To The Editor:
Regarding "Acute Total Hip Arthroplasty for Selected Displaced Acetabular Fractures. Two to Twelve-Year Results" (2002;84:1-9), by Mears and Velyvis, the indications for total hip arthroplasty for the treatment of acute acetabular fractures and the most reliable technique for achieving fixation of the socket are unclear. Why did the authors decide that intra-articular fractures with ten or more fragments had a poor prognosis? Why not five or eight fragments? On what radiographic view were these fragments counted? In the radiographs of the only patient under the age of thirty years who had a both-column fracture (Figs. 2-A, 2-B, and 2-C), I fail to recognize the multiple fragments mentioned in Table III. In the Discussion, the authors cited the experience of Letournel and Judet, who reported unfavorable outcomes associated with fractures of the anterior wall, the transverse-posterior wall, and the posterior wall-posterior column in comparison with the outcomes of other fracture patterns. However, if one compares Mears and Velyvis's rate of good to excellent results after total hip replacement for the treatment of acetabular fractures (79%) with the rates of such results after open reduction and internal fixation of fractures of the anterior column (78%) and the transverse-posterior wall (73%) 1 , the results of open reduction with internal fixation are not substantially inferior to those of total hip replacement. Since the fracture classification was never meant to be prognostic of outcome but a tool for better understanding the anatomy of the fracture pattern and planning its reduction 2 , the accuracy of the reduction should remain the standard in predicting outcome 3 .
As discussed by Mears and Velyvis, malreduction increases the risk of misplacement of internal fixation. How much nonanatomical reduction of the acetabular fracture is acceptable to minimize the risk of fixation misplacement? With regard to the fixation of acetabular fractures, why do the authors attach so much importance to the medial wall as the primary area for socket stability? Mears and Velyvis advocated cable fixation as an efficient means to achieve this medialization although the majority of the acetabular cups in their series demonstrated an average of 2 mm of vertical displacement and 3 mm of medial displacement at six weeks postoperatively, when the fracture had healed. This emphasis on the medial wall as the primary site for socket stability is even more difficult to understand since most of the literature on acetabular fixation in the presence of deficient bone stock, which occurs in the setting of revision hip surgery 4 and in the treatment of dysplastic hips 5 , has emphasized the importance of the rim fit, with some studies advocating perforation of the medial wall to achieve it.
Although I would agree that the results appear favorable for elderly patients, use of this treatment choice should be approached with extreme caution in the case of younger patients. At a mean follow-up of only 8.1 years, 19% of the hips had radiolucencies and 21% also had acetabular osteolysis. Finally, with 16% of the hips demonstrating mean wear of >2 mm at a mean follow-up of 8.1 years (i.e., >0.2 mm of wear per year), the risk of revision surgery is not negligible 6 .
I thank Dr. Mears and Dr. Velyvis for sharing their valuable clinical experience with these injuries, which are particularly difficult to treat.
D.C. Mears and J.H. Velyvis reply:
We thank Dr. Beaulç?¦or his interest in our study. We acknowledge that the indications for a total hip arthroplasty in the setting of an acute displaced acetabular fracture remain controversial. The critical issue seems to be our contention that certain displaced acetabular fractures possess complicating features such that an open reduction and internal fixation is not likely to produce a favorable clinical outcome. In their detailed report, Letournel and Judet 1 documented the clinical outcomes of open reduction and internal fixation according to selected acetabular fracture patterns. Whereas the rates of excellent, very good, and good results following treatment of transverse fractures totaled 95%, the corresponding rates of these results were 67% for treatment of anterior column fractures and 47% for posterior column-posterior wall fractures. Eighty-eight percent of the patients aged ten to twenty-nine years achieved an excellent or good result, whereas 68% of those aged sixty to sixty-nine years had such results. Similarly, Matta 3 reported a comparable age-related trend, with excellent or good results in 81% of his patients under forty years of age but in only 68% of those over the age of forty years. These authors did not provide a similar tabulation of clinical results for patients over the age of sixty years. In procedures of such recognized complexity, the expected results achieved by less experienced surgeons are unlikely to be superior to those reported by these authors. Furthermore, given the rapidly aging population, increasing numbers of older patients are likely to sustain these fractures and, thereby, further confound the anticipated clinical outcomes. Once a subset of patients in whom an open reduction has less than a 50% likelihood of an excellent or good result is identified, despite the fact that the patients are undergoing a major operation with the risk of diverse postoperative complications, therapeutic alternatives merit serious consideration. Once this hypothesis is accepted, the specific criteria for implementation of an alternative treatment merit review. The principal alternatives remain initial conservative treatment or an acute total hip arthroplasty.
In our paper, we defined the indications that we used for an acute total hip arthroplasty. Since the available literature has not identified such a characterization of the factors that led to poor outcomes after open reduction and internal fixation, the criteria in our series are based solely upon our prior clinical experience, along with that encountered in our review of other studies 1,3 .
For our patients, we employed preoperative guidelines based upon multiple radiographic views and computed tomographic scans to determine the potential value of an acute total hip arthroplasty; ultimately the decision rested upon intraoperative scrutiny of the hip joint. While the principal indications were extensive impaction of the femoral head and impaction of the acetabulum involving the weight-bearing dome, the degree of comminution that resulted in ten or more intra-articular fragments was a relative indication for an acute total hip arthroplasty. A minority of our patients who did not have an extensive area of impaction had notable intra-articular comminution, typically in excess of ten bone fragments, as assessed intraoperatively. We agree that plain radiographs limit the actual amount of osseous comminution that can be seen, while computed tomographic scans over-exaggerate it.
We agree that Letournel and Judet's classification 1 of acetabular fractures was intended primarily to serve as an anatomical guide to the region of injury and, thereby, as an operative planning tool, not as an outcome measure. Nevertheless, Letournel and Judet chose to characterize their clinical and radiographic results according to the type of injury. We have merely reported the extraordinary variation of our clinical results according to the type of injury. Letournel and Judet were well aware of the impact that selected injury patterns, patient age, the presence of osteoporotic bone, and other factors could have on the likelihood of a favorable clinical result. Nevertheless, they employed this preoperative information primarily to select a specific strategy of operative fixation. While Letournel and Judet chose to focus on the accuracy of reduction as the basis for predicting the anticipated clinical outcome, a critical review of their results illustrates the marked limitations of this time-honored concept, especially when applied to results in elderly patients in whom frequent complications of marked impaction of the femoral head and acetabulum are not subject to anatomical restoration with the use of currently available techniques. Admittedly, for younger patients and others with dense bone and relatively simple fracture patterns, their concept remains valid.
Our study demonstrated, at a mean follow-up of 8.1 years, a favorable outcome after acute total hip arthroplasty for the treatment of acetabular fractures that we believe would have had an extremely poor result after open reduction and internal fixation.
With respect to the adjunctive use of internal fixation to buttress the cup, clearly the critical site or sites of support depend upon the fracture pattern. Whereas a posterior wall fracture with extensive marginal impaction can be treated with augmentation and buttressing of the posterior rim or wall alone, a both-column fracture requires much more extensive immobilization. Our focus upon the medial wall pertains to the most typical geriatric fracture patterns that we have encountered, namely, both-column, anterior column-posterior hemitransverse, and fractures of the anterior wall that involve the quadrilateral surface of the ilium. Unless both sites in these fracture patterns are effectively stabilized, a marked secondary displacement of the cup with loosening is likely to ensue. The cabling technique that we used provided a minimally invasive strategy to immobilize these diverse injury patterns. The general principle of the technique was to restore a stