Participants
From May 2015 to the end of July 2019, patients in this cross-sectional study were selected from among 276 patients who underwent ACL reconstruction (Figure 1). Inclusion criteria were: primary/unilateral anatomical double-bundle reconstruction using either hamstring tendon autograft alone or gracilis tendon harvested in addition to hamstring tendon; age ≥16 years and ≤40 years at testing; postoperative rehabilitation with the same protocol used in the sports physical therapy department; participation in sports involving cutting, pivoting, and jump-landings (basketball, soccer, volleyball, badminton, tennis, or frisbee) before injury; and participating in the training sessions of the same sport played before surgery at approximately 6 months after reconstruction.
Patients were excluded if they had: past experience with ACL or meniscus injury or surgery on the other side; experienced injury that affected physical function in the lower back or lower limb; undergone reconstruction of other ligaments of the knee in addition to ACL; hoped to return to high-contact sports with tackling, such as football and rugby, martial arts sports that require pairing and throwing, such as judo and wrestling, striking sports such as boxing, and snow and ice sports such as skiing and ice hockey; did not go to the hospital or participate in sports for social reasons such as relocating or becoming pregnant after reconstruction; had comorbidities that hindered a return to sports; or had missing values. The institutional review board at our institution approved the study design (approval number: M2019-019). All participants provided written, informed consent.
Postoperative rehabilitation
The postoperative rehabilitation protocol was the same for all patients. However, patients who underwent repair of the middle posterior segment of the meniscus were prohibited from deep squatting until 3 months after surgery. Patients were permitted to begin isometric quadriceps exercises as tolerated from the day after ACL reconstruction. Using a knee brace (Straighten Position Knee-Joint Immobilizer; ALCARE, Tokyo, Japan) and crutches, partial weight-bearing (20 kg) was permitted on the first day after reconstruction, and gradually increased to the maximum body weight of each patient. Use of the knee brace and crutches was discontinued at 4 weeks after reconstruction. Range-of-motion exercises from full extension to 120° of flexion were started on the second day after reconstruction. Closed kinetic chain exercises such as weight shifting and squatting were started 1–2 weeks after reconstruction. Open kinetic exercises such as seated knee extension using a machine load and a perturbation training program were started 3 months after reconstruction.
Running exercises were started in athletes who had cleared the criterion of limb symmetry index (LSI) of 65% of the knee extensor strength, as measured by the Biodex Multi-Joint Testing and Rehabilitation System (BDX-4; Biodex Medical Systems, New York, NY, USA) at 3 months after reconstruction. Speed and distance of running were gradually increased for joint effusion and symptoms of each patient. Once 80% of subjective full-speed running ability was achieved, athletic exercises related to the desired sporting activities were initiated with detailed instructions. All athletic exercises were specific to each patient, depending on the type of sport and position played. Participation in training sessions for the sports to which individual patients aimed to return was allowed from 6 months after reconstruction, as long as the patient showed no problematic symptoms in the joint and had sufficient knee isokinetic flexion/extension strength at 60°/s (LSI >80%) and had anterior OLH distance (LSI >80%) after the specified athletic training had been completed 22, 23.
Measurements
Demographic characteristics, modified Tegner activity scale score before injury, days from injury to reconstruction, days after reconstruction, and meniscus surgery procedure were taken from medical records. OLH distances and knee strength in the operated (involved) limb and non-operated (uninvolved) limb were measured on the same day, which varied by patient. All physical function tests were conducted by 4 physiotherapists, each with more than 10 years of clinical experience in rehabilitation and conditioning of patients after ACL reconstruction.
Sex was determined based on medical records. Height and weight were measured on the same testing day, and body mass index (BMI) was calculated. Oral questioning was used to confirm whether the knee that had undergone reconstruction was on the dominant side. The dominant limb was defined as the leg used to kick a ball to maximal distance 24. The level of sports activity before injury was graded using the modified Tegner activity scale 25.
The date of injury and date of reconstruction were confirmed by the patient and from medical records, and the number of days from date of injury to date of reconstruction was calculated. The number of days after reconstruction was the number of days from surgery to testing.
Meniscus injuries and treatments were confirmed from detailed records of arthroscopic findings during reconstruction. The injured segment (anterior, middle, or posterior), injury type (longitudinal, radial, or horizontal) and treatment method (suture, centralization, or partial meniscectomy) were confirmed. Participants were defined as being treated regardless of the method used.
OLH distances in the three directions (anterior, lateral, and medial) were measured in random order according to previous research 26, 27. Participants stood on one leg and were instructed to hop as far as possible and land on the same leg. The longest distance of 3 trials was recorded for each leg and each direction. The test was considered successful if the landing was stable. If the patient landed with early touchdown of the contralateral limb, which represented loss of balance, or took additional hops after landing, the hop was repeated. Patients were initially given a verbal description of the test, and were allowed to perform as many practice trials as desired, until they felt confident about the test. Three trials were performed for each leg, always starting with the non-surgical limb. For anterior OLH, the distance between the front end of the toe at starting position and the trailing edge of the heel at landing position was measured 28. For lateral OLH, the distance between the lateral side of the foot at starting position and the medial side of the foot at landing position was measured. For medial OLH, the distance between the medial side of the foot at starting position and the lateral side of the foot at landing position was measured. Results are represented as the distance-to-height ratio. Total OLH distance was calculated by standardizing the total of the three-direction OLH distances by height. Intraclass correlation coefficient (ICC) case 1 was calculated to examine the reproducibility of OLH distances in three directions in the involved limb and uninvolved limb of 10 athletes who met the same inclusion criteria as in this study. As a result, the ICCs of the single measurement values of the involved limb and uninvolved limb were in the range of 0.91–0.99 and 0.91–0.96, respectively, showing almost perfect reproducibility 29.
The Biodex Multi-Joint Testing and Rehabilitation System was used to evaluate the isokinetic strength of the knee in extension/flexion. To minimize compensatory movements during testing, participants were seated and secured with padded straps around the thigh, pelvis, and torso. The femoral condyle of the tested limb was aligned with the rotation axis of the torque meter. Participants performed 3–5 repetitions of submaximal knee extension/flexion to familiarize themselves with the testing motion. To determine the strength of knee extension/flexion, participants performed 5 consecutive concentric contractions of extension/flexion at 60°/s and 180°/s. Peak torque within the 5 trials was extracted and normalized by body weight. The uninvolved limb was tested first. Results were represented by peak torque-to-weight ratio. The ICC of concentric peak torque of the knee using the Biodex system was close to 0.90, indicating high to very high reproducibility 30.
Statistical analysis
For sample size calculation, a priori power analysis and a previous study with similar outcome measurements revealed that at least 42 participants were required to achieve an alpha of 0.05 and a power of 0.90 16, 31. Due to potential attrition, a total of 47 subjects were recruited and tested in the current investigation. The normality of each variable was confirmed by the Shapiro-Wilk test. Simple regression analyses using Spearman's rank correlation coefficient were performed to clarify the relationship between OLH distance and knee strength. The magnitude of the effect size was determined to be strong for ³0.50 and medium for ³0.30 but <0.50 32. All data were analyzed with the Statistical Package for the Social Sciences for Windows (version 21.0; IBM Corp., New York, NY, USA). Values of P <.05 were used to indicate statistical significance.