Functional outcome after transphyseal anterior cruciate ligament reconstruction in young patients with open growth plates

THEKNE-02323; No of Pages 12 The Knee xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

The Knee

Functional outcome after transphyseal anterior cruciate ligament reconstruction in young patients with open growth plates C. Holwein a,⁎, S. Hinterwimmer a,b, H.O. Mayr c, L. Lämmle d, P.U. Brucker a,b, E.O. Münch e, A.B. Imhoff a a

Abteilung für Sportorthopädie, Klinikum Rechts der Isar, Technische Universität München, Ismaninger Str. 22, 81675 München, Germany OrthoPlus München, Alte Börse Lenbachplatz 2a, 80333 München, Germany c Universitätsklinikum Freiburg, Klinik für Orthopädie und Unfallchirurgie, Muskuloskelettales Forschungslabor Hugstetter Straße 55, 79106 Freiburg, Germany d Department Psychologie, Medical School Hamburg, Kaiserkai 1, 20457 Hamburg, Germany e OCM Klinik, Orthopädische Chirurgie München, Steiner Str. 6, 81369 München, Germany b

a r t i c l e

i n f o

Article history: Received 16 December 2015 Received in revised form 23 August 2016 Accepted 1 September 2016 Available online xxxx Keywords: ACL Children Transphyseal Rotational laxity Return to sport

a b s t r a c t Background: This study evaluates sports ability, rotational laxity and potential growth changes in children after transphyseal ACL reconstruction with metaphyseal fixation technique, considering physis biology by placing drill holes vertically in the femoral anatomic origin in order to reduce volumetric injury to the physis. Methods: In this retrospective trial of 42 patients data were collected. Thirty-seven were reviewed measuring rotational laxity and anteroposterior tibial translation using the Laxitester (ORTEMA Sport Protection, Markgroeningen, Germany) and the KT1000. Clinical examination was evaluated with the IKDC 2000 knee examination form. Leg axis was determined with digital photography and leg length was assessed clinically. Sports ability was assessed with questionnaires including subjective IKDC, Tegner Activity Scale, Activity Rating Scale and a questionnaire on sports and level of sports. Results: Mean follow-up was 24.9 months. Mean age at surgery was 13.2 years in boys and 13.1 years in girls. IKDC 2000 grading was A or B in 28 patients and C in nine patients. Significant increased anterior tibial translation was observed in neutral position and in external tibia rotation. No growth abnormalities were seen. Fifty-seven percent of the patients were able to participate in competitive sports at follow-up. Conclusion: Transphyseal ACL reconstruction with metaphyseal fixation in children with open growth plates can be done with low risk of growth changes. Return to competitive sports is possible although low rotational laxity still exists. Level of evidence: IV. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Intraligamentous anterior cruciate ligament (ACL) tears cause a dilemma for the patient as well as for the physician: On the one hand, a delay of surgical treatment until physeal closure includes the risk of eventual prearthrotic damage caused by a

⁎ Corresponding author. E-mail address: [email protected] (C. Holwein).URL: http://www.sportortho.de (C. Holwein).

http://dx.doi.org/10.1016/j.knee.2016.09.001 0968-0160/© 2016 Elsevier B.V. All rights reserved.

Please cite this article as: Holwein C, et al, Functional outcome after transphyseal anterior cruciate ligament reconstruction in young patients with open growth plates, Knee (2016), http://dx.doi.org/10.1016/j.knee.2016.09.001

2

C. Holwein et al. / The Knee xxx (2016) xxx–xxx

persisting unstable knee. In addition to reported meniscal and cartilage injuries, patients treated conservatively, usually are not able to return to their preinjury activity levels [1–5]. On the other hand, contemporary ACL reconstruction can lead to growth disturbances by harming the open growth plates with both growth arrest as well as overgrowth [6–9]. Nevertheless, recent studies underline the low risk of growth changes following transphyseal ACL reconstruction [10–13] and report only on five of 43 immature patients demonstrating a magnetic resonance imaging (MRI)-based focal physeal disruption which did not cause clinically apparent growth disturbance [14]. For minimizing the risk of growth disturbances some authors recommend reducing volumetric damage to the physis by placing vertical drill holes across the growth plates especially at the femur while performing a transphyseal ACL reconstruction [15]. But how such a vertical placed tunnel at the femur influences the rotational laxity in children has not been described yet [16]. This led us to the following questions to be answered: Is a transphyseal ACL reconstruction able to restore a stable knee, especially in rotation of the lower limb, comparable to the healthy knee? Does a transphyseal reconstruction with metaphyseal fixation technique lead to growth abnormalities? Can children return to their preinjury sporting level after transphyseal ACL reconstruction? In our current retrospective study we are describing the functional outcome of transphyseal ACL reconstruction in skeletally immature patients with hamstring grafts and metaphyseal fixation. We hypothesised that the affected knee would show equal stability, no growth disturbances, and patients would have good results in return to sport.

2. Methods This study was authorised by the responsible ethics committee, reference code: 69/15s.

2.1. Patients' data and selection Between October 2006 and April 2010 in total 42 (28 males, 14 females) young patients with radiological open physes received transphyseal ACL reconstruction in a single-bundle technique with hamstring grafts and metaphyseal fixation. Physeal growth remaining was identified according to Prince et al. by classifying the cartilage sign of the physis in the T2 weighted magnetic resonance imaging (MRI). Three patients had a rerupture, each because of a new distorsional trauma. One patient suffered a new torsional trauma while playing handball at nine months after initial reconstruction. One patient got reinjured after seven months while playing burning ball (game similar to baseball). One patient got reinjured in school sports. Two patients with ACL reconstruction on the contralateral knee could not be included in the survey. So 37 patients were retrospectively examined. From these 20 patients showed cartilage signal intensity present throughout the physis as complete open physes and 17 discontinuous cartilage signal intensity as partially open physes [17].

