- Email: [email protected]
Anterior cruciate ligament reconstruction using four-strand semitendinosus and gracilis tendon grafts and metal interference screw fixation
Anterior Cruciate Ligament Reconstruction Using Four-Strand Semitendinosus and Gracilis Tendon Grafts and Metal Interference Screw Fixation Philippe Colombet, M.D., Ph.D., Michel Allard, M.D., Vincent Bousquet, M.D., Christophe de Lavigne, M.D., Pierre-Henri Flurin, M.D., and Christophe Lachaud, P.T.
Purpose: The purpose of this study was to determine the outcome of 200 anterior cruciate ligament (ACL) reconstructions using hamstring tendons. Type of Study: This is a case series reporting on 200 endoscopic procedures for reconstruction of the ACL. Methods: This study included patients over 18 years old with a healthy controlateral knee, intact posterior cruciate ligament, and without any peripheral surgical procedure or cartilage injury. Patients having undergone prior ligament reconstruction were excluded from the study. The minimum follow-up was 1 year. Clinical review allowed for documentation of International Knee Documentation Committee (IKDC), KT-1000 arthrometer laxity measurement, and isokinetic dynamometric analysis. Results: Quadriceps and hamstring muscle strength loss was less than 17%. At review, overall IKDC evaluation found that 50% of patients graded A, 44% graded B, and 6% were C or D. The differential anterior laxity was graded A for 157 patients with a median of 1 mm. Of the 113 high-performance athletes, 98 (86%) had resumed a preinjury level of sporting activity. Conclusions: Endoscopic reconstruction using 4-strand hamstring autograft may be considered safe, reliable, and reproducible. Preliminary outcome is fulfilling and this technique corresponds completely to therapeutic fields regarding ACL reconstructions. Key Words: Anterior cruciate ligament reconstruction—Hamstring autograft—Interference screw fixation—Endoscopic method.
T
he surgical treatment of knee instability secondary to anterior cruciate ligament (ACL) deficiency has evolved during the 20th century. ACL reconstruction was first performed in 1917 by Hey Groves1,2 using a strip of fascia lata. A variety of different donor grafts have since been used. Campbell3 reported the use of the medial third of the patellar tendon and Macey4 used the semitendinosus tendon with a distally attached tibial bone plug. These techniques have evolved over time with improved harvest-
From the Centre de Chirurgie Orthope´dique et Sportive, Bordeaux-Me´rignac, Me´rignac, France. Address correspondence and reprint requests to Philippe Colombet, M.D., Ph.D., Centre de Chirurgie Orthope´dique et Sportive, Bordeaux-Me´rignac, 9 Rue Jean Moulin, 33700 Me´rignac, France. E-mail: [email protected] © 2002 by the Arthroscopy Association of North America 0749-8063/02/1803-2795$35.00/0 doi:10.1053/jars.2002.30637
232
ing techniques, graft preparation, more thorough understanding of joint biomechanics and the role of isometricity in graft function, and the emergence of endoscopic surgery. The bone–patellar tendon– bone graft harvested from the middle third of the patellar tendon may still be regarded as the gold standard. Satisfactory results have been extensively recorded by Deehan et al.5 However, although excellent stability was achieved, a subgroup of patients experienced persistent donor-site morbidity and delay in return of full range of movement, with a small percentage of patients experiencing a slow return to work. Therefore, in 1996, this unit began to use a 4-strand hamstring autograft. It was felt that this would minimize quadriceps mechanism morbidity, thereby decreasing operative trauma with an expectation of reduced secondary complications, while still providing excellent knee stability. This report outlines our early experience to date in the first 200 patients.
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 18, No 3 (March), 2002: pp 232–237
FOUR-STRAND STG GRAFTS AND SCREW FIXATION PATIENTS AND METHODS This prospective study involved a series of 200 consecutive patients (1 center, 4 surgeons) who underwent surgery between July 1997 and September 1998. All patients were evaluated preoperatively, and at 6 months and 1 year postoperatively, using the CLAS system (Competitor Leisure Active Sedentary) for allocation of patients into 1 of 4 categories according to highest level of sporting activity. These would be equivalent to the 4 subgroups in the IKDC rating system and a full IKDC evaluation form. Laxity was measured using a KT-1000 Arthrometer (MedMetric, San Diego, CA) at 30 lb. Inclusion criteria included patients over 18 years of age with a healthy contralateral knee who had not had any prior peripheral surgical procedure, an intact posterior cruciate ligament, and no cartilage injury identified at initial arthroscopy. Patients having undergone prior ligament reconstruction were excluded from the study. Among 566 patients operated on between July 1997 and September 1998, the first 200 patients have been evaluated at a minimum 1-year follow-up. The mean follow-up period was 15 months (range, 12 to 25 months). The mean age of the patients was 26 years (range, 18 to 52 years). There were 67 women and 133 men. No bilateral procedures were performed. The right knee was involved in 101 cases and the left knee in 99 cases. The patient population consisted of 113 high-performance athletes (patients involved in national and international sporting events), 71 athletes involved in recreational sports, 9 active patients, and 7 sedentary