Treating ACL injuries in young moderately active adults

BMJ 2013; 346 doi: http://dx.doi.org/10.1136/bmj.f963 (Published 13 February 2013) Cite this as: BMJ 2013;346:f963
  1. Bruce A Levy, associate professor,
  2. Aaron J Krych, assistant professor ,
  3. Diane L Dahm, associate professor,
  4. Michael J Stuart, professor
  1. 1Department of Orthopedic Surgery and Sports Medicine, Mayo Clinic, Rochester, MN 55905, USA
  1. B A Levy

Individual patient factors should guide the decision on whether to surgically reconstruct

Currently, the clinical indication for anterior cruciate ligament (ACL) reconstruction takes into account both patient related and knee related factors. In an effort to better identify these factors, Frobell and colleagues performed a randomized clinical trial comparing two treatment strategies for young moderately active patients with a torn ACL. Patients were randomized to “acute” ACL reconstruction (within 10 weeks) or “optional delayed” ACL reconstruction (rehabilitation with the option for delayed surgery).1 In 2010, the authors published two year outcomes that reported no difference between the two treatment arms in the primary outcome (knee injury and osteoarthritis outcome score; KOOS) or secondary outcomes of SF-36 and Tegner activity scale (TAS). However, supplementary data provided in an appendix reported 29 subsequent meniscus tears in the “optional delayed” group, and by two years almost 40% of the patients in this group had already crossed over to ACL reconstruction surgery.

In the linked study (doi:10.1136/bmj.f232), Frobell and colleagues report the five year outcomes for the trial.2 Similar to the two year results, the authors found no significant differences between the two treatment arms in primary (KOOS score) or secondary outcomes (TAS, meniscal surgery, radiographic osteoarthritis). The main strengths of this study are the randomized clinical trial design and excellent mid-term follow-up. This RCT required a high level of dedication and organization. The authors captured all crucial data points, including clinical and functional outcomes scores, as well as radiographic analysis for osteoarthritis.

However, in our opinion the greatest weakness of the study is the intention to treat analysis of the data. Half of the patients (51%) initially treated with rehabilitation elected to proceed with ACL reconstruction surgery but were analyzed in the rehabilitation group. In a randomized study, intention to treat analysis preserves randomization and minimizes bias, but in this study it also provides a conservative estimate of the effect of surgical treatment. Frobell and colleagues recognized this shortcoming and further subdivided their data into the “as treated group” and the “full analysis set.” The as treated group was separated into three distinct subgroups: early ACL reconstruction, delayed ACL reconstruction, and rehabilitation alone. In the as treated analysis, there was still no difference between treatment groups in the primary outcome (KOOS). However, the secondary analysis showed superior mechanical stability of the knee in the early and delayed ACL reconstruction groups.

The findings of this study should be treated with caution because fewer than 25% of the “active adults” were able to return to their pre-injury activity level regardless of treatment, which is inconsistent with the published literature. Furthermore, only 76% of the early ACL reconstruction group and 60% of the delayed reconstruction group had a normal pivot shift test. Regardless of treatment, all patients had a median Tegner score of only 4, which is equivalent to moderately heavy labor (a score of 5 indicates that the patient is able to participate in recreational sports). A recent meta-analysis reported an 82% rate of return to sport after ACL reconstruction.3 By comparison, the rate of return to sport in the current study is low, and these relatively poor results in all treatment groups bring into question the generalizability of the conclusions to other patient populations, particularly high demand athletes.

In the two year follow-up report, significantly more meniscal procedures were performed in the “optional delayed” group. In the current five year follow-up study, the authors report that subsequent meniscal procedures were “similar” when analyzed by the treatment actually received. However, it is difficult to interpret the data because patients who underwent “extensive meniscal fixation” were excluded from the study, yet meniscus tears in both groups were treated with “partial resection or fixation when indicated.” In addition, the investigators found that surgery on the same meniscus was less common in patients treated with early ACL reconstruction than in those treated with initial rehabilitation. Selection bias according to the extent of meniscus damage, the potentially increased risk of meniscus repair failure in an ACL deficient knee, and improved healing rates of meniscus repair performed in conjunction with ACL reconstruction need to be considered.4

Frobell and colleagues’ study did not show a difference in the prevalence of radiographic osteoarthritis at five years after injury. This observation may be a function of the initial ACL injury and may be influenced by the 51% crossover rate. In addition, the five year endpoint may be too early to detect arthritic changes, as indicated by the low rates (3-16%) of arthritis identified in this study. Patients who did not undergo ACL reconstruction had significantly more instability, which could put them at higher risk for meniscus tears and deterioration of articular cartilage in the long term.5 6 7

This trial has shown—at both two year and five year endpoints—that some moderately active patients with a torn ACL but no associated collateral ligament or extensive meniscal damage can function well with an ACL deficient knee. However, it is difficult to predict which patients will have symptoms of instability that require surgery. We maintain that surgical decision making should be individualized, taking into consideration the specifics of the particular injury, the individual needs of the patient, and the anticipated demands that the patient will place on the knee.8


Cite this as: BMJ 2013;346:f963


  • Research, doi:10.1136/bmj.f232
  • Competing interests: We have read and understood the BMJ Group policy on declaration of interests and declare the following interests: no funding was received to write this editorial; BAL and MJS have received consultancy fees and royalties from Arthrex.

  • Provenance and peer review: Commissioned; not externally peer reviewed.