2.2. Surgical data and technique All patients underwent full arthroscopic surgical treatment with a single-bundle reconstruction and metaphyseal fixation by highly specialised knee orthopaedic surgeons. Additional medial meniscus surgery was needed eight times, at the lateral meniscus 11 times and one time both. Meniscal surgery included seven refixations and 14 partial resections. As transplants hamstring tendon autografts were used. Femoral drill holes were placed anatomically closely to the border of the posterolateral bundle insertion at the anteromedial bundle insertion site to gain more off-set to the perichondral structures of the physes [18,19] (Figure 10). Both tunnels were drilled more vertically to reduce volumetric damage to the physes [15]. Femoral fixation was performed metaphyseally with Endobutton (Smith & Nephew, Andover, MA; n = 21), Rigidfix pins (Johnson & Johnson Medical GmbH, DePuy Mitek Division, Norderstedt, Germany); n = 17) or bioabsorbable interference screws (Arthrex, Naples, FL, USA; n = 4). In the four cases of femoral screw fixation, the screws were placed beyond the growth plate, so that no hardware was located at the level of the physis. At the tibia, the drillholes were positioned more vertically. The tibial fixation of the graft was achieved using a bioabsorbable interference screw below the physis (n = 42). Overall, the aim was to minimise the cross section of the drillholes at the level of the physis and to place the femoral and tibial tunnel apertures anatomically.

2.3. Clinical scoring At the time of follow-up, the patients were asked for their overall satisfaction and if they would undergo surgery again. The pain was quantified with the Visual Analogue Scale (VAS) [20]. In addition to the International Knee Documentation Committee (IKDC) subjective score [21], the Lysholm Score [22] and the Tegner Activity Scale [23], two more sports specific questionnaires were utilised: the Activity Rating Scale by Marx et al. [24] and the Sports and Activity Questionnaire [25] which evaluates the different disciplines of sports and the level of sports activity (high end level, superior level, competitive level, recreational level, no considerable activity). Please cite this article as: Holwein C, et al, Functional outcome after transphyseal anterior cruciate ligament reconstruction in young patients with open growth plates, Knee (2016), http://dx.doi.org/10.1016/j.knee.2016.09.001

C. Holwein et al. / The Knee xxx (2016) xxx–xxx

3

2.4. Clinical examination Knee examination findings were quantified using the IKDC 2000 form [26]. Static knee joint laxity was assessed in the anterior direction using the KT1000 arthrometer (MEDmetric, San Diego, CA) in neutral position. In addition, anterior laxity in a rotated position of the tibia was evaluated in a standardised internal and external rotations using a validated instrument (Laxitester, Ortema Sport Protection, Markgröningen, Germany), details in Appendix A [16].

2.5. Biological growth evaluation Body height, leg length, and leg axis were measured clinically. In detail, the leg length was measured clinically from the anterior superior iliac spine to the medial malleolus [27]. The leg axis was performed according to Schmitt et al. [28]. Therefore, standardised digital photographs of both legs were performed. The leg axis was declared as the angle between the midpoint of the widest width femorally, the centre of the knee, and the centre of ankle. The final value of the leg axis was averaged after three measurements.

2.6. Statistical methods The Statistical Package for Social Sciences (SPSS, version 21, SPSS Inc., Chicago, IL) was used for statistical analysis. The effect sizes were ruled out with G*Power Version 3.1.3 (©Franz Faul, 2006; values b 0.5 small effect, medium effect between 0.5 and 0.8, strong effect N 0.8). The descriptive statistics were calculated as mean, range, and standard deviation (SD). Normal deviation was assessed with the Kolmogorov–Smirnov-Test. In case of normal distribution Pearson's correlation coefficient was calculated; this was the case for ‘follow-up time’ and ‘objective laxity’ as well as for ‘follow-up time’ and ‘return to preinjury sports level’. Correlation between bone maturity according to Prince's growth plate signal [17], growth, leg length difference and axial deviation were calculated with Kendall's Tau-b as normal distribution was not given. For comparison of the clinical scores (VAS, Tegner, Activity Rating Scale, Sports and Activity Questionnaire), KT1000 measurements, leg length, and leg axis the paired t-test was used. For comparing outcomes depending on additional meniscal surgery and femoral fixation the Mann–Whitney-U-Test and the Kruskal–Wallis-Test were used because normal deviation was not given for these variables. The statistical significance was set at p b 0.05.

3. Results 3.1. Patient demographics Mean follow-up was 24.9 months (range 9.3 to 50.1, SD ± 11.4). Mean age at surgery was 13.2 years (range 9.8 to 15.9, SD ± 1.6) in boys and 13.1 years (range 11.0 to 14.6, SD ± 1.0) in girls. Time gap between injury and surgery was 4.2 months in mean (range 0.9 to 20.6, SD ± 4.2).

3.2. Surgical data results Analysing postoperative radiographs in the sagittal plane (35/37 available), femoral drillholes were placed in the first (19) and in the second quarter (16) drawn below the Blumensaat's line from posterior to anterior, using the ACL-Ruler [29]. Tibial graft positioning was completed in the second quadrant from anterior to posterior of the lateral tibia plateau in 33 cases and just into the third quadrant in two cases [30].

