6
Introduction Experimental studies have suggested that fixation of the bone-patellar tendon-bone (B-PT-B) graft at the level of the tibial plateau (anatomic fixation) improves graft isom- etry and objective stability compared with a more distal fixation within or outside the tibial tunnel. Morgan et al. [8] measured in vivo length changes of the distance be- tween the anterior cruciate ligament (ACL) graft insertion sites in 25 knees. The knee was flexed from 0° to 140°. Length changes were 1 mm on average when measured at the plateau level and 6 mm when measured outside the tibial tunnel, 13 cm distal to the plateau. Ishibashi et al. [5] measured knee stability using robotic technology in a series of ACL reconstructions in porcine knees. The graft was fixed at a common femoral position and at three dif- ferent levels in the tibial tunnel: anatomic (tibial plateau level), mid-tibial tunnel and outside the tibial tunnel. The anterior tibial translation was measured at 60° and 90° of knee flexion when applying a force of 110 N. Anterior tib- ial displacement was significantly lower with anatomic graft fixation. To evaluate clinically the effectiveness of anatomic tib- ial fixation, we studied 60 consecutive ACL reconstruc- tions performed using the B-PT-B graft. The patients were alternately assigned to group A (“anatomic fixation”, at the level of the tibial plateau) and group B (“non- anatomic fixation”, distal to the plateau level). Patients and methods Between May 1995 and April 1996, 60 chronic isolated ACL le- sions were reconstructed by the same surgeon (P.A.) using an au- tologous B-PT-B graft. An arthroscopically assisted technique was Abstract We prospectively com- pared two series of 30 anterior cruci- ate ligament (ACL) reconstructions each where the bone-patellar tendon- bone graft was alternately fixed at the level of the tibial plateau (group A; anatomic fixation) or distal to the plateau level (group B; non-anatomic fixation). In group A, a 35-mm-long tibial tuberosity bone block was har- vested. The distal 10–15 mm were resected and fixed proximally to the undersurface of the tendon to shorten it. After an average 18 months’ fol- low-up, there were no significant dif- ferences between the two groups concerning subjective evaluation, symptoms, range of motion and ob- jective stability. Tibial tuberosity pain was more frequent in group A (53% vs 17%, P = 0.01). Radiogra- phic evaluation showed that tibial tunnel enlargement was less frequent in group A (23% vs 43%, P = 0.02). There was no correlation between tunnel enlargement and objective sta- bility. In conclusion, fixation of the graft at the tibial plateau level did not improve objective stability in this study. Because of the greater techni- cal difficulty and occurrence of tibial tuberosity pain, this technique is not recommended. Key words Knee ligaments · Reconstruction · Patellar tendon · Tibial fixation Knee Surg, Sports Traumatol, Arthrosc (1998) 6 [Suppl 1] : S43–S48 © Springer-Verlag 1998 Paolo Aglietti Giovanni Zaccherotti Alfred J. V. Simeone Roberto Buzzi Anatomic versus non-anatomic tibial fixation in anterior cruciate ligament reconstruction with bone-patellar tendon-bone graft P. Aglietti (Y) · G. Zaccherotti · A. J. V. Simeone · R. Buzzi II Clinica Ortopedica, Università di Firenze, Largo P. Palagi, 1, I-50139 Florence, Italy

Anatomic versus non-anatomic tibial fixation in anterior cruciate ligament reconstruction with bone-patellar tendon-bone graft

  • Upload
    roberto

  • View
    215

  • Download
    2

Embed Size (px)

Citation preview

Page 1: Anatomic versus non-anatomic tibial fixation in anterior cruciate ligament reconstruction with bone-patellar tendon-bone graft

