Advertisement for orthosearch.org.uk
Results 1 - 14 of 14
Results per page:
Applied filters
Include Proceedings
Dates
Year From

Year To
Orthopaedic Proceedings
Vol. 93-B, Issue SUPP_III | Pages 357 - 357
1 Jul 2011
Starantzis K Lendi A Kondovazainitis P Koulalis D Mastrokalos D
Full Access

Evaluation of transtibial aiming of the femoral tunnel at its anatomical position in arthroscopical ACL reconstruction.

43 ACL reconstructions with hamstrings’ graft were studied. First, the femoral tunnel was drilled through the anteromedial portal at 09.30–10.00 (14.00–14.30 resp.) and then the tibial tunnel (av. anteroposterior angle: 63,5°, sagittal: 64,2°) at the same diameter with simoultaneous radiological documentation. Then, with a femoral aiming device, we tried to put a K-wire at the center of the drilled femoral tunnel. Fotographic documentation took place. In 20 cases the diameter of the tunnels was 7mm, in 11, 7,5mm, in 7, 8mm, in 3, 8,5mm and in 1, 9mm. Evaluation of all radiological and photographic material from 2 observers followed, according to the deviation of the transtibial K-wire from the center of the femoral tunnel.

38 ACL reconstructions were evaluated. It was shown that in 11 cases the transtibial K-wire was in the femoral tunnel (28,9%) (in 7 with a diameter of 7mm., in 2 with 7,5mm. and in 2 with 8mm.). The K-wire was in 23 cases (60,5%) at the perimeter or out of the femoral tunnel (in 11, with a diameter of 7mm., in 8 with 7,5mm., in 4 with 8mm., in 3 with 8,5mm. and in 1 with 9mm.). There was no correlation with the angles of the tibial tunnel or the age of the patients.

Transtibial aiming of the femoral tunnel at its anatomical position is very difficult and there is no correlation of the transtibial deviation with the diameter of the tibial tunnel.


Orthopaedic Proceedings
Vol. 93-B, Issue SUPP_III | Pages 357 - 358
1 Jul 2011
Kondovazainitis P Starantzis K Lendi A Koulalis D Mastrokalos D
Full Access

The goal of this prospective study was to evaluate the results of arthroscopic meniscal repair with the FasT-Fix repair system. Type of study: Prospective case series. Methods: 83 meniscal repairs with the FasT-Fix meniscal repair system in 80 patients with a mean age of 29 years were performed between 2004 and 2008. Concurrent anterior cruciate ligament (ACL) reconstruction was performed in 70% of the cases. All tears were longitudinal and located in the red/red or red/white zone. Criteria for clinical success included absence of joint line tenderness, locking, swelling, and a negative McMurray test. Clinical evaluation included also the Lyscholm knee score, and KT-1000 arthrometry. Results: The average follow-up was 38 months (range, 10–61 months). Six of 83 repaired menisci (7.23%) were considered failures according to our criteria. Therefore, the success rate was 92.77%. Time required for meniscal repair averaged 15 minutes. Postoperatively, the majority of the patients had no restrictions in sports activities. 92% had an excellent or good result according to the Lysholm knee score. Four patients had a restriction of knee joint motion postoperatively, and an arthroscopic arthrolysis was performed in one of them. Analysis showed that, age, length of tear, simultaneous ACL reconstruction, chronicity of injury, and location of tear did not affect the clinical outcome. Conclusions: Our results, shows that arthroscopic meniscal repair with the FasT-Fix repair system provides a high rate of meniscus healing and offers reduction of both the risk of serious neurovascular complications and operative time.


Orthopaedic Proceedings
Vol. 93-B, Issue SUPP_III | Pages 373 - 373
1 Jul 2011
Koulalis D Kendoff D Mustafa C Di Benedetto P Cranchi C Mastrokalos D Pearle A
Full Access

Measurement of precision in positioning multiple autologous osteochondral transplantation in comparison to the conventional free hand technique.