Table 1 Overview subjective score data results. Subjective score data results

Mean

Min

Max

SD

Significance

Effect size

VAS preoperative VAS at follow-up IKDC subjective at follow-up Lysholm at follow-up Tegner preinjury Tegner at follow-up ARS preinjury ARS at follow-up sport hours per week preinjury sport hours per week at follow-up

4.9 0.7 92.9 93.9 7.8 7.2 12.1 11.4 6.2 5.8

0 0 73.6 72.0 2.0 2.0 2.0 0 0 0

9 6 100.0 100.0 10.0 10.0 16.0 16.0 18.0 17.5

2.7 1.4 7.7 7.7 1.6 1.9 3.3 4.3 3.9 4.0

p b 0.05

dz = 0.18

p b 0.05

dz = 0.49

p = 0.266

dz = 0.18

p b 0.54

dz = 0.10

Please cite this article as: Holwein C, et al, Functional outcome after transphyseal anterior cruciate ligament reconstruction in young patients with open growth plates, Knee (2016), http://dx.doi.org/10.1016/j.knee.2016.09.001

4

C. Holwein et al. / The Knee xxx (2016) xxx–xxx

Figure 1. Tegner activity score. Number of patients in the different Tegner activity levels at preinjury time and at follow-up.

3.3. Score data results All patients were satisfied with the surgical outcome (65% deeply contented, 35% contented). Except one patient every other patient would undergo the same procedure again. Subjective score data results including IKDC, Lysholm and Activity Rating Scale are shown in Table 1. There was a significant pain reduction in the Visual Analogue Scale from preoperatively to the time at follow-up (p b 0.05, effect size 1.2). Tegner Activity Score showed a slight reduction from 7.8 (range 2.0 to 10.0, SD ± 1.6) before the injury to 7.2 (range 2.0 to 10.0, SD ± 1.9) at follow-up (p b 0.05, effect size 0.5) (Figure 1). An overview of the sporting level over the course of time is shown in Figure 2. At follow-up 35 patients indicated that their sports ability at least remained the same, although only 27 reported that they could return to their preinjury level. Mean follow-up from patients who at least returned to their preinjury level was 27.6 (range 9.3 to 50.1, SD ± 11.2) and from those who did not 17.7 (range 10.5 to 32.9, SD ± 7.3). A slight reduction was seen concerning patients' time passed in doing sports, reaching on average 5.8 h per week (range 0 to 17.5, SD ± 4.0), 57% were able to join at least competitive sports. Before injury, time of doing sports per week was 6.2 h (range 0 to 18, SD ± 3.9) with

Figure 2. Sporting level over the course of time. Number of patients in the different sporting levels in their life time, one year before surgery and at follow-up.

Figure 3. Type of sports over the course of time. The number of patients and their type of sports over the course of time. At follow-up pivoting sports like soccer and skiing are still possible, although a slight reduction is seen.

Please cite this article as: Holwein C, et al, Functional outcome after transphyseal anterior cruciate ligament reconstruction in young patients with open growth plates, Knee (2016), http://dx.doi.org/10.1016/j.knee.2016.09.001

C. Holwein et al. / The Knee xxx (2016) xxx–xxx

5

Table 2 KT1000 measurements comparing operated and contralateral leg. KT1000 measurements [mm]

Mean

Min

Max

SD

Significance

Effect size

ATT neutral operated leg 30 lb ATT neutral contralateral 30 lb ATT neutral operated leg max ATT neutral contralateral max ATT internal rotation operated leg 30 lb ATT internal rotation contralateral 30 lb ATT internal rotation operated leg max ATT internal rotation contralateral max ATT external rotation operated leg 30 lb ATT external rotation contralateral 30 lb ATT external rotation operated leg max ATT external rotation contralateral max

3.8 2.6 5.4 3.7 4.4 3.7 5.6 4.4 3.8 3.0 5.5 4.1

1.3 0.7 1.9 0.8 1.0 0.7 1.6 1.0 1.5 1.0 2.1 1.1

7.7 5.5 9.4 6.3 8.2 8.9 9.6 9.5 7.3 5.9 9.3 7.8

1.6 1.0 1.8 1.3 1.6 1.7 1.8 1.8 1.5 1.2 1.8 1.5

p b 0.05

dz = 0.74

p b 0.05

dz = 0.89

p = 0.075

dz = 0.31

p b 0.05

dz = 0.50

p b 0.05

dz = 0.47

p b 0.05

dz = 0.59

70% at least in competitive sports. However time of doing sports before trauma and at follow-up was not significantly reduced (p = 0.542, effect size 0.1). A significant correlation (r = 0.37, p b 0.05) between return to preinjury sport level and follow-up time could be found. Patients with a longer follow-up were rather likely to return to their preinjury sporting level. However a correlation between follow-up time and final IKDC 2000 grading could not be found (r = 0.03, p = 0.85). An extraction of exercised type of sports is shown in Figure 3. High demand sports such as soccer and skiing are frequently exercised at follow-up, e.g., 28 patients played soccer one year before surgery and 23 patients at follow-up. 3.4. Clinical examination results IKDC 2000 objective was grade A in five patients, grade B in 23 and grade C in nine patients. Lachman was normal in 31 cases and nearly normal in six cases. Pivot shift test was negative in 20 patients, 14 showed a gliding and in three patients a clunk was noticeable. KT1000 measurements brought up the following results (Table 2 and Figure 4). There was a significant difference between anterior tibial translation with 30 lb of the affected and the contralateral knee in neutral position and in standardised external rotation. Values for anterior tibial translation (ATT) in standardised internal rotation showed no significant difference. There were only four patients reaching a side to side difference greater than three millimetres in neutral position, with the highest difference being five millimetres. This patient was seen for another control one year after the first follow-up and three years after reconstruction with then showing a side to side difference in neutral position of only 2.6 mm. A side to side difference in standardised internal rotation higher than three millimetres was seen in eight patients with the highest being five millimetres performed by the same patient mentioned previously. This one also showed the highest side to side difference in external rotation. Like in neutral position the findings also decreased one year later. In external rotation only one other patient showed side to side differences in exceeding three millimetres. No patient was graded C because of a static laxity finding. Additional values of ATT with an applied manual maximum force are also shown in Table 2. The degree of documented internal rotation (IRO) showed no significant difference (p = 0.186, effect size 0.2). It was 26.7° in mean (range 10.0 to 45.0, SD ± 9.0) for the affected knee and 25.6° (range 10.0 to 45.0, SD ± 8.1) for the contralateral knee.