patients. Preoperative alignment was documented as varus in 115 knees, neutral in 63, and valgus in 22 knees. Forty-two patients underwent surgical reconstruction within 3 months of injury with the remaining 158 patients having chronic instability. Twenty-four patients had a history of meniscectomy and a further 20 patients had undergone arthroscopic examination without formal resection. Surgical Technique A longitudinal 4-cm long proximal tibial incision was made, and both tendons (semitendinosus and gracilis) were harvested using a tendon stripper (Linvatec, Largo, FL). All remaining muscular debris was cleaned away from the graft. The 4 tendon strands were sewn to each other. Two arthroscopic portals (anteromedial and anterolateral patellar portals) were made, providing access to the joint. The intercondylar notch was cleaned with a 5.5-mm full-radius shaver (Dyonics, Smith & Nephew, Andover, MA). Using a
233
curette, the cortical median portion of the lateral femoral condyle was cleaned of all ligamentous debris, which provided full visualization of its posterior cortical margin. A bone spike was used to create a pilot hole located within 5 mm of the posterior border of the femoral condyle. The knee was flexed to 130° and a 3-cm femoral socket created in an inside-out direction to accommodate the previously harvested graft. A pull-through suture was placed into the eye of a Beath pin, passed through the femoral socket, and out through the lateral thigh. The knee was flexed to 90° and a guide pin was inserted with a guidewire that ensured accurate intra-articular positioning of the tibial tunnel (Acufex, Smith & Nephew, Andover, MA). The bone tunnel was fashioned over this guidewire, using the original graft harvesting incision and a reamer of the same diameter as that of the graft. The pull-through suture was grasped proximally and the graft pulled in a retrograde manner through the tibial tunnel and docked into the femoral socket. The knee was flexed again to 130°, and a 7- ⫻ 25-mm softthreaded, perforated, round-headed interference screw (RCI; Smith & Nephew) passed over a guide pin, via the portal using the anteromedial previously used for drilling. This ensured correct alignment of the graft and the interference screw. The knee was brought back to 90° of flexion, tension on the graft was maintained throughout by pulling distally, and a second interference screw of equivalent diameter was passed into the tibial tunnel close to the proximal portion of the tibial tunnel. Rehabilitation Full weight bearing and rehabilitation were commenced immediately. Patients were discharged after 2 days. Patients were directed toward achieving full extension using closed-chain exercises. All patients were requested to wear a dynamic range-of-movement knee orthosis (Largarigue, Bordeaux, France) for 6 weeks. This allowed for 0° to 60° of flexion for the first 2 weeks and then 0° to 90° for the remaining 4 weeks. Cybex tests were performed 5 months after surgery in professional athletes for muscle assessment. This test compared the affected knee with the healthy side and consisted of a series of managed cycles of arc of motions at differing speeds. The relative movement of the tibia on the femur at these speeds was determined with a measurement being taken as the mean of 5 repetitions.
234
P. COLOMBET ET AL.
Review Patients were assessed at 6 months and at 1 year after surgical reconstruction by independent observers. At these time points, subjective symptoms were recorded and clinical examination performed to allow for completion of the IKDC score. Each patient was examined using the KT-1000 (at 30 lb displacement force) to determine side-to-side knee laxity. In addition, we completed laxity assessments for 125 of our patients using the Telos system (Telos, Hungen, Germany). These recordings were taken at 6 and 12 months after surgery. Side-to-side quadriceps circumference was taken as a reading 10 cm from the superior pole of the patella. Statistical Analysis For data analysis, the unpaired Student t test was used to compare continuous data. The correlation test was used to compare ordinal data from the 2 treatment groups; P ⬍ .05 was regarded as significant. RESULTS Operative Intervention All procedures were undertaken using the same technique in a single center under the care of 4 surgeons. The patients participated in a standard rehabilitation program. Forty medial meniscectomies and 33 lateral meniscectomies were also carried out at the time of surgical reconstruction. A further 26 meniscal tears were confirmed as stable and left intact. Graft diameters were as follows: in 20 patients, 6.5 mm; in 35 patients, 7 mm; in 59 patients, 7.5 mm; in 46 patients, 8 mm; and in 40 patients, 8.5 mm. The median time lost from work was 75 days (range, 9 to 307 days). Median time to return to sports was 236 days (range, 57 to 354 days). Of the 113 competitive athletes, 98 (86.73%) returned to their previous level, 12 patients returned to “leisure” activity (class L), and 3 did not resume sports (Fig 1). Seventy-one patients were class L before surgery and we had 73 patients in class L at the time of evaluation (10 stopped sports activities so it was 71 at first plus 12 from the competitor class minus the 10 who became only active, thus 73). We had 22 active patients at 1 year and 7 sedentary patients the same as preoperatively. The quadriceps was judged normal in 130 cases, mild amyotrophy (1 to 3 cm side-to-side difference) was found in 66 cases, and severe amyotrophy (side-
FIGURE 1.
Activity change.
