Distal radial fractures account for 20% of fractures among Medicare enrollees1 and are second only to hip fractures in terms of incidence in the elderly population. These fractures represent approximately one-sixth of all fractures that are treated in emergency departments2. The lifetime risk of sustaining a distal radial fracture is 15% for women and 2% for men2.
The current body of literature regarding distal radial fracture care is relatively large, but it is inadequate. The Cochrane Collaboration, which evaluates the literature for evidence-based practices, reviewed distal radial fracture treatment and concluded that there is not enough evidence to decide best practices regarding (1) conservative interventions, (2) surgical interventions, (3) anesthetic considerations, (4) rehabilitation, and (5) closed reduction techniques3-7.
There has been recent enthusiasm regarding open reduction and internal fixation of these fractures with use of a volar-inserted locked plate8-12. There is a feeling among some surgeons that the use of open reduction and internal fixation with these locked plates is associated with a lower complication rate, better function, and greater patient satisfaction than percutaneous fixation, particularly external fixation, for the treatment of distal radial fractures12.
During the administration of Part II (the oral examinations) of the American Board of Orthopaedic Surgery (ABOS) over the past several years, it has been observed that candidate orthopaedic surgeons appear to be stabilizing distal radial fractures preferentially with open treatment rather than with percutaneous fixation. We decided to explore this phenomenon with use of the ABOS Part II database.
The purposes of the present study were (1) to document the changing use of open treatment as compared with percutaneous fixation for the operative treatment of distal radial fractures and (2) to evaluate the surgeon self-reported outcomes and complication rates associated with each of these surgical options.
In preparation for the oral certification process for the ABOS, candidates submit a list of all surgical cases performed over a defined six-month period (July through December) within their preceding twenty-two-month period of practice. Seventy percent of these candidates take the oral examination within two years after completion of their training (residency or fellowship), and 98% take the examination within five years after completion of their training. All data are entered into a secure Internet-based database. As the examination occurs in July of the following year, long-term outcomes are not available.
The collected information included the date of the procedure, the geographic location of the candidate's practice, ICD-9 (International Classification of Diseases, Ninth Revision) diagnosis codes, Current Procedural Terminology (CPT) surgical procedural codes, de-identified patient demographic data such as age and sex, surgical and medical complications, and four surgeon-reported outcome measures (pain, deformity, function, and patient satisfaction).
The scales that were used for the outcome measures were as follows. Pain was described as increased, unchanged, decreased, or absent; deformity was described as increased, unchanged, improved, or normal; function was described as decreased, unchanged, improved, or normal; and patient satisfaction was described as poor, fair, good, or excellent. These outcomes are physician-derived and generic as they are meant to apply to the full spectrum of orthopaedic procedures and to give a very general sense of the success or failure of a procedure. These scales are provided to candidates by means of drop-down menus during the online submission of case data; however, candidates are not provided with specific definitions.
Each candidate is allowed to enter these outcomes on the basis of his or her own subjective patient assessment. There are no patient-derived (self-reported) data, and the assessment period for each patient can vary from a few weeks to a few months. All cases are included, even those lacking postoperative follow-up. Complications are also physician-reported and are entered from drop-down menu lists containing medical and surgical categories. In 2004, the category of "surgical unspecified" was added to the complication list to capture the occurrence of rare adverse outcomes associated with the surgical procedure.
The ABOS Part II database was searched for all distal radial fractures (ICD-9 codes 813.42, 813.44) for the years 1999 through 2007. The procedures that were performed for the treatment of distal radial fractures were separated into two groups on the basis of CPT codes: (1) those involving percutaneous skeletal fixation with or without external fixation (CPT code 25611 [in 2007, this CPT code was changed to 25606]) and (2) those involving open treatment with or without internal or external fixation (CPT code 25620 [in 2007, this CPT code was changed to 25607-25609]). It should be noted that the database provides no information regarding surgical approach (volar or dorsal) or fixation method (plate or fragment-specific). The relative proportions of the two procedures were compared for each calendar year. The United States was divided into six geographic regions, loosely based on the regions of the United States Census Bureau, for a geographic analysis13. Comparisons were made between the two procedural groups for all outcome measures, including complications.
