Quantitative evaluation of knee instability and muscle strength after anterior cruciate ligament reconstruction using patellar and

quadriceps tendon KAZUNORI YOSHIE

YASUDA,*† MD, PhD, YASUMITSU OHKOSHI,* MD, TANABE,‡ RPT, AND KIYOSHI KANEDA,* MD, PhD

*Department of Orthopaedic Surgery, Hokkaido University School of Medicine, University Hospital and Noboribetsu Branch Hospital, and the ‡Section of Rehabilitation, Shin-Sapporo Orthopedic Hospital, Sapporo, Japan From the

ABSTRACT

struction, including the patellar and quadriceps

don,’ z6 the iliotibial

Anterior cruciate ligament reconstruction using an autologous graft harvested from the central one-third of the patellar and quadriceps tendon was performed in 65 knees of 65 patients who were followed from 3 to 7 years. Mean anterior laxity of both knees was measured before and after surgery in each patient using the Styker Knee Laxity Tester. At 30° of knee flexion, 58 patients (89%) had differences of less than 2.5 mm between the operated and unoperated knees. Quadriceps strength was measured with the Cybex II and was less than 50% of the uninjured knee at 3 months after surgery. In men, quadriceps strength returned to 78% of normal at 1 year and 85% at final followup. These values were equal to the preoperative level. In women, the quadriceps strength at final followup was 70%, significantly lower than preoperative strength. Hamstring strength recovered to equal the normal strength. Although anterior cruciate ligament reconstruction using one-third of the patellar and quadriceps tendon achieves stability, postoperative quadriceps weakness is a disadvantage. This weakness may be caused by impairment of the knee extensor mechanism resulting from harvesting the graft. We do not currently recommend this technique for anterior cruciate ligament reconstruction.

Several autologous grafts have been used for ACL

ten-

and the semitendinosus tendon.&dquo; In 1978, Marshall et a1.15° ls developed a method to obtain sufficient length in the graft by using the central onethird of the patellar and quadriceps tendon. We modified their technique in 1982 to obtain a graft with an even diameter, making the patellar portion stronger. We have reported an arthroscopic follow-up study on the morphologic maturation of the graft using this type of ACL reconstruction.23 In the early postoperative period, the graft is enveloped in thick synovial tissue with abundant capillary blood vessels. This synovial tissue gradually decreases, resulting in long-term maintenance of the ligamentous tissue enveloped within a thin synovial tissue. Arthroscopic biopsy at the 1- to 2-year followup demonstrated that the graft had matured as ligamentous tissue, but long-term results are not yet available. An essential factor in determining the success of ACL reconstruction is the quantitative evaluation of knee instability. Also important is the determination of the impairment of knee function resulting from the harvest of the graft because the harvest makes an injury in a part of the knee extensor mechanism. One method of evaluating this impairment is to measure the quadriceps and hamstring strength before and after surgery. There are few published studies of muscle strength after patellar and quadriceps tendon substitution techniques. Muscle strength in the early postoperative phase, especially the first 6 months, has rarely been tested because it is generally believed that quadriceps muscle contraction exerts an anterior drawer force on the tibia. We determined that isolated quadriceps muscle contraction with the knee in 70°

recon-

t Address correspondence and reprnt requests to Kazunon Yasuda, MD, PhD, Department of Orthopaedic Surgery, Hokkaido Umversity School of Medicme, Kita 14-~0, Nishi 5-chome, Sapporo, 060, Japan. 471

tract,8,18

472

of flexion exerted a posterior drawer force on the a safe method to use in measuring quadriceps torque in the early postoperative phase and allowed us to conduct a prospective study to measure the strength of the quadriceps and hamstring muscles at various periods after ACL reconstruction throughout the 3- to 7year followup. We sought to learn whether 1) this type of ACL reconstruction improved postoperative knee stability in the ACL-deficient knee, 2) quadriceps and hamstring strength returned to normal levels, and 3) this procedure or more

tibia.21 This provided

was

useful for ACL reconstruction.