to-side difference ⬎3 cm) in 4 cases. There were 191 patients who did not report symptoms pertaining to the extensor mechanism. However, 4 patients had patellar tendinitis, 1 had quadriceps tendinitis, and 4 had a fasciitis of the fascia lata. Seven patients experienced slight discomfort at the donor site. This morbidity did not compromise participation in the postoperative rehabilitation program. At final review, 149 patients were very satisfied and 45 were satisfied with their operation. However, 6 patients remained dissatisfied with the outcome of surgery because of their failure to attain the desired level of function. IKDC Evaluation One hundred fifty-nine patients returned to their preinjury level of sporting activity. According to the patients’ subjective assessment, 152 had the feeling that their knee was normal (grade A), 42 that it was almost normal (grade B), and 6 that it was abnormal (grade C). Twenty-seven patients (13.5%) continued to experience pain with moderate activity, and 3 (1.5%) with activities of daily living. Eleven patients (5.5%) had effusion with exercise. Six patients did not feel safe with moderate activity, and 2 felt unstable with mild activity. In the symptom parameter group, 150 patients were graded A, 44 B, 3 C, and 3 D. In 194 cases, there was a difference of less than 5° in sideto-side loss of extension. In 181 cases, flexion was identical (⫾ 5°) on both sides, in 18 there was a loss of flexion between 6° and 15°, and in 1 there was a
FOUR-STRAND STG GRAFTS AND SCREW FIXATION
235
TABLE 1. Postoperative IKDC Grade
Patient subjective assessment Symptom group Range-of-motion group Ligament examination group Final evaluation
A
%
B
%
152 150 179 143 100
76 75 89.5 71.5 50
42 44 19 50 88
21 22 9.5 25 44
loss of 20°. In the range of motion parameter group, 179 patients were graded A, 19 B, 1 C, and 1 D. As evaluated with the KT-1000 at 30 lb, 157 patients were graded A with a median of 1 mm, 38 B with a median of 3 mm, 4 C with a median of 6 mm, and 1 D with a median of 12 mm. Pivot-shift testing found that 181 patients were grade 0, 15 patients were grade 1, 3 patients were grade 2, and 1 patient was grade 3. In the laxity parameter group, instrumented laxity testing found that 143 patients were grade A, 50 were grade B, 5 were grade C, and 2 were grade D, according to IKDC criteria. One-Leg Hop Test The one-leg hop test result was over 90% (grade A) in 167 patients, between 75% and 90% (grade B) in 30, between 50% and 75% (grade C) in 2, and under 50% (grade D) in 1 patient. Results of the one-leg hop test are much better at 1 year than they were at 6 months. There were 46.5% grade A, 37.7% grade B, 11.9% grade C, and 3.77% grade D at 6 months versus 83.5% grade A, 15% grade B, 1% grade C, and 0.5% grade D at 1 year (Table 1). Return to Sports Activities Of the 113 high-performance athletes, 12 resumed recreational sports activities, and 3 gave up sports altogether. None of these patients had residual pain and all recovered their muscle strength 1 year after surgery and returned to sports at an average of 7.7 months (range, 3 to 10 months). Over 86% of the high-performance athletes and 86% of athletes involved in recreational sports resumed the same level of sports activity. Complications Recorded complications included 1 deep infection, 2 superficial infections, 1 hematoma treated by simple aspiration, 2 hematomas evacuated surgically, 5 cases of algodystrophy, 1 knee requiring manipulation un-
A⫹B
%
C
%
D
%
194 194 198 193 188
97 97 99 96.5 94
6 3 1 5 9
3 1.5 0.5 2.5 4.5
3 1 2 3
0 1.5 0.5 1 1.5
der general anesthesia, 2 arthroscopic arthrolyses, and 1 cyclops syndrome. There has been a very low incidence of hematoma formation, although no postoperative drainage was performed (2 surgical evacuations), a very low rate of algodystrophy (5 cases), and good restoration of range of motion since we performed 2 arthroscopic arthrolyses for a cyclops syndrome and articular adhesions. There was only 4.5% of residual tendinitis at 1 year (1 fascia lata, 3 hamstring, 4 patellar tendon, and 1 quadriceps). Three grafts failed with 6 mm, 7 mm, and 12 mm of laxity, respectively. There was a marked rotational jerk. One hyperlax young woman (a handball player) was injured at home within the first 3 postoperative weeks. One patient gradually recovered his normal laxity, but the graft had completely disappeared at resumption of activities. The third patient sustained a new trauma at resumption of his sports activities. DISCUSSION This study examined the outcome of surgical reconstruction for a consecutive series of patients with isolated ACL deficiency. All procedures were performed in a single center using a similar technique by 1 of 4 experienced surgeons. All patients underwent a prescribed rehabilitation program. All review assessments were carried out by an independent observer, thereby minimizing susceptibility bias. According to published reports, our results seem to be satisfactory. In 1999, Corry et al.6 published the 2-year follow-up results of a group of patients who had undergone surgical reconstruction using 4-strand hamstring autograft. In this group, 72% were graded as A, with the KT 1000 at 89 N identifying a median laxity of 1.7 mm as against 1.4 mm in our series. Comparison of laxity by gender of the patient found the average laxity in males to be 1.1 mm versus 0.9 mm for Corry et al., and the average laxity in females to be 1.7 mm versus 2.5 mm for Corry. We have also found a significant difference between male and female patients (P ⫽ .04, t test), although 66.5% of the
236
P. COLOMBET ET AL.
patients in our series were male compared with 52% in the study performed by Corry. Similar results were found for IKDC overall evaluation between the 2 series. Corry reported 40% grade A, 53% grade B, 5% grade C, and 2% grade D. Equally, our results compare favorably with those of Aglietti et al.7 who evaluated 30 patients 2 years after reconstruction using hamstring tendon graft and suspensory fixation. This prospective study reported that 20% of the patient cohort had laxity greater than 5 mm as measured with the KT-1000 at 30 lb, whereas this figure was only 2.5% in our study. However, at the time of surgery, more than 1 method of fixation was used. Indeed, in some patients, simple suture methods were used, which could account for the discrepancy in outcome. Recent biomechanical studies performed by Giuera et al.8 and Brown and Skahr9 have shown the superiority of bone-tendon-bone fixation with interference screws. When interference screws are used for fixation of the STG tendon graft, care should be taken to avoid aggressive physical therapy. On the femoral side, fixation strength is up to 400 to 500 N. This is due to the fact that bone density is usually good and the graft is permanently under tension because it is not aligned with the longitudinal axis of the socket. In contrast, on the tibial side, fixation fails on the first day at a cyclic load of 450 N applied parallel to the long axis of the bone tunnel. However, it should be noted that secondary fixation through incorporation of the graft enhances primary fixation; with the STG tendon graft, this has been shown to take place between 8 and 10 weeks after implantation.10 There are other methods of hamstring tendon fixation that might make possible a more aggressive rehabilitation as performed in specialized rehabilitation centers: cortical or corticocancellous fixation. However, interference screws provide anatomic fixation, thus avoiding deformation of the tunnels due to bungee and windshield-wiper effects. Interference screws for hamstring tendon fixation are now available in several diameters and lengths (7, 8, and 9 mm diameters; 25 and 30 mm lengths), and a reverse pitch screw has been developed for right knees, allowing for better positioning of the graft within the femoral socket. Tibial fixation has been improved by the use of 9 ⫻ 25 mm screws: Magen et al.10 showed that yield point is over 700 N and stiffness is 220 N/mm. Furthermore, these authors showed that slippage of a 9 ⫻ 25 mm RCI interference screw implanted in pig tibial bone is less than 1 mm at a load of 500 N. Preservation of the extensor mechanism does decrease the incidence of secondary complications.