Introduction

Experimental studies have suggested that fixation of thebone-patellar tendon-bone (B-PT-B) graft at the level ofthe tibial plateau (anatomic fixation) improves graft isom-etry and objective stability compared with a more distalfixation within or outside the tibial tunnel. Morgan et al.[8] measured in vivo length changes of the distance be-tween the anterior cruciate ligament (ACL) graft insertionsites in 25 knees. The knee was flexed from 0° to 140°.Length changes were 1 mm on average when measured atthe plateau level and 6 mm when measured outside thetibial tunnel, 13 cm distal to the plateau. Ishibashi et al.[5] measured knee stability using robotic technology in aseries of ACL reconstructions in porcine knees. The graftwas fixed at a common femoral position and at three dif-ferent levels in the tibial tunnel: anatomic (tibial plateau

level), mid-tibial tunnel and outside the tibial tunnel. Theanterior tibial translation was measured at 60° and 90° ofknee flexion when applying a force of 110 N. Anterior tib-ial displacement was significantly lower with anatomicgraft fixation.

To evaluate clinically the effectiveness of anatomic tib-ial fixation, we studied 60 consecutive ACL reconstruc-tions performed using the B-PT-B graft. The patients werealternately assigned to group A (“anatomic fixation”, atthe level of the tibial plateau) and group B (“non-anatomic fixation”, distal to the plateau level).

Patients and methods

Between May 1995 and April 1996, 60 chronic isolated ACL le-sions were reconstructed by the same surgeon (P.A.) using an au-tologous B-PT-B graft. An arthroscopically assisted technique was

Abstract We prospectively com-pared two series of 30 anterior cruci-ate ligament (ACL) reconstructionseach where the bone-patellar tendon-bone graft was alternately fixed atthe level of the tibial plateau (groupA; anatomic fixation) or distal to theplateau level (group B; non-anatomicfixation). In group A, a 35-mm-longtibial tuberosity bone block was har-vested. The distal 10–15 mm wereresected and fixed proximally to theundersurface of the tendon to shortenit. After an average 18 months’ fol-low-up, there were no significant dif-ferences between the two groupsconcerning subjective evaluation,symptoms, range of motion and ob-jective stability. Tibial tuberosity

pain was more frequent in group A(53% vs 17%, P = 0.01). Radiogra-phic evaluation showed that tibialtunnel enlargement was less frequentin group A (23% vs 43%, P = 0.02).There was no correlation betweentunnel enlargement and objective sta-bility. In conclusion, fixation of thegraft at the tibial plateau level didnot improve objective stability in thisstudy. Because of the greater techni-cal difficulty and occurrence of tibialtuberosity pain, this technique is notrecommended.

Key words Knee ligaments · Reconstruction · Patellar tendon · Tibial fixation

Knee Surg, Sports Traumatol, Arthrosc(1998) 6 [Suppl 1] :S43–S48

© Springer-Verlag 1998

Paolo AgliettiGiovanni ZaccherottiAlfred J. V. SimeoneRoberto Buzzi

Anatomic versus non-anatomic tibial fixation in anterior cruciate ligament reconstruction with bone-patellar tendon-bone graft

P. Aglietti (Y) · G. Zaccherotti ·A. J. V. Simeone · R. BuzziII Clinica Ortopedica, Università di Firenze, Largo P. Palagi, 1, I-50139 Florence, Italy

Page 2: Anatomic versus non-anatomic tibial fixation in anterior cruciate ligament reconstruction with bone-patellar tendon-bone graft

used. Patients with combined medial, lateral or posterior instabilityand those with an instability of the opposite knee were excluded.

The patients were alternately assigned to group A (anatomicfixation) or group B (non-anatomic fixation). Comparison of thegeneral data showed that the two groups were similar in terms ofage, sex, length of interval between injury and surgery, preopera-tive KT-1000 anterior tibial translation [1], meniscal status andcartilage pathology (Table 1).

All 60 patients were evaluated using the IKDC Knee LigamentStandard Evaluation Form [4]. This form separately evaluates sub-jective assessment, symptoms (pain, swelling, giving-way), objec-tive stability and range of motion (ROM). Each parameter is clas-sified as normal, nearly normal, abnormal or severely abnormal.The final result is determined by the worst result achieved in eachcategory.