MATERIALS AND METHODS Between 1982 and 1985, 87 symptomatic ACL-deficient knees in 87 patients were reconstructed using the autogenous patellar and quadriceps tendon. Sixty-five patients, 36 men and 29 women, were available for follow-up evaluation. The ages of the follow-up group at the time of surgery ranged from 18 to 39 years (average, 23). Followup ranged from 3 to 7 years (average, 4.5). Preoperatively all 65 patients were limited in sports or daily activities because of frequent episodes of giving way of the knee. Preoperative arthroscopic examination found marked attenuation or absence of the ACL in all of the knees. In addition to the ACL tear, 35 patients (53%) had medial collateral ligament (MCL) insufficiency and 5 patients (8%) had insufficiency of the lateral structures in their knees. Meniscal injury was found in 51 patients (78%): the medial meniscus was torn in 26 knees (40%), the lateral meniscus in 15 (23%), and both menisci in 10 (15%). In those knees that had no significant medial or lateral instability, intraarticular ACL reconstruction alone (without extraarticular augmentation) was performed. Other procedures performed were MCL reconstruction using the semitendinosus tendon in 28, lateral extraarticular augmentation by the Losee et aI.14 method in 3, arthroscopic partial meniscectomy in 29, and meniscal suture in 22.

Surgical technique and

rehabilitation

We modified the original graft procedure by using the following substitution technique. A rectangular graft approximately 13 cm long was harvested from the quadriceps and patellar tendon and left attached to the tibial tubercle. The graft was as wide as the middle third of the patellar tendon. The patellar portion of the graft, with the periosteum, was harvested from the bone with a sharp scalpel. The quadriceps tendon portion of the graft was twice the thickness of the patellar portion; to preserve the suprapatellar pouch, the full thickness of the quadriceps tendon was not harvested. The quadriceps tendon portion of the graft was separated with a knife into two flaps on the frontal plane, ending just short of the patellar portion (Fig. lA). One of the flaps was retroflexed and firmly sutured to the patellar portion and a part of the patellar tendon to reinforce the thin patellar portion. This provided a rectangular graft of fairly even thickness and diameter after tubing (Fig. 1, B and C). The joint cavity was approached through the slit in the

Figure 1. Surgical procedure for reinforcing graft. A, quadriportion of the graft is split into two segments; B, the top segment is attached to the patellar portion of the graft to cover and augment that thinner area; C, the tubing results in a graft of uniform size and strength. ceps

patellar tendon made by harvesting the graft. A tibial tunnel was made beginning just superior to the patellar tendon attachment and directed toward the medial intercondylar tubercle. A femoral drill hole was made using the standard technique reported by Clancy et al.~ The isometric position of the tunnel was confirmed by passing a trial graft through the femoral and tibial tunnels. Finally, the graft was securely fixed to the femur with a spiked staple under maximum tension (Fig. 2). The knees of the first 35 patients in our study were immobilized in plaster casts for 4 weeks with the knees in 30° of flexion. For the other patients, custom-made functional knee braces were applied and continuous passive motion was begun the day after surgery. Partial weightbearing was encouraged in the brace at 6 weeks after surgery, and total weightbearing was allowed at 10 weeks. Extension of the knee was limited to 20° for the first 6 weeks and to 10° for another 6 weeks; extension and flexion were not limited after this time.’o Based on the results of previous biomechanical studies, quadriceps muscle exercise was encouraged, using simultaneous contraction of the quadriceps and hamstrings with the knee at 30° of flexion22 and isolated quadriceps contraction with the knee at 70° to 90° of flexion.21 The isolated contraction of the hamstrings was encouraged at various angles of knee flexion.

473

final followup (3 to 7 years). Torque was expressed as the ratio of the injured knee to the uninjured knee because absolute values of the torque were extremely variable. Statistical analysis was performed by using Student’s ttest, the chi-square test, and the Fisher-Behrens test.

RESULTS The subjective

Figure 2. The graft is securely attached a spiked staple.

to the femur

by use

of

Postoperative evaluation

Subjective evaluation was performed using the scoring system of Noyes et aI.17 Clinical evaluation was performed using manual tests such as the Lachman, anterior drawer, and jerk tests. The results of these tests were categorized three

ways23: 1) negative, when the results were the same as those of the normal side; 2) positive/negative, when the difference between the injured and uninjured knees was slight and laxity might be considered as normal; and 3) positive, when the difference between the injured and uninjured knees was significant. This simplified grading system was intended to make the evaluation more objective.

performed by measuring range (Stryker Knee Laxity Tester, Stryker Corp, Kalamazoo, MI), and quadriceps and hamstring torque (Cybex II, Lumex Inc, Ronkonkoma, NY).