Figure 4. KT1000 — anterior tibial translation with 30 lb. Anterior tibial translation (ATT) in neutral position (NT) and in external rotation showed a significant difference comparing the operated and the healthy leg (p b 0.05).

Please cite this article as: Holwein C, et al, Functional outcome after transphyseal anterior cruciate ligament reconstruction in young patients with open growth plates, Knee (2016), http://dx.doi.org/10.1016/j.knee.2016.09.001

6

C. Holwein et al. / The Knee xxx (2016) xxx–xxx

Figure 5. Leg length at follow-up comparing both sides. Leg length comparing both sides showed high correlation (r = 0.997).

However documented external rotation (ERO) was significantly higher in the operated knee and reached 43.9° (range 10.0 to 65.0, SD ± 10.4) and 40.0° (range 15.0 to 60.0, SD ± 12.7) in the healthy knee (p b 0.05, effect size 0.4). 3.5. Biological growth results Patients' body height at reconstruction increased on average 6.7 cm (range 0 to 23, SD ± 6.4) until follow-up. Leg length and axis showed high correlations for the values between both sides (r = 0.997 and r = 0.879) (Figures 5–6). Leg lengths on both lower extremities were 88.8 cm (range 73.5 to 107.5, SD ± 7.1) vs. 89.0 cm (range 74 to 108.5, SD ± 7.2). A discrepancy of more than one centimetre (M −0.2, range −1 to 0.5 cm, SD ± 0.48) could not be found. The operated leg showed 1.2° valgus alignment in mean (range −8.3 to 8.0, SD ± 3.3), the other side 0.5° (range −8.8 to 6.8, SD ± 3.5). There was no valgus or varus deformity more than three degrees in 34 patients. Three patients showed varus (1) or valgus (2) malalignment between three degrees and 4.5°. There was a significant moderate to high correlation between growth and preoperative growth plate signal on MRI (τb = −0.57). Correlations between growth or growth plate signal with leg length difference (τb = 0.01 and τb = − 0.18) and axial deviation (τb = 0.11 and τb = − 0.15) were small and not significant. 3.6. Outcome results depending on additional meniscal surgery or femoral fixation Concerning outcome results in ‘subjective IKDC’, ‘Lysholm’, ‘Tegner Activity Score’, ‘Activity Rating Scale’, ‘sporting level’, ‘sport hours per week’ and KT1000 measurements no significant differences between patients who additionally received meniscal surgery or not could be found (p b 0.05). The effect size was low too (dz ranging from 0.01 to 0.3).

Figure 6. Leg axis at follow-up comparing both sides. Leg axis comparing both sides showed high correlation (r = 0.879).

Please cite this article as: Holwein C, et al, Functional outcome after transphyseal anterior cruciate ligament reconstruction in young patients with open growth plates, Knee (2016), http://dx.doi.org/10.1016/j.knee.2016.09.001

C. Holwein et al. / The Knee xxx (2016) xxx–xxx

7

Figure 7. KT1000 — side to side differences in ATT with 30 lb in NT between femoral fixation groups. Side to side differences in anterior tibial translation (ATT) with 30 lb in neutral position (NT). A medium effect (dz = 0.77) could be seen for higher values in the screw group comparing to the button group.

The different femoral fixation groups (button n = 19, pin n = 14 and screw n = 4) had almost the same follow-up (button 22.1 months ± 12.2; pin 28.4 months ± 10.8 and screw 26.0 months ± 6.6). Follow-up time in the button group was a little bit shorter than in the pin group (p = 0.07; dz = 0.57). Although patients with screw fixation showed lower values in ‘subjective IKDC’ 89.9 ± 3.3 than patients in the button 93.4 ± 7.8 (p = 0.12; dz = 0.63) or pin group 93.3 ± 8.8 (p = 0.23; dz = 0.78) they spend more hours per week doing sports 8.3 h ± 2.0 than patients in the other groups (button 6.7 h ± 4.3 (p = 0.22; dz = 0.51) and pin 4.8 h ± 3.0 (p b 0.05; dz = 1.42)). Sport level was a little lower in the screw group (competitive level) compared to the pin group (superior level − competitive level) (p = 0.5; dz = 0.58). Values for ATT side to side differences in neutral position and internal rotation did not reach statistical significance comparing the groups (Figures 7 and 8) but ATT in external rotation was significantly higher in the pin group compared with the button group (p b 0.05; dz = 1.0) (Figure 9). Standardised internal or external rotation values were not significantly different between the groups. Only two medium effect sizes could be detected. One comparing side to side differences in internal rotation between pin group and screw group (p = 0.65; dz = 0.5) and the other comparing the external rotation between button group and pin group (p = 0.22; dz = 0.53). The operated legs in the screw group tended to be a little shorter −0.4 cm ± 0.2 than in the button group −0.1 cm ± 0.5 (p = 0.16; dz = 0.88). Operated legs compared with contralateral in the screw fixation group were more valgus like 1.9° ± 1.5 comparing to the button group 0.5° ± 1.7 (p = 0.37; dz = 0.88) or pin group 0.8° ± 1.6 (p = 0.44; dz = 0.73).

Figure 8. KT1000 — side to side differences in ATT with 30 lb in IRO between femoral fixation groups. Side to side differences in anterior tibial translation (ATT) with 30 lb in internal rotation (IRO). No effects (dz = 0) between each group could be found.