All patients were studied radiographically at follow-up. We ob-tained a standing anterior-posterior (A-P) view and a lateral (LL)view in full passive extension. Both images were centered with animage amplifier. Tibial tunnel enlargement was studied accordingto Peyrache et al. [10]. The diameter of the tunnel was measured atthe proximal (intra-articular tunnel exit), middle and distal locationon the A-P and LL views. Tunnel enlargement was classified as“cone type” if the proximal diameter was greater than the middle,and the middle greater than the distal. Tunnel enlargement wasclassified as “cavity type” when the proximal and distal diameterswere less than the diameter of the middle part of the tunnel. Tun-nel enlargement was also quantified as a ratio between the largestand the narrowest diameter of the tunnel.

The distance of the proximal aspect of the tibial bone blockfrom the tibial plateau level was measured in millimeters on theLL view (Fig.1).

The results were stored in a database and investigated for sta-tistical significance using the chi-square test and the non-pairedStudent’s t-test. The minimum level of significance was P = 0.05.

Surgical technique

The middle third of the patellar tendon (9–11 mm in width) washarvested with bone blocks at each end. In group A the tibial bone

S44

Table 1 General data on study subjects

Group A Group B(anatomic (non-anatomicfixation) fixation)

Age (years) 25 24(range 16–37) (range 14–40)

GenderMale 25 (83%) 25 (83%)Female 5 (17%) 5 (17%)

Interval injury-surgery 28 24(months) (range 3–110) (range 3–100)

Preoperative anterior tibial 6.40 5.29translation (mm)a (range 4–12) (range 4–9)

Meniscal statusb

Medial meniscectomy 10 6Lateral meniscectomy 9 9Bilateral meniscectomy 1 1Normal menisci 10 14

Cartilage pathologyc

Medial femoral condyle 1 2Lateral femoral condyle 1 1Patella – –Absent 28 27

a Measured with the KT-1000 arthrometer as side-to-side differ-ence at 132 N forceb Including previous surgery and meniscal surgery performed at thetime of anterior cruciate ligament (ACL) reconstructionc Deep fibrillation or erosion down to bone (≤ 1 cm in diameter)

Fig.1 Measurement of the position of the bone block within thetibial tunnel. A line tangent to the medial tibial plateau is drawn.The distance between this line and the proximal aspect of the boneblock is measured along the axis of the tibial tunnel

Fig.2A–C Preparation of the bone-patellar tendon-bone graft ingroup A knees (anatomic fixation). A The 35-mm-long tibialtuberosity bone block is predrilled near its proximal and distal endwith a 1.5 mm drill bit. The distal 10–15 mm are resected with anoscillating saw and cleared of the soft-tissue attachments. B Thebone graft is transposed to the undersurface of the tendon and fixedwith a 0.7 mm stainless steel wire to the bone block itself. C A cir-cumferential no. 2 Vycril suture (Ethicon, Pomezia, Italy) is addedto secure the bone graft to the tendon. A stainless steel wire ispassed through the patellar bone block

25 mm 35 mm

C

A

B

Page 3: Anatomic versus non-anatomic tibial fixation in anterior cruciate ligament reconstruction with bone-patellar tendon-bone graft

block measured 35 mm in length, while in group B it was 25 mm.The patellar bone block measured 25 mm in length in both groups.The length of the patellar tendon (from bone block to bone block)was measured using a ruler. It was 44 mm on average in group A(range 38–53 mm) and 46 mm in group B (range 40–54 mm).

