Objectives: Anterior cruciate ligament reconstruction (ACLR) is one of the most common orthopaedic procedures performed in the United States each year. It has been shown to be an effective procedure with have relatively good surgical and patient-reported outcomes (PROs). Historically, there have been concerns that bone-patellar tendon-bone (BTB) autograft can be associated with more postoperative anterior knee pain compared to other graft types due to donor site morbidity. Many surgeons and patients elect to use hamstring autograft to mitigate this potential concern. Previous work has evaluated anterior knee pain with visual analog scales and other instruments; however, none have specifically used the Knee Injury and Osteoarthritis Outcome Score (KOOS) pain subscale. The incidence of anterior knee pain by graft type has yet to be fully evaluated. The primary aim of this study was to compare differences in anterior knee pain between patients undergoing primary ACLR with BTB or hamstring autograft at 2-year follow-up. Secondary aims include evaluating KOOS and Single Assessment Numeric Evaluation (SANE) differences between graft types. Methods: A retrospective review of a prospectively collected PRO database was conducted for patients who underwent primary ACLR between 2009 and 2021. Patients were excluded if they underwent any nonprimary ACLR procedures other than meniscal repair or meniscectomy, received a graft type besides BTB or hamstring autograft, had incomplete PROs, experienced reoperation or graft failure, or were not between the ages of 14 and 50. Demographic information including age, sex, body mass index (BMI) was collected, as well as surgical characteristics such as American Society of Anesthesiology (ASA) score, concomitant meniscal procedures, and graft type (BTB or hamstring autograft). Outcomes collected included 2-year KOOS, the KOOS pain subscale, and SANE. The KOOS pain subscale was used to evaluate any anterior knee pain differences between the graft types. Statistical analysis consisted of descriptive analyses (counts, means, ranges, etc.), independent sample t-tests, chi-squared tests, as well as analysis of variances (ANOVAs). The level of statistical significance was set at p≤0.05. Results: A total of 411 patients were included in the final analysis. Most patients were female (n=238, 57.9%), had an average age of 26.2±10.6, and a BMI of 25.5±4.5. 80.5% of patients (n=331) had an ASA of 1. There were 237 (57.7%) BTB and 174 (42.3%) hamstring autograft ACLRs. Approximately half of patients had a concurrent meniscus tear (49.9%, n=205). There were 176 patients who underwent a meniscus operation (43.8%) consisting of having a meniscectomy (27.7%, n=114) or undergoing a repair (19.2%, n=79). The average two-year change in KOOS, KOOS pain, and SANE were +17.8, +13.8, +26.0, respectively. Female sex was found to be associated with larger improvements in KOOS (p=0.044), and KOOS pain (p=0.003) (Table 1). There were no statistically significant differences between sex or ASA score between the graft types (p=0.172, p=0.069). BMI was found to be statistically different between BTB and hamstring patients, (25.1 vs. 26.2, p=0.017, respectively) although it was not associated with changes in KOOS, KOOS pain, or SANE. There were no differences in the proportion of concomitant meniscus tears or operations between the graft types (p≥0.087). The 2-year improvement in the KOOS pain subscale from preoperative baseline was not found to be significantly different between the groups (BTB: 12.5 vs. hamstring: 15.2, p=0.137). The 2-year KOOS pain value also did not differ by graft type (BTB: 91.6 vs. hamstring: 90.6, p=0.406). The 2-year change in SANE also did not differ between the grafts (p=0.815). The baseline overall KOOS and 2-year change in KOOS were higher in BTB autograft patients compared to hamstring autograft patients (68.2 vs. 63.3, p<0.001; 16.4 vs.19.7, p=0.040, respectively). Overall KOOS at final follow-up did not differ between grafts (84.6 vs. 83.0, p=0.160, respectively). Conclusions: ACLR with BTB autograft was not found to be associated with worse anterior knee pain scores compared to hamstring autograft, with most patients achieving good outcomes in SANE, KOOS and KOOS pain at 2-year follow-up. Although patients who received hamstring grafts demonstrated larger improvements in KOOS at 2-years, they also started at a lower KOOS score on average compared to BTB patients. Patients in both groups achieved similar 2-year outcomes indicating that any differences may be due to ceiling effects of the PRO. During graft selection for ACLR, surgeons and patients can expect similar improvement in pain levels between BTB and hamstring autograft and should not choose one graft over the other based on concerns related to anterior knee pain differences.
Read full abstract