Objective evaluation was of motion, anterior laxity

Anterior laxity was defined as tibial translocation under an anterior drawer force of 133 N. Anterior laxity was measured in both knees before and after surgery at 30° and 90° of knee flexion. To determine normal anterior laxity values in the Japanese population, we measured the normal knees of 60 young adults (30 men and 30 women). A difference of anterior laxity in the injured and uninjured knees was considered normal when less than 2.5 mm at 30° of knee flexion and less than 2.0 mm at 90°. These upper limit values were calculated statistically using the 95% confidence level. Maximum isometric torque of the quadriceps and the hamstrings was measured with the knee in 70° of flexion. This angle was chosen because we had determined that contraction of the quadriceps in this position did not place anterior drawer forces on the tibia .2’ This measurement was performed at 3, 6, 9, and 12 months after surgery and at

scores improved from an average of 28.4 points (SD, 9.4) preoperatively to 39.4 (SD, 7.4) postoperatively. Thirty-eight patients (58%) had scores of 40 to 50, 22 (34%) had scores of 30 to 39, and 5 (8%) had scores of 29 or lower. In preoperative clinical evaluation, the Lachman, anterior drawer, and jerk tests were positive in all of the patients. Postoperatively, they were positive in only 5 (8%), 5 (8%), and 4 (6)%, respectively (Table 1). The average postoperative range of knee motion was 1.0° (SD, 2.5°) to 143.4° (SD, 13.3°) of flexion. There was no significant difference between ranges of motion in patients with positive Lachman tests and those with negative ones. Seven patients (11%) could not perform terminal knee extension of 10°. Sixty patients (92%) could squat without discomfort; however, 34 patients (53%) could not sit on the floor in Japanese fashion (knees bent with the toes directly beneath the body) for more than 5 minutes. The average difference in anterior laxity between the injured and the uninjured knees at 30° of flexion improved significantly from 6.5 mm preoperatively to 1.5 mm postoperatively (Table 2). The postoperative difference was not statistically different from the normal value. At 90° of flexion, the mean difference improved significantly from 3.3 mm preoperatively to 1.4 mm postoperatively; however, the postoperative average was significantly different from that of the normal control. Comparing the postoperative anterior laxity of the injured knee to the uninjured knee, 58 knees (89%) were considered normal (differences of less than 2.5 mm) at 30° of flexion, and 54 (83%) were considered normal (differences of less than 2.0 mm) at 90° of flexion (Table 2). The mean preoperative isometric torque of the quadriceps

muscle was 79% in men and 78% in women. At 3 months after surgery, this dramatically decreased to 50% in men and 33% in women (Table 3), then increased gradually. At 12 months after surgery, the torque increased to 78% in men, but to only 66% in women, significantly less than the preoperative level. Compared to the uninjured knee, the mean torque in men was 85% at final followup. This value indicated some weakness, but lacked statistical significance. In women, the final follow-up torque of 70% was signifiTABLE 1

Postoperative evaluation of knee instability using manual tests

474 TABLE 2 Evaluation of anterior

laxity

° These values were decided using 95% confidence level in the normal population (see text). b Comparison between the postoperative and the preoperative values. Comparison between the postoperative and the control values. The control value in the normal population (see text) is 1.0 ± 0.7 at 30° of flexion and 0.8 ± 0.7 at 90°. c

TABLE 3 Isometric muscle

strength of the quadriceps and hamstringsa

° The torque in the injured side is shown as the ratio to that in the b Significantly lower than the preoperative value (P < 0.01). ‘ Not significantly lower than the preoperative value. d Significantly higher than the preoperative value (P < 0.01).

cantly lower than that measured preoperatively. The torque in women was significantly less than that in men at each period after reconstruction (Table 3). The mean preoperative isometric torque of the hamstrings muscle was 82% in men and 80% in women. At 3 months after surgery, this decreased to 57% in men and to 62% in women (Table 3), then increased gradually. Torque recovered to 90% in men and 89% in women at 12 months after surgery, which was significantly higher than preoperative levels. There was no significant difference between men and women. Torque at the final follow-up period was 97% in men and 96% in women, equal to normal levels

(Table 3). Of 37 patients who had participated in competitive sports before injury, 15 returned to the same sports, 19 changed to recreational sports, and 3 gave up sports. Of 28 patients who had participated in recreational sports before injury, 22 returned to the same sports, and 6 gave up sports. The only complications experienced by patients in this study were problems with marginal skin necrosis in two

patients. There were

no

acute

or

delayed infections.

