Summary
The DOSTAR Study: A Double Blinded Randomised Controlled Trial Comparing Dual and Single Tendon Hamstring Graft Anterior Cruciate Ligament Reconstruction
Abstract
Introduction
In spite of the popularity of hamstring graft Anterior Cruciate Ligament (ACL) reconstruction, hamstring tendon harvest has been linked to reduced strength, donor site pain and muscle strains after surgery, whilst the impact of additional Gracilis tendon harvest is unclear. Traditional graft fixation methods using a tibial interference screw generally require both Semitendinosus and Gracilis tendons, but newer techniques allow for a short single tendon (Semitendinosus) graft and suspensory fixation on the tibia with the theoretical benefit of minimising hamstring donor site morbidity and graft laxity. This study aims to prospectively compare outcomes between single-tendon (ST) and dual-tendon (DT) hamstring grafts for ACL reconstruction.
Methods
134 patients (ST=65, DT=69) patients were included, powering the study to reveal differences in donor site morbidity at an alpha of 0.05 with 80% power. All patients underwent hamstring ACL reconstruction by the senior author (PD) using uniform general anaesthesia, adductor canal block, local anaesthetic infiltration and minimum 8mm diameter grafts. Suspensory fixation (Fixed Loop Endobutton) was used for femoral fixation, while tibial fixation utilised an interference screw (BioRCI) in the DT group and adjustable loop (UltraButton) in the ST group. All other surgical techniques and rehabilitation were uniform, and the patient and therapist were blinded as to graft choice and fixation method.
35 patients have reached 24 months, 90 patients 12 months; while 6 month post-operative outcomes are presented here for 114 patients (ST=53; DT=61). Patient-reported outcome measures (PROMS) were collected including IKDC2000, Lysholm and Modified Cincinnati Scores as well as Visual Analog Scale for Pain Frequency (VAS-F) and Severity (VAS-S). A validated donor site morbidity score was utilised along with assessment of anterior knee laxity using KT-1000. Single hop for distance and isokinetic knee extensor and flexor peak torques were analysed. Primary analysis was by generalised linear mixed models, with Mann Whitney U Tests used to assess group differences.
Results
Patient characteristics including age, sex and concomitant meniscal repairs were matched between groups at baseline. Mean graft size was significantly larger in the DT group (ST=8.45mm; DT=8.95mm, p=0.001). All PROMs exhibited a significant effect of time (p=0.05), reflecting improvements in both groups post-operatively. There were no differences in strength or functional outcomes between the groups, namely knee flexion torque (p=0.113), knee extensor torque (p=0.286) or hop tests (p=0.235). Significant differences emerged in donor site morbidity, with the ST group showing markedly lower donor site morbidity to 6 months (p=0.001) compared to the DT group. In contrast there were no significant differences in side to side KT-1000 laxity, ACL-RSI, and Noyes scores between groups at the same time point (p>0.05).
Conclusion
To the best of our knowledge this study provides the first demonstration in a prospective, randomised trial of significantly reduced donor site morbidity using single-tendon hamstring ACL grafts with adjustable suspensory tibial fixation. This results in a smaller mean graft diameter, but there is no compromise on other early post-operative outcomes including PROMs, knee laxity and functional assessments. Ongoing longer-term clinical and MRI follow-up is continuing to further validate these results.