Perioperative Morbidity Associated with ACL Reconstruction: A Comparison of Risk Between Soft-Tissue Allografts and Autogenous Grafts in 413 Patients (SS-66)

Abstract

<h3>Introduction</h3> To compare risk of perioperative complication events associated with allogenic and autogenic graft procedures during routine follow-up for six months after index anterior cruciate ligament (ACL) reconstruction surgery. <h3>Methods</h3> We used a retrospective cohort design. The patient cohort was identified through billing record queries for specific diagnostic and procedure codes for primary ACL reconstruction. Eligible patients were age 12 to 60 years who underwent ACL reconstruction from January 1, 2003-June 30, 2009 in the surgical practice of one of the authors. ACL reconstruction was performed using a traditional trans-tibial single tunnel approach. Fixation was primarily EndoButton for the femora and either interference screw or bicortical fixation (Washer-loc) on the tibia. Information on patient demographics, graft type, and surgical procedures were abstracted from medical records. Each patient was followed for six months from the date of his/her surgery. Morbidity was defined as complications occurring during each patient's follow-up period. Complications were identified from billing record queries for specific diagnostic and treatment codes. Risk of complication among patients receiving allograft tissue was compared to risk among patients receiving autograft tissue. We used the exact Cochran-Mantel-Haenszel method to estimate risk ratios (RR) and 95% confidence intervals (CI) as the measure of association between graft type and morbidity risk. <h3>Results</h3> The cohort included 413 patients. The average age of the patients was 33 (±12) years, 65% were male, and 66% received allograft (tibialis anterior, hamstring, patellar, or Achilles tendons) tissue while the remainder received autograft tissue (quadrupled semitendinosus and gracilis or bone-patellar tendon-bone). Graft selection occurred primarily based on patient preference. Patients receiving allograft tissue were on average older and with higher body mass index (BMI) compared to those who received autograft tissue, and were more likely to be female or married. During the six-month follow-up, 22 patients (5%) experienced a complication. Morbidity risk was 7% among patients receiving allograft tissue and 2% among patients receiving autograft tissue. Complications occurred less often among men (3%) than women (10%) (p=0.01), but did not according to any other patient characteristic. After adjustment for patient sex, the morbidity risk was 3-fold greater comparing allograft with autograft tissue use, with RR=3.0 (95% CI: 0.9-9.7) (p=0.06). Non-infectious complications, such as painful orthopedic fixation device and subsequent removal, were observed as the primary complication among 14 (3%) patients, 12 of whom received allograft tissue (adjusted RR=2.7, 95% CI: 0.6-12.1). Infections leading to graft removal were the primary event observed among 8 patients (2%), 7 of whom received allograft tissue (adjusted RR=3.5, 95% CI: 0.5-25.4). <h3>Conclusion</h3> Within 6 months after a first ACL reconstruction surgery, the risk of complication, including infection or device removal, was 5% in the entire cohort. However, morbidity risk was 3 times greater among patients receiving allograft tissue than among those receiving autograft tissue. Risk was elevated by a similar magnitude for non-infectious and infectious complications. These results may help patients and surgeons weigh the short term risks and benefits of choosing allograft or autograft tissue for use in ACL reconstruction.