Moreover, harvesting of hamstring tendons has not been found to have any detrimental effects, as confirmed by the Cybex test. Overall results show that loss of quadriceps strength decreases from 17.29% to 16.08% as motion speed increases, loss of hamstring strength decreases from 10.50% to 5.23% as motion speed increases, and hamstring/quadriceps ratio is slightly high at a low speed (84.73%) and at 180°/sec (88%). At 1-year follow-up, there is a mild loss of quadriceps and hamstring strength. Therefore, it appears that harvesting of the gracilis and semitendinosus tendons does not adversely affect the muscle output in flexion. The radiographic study of bone tunnels identified width (36%), tapering (9%), and bulging (9%) on the tibial side, with width (22%), tapering (8%), and bulging (10%) on the femoral side. These changes were attributable both to biologic factors (synovectomy and osteolysis) and mechanical factors, thus emphasizing the importance of anatomic fixation. These findings are consistent with those of Insalata et al.11 (28% femoral changes, 25% tibial changes) and Jansson et al.12 (23% femoral changes, 23% tibial changes). We believe that the KT-1000 measurements tended to optimize the outcome scores. We then made a comparative study of laxity as measured using the KT-1000 at 30 lb and the Telos at 245 N (differential radiographic displacement of the tibia relative to the femur under a load of 245 N, between the healthy and the affected side), in 107 patients. Mean displacement was 1.34 mm with the KT-1000 at 30 lb versus 4.02 mm with the Telos. The 2.68-mm discrepancy between the 2 methods of measurement is very close to that reported by Jardin et al.13 between the KT-1000 and the Telos (2.54 mm). We believe that these radiographic measurements are much more realistic. This is the whole problem of choosing the appropriate method to evaluate laxity. The arthrometric laxity score had a significant impact on the overall IKDC score. The median instrumented side-to-side difference was 0.7 mm for patients whose score placed them in group A, 3.4 mm for patients in group B, 6.3 mm for those in group C, and 12 mm for those in group D. We did not find a correlation between level of sporting activity and overall IKDC score (P ⫽ .11, -squared test) or with final instrumented laxity measurements. There was no difference in overall IKDC assessment between reconstruction for acute ACL disruption (49% overall IKDC group A, 46% group B) and chronic instability (55% overall group A, 38% group B) (P ⫽ .3, -squared test). Laxity measurements were similar for
FOUR-STRAND STG GRAFTS AND SCREW FIXATION these 2 groups of patients with a side-to-side difference of 1.2 mm in the acutely reconstructed group and 1.4 mm in the chronic instability group. Although Lachman testing showed that more than 98% of patients were grade A or B at both 6 and 12 months after surgery, there was a trend toward increasing laxity on the operated limb between these 2 time points (overall median laxity 1.1 mm at 6 months and 1.3 mm at 12 months) but without statistical significance. In conclusion, ACL reconstruction using hamstring tendon fixation with interference screws, a postoperative dynamic range-of-motion knee mobile brace, and a nonaggressive rehabilitation protocol, is for us a dependable and reproducible method of treatment for high-performance athletes. However, the tendon fixation method may be improved if a more aggressive rehabilitation protocol is to be used. REFERENCES 1. Hey Groves EW. The cruciate ligaments of the knee joint. Their function, rupture, and operative treatment of the same. Br J Surg 1920;7:505-515. 2. Hey Groves EW. Operation for the repair of cruciate ligament. Lancet 1917;2:674-675. 3. Campbell WC. Repair of the ligaments of the knee: Report of a new operation for the repair of the anterior cruciate ligament. Surg Gynecol Obstet 1936;62:964-968. 4. Macey HB. A new operative procedure for repair of ruptured cruciate ligament of the knee joint. Surg Gynecol Obstet 1939; 69:108-109.
237
5. Deehan DJ, Salmon L, Webb V, Davies A, Pinczewski LA. The natural history of endoscopic anterior cruciate ligament reconstruction with ipsilateral patellar tendon autograft. A prospective longitudinal five year study. J Bone Joint Surg Br 2000;82:984-991. 6. Corry IS, Webb JM, Clingeleffer AJ, Pinczewski LA. Endoscopic ACL reconstruction comparing 4 strands hamstring tendon with patellar tendon autograft: Two years results. Am J Sport Med 1999;27:444-454. 7. Aglietti P, Buzzi R, Zaccherotti G, De Biase P. Patellar tendon versus doubled semitendinosus and gracilis tendons for anterior cruciate ligament reconstruction. Am J Sports Med 1994; 22:211-218. 8. Giurea M, Zorilla Amis AA, Aircroth P. Comparative pull-out and cyclic-loading strength tests of anchorage of hamstring tendons grafts in anterior cruciate ligaments reconstruction. Am J Sports Med 1999;27:621-625. 9. Brown CH Jr, Skahr JH. Endoscopic anterior cruciate ligament reconstruction using quadruple hamstring tendons and endobutton femoral fixation. Tech Orthop 1998;13:281-298. 10. Magen HE, Howell SM, Hull ML. Structural properties of six tibial fixation methods for anterior cruciate ligament soft tissue graft. Am J Sports Med 1999;27:35-43. 11. L’Insalata JC, Klatt B, Fu FH, Harner CD. Tunnel expansion following anterior cruciate ligament reconstruction: A comparison of hamstring and patellar tendons autograft. Knee Surg Sports Traumatol Arthrosc 1997;5:234-238. 12. Jansson KA, Harilainen A, Sandelin J, et al. Bone tunnel enlargement after anterior cruciate ligament reconstruction with hamstring autograft and endobutton fixation technique. Knee Surg Sports Traumatol Arthrosc 1999;7:290-295. 13. Jardin C, Chantelot C, Migaud H, Gougeon F, Deboucker MJ, Duquennoy A. Low accuracy of KT-1000 arthrometer versus Telos radiographic measurements to assess knee anterior laxity after ACL graft intra and interobserver reproducibility of KT1000. Rev Chir Orthop 1999;85:708-712.
