“Hmmm…. Maybe I’m operating on too many ankle fractures.” That was my first thought as I read the abstract of the recent Willett et al. study in JAMA. They conducted a well-designed, randomized controlled trial that compared operative and nonoperative treatment of unstable ankle fractures, using the Olerud-Molander Ankle Score at 6 months postoperatively as the primary outcome measure.
On the surface, it appeared as though patients who were treated nonoperatively with close contact casting did just as well as those who underwent operative intervention. This seemed to be not only the case with the primary outcome measure, but also with secondary outcomes such as quality of life, pain, and patient satisfaction. “Do less” appeared to be the main message of the abstract. However, I became more skeptical after critically reading the entire article.
First off, the study was designed to determine differences between treatment groups, not to prove that they were equivalent. Finding no difference is not the same as showing equivalence, and the article did the former, not the latter.
There are also a few things about the study that may limit the wide applicability of the findings and provide some solace to surgeons like me who feel that fixing unstable ankle fractures provides superior outcomes. First, only initial radiographs were used to determine who had unstable ankle fractures. Stress radiographs were an exclusion criterion, so for the many ankle fractures that require such imaging to determine instability, the results from Willett et al. may not apply.
Second, the study was designed to compare these treatments in older adults. The mean ages of operative and nonoperative groups were 69.8 and 71.4 years old, respectively, and almost 75% of both groups were female. While bone density was not measured in either group, it is likely that many patients included in this study had osteoporotic disease, which introduces another potential variable when interpreting the findings.
Furthermore, nearly 20 percent of all patients who initially were treated with casting developed some type of complication that required conversion to surgical fixation. This finding, plus the fact that all casts were applied by surgeons in the operating room with patients under general or spinal anesthesia, suggests that treating unstable ankle fractures with surgical fixation in a single visit would perhaps provide the most definitive treatment.
So, I will probably continue to offer patients with unstable ankle fractures surgical fixation. I have never tried the close contact casting that was described in the article, and I suspect, despite the authors’ claim of evidence to the contrary, that there is a significant learning curve associated with that technique. If about one out of every five patients I perform casting on as definitive treatment ends up needing additional procedures, I am not sure I have done the patient justice. While this study provides interesting evidence and may apply to a small subset of my older patients, I think it has limited applicability in other patients who present with unstable ankle fractures.
Chad Krueger, MD is a military orthopaedic surgeon at Womack Army Medical Center in Fort Bragg, North Carolina.
* * * *
The Willett et al. study in JAMA indicates that some patients older than 60 years with unstable ankle fractures can be treated by modified casting alone, without the need for operative stabilization and fixation. The study protocol allowed conversion to surgery among patients randomized to casting if reduction was not possible during the initial procedure or was lost within the first 3 weeks.
One hundred surgeons applied close contact casting at 24 major trauma centers and general hospitals in the UK. After 6 months, the mean Olerud-Molander Ankle Score was 66.0 in the surgery group vs 64.5 in the casting group—no significant difference in the primary outcome.
Secondary outcomes showed that the rate of radiographic malunion was 15% in the casting group compared with 3% in the surgery group. Conversion from casting to operative treatment was high: of the 311 patients randomized to casting, 70 (23%) were ultimately treated by internal fixation, including 18 never treated with close contact casting and 52 who lost reduction and required conversion to internal fixation. Rates of infection and wound complications were 10% in the surgical group versus 1% in the casting group. Additional operating room procedures were required in 6% of the surgery group and 1% of the casting group. Casting required less operating room time compared with surgery.
The overall similarity in clinical outcomes in this study challenges the importance of restoring exact ankle-joint congruence in older adults and suggests that function and pain are not as closely related to malunion as many clinicians believe. Neither method yielded an entirely satisfactory outcome in older adults. In older patients with lower demand, shorter life expectancy, lesser bone and tissue quality, and diminished capacity for healing, the rates of delayed or infected wound healing and loss of implant fixation are greater.
Casting may be an imperfect alternative to surgery particularly in developing countries. One must remember, however, that plaster technique is an art. Achieving the successful outcomes with close contact casting as described by Willett et al. is likely to pose a learning curve. Further studies are needed to identify which specific patients are most and least likely to benefit from casting.
Shahriar Rahman, MS is a consultant orthopaedic surgeon at the Ministry of Health & Family Welfare in Bangladesh.