Commentary Single-event multilevel surgery (SEMLS), consisting of multiple, patient-specific soft-tissue and osseous surgical operations, has become the standard of care to improve or maintain gait function in ambulant persons with cerebral palsy (CP). Despite support from 1 pilot randomized controlled trial and numerous cohort studies, key questions remain unanswered1,2. These include the effects of interventions before and after SEMLS and the best method of achieving optimum dynamic knee function, especially correction of flexed-knee gait. The long-term outcome study of distal femoral extension osteotomy combined with patellar tendon advancement (DFEO + PTA) by Boyer and colleagues is a major effort in understanding the interplay between this knee surgery and the effects of time, growth, and previous and additional interventions on outcomes compared with more conventional treatment. It was not a perfect study design; a randomized surgical trial with long-term follow-up would have been superior but also nearly impossible on both ethical and practical grounds. However, the surgical technique for DFEO + PTA has been developed and standardized by the Gillette group, the follow-up was long-term (median, 13 years), and the battery of outcome measures covered the spectrum of the International Classification of Functioning, Disability and Health (ICF). Furthermore, the comparison group, although not perfectly matched on baseline characteristics and surgeries prior to baseline, underwent procedures that are familiar to most surgeons. For many, this will be a useful and practical comparative study. The authors had 3 hypotheses, namely that DFEO + PTA would result in better knee kinematics than conventional surgery, the DFEO + PTA group would score better on questionnaires addressing ICF domains, and short-term gains in gait after DFEO + PTA would be maintained. Only the first hypothesis was confirmed, which is an important finding. Perhaps the reason the 2 groups were not different in regard to ICF domains was the relapse in gait correction, as shown by the gait deviation index, at long-term follow-up. The authors suggest several other possibilities, including that the DFEO + PTA group showed more severe involvement at baseline, the Americans with Disabilities Act levels the playing field in terms of activities and participation, and the natural history of CP overpowers our interventions. These are all plausible explanations and are very important for all surgeons in this field to consider. However, the issue of prior gastrocnemius-soleus complex lengthening may not have received sufficient attention. The body support moment, which allows persons with CP to maintain upright posture, is contributed to by the ankle plantar flexors, the knee extensor mechanism, and the hip extensors3. The majority of patients had prior lengthening of both the ankle plantar flexors and the hamstrings, thereby weakening 2 key muscle groups4, which might have contributed to a more extended posture at the ankle, knee, and hip. In addition, the DFEO + PTA group had 3 times more injections of botulinum toxin A and more procedures prior to the baseline gait analysis than did the non-DFEO + PTA group. In this regard, populations of persons with CP attending gait laboratories in North America seem quite different from those in other parts of the world. In Australia, most patients with CP presenting for gait analysis and SEMLS have had no prior surgery1,2. Accordingly, the prevalence of iatrogenic flexed-knee gait may be less; when it is present, it may be less severe; and it seems to be more amenable to long-term, stable correction by less invasive surgery. It may be asking too much of DFEO + PTA to maintain knee extension if plantar flexion-knee extension coupling has been compromised by previous lengthening and weakening of the gastrocnemius-soleus complex. The effect of impaired selective motor control may also have an impact on the persistence of flexed-knee gait5. In 1 center, an active campaign to proscribe isolated gastrocnemius-soleus complex lengthening in children with diplegia was followed by a large, sustained reduction in the prevalence of severe crouch gait6. Furthermore, in the children who later progressed to SEMLS, improvements at short-term follow-up were maintained at a mean of 9 years of follow-up7. In these centers, DFEO + PTA may be less frequently indicated. When it is utilized, patellar tendon shortening (PTS) is a viable and simpler alternative to PTA8. PTS has become a safe and reproducible procedure with the advent of strong, synthetic materials that are able to withstand the pull of the knee extensor mechanism8. However, we know of no long-term studies of DFEO + PTS. Boyer and colleagues are to be congratulated on a landmark study, which will provide surgeons in the field with important lessons and much to consider in their practice of gait-correction surgery for persons with CP.
