Summary

Knee pain reported over 5 years following ACLR with BPTB vs. HT autografts was generally mild in severity and had no effect on knee function or activity levels in either graft group.

Abstract

Background

Persistent, anterior knee pain is a potential complication of anterior cruciate ligament reconstruction (ACLR). The exact origin of this knee pain is unclear although bone patellar-tendon bone (BPTB) autograft use has been identified as a potential risk factor. However, few studies have comprehensively analysed the incidence, severity, and impact of knee pain on function over time.

Purpose

This study prospectively and longitudinally documented the incidence, severity and location of knee pain over 5 years in a cohort of patients who underwent primary ACLR. The main objective was to compare the incidence and severity of pain between BPTB and Hamstring (HT) autografts. Secondary objectives included identifying predictive factors for pain at each time point and assessing the impact of pain on knee function and activity levels.

Methods

This study included a prospective cohort of 1407 patients (1145 BPTB and 262 HT) aged 13-45 years who underwent primary ACLR between 2014-2016. The presence, location and severity of knee pain was assessed at 6-months, 1-year, 2-years and 5-years post-operatively. Pain severity was recorded on a numerical rating scale (NRS) and classified into categories (1-3 = mild, 4-6 = moderate, 7-10 = severe). Functional outcomes were evaluated using IKDC and Marx at each time point, with additional KOOS and WOMAC scores at 5-years. Preoperative and intraoperative characteristics were analysed for their influence on knee pain at each time point.

Results

The BPTB group reported pain more frequently at 6-months (40% vs. 25%, p < .001) but not at 1-year and 2-years follow-up. Among those reporting pain at 6-months, 62.5% of BPTB patients and 76% of HT patients classified their pain as mild. At 5-years, the BPTB group reported a higher frequency of pain (31% vs. 24%, p < .032). However, the mean KOOS pain score (95.3 vs. 95.5, p = .31) and WOMAC pain score (99.6 vs. 99.5, p = .725) at 5-years were comparable for both autografts respectively. There was no difference in IKDC score or Marx score between graft types at any time point.
At 6-months, other factors associated with increased pain included female sex (OR=1.41, p < .035) and older age (OR=1.02, p < .014). At 2-years, medial femoral condyle chondral pathology was associated with pain (OR=1.61, p < .022). At 5-years, medial femoral condyle chondral pathology (OR=1.70, p < .020) and older age (OR=1.06, p < .001) were significant predictors of pain.

Conclusions

While knee pain was more frequently linked to BPTB grafts, the difference between BPTB and HT grafts was not significant at 1- and 2-years follow-up. Most pain reported at each time point was mild and had no effect on knee function or activity level. These findings suggest that reporting the presence of pain measured solely using numerical rating scales (NRS) or individual patient reported outcome measures may not accurately capture clinically relevant and “consequential” knee pain, which is persistent and significantly impacts patient activity and quality of life.