Purpose: The purpose of this study was to determine the outcome of 200 anterior cruciate ligament (ACL) reconstructions using hamstring tendons. Type of Study: This is a case series reporting on 200 endoscopic procedures for reconstruction of the ACL. Methods: This study included patients over 18 years old with a healthy controlateral knee, intact posterior cruciate ligament, and without any peripheral surgical procedure or cartilage injury. Patients having undergone prior ligament reconstruction were excluded from the study. The minimum follow-up was 1 year. Clinical review allowed for documentation of International Knee Documentation Committee (IKDC), KT-1000 arthrometer laxity measurement, and isokinetic dynamometric analysis. Results: Quadriceps and hamstring muscle strength loss was less than 17%. At review, overall IKDC evaluation found that 50% of patients graded A, 44% graded B, and 6% were C or D. The differential anterior laxity was graded A for 157 patients with a median of 1 mm. Of the 113 high-performance athletes, 98 (86%) had resumed a preinjury level of sporting activity. Conclusions: Endoscopic reconstruction using 4-strand hamstring autograft may be considered safe, reliable, and reproducible. Preliminary outcome is fulfilling and this technique corresponds completely to therapeutic fields regarding ACL reconstructions. Key Words: Anterior cruciate ligament reconstruction—Hamstring autograft—Interference screw fixation—Endoscopic method.
T
he surgical treatment of knee instability secondary to anterior cruciate ligament (ACL) deficiency has evolved during the 20th century. ACL reconstruction was first performed in 1917 by Hey Groves1,2 using a strip of fascia lata. A variety of different donor grafts have since been used. Campbell3 reported the use of the medial third of the patellar tendon and Macey4 used the semitendinosus tendon with a distally attached tibial bone plug. These techniques have evolved over time with improved harvest-
From the Centre de Chirurgie Orthope´dique et Sportive, Bordeaux-Me´rignac, Me´rignac, France. Address correspondence and reprint requests to Philippe Colombet, M.D., Ph.D., Centre de Chirurgie Orthope´dique et Sportive, Bordeaux-Me´rignac, 9 Rue Jean Moulin, 33700 Me´rignac, France. E-mail: [email protected] © 2002 by the Arthroscopy Association of North America 0749-8063/02/1803-2795$35.00/0 doi:10.1053/jars.2002.30637
232
ing techniques, graft preparation, more thorough understanding of joint biomechanics and the role of isometricity in graft function, and the emergence of endoscopic surgery. The bone–patellar tendon– bone graft harvested from the middle third of the patellar tendon may still be regarded as the gold standard. Satisfactory results have been extensively recorded by Deehan et al.5 However, although excellent stability was achieved, a subgroup of patients experienced persistent donor-site morbidity and delay in return of full range of movement, with a small percentage of patients experiencing a slow return to work. Therefore, in 1996, this unit began to use a 4-strand hamstring autograft. It was felt that this would minimize quadriceps mechanism morbidity, thereby decreasing operative trauma with an expectation of reduced secondary complications, while still providing excellent knee stability. This report outlines our early experience to date in the first 200 patients.
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 18, No 3 (March), 2002: pp 232–237
FOUR-STRAND STG GRAFTS AND SCREW FIXATION PATIENTS AND METHODS This prospective study involved a series of 200 consecutive patients (1 center, 4 surgeons) who underwent surgery between July 1997 and September 1998. All patients were evaluated preoperatively, and at 6 months and 1 year postoperatively, using the CLAS system (Competitor Leisure Active Sedentary) for allocation of patients into 1 of 4 categories according to highest level of sporting activity. These would be equivalent to the 4 subgroups in the IKDC rating system and a full IKDC evaluation form. Laxity was measured using a KT-1000 Arthrometer (MedMetric, San Diego, CA) at 30 lb. Inclusion criteria included patients over 18 years of age with a healthy contralateral knee who had not had any prior peripheral surgical procedure, an intact posterior cruciate ligament, and no cartilage injury identified at initial arthroscopy. Patients having undergone prior ligament reconstruction were excluded from the study. Among 566 patients operated on between July 1997 and September 1998, the first 200 patients have been evaluated at a minimum 1-year follow-up. The mean follow-up period was 15 months (range, 12 to 25 months). The mean age of the patients was 26 years (range, 18 to 52 years). There were 67 women and 133 men. No bilateral procedures were performed. The right knee was involved in 101 cases and the left knee in 99 cases. The patient population consisted of 113 high-performance athletes (patients involved in national and international sporting events), 71 athletes involved in recreational sports, 9 active patients, and 7 sedentary patients. Preoperative alignment was documented as varus in 115 knees, neutral in 63, and valgus in 22 knees. Forty-two patients underwent surgical reconstruction within 3 months of injury with the remaining 158 patients having chronic instability. Twenty-four patients had a history of meniscectomy and a further 20 patients had undergone arthroscopic examination without formal resection. Surgical Technique A longitudinal 4-cm long proximal tibial incision was made, and both tendons (semitendinosus and gracilis) were harvested using a tendon stripper (Linvatec, Largo, FL). All remaining muscular debris was cleaned away from the graft. The 4 tendon strands were sewn to each other. Two arthroscopic portals (anteromedial and anterolateral patellar portals) were made, providing access to the joint. The intercondylar notch was cleaned with a 5.5-mm full-radius shaver (Dyonics, Smith & Nephew, Andover, MA). Using a
233
