Long-term clinical, functional, physical activity, and radiographic outcomes a er anterior cruciate ligament reconstruction or rehabilitation alone

Abstract

Introduction: Current literature does not demonstrate superior outcomes after anterior cruciate ligament (ACL) reconstruction (ACLR) compared with progressive rehabilitation alone. ACL injured patients are, however, a heterogeneous population and treatment do not fit into a onesize-fits-all paradigm. Also, the choice between ACLR and rehabilitation alone is only one part of ACL treatment: pre-and postoperative rehabilitation, surgical indications, and patient education are also important. We, therefore, need studies that evaluate treatment algorithms applicable in clinical practice to further inform treatment choices. Further, factors other than treatment choice may explain the heterogeneity in outcomes after ACL injury. We need to improve the knowledge of how patients’ developmental trajectories differ, and factors associated with successful outcomes after both ACLR and rehabilitation alone to optimize individualized treatment.

The overall aims of this dissertation were to (I) evaluate the five-year outcomes of the treatment algorithm used in the Delaware-Oslo ACL cohort study and (II) explore subgroups of ACL injured patients and identify factors associated with outcomes.

Methods: The four papers included in this dissertation are based on two separate research projects. Papers I, II, and IV origin from the Delaware-Oslo ACL cohort study - a prospective cohort study including 276 patients with a first-time ACL injury. All patients followed a specific treatment algorithm: Patients participated in an informed shared decision-making process about treatment, including education, and they concurrently underwent five weeks of progressive neuromuscular and strength training exercises followed by clinical testing. In papers I and II, we described treatment choices and five-year clinical (patient-reported outcome measures, PROMs; knee pain; new knee injuries), functional (muscle strength; single-legged hop tests), physical activity (sports participation; Marx activity rating scale), and radiographic outcomes (tibiofemoral osteoarthritis, OA; radiographic features). Further, we used one-way analysis of variance, chisquare tests, and Fisher exact tests to compare outcomes among patients who chose (1) early ACLR (≤ six months) with pre- and postoperative rehabilitation, (2) delayed ACLR (> six months) with pre- and postoperative rehabilitation, or (3) progressive rehabilitation alone. Paper III was a systematic review on prognostic factors for PROMs and physical activity two to ten years after ACL reconstruction or injury. We selected only studies with appropriate study designs and statistical analyses, performed a rigorous risk of bias assessment using the Quality in Prognosis Study (QUIPS) risk of bias tool, and judged the quality of evidence for each prognostic factor according to the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach. In paper IV, we used group-based trajectory modeling to identify subgroups who share common trajectories of five-year PROMs, and multinomial logistic regression to assess associations with trajectory affiliation.

Main results: Within five years (80% follow-up rate), 64% of the patients had undergone early ACLR, 11% delayed ACLR, and 25% progressive rehabilitation alone (papers I and II). The rehabilitation alone group were older, had less concomitant medial meniscus injuries, and participated in less level-I versus level-II sports preinjury compared to the early and delayed ACLR groups. We found good clinical, functional, physical activity, and radiographic outcomes following our treatment algorithm: Among the whole cohort, 79% to 85% scored above a threshold for a patient acceptable symptom state for different PROMs (83%-87% after early ACLR, 65%-78% after delayed ACLR, and 77%-88% after rehabilitation alone), and >95% were still active in some kind of sports (paper I). Only 6% and 4% had tibiofemoral OA in the index and contralateral knee, respectively, and only 6% had knee pain in the index knee (paper II). No outcomes statistically significantly differed among the three treatment groups (papers I and II). The systematic review (paper III) found moderate certainty evidence for concomitant meniscus and cartilage injuries as prognostic factors for worse PROMs two to ten years after ACLR. Other prognostic factors had very low certainty. There was a lack of studies on patients treated with rehabilitation alone and 60% of the included studies were at high risk of bias. We identified four distinct five-year trajectories of PROMs – Low (8.9%), Moderate (50.2%), High (37.5%), and High before declining (3.4%) - indicating four subgroups of ACL injured patients. Concomitant meniscus injuries and new knee injuries were important characteristics of the unfavorable Low and High before declining trajectories, respectively. Factors associated with belonging to the High trajectory were mainly related to having better functional outcomes early after injury (paper IV).

Conclusions: We found good five-year clinical, functional, physical activity, and radiographic outcomes following the treatment algorithm used in the Delaware-Oslo ACL cohort study. There were no statistically significant differences in outcomes among patients who underwent early ACLR, delayed ACLR, and progressive rehabilitation alone (papers I and II). The systematic review (paper III) identified concomitant meniscus and cartilage injuries as prognostic factors for worse PROMs long-term after ACLR. We identified four distinct trajectories of five-year patientreported knee function, whereof 9 of 10 patients belonged to the favorable Moderate and High trajectories. In accordance with paper III, concomitant meniscus injuries were associated with affiliation to the unfavorable Low trajectory (paper IV).