This research protocol was reviewed by the institutional review board of the Dartmouth-Hitchcock Medical Center and was judged to be exempt from requiring consents. Candidates applying for this examination are informed by the ABOS that the de-identified data that they submit as part of the voluntary process of Board certification may be used for research purposes.
Statistical Methods
Parametric comparisons were conducted with use of unpaired two-tailed t tests for two-sample comparisons. Fisher exact tests were performed for contingency tables. The Mann-Whitney U test was performed on the outcome data elements. The p values indicated in the results for the pain, function, deformity, and satisfaction scales are the results of the Mann-Whitney U tests performed on each ordinal scale for the four ordinal values. The level of significance for all tests was set at p < 0.05.
Over the nine-year period (1999 to 2007), a total of 12,061 distal radial fractures (ICD-9 813.42 and 813.44) were treated surgically by candidates during the case-collection periods. A total of 6714 candidates submitted case lists, of whom 3621 performed no operative fixation for this type of fracture. Therefore, the 12,061 fractures were treated by 3093 candidates, who treated an average of four (range, one to thirty-nine) distal radial fractures during their six-month case-collection period. The number of candidates performing distal radial fracture surgery varied by year. Three hundred and twenty-four candidates performed these procedures in 1999; the number of candidates performing these procedures peaked at 400 in 2002 and dropped to 299 in 2007. However, the number of distal radial fractures treated operatively per candidate increased almost every year. The average number of distal radial fractures treated operatively per candidate was 3.2 in 1999, 3.4 in 2002, and 5.2 in 2007 (Table I).
Percutaneous fixation was used in 5510 cases, and open treatment was used in 6551 cases. For the entire cohort, women were more likely to undergo percutaneous fixation than men were (48% compared with 43%; p < 0.0001). Men undergoing percutaneous fixation were younger than those undergoing open treatment (thirty-five compared with forty years; p < 0.0001); there was no age difference in women.
For 2004 to 2007, we are able to match candidate cases with each candidate's declared subspecialty. For this four-year period, the percentage of fractures that were stabilized with open treatment was significantly greater for hand-fellowship-trained surgeons (84%; 1747 of 2085) than for non-fellowship-trained surgeons (57%; 2131 of 3722) (p < 0.0001). Although the percentage of fractures that were stabilized by means of open treatment increased each year for both groups of surgeons, the percentage was always higher for hand-fellowship-trained surgeons. For hand-fellowship-trained surgeons, the percentage of fractures that were stabilized by means of open treatment was 70%, 84%, 87%, and 90% for 2004, 2005, 2006, and 2007, respectively. For non-hand-fellowship-trained surgeons, the corresponding percentages for these years were 41%, 54%, 62%, and 73%.
From 1999 to 2007, there was a distinct shift in surgical technique. Nationwide, the use of open treatment increased from approximately 42% in 1999 to 81% in 2007 (p < 0.0001) (Table I, Fig. 1). While this transition from percutaneous fixation to open treatment was seen in nearly all regions of the country, it did not occur at the same pace or to quite the same extent in all regions. The Northeast had the highest percentage of open treatment in 1999, 2001, 2005, and 2007; the Southwest had the lowest percentage in 2001 and 2007.
In terms of physician-perceived outcomes, pain at the time of follow-up was absent for 36% of all patients who had been managed with percutaneous fixation compared with 32% of those who had been managed with open treatment (p < 0.0001). Seventeen percent of the patients who had been managed with percutaneous fixation had normal function, compared with 12% of those who had been managed with open treatment (p < 0.0001). Thirty percent of patients who had been managed with percutaneous fixation had no deformity, compared with 32% of those who had been managed with open treatment (p = 0.0058). Thirty-seven percent of the patients who had been managed with percutaneous fixation had excellent satisfaction, compared with 36% of those who had been managed with open treatment (p = 0.0909).