Please cite this article as: Holwein C, et al, Functional outcome after transphyseal anterior cruciate ligament reconstruction in young patients with open growth plates, Knee (2016), http://dx.doi.org/10.1016/j.knee.2016.09.001

8

C. Holwein et al. / The Knee xxx (2016) xxx–xxx

Figure 9. KT1000 — side to side differences in ATT with 30 lb in ERO between femoral fixation groups. Side to side differences in anterior tibial translation (ATT) with 30 lb in internal rotation (ERO). There were high and medium effects in that values for the pin group were higher than in the button (dz = 1.03) or in the screw group (dz = 0.69). Values for the button group were lower than in the screw group (dz = 0.52).

4. Discussion 4.1. Discussion of patients' data In our patient population time gap between injury and surgery is long. A reason might be the unawareness of the possibility to reconstruct the ACL in skeletally immature patients. Our patients often reported that it takes a long time to find a physician to treat their injury. In fact in our retrospective study design we could not determine skeletal age according to the Tanner–Whitehouse staging [31]. However we built our skeletally immature patients' selection on the apparent cartilage sign in T2 weighted preoperative MRI according to Prince et al. and therefore on a potential growth remaining in an active physis [17]. The reported correlation between growth and growth plate signal on MRI confirmed quality of this parameter. Whether the high number of meniscal injuries, 21 out of 42, results from the delay or is caused by initial trauma, remains unclear. Nevertheless Mizuta et al., Lawrence et al. and Millett et al. also found an increase of meniscal injuries in delaying surgery or conservative treatment [1–3]. In a cohort of 130 patients A.F. Anderson and C.N. Anderson even showed not only meniscal injuries but also further cartilage damage with increased time to surgery. With episodes of instability this risk was higher [5]. Only Woods and O'Connor found no higher occurrence in secondary damage in a group in which ACL reconstruction was delayed until physeal closure compared to matured adolescent groups with a simular delay. The sticking point was a strictly conservative treatment with bracing and waiver of participating in team sports. For that a high compliance was necessary [32]. As far as we could recognise in our cases, this way of therapy does not go along with patients' sport expectations. 4.2. Discussion of score results Pain reduction and good subjective score results are comparable to the recent published values, for example with Hui et al. where Lysholm Score reached 97 points and IKDC subjective was 96.0. Here a group of 16 prepubescent patients were reviewed with a mean follow-up of 25 months [12]. Similar results were published by Cohen et al., in a cohort of 26 patients they found a mean IKDC subjective of 91.5 ± 5.7 and an average Lysholm score of 93.5 ± 4. In this population also three cases of traumatic reruptures were mentioned [33]. Concerning sports activity and ability we could see a discrepancy between what the patients think they could do and what they are really doing. A reason for that might be the significant shorter follow-up time for those who did not return to their preinjury level. In contrast to our results Moksnes et al. who performed a conservative treatment with muscle strengthening until physeal closure reported a significant reduction of patients' activity level in a two year follow-up [4]. In 2011 in a meta-analysis Ardern et al. found out that after 12 months only 33% of the observed ACL reconstructed patients, including all ages, returned to competitive sports and another 33% had only attempted training and/or modified competition [34]. They summarised values from other studies amounting to 64% returning to competitive sports. After our follow-up comparably 57% of our young patients returned to competitive sports. Long ago Smith et al. reported 81% having returned to competition within 12 months in a review of 109 patients aged 21 in mean [35]. Feller and Webster summarised that there are many factors for return to preinjury sporting level or not [36]. Of course surgical details are not far to seek, but other reasons like rehabilitation, social and psychological factors should be taking into account too. Precisely young patients are exposed to changes in their hobbies and social activities. This could be a reason for not returning to Please cite this article as: Holwein C, et al, Functional outcome after transphyseal anterior cruciate ligament reconstruction in young patients with open growth plates, Knee (2016), http://dx.doi.org/10.1016/j.knee.2016.09.001