DISCUSSION This study demonstrated that this modified ACL reconstruction technique provided favorable results to the ACL-deficient knee concerning postoperative knee stability. Kornblatt et al.11 reported their results using the original Marshall technique and found that, at 5.5 years after surgery, pivot shift tests were positive in 45% of the 38 patients examined. They concluded that the graft is relatively weak. In the

noninjured side. Numbers shown are the

average ± SD.

Marshall technique, the patellar portion of the graft consists of the thin tendinous membrane and periosteum.&dquo; In our modified method, that portion is reinforced with the wider and thicker quadriceps tendon (Fig. 1). This study demonstrated that the reinforcement technique using the larger volume of tendinous tissue was effective to improve knee stability. However, harvest of the tendinous tissue for the reinforcement from the quadriceps tendon has to increase impairment of the knee extensor mechanism. This is a

disadvantage of this procedure. Muscle strength is a significant factor in the return to vigorous activity after surgery. Despite the encouraging results of muscle exercise immediately after reconstruction, muscle torque of the quadriceps and hamstrings remarkably decreased in early the postoperative period. Hamstring torque returned to almost normal levels at 12 months, but quadriceps recovery was unsatisfactory, especially in women where quadriceps torque had not returned to preoperative levels at the final followup. In men, the quadriceps torque returned to preoperative levels at 12 months, but did not equal that of the uninjured knee even at final followup. We believe this quadriceps weakness may be caused by impairment of the knee extensor mechanism resulting from harvesting of the graft. The removal of the middle third of the tendon may reduce the efficiency of extension force transmission. Impairment of the neural mechanism in the tendon or surrounding tissues also may suppress activity of the anterior horn cells of the quadriceps and the hamstrings through some reflex mechanisms. We conclude that postoperative weakness of the quadriceps is a disadvantage of our method. We believe the use of

475

procedure increased postoperative weakness of our patients. Arthroscopically-assisted techniques may help pre-

icine, Burlington, Vermont, manuscript.

vent muscle weakness. Clancy et al.5 described a technique using a patellar tendon graft with a bone block at each end. This is currently a

REFERENCES

an

open

standard technique for ACL reconstruction. Tibone and Antich19 reported that after bone-patellar tendon-bone graft procedure, quadriceps torque in the operated knee was 83% of that in the nonoperated knee, and hamstring torque was 95%. Harter et al.~reported that quadriceps torque in the ACL-reconstructed knee was 84% of the uninjured knee and hamstring torque was 93%. While we found no reports of muscle strength after ACL reconstruction using the patellar and quadriceps tendon, these other results are similar to ours. Therefore, we conclude that quadriceps weakness may be a disadvantage of any ACL reconstruction in which a part of the knee extensor apparatus is used as a graft. All of our patients had marked decreased in quadriceps strength 3 months after surgery. The determination of quadriceps muscle torque in the early postoperative phase is important to estimate final muscle strength and to modify exercise programs to obtain maximum strength. Currently, hamstring tendons have been preferred as a substitute for ACL reconstruction because harvesting them does not injure the knee extensor apparatus. The rate of return to

for his critical advice

on

this

Alm A Survival of part of patellar tendon transposed for reconstruction of anterior cruciate ligament Acta Chir Scand 139 : 443-447, 1973 2 Amiel D, Kleiner JB, Akeson WH The natural history of the anterior cruciate ligament autograft of patellar tendon origin Am J Sports Med 14 4491

462, 1986 3

4

Arnoczky SP, Tarvin GB, Marshall JL Anterior cruciate ligament replacement using patellar tendon An evaluation of graft revascularization in the dog J Bone Joint Surg 64A 217-224, 1982 Chiroff RT Experimental replacement of the anterior cruciate ligament A histological and microradiographic study J Bone Joint Surg 57A 11241127,1975

Nelson DA, Reider B, et al Anterior cruciate ligament reconstruction using one-third of the patellar ligament augmented by extraarticular tendon transfers J Bone Joint Surg 64A. 352-359, 1982 6 Eriksson E Reconstruction of the anterior cruciate ligament Orthop Clin North Am 7 167-179, 1976 7 Harter RA, Osterning LR, Singer KM, et al Long-term evaluation of knee stability and function following surgical reconstruction for anterior cruciate : 434-443, 1988 ligament insufficiency Am J Sports Med 16 8 Hey Groves EW The crucial ligaments of the knee-joint Their function, of the and the treatment same Br J Surg 7 505-515, rupture, operative 1920 9 Johnson RJ, Eriksson E, Haggmark T, et al Five- to ten-year follow-up evaluation after reconstruction of the anterior cruciate ligament Clin Orthop 183 122-140,1984 10 Jones KG: Reconstruction of the anterior cruciate ligament A technique using the central one-third of the patellar ligament J Bone Joint Surg 45A. 5