curette, the cortical median portion of the lateral femoral condyle was cleaned of all ligamentous debris, which provided full visualization of its posterior cortical margin. A bone spike was used to create a pilot hole located within 5 mm of the posterior border of the femoral condyle. The knee was flexed to 130° and a 3-cm femoral socket created in an inside-out direction to accommodate the previously harvested graft. A pull-through suture was placed into the eye of a Beath pin, passed through the femoral socket, and out through the lateral thigh. The knee was flexed to 90° and a guide pin was inserted with a guidewire that ensured accurate intra-articular positioning of the tibial tunnel (Acufex, Smith & Nephew, Andover, MA). The bone tunnel was fashioned over this guidewire, using the original graft harvesting incision and a reamer of the same diameter as that of the graft. The pull-through suture was grasped proximally and the graft pulled in a retrograde manner through the tibial tunnel and docked into the femoral socket. The knee was flexed again to 130°, and a 7- ⫻ 25-mm softthreaded, perforated, round-headed interference screw (RCI; Smith & Nephew) passed over a guide pin, via the portal using the anteromedial previously used for drilling. This ensured correct alignment of the graft and the interference screw. The knee was brought back to 90° of flexion, tension on the graft was maintained throughout by pulling distally, and a second interference screw of equivalent diameter was passed into the tibial tunnel close to the proximal portion of the tibial tunnel. Rehabilitation Full weight bearing and rehabilitation were commenced immediately. Patients were discharged after 2 days. Patients were directed toward achieving full extension using closed-chain exercises. All patients were requested to wear a dynamic range-of-movement knee orthosis (Largarigue, Bordeaux, France) for 6 weeks. This allowed for 0° to 60° of flexion for the first 2 weeks and then 0° to 90° for the remaining 4 weeks. Cybex tests were performed 5 months after surgery in professional athletes for muscle assessment. This test compared the affected knee with the healthy side and consisted of a series of managed cycles of arc of motions at differing speeds. The relative movement of the tibia on the femur at these speeds was determined with a measurement being taken as the mean of 5 repetitions.
234
P. COLOMBET ET AL.
Review Patients were assessed at 6 months and at 1 year after surgical reconstruction by independent observers. At these time points, subjective symptoms were recorded and clinical examination performed to allow for completion of the IKDC score. Each patient was examined using the KT-1000 (at 30 lb displacement force) to determine side-to-side knee laxity. In addition, we completed laxity assessments for 125 of our patients using the Telos system (Telos, Hungen, Germany). These recordings were taken at 6 and 12 months after surgery. Side-to-side quadriceps circumference was taken as a reading 10 cm from the superior pole of the patella. Statistical Analysis For data analysis, the unpaired Student t test was used to compare continuous data. The correlation test was used to compare ordinal data from the 2 treatment groups; P ⬍ .05 was regarded as significant. RESULTS Operative Intervention All procedures were undertaken using the same technique in a single center under the care of 4 surgeons. The patients participated in a standard rehabilitation program. Forty medial meniscectomies and 33 lateral meniscectomies were also carried out at the time of surgical reconstruction. A further 26 meniscal tears were confirmed as stable and left intact. Graft diameters were as follows: in 20 patients, 6.5 mm; in 35 patients, 7 mm; in 59 patients, 7.5 mm; in 46 patients, 8 mm; and in 40 patients, 8.5 mm. The median time lost from work was 75 days (range, 9 to 307 days). Median time to return to sports was 236 days (range, 57 to 354 days). Of the 113 competitive athletes, 98 (86.73%) returned to their previous level, 12 patients returned to “leisure” activity (class L), and 3 did not resume sports (Fig 1). Seventy-one patients were class L before surgery and we had 73 patients in class L at the time of evaluation (10 stopped sports activities so it was 71 at first plus 12 from the competitor class minus the 10 who became only active, thus 73). We had 22 active patients at 1 year and 7 sedentary patients the same as preoperatively. The quadriceps was judged normal in 130 cases, mild amyotrophy (1 to 3 cm side-to-side difference) was found in 66 cases, and severe amyotrophy (side-
FIGURE 1.
Activity change.
to-side difference ⬎3 cm) in 4 cases. There were 191 patients who did not report symptoms pertaining to the extensor mechanism. However, 4 patients had patellar tendinitis, 1 had quadriceps tendinitis, and 4 had a fasciitis of the fascia lata. Seven patients experienced slight discomfort at the donor site. This morbidity did not compromise participation in the postoperative rehabilitation program. At final review, 149 patients were very satisfied and 45 were satisfied with their operation. However, 6 patients remained dissatisfied with the outcome of surgery because of their failure to attain the desired level of function. IKDC Evaluation One hundred fifty-nine patients returned to their preinjury level of sporting activity. According to the patients’ subjective assessment, 152 had the feeling that their knee was normal (grade A), 42 that it was almost normal (grade B), and 6 that it was abnormal (grade C). Twenty-seven patients (13.5%) continued to experience pain with moderate activity, and 3 (1.5%) with activities of daily living. Eleven patients (5.5%) had effusion with exercise. Six patients did not feel safe with moderate activity, and 2 felt unstable with mild activity. In the symptom parameter group, 150 patients were graded A, 44 B, 3 C, and 3 D. In 194 cases, there was a difference of less than 5° in sideto-side loss of extension. In 181 cases, flexion was identical (⫾ 5°) on both sides, in 18 there was a loss of flexion between 6° and 15°, and in 1 there was a
FOUR-STRAND STG GRAFTS AND SCREW FIXATION
235
TABLE 1. Postoperative IKDC Grade
Patient subjective assessment Symptom group Range-of-motion group Ligament examination group Final evaluation
A
%
B
%
152 150 179 143 100
76 75 89.5 71.5 50
42 44 19 50 88
21 22 9.5 25 44
