Adolescent ACL Injury

Transcription

1 2bach.qxd 3/28/2006 2:59 PM Page 79 Adolescent ACL Injury Treatment Considerations Raymond Pavlovich, Jr, MD Steven H. Goldberg, MD Bernard R. Bach, Jr, MD ABSTRACT: Anterior cruciate ligament (ACL) injury in a skeletally immature patient presents unique treatment challenges. In many cases, conservative treatment with bracing and physical therapy fails, resulting in recurrent instability, pain, swelling, and meniscal and chondral injury. The goal of surgical reconstruction is to recreate ACL stability without causing growth plate arrest, leg-length discrepancy, or angular deformity. Patient characteristics such as skeletal age, Tanner stage, onset of menses, family member height, growth spurt, recent change in foot size, and growth charts can help the surgeon approximate the degree of skeletal maturity and aid in selecting the timing and safest type of reconstruction. Numerous surgical techniques, ranging from an extra-articular reconstruction to intra-articular graft passage without physeal violation to standard transtibial and transfemoral tunnel placement with physeal violation, have been popularized. The majority of existing studies are retrospective case series, describing a particular author s specific technique experience. This article reviews the basic science and clinical literature, presents a treatment algorithm, and provides several case studies. [J Knee Surg. 2004;17:79-93.] INTRODUCTION Anterior cruciate ligament (ACL) injuries in the active, skeletally immature patient present a difficult problem. Nonoperative treatment in sedentary individuals is a viable option. However, the natural history of the ACL-deficient knee in younger age groups parallels that of the adult population and is characterized by recurrent instability, meniscal tears, and chondral injury. Surgical treatment involving extraphyseal reconstruction often has less than desirable results, whereas transphyseal fixation presents the potential of growth plate disturbance and subsequent limb malalignment. Previous studies of midsubstance tears in the skeletally immature athlete suggest they are uncommon and Drs Pavlovich, Goldberg, and Bach are from the Department of Orthopedic Surgery, Rush University Medical Center, Chicago, Ill. Reprint requests: Bernard R. Bach, Jr, MD, Dept of Orthopedic Surgery, Rush University Medical Center, 1725 W Harrison St, Ste 1063, Chicago, IL occur with a 1%-3% incidence in children aged 14 years. 15,31 Many authors maintain that tibial spine avulsion is more likely due to the added strength of the ACL versus the epiphyseal plate. However, a recent literature review revealed an increase in midsubstance tears, possibly due to increased participation in organized athletics in younger age groups, increased awareness of such injuries, and improved diagnostic abilities and techniques (eg, magnetic resonance imaging [MRI] and KT-1000). Traditionally, nonoperative care has been the mainstay of treatment in this age group. Education, activity modification, and rehabilitation, combined with bracing, have been traditionally recommended. However, high noncompliance rates combined with functional instability place these patients at high risk for recurrent knee injury. Studies have demonstrated an increased incidence of instability and meniscal tears despite rehabilitation and bracing. Fear of premature growth plate closure and angular deformities and limb-length inequalities have prompted the investigation of several nonanatomic, extraphyseal ACL reconstruction techniques as well as several alternative transphyseal techniques. Many of these studies involving 79

2 2bach.qxd 3/28/2006 2:59 PM Page 80 THE JOURNAL OF KNEE SURGERY April 2004/Vol 17 No 2 TABLE 1A LITERATURE REVIEW Avg Chronologic Physiologic Age Study Growth Study Age (y) (range) Determination Size (n) Plate Aichroth et al 1 13 (11-15) Tanner, family growth wide open characteristics 21 closing Andrews et al (10-15) Bone age left hand, Tanner, 8 Open family growth characteristics lower-limb growth indices Aronowitz (11-15) Bone age left hand, age 14 y 19 Open to undergo surgery Brief (14-28) None 9 NS Graf et al ( ) NS 4 Open Janarv et al NS 28 Open Lipscomb & NS wide open Anderson partially open Lo et al (8-14) NS but minimum expected 5 5 wide open growth was 5 cm Matava & 14.7 Secondary sexual characteristics 8 NS Siegel 37 McCarroll et al (10-14) Bone age left hand, family growth, 10 Open Tanner McCarroll et al (13-15) Tanner, family growth, growth 60 NS spurt occurrence, degree of physeal opening on radiograph Micheli et al (2-14) Bone age left hand 8 NS Parker et al ( ) NS 6 All open Pressman et al (5-17) NS 23 7 open 11 closing/closed Robert & 9.9 (9.3-13) NS 8 NS Bonnard 47 Abbreviations: ACL=anterior cruciate ligament, allo=allograft, auto=autograft, avg=average, BPTB=bone-patellar tendon-bone, EA=extra-articular, IA=intra-articular, ITB=iliotibial band, NA=not available, NR=not relevant, NS=not specified, OTT=over-the-top, PT=patellar tendon, and ST/G=semitendinosus gracilis. the skeletally immature patient are limited due to difficulties in classifying growth potential at the time of injury and treatment (Table 1A). 3,9,19,22,33,37,45,46 Differences in gender and method of reporting age (skeletal versus chronologic) present investigators with a challenge. Various methods of estimating the degree of skeletal maturity include the determination of skeletal age, gender, parental and sibling growth characteristics, degree or stage of secondary sexual characteristic growth, occurrence of a growth spurt, changes in shoe size, and postinjury height changes at follow-up. Background The differential diagnosis of knee pain and hemarthrosis in the skeletally immature patient includes ligamentous tear, tibial spine avulsion fracture, meniscal tear, patellar sleeve fracture, distal femoral physeal fracture, capsular tear, and osteochondral fracture. Anterior knee instability can arise from ligamentous or bony sources. The differential diagnosis of anterior instability includes an ACL injury, congenital ACL deficiency, excessive hyperelasticity, and periarticular fractures 80