The articular surfaces of 6 cadaveric condyles (medial – lateral) were used. The knee was referenced by a navigation system (Praxim). The pins carrying the navigation detectors were positioned to the femur and to the tibia. The grafts were taken from the donor side (measurement I) with the special instrument which carried the navigation detectors. The recipient site was prepared and the donor osteochondral grafts were forwarded to the articular surface (II). The same procedure took place without navigation. The articular surface congruity was measured with the probe (measurement III)

The angle of the recipient plug removal (measurement I) with the navigation technique was 3,27° (SD 2,05°; 0°–9°). The conventional technique showed 10,73° (SD 4,96°; 2°–17°). For the recipient plug placement (measurement II) under navigated control a mean angle of 3,6° (SD 1,96°; 1°–9°) was shown, the conventional technique showed results with a mean angle of 10,6° (SD 4,41°; 3°–17°). The mean depth (measurements III) under navigated control was 0,25mm (SD 0,19mm; 0mm–0,6mm). With conventional technique the mean depth was 0,55mm (SD 0,28mm; 0,2mm –1,1mm).

The application of navigation showed that complications like diverging of the grafts leading to breakage or loosening as well as depth mismatch which can lead to grafts sitting over or under the articular surface can be avoided providing better results in comparison to the free hand procedure


Orthopaedic Proceedings
Vol. 92-B, Issue SUPP_II | Pages 324 - 324
1 May 2010
Mastrokalos D Koulalis D Zachos K Pyrovolou N Kontovazenitis P Lendi A Karaliotas G Sakellariou V Pandos P
Full Access

Purpose: The goal of this prospective study was to evaluate the results of arthroscopic meniscal repair with the FasT-Fix repair system.

Type of study: Prospective case series.

Methods: 83 meniscal repairs with the FasT-Fix meniscal repair system in 80 patients with a mean age of 29 years were performed between 2001 and 2004. Concurrent anterior cruciate ligament (ACL) reconstruction was performed in 70% of the cases. All tears were longitudinal and located in the red/red or red/white zone. Criteria for clinical success included absence of joint line tenderness, locking, swelling, and a negative McMurray test. Clinical evaluation included also the Lyscholm knee score, and KT-1000 arthrometry.

Results: The average follow-up was 38 months (range, 24–61 months). Six of 83 repaired menisci (7.23%) were considered failures according to our criteria. Therefore, the success rate was 92.77%. Time required for meniscal repair averaged 15 minutes. Postoperatively, the majority of the patients had no restrictions in sports activities. 92% had an excellent or good result according to the Lysholm knee score. Four patients had a restriction of knee joint motion postoperatively, and an arthroscopic arthrolysis was performed in one of them. Analysis showed that, age, length of tear, simultaneous ACL reconstruction, chronicity of injury, and location of tear did not affect the clinical outcome.

Conclusions: Our results, shows that arthroscopic meniscal repair with the FasT-Fix repair system provides a high rate of meniscus healing and offers reduction of both the risk of serious neurovascular complications and operative time.


Orthopaedic Proceedings
Vol. 91-B, Issue SUPP_I | Pages 143 - 143
1 Mar 2009
Koulalis D Schultz W Mastrokalos D Zachos K Karaliotas G Menelaou M Liberis I
Full Access

Aim of study: Comparison of clinical and radiological results of the applied methods.

Material and method: 20 patients with osteochondritis dissecans of the talus were treated. Autologous chondrocyte transplantation was applied to a group of 10 patients (Group A) and autologous osteochondral transplantation to the rest 10 patients (Group B).Group A: Average-age=30,8 years, -follow up time= 33,6 (12–48) months, -lesion size= 20×16,2 (35–15 × 25–15) mm and depth =7 (20–5) mmGroup B: Average –age =33, 7 years, follow up time =32,4 (12–48) months, – lesion size=16,5×15 (25–10 × 20–10) mm and depth=4 (5–3) mm.All patients underwent clinical und radiological investigation and the symptoms were classified in accordance with the Finsen classification. Group A was treated with autologous chondrocyte and Group B with autologous osteochondral transplantation. Osteotomy of the medial or lateral malleolus was necessary by 6 Group A – patients and 4 Group B – patients. Postoperative treatment : Non-weight bearing for 6 weeks, continuous passive motion of the joint, clinical and radiological follow up 3rd, 6th, 12th and yearly basis

Results: Postoperatively the average Finsen score showed for Group A an improvement from 3,5 (very bad) to 1,1 (excellent) and for Group B from 3,6 (very bad) to 1 (excellent). MRI follow up showed defect coverage by 7 patients of Group A and 9 patients of Group B after 12 months. Second look arthroscopy was performed by 4 patient of Group A and 3 patients of Group B showing full coverage of the defect site. Complications: persisting swelling for an average time of 2,7 months in Group A and 4,3 months in Group B. Additionally 10 degrees loss of joint extension in 4 patients of Group A and 3 patients of Group B.