C. Holwein et al. / The Knee xxx (2016) xxx–xxx

9

preinjury sports. Nevertheless the fear of a reinjury should not be underestimated in the young. A different follow-up time might reduce our amount too, because some patients mentioned the intention to return soon to their seasonal sport activity depending on the actual season of the year in which the date of review fell. Especially for patients with a short follow-up this might be a reason for not returning to preinjury level at follow-up. Finally we can say that not all patients returned to their preinjury activity level. However some are thinking that they could. Nevertheless high level sports activity like soccer and skiing are still possible. 4.3. Discussion of clinical examination results In the IKDC 2000 form nine patients were graded C. Three were graded C because they complained of hyp- and dysesthesia at the donor site. Further along, two of them had a lower hyperextension compared to the contralateral knee. And three of these nine had really an abnormal pivot shift test and one a greater external rotation in the operated knee comparing both sides. Pivot shift test findings indicate that a residual dynamic laxity in some patients exists. But our further clinical experience shows us that this residual dynamic laxity decreases with skeletal maturation which has to be proven by further studies with a longer follow-up. This fact could be caused by a more prestressed graft during extremity growth. Biomechanical properties of graft tendons during growth might depend on viscoelasticity and on ligamentisation [37–39]. Bollen et al. found a graft lengthening without a diminished diameter in five patients who grew on average 17 cm [40]. However longer does not mean more strength. Like in our study their clinical experiences also showed more strength graft and decreased knee laxity with skeletal maturation. A prospective clinical trial measuring ATT over the time until physeal closure could solve this unanswered question. Lachman test findings also brought up residual static laxity findings. However our results are comparable to those documented in adults with an even longer follow-up. Janssen et al. found a mean anterior tibial translation of 2.3 mm ± 2.9 in mean in a cohort of 100 adult patients after 10 years of follow-up [41]. There was a significant side to side difference in ATT in neutral position and in external rotation. But there were only eight patients reaching a side to side difference greater than three millimetres in KT1000 measurements and no patient was graded C because of a static laxity finding. Unfortunately other studies documented the side to side difference with only an applied manual maximum force with the KT1000 which is unusual [11,12]. For a better comparison both the standard force of 134 N (or 30 lb) and the manual maximum force as recommended in the IKDC Evaluation Form should be applied. For that a comparison with these similar studies is difficult. However Streich et al. found comparable values for ATT with 1.8 mm ± 1.4 in neutral position, and also a similar range from −1 mm to five millimetres [42]. Like Mayr et al. we could not find a significant deviation for internal rotation with a standard applied force on the operated side compared with the unaffected knee [16]. The significant higher external rotation in the operated knee might be a sign for the femoral vertically tunnel drilling across the physis. Although the tunnel entrance is located anatomically, the fibres of the graft are orientated in a straight way, when tunnels are drilled more vertically. Other reasons like collateral ligament laxity or meniscal damages were not remarkable in patients with increased external rotation. However the ACL is the primary restraint not only against ATT but also against external tibial rotation under the femur. So our results show that a transphyseal ACL reconstruction with metaphyseal fixation does not completely restore a stable knee compared with the healthy knee. A small dynamic or static laxity in the affected knees still exists. 4.4. Discussion of growth results We intended to show that there would be no growth disturbances with a transphyseal reconstruction technique. In detail a leg length difference of more than one centimetre could not be found. Three patients showed varus (1) or valgus (2) malalignment between three degrees and 4.5°. So we could demonstrate that with a mean growth of 6.7 cm ± 6.4 cm neither leg length differences nor excessive axial deviation occurred. Until now the reported number of growth changes is low and often caused by surgical deficits including hardware across the physes [6–8,14,43,44] although a physeal disruption in the MRI sometimes can be seen [14]. For example Kocher et al. reported six cases of distal femoral valgus deformity because of a bony bar caused by fixation implants or the bone plug of a patellar tendon graft. A femoral valgus deformity was also seen in combination with a large drill hole (12 mm size) and an over-the-top positioned graft. Only two cases of valgus axis deviation were reported without a bony bar. They also found two cases with leg length discrepancy, one with overgrowth and one with leg shortening requiring appropriate epiphysiodesis. At the tibial side a formation of a genu recurvatum was found in three patients because of closure of the tibial tubercule apophysis [6]. As Schmitt et al. we used digital photographs to determine knee joint alignment. With increasing body mass index (BMI) this measurement gets uncertain [28]. Yet in our study mean patient's BMI was 21.3 (range 16.2 to 28.7, SD ± 3.4). Standard leg radiographs might be more exact but would cause more radiation exposition. With reference to growth disturbance, tunnel placement and graft fixation are the most considerable aspects during ACL reconstruction. Several studies described that a volumetric injury to the physes of larger than 7–10% can lead to growth abnormalities. Following Shea et al. a drillhole of nine millimetres placed through the femoral physis would remove a volume of 5.4% ± 1.86% in mean and 3.76% ± 1.23% respectively at the tibia physis [15]. Kercher et al. found that increasing the tunnel angle would decrease the injured physeal volume, about 0.2% each five degrees. But a larger effect would be to reduce the drill diameter, on average 1.1% every one millimetre [45]. For that we choose a diameter of seven millimetres in mean (range from five millimetres to nine millimetres) and tried to achieve that the drillholes were placed in a higher angle to the physes. Only Please cite this article as: Holwein C, et al, Functional outcome after transphyseal anterior cruciate ligament reconstruction in young patients with open growth plates, Knee (2016), http://dx.doi.org/10.1016/j.knee.2016.09.001

10

C. Holwein et al. / The Knee xxx (2016) xxx–xxx

Figure 10. Femoral tunnel placement — recommendation. Femoral tunnel placement as a compromise between am and pl insertion with a drill direction away from the periphery of the physes.

one patient got drillholes with nine millimetres of size without presenting any abnormalities at follow-up. While performing the tunnel placement the physician avoided to drill with high speed which would probably lead to thermal damage of the physis. Eccentric damage to the growth plate leads to more axial deformation than an injury to the central portion of the growth plate [8,18,46]. That is why the femoral tunnel placement, situated more peripheral, is challenging. Knowing that the femoral origin of the ACL is located only 2.92 mm ± 0.68 away from the femoral physis [47], a femoral tunnel which is too dorsal might interfere with the important ossification groove of Ranvier and the perichondral ring of LaCroix at the periphery of the physes [18,19]. Additionally a sheep model has shown that there are only minimal growth changes when filling the channels crossing the physes with graft tissue [18]. Taking this into account we entered the femoral drillhole closely to the border of the posterolateral bundle insertion at the anteromedial bundle insertion site and only used metaphyseal fixations filling the tunnels with graft tissue (Figure 10).

4.5. Discussion of outcome results depending on additional meniscal surgery or femoral fixation Meniscal injuries with additional surgery did not seem to influence patients' outcome at least in our sample size. It was not the aim of this study to determine differences between femoral fixation techniques but some findings should be discussed. Interestingly femoral button fixation seemed to be better than screw or pin fixation concerning values for ATT in side to side differences in external rotation, even screw fixation seemed to be better than pin fixation. This fact should be assessed in a new study. In the outcome scores there was no obvious trend between the groups. The fact that patients with femoral screw fixation spend more hours per week participating in sports than those in the pin group which might be more caused by hazard because no other trend in the other scores was detectable. Even though we exclusively observed low values for the differences between screw fixation with the other two groups concerning leg length and axis, we could see the trend to a shortening and valgization. So the screw although situated above the physes might induce a slower growing at the lateral femur condyle physes. But this fact must be reevaluated with a taller sample size than ours with n = 4. Please cite this article as: Holwein C, et al, Functional outcome after transphyseal anterior cruciate ligament reconstruction in young patients with open growth plates, Knee (2016), http://dx.doi.org/10.1016/j.knee.2016.09.001