In group A the tibial bone block was predrilled at its proximaland distal aspect with a 1.5 mm drill bit. The distal 10–15 mm ofthe tibial bone block was resected with an oscillating saw and su-tured to the undersurface of the tendon proximal to the tibial boneblock. The length of the bone graft to be resected depended on thepatellar tendon length. We attempted to achieve a final tendonlength of 30–32 mm. If the patellar tendon was, for instance, 42mm, we resected a 12 mm long cortical graft. This cortical bonegraft was fixed to the tibial bone block with 0.7 mm stainless steelwire (Fig.2). After this procedure the length of the patellar tendonwas reduced from 44 to 33 mm on average (range 29–36 mm) (Fig.3). In group B the B-PT-B graft was not shortened, and a 0.7 mmstainless steel wire was passed through each bone block.

A limited notchplasty was performed under arthroscopic controlin order to achieve an anterior notch width of at least 18–20 mmand a smooth lateral wall. The femoral and tibial tunnels were sim-ilarly prepared in both groups. For the femoral tunnel a K-wire wasintroduced through the anteromedial portal with the knee in maxi-

mum flexion. We aimed at a position at 11 o’clock for the right kneeand 6–7 mm anterior to the “over the top”. A second K-wire wasintroduced into the tibia using the “Pinn-ACL” guide (Linvatec,Largo, Fla.). We tried to place the K-wire on a line between the an-tero-medial tibial spine and the posterior aspect of the anterior hornof the lateral meniscus, one-third of the distance from the spine [6].

The position of the femoral and tibial K-wires was checked in-traoperatively with an image amplifier in the lateral view with theknee in full extension. The aim was to position a tibial K-wire par-allel and posterior to the Blumensaat line and a femoral K-wire inthe posterior 30% of the width of the condyles measured along theBlumensaat line.

A 25-mm-long femoral hemi-tunnel was produced by over-drilling the K-wire through the anteromedial portal with a cannu-lated and calibrated reamer (Acufex, Microsurgical, Manfield,Mass.) and the knee in maximum flexion. The tibial tunnel wascreated using a cannulated drill bit. The graft was introduced intothe joint until the patellar bone block was seated within thefemoral tunnel. Fixation on the femoral side was accomplished bytying the stainless steel wire over a screw and washer through asmall lateral incision. In group A tibial fixation was performedwith an 8 × 30 mm interference screw which engaged both the boneblock and the bone graft (Figs.4 and 5). In group B a 8 × 25 mm

S45

Fig. 3 The prepared bone-patellar ten-don-bone graft for group A knees(anatomic fixation). The length of thetendon has been reduced to 30 mm

Fig. 4 Graft fixation in group A(anatomic fixation). Proximal fixationis achieved by a stainless steel wire tiedaround a screw and washer. Distal fixa-tion is achieved by an interferencescrew advanced to the tibial plateaulevel which engages both the tibialbone block and the cortical graft. Fixa-tion is reinforced with a stainless steelwire

Fig. 5 Arthroscopic view of theanatomic fixation on the tibia. The in-terference screw, the cortical bone graftand the patellar tendon are shown

Fig. 6 Graft fixation in a group B knee(non-anatomic fixation). Fixation isachieved by a stainless steel wire proxi-mally and interference screw and wiredistally. The fixation level is distal tothe tibial plateau

3 5

4 6

Page 4: Anatomic versus non-anatomic tibial fixation in anterior cruciate ligament reconstruction with bone-patellar tendon-bone graft

interference screw was utilized (Fig.6). Tibial fixation was rein-forced by securing the stainless steel wire around a screw andwasher in both groups.

Patellar and tibial bone defects were filled using autologouscancellous bone obtained from drilling of the tibial tunnel. A fibringlue (Tissucol, Immuno, Pisa, Italy) layer was placed over thebone defects and within the tibial tunnel to seal it.

Postoperatively the patients used a brace (ELS, Donjoy, Carls-bad, Calif.) for 3–4 weeks. The brace was locked in full extensionduring the night. Before discharge, the patients had to achieve fullextension and at least 90° of flexion. Partial weight-bearing was al-lowed as tolerated and progressed to full weight-bearing, usuallyby the end of the first month. The same rehabilitation protocol withemphasis on closed kinetic chain exercises was recommended inboth groups [9, 14].