3 2bach.qxd 3/28/2006 2:59 PM Page 81 Adolescent ACL Injury Reconstruction Graft Type Tibial Femoral Other ST/G auto Transphyseal, titanium Transphyseal, titanium None stirrup positioned away screw aimed away from physis from physis 5 freeze-dried fascia lata allo, Transphyseal 7-mm tunnel, OTT fixed w/ staples Some ACLs repaired 1 fresh-frozen Achilles allo more vertical fixed w/ or tied over washer 2 irradiated Achilles allo staples or tied over washer Cryopreserved Achilles allo Transphyseal 9-10 mm tunnel Transphyseal 9-10 mm None fixed by press-fitting tunnel fixed with staples ST/G auto released at Passed under anterior OTT fixed w/ staples EA ITB tenodesis musculotendinous junction, horn medial meniscus attached distally 2 modified Andrews ITB NR NR None tenodesis, 2 ST passed IA Tibial tunnel OTT fixed to femur 12 ST auto, 4 PT auto, 4 data NA Transepiphysis OTT fixed w/ staples None ST/G auto Transphyseal 6.4-mm tunnel Transepiphyseal 8-mm 3 Ellison and 18 tunnel (two incision Losee EA technique) sutured to augmentations periosteum 3 ST/G auto, 2 quadriceps More vertical transphyseal OTT fixed w/ staples None and PT auto 6-mm tunnel; tendons attached distally, supplemental fixation NS ST/G auto Transphyseal 7-9 mm tunnel Transphyseal 7-9 mm None fixed w/ suture over tunnel fixed w/ button button or staple w/ or w/o staple 3 AO ITB tenodesis, 7 modified NR NR NS Andrews ITB tenodesis BPTB auto Transphyseal fixed w/ Transphyseal fixed w/ 2 had prior EA interference screw interference screw reconstruction ITB Trough in metaphysis made, ITB passed behind OTT Posterolateral corner ITB passed deep to inter- position, into the joint repair meniscal ligament, sutured and sutured to femur to tibia 5 ST/G auto, 1 doubled loop Groove in tibial ephiphysis, Groove in distal femur in 3 ACLs repaired ST auto 5 were left attached, in OTT position fixed w/ 1 stapled staples 6 ST auto Transphyseal Transphyseal in 20, None 17 BPTB auto fixation NS OTT in 3 Free PT Epiphyseal groove Outside-in extraphyseal None fixed with staples tunnel, fixed with interference screw (Figure 1). The collagen fibers in the skeletally immature ACL form a strong connection between ligament, perichondrium, and epiphyseal cartilage, whereas in the adult ACL, weaker Sharpey s fibers form the link between ligament and bone. 11 This difference in ligamentous attachment between the adult and child may play a role in the pattern of knee injury seen in each age group. Kellenberger and von Laer 27 observed that tibial spine avulsion occurred 80% of the time in children aged 12 years whereas midsubstance tears occurred 90% of the time in children aged 12 years. The age dependency of ACL injury in skeletally immature children was further demonstrated in another retrospective review of 1273 children aged 16 years presenting to a trauma center over a 10-year period. 52 The authors reported no ACL tears in children aged 10 years, whereas they found occasional tears in children aged years, and an increased frequency in children aged years. 52 Rare traumatic ACL injury has occurred in children aged 10 years. 33,41,46,47 81

4 2bach.qxd 3/28/2006 2:59 PM Page 82 THE JOURNAL OF KNEE SURGERY April 2004/Vol 17 No 2 Figure 1. High-resolution radiograph of a proximal canine tibia 2 weeks after drilling across the growth plate. The longitudinal arrangement of the trabecular bone in the rapidly distracting metaphyseal bone is similar to that observed during distraction osteogenesis. (Reprinted with permission from 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: Copyright American Orthopaedic Society for Sports Medicine.) Figure 2. Photomicrogaph of the tibial section seen in Figure 1. Note the bony bridge that spans the growth plate (hematoxylin-eosin, original magnification 40). (Reprinted with permission from 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: Copyright American Orthopaedic Society for Sports Medicine.) Diagnosis The diagnosis of ACL tears should be based primarily on clinical history and physical examination. The injury mechanism can include a noncontact twisting injury, valgus blow to the leg with the foot fixed on the ground, or forced knee hyperextension. Usually an audible pop is heard at injury. Acutely, the patient presents with pain, knee effusion within hours of injury, decreased range of motion, and difficulty bearing weight. Chronically, a child may present with recurrent effusion and recurrent instability. Clinically, the diagnosis is established with an asymmetric anterior drawer, Lachman test, pivot shift test, or KT-1000 side-to-side differences 3 mm. Joint line tenderness and provocative meniscal rotation tests (eg, McMurray test, Apley compression test) should be performed to evaluate for associated meniscal tears. Magnetic resonance imaging of the knee is highly sensitive and specific in evaluating adult knee pathology. However, MRI does not provide increased diagnostic ability over clinical examination in children. 29 In one study, MRI and clinical examination both had a sensitivity of 71% and specificity of 92%. Other studies indicated the sensitivity may be as low as 50% and specificity as low as 64%. 28,36,50,55 Variations in developmental anatomy, less experience interpreting MRI in skeletally immature patients, smaller anatomic structures, and a higher incidence of partial ligamentous injuries may contribute to the lower sensitivity and specificity seen in children. 29 Additionally, one study showed meniscal signal abnormalities in 6 of 13 patients, all with normal menisci on arthroscopic examination. This confirmed other reports that increased meniscal signal in children has a high falsepositive rate and may be due to increased vascularity compared to the adult meniscus. 52 McDermott et al 40 emphasized the sensitivity, specificity, and accuracy of MRI is in part age-dependent. They reported that MRI evaluation of adolescents (children aged 15 years) yielded similar results as adult patients, whereas MRI evaluation of children aged 15 years had decreased sensitivity, specificity, positive predictive value, and accuracy for all knee injuries. 40 Due to the difficulties in interpreting knee MRI in younger patients, the clinician should be aware of its limitations when weighing its importance in the diagnostic scheme. Arthroscopic evaluation and evaluation under anesthesia are valuable diagnostic adjuncts in the young patient with a hemarthrosis when the clinical examination is equivocal. In one study, in which arthroscopy was performed in skeletally immature patients with a hemarthrosis, 47% had an ACL tear. Furthermore, examination under anesthesia for evidence of laxity or a pivot shift prior to arthroscopy also provides valuable clinical information to aide in the diagnosis of an ACL injury. In the majority of cases, arthroscopy is not necessary to diagnose an ACL injury. Natural History In the adult, evidence indicates that active patients with an ACL-deficient knee will experience repeated episodes of instability resulting in chondral injury, meniscal tears, recurrent swelling, and pain. 3,8,10,12,25,54 The natural history of the ACL-deficient knee in the younger population paral- 82