Clancy WG,

925-932,1963 Kornblatt I, Warren RF, Wickiewicz TL Long-term followup of anterior cruciate ligament reconstruction using the quadriceps tendon substitution

vigorous sports activities was unsatisfactory in our patients, although knee stability was successfully reconstructed postoperatively. We had few professional or semiprofessional athletes in this study. Most patients gave graduation from high school or a university as the reason for changing from competitive to recreational sports. According to Tibone and Antich,19 return to competitive sports is not a valid criterion for determining the longterm results of ACL reconstruction. However, we believe that muscle weakness is responsible for the low percentage of our patients who returned to competitive sports activities

11

after surgery. Although several studiesl-4°9 indicate that autogenous tendon grafting is currently the most reliable method of ACL reconstruction, there are several disadvantages: harvest of the graft impairs the donor site and lengthy rehabilitation is necessary. Larson 12 stated it well: &dquo;We still have no elegant solution.&dquo; Further improvements in technique are still required for methods of ACL reconstruction. Our modified reconstruction technique using the central third of the patellar and quadriceps tendon improves postoperative knee stability, but results in quadriceps weakness. Therefore, we do not recommend this procedure for ACL reconstruction.

16 Marshall JL, Warren RF, Wickiewicz TL, et al The anterior cruciate ligament A technique of repair and reconstruction Clin Orthop 143 97-

ACKNOWLEDGMENT The authors thank Robert J. Johnson, MD, Department of Orthopaedics, The University of Vermont, College of Med-

for chronic anterior cruciate

ligament insufficiency

Am J

Sports

Med 16:

444-448,1988 12

Larson RL Future of prosthetic ligament reconstruction, in Friedman MJ, Ferkel RD (eds) Prosthetic Ligament Reconstruction of the Knee Philadelphia, WB Saunders Co, 1988, pp 209-211 13 Lipscomb AB, Johnston RK, Snyder RB, et al Secondary reconstruction of anterior cruciate ligament in athletes by using the semitendinosus tendon Am J Sports Med 7 81-84, 1979 14 Losee RE, Johnson TR, Southwick WO, et al Anterior subluxation of the lateral tibial plateau A diagnostic test and operative repair J Bone Joint 15

Surg 60A 1015-1030, 1978 Marshall JL, Rubin RM, Wang JB, et al The anterior cruciate ligament The diagnosis and treatment of its injuries and their serious prognostic implications Orthop Rev 7(10) 35-46, 1978 106, 1979

17

Mooar PA, Matthews DS, et al The symptomatic anterior cruciate-deficient knee Part I The long-term functional disability in athletically active individuals J Bone Joint Surg 65A 154-162, 1983 18 O’Donoghue DH A method for replacement of the anterior cruciate ligament of the knee Report of twenty cases J Bone Joint Surg 45A. 905-

Noyes FR,

924,1963 19 Tibone JE, Antich TJ A biomechanical analysis of anterior cruciate ligament reconstruction with the patellar tendon. A two year followup Am J Sports Med 16 332-335,1988 20 Yasuda K, Aoki Y, Kurosawa H, et al Rehabilitation of the anterior cruciate ligament reconstruction Rinsho Seikei Geka 23 837-846, 1988 21 Yasuda K, Sasaki T Exercise after anterior cruciate ligament reconstruction The force exerted on the tibia by the separate isometric contraction of the quadriceps or the hamstrings Clin Orthop 220. 275-283, 1987 22 Yasuda K, Sasaki T Muscle exercise after anterior cruciate ligament reconstruction Biomechanics of the simultaneous isometric contraction method of the quadriceps and the hamstrings Clin Orthop 220 266-274, 1987 23 Yasuda K, Tomiyama Y, Ohkoshi Y, et al Arthroscopic observations of autogeneic quadriceps and patellar tendon grafts after anterior cruciate : 217-224, 1989 ligament reconstruction of the knee Clin Orthop 246