My Clavicle Fracture Definitely Warrants Surgery—Or Not So Much?

Commentary It is clear that Liu et al. undertook a challenging task in assessing the cost-effectiveness of clavicle fracture surgery by using decision-analysis methodology. The principal conclusion is that internal fixation of a midshaft clavicle fracture is cost effective as long as “clinical benefits” last for greater than 3 years after surgery. However, this study has additional value for a reason perhaps unintended by the author. It serves as an example of how statistics and esoteric analysis can lead to a conclusion that, while appearing to be mathematically robust, may challenge common sense and the experience of clinical practice. The first difficultly with this project is its dependence on complex statistical and economic methodology. It is unlikely that most practicing orthopaedic surgeons are familiar with the intricacies of Markov models, Monte Carlo simulations, or rollback analysis. The temptation is to take the author’s word for it. The average reader will likely conclude that operative treatment of clavicle fractures is cost effective. What exactly this means, however, is buried within the statistical jargon of the text. For example, the “health utility” of operative treatment is said to be greater than that of nonoperative care. But what exactly does this mean? What are the reasons for the increased health utility of the operatively treated subgroup? Is it less pain (for how long?), is it a faster return to work (how much faster?), is it a better or faster return to sports (how much better or faster?), or is it just better subjective patient satisfaction—or perhaps a combination of these factors? Readers must look up references and decide for themselves, which most will not do. Reference 23, an article by Potter et al., is cited 3 separate times to show that delayed treatment of a clavicle fracture has an inferior clinical outcome—and less health utility. However, reviewing the actual article by Potter et al. reveals a somewhat different and more subtle conclusion: “Late reconstruction of nonunion and malunion after displaced midshaft fractures of the clavicle is a reliable and reproducible procedure that results in restoration of objective muscle strength similar to that seen with immediate fixation . . . Both groups rated their satisfaction with the procedure as excellent.”1 There are additional troubling details to consider. How can the cost of nonoperative treatment be $12,377? Even taking into consideration the cost of treating nonunion in the 9.3% of cases in which the fracture does not heal, this number seems illogically high. What is the negative health utility for patients who have surgical complications, such as osteomyelitis or wound breakdown? Several papers have suggested that the reoperation rate following internal fixation of the clavicle is as high as 25% and that not all of these secondary surgical procedures are simply for implant removal2-7. It seems impossible that secondary surgery rates of 25% have no influence on the cost-effectiveness of primary surgical care. The consequences of symptomatic nonunion of a nonoperatively treated clavicle fracture do not include the risks of infection or neurovascular injury, whereas operative intervention that goes awry can seriously and permanently impair the patient. As the authors note, after the 2007 Canadian Orthopaedic Trauma Society (COTS) study was published, clavicle fracture surgery rates increased by 700%, and it is not clear if this was either justified or executed with a reasonably low complication rate8. There are certainly indications for such surgical care, but a lot of published data suggest that the most critical factors relate to patient preferences, lifestyle considerations, personal expectations, and surgeon judgment. It is also worth mentioning that return to work in just 1 week after implant removal, as the authors assume, is completely at odds with clinical reality. It might be necessary to protect the clavicle against stress for at least 6 weeks after removal of a clavicle fracture plate, and none of these patients could return to full-weight-bearing (or lifting) work activities at just 1 week. Compared with a patient who underwent implant removal, the reference (nonoperatively treated) patient would gain at least 6 weeks of wage benefits, even if the patient who had the implant removal went back to work with restrictions for the first 6 postoperative weeks. There are many jobs (construction, police, firefighter, etc.) to which employees cannot return until they are rated by their physician as being “100%” able to perform all work-related activities. The unrealistic return-to-work assumption made by Liu et al. had a direct effect on the cost comparison between the operatively and nonoperatively treated patients, in that it maximized the cost for the latter while minimizing the cost for the former. There is a risk that readers will look at the abstract and conclusions of this paper and presume that there is now more justification than ever to operate on a fractured clavicle. Our patients critically depend on our judgment; nothing less than the public’s trust is at stake. So ask yourself a simple question: If this were my midshaft displaced clavicle fracture, what treatment would I choose? I myself would not envision operative fixation as having an impact on my ability to work or pursue my hobbies of dog training, golf, and woodworking. And when discussing the treatment choices for my patients, I do not anticipate mentioning the impact of stochastic methods, iterative algorithms, and financial engineering. Giving good advice as an orthopaedic surgeon depends the most on knowing your patient as a unique person. Strategy tables and tornado charts will only get us and our patients so far.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.