The last 50 years of pediatric fracture management may have included a substantial misunderstanding of Salter and Harris’s original findings.
In a recent issue of SMART EM we went under the evidence sea and into the depths of pediatric orthopedics, where there is a question trainees (and parents) with overdeveloped common sense have been asking for a half century: why do we immobilize bones that are clinically and radiographically normal?
The infamous Salter-Harris type I fracture has, in many settings, become the most common type of pediatric ‘fracture’. And yet the diagnosis is, in the overwhelming majority of cases, provisional… preliminary... potential. That is because Salter-Harris I fractures are typically diagnosed in the setting of normal x-rays and subjective bony tenderness at a growth plate. In other words, when there is no objective evidence of fracture.
Turns out Drs. Robert Salter and Robert Harris may have been thinking about something quite different. In the seminal 1963 publication outlining their classification system the two pediatric orthopedists offer an explanation of type I fractures as “resulting from a shearing or avulsion force”, and “most commonly associated with birth injuries.” They go on to describe their management, saying “reduction is not difficult.” The paper includes 12 reproduced radiographs and one illustration of type I fractures. All of these include visually conspicuous separation of the epiphysis from the metaphysis in the absence of bony fracture.
In other words Salter and Harris were describing instances of conspicuous displacement and total separation of the growth plate, and at no point do they mention or even suggest the possibility of radiographically negative variants.
Of course, it is always possible that the authors meant for the management of this type of injury to be extended and extrapolated to non-displaced growth plate injuries. If so, this would have been an attempt to prevent growth disturbances and, ultimately, functional or esthetic sequelae, since this was the intent of their classification system. Which led us to another question of evidence: how often does a pediatric bone injury result in functional or cosmetic sequelae when nondisplaced, radiographically negative growth plate injuries are not immobilized?
The answer to this question was mercifully unambiguous: never. At least, to our knowledge, no such instance has ever been reported in the medical literature. In contrast, a number of studies have now documented the management of presumed Salter-Harris type I injuries with early mobilization, functional splinting, or as-desired orthopedic supports. In all cases outcomes are excellent, while revisits and splint complications are essentially vanquished.
The bottom line is that we were able to find no literature to suggest that immobilization is beneficial for Salter-Harris type I fractures. Therefore particularly in the case of those that appear to show no signs of clinical injury that would benefit from immobilization (minimal swelling, no deformity, able to weight bear, etc.), there seems to be no benefit to splinting or casting.
Perhaps, then, the last fifty years of Salter-Harris type I fracture management has really just been a misunderstanding. Have a listen, see what you think. Maybe those trainees and parents, with their silly common sense, have a point after all.
David H. Newman, MDAuthor of Hippocrates’ Shadow: Secrets From The House Of Medicine