loss of 20°. In the range of motion parameter group, 179 patients were graded A, 19 B, 1 C, and 1 D. As evaluated with the KT-1000 at 30 lb, 157 patients were graded A with a median of 1 mm, 38 B with a median of 3 mm, 4 C with a median of 6 mm, and 1 D with a median of 12 mm. Pivot-shift testing found that 181 patients were grade 0, 15 patients were grade 1, 3 patients were grade 2, and 1 patient was grade 3. In the laxity parameter group, instrumented laxity testing found that 143 patients were grade A, 50 were grade B, 5 were grade C, and 2 were grade D, according to IKDC criteria. One-Leg Hop Test The one-leg hop test result was over 90% (grade A) in 167 patients, between 75% and 90% (grade B) in 30, between 50% and 75% (grade C) in 2, and under 50% (grade D) in 1 patient. Results of the one-leg hop test are much better at 1 year than they were at 6 months. There were 46.5% grade A, 37.7% grade B, 11.9% grade C, and 3.77% grade D at 6 months versus 83.5% grade A, 15% grade B, 1% grade C, and 0.5% grade D at 1 year (Table 1). Return to Sports Activities Of the 113 high-performance athletes, 12 resumed recreational sports activities, and 3 gave up sports altogether. None of these patients had residual pain and all recovered their muscle strength 1 year after surgery and returned to sports at an average of 7.7 months (range, 3 to 10 months). Over 86% of the high-performance athletes and 86% of athletes involved in recreational sports resumed the same level of sports activity. Complications Recorded complications included 1 deep infection, 2 superficial infections, 1 hematoma treated by simple aspiration, 2 hematomas evacuated surgically, 5 cases of algodystrophy, 1 knee requiring manipulation un-
A⫹B
%
C
%
D
%
194 194 198 193 188
97 97 99 96.5 94
6 3 1 5 9
3 1.5 0.5 2.5 4.5
3 1 2 3
0 1.5 0.5 1 1.5
der general anesthesia, 2 arthroscopic arthrolyses, and 1 cyclops syndrome. There has been a very low incidence of hematoma formation, although no postoperative drainage was performed (2 surgical evacuations), a very low rate of algodystrophy (5 cases), and good restoration of range of motion since we performed 2 arthroscopic arthrolyses for a cyclops syndrome and articular adhesions. There was only 4.5% of residual tendinitis at 1 year (1 fascia lata, 3 hamstring, 4 patellar tendon, and 1 quadriceps). Three grafts failed with 6 mm, 7 mm, and 12 mm of laxity, respectively. There was a marked rotational jerk. One hyperlax young woman (a handball player) was injured at home within the first 3 postoperative weeks. One patient gradually recovered his normal laxity, but the graft had completely disappeared at resumption of activities. The third patient sustained a new trauma at resumption of his sports activities. DISCUSSION This study examined the outcome of surgical reconstruction for a consecutive series of patients with isolated ACL deficiency. All procedures were performed in a single center using a similar technique by 1 of 4 experienced surgeons. All patients underwent a prescribed rehabilitation program. All review assessments were carried out by an independent observer, thereby minimizing susceptibility bias. According to published reports, our results seem to be satisfactory. In 1999, Corry et al.6 published the 2-year follow-up results of a group of patients who had undergone surgical reconstruction using 4-strand hamstring autograft. In this group, 72% were graded as A, with the KT 1000 at 89 N identifying a median laxity of 1.7 mm as against 1.4 mm in our series. Comparison of laxity by gender of the patient found the average laxity in males to be 1.1 mm versus 0.9 mm for Corry et al., and the average laxity in females to be 1.7 mm versus 2.5 mm for Corry. We have also found a significant difference between male and female patients (P ⫽ .04, t test), although 66.5% of the
236
P. COLOMBET ET AL.
patients in our series were male compared with 52% in the study performed by Corry. Similar results were found for IKDC overall evaluation between the 2 series. Corry reported 40% grade A, 53% grade B, 5% grade C, and 2% grade D. Equally, our results compare favorably with those of Aglietti et al.7 who evaluated 30 patients 2 years after reconstruction using hamstring tendon graft and suspensory fixation. This prospective study reported that 20% of the patient cohort had laxity greater than 5 mm as measured with the KT-1000 at 30 lb, whereas this figure was only 2.5% in our study. However, at the time of surgery, more than 1 method of fixation was used. Indeed, in some patients, simple suture methods were used, which could account for the discrepancy in outcome. Recent biomechanical studies performed by Giuera et al.8 and Brown and Skahr9 have shown the superiority of bone-tendon-bone fixation with interference screws. When interference screws are used for fixation of the STG tendon graft, care should be taken to avoid aggressive physical therapy. On the femoral side, fixation strength is up to 400 to 500 N. This is due to the fact that bone density is usually good and the graft is permanently under tension because it is not aligned with the longitudinal axis of the socket. In contrast, on the tibial side, fixation fails on the first day at a cyclic load of 450 N applied parallel to the long axis of the bone tunnel. However, it should be noted that secondary fixation through incorporation of the graft enhances primary fixation; with the STG tendon graft, this has been shown to take place between 8 and 10 weeks after implantation.10 There are other methods of hamstring tendon fixation that might make possible a more aggressive rehabilitation as performed in specialized rehabilitation centers: cortical or corticocancellous fixation. However, interference screws provide anatomic fixation, thus avoiding deformation of the tunnels due to bungee and windshield-wiper effects. Interference screws for hamstring tendon fixation are now available in several diameters and lengths (7, 8, and 9 mm diameters; 25 and 30 mm lengths), and a reverse pitch screw has been developed for right knees, allowing for better positioning of the graft within the femoral socket. Tibial fixation has been improved by the use of 9 ⫻ 25 mm screws: Magen et al.10 showed that yield point is over 700 N and stiffness is 220 N/mm. Furthermore, these authors showed that slippage of a 9 ⫻ 25 mm RCI interference screw implanted in pig tibial bone is less than 1 mm at a load of 500 N. Preservation of the extensor mechanism does decrease the incidence of secondary complications.