5 2bach.qxd 3/28/2006 2:59 PM Page 83 Adolescent ACL Injury 3 4 Figure 3. High-resolution radiograph of a proximal canine tibia 4 months after placement of a fascia lata graft through a drill hole made across the open growth plate. No bony formation is noted across the growth plate in the bone tunnel. (Reprinted with permission from 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: Copyright American Orthopaedic Society for Sports Medicine.) Figure 4. Photomicrograph of the section seen in Figure 4. The graft crosses the growth plate, and no bony formation is noted across the physis. Note that the growth plate immediately adjacent to the graft-filled tunnel appears normal (hematoxylin-eosin, original magnification 40). (Reprinted with permission from 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: Copyright American Orthopaedic Society for Sports Medicine.) lels that of the adult population. Numerous studies have documented poor results with long-term conservative management including physical therapy and bracing. 5,19,39,46 Compliance with activity modification often is difficult to achieve in adults and may be even more difficult in children. Furthermore, committing a young patient to years of activity modification and bracing may be unrealistic. In most case series, 19,26 a high incidence of recurrent instability, pain, swelling, and new meniscal tears occurs in skeletally immature patients treated nonoperatively for a complete ACL tear. In one study, all 7 patients with an average age of 14.8 years who were treated conservatively had fair or poor results. Four patients had definitive radiographic evidence of arthritis at follow-up. 26 In another study, 8 patients with an average age of 14.5 years and open physes were treated with supervised physical therapy and bracing. Seven of 8 patients experienced new meniscal tears at an average of 15 months from initial injury. 19 Additionally, most children treated conservatively have exhibited difficulty in returning to their preinjury level of sports participation, as reported by Mizuta et al 42 where only 1 of 18 patients returned to the preinjury level of athletics and 11 of 18 patients had radiographic Fairbanks criteria of arthritis at minimum 36-month follow-up. 42 In light of the difficulties involved with conservative treatment of ACL injuries in active children, many clinicians have advocated surgical intervention. In adults, ACL reconstruction principles involve use of strong grafts, anatomically created tunnels, and rigid fixation. In a patient with open physes, creation of femoral or tibial bone tunnels would violate the physes if placed in an anatomic position. Concerns of possible growth disturbance as a result of physeal drilling and transphyseal graft placement has led many clinicians to avoid such a procedure in the skeletally immature patient. These concerns have prompted numerous animal investigations of physeal growth characteristics. These studies suggest that physeal defect size, filling of the defect with soft tissue, and magnitude of tension placed across the physis may significantly affect growth plate function. 16,21,35 Additionally, abnormal growth may result in the absence of bony bridge formation. 16 Animal Studies Stadelmaier et al 49 examined the effect of soft-tissue grafting on the prevention of physeal bar formation. Eight canines underwent transphyseal drilling of the tibia and femur with or without placement of a fascia lata autograft in the drill hole. Drill hole size was calculated to represent the same ratio of hole size to physeal surface area as would be expected in a skeletally immature child. All four nongrafted canines developed bony physeal bridging, whereas the grafted subjects showed no evidence of bridge formation or alteration of growth plate morphology on either radiographic or histologic analysis (Figures 1-4). This study illustrates the importance of completely filling the drill hole with soft tissue to prevent bony bar formation. Other investigators have attempted to quantitate the degree of physeal destruction that results in growth disturbance. Makela et al 35 showed that growth disturbance, osseous bridging, and epiphyseal cartilage changes were not observed when drilling across 13% of the transverse diameter (3% of total cross-sectional area) of the distal 83

6 2bach.qxd 3/28/2006 2:59 PM Page 84 THE JOURNAL OF KNEE SURGERY April 2004/Vol 17 No 2 femoral physis in rabbits, but the above changes were observed when drilling across 20% of the transverse diameter (7% of total cross-sectional area). Guzzanti et al 21 reported the effect of transphyseal reconstruction of the ACL in skeletally immature rabbits using semitendinosus tendon autograft and proximal femoral periosteal suture fixation. The procedure resulted in damage to 3% of the cross-sectional area of the femoral physis and 4% of the cross-sectional area of the tibial physis. Histologic examination did not reveal evidence of epiphysiodesis, but 2 of 21 rabbits developed tibia valgus and 1 tibia was shortened. These studies demonstrate a critical threshold of cross-sectional physeal injury that results in growth abnormalities. From a practical standpoint, a more vertically oriented tibial tunnel results in less transphyseal violation. The physis responds differently to force depending on the direction in which the force is applied. The Delpech principle states a physis will respond to tension with increased growth. 16 Conversely, the Heuter-Volkman principle states the application of compressive force perpendicular to the physis will inhibit longitudinal growth. 6 A recent study involving transphyseal placement of fascia lata autograft in the knees of skeletally immature beagles followed by graft tensioning to 80 N resulted in statistically significant distal femoral valgus deformity and proximal tibial varus deformity. 16 These growth disturbances occurred in the absence of radiographic or histologic evidence of physeal bar formation, leading the authors to caution against transphyseal reconstruction in the skeletally immature patient population. 16 Modifications in surgical technique may help minimize physeal damage if a transphyseal reconstruction is necessary in a skeletally immature patient. As animal studies indicate a critical threshold of physeal damage must occur before growth disturbance results, tunnel size can be decreased to 6-7 mm to limit the physeal surface area that is damaged by drilling and graft placement. 4,33 To minimize damage to the physis, more vertically oriented tibial and femoral tunnels will result in smaller, circular physeal defects than more horizontally placed tunnels, which will produce larger, elliptical defects. 4,33 Additionally, choosing an all soft-tissue graft over bone-tendon-bone grafts helps minimize physeal bar formation. 49 Estimation of Bone Age and Deformity Animal models have generated insight into understanding physeal physiology and response to trauma, but ultimately it is unknown exactly how much physeal damage can occur in humans without producing adverse effects. To assist physicians in decision making, researchers have investigated means of predicting skeletal growth. Wester et al 53 developed a technique of estimating maximal angular deformity and shortening if complete growth arrest occurs at surgery. They reported that remaining physeal growth is almost entirely determined by skeletal age and physeal diameter and is independent of overall leg length. Greater angular deformity is expected in patients with a smaller physeal transverse diameter. Using trigonometry, they generated graphs that predict angular deformity and shortening, which can be used as a rough approximation to aid in surgical timing and help counsel patients and families on the risks of surgery. The distal femur and proximal tibia are responsible for 65% of the total leg length, and thus damage to either physis significantly impacts future limb growth. Other methods of estimating leg-length inequality can be used based on the White-Menelaus rule, which states the distal femur provides 3/8-inch of growth per year whereas the proximal tibia provides 1/4-inch of growth per year. Finally, tables from Green and Anderson can be used to aid in estimation of remaining growth potential. 2 In a personal communication, Kuo (K. Kuo, MD) stated that in a patient with 1 year of estimated growth remaining, a growth abnormality does not appear for 6 months. However, patients with greater amounts of growth remaining may manifest deformities more quickly. This is in agreement with Timperlake et al, 51 who stated bony bridge formation after percutaneous epiphysiodesis occurred within 4-6 months. It is critical to accurately estimate remaining growth to make decisions about surgical timing and type of surgical reconstruction. Most methods of determining remaining growth depend on accurate determination of bone age. 41 Skeletally immature patients with an ACL injury should undergo a posteroanterior (PA) radiograph of the left hand and wrist. Bone age can then be determined by comparing this radiograph to standard images in the Radiographic Atlas of Skeletal Development of the Hand and Wrist. 20 Aside from bone age, numerous secondary factors can also be used to ascertain the degree of skeletal maturity including physeal width on plain radiographs, Tanner stage, parental and sibling height, recent foot growth, growth spurt occurrence, and menses onset. 23,32,39 In females, skeletal growth generally occurs for 2 years following menarche. It is important to remember that skeletal maturity in females (14 years) differs from that in males (16 years). All of the above clinical information should be used to help the surgeon obtain an impression of the degree of skeletal immaturity to counsel the patient and family and recommend treatment options. For example, a 12-year-old premenarchal patient with a skeletal age of 11 years will be treated differently than a 12-year-old mature female whose menses began at age 10 years. SURGICAL TECHNIQUES Primary ACL Repair Primary unaugmented ACL repair has been shown to 84

