee. In addition, the KT-1000 arthrometer (MEDmetric, San Diego, CA) was used to assess for anterior translation. The KT-1000 tests were performed in 30° of flexion with a standard anterior force of 30 lb (134 N) to measure the total anteroposterior translation.
Bilateral, double-leg posteroanterior weight-bearing radiographs at 35° to 45° of flexion were used to evaluate narrowing at the medial and lateral joint spaces. Degenerative changes were graded according to the Ahlbäck radiologic classification of arthritis.
The Merchant view at 45° was used to document patellofemoral narrowing. The femoral and tibial bone tunnels were measured on the AP and lateral radiographs at 3 months, 1 year, 2 years, and last follow-up. To evaluate for bone tunnel enlargement, the sclerotic margins of the femoral and tibial tunnels were measured at the widest dimension of the tunnel.
Thigh atrophy was defined as the difference in thigh circumference between the operative and nonoperative knees at a point 10 cm proximal to the superior pole of the patella. The Cybex 340 dynamometer (Cybex, New York, NY) was used to measure flexor and extensor muscle strength. Peak extension torque and side-to-side ratio were measured.
Statistical Analysis
The Mann-Whitney U test was performed to compare the results between groups. The Wilcoxon signed rank test was used to compare preoperative and postoperative results of each group. The level of statistical significance was set as P < .05.
Results
A total of 78 patients (38 men and 40 women; mean age, 32.8 years [range, 18 to 45 years]) underwent arthroscopic ACL reconstruction using hamstring autografts. Seventy-three patients had complete follow-up documentation for a minimum of 2 years and were included in the final outcome analyses. The mean follow-up period was 38 months (range, 24 to 61 months). Five patients could not complete regular follow-up evaluation because of a move abroad (n = 2), a move to another city (n = 1), or other trauma that made assessment difficult (n = 2). Other trauma included a contralateral tibial shaft fracture (due to another motorcycle accident 7 months after ACL surgery) and an ipsilateral lateral malleolar fracture (after another sports injury 19 months after ACL surgery).
The mechanisms of injury included 60 sports injuries, 12 motor vehicle accidents, and 1 fall on the knee. The mean interval between the injury and the operation was 5.4 months (range, 2 to 30 months). Sex, age, isolated ACL rupture, combined meniscal tears, and follow-up period showed no significant differences between the 2 groups; a significant difference was found for mean operative time (Table 1).
Articular cartilage damage was identified in 6 patients. For femoral chondral defects measuring less than 2 cm, microfracture was performed. For patients with partial-thickness chondral lesions, abrasion chondroplasty was performed. Four patients underwent microfracture procedures, and 2 underwent abrasion chondroplasty to smooth irregular surfaces.
Stability
Preoperatively, there was no significant difference in the number of positive cases on the Lachman test between the groups. Postoperatively, the Lachman test was negative in 37 cases (92.5%) in group A and 30 cases (90.9%) in group B. There were no significant differences in the pivot-shift test between the groups preoperatively and postoperatively (Table 2).
The anterior stability of the knee was measured with the KT-1000 arthrometer at a force of 134 N. The difference between the injured knee and the contralateral knee was recorded. There were no significant differences in anterior knee stability between groups A and B, either preoperatively or at the last follow-up (Table 3).
Table 3KT-1000 Arthrometer Measurements Preoperatively and at Last Follow-Up
Preinjury, preoperative, and postoperative activity levels were graded according to the Tegner scoring system (Table 4). In group A the mean preinjury score was 5.9 (range, 3 to 9), decreasing to 2.8 preoperatively (range, 2 to 5) and increasing to 6.2 postoperatively. In group B the mean preinjury score was 5.6 (range, 3 to 9), decreasing to 2.9 preoperatively (range, 2 to 5) and increasing to 6.1 postoperatively.
Table 4Clinical Scores Preoperatively and at Last Follow-Up
The Lysholm scores at last follow-up showed that there were improvements in daily living in both groups. Preoperatively, the Lysholm score was 63.1 in group A and 64.4 in group B, and the difference was not significant. Postoperatively, the Lysholm score was 92.3 in group A and 91.5 in group B, and the difference was not significant.