Conclusion: Autologous chondrocyte transplantation as well as autologous osteochondral transplantation present very good clinical results. As methods of articular surface restoration they have their advantages and disadvantages playing an important role in chosing one of them, in combination always with the surgeons philosophy. The existence of these methods is important in influencing the course of the ankle joint towards a good prognosis, in cases of osteochondral lesion. Further investigation is necessary.


Orthopaedic Proceedings
Vol. 88-B, Issue SUPP_I | Pages 112 - 112
1 Mar 2006
Mastrokalos D Kotsovolos E Hantes E Paessler H
Full Access

Aim: To compare two arthroscopic all-inside methods of meniscal refixation (Fast-FixTM by Smith and Nephew and Clearfix screw by Innovasive Devices Inc.) in a prospective study.

Method: 85 patients (mean age 32.7 years) having 87 meniscal repairs (Group C: 27 with Clearfix screw and Group F: 60 with Fast-FixTM) were included in the study.

Ligament stabilizing procedures were done in 46 (54,1%) patients who had ACL deficient knees (18 reconstructions in Group C and 38 in Group F). Only longitudinal lesions in the red/red or red/white zone were repaired. Follow-up averaged 12.3 months with a range from 6 to 25 months. Only longitudinal lesions in the red/red or red/white zone were repaired. Patients were evaluated using clinical examination, the “OAK” knee evaluation scheme and Magnetic Resonance Imaging. Criteria for clinical success included absence of joint line tenderness, swelling and a negative McMurray test.

Results: 10 out of 87 repaired menisci (11.5%) were considered as failures according to the above mentioned criteria (3 in Group C (11,1%) and 7 in Group F (11,6%)). According to the “OAK” knee evaluation scheme 68 patients (80.%) had excellent or good result (Group C: 20 (80%), Group F: 48 (80%)). Magnetic resonance imaging, however, showed persisting grade III or IV lesions in 41 (47,1%) of 87 patients with successful result (Group C: 13 (47%), Group F: 28 (46,6%)). Postoperatively, we had 10 complications (11,3%) which were not directly associated with the meniscal repair device (Group C: 2 (3,7%), Group F: 7 (11,6%)).

Conclusion: Risk factors for failure of meniscus repair are chronicity of injury, location of tear more than 3 mm from the meniscosynovial junction and meniscus side (medial). At all events, both methods seem to be very promising because of their efficasy, safety and ease to use.


Orthopaedic Proceedings
Vol. 86-B, Issue SUPP_III | Pages 234 - 234
1 Mar 2004
Paessler H Rossis J Mastrokalos D Kotsovolos I
Full Access

Aim: To evaluate whether a guiding pin for a femoral tunnel could be positioned through the tibial tunnel into the center of the anatomical ACL attachment. Methods: 77 knees underwented arthroscopic ACL reconstruction with hamstrings. The femoral tunnel was drilled through an anteromedial portal at the center of the anatomic insertion at about 10.00 resp.14.00 position. Tibial tunnel (mean diameter 7.55 ± 0.54 mm) was drilled using a guide inserted at 90 degrees of knee flexion. Then, through the tibial tunnel, a 4mm offset femoral drill guide was positioned as close as possible to the femoral tunnel and a 2.5 mm guide wire was drilled. The position of the guide wire was photographed arthroscopically and the deviation was measured as the distance between the center of the femoral tunnel and the guide wire. Results: The mean deviation was 4.50 ± 1.54 mm (p = 0.00000004) In 74 knees (96.1%) the guidewire did not reach the femoral tunnel. Only in 3 knees it reached the superomedial edge of the femoral tunnel. No statistical relationship was found between deviation and tibial tunnel inclination angles or tibial tunnel diameter. Conclusions: Transtibial femoral tunnel drilling does not reach the anatomic site of the ACL insertion, even with larger tibial tunnels (for hamstring grafts up to 8.5 mm). Transtibial tunnel drilling should be replaced by drilling through the anteromedial portal at least for tunnels with diameters < 9 mm.