7 2bach.qxd 3/28/2006 2:59 PM Page 85 Adolescent ACL Injury have a low probability of long-term success with activity limitation and instability as the rule. During injury, in addition to the gross mechanical tearing of the ligament, a more diffuse, microscopic deformation and stretching of individual fibers occurs. Fiber elongation results in a functional lengthening of the tendon on suture reapproximation. When the ACL is torn, a loss of the vascular supply also occurs due to tearing of branches of the middle genicular artery and the medial and lateral inferior genicular arteries. Finally, the normally extrasynovial ACL becomes intrasynovial as a result of the tearing of its protective synovial sheath. The combination of the collagen fiber stretching, loss of blood supply, and damage to the synovial sheath makes ACL injury healing unpredictable and unlikely. 23 A prospective study of acute repair of the ACL compared to acute repair with augmentation was performed on adults in 1990 and is presented for a historical perspective. 24 The augmentation consisted of placing a distally attached strip of patellar retinaculum through tibial and femoral drill holes with imbrication to the repaired ACL. In the repair group, 16 of 22 patients decreased their activity level, 15 of 22 had 2 on anterior drawer testing, 10 of 22 had a positive Lachman test, and 8 of 22 had a positive pivot shift test at follow-up. The augmentation group had improved results in all of the above categories. The unacceptable levels of knee instability and limitation of activity in the repair group is consistent with other studies of acute ACL repair. 18,34,44 Furthermore, Cross et al 13 showed that failure of the repaired ligament occurred over time with increased anterior tibial translation developing between 2- and 7-year follow-up. Results of ACL repair in the skeletally immature population mirrors that of the adult population. DeLee and Curtis 15 treated three patients, aged 12 years, with primary unaugmented ACL repair, and at 20-month followup, all three had clinically significant laxity and abnormal Lachman testing. Engebretsen et al 17 also attempted unaugmented primary repair in children and reported 8 of 8 patients had to modify their activities and 5 of 8 patients had significant instability. In light of these poor results, most surgeons have directed surgical attempts towards reconstruction. Surgical Reconstruction The literature review regarding ACL reconstruction in the skeletally immature patient is challenging due to several factors. The overall low incidence of these injuries limits study size and power. Limited preoperative assessment of skeletal maturity and potential for future growth combined with the grouping of patients of various ages makes interpretation of outcomes difficult. Authors often have failed to delineate between pre-and postpubescent patients. Reporting of chronologic rather than skeletal age has been a major weakness of some studies. Diverse surgical techniques and varying postoperative evaluations add to the complexity in developing conclusions about efficacy and safety. Ideally, postoperative scanograms would most accurately detect any leg-length discrepancy. The above limitations should be considered when critically reviewing the literature regarding ACL injury treatment in the skeletally immature patient. Extraphyseal Reconstruction. To avoid physeal violation and a possible growth disturbance, several investigators have described nonanatomic, nonisometric, extraphyseal techniques aimed at restoring knee stability. 14,15 DeLee and Curtis 15 recommended using the iliotibial band by releasing it proximally and passing it through the femoral notch, proximal to the distal femoral physis and then under the lateral collateral ligament. The iliotibial band is then fixed to Gerdy s tubercle with a screw. 15 Bergfeld described using a portion of the patellar tendon, which is detached proximally and passed beneath the transverse ligament ( tomato stake procedure). It is then attached in the over the top position. 14 Drez modified Bergfeld s technique by fashioning a groove in the tibial epiphysis and lateral femoral condyle, allowing the graft to lie in a more anatomic position. 14 Several extraphyseal reconstruction cases series have been published with most authors reporting mixed results (Tables 1A and 1B). 9,19,22,38,41,45,47 McCarroll et al 38 reported midsubstance ACL tears in 10 patients and Graf et al 19 reported 2 patients who underwent extra-articular reconstruction without violating either physes. No leglength discrepancies or angular deformities were reported. Postoperative instability physical examination data and standardized knee score outcomes were not specified or reported. Most patients were able to return to prior activity (55 [92%] of 60), and only 3 (5%) patients had a KT mm degree of instability. Three (25%) of 12 patients in these studies sustained postoperative meniscal and graft tears requiring revision ACL or meniscal surgery. Although minimizing the risk of growth-plate injury, at the expense of continued instability and recurrent injury, these techniques are limited to prepubescent patients who have failed conservative treatment. To improve postoperative stability allowing for a return to sports while minimizing the risk for growth abnormality, several authors have developed ACL reconstruction techniques that use an intra-articular graft passage without physeal violation (Figure 5). Brief 9 reported 9 patients, Janarv et al patients, Micheli et al 41 8 patients, Parker et al 45 6 patients, and Robert et al 47 8 patients. No leg-length discrepancies or angular deformities were reported at followup. Physical examination for instability revealed 3-mm maximum manual difference with KT-1000 testing in most cases, a Lachman grade of 0 or 1, and a negative pivot shift phenomenon in instances where it was tested. 4,9,22,31,33,39,47 Lysholm scores ranged from in three studies. 22,41,45 85

8 2bach.qxd 3/28/2006 2:59 PM Page 86 THE JOURNAL OF KNEE SURGERY April 2004/Vol 17 No 2 TABLE 1B LITERATURE REVIEW Follow-Up KT-1000 Leg-Length Angular Side-to-Side Study Avg (mo) Discrepancy Deformities Difference (mm) Aichroth et al 1 49 None None NS Andrews et al tibia 10 mm shorter; 1 tibia None 5 3 mm, mm 8 mm longer; 5/8 femurs different by 7 mm; 2 femurs shorter by 10 and 11 mm; 1 femur 10 mm Aronowitz et al 7 25 Avg 0.12 cm shorter w/ range None 20-lb force avg 1.7 mm 0.7 cm shorter to 0.5 cm longer (range 0-3) Brief 9 (36-78) NS NS Only tested 5 patients: 4 max man 3 mm 1 max man 5 mm Graf et al None None NS Janarv et al None None Avg at 89 N 1.5 mm Lipscomb & 35 No preop data, postop equal Femorotibial valgus Avg at 20-lb force 1.8 Anderson 31 in 7; differences of 5 mm angulation equal in 20, mm (range: 7 to 3.5) in 10; operative limb longer 2 had 1 difference, in 6 and shorter in 4; difference 1 had 2 difference, of 5-10 mm in 5 w/ operative 1 had 3 difference limb longer in 3, shorter in 2; 1.3 cm longer in 1 and 2 cm longer in 1 Lo et al No subjective complaints No significant deformities Mean 1 mm max man avg mm mean difference 1.0 none w/ difference (range: 5 to 4 mm) (range: 2 to 3) 3 mm Matava & 32 Overall leg-length avg 1 mm None 5 5 at 30 lb 3 mm Siegel 37 (range: 7 to 7 mm); femur 1 at 30 lb 8 mm avg 1 mm (range: 7 to 2 had contralateral 5 mm); tibia, avg 1 mm injuries not compared (range: 3 to 3 mm) McCarroll et al None None NS McCarroll et al None 1 cm None mm max man, 4-5 mm max man in 6 5 mm max man in 3 Micheli et al year old with congenitally None Avg max man 1.1 mm absent ACL developed 2-cm (range: 2 to 4.5 mm) discrepancy thought to be due to congenital disorder and unrelated to surgery Parker et al All operative legs 1 cm None Max man 3.6 mm longer Pressman et al None 1 cm None ST max man 6.1 mm; PT max man 3.5 mm Robert & 42 No difference in 4; 0-5-cm None Max man 0 mm in 2 Bonnard 47 difference in 3; 0.7-cm max man 1 mm in 3 difference in 1 max man 3 mm in 3 Abbreviations: ACL=anterior cruciate ligament, avg=average, dis=displacement, EA=extra-articular, HSS=Hospital for Special Surgery, IA=intra-articular, IKDC=International Knee Documentation Committee, max man=maximum manual, NS=not specified, postop=postoperative, preop=preoperative, PT=patellar tendon graft, and ST=semitendinosus graft. 86