Results

The results were evaluated at an average follow-up of 19months in group A (range 15–24 months) and 17 monthsin group B (range 13–24 months) by a single independentexaminer (G.Z.). There were no significant intra-operativeor postoperative complications.

At the time of follow-up, 13 (43%) of the group A pa-tients participated in level I activities (cutting, jumping,football, soccer), 9 (30%) in level II (downhill skiing, ten-nis and heavy manual work) and 8 (27%) in level III (lightmanual work, jogging, running). Sixteen (53%) of thegroup B patients participated in level I activities, 10(33%) in level II and 4 (13%) in level III. Twenty-five(83%) of the group A patients and 28 (93%) of group Breturned to their preinjury activity level. Two patients ofgroup A reduced their activity level because of the oper-ated knee and 3 because of reasons not related to the knee.Two patients in group B reduced the activity level becauseof reasons not related to the operated knee. There were no

statistically significant differences in activity level be-tween groups A and B.

One patient (3%) in group A complained of significantpain, and 2 (7%) were subjectively dissatisfied (Table 2).A side-to-side difference > 5 mm in anterior tibial dis-placement at 132 N was present in 4 patients (14%) ofgroup A and in 2 (7%) of group B. A positive 2+ pivot-shift (“clunk”) was noted in 2 patients (7%) of group Aand in 1 (3%) of group B. The final result was unsatisfac-tory in 5 knees (17%) of group A and 2 (7%) of group B.These differences between groups A and B were not sta-tistically significant.

Pain over the tibial tuberosity (Table 3) during kneel-ing was moderate in 5 (17%) knees of group A and 2 (7%)of group B (P = 0.01). There were no differences betweengroups A and B regarding patellar crepitation and quadri-ceps strength evaluated with the “one-leg hop test” [2].

Radiographic measurements of the distance from theproximal margin of the tibial bone block to the tibialplateau line was possible in 50 of the 60 patients. In groupA knees the bone block was 1.2 mm proximal to the tibialplateau line on average (ranging from 7 mm proximal to 5mm distal). In group B the bone block was 13.4 mm dis-tal to the plateau line on average (ranging from 5 to 20mm distal) (P < 0.001). If the 10-mm bone block was notincluded in the measurements of group A, there were nosignificant differences between the two groups (14.2 mmvs 13.4 mm distal to the plateau line on average in groupsA and B, respectively).

Tibial tunnel shape modifications occurred in 7 (23%)knees of group A (2 cone type and 5 cavity type) and in 13(43%) of group B (3 cone type and 10 cavity type) (P =0.02). The percentage of tunnel enlargement was 3.1% onaverage (range 0%–20%) in group A and 9.6% (range

S46

Table 2 Clinical evaluation atfollow-up (IKDC form [4]) Normal Nearly normal Abnormal Severely abnormal

Group Group Group Group Group Group Group GroupA B A B A B A B

Subjectiveassessment 14 17 14 13 2 – – –

Pain 24 23 5 7 1 – – –

Swelling 28 29 2 1 – – – –

Giving-way 28 28 1 2 1 – – –

Extension loss(0°–2°) (3°–5°) (6°–10°) (> 10°)

29 29 1 1 – – – –

Flexion loss(0°–5°) (6°–15°) (16°–25°) (> 25°)

30 30 – – – – – –

KT-1000 ssd(0–2 mm) (3–5 mm) (6–10 mm) (> 10 mm)

16 17 10 11 4 2 – –

Pivot-shift(Absent) (1+, glide) (2+, clunk) (3+, gross)

22 25 6 4 2 1 – –

Final score 9 9 16 19 5 2 – –

Page 5: Anatomic versus non-anatomic tibial fixation in anterior cruciate ligament reconstruction with bone-patellar tendon-bone graft

0%–50%) in group B (P = 0.01). There was no correlationbetween tibial tunnel shape modifications and objectiveknee stability.