The overall complication rate was higher for patients who had been managed with percutaneous fixation than for those who had been managed with open treatment (14.0% compared with 12.3%; p = 0.006) (Table II). Significant differences were found in the infection rate and the nerve palsy and/or injury rate between the two techniques. There was a higher infection rate in association with use of percutaneous fixation (5.0% compared with 2.6%; p < 0.0001) and a higher nerve palsy and/or injury rate in association with open treatment (2.0% compared with 1.2%; p = 0.001). No other differences in complication rates were found between the two techniques.
We found a distinct shift in surgical techniques for distal radial fracture fixation that occurred over a nine-year period in the group of surgeons who were candidates for ABOS Part II certification. In 1999, the majority (58%) of distal radial fractures were treated with percutaneous fixation; by 2007, 81% of such fractures underwent open treatment. It is unclear why surgical treatment for distal radial fractures changed so substantially in such a short period. There are many possible influences, and the cause is probably multifactorial. This dramatic shift may have been driven by the introduction and reported good results for volarly inserted locked plates for the treatment of distal radial fractures9-11. Many surgeons believe that open reduction and locked volar plate fixation provides more stable fixation and allows earlier range of wrist motion than percutaneous fixation does12.
It may be that younger surgeons are responding to a change in training and that more recent graduates are being trained preferentially in open reduction and plate fixation techniques for distal radial fractures. Whether that is true is unknown, and, if it is true, it merely pushes the "why?" question back to the educators. There seems to be a certain intrinsic attraction to newer technologies related to fixation devices and surgical techniques. Younger surgeons may perceive certain pressures to offer new techniques to a medical market that is constantly searching for the latest in technological advancement.
Other possible reasons why surgeons have shifted from percutaneous fixation to open treatment are that (1) the difficulty of caring for fractures that have been treated with percutaneous fixation can lead to a concern about pin-track infection, (2) the sight of the percutaneous fixation may be more frightening to the average patient than the site of a sutured wound is, (3) percutaneous fixation can interfere with clothing and is cumbersome for the patient, (4) it is generally easier to begin early motion after open treatment than after percutaneous fixation, and (5) open treatment may obviate the need for implant removal.
It is unlikely that physician reimbursement is driving the change. According to the 2006 Resource-Based Relative Value Scale, percutaneous fixation (CPT code 25611) is worth 18.09 total relative value units (RVU) whereas open treatment is worth 17.24 total RVU. Using a Medicare conversion factor of approximately $38 per RVU, this resulted in a pay differential of approximately $32 in favor of percutaneous fixation. If a surgeon believed that either treatment was acceptable for a particular fracture, it is unlikely that he or she would choose the procedure that provided poorer reimbursement, even if the difference was very small, as in this case.
In terms of physician-perceived outcomes, a higher percentage of patients who had been managed with percutaneous fixation had no pain and normal function but some deformity as compared with those who had had open treatment. Although most of these differences were significant (p < 0.01), probably as a result of the size of the cohort, the magnitude of these differences was small and therefore probably has little, if any, clinical importance.
The literature on outcomes after distal radial fractures is controversial, with only a few comparative studies. Grewal et al. performed a prospective, randomized, controlled study of sixty-two patients in which dorsal plate fixation was compared with external fixation with limited reduction and pin fixation14. There were no significant differences between the groups in terms of patient-based outcome scores; however, the dorsal plate group had significantly weaker grip strength and a higher number of complications, especially tendinitis and the need for hardware removal. Kreder et al. performed a prospective randomized study of 179 patients in which external fixation with indirect reduction and percutaneous pin fixation was compared with open reduction and internal fixation; although patient-based outcome scores were no different between the groups at two years, the external fixation group had better outcomes at the six-month interval15. Wright et al. performed a case-control study of thirty-two patients in which the use of a volar fixed-angle plate was compared with external fixation; significantly better radiographic outcomes were found in association with use of the plate, but there was no difference between the groups in terms of patient-based outcomes16. In a recent prospective, multicenter, randomized study of 137 patients with intra-articular fractures of the distal part of the radius, Leung et al. reported that those who had been managed with plate fixation had a better cosmetic result and function at two years (as rated with the Gartland and Werley scoring system) than did those who had been managed with external fixation combined with percutaneous pin fixation8.
