In this month's Journal, SooHoo et al. report on short-term and intermediate-term complications of ankle fractures that required readmission to a hospital. The authors identified more than 57,000 patients who had been treated as inpatients for ankle fracture in California hospitals during an eleven-year period. From this group, they examined the rates of infection, reoperation, and pulmonary embolism. Data were available for 30,728 patients at a follow-up of five years. The data from this subgroup were queried, specifically with regard to the prevalence of reoperation for the performance of ankle fusion or ankle arthroplasty. The overall rate of short-term complications was low, with a wound infection rate of 1.44%, a pulmonary embolism rate of 0.34%, and a surgical revision rate of 0.82%. A mortality rate of 1.07% was surprisingly high. Not surprisingly, patients with severe medical comorbidities were the patients most likely to have complications.
The investigators used a method that is not commonly used in orthopaedics but that is often used in health-sciences research—mining data in a registry for very large numbers of a common illness or treatment and comparing rates of complications among different groups. The concept is simple; the methods are complex. This method is not often employed in orthopaedic studies because it obscures much detail and does not assess factors that technically oriented surgeons often think are important, such as surgical technique, implant strength, and treatment or rehabilitation strategies. Also, this methodology does not help determine which treatment is best or who requires treatment, and it does not discriminate between problem fractures and simple fractures. Lastly, it is dependent upon the coding accuracy of clerical support staff who may be unfamiliar with patient-care issues.
What it does do, however, is provide data from a completely disinterested investigator mining a completely objective, if imperfect, data source. It provides reliable quantitative data that, in this case, support our qualitative observations. The California discharge data are compiled by California's Office of Statewide Health Planning and Development (OSHPD). The mission statement of the OSHPD is "to promote healthcare accessibility through leadership in analyzing California's healthcare infrastructure,… providing information about healthcare outcomes…"1 Like other government agencies that are struggling with the cost of providing medical care to a population that is older and has more treatable illnesses than at any time in history, the OSHPD is looking for the treatments that are most cost-effective. That should be good for orthopaedics. It is unlikely that there are many treatments that are more cost-effective than open reduction and internal fixation of an ankle fracture, or total hip replacement, or intramedullary repair of a fracture of the femoral shaft. However, this database cannot differentiate between good and suboptimal techniques or between an anatomic reduction and a suboptimal one.
What should the readership of JBJS take from this study? It is worth reading in detail, including the appendices, because it includes information of value. Such information includes the facts, for instance, that a seventy-five-year-old patient with complicated diabetes and an open ankle fracture can be advised that there is a strong likelihood of complications, that the data for complicated diabetes show a worrisome trend for the future, and that a patient without diabetes can be advised that there is less than a 1% chance of end-stage ankle arthritis in the five years after the fracture. In addition, the authors' finding that patients who underwent surgery for ankle fracture had a one-in-three-hundred chance of sustaining a pulmonary embolism raises the question of whether prophylactic therapy, with its attendant morbidity, is justified.
It is worthwhile for surgeons to pay attention to the data that is being collected with regard to their patients, and to ensure, for instance, that a diagnosis of diabetes and peripheral neuropathy is documented. Failure to code comorbidities may make the treatment seem excessive in comparison with the expected complication rate.
It is important that orthopaedic surgeons conduct the kind of research done by SooHoo et al. so that the collected data will reflect the factors that we observe to be important. For a young orthopaedic surgeon who wants to contribute to orthopaedic research in the future, a background in biostatistics and epidemiology may be more helpful than a background in biomechanics, genetics, or biochemistry.
Overall, the data reviewed by SooHoo et al. support our general belief that treatment of ankle fractures is safe and effective. The importance of understanding how the outside world looks at the care we provide is an equally worthwhile message of this manuscript.
*The author did not receive any outside funding or grants in support of his research for or preparation of this work. Neither he nor a member of his immediate family received payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity.
1. Equitable Healthcare Accessibility for California. Office of Statewide Health Planning and Development. http://www.oshpd.ca.gov/General_Info/Mission_and_Values.html#heading. Accessed 2009 Apr 23.