C. Holwein et al. / The Knee xxx (2016) xxx–xxx

11

5. Conclusion Few studies describe clinical outcomes after transphyseal ACL reconstruction in patients with open physes [10–13,48,49]. To our knowledge there is no study so far reporting return to sport in detail and rotational laxity in combination with anterior translation for this patient population. Based on the knowledge of consequential damage to cartilage and meniscus we are performing a contemporary transphyseal reconstruction with metaphyseal fixation when clinical findings and symptoms confirm an ACL-tear. Our study shows promising results in skeletally immature patients undergoing transphyseal ACL reconstruction using hamstring grafts and a metaphyseal fixation. Among 42 patients there were only three reruptures. Subjective stability and objective laxity are comparable to adults. Return to previous sport activity level is possible. Severe growth disturbances were not seen. Conflict of interest statement The corresponding author declares that he has no conflict of interest. The senior author has a consulting contract with the following companies: - Arthrex Inc., 370 Creekside Blvd. Naples, Florida 34108, USA - medi GmbH & Co. KG, Medicusstr. 1, 95448 Bayreuth, Germany Appendix A Description of Laxitester measurement with KT1000 according to Mayr et al. [16] In detail, the patient flexed his knee to 30° and dorsiflexed the ankle for prevention of supination or pronation. To avoid varus and valgus movements in the knee stress posts were used. Anterior tibial translation (ATT) was measured in three positions: in neutral position, in internal and in external tibial rotation. Internal and external tibial rotations, which were documented in five degree steps, were performed applying a torque of two nanometres. For ATT measuring the applied forces were 30 lb and manual maximum. For final IKDC objective grading values assessed with 30 lb were crucial. References [1] Mizuta H, et al. The conservative treatment of complete tears of the anterior cruciate ligament in skeletally immature patients. J Bone Joint Surg Br 1995;77(6): 890–4. [2] Lawrence JT, Argawal N, Ganley TJ. Degeneration of the knee joint in skeletally immature patients with a diagnosis of an anterior cruciate ligament tear: is there harm in delay of treatment? Am J Sports Med 2011;39(12):2582–7. [3] Millett PJ, Willis AA, Warren RF. Associated injuries in pediatric and adolescent anterior cruciate ligament tears: does a delay in treatment increase the risk of meniscal tear? Arthroscopy 2002;18(9):955–9. [4] Moksnes H, et al. Functional outcomes following a non-operative treatment algorithm for anterior cruciate ligament injuries in skeletally immature children 12 years and younger. A prospective cohort with 2 years follow-up. Br J Sports Med 2013;47(8):488–94. [5] Anderson AF, Anderson CN. Correlation of meniscal and articular cartilage injuries in children and adolescents with timing of anterior cruciate ligament reconstruction. Am J Sports Med 2015;43(2):275–81. [6] Kocher MS, et al. Management and complications of anterior cruciate ligament injuries in skeletally immature patients: survey of the Herodicus Society and the ACL Study Group. J Pediatr Orthop 2002;22(4):452–7. [7] Chotel F, et al. Growth disturbances without growth arrest after ACL reconstruction in children. Knee Surg Sports Traumatol Arthrosc 2010;18(11):1496–500. [8] Robert HE, Casin C. Valgus and flexion deformity after reconstruction of the anterior cruciate ligament in a skeletally immature patient. Knee Surg Sports Traumatol Arthrosc 2010;18(10):1369–73. [9] Kumar S, Ahearne D, Hunt DM. Transphyseal anterior cruciate ligament reconstruction in the skeletally immature: follow-up to a minimum of sixteen years of age. J Bone Joint Surg Am 2013;95(1), e1. [10] Kim SJ, Shim DW, Park KW. Functional outcome of transphyseal reconstruction of the anterior cruciate ligament in skeletally immature patients. Knee Surg Relat Res 2012;24(3):173–9. [11] Redler LH, et al. Anterior cruciate ligament reconstruction in skeletally immature patients with transphyseal tunnels. Arthroscopy 2012;28(11):1710–7. [12] Hui C, et al. Outcome of anatomic transphyseal anterior cruciate ligament reconstruction in Tanner stage 1 and 2 patients with open physes. Am J Sports Med 2012;40(5):1093–8. [13] Calvo R, et al. Transphyseal anterior cruciate ligament reconstruction in patients with open physes: 10-year follow-up study. Am J Sports Med 2015;43(2): 289–94. [14] Yoo WJ, Kocher MS, Micheli LJ. Growth plate disturbance after transphyseal reconstruction of the anterior cruciate ligament in skeletally immature adolescent patients: an MR imaging study. J Pediatr Orthop 2011;31(6):691–6. [15] Shea KG, et al. Volumetric injury of the physis during single-bundle anterior cruciate ligament reconstruction in children: a 3-dimensional study using magnetic resonance imaging. Arthroscopy 2009;25(12):1415–22. [16] Mayr HO, et al. Validation of a measurement device for instrumented quantification of anterior translation and rotational assessment of the knee. Arthroscopy 2011;27(8):1096–104. [17] Prince JS, Laor T, Bean JA. MRI of anterior cruciate ligament injuries and associated findings in the pediatric knee: changes with skeletal maturation. AJR Am J Roentgenol 2005;185(3):756–62. [18] Seil R, Pape D, Kohn D. The risk of growth changes during transphyseal drilling in sheep with open physes. Arthroscopy 2008;24(7):824–33. [19] Ballock RT, O'Keefe RJ. The biology of the growth plate. J Bone Joint Surg Am 2003;85-A(4):715–26. [20] Aitken RC. Measurement of feelings using visual analogue scales. Proc R Soc Med 1969;62(10):989–93. [21] Higgins LD, et al. Reliability and validity of the International Knee Documentation Committee (IKDC) subjective knee form. Joint Bone Spine 2007;74(6):594–9. [22] Lysholm J, Gillquist J. Evaluation of knee ligament surgery results with special emphasis on use of a scoring scale. Am J Sports Med 1982;10(3):150–4. [23] Tegner Y, Lysholm J. Rating systems in the evaluation of knee ligament injuries. Clin Orthop Relat Res 1985;198:43–9. [24] Marx RG, et al. Development and evaluation of an activity rating scale for disorders of the knee. Am J Sports Med 2001;29(2):213–8. [25] Naal FD, et al. Return to sports and recreational activity after unicompartmental knee arthroplasty. Am J Sports Med 2007;35(10):1688–95. [26] Hefti F, et al. Evaluation of knee ligament injuries with the IKDC form. Knee Surg Sports Traumatol Arthrosc 1993;1(3–4):226–34. [27] Beattie P, et al. Validity of derived measurements of leg-length differences obtained by use of a tape measure. Phys Ther 1990;70(3):150–7.