9 2bach.qxd 3/28/2006 2:59 PM Page 87 Adolescent ACL Injury Subsequent Lachman Pivot Shift Activity Level Final Rating Surgery/Reinjury 16 grade 0, 27 grade 0, NS IKDC: 21 grade A, 3 reruptures, 1 from a 20 grade 1 15 grade 1, 15 grade B, fall and 2 from sports- 8 grade 2, 2 grade 2, 8 grade C, related injuries; 3 grade 3 3 grade 3 3 grade D reconstruction NS 5 grade 0 NR 1 patient limited running and 6 excellent, 4 tibial hardware 3 grade 1 turning 1 good, 1 fair removals, 1 arthroscopic release, 1 arthroscopic chondroplasty; 1 ACL reinjury 5 grade 1 All grade 0 16/19 returned to prior level Lysholm avg 97 2 femoral hardware removals; 1 meniscectomy 9 grade 1 NS None returned to prior level w/o 8 satisfied NS bracing 1 dissatisfied 2 grade 1 2 grade 0 4 returned to prior level, 2 EA NS 1 meniscal repair and 2 not 2 not reported reconstructions suffered sympto- IA ACL reconstruction; reported matic instability and new meniscal 1 bilateral partial tears, 2 hybrid asymptomatic meniscectomy with IA ACL reconstruction NS NS Tegner 7 in 9 w/ 2 having lower Lysholm avg patient w/ meniscal activity than desired; Tegner 6 repair and subsequent in 6 w/ 1 having lower activity reinjury with ACL retear than desired 11 recorded as 1 positive, 15 w/ return to sport at prior HSS score 45.9 NS abnormal 2 moderate, level; 3 with return to sport w/ (50-pt scale) 3 mild antero- decreased performance; 5 16 excellent medial returned to all sports except 7 good, 1 fair rotatory football; 1 limited to jogging instability 4 grade 0 All grade 0 4 returned to prior level; 1 could IKDC score: NS 1 grade 1 only perform level III activity 4 Grade A (due to postop patellar osteo- 1 Grade C chondral fracture) 6 grade 0 1 grade 1, All 8 returned to sports, 3 at same Cincinnati knee 1 recurrent tear; 1 closed 1 4-mm dis others NS level, 5 returned to decreased rating: avg 98 manipulation; 1 arthro- 1 6-mm dis level for reasons unrelated scopic lysis of to injury adhesions NS NS 4 patients with mild instability, NS 1 revision ACL with surgery needed; 5 asymptomatic PT graft and medial w/ unrestricted activity meniscal repair NS NS 55/60 returned to prior level NS 3 ACL tears, 1 meniscal tear, 2 arthroscopic releases 6 grade 0 All grade 0 All returned to sports Lysholm avg 97.4 NS 2 grade 1 4 grade 0 All grade 0 4 returned to prior level w/ Lysholm avg 95.2 NS 2 grade 1 protective brace, 1 denied instability but did not return to HSS score 96.6 sports to avoid reinjury (100-pt scale) ST avg 1.9, ST avg 1.4, NS Lysholm: ST 78.6 NS PT avg 1.1 PT avg 0.85 PT 88.4 NS NS Tegner 5 in 1, Tegner 6 in 3 IKDC score: NS Tegner 8 in 1, Tegner 9 in 3 5 grade A, 3 grade B 87

10 2bach.qxd 3/28/2006 2:59 PM Page 88 THE JOURNAL OF KNEE SURGERY April 2004/Vol 17 No 2 A B C D Figure 5. Anterior cruciate ligament reconstruction as proposed by Micheli (A). Iliotibial band harvesting (B). The curved graft passer is placed around the lateral femoral condyle and through the notch (C). Graft passage around the lateral condyle and over-the-top position and through the notch (D). Completed reconstruction with the graft sutured to the superior aspect of the lateral femoral condyle and periosteum medial to the tibial tubercle. Figure 6. Graft placement in the over-the-top position of the lateral femoral condyle fixated within a cortical trough in the femur. Four of 5 studies did not comment on reinjury or need for further surgery. 9,41,45,47 One study reported recurrent meniscal and ACL tears requiring surgery. These studies suggest that current intra-articular graft passage techniques may minimize growth-plate injury and have an improved functional outcome compared to an extra-articular reconstruction. A nonanatomic intra-articular reconstruction is, however, inferior to anatomic transtibial and transfemoral tunnel placement used in skeletally mature patients. Most reports of extra-articular, nonanatomic knee stabilization in the pediatric population with ACL ruptures reveal several deficits. Often, active patients have continued instability resulting in activity modification and are unable to return to preinjury levels of sports participation. Additionally, continued instability may lead to progressive meniscal and articular cartilage pathology. Due to these limitations, these procedures often are selected for active, immature patients with a significant potential for future growth who have failed conservative treatment with the knowledge that future revision surgery may be necessary at skeletal maturity. Partial Transphyseal Reconstruction. Several reports have been made involving combined extraphyseal and transphyseal techniques in skeletally immature individuals. 4,31,33,43 Some authors prefer extraphyseal femoral fixation to avoid an eccentrically placed transphyseal tunnel. One reason is the distal femur s large contribution to overall leg length (37%) 43 in addition to the potential for subsequent angular deformity that exists with peripheral femoral tunnel placement. Many authors use transphyseal tibial tunnels as the proximal tibia contributes less to overall leg length (28%) 43 and the tunnel is more centrally placed (Figure 6). Three studies, with a total of 37 patients, were performed by Andrews et al 4 (n=8), Lipscomb et al 31 (n=24), and Lo et al 33 (n=5) using this strategy (Tables 1A and 1B). To minimize physeal damage, small transphyseal tibial tunnels measuring 6-7 mm were made in a more vertical orientation. Grafts were placed in the over-thetop position on the femur without physeal violation. Additionally, two of the three studies used extra-articular augmentations or ACL repairs in certain cases. Overall, only 2 patients had 1-cm leg-length discrepancy with smaller discrepancies noted in many of the other patients. However, Rush and Steiner 48 showed that 77% of the general population has a 7-mm leg-length discrepancy. Most studies, unfortunately, did not measure leg lengths preoperatively. No significant angular deformities were reported. Instability testing revealed a maximum manual difference of 3 mm on KT-1000 testing in most patients with a mixture of grades 0 and 1 on pivot shift and Lachman testing. The majority of patients had good to excellent results and were able to return to unrestricted 88