Discussion

Experimental studies have suggested that anatomic fixa-tion of the ACL graft on the tibia may improve objectivestability [5, 8]. Achievement of anatomic fixation on boththe femur and the tibia is difficult using a B-PT-B graftbecause of the mismatch between graft length and inser-tion site distance. The average distance between the intra-articular femoral and tibial tunnel holes has been reportedto be 26 (± 3) mm, while the average patellar tendonlength was 48 (± 6) mm [13]. The average length of thepatellar tendon was 45 mm in our series.

A surgical technique was developed to “shorten” theB-PT-B graft and achieve an anatomic fixation at theplateau level. A longer 35-mm bone block was harvestedfrom the tibia. A 10–15 mm cortical bone graft was re-moved from the tibial bone block and transposed to theundersurface of the tendon. An interference screw was ad-vanced in the tibial tunnel to engage both the bone blockand the cortical graft.

The effectiveness of this procedure was clinicallytested in a group of 60 ACL reconstructions with knees al-ternately allotted to the “anatomic” or “non-anatomic”fixation group. At an average follow-up of 18 months, ob-jective stability was comparable in the two groups. Therewere no significant differences in range of motion andsymptoms. However, the patients with anatomic fixationexperienced more tibial tuberosity pain while kneelingcompared with knees with non-anatomic fixation (53% vs17%, P = 0.01). This fact can be probably related to thelarger size of the tibial tuberosity defect. The only othersignificant difference between the two groups was the in-cidence of radiographic tibial tunnel modification, whichwas significantly (P = 0.02) higher in the knees with non-anatomic fixation (43%) compared with those with

anatomic fixation (23%). We were unable to documentany relationship between anterior tibial displacement andtibial tunnel modification as has been found by others[10]. Tibial tunnel modifications have been well docu-mented following allograft replacement of the ACL [3, 7,11]. Other studies have reported bone tunnel modifica-tions using an autogenous B-PT-B graft [3, 10, 12]. Theaetiology of this phenomenon is unknown and probablymultifactorial. Possible explanations for tunnel enlarge-ment include an immune response with osteolysis in allo-grafts, stress shielding proximal to the interference screw,an inflammatory response by the synovium within thetunnel, and resorption of the necrotic bone induced bytunnel drilling [10]. Motion of the graft within the tunnel(“windshield wiper effect”) may also induce bone resorp-tion [8]. The reduced incidence of tibial tunnel modifica-tions in the knees with anatomic fixation could be attrib-uted to the limited inflammatory response by the syn-ovium within the tunnel or to the decreased “windshieldwiper effect” due to tunnel occlusion.

It is not clear why anatomic fixation did not improveobjective stability. If anatomic fixation works because ofthe reduced length and increased stiffness of the graft, ourtechnique may have failed to effectively shorten the ten-don, which may slide between the cortical graft and thewall of the tunnel. Furthermore, shortening was only 11mm on average. The experimental studies [5] documenteda significant improvement in knee stability comparing fix-ation at the tibial plateau level and fixation outside the tib-ial tunnel, that is 3–4 cm distal to the plateau level. Otherprocedures such as bone-hamstring-bone composite auto-graft, the “flipped” B-PT-B graft and the quadriceps ten-don graft have been proposed [8] to achieve anatomic fix-ation on both femur and tibia. We have limited experiencewith the “flipped” patellar tendon graft. With an averagelength of the patellar tendon of 45 mm and with 30 mm asthe desired length, the length of the bone block to beflipped had to be 15 mm. Trimming of the bone block to15 mm long seemed to be excessive. For this reason wedeveloped the new technique described here.

