To the Editors:

I read with interest the article by Ageberg et al (1), published in a recent issue of Arthritis Care & Research, which presented cross-sectional observations from a randomized controlled trial (RCT) in which patients, after an anterior cruciate ligament (ACL) injury, were randomized to either training and surgical reconstruction, or to training only. The objective was to study muscle strength and functional performance 2 to 5 years after injury. The outcome was assessed by 3 muscle power tests (as measures of lower extremity muscle strength) and 3 hop tests (as measures of functional performance). The authors concluded that the lack of differences between the patients treated with training and surgical reconstruction and those treated with training only indicates that reconstructive surgery is not a prerequisite for restoring muscle function.

A crucial question to consider is why cruciate ligaments are reconstructed. Most, if not all, knee surgeons performing ACL reconstruction would say that the operation is performed to restore functional knee stability, not to increase muscle strength or to improve hop tests. Very few, if any, would use a reconstruction to prevent the development of knee osteoarthritis (OA). Accordingly, the main indication for ACL reconstruction is symptomatic instability. Although the study subjects were drawn from a RCT, and the authors highlight this fact as their main strength (i.e., random allocation of the treatment), it is concerning that the authors do not comment on the discrepancy with respect to meniscal tears at baseline (20 of 36 for the surgical group versus 5 of 18 for the training-only group; no information is provided on how these tears were diagnosed or if meniscectomies were performed), and contralateral ACL injuries (9 out of 36 for the surgical group versus 2 out of 18 for the training-only group), the higher share being in those treated with surgical reconstruction as shown in Table 1 of the article by Ageberg et al. The reason(s) for this discrepancy is unclear, but must be interpreted by the reader as due to either selection bias (41% drop out), unsuccessful randomization in the RCT, crossover (i.e., some patients who were randomly allocated to the training arm of the RCT had ACL reconstruction at a later stage due to treatment failure), or any combination of these. Data suggest that the surgically reconstructed group has had more substantial knee trauma and, if crossover patients are included, most likely have had the most instability and lack of confidence with their knee. In addition, the discrepancy of meniscal tears between groups and the higher number of contralateral ACL injuries in the subjects with reconstruction suggests that using the Limb Symmetry Index (calculated by dividing the result for the injured leg by that of the uninjured leg and multiplying by 100) to compare the surgical treatment with the nonsurgical treatment group may be less adequate. The lack of consideration of these facts leaves me puzzled about the validity and clinical relevance of the author's conclusion.

The authors also speculate that the impaired muscle function in the study subjects is the cause of OA development after ACL injury. However, the number of meniscal tears, mainly in the reconstructed knees, makes this hypothesis for OA development questionable. Meniscal tears and meniscectomy are strong risk factors for knee OA (2, 3). Recently, we have shown that in ACL-injured subjects, only those who were meniscectomized developed knee OA (4). Indeed, we agree that that there is no evidence showing that reconstruction of the ACL prevents or reduces the risk of OA. However, how different factors contribute to OA in the ACL-injured knee is not well studied. It is most likely that too many ACL reconstructions are made, although references to support the statement, “∼50% of patients in Sweden and 90% of patients in the US with ACL deficiency have surgical reconstruction,” would be appropriate.

In deciding treatment in ACL deficiency, patients with symptomatic unstable knees should be individually assessed regarding type and frequency of current physical activity as well as expected future activity. To study the outcome of ACL reconstruction, we urge tests that examine the exclusive properties of a reconstructed cruciate ligament with respect to functional knee stability.

Leif Dahlberg MD, PhD*, * Lund University, Lund, Sweden.