Facing Challenges in Osteochondritis Dissecans Healing Research

In their study, Segal et al. explored the association between patient characteristics and osteochondritis dissecans (OCD) lesion factors before surgery and the observation of ≥90% bone integration at 6 months after surgery, referred to as a “fast” healing outcome. Using prospectively collected data of 79 cases, the researchers found that OCD bone density of ≥70% relative to the parent bone at the preoperative timepoint was associated with the fast healing outcome, with specificity of 66% and sensitivity of 87%, whereas patient weight and age, the lesion stability, lesion dimensions, and extent of edema were not independently predictive of the OCD bone healing outcome in a multivariable model. Of note, the OCD bone density change from the preoperative measurement to the 6-month postoperative measurement was not different between the slow healing group (<90% bone integration) and the fast healing group (≥90% bone integration), implying that “more bone” in the OCD fragment before surgery was associated with “more bone” at the end point, but the biology of the healing process was not necessarily different between the slow and fast OCD healing groups. The conclusion of the study can thus be valuable for counseling patients when discussing their chances to return to unrestricted activities within 6 months on the basis of simple preoperative radiographic measurements. Yet this does not seem to be the end of a story but rather a beginning that motivates further research on the topic. Three thoughts can be illuminated in this regard. First, from a mathematical point of view, the predictive ability of a specific outcome should optimally be reported using 4 values: (1) positive predictive value, (2) negative predictive value, (3) sensitivity value, and (4) specificity value1. Among these, the positive predictive value is the statistical metric that should preferably be used to answer the question: If the factor is present, will the outcome occur?1 This value quantifies the probability of the outcome given the presence of the factor in an individual patient. However, the sensitivity value is important for ruling out a disease (with high sensitivity meaning that there are few false negatives), and the specificity value is important for ruling in a disease (with high specificity meaning that there are few false positives). Accordingly, for the patient who is scheduled for OCD surgery and is eager to know the likelihood of experiencing fast healing, there is an interest in the positive predictive value of the preoperative radiographic measurement “OCD progeny/parent bone density ratio ≥70%” to predict the outcome of ≥90% bone healing at the 6-month follow-up. Considering the specificity value reported in this study, which was only 66%, this implies that a substantial proportion of the patients with preoperative bone density of ≥70% are still likely to experience <90% density at the 6-month follow-up. For research purposes, this practically means that other factors are interrelated with the density outcome measure in this disease. Accordingly, let’s reflect on the next thought. Second, considering the biological process that is taking place at the progeny-parent bone interface, it may be naïve to limit assessment of this process to merely counting bone trabeculae. Instead, there is likely a complex interplay of mechanical and biological factors2. From a mechanical standpoint, this primarily relates to OCD fragment stability and the repetitive forces applied to it. It seems challenging to quantify how much stability was achieved in each case during surgery. Furthermore, too much stability can be as bad as too little stability when aiming for optimal bone integration. From a biological standpoint, factors involved in this process of bone healing, including hematoma formation, inflammation, stem cell differentiation, growth factors such as bone morphogenetic proteins, and cytokines, are enormously complex. None of these was measured in this study. This complex mechanical-biological interplay can add to the inherent limitations of such a study and can explain the limited specificity value of the study for ruling in the outcome measure of fast OCD osseous healing. Third, the articular hyaline cartilage that covers OCD fragments is often involved3. The mechanical quality of this structure is crucial for the successful return to unrestricted activities, but it was not explored in this study. In summary, this study reveals the tip of an iceberg that awaits further interplay between orthopaedic and biological research.