S47

Table 3 Other clinical param-eters (IKDC form [4]) Normal Nearly normal Abnormal Severely abnormal

Group Group Group Group Group Group Group GroupA B A B A B A B

Patellar crepitation (absent) (mild) (moderate) (severe)25 21 5 8 – 1 – –

NS

Tibial tuberosity pain (absent) (mild) (moderate) (severe)(while kneeling) 14 25 11 3 5 2 – –

P = 0.01

“One-leg hop test” (≥ 90%) (89%–76%) (75%–50%) (< 50%)18 17 10 13 2 – – –

NS

Page 6: Anatomic versus non-anatomic tibial fixation in anterior cruciate ligament reconstruction with bone-patellar tendon-bone graft

In conclusion, improvement of objective stability in B-PT-B ACL reconstruction is attractive. Despite some ex-perimental evidence, the use of a short B-PT-B graft withanatomical fixation (at the tibial plateau level) did not im-prove stability in our prospective randomized study. Weused a tibial cortical graft to “shorten” the tendon and oc-

clude the tunnel. This seems to prevent tibial tunnel en-largement. However, harvesting a longer tibial bone blockcauses increased tibial tuberosity discomfort while kneel-ing. For these reasons and because of the increased tech-nical difficulty, this procedure cannot be recommended.

S48

1.Daniel DM, Stone ML (1990) KT-1000 anterior-posterior displacementmeasurements. In: Daniel D.M, et al(eds) Knee ligaments: structure, func-tion injury and repair. Raven Press,New York, pp 427–447

2.Daniel DM, Stone ML, Riehl B, MooreMR (1988) The one leg hop for dis-tance. Am J Knee Surg 1 :212–213

3.Fahey M, Indelicato PA (1994) Bonetunnel enlargement after anterior cruci-ate ligament replacement. Am J SportsMed 22 :410–414

4.Hefti F, Müller W, Jakob RP, StäubliH-U (1993) Evaluation of knee liga-ment injuries with the IKDC form.Knee Surg Sports Traumatol Arthrosc 1 :226–230

5. Ishibashi Y, Rudy TW, Kim HS, FuFH, Woo SL-Y (1995) The effect ofACL graft fixation level on knee stabil-ity. Arthroscopy 11 :373

6. Jackson DW, Gasser SI (1994) Tibialtunnel placement in ACL reconstruc-tion. Arthroscopy 10 :124–131

7.Linn RM, Fischer DA, Smith JP, et al(1993) Achilles tendon allograft recon-struction of the anterior cruciate liga-ment-deficient knee. Am J Sports Med21 :825–831

8.Morgan CD, Kalman VR, Grawl DM(1995) Isometry testing for anteriorcruciate ligament reconstruction revis-ited. Arthroscopy 11 :647–659

9.Noyes FR, Mangine RE, Barber S(1987) Early knee motion after openand arthroscopic anterior cruciate liga-ment reconstruction. Am J Sports Med15 :149–157

10.Peyrache MD, Djian P, Christel P,Witvoet J (1996) Tibial tunnel enlarge-ment after anterior cruciate ligamentreconstruction by autogenous bone-patellar tendon-bone graft. Knee SurgSports Traumatol Arthrosc 4 :2–8

11.Roberts TS, Drez D, McCarthy W, etal (1991) Anterior cruciate ligament re-construction using freeze-dried, ethyl-ene oxide-sterilized, bone-patellar ten-don-bone allografts: two year results inthirty-six patients. Am J Sports Med 19 :35–41

12.Schulte K, Majewski M, Irrgang JJ, FuFH, Harner CD (1995) Radiographictunnel changes following arthroscopicACL reconstruction autograft versusallograft. Arthroscopy 11 :372–373

13.Shaffer B, Gow W, Tibone JE (1993)Graft-tunnel mismatch in endoscopicanterior cruciate ligament reconstruc-tion: a new technique of intraarticularmeasurement and modified graft har-vesting. Arthroscopy 9 :633–646

14.Shelbourne KD, Nitz P (1990) Accel-erated rehabilitation after anterior cru-ciate ligament reconstruction. Am JSports Med 18 :292–300

References