IKDC Scores
Knee Function by Patient Subjective Assessment
In group A, 14 patients (35.0%) had an abnormal status and 21 (52.5%) had a severely abnormal status preoperatively based on self-assessment. In group B, 12 patients (36.4%) had an abnormal status and 18 (54.5%) had a severely abnormal status preoperatively based on self-assessment.
At the last follow-up assessment after reconstruction, 33 group A patients (82.5%) had a normal status and 1 (2.5%) had an abnormal status and 26 group B patients (78.8%) had a normal status and 2 (6.1%) had an abnormal status based on self-assessment.
Symptoms
At last follow-up, 2 patients (5.0%) in group A and 1 patient (3.0%) in group B reported pain during moderate or strenuous activities; in addition, 1 patient (3.0%) in group B reported swelling during moderate or strenuous activity. Moreover, 1 patient (2.5%) in group A displayed symptoms of partial giving way during moderate or strenuous activities, and 1 patient (3.0%) in group B reported occasional full giving way during moderate or strenuous activities.
Range of Motion
ROM was the same for both groups at last follow-up. Normal knee ROM was found in 39 patients (97.5%) in group A and 32 (97.0%) in group B. Each group had 1 patient with nearly normal ROM.
Ligament Laxity
The KT-1000 arthrometer measurements were also graded according to the IKDC form. The results between the 2 groups were similar (no significant difference was noted) (Table 5).
Table 5Evaluation of Knee by KT-1000 Arthrometer Measurements According to IKDC Form Preoperatively and at Last Follow-Up
At last follow-up, 37 patients (92.5%) in group A and 30 patients (90.9%) in group B returned to the same preinjury daily activities and sports activities.
Patellofemoral Crepitus
Patellofemoral crepitus with mild pain was noted in 1 patient (2.5%) in group A and 1 patient (3.3%) in group B.
Donor-Site Morbidity
Two patients (5%) in group A and 1 patient (3.3%) in group B sustained some numbness at the donor site.
Radiographic Findings
Grade I osteoarthritis according to the Ahlbäck classification was detected in 7 patients (17.5%) in group A and 5 patients (15%) in group B (Fig 4).
Fig 4Anteroposterior and lateral radiographs show the femoral and tibial tunnels.
Twenty-nine patients (72.5%) in group A and 25 patients (75.7%) in group B achieved a distance, using the injured leg, that was 90% or more of that achieved with the normal leg. Eight patients (20.0%) in group A and 6 patients (18.1%) in group B achieved a distance of 76% to 89% with the injured leg compared with the normal leg. Both groups had 1 patient who could only achieve less than 50% of the distance as compared with the normal leg.
Overall Rating
Our results indicate that there was no significant difference in the IKDC subjective knee evaluation form score preoperatively and postoperatively (P > .05) (Table 4).
Thigh Muscle Atrophy and Muscle Strength
There were no significant differences in thigh girth difference at last follow-up between groups A and B. The thigh girth difference averaged 0.9 cm in group A and 0.8 cm in group B (the difference was not significant). Cybex dynamometer studies showed no significant differences in extensor and flexor strength ratios at last follow-up between the 2 groups (Table 6).
Table 6Thigh Muscle Atrophy and Isokinetic Strength
For the femoral tunnel, the difference in the diameters at last follow-up minus the diameters at 1 week postoperatively was 0.36 ± 0.04 mm in group A and 1.10 ± 0.17 mm in group B (P = .021). Femoral tunnel expansion of more than 3 mm was identified in 1 patient (2.5%) in group A and 6 patients (18.4%) in group B.
For the tibial tunnel, the difference in the diameters at last follow-up minus the diameters at 1 week postoperatively was 0.66 ± 0.31 mm in group A and 0.78 ± 0.42 mm in group B (P > .05).
Complications and Problems With Femoral Knot/Press-Fit Method
Four patients in group A and 2 patients in group B complained about skin irritation around the Mersilene tape tie at the tibial incision. Only 1 patient asked for removal of the Mersilene tape tie; we removed the Mersilene tape tie with the patient under local anesthesia at 3 years after reconstruction.
The patient with a poor result in group B was a 26-year-old woman who had frequent knee instability and knee pain during daily activity and had a side-to-side difference of 13 mm. This patient sustained graft rupture according to magnetic resonance imaging examination.
In group A the hamstring tendon graft was noted to be too thin in 3 patients. The graft was too thin to create the ample knot to be stuck in the bottleneck femoral tunnel. We used gracilis tendon to tie a supplemented knot on the knot of the semitendinosus tendon, suturing with 4 stitches using No. 2 Ethibond to form 1 ample knot (Fig 5).