Please cite this article as: Holwein C, et al, Functional outcome after transphyseal anterior cruciate ligament reconstruction in young patients with open growth plates, Knee (2016), http://dx.doi.org/10.1016/j.knee.2016.09.001

12

C. Holwein et al. / The Knee xxx (2016) xxx–xxx

[28] Schmitt H, et al. Determining knee joint alignment using digital photographs. Knee Surg Sports Traumatol Arthrosc 2008;16(8):776–80. [29] Sommer C, Friederich NF, Muller W. Improperly placed anterior cruciate ligament grafts: correlation between radiological parameters and clinical results. Knee Surg Sports Traumatol Arthrosc 2000;8(4):207–13. [30] Harner CD, et al. Anterior cruciate ligament reconstruction: endoscopic versus two-incision technique. Arthroscopy 1994;10(5):502–12. [31] Medicus H, Gron AM, Moorrees CF. Reproducibility of rating stages of osseous development. (Tanner–Whitehouse system). Am J Phys Anthropol 1971;35(3): 359–72. [32] Woods GW, O'Connor DP. Delayed anterior cruciate ligament reconstruction in adolescents with open physes. Am J Sports Med 2004;32(1):201–10. [33] Cohen M, et al. Transphyseal anterior cruciate ligament reconstruction in patients with open physes. Arthroscopy 2009;25(8):831–8. [34] Ardern CL, et al. Return to sport following anterior cruciate ligament reconstruction surgery: a systematic review and meta-analysis of the state of play. Br J Sports Med 2011;45(7):596–606. [35] Smith FW, et al. Subjective functional assessments and the return to competitive sport after anterior cruciate ligament reconstruction. Br J Sports Med 2004; 38(3):279–84. [36] Feller J, Webster KE. Return to sport following anterior cruciate ligament reconstruction. Int Orthop 2013;37(2):285–90. [37] Ciccone 2nd WJ, et al. Viscoelasticity and temperature variations decrease tension and stiffness of hamstring tendon grafts following anterior cruciate ligament reconstruction. J Bone Joint Surg Am 2006;88(5):1071–8. [38] Hoher J, et al. Mechanical behavior of two hamstring graft constructs for reconstruction of the anterior cruciate ligament. J Orthop Res 2000;18(3):456–61. [39] Mayr HO, et al. Graft-dependent differences in the ligamentization process of anterior cruciate ligament grafts in a sheep trial. Knee Surg Sports Traumatol Arthrosc 2012;20(5):947–56. [40] Bollen S, et al. Changes in the four-strand hamstring graft in anterior cruciate ligament reconstruction in the skeletally-immature knee. J Bone Joint Surg Br 2008; 90(4):455–9. [41] Janssen RP, et al. Anterior cruciate ligament reconstruction with 4-strand hamstring autograft and accelerated rehabilitation: a 10-year prospective study on clinical results, knee osteoarthritis and its predictors. Knee Surg Sports Traumatol Arthrosc 2013. [42] Streich NA, et al. Transphyseal reconstruction of the anterior cruciate ligament in prepubescent athletes. Knee Surg Sports Traumatol Arthrosc 2010;18(11):1481–6. [43] Zimmerman LJ, et al. Symmetric limb overgrowth following anterior cruciate ligament reconstruction in a skeletally immature patient. J Pediatr Orthop B 2015; 24(6):530–4. [44] Shifflett GD, et al. Growth arrest following ACL reconstruction with hamstring autograft in skeletally immature patients: a review of 4 cases. J Pediatr Orthop 2015. [45] Kercher J, et al. Anterior cruciate ligament reconstruction in the skeletally immature: an anatomical study utilizing 3-dimensional magnetic resonance imaging reconstructions. J Pediatr Orthop 2009;29(2):124–9. [46] Stadelmaier DM, et al. The effect of drilling and soft tissue grafting across open growth plates. A histologic study. Am J Sports Med 1995;23(4):431–5. [47] Behr CT, Potter HG, Paletta Jr GA. The relationship of the femoral origin of the anterior cruciate ligament and the distal femoral physeal plate in the skeletally immature knee. An anatomic study. Am J Sports Med 2001;29(6):781–7. [48] Maffulli N, Del Buono A. Anterior cruciate ligament tears in children. Surgeon 2013. [49] Preiss A, et al. Rupture of the anterior cruciate ligament in growing children: surgical or conservative treatment? A systematic review. Unfallchirurg 2012; 115(9):848–54.

Please cite this article as: Holwein C, et al, Functional outcome after transphyseal anterior cruciate ligament reconstruction in young patients with open growth plates, Knee (2016), http://dx.doi.org/10.1016/j.knee.2016.09.001