Orthopaedic Proceedings
Vol. 86-B, Issue SUPP_III | Pages 246 - 246
1 Mar 2004
Springer J Mastrokalos D Kilger R Paessler H
Full Access

Aim: Goal of this prospective, randomized study is the functional evaluation of two different techniques of ACL reconstruction by using bone-patellatendon (BPT) vs. hamstrings (ST/G). Methods: 62 ACL-insufficient patients (33 women/29 men) took part in this study. 31 (mean:29,8 y.) underwent ACL-reconstruction using BPT (GI). 31 (mean: 34,2 y.) patients underwent ACL-reconstruction using ST/G (GII). Both techniques were press-fit and implant-free. We used accelerated rehabilitation for both groups. Patients were evaluated by IKDC, Lysholm and Tegner score, KT 1000, one-leg-hop, isokinetics, internal torque, kneeling- and knee-walking-test, 1 day preop., and 3, 6 and 12 months postop. Results: One year results of GII were according to IKDC-score (GII: 30 patients= A and B vs. GI: 24 patients= A and B), Lysholm-score (GII: 95,61 vs. GI: 90,87 (p=0.017) and Tegner-score (GII: 7,07 vs. GI: 6,61 (p=0.00)) better than those of GI. The Evaluation of the strength of hamstrings using isokinetics showed significant differences: GII: 90,34 Nm vs. 99,19 Nm in GI, (p=0.008). However results concerning the internal torque evaluation were not significant. The one-leg-hop comparing injured and non-injured leg resulted in significant differences: GII: 96% vs. GI: 91%, (p=0.012). Results in GI were significantly worse than in GII at kneelling and kneewalking-testings ((p=0.00)(p=0.00)). Conclusion: All scoring, clinical and functional evaluations, except isokinetic hamstring evaluation, showed a hamstring’s supperiority in ACL reconstruction.


Orthopaedic Proceedings
Vol. 86-B, Issue SUPP_III | Pages 246 - 246
1 Mar 2004
Mastrokalos D Springer J Kotsovolos I Paessler H
Full Access

Aim: To evaluate activity level and remaining symptoms concerning the donor site morbidity in patients having anterior cruciate ligament reconstruction (ACLreconstruction) with either ipsilateral or contralateral bone-patellar-tendon graft (BPT-graft). Methods: 100 patients aged from 18 to 49 years (mean 34) having an ACL-reconstruction with BPT-graft (with one bone block) from 1997 to 1999 were included in this study. In 52 of them a BPT-graft from the ipsilateral side was used (Group I). In 48 the contralateral BPT was used (Group II). A questionaire, including Cincinnati-, Tegner-activity score and special questions concerning persisting symptoms at the donor site, such as tenderness, numbness, kneeling pain and knee-walking pain, was sent to all patients. Results: The average Cincinnati Scoring was 85,2 in Group I and 86,3 in Group II. There was no statistical significance in Tegner scoring between the two groups. In Group I, 37,5% of the patients refered local tenderness, 59,6% kneeling pain and 82,65% knee-walking pain. According to contralateral leg in Group II, 37,5% of the patients refered local tenderness, 52% kneeling pain and 62,5% knee-walking pain vs. 8,3%, 25% and 25% respectively in the ACL reconstructed knee. Conclusions: This study showed that there are no benefits if the contralateral BPT graft is used, because all symptoms concerning donor site morbidity are shifted from the injured into the healthy knee if the graft is taken from the contralateral side.


Orthopaedic Proceedings
Vol. 86-B, Issue SUPP_II | Pages 176 - 176
1 Feb 2004
Mastrokalos D Kotsovolos I Paessler H
Full Access

Aim: To compare the donor site morbidity after anterior cruciate ligament (ACL) reconstruction using ipsilateral and contralateral bone-patellar-tendon (BPT) autograft.

Patients and Method: 100 patients underwent ACL- reconstruction with autologous BPT graft between 1997 and 1999. ACL-reconstruction was performed using ipsilateral (n= 52) and contralateral (n= 48) BPT-autograft. The average follow-up was 39,2 months. Donor site morbidity was evaluated by using a questionaire, computerised historical data, KT-2000 measurments, Cincinnati- (CKS) and Tegner-Score.