Fig 5When the hamstring tendon was too thin to create an ample knot to be stuck in the bottleneck femoral tunnel, we used gracilis tendon to tie a supplemented knot on the knot of the semitendinosus tendon, suturing with 4 stitches using No. 2 Ethibond to form 1 ample knot (1 knot and 2 loops).
A posterior cortical breach of the bottleneck tunnel was noted under arthroscopic visualization in 2 patients in group A. We still used the knot/press-fit method in these 2 patients, and it did not affect clinical outcomes.
Discussion
Outcomes from various femoral fixation methods for ACL reconstruction using hamstring tendons may be affected by factors, such as uncertain tendon-bone healing, undetermined windshield-wiper effect between the tendon and tunnel, and use of different fixation devices. It takes approximately 12 weeks for the soft tissue–like hamstring tendon to heal to the osseous tunnel.
Soft tissue graft interference fit fixation: Observations on graft insertion site healing and tunnel remodeling 2 years after ACL reconstruction in sheep.
Tendon healing in a bone tunnel. Part II: Histologic analysis after biodegradable interference fit fixation in a model of anterior cruciate ligament reconstruction in sheep.
Thus a secure fixation technique is needed to withstand the forces on the graft resulting from current rehabilitation protocols that allow for unrestricted ROM, weight bearing, and early return to athletic activity after ACL reconstruction.
Several biomechanical studies have compared the methods of soft-tissue fixation in the femoral tunnel, with varying results.
Tunnel widening following anterior cruciate ligament reconstruction using hamstring autograft: A comparison between double cross-pin and suspensory graft fixation.
They found no clear biomechanical advantage of any current method of fixation over another. Suspensory fixation or cross-pin methods have disadvantages such as (1) increased early motion of the graft within the tunnel, which does not promote stable healing; (2) more synovial fluid in the bone tunnel, increasing possible negative effects of cytokines; and (3) the potential occurrence of the “bungee effect,” whereby the graft moves longitudinally within the tunnel because of fixation at a point away from the tunnel entrance. The disadvantages of interference screws include possible graft irritation caused by introducing the screw, a reduced bone-tendon interface, and reported slippage of the graft causing clinical failure. A lack of replacement of bioabsorbable interference screws by bony tissue even after an extended period has been reported, as has inflammation of the synovium elicited by a foreign-body reaction.
Tunnel widening after hamstring anterior cruciate ligament reconstruction is influenced by the type of graft fixation used: A prospective randomized study.
Bioabsorbable interference screw versus bioabsorbable cross pins: Influence of femoral graft fixation on the clinical outcome after ACL reconstruction.
Cross-pin femoral fixation versus metal interference screw fixation in anterior cruciate ligament reconstruction with hamstring tendons: Results of a controlled prospective randomized study with 2-year follow-up.
to compare the knee kinematics of the intact knee, ACL-deficient knee, EndoButton-reconstructed knee, and knot/press-fit–reconstructed knee. The load at failure of the knot/press-fit complex was 540 ± 97.7 N, and the stiffness was 37.8 ± 9.6 N/mm. Woo and colleagues concluded that the knot/press-fit technique was a reliable alternative for femoral fixation of hamstring tendon grafts in ACL reconstruction.
The difference in femoral tunnel enlargement in group A was significantly smaller than that in group B in this study. A possible reason might be that the femoral fixation of the hamstring graft was close to the insertion site of the native ACL, which was close to the joint. In addition, the femoral tunnel opening was also filled with tendon knots; this construction potentially prevented some synovial fluid from entering the bone tunnels, creating a theoretically better environment for tendon-bone healing. Paessler and Mastrokalos
Anterior cruciate reconstruction using semitendinosus and gracilis tendons, bone patellar tendon, or quadriceps tendon-graft with press-fit fixation without hardware. A new and innovative procedure.
However, the femoral knot/press-fit technique for ACL reconstruction has not become popular because of potential problems. A frequently encountered problem is the short length of the hamstring graft. Another problem is that the hamstring graft is too thin to make an ample knot to be stuck in the bottleneck tunnel. In addition, in the original method of Paessler and Mastrokalos,
Anterior cruciate reconstruction using semitendinosus and gracilis tendons, bone patellar tendon, or quadriceps tendon-graft with press-fit fixation without hardware. A new and innovative procedure.
a complete set of specially designed instruments was also required to create a bottleneck femoral tunnel. We modified their method for creating the bottleneck femoral tunnel in a previous clinical study about posterior cruciate ligament reconstruction.
