Invited Commentary related to: Cost-Utility Analysis of Reconstruction Compared With Primary Amputation for Patients With Severe Lower Limb Trauma in Colombia
Significant medical and surgical advances have improved the ability of physicians and surgeons to reconstruct severely injured lower extremities. Although the concurrent development of bioengineering advances in below-the-knee prosthetics and integrated orthotics has resulted in much better function for these patients, these advances have also contributed to a therapeutic dilemma for treating surgeons.5 Determining which treatment algorithm is best for the injured patient—for the short and the long term, in a relatively short period of time, and under stressful circumstances—is one of the most difficult decisions orthopedic trauma surgeons are asked to make. In each case, the surgeon must act on what he or she believes to be the best available information and in the service of the best interests of his or her patient. To do this, surgeons must stay abreast of the scientific findings related to the functional and technical outcomes of these treatments and the long-term financial ramifications associated with them.
To satisfy the objectives of this study, the authors built a Markov model (a model of randomly changing systems in which it is assumed that future states depend only on the current state and not on the events that occurred before it) that included different short-term and long-term states that represented the main events that a patient could experience after treatment, with a 50-year time horizon for the base case.6 The health outcome selected was quality-adjusted life years. Somewhat surprisingly, the authors found that, when limb reconstruction was compared with primary amputation, it was the dominant strategy; in other words, reconstruction provided more quality-adjusted life years, and it did so at a lower cost. This result changed only when the time horizon was less than 9 years, when the probability of a secondary amputation was greater than 57%, or when the probability of infection was greater than 59%.
Several important limitations of the current study are important to note. First, considerable heterogeneity was present in the outcomes of the studies used to create this model.7 Second, the data used in this model were taken from studies of patients from countries other than Colombia because no other information about the behavior of Colombian patients with these injuries is available.