Results: KT-2000 evaluation showed an average side to side difference of 0,6 mm in both groups. There were no significant differences between the two groups concerning CKS or Tegner-Score. In the ipsilateral donor knee, 59,6% of the patients had local tenderness, 69,2% kneeling-pain, 76,9% knee-walking-pain and 75% numbness. In the contralateral donor knee, 58,3% of the patients had local tenderness, 70,8% kneeling-pain, 70,8% knee-walking-pain and 85,4% numbness. In the ACL reconstructed knee 6,3% had local tenderness,6,3% kneeling-pain, 10,4% knee-walking-pain and 64,6% numbness.

There was one rupture of the patellar tendon and one patient with chronic patellar tendinitis.

Conclusion: The contralateral BPT graft appears to present no advantage over the ipsilateral as all symptoms concerning donor site morbidity are shifted from the injured into the healthy knee and return to activity is not more rapid.


Orthopaedic Proceedings
Vol. 86-B, Issue SUPP_II | Pages 181 - 181
1 Feb 2004
Kotsovolos I Mastrokalos D Kilger R Thermann H Paessler H
Full Access

Aim: Our aim was to evaluate the accuracy and reliability of both MRI and CT in estimating the patello-femoral alignement with the TT/TG (tibial tuberosity/trochlear groove) distance in 14 healthy probants.

Method: The TT/TG distance has been estimated in 28 healthy knees of 14 probants, 8 women and 6 men (age: 24 – 42) with a) MRI in an Esaote Arthroscan 0,2 Tesla Tomographer and b) CT in a General Electric Tomographer. This parameter was estimated in both imaging devices, first by overprojecting a tibial axial image through the tibial tuberosity onto an axial image through both femoral condyles and then by measuring the distance between the deepest point of the trochlear groove and the most prominent point of the TT. The parameter has been estimated twice by 3 well trained independent observers. The statistical evaluation was done with an unifactorial analysis of variance (ANOVA).

Results: Our results showed a good reproducibility (> 95%) of the TT/TG measurement in both methods: The intraobserver reliability was in CT, 0,008 ± 0,005mm and in MRI 0,03 ± 0,0017mm respectively. The interob-server reliability was 0,046 for the CT and 0,66 for the MRI. Interesting was that the average value of TT/TG by measuring with MRI (14 mm) was 3 mm less than the one measured with CT (17mm).

Conclusion: We concluded that in spite of the difference of the average values between MRI and CT the evaluation of the TT/TG parameter by means of MRI could be a good method for estimating this parameter thus avoiding radiation uptake.


Orthopaedic Proceedings
Vol. 85-B, Issue SUPP_III | Pages 230 - 230
1 Mar 2003
Mastrokalos D Springer J Rossis J Thermann H Paessler H
Full Access

Introduction: The goal of this prospective, randomized study is the functional evaluation of two different techniques of ACL reconstruction by using the bone-patella tendon (BPT) vs. hamstrings (ST/G).62 ACL-insufficient patients without any concommitent sport injuries took part in a prospective randomized study.

Material and Methods: Eighteen men and 13 women underwent ACL-reconstruction with BPT graft (Group I) with average age 29,87 (16–46) years. 31 patients (16 women and 15 men) with average age 34,23 (16–55) years underwent ACL-reconstruction using a quadrupled hamstrings graft (Group II). Both techniques were performed by using a press-fit and implant-free technique. We used the same accelerated rehabilitation protocol for both groups. The patients were evaluated by IKDC, Lysholm and Tegner scoring systems, KT 1000, one leg hop, isokinetics, internal torque, kneeling and knee walking test, 1 day preoperatively (VU), and 3 (NU I), 6 (NU II) and 12 months (NU III) postoperatively.

Results: One year postop.the results of Group II (30 patients classified as A and B) were according to IKDC scoring system better than those in Group I (Group I: 24 patients classified as A and B). We had similar results according to Lysholm-scoring evaluation (Group 11:95,61 points vs. 90,87 in Group I (p=0.017)) and Tegner-scoring (Group II: 7,07 vs. 6,61 in Group I (p=0.00)). According to the KT 1000 stability evaluation, there was no statistical significant difference between injured and uninjured knees in both groups. The evaluation of the strength of the hamstrings by isokinetics in both groups showed statistical significant differences (Group II: 90,34 Nm vs. 99,19 Nm in Group I, (p=0.008)). However, our results concerning internal torque evaluation were not statistical significant. The evaluation of one leg hop by comparing injured and non-injured leg showed a significant difference between group II and group I (Group II: 96% vs. 91% in Group I, (p=0.012)). We had worse results in Group I vs. Group II at kneelling and knee-walking-testings ((p=0.00)(p=0.00)), concerning the anterior knee pain.