In this modified technique for posterior cruciate ligament reconstruction, a femoral bottleneck tunnel can be created using traditional instruments or only 1 stepped impactor. We also used the modified technique in this ACL-reconstructed series.
In the original method of Paessler and Mastrokalos,
Anterior cruciate reconstruction using semitendinosus and gracilis tendons, bone patellar tendon, or quadriceps tendon-graft with press-fit fixation without hardware. A new and innovative procedure.
no hardware was needed for tibial fixation. They used single tibial fixation without interference screws (only a Mersilene tape tie over a bone bridge). Although they reported no widening of the femoral tunnels, widening of the tibial tunnels was noted in their series clinically.
Anterior cruciate reconstruction using semitendinosus and gracilis tendons, bone patellar tendon, or quadriceps tendon-graft with press-fit fixation without hardware. A new and innovative procedure.
failure occurred at the tibial insertion rather than at the knot/press-fit fixation site in 4 of 8 specimens (50%). For these reasons, we used dual tibial fixations with an additional interference screw at the tibial tunnel opening. Theoretically, this dual fixation could eliminate an unwanted windshield-wiper effect and minimize the effect of uncertain tibial bone quality and shortened length of the hamstring tendon.
Limitations
Our study had several limitations. First, the described method could not be applied to double-bundle reconstruction until now because of possible tunnel collapse if 2 bottleneck femoral tunnels are created. In addition, there was no objective proof to support claims that relative waterproof construction and favorable tendon-to-bone healing occur with knot/press-fit fixation. Further in vivo studies should be conducted to elucidate this point. Furthermore, a short length of the hamstring tendon was noted in the technique. Though not confirmed, the graft length does not appear to affect outcomes if good initial fixation, such as dual tibial fixation, and improved tendon-to-bone healing are achieved. Future biomechanical and histologic investigations should be conducted regarding this issue. In addition, this was a retrospective nonrandomized clinical study with a small case number, and selection bias is a possibility.
Conclusions
In this nonrandomized study, femoral knot/press-fit ACL reconstruction did not appear to provide increased anterior instability compared with that of conventional femoral interference screw ACL reconstruction. Favorable outcomes with regard to knee stability and patient satisfaction were achieved in most of our ACL-reconstructed patients using femoral knot/press-fit fixation with hamstring tendon autograft.
A prospective randomized study of anterior cruciate ligament reconstruction: A comparison of patellar tendon and quadruple-strand semitendinosus/gracilis tendons fixed with bioabsorbable interference screws.
Comparative outcome analysis of malpositioned and properly positioned fixation groups after hamstring autograft ACL reconstruction with femoral cross-pin fixation.
Anterior cruciate reconstruction using semitendinosus and gracilis tendons, bone patellar tendon, or quadriceps tendon-graft with press-fit fixation without hardware. A new and innovative procedure.
Soft tissue graft interference fit fixation: Observations on graft insertion site healing and tunnel remodeling 2 years after ACL reconstruction in sheep.
Tendon healing in a bone tunnel. Part II: Histologic analysis after biodegradable interference fit fixation in a model of anterior cruciate ligament reconstruction in sheep.
Tunnel widening following anterior cruciate ligament reconstruction using hamstring autograft: A comparison between double cross-pin and suspensory graft fixation.
Tunnel widening after hamstring anterior cruciate ligament reconstruction is influenced by the type of graft fixation used: A prospective randomized study.
Bioabsorbable interference screw versus bioabsorbable cross pins: Influence of femoral graft fixation on the clinical outcome after ACL reconstruction.
Cross-pin femoral fixation versus metal interference screw fixation in anterior cruciate ligament reconstruction with hamstring tendons: Results of a controlled prospective randomized study with 2-year follow-up.
00202-3&id=gr1.jpg){kind=link}
00202-3&id=gr2.jpg){kind=link}
00202-3&id=gr3.jpg){kind=link}
00202-3&id=gr4.jpg){kind=link}
00202-3&id=gr5.jpg){kind=link}