Moreover, harvesting of hamstring tendons has not been found to have any detrimental effects, as confirmed by the Cybex test. Overall results show that loss of quadriceps strength decreases from 17.29% to 16.08% as motion speed increases, loss of hamstring strength decreases from 10.50% to 5.23% as motion speed increases, and hamstring/quadriceps ratio is slightly high at a low speed (84.73%) and at 180°/sec (88%). At 1-year follow-up, there is a mild loss of quadriceps and hamstring strength. Therefore, it appears that harvesting of the gracilis and semitendinosus tendons does not adversely affect the muscle output in flexion. The radiographic study of bone tunnels identified width (36%), tapering (9%), and bulging (9%) on the tibial side, with width (22%), tapering (8%), and bulging (10%) on the femoral side. These changes were attributable both to biologic factors (synovectomy and osteolysis) and mechanical factors, thus emphasizing the importance of anatomic fixation. These findings are consistent with those of Insalata et al.11 (28% femoral changes, 25% tibial changes) and Jansson et al.12 (23% femoral changes, 23% tibial changes). We believe that the KT-1000 measurements tended to optimize the outcome scores. We then made a comparative study of laxity as measured using the KT-1000 at 30 lb and the Telos at 245 N (differential radiographic displacement of the tibia relative to the femur under a load of 245 N, between the healthy and the affected side), in 107 patients. Mean displacement was 1.34 mm with the KT-1000 at 30 lb versus 4.02 mm with the Telos. The 2.68-mm discrepancy between the 2 methods of measurement is very close to that reported by Jardin et al.13 between the KT-1000 and the Telos (2.54 mm). We believe that these radiographic measurements are much more realistic. This is the whole problem of choosing the appropriate method to evaluate laxity. The arthrometric laxity score had a significant impact on the overall IKDC score. The median instrumented side-to-side difference was 0.7 mm for patients whose score placed them in group A, 3.4 mm for patients in group B, 6.3 mm for those in group C, and 12 mm for those in group D. We did not find a correlation between level of sporting activity and overall IKDC score (P ⫽ .11, -squared test) or with final instrumented laxity measurements. There was no difference in overall IKDC assessment between reconstruction for acute ACL disruption (49% overall IKDC group A, 46% group B) and chronic instability (55% overall group A, 38% group B) (P ⫽ .3, -squared test). Laxity measurements were similar for
FOUR-STRAND STG GRAFTS AND SCREW FIXATION these 2 groups of patients with a side-to-side difference of 1.2 mm in the acutely reconstructed group and 1.4 mm in the chronic instability group. Although Lachman testing showed that more than 98% of patients were grade A or B at both 6 and 12 months after surgery, there was a trend toward increasing laxity on the operated limb between these 2 time points (overall median laxity 1.1 mm at 6 months and 1.3 mm at 12 months) but without statistical significance. In conclusion, ACL reconstruction using hamstring tendon fixation with interference screws, a postoperative dynamic range-of-motion knee mobile brace, and a nonaggressive rehabilitation protocol, is for us a dependable and reproducible method of treatment for high-performance athletes. However, the tendon fixation method may be improved if a more aggressive rehabilitation protocol is to be used. REFERENCES 1. Hey Groves EW. The cruciate ligaments of the knee joint. Their function, rupture, and operative treatment of the same. Br J Surg 1920;7:505-515. 2. Hey Groves EW. Operation for the repair of cruciate ligament. Lancet 1917;2:674-675. 3. Campbell WC. Repair of the ligaments of the knee: Report of a new operation for the repair of the anterior cruciate ligament. Surg Gynecol Obstet 1936;62:964-968. 4. Macey HB. A new operative procedure for repair of ruptured cruciate ligament of the knee joint. Surg Gynecol Obstet 1939; 69:108-109.
237
5. Deehan DJ, Salmon L, Webb V, Davies A, Pinczewski LA. The natural history of endoscopic anterior cruciate ligament reconstruction with ipsilateral patellar tendon autograft. A prospective longitudinal five year study. J Bone Joint Surg Br 2000;82:984-991. 6. Corry IS, Webb JM, Clingeleffer AJ, Pinczewski LA. Endoscopic ACL reconstruction comparing 4 strands hamstring tendon with patellar tendon autograft: Two years results. Am J Sport Med 1999;27:444-454. 7. Aglietti P, Buzzi R, Zaccherotti G, De Biase P. Patellar tendon versus doubled semitendinosus and gracilis tendons for anterior cruciate ligament reconstruction. Am J Sports Med 1994; 22:211-218. 8. Giurea M, Zorilla Amis AA, Aircroth P. Comparative pull-out and cyclic-loading strength tests of anchorage of hamstring tendons grafts in anterior cruciate ligaments reconstruction. Am J Sports Med 1999;27:621-625. 9. Brown CH Jr, Skahr JH. Endoscopic anterior cruciate ligament reconstruction using quadruple hamstring tendons and endobutton femoral fixation. Tech Orthop 1998;13:281-298. 10. Magen HE, Howell SM, Hull ML. Structural properties of six tibial fixation methods for anterior cruciate ligament soft tissue graft. Am J Sports Med 1999;27:35-43. 11. L’Insalata JC, Klatt B, Fu FH, Harner CD. Tunnel expansion following anterior cruciate ligament reconstruction: A comparison of hamstring and patellar tendons autograft. Knee Surg Sports Traumatol Arthrosc 1997;5:234-238. 12. Jansson KA, Harilainen A, Sandelin J, et al. Bone tunnel enlargement after anterior cruciate ligament reconstruction with hamstring autograft and endobutton fixation technique. Knee Surg Sports Traumatol Arthrosc 1999;7:290-295. 13. Jardin C, Chantelot C, Migaud H, Gougeon F, Deboucker MJ, Duquennoy A. Low accuracy of KT-1000 arthrometer versus Telos radiographic measurements to assess knee anterior laxity after ACL graft intra and interobserver reproducibility of KT1000. Rev Chir Orthop 1999;85:708-712.