11 2bach.qxd 3/28/2006 2:59 PM Page 89 Adolescent ACL Injury activity. Some patients reported no decreased performance without evidence of instability. One ACL graft reinjury was reported. All Transphyseal Reconstruction. Several studies evaluated ACL reconstruction using both tibial and femoral transphyseal tunnels for graft fixation 1,7,37,39,46 (Figure 7). Patients in these studies were typically aged between 13 and 14 years and were more likely to have partially open rather than wide open physes. Patients with wide open physes typically were reconstructed using an all soft-tissue graft whereas those exhibiting characteristics consistent with a more skeletally mature age, as well as the start of physeal closure, were treated with bonepatellar tendon-bone autografts. Aichroth et al 1 (n=47), Aronowitz et al 7 (n=19), Matava and Siegel 37 (n=8), McCarroll et al 39 (n=60), and Pressman et al 46 (n=23) represent collectively 157 patients with tibial and femoral transphyseal ACL reconstructions. No leg-length inequalities were 1 cm and no angular deformities were reported in these studies. Manual instability testing showed that KT-1000 maximum manual differences were typically 3 mm, with 5 mm seen in few patients (Table 1B). The hamstring reconstructions tended to have a slightly increased anterior translation on KT-1000 testing (Table 1B). The majority of Lachman and pivot shift tests were grade 0 or 1, with a few grade 2 and 3 results (Table 1B). Several knee score systems were used, and the majority of patients returned to full activities (Table 1B). Seven recurrent ACL tears, 2 meniscal tears, and 4 cases of arthrofibrosis requiring arthroscopic debridement occurred (Table 1B). A report of a growth abnormality as a result of transphyseal ACL reconstruction was published in A 14-year-old boy underwent reconstruction with a double-stranded semitendinosus graft anchored with a femoral cannulated screw and tibial staples and cancellous bone plug placement in both tunnels. Neither Tanner stage, bone age, family member heights, or an adolescent growth spurt were recorded prior to surgery. Two years later, despite a stable knee, the patient developed a progressive valgus knee deformity requiring an openingwedge femoral osteotomy with tricortical bone allograft interposition and contralateral distal femoral epiphysiodesis. One year later, the patient was reportedly doing well without further deformity. Despite the low incidence of significant leg-length discrepancies or angular deformities in the majority of case series, the angular deformity that occurred in this patient emphasizes the theoretical and real concerns of transphyseal ACL reconstruction in skeletally immature patients. Treatment Algorithm Based on the senior author s (B.R.B.) experience Figure 7. Guide pin placement used for reaming the transphyseal tibial tunnel in ACL reconstruction. The tibial entrance site is on the flare of the medial tibial metaphysis. If the tibial tunnel is too steep, the femoral tunnel placement may be made too anteriorly with the knee flexed 90. Similarly, if an appropriate tibial tunnel angle is selected (45-50 ) but the knee is inadequately flexed, posterior cortical blowout may occur. (Reprinted with permission from Hardin GT et al. Endoscopic single-incision anterior cruciate ligament reconstruction using patellar tendon autograft: surgical technique. Am J Knee Surg. 1992;5: Copyright SLACK Incorporated.) and review of the available literature, the following treatment algorithm has been developed for skeletally immature patients with evidence of symptomatic ACL deficiency. Initially, the patient s current skeletal age is determined based on a PA radiograph of the left hand and wrist. Data from Anderson and Green suggests minimal remaining femoral and tibial growth in children with a skeletal age of 14 in females and 16 in males (Figure 8). By subtracting their current skeletal age, time to lower extremity growth completion is estimated. For example, a female patient with a skeletal age of 12 would have an estimated 2 years of remaining lower extremity growth. Thus, a hamstring autograft reconstruction using extraphyseal tibial and femoral graft fixation is recommended. Radiographic appearance of physes, age at menarche, Tanner stage, a growth spurt, and familial growth characteristics are used to further improve the estimation of skeletal maturity. Patients with gross instability, unwillingness to comply with activity modification, or both are treated according to the algorithm in Table 2. Case 1 CASE REPORTS A 12-year-old prepubescent boy presented with instability, catching and locking, pain, and swelling. At prior arthroscopy, an ACL mid-substance tear and displaced 89

12 2bach.qxd 3/28/2006 2:59 PM Page 90 THE JOURNAL OF KNEE SURGERY April 2004/Vol 17 No 2 TABLE 2 ALGORITHM FOR ACL RECONSTRUCTION IN THE SKELETALLY IMMATURE PATIENT Skeletal Reconstruction Age Gender Graft Femoral Tibial 12 Male/Female HS OTT Transepiphyseal 13 Male/Female HS OTT Transphyseal tunnel with extraphyseal fixation 14 Male HS OTT Transphyseal tunnel with extraphyseal fixation Female HS/BTB FT Transphyseal tunnel 15 Male HS OTT or FT Transphyseal tunnel with extraphyseal fixation Female HS/BTB FT Transphyseal tunnel 16 Male/Female HS/BTB FT Transphyseal tunnel Abbreviations: BTB=bone tendon bone, FT=femoral tunnel, HS=hamstring, and OTT=over the top. peripheral bucket-handle lateral meniscal tear were noted. He had injured his knee 2 years prior to presentation while attempting to jump over a stream. His mother was 5 8 tall and his father was 6 0. The patient had not yet undergone a growth spurt. On physical examination, he was 5 1 tall, weighed 95 lbs, and was consistent with a Tanner stage 2. Joint line tenderness, abnormal meniscal rotation signs, painful range of motion, and an effusion were noted. A grade 3 Lachman, grade 2 anterior drawer, and grade 2 pivot shift were noted. KT-1000 testing measurements of 11, 13, and 14 mm were found for the symptomatic knee and 5, 5.5, and 6 mm for the contralateral knee at 69 N, 89 N, and maximum manual translation, respectively. Radiographs exhibited grossly open physes. Due to the patient s symptoms and a large repairable meniscal tear, meniscal repair using an inside-out technique and ACL reconstruction using hamstring autograft was performed. Reconstruction was performed using an extraphyseal technique with transepiphyseal tibial placement (under the medial meniscus) and over-the-top femoral placement secured extra-articular fixation (staples/screw and post) (Figure 9). At 14-month follow-up, KT-1000 measurements were 2.5, 3.5, and 4 mm on the affected knee. At 6-year followup, the patient had grown 12 inches in height, reported no instability, and ACL tests were normal. No symptoms suggested meniscal repair failure. Radiographs and physical examination revealed no evidence of growth abnormality or angular deformity. Case 2 A 12-year-old female soccer player presented after a prior slide tackling injury that resulted in persistent pain and instability with cutting activities. Menarche had occurred 10 months prior to presentation. On physical examination the patient was 5 1 tall, weighed 100 lbs, had secondary sexual characteristics of a Tanner stage 4, and was close in approximation in height to her mother. She exhibited a full range of motion, a grade 3 Lachman, and a grade 1 pivot shift. KT measurements were 8, 10, and 14 mm on the affected side and 3, 4, and 6 mm on the contralateral knee at 69 N, 89 N, and maximum manual testing, respectively. Radiographs revealed closing femoral and tibial physes with an open tibial apophysis. Skeletal age was approximately 14.5 years. Magnetic resonance imaging revealed midsubstance ACL disruption with a characteristic bone bruise pattern. Due to symptomatic instability and an unwillingness to refrain from sports, ACL reconstruction was performed. As she was a petite girl with an open tibial tubercle apophysis, a bone-patellar tendon-bone allograft was used to avoid injuring the apophysis and to allow for an adequate sized graft. At last follow-up, approximately 13 months post-reconstruction, she had returned to soccer with no symptoms and KT-1000 measurements were 5, 6, and 7 mm on the affected knee, respectively, without evidence of growth abnormality or angular deformity. Case 3 A 12-year-old female softball player presented after a collision while sliding into third base, resulting in immediate knee pain and subsequent inability to play. She had experienced menarche approximately 1.5 years prior. Physical examination demonstrated a full range of motion, grade 2 Lachman, and grade 2 pivot shift testing. KT-1000 measurements were 5, 7, and 9.5 for the affected knee and 3.5, 4, and 7 mm for the asymptomatic limb at 69 N, 89 N, and maximum manual testing, respectively. Tanner stage appeared to be that of a late 3 or early 4. Radiographs of the knee revealed a closing tibial and closed femoral growth plates. Her skeletal age was determined to be closer to that of a 14-year-old. Magnetic resonance imaging revealed a midsubstance ACL injury with a characteristic bone bruise pattern. Bracing and activity modification were used initially, however she was an active athlete, who was unable to comply with this treatment, thus ACL reconstruction was performed using a bone-patellar tendon-bone autograft at age 13 years. At latest follow-up 1 year postoperatively, she had returned 90