Conclusion: From our results concerning IKDC, Lysholm, Tegner, kneeling and knee walking tests it seems that hamstrings can be recommended for ACL reconstruction. The isokinetic evaluation of hamstrings showed a statistical significant deficiency compaired to the BPT-group. But this result could not be confirmed with the internal torque evaluation and “one leg hop”-testing.


Orthopaedic Proceedings
Vol. 85-B, Issue SUPP_III | Pages 230 - 230
1 Mar 2003
Mastrokalos D Rossis J Jiakuo Y Paesssler H
Full Access

Introduction: The aim of this study was to evaluate whether a guiding pin for a femoral tunnel could be positioned through the tibial tunnel into the center of the anatomical ACL attachment.

Material and Methods: We studied 77 knees who underwented arthroscopic ACL reconstruction with hamstrings. The femoral tunnel was drilled through an antero-medial portal at the center of the anatomic insertion at about 10 resp. 2 o’clock position. Tibial tunnel (mean diameter 7.55 ± 0.54 mm ) was drilled at 90° of knee flexion. The aiming point was on a line, being a “prolongation” of the posterior border of the anterior horn of the lateral meniscus and at exactly 60% of the distance from the end of the anterior horn of the lateral meniscus and the medial tibial spine. Then, through the tibial tunnel, a 4mm offset femoral drill guide was positioned as close as possible to the femoral tunnel and a 2.5 mm guide wire was drilled. The position of the guide wire was photographed arthroscopically and the deviation was measured as distance between the center of the femoral tunnel and guide wire.

Results: The mean angle of the tibial tunnel in the coronar plain was 27,53° and in the sagittal plain 25,84°, both according to the longitudinal axis of the tibia. In 74 knees ( 96. 1 % ) the guidewire did not reach the femoral tunnel. Only in 3 knees it reached the superomedial edge of the femoral tunnel. The mean deviation was 4.50 ± 1.54 mm (p = 0.00000004 ). No statistical relationship was found between deviation and tibial tunnel inclination angles or tibial tunnel diameter.

Conclusion: Transtibial femoral tunnel drilling did not reach the anatomic side of the ACL insertion in most of the cases, even with larger tibial tunnels (for hamstring grafts up to 8.5 mm). Therefore we recommend tibial tunnel drilling through the anteromedial portal.


Orthopaedic Proceedings
Vol. 85-B, Issue SUPP_III | Pages 230 - 230
1 Mar 2003
Mastrokalos D Rossis J Yu J Thermann H Paessler H
Full Access

Aim of the study: To evaluate whether additional aperture fixation to distal fixation (two-point fixation or 2P) may diminish tibial tunnel enlargement compared to distal fixation alone (one-point fixation or1P).

Methods: Two groups of patients were evaluated. In group 1P (44pts.) the grafts were fixed on the tibial side, using two 4 mm Mersilene tapes tied over a 10 mm bone bridge created in the anterior cortex. In group 2P, an additional fixation, using 1 absorbable cross pin in 27 pts. and 2 absorbable cross pins (Rigid fix, Mitek) in 24 pts., was performed. AP and lateral radiographs as well as patients evaluation were performed 3, 6 and 12 months post-op. Tibial tunnel size was measured at the widest diameter in both AP and lateral radiographs.

Results: Tunnel enlargement occurred in both groups at 3 months post-op, but not thereafter. In the lateral radiographs, tunnel enlargement was significantly less in group 2P compared to group 1P (p< 0.05). No statistical relationship was found between tunnel enlargement and gender, age, IKDC, and KT 1000 side-to-side difference.

Conclusion: Additional tibial graft fixation with cross-pins seems to diminish tibial tunnel enlargement in the sagital plane. This may be explained by the fact that the cross pins inserted in the coronal plane, reduce tibial graft movements mainly in the sagital plane. The study supports the hypothesis that tibial graft micro-movements during the period of tendon healing to bone (up to 3m) play a role in tibial tunnel enlargement after ACL reconstruction using hamstrings.