In terms of surgeon self-reported complications, there was a significantly higher overall complication rate in the percutaneous fixation group than in the open treatment group (14.0% compared with 12.3%). The percutaneous fixation group had a significantly higher infection rate (5.0% compared with 2.6%), whereas the open treatment group had a significantly higher nerve palsy and/or injury rate (2.0% compared with 1.2%). No other differences in the complication rates were found between the two groups. These results can be compared with those reported by other authors. In a randomized clinical trial in which spanning external fixation was compared with locked volar plate fixation for the treatment of distal radial fractures, Egol et al. reported that the overall complication rate was substantially higher for the group treated with external fixation than for the group treated with locked volar plate fixation (9.8% compared with 4%)17.
The present study had several limitations. It involved the use of a retrospective database, with all of the problems inherent with this methodology. Similar to most database projects, users cannot independently verify the accuracy of the data, its standardization, or its input. The ABOS database does not include detailed clinical information such as medication history, the severity of the associated comorbidities, lifestyle factors, body mass index, or radiographic information such as fracture displacement, comminution, or the degree of osteoporosis. Therefore, we do not have all of the information necessary to fully evaluate the appropriateness of specific interventions or to control for all relevant patient factors that may affect the outcome after distal radial fracture fixation.
Although we had intended to compare the use of external fixation with the use of locked volar plate fixation for the stabilization of distal radial fractures, the CPT codes that are available are not specific with regard to the choice of implant. The two codes used are much more generic: CPT code 25611 represents percutaneous skeletal fixation of a distal radial fracture with or without external fixation, and CPT code 25620 represents open treatment of a distal radial fracture with or without internal or external fixation. Therefore, there is no way to determine specifically what type of percutaneous fixation or open treatment was utilized. Although we believe that the first code, 25611, usually refers to external fixation, it can also be used to indicate Kirschner-wire fixation. Similarly, although it has been our observation over the last several years that the code 25620 usually refers to locked volar plate fixation, it could refer to nonlocked volar plating, dorsal plate fixation, or fragment-specific fixation.
Another potential limitation is that the data reflect the experience of a homogeneous, young group of orthopaedic surgeons. It might be hazardous to generalize any of the findings to the entire body of practicing orthopaedic surgeons. The available data only include information on the distal radial fractures that were treated operatively; no information is provided on the (probably large) number of fractures treated nonoperatively. Therefore, there is no way to determine the total proportion of distal radial fractures that were treated surgically.
A major limitation of the present study is the portion that relies on outcome data entered by the candidates themselves. As noted above, the four outcome categories (pain, function, deformity, and patient satisfaction) are not clearly defined, are not scientifically validated, and are dependent on subjective judgments that are self-reported by the candidate surgeons. That is, because the surgeons use self-applied outcome scales and may have a stake in better outcomes, their reporting may not have been as accurate as that of an unbiased observer. Finally, the duration of follow-up was short and was always less than one year. Although all of these limitations are valid, they apply equally to patients managed with either surgical technique.
The criticism of subjectivity and bias also may apply to the surgeons' reporting of complications. As patient follow-up was variable and short, it is quite possible that not all complications were detected; the final complication rates could be substantially higher as patients have longer follow-up. Those limitations would also apply equally to both patient groups.
However, because of the high-stakes nature of this examination and the candidates' awareness that they will have to present and defend charts and radiographs for a certain number of cases, we believe that candidates likely go to great lengths to be accurate and honest.
In any event, the primary findings of the present study do not depend on the outcome scales. The complications lists are relatively specific and objective, although they do include the category of "surgical unspecified" to capture adverse occurrences related to surgery that may be rare or not on the list.
In conclusion, a striking shift in fixation strategy for distal radial fractures has occurred in the United States over a brief period of time among younger orthopaedic surgeons. This shift has occurred despite a lack of improvement in surgeon-perceived functional outcomes. Regardless of the fixation method (percutaneous or open), the rate of complications associated with operative fracture treatment was 12% to 14%. Additional studies should be performed to determine the optimum type of fixation for the treatment of distal radial fractures. 