13 2bach.qxd 3/28/2006 2:59 PM Page 91 Adolescent ACL Injury Figure 8. Growth chart that may be used as a guide in estimating the amounts of growth that may be inhibited in the distal end of the normal femur or proximal end of the normal tibia by epiphyseal arrest in the skeletal ages indicated on the base line. (Reprinted with permission from Anderson M et al. Growth and predictions of growth in the lower extremities. J Bone Joint Surg Am. 1963;45:1-14. Copyright The Journal of Bone and Joint Surgery Incorporated.) to her preinjury level of athletics and was asymptomatic. KT-1000 measurements were 4, 5, and 7 on the affected knee. Her growth plates had fully closed, and no growth abnormality or angular deformity were noted. Case 4 A 14-year-old female gymnast presented after a twisting injury to her left knee during completion of an extension flip in which her knee gave way causing her to fall. She experienced intense pain, had difficulty ambulating, and quickly developed swelling. She had not yet experienced menarche and appeared to be an early Tanner stage 3. Further examination revealed a knee effusion with limited range of motion. Grade 2 Lachman and grade 1 pivot shift testing were noted. KT-1000 testing measured 6, 8, and 14 mm on the affected knee with values of 3, 4, and 5 mm on the asymptomatic limb at 69 N, 89 N, and maximum manual testing, respectively. Radiographs revealed open growth plates and a skeletal age of 13 years (Figures 10 and 11). Magnetic resonance imaging revealed a midsubstance ACL injury (Figures 12). Physical therapy and bracing were prescribed, but despite achieving full range of motion and symmetric lower limb strength, symptomatic instability was reported 3 months postinjury. 91

14 2bach.qxd 3/28/2006 2:59 PM Page 92 THE JOURNAL OF KNEE SURGERY April 2004/Vol 17 No Figure 9. Case 1. Intraoperative lateral radiograph of a guide pin placed in the epiphyseal tibial position directed towards the over-the-top femoral position. Figure 10. Case 4. AP radiograph of the affected knee depicting open proximal tibial and distal femoral physes. Figure 11. Case 4. PA radiograph of the patient s hand demonstrating characteristics consistent with a skeletal age of 12 years. Figure 12. Case 4. Sagittal MRI demonstrates open physes. Note the absence of the ACL. Figure 13. Case 4. Postoperative AP radiograph of the affected knee displaying transphyseal tibial tunnel and hardware fixation. Anterior cruciate ligament reconstruction was performed using a hamstring (semitendinosus and gracilis) allograft placed in the over-the-top femoral and transphyseal tibial positions with extraphyseal fixation (soft-tissue screws and washers in combination with staples) (Figure 13). CONCLUSION Midsubstance ACL tears in skeletally immature patients, although rare, occur and present unique treatment considerations. A careful assessment of the patient s potential remaining growth and ability to comply with nonoperative and operative protocols must be taken. Nonanatomic, extraphyseal attempts at reconstruction appear to decrease the risk for subsequent growth abnormality, but often are suboptimal in restoring ACL function. These reconstructions are primarily reserved for prepubescent individuals with larger potentials for growth. When the patient nears skeletal maturity, more anatomic, transphyseal reconstructions are recommended, as they provide the optimal stability with minimal risk of growth disturbance. REFERENCES 1. Aichroth PM, Patel DV, Zorrilla P. The natural history and treatment of rupture of the anterior cruciate ligament in children and adolescents. A prospective review. J Bone Joint Surg Br. 2002;84: Anderson M, Green WT, Messner MB. Growth and predictions of growth in the lower extremities. Am J Orthop. 1963;45: Andersson AC. Knee laxity and function after conservative treatment of anterior cruciate ligament injuries. A prospective study. Int J Sports Med. 1993;14: Andrews M, Noyes FR, Barber-Westin SD. Anterior cruciate ligament allograft reconstruction in the skeletally immature athlete. Am J Sports Med. 1994;22: Angel KR, Hall DJ. Anterior cruciate ligament injury in children and adolescents. Arthroscopy. 1989;5: Arkin AM, Katz JF. The effects of pressure on epiphyseal growth; the mechanism of plasticity of growing bone. J Bone Joint Surg Am. 1956;38: Aronowitz ER, Ganley TJ, Goode JR, Gregg JR, Meyer JS. Anterior cruciate ligament reconstruction in adolescents with open physes. Am J Sports Med. 2000;28: Bonamo JJ, Fay C, Firestone T. The conservative treatment of the anterior cruciate deficient knee. Am J Sports Med. 1990;18: Brief LP. Anterior cruciate ligament reconstruction without drill holes. Arthroscopy. 1991;7: Buckley SL, Barrack RL, Alexander AH. The natural history of conservatively treated partial anterior cruciate ligament tears. Am J Sports Med. 1989;17: Busch MT. Sports medicine. In: Morrissy RT, Weinstein SL, eds. Lovell and Winter s Pediatric Orthopaedics. Philadelphia, Pa: Lippincott-Raven; 1996: Clancy WG Jr, Ray JM, Zoltan DJ. Acute tears of the anterior cruciate ligament. Surgical versus conservative treatment. J Bone Joint Surg Am. 1988;70: Cross MJ, Wootton JR, Bokor DJ, Sorrenti SJ. Acute repair of injury to the anterior cruciate ligament. A longterm followup. Am J Sports Med. 1993;21: DeLee JC. Ligamentous Injury of the Knee. Philadelphia, Pa: WB Saunders Co; DeLee JC, Curtis R. Anterior cruciate ligament insufficiency in children. Clin Orthop. 1983;172: Edwards TB, Greene CC, Baratta RV, Zieske A, Willis RB. The effect of placing a tensioned graft across open growth plates. A gross and histologic analysis. J Bone Joint Surg Am. 2001;83: Engebretsen L, Svenningsen S, Benum P. Poor results of anterior cruciate ligament repair in adolescence. Acta Orthop Scand. 1988;59: Feagin JA Jr, Curl WW. Isolated tear of the anterior cruciate ligament: 5-year followup study. Clin Orthop. 1996;325: Graf BK, Lange RH, Fujisaki CK, Landry GL, Saluja RK. Anterior cruciate ligament tears in skeletally immature patients: meniscal pathology at presentation and after attempted conservative treatment. Arthroscopy. 92

