Hallux valgus remains, in many ways, the flagship condition of orthopaedic foot and ankle practice, occupying prominent early chapters in standard textbooks and continuing to fill journals at all levels of academic discourse. It also continues to be a largely soluble problem—except when it isn’t. The relatively high overall satisfaction rates reported in Level-IV follow-up studies, seemingly regardless of specific technique, render the analysis of the small number of clinical failures difficult. The challenge in all studies of the condition has transformed from that of painting a broad picture of success in the many to understanding what failed in the maddening few. The paper by Bock et al. is not revolutionary, nor is it fully generalizable beyond the Scarf osteotomy, but it does hold incremental lessons large and small toward this end.

The first incremental lesson is simple. The patients undergoing the procedure perceived themselves to be better for having done so, and that perception lasted, on balance, at least ten years. The chief psychometric instrument used in the study, the American Orthopaedic Foot & Ankle Society (AOFAS) scoring system, prohibits any finer analysis of this statement. The score is nonvalidated and contains subscores that are likely highly interrelated; it cannot be presumed to behave in a linear fashion or produce normally distributed data. The authors have appropriately recognized these limitations and modified their analyses accordingly. Median scores rather than means are reported for these data, and nonparametric statistical methods are used. Additionally, the usual limitations of a Level-IV study apply: the patients have themselves chosen to undergo the procedure and surgeon selection bias is inevitable. Less powerful statistics and a less powerful methodology lead to a less powerful conclusion; this group of selected patients perceived themselves to be better than they were preoperatively. How much so remains indeterminate.

The second lesson is more surprising. Radiographic recurrence was unmasked by the long-term follow-up at a remarkably high rate. Recurrent deviation of the toe proved to be a necessary but not sufficient condition for clinical failure: while nearly all patients with long-term pain had a recurrence, not all recurrences were painful.