15 2bach.qxd 3/28/2006 2:59 PM Page 93 Adolescent ACL Injury 1992;8: Greulich WW, Pyle SI. Radiographic Atlas of Skeletal Development of the Hand and Wrist. Stanford, Calif: Stanford University Press; Guzzanti V, Falciglia F, Gigante A, Fabbriciani C. The effect of intra-articular ACL reconstruction on the growth plates of rabbits. J Bone Joint Surg Br. 1994;76: Janarv PM, Nystrom A, Werner S, Hirsch G. Anterior cruciate ligament injuries in skeletally immature patients. J Pediatr Orthop. 1996;16: Johnston DR, Ganley TJ, Flynn JM, Gregg JR. Anterior cruciate ligament injuries in skeletally immature patients. Orthopedics. 2002;25: Jonsson T, Peterson L, Renstrom P. Anterior cruciate ligament repair with and without augmentation. A prospective 7-year study of 51 patients. Acta Orthop Scand. 1990;61: Kannus P, Jarvinen M. Conservatively treated tears of the anterior cruciate ligament. Long-term results. J Bone Joint Surg Am. 1987;69: Kannus P, Jarvinen M. Knee ligament injuries in adolescents. Eight year follow-up of conservative management. J Bone Joint Surg Br. 1988;70: Kellenberger R, von Laer L. Nonosseous lesions of the anterior cruciate ligaments in childhood and adolescence. Prog Pediatr Surg. 1990;25: King SJ, Carty HM, Brady O. Magnetic resonance imaging of knee injuries in children. Pediatr Radiol. 1996;26: Kocher MS, DiCanzio J, Zurakowski D, Micheli LJ. Diagnostic performance of clinical examination and selective magnetic resonance imaging in the evaluation of intraarticular knee disorders in children and adolescents. Am J Sports Med. 2001;29: Koman JD, Sanders JO. Valgus deformity after reconstruction of the anterior cruciate ligament in a skeletally immature patient. A case report. J Bone Joint Surg Am. 1999;81: Lipscomb AB, Anderson AF. Tears of the anterior cruciate ligament in adolescents. J Bone Joint Surg Am. 1986;68: Lo IK, Bell DM, Fowler PJ. Anterior cruciate ligament injuries in the skeletally immature patient. Instr Course Lect. 1998;47: Lo IK, Kirkley A, Fowler PJ, Miniaci A. The outcome of operatively treated anterior cruciate ligament disruptions in the skeletally immature child. Arthroscopy. 1997;13: Lysholm J, Gillquist J, Liljedahl SO. Long-term results after early treatment of knee injuries. Acta Orthop Scand. 1982;53: Makela EA, Vainionpaa S, Vihtonen K, Mero M, Rokkanen P. The effect of trauma to the lower femoral epiphyseal plate. An experimental study in rabbits. J Bone Joint Surg Br. 1988;70: Mandelbaum BR, Finerman GA, Reicher MA, et al. Magnetic resonance imaging as a tool for evaluation of traumatic knee injuries. Anatomical and pathoanatomical correlations. Am J Sports Med. 1986;14: Matava MJ, Siegel MG. Arthroscopic reconstruction of the ACL with semitendinosus-gracilis autograft in skeletally immature adolescent patients. Am J Knee Surg. 1997;10: McCarroll JR, Rettig AC, Shelbourne KD. Anterior cruciate ligament injuries in the young athlete with open physes. Am J Sports Med. 1988;16: McCarroll JR, Shelbourne KD, Porter DA, Rettig AC, Murray S. Patellar tendon graft reconstruction for midsubstance anterior cruciate ligament rupture in junior high school athletes. An algorithm for management. Am J Sports Med. 1994;22: McDermott MJ, Bathgate B, Gillingham BL, Hennrikus WL. Correlation of MRI and arthroscopic diagnosis of knee pathology in children and adolescents. J Pediatr Orthop. 1998;18: Micheli LJ, Rask B, Gerberg L. Anterior cruciate ligament reconstruction in patients who are prepubescent. Clin Orthop. 1999;364: Mizuta H, Kubota K, Shiraishi M, Otsuka Y, Nagamoto N, Takagi K. The conservative treatment of complete tears of the anterior cruciate ligament in skeletally immature patients. J Bone Joint Surg Br. 1995;77: Moseley CF. Leg length discrepancy and angular deformity of the lower limbs. In: Morrissy RT, Weinstein SL, eds. Lovell and Winter s Pediatric Orthopaedics. Philadelphia, Pa: Lippincott-Raven; 1996: Odensten M, Lysholm J, Gillquist J. Suture of fresh ruptures of the anterior cruciate ligament. A 5-year followup. Acta Orthop Scand. 1984;55: Parker AW, Drez D Jr, Cooper JL. Anterior cruciate ligament injuries in patients with open physes. Am J Sports Med. 1994;22: Pressman AE, Letts RM, Jarvis JG. Anterior cruciate ligament tears in children: an analysis of operative versus nonoperative treatment. J Pediatr Orthop. 1997;17: Robert H, Bonnard C. The possibilities of using the patellar tendon in the treatment of anterior cruciate ligament tears in children. Arthroscopy. 1999;15: Rush WA, Steiner HA. A study of lower extremity length inequality. AJR Am J Roentgenol. 1946;56: Stadelmaier DM, Arnoczky SP, Dodds J, Ross H. The effect of drilling and soft tissue grafting across open growth plates. A histologic study. Am J Sports Med. 1995;23: Stanitski CL. Correlation of arthroscopic and clinical examinations with magnetic resonance imaging findings of injured knees in children and adolescents. Am J Sports Med. 1998;26: Timperlake RW, Bowen JR, Guille JT, et al. Prospective evaluation of fifty-three consecutive percutaneous epiphysiodeses of the distal femur and proximal tibia and fibula. J Pediatr Orthop. 1991;11: Wessel LM, Scholz S, Rusch M, et al. Hemarthrosis after trauma to the pediatric knee joint: what is the value of magnetic resonance imaging in the diagnostic algorithm? J Pediatr Orthop. 2001;21: Wester W, Canale ST, Dutkowsky JP, Choi IH. Prediction of angular deformity and leg-length discrepancy after anterior cruciate ligament reconstruction in skeletally immature patients. J Pediatr Orthop. 1994;14: Wittenberg RH, Oxfort HU, Plafki C. A comparison of conservative and delayed surgical treatment of anterior cruciate ligament ruptures. A matched pair analysis. Int Orthop. 1998;22: Zobel MS, Borrello JA, Siegel MJ, Stewart NR. Pediatric knee MR imaging: pattern of injuries in the immature skeleton. Radiology. 1994;190:

16 2bach.qxd 3/28/2006 2:59 PM Page 94 THE JOURNAL OF KNEE SURGERY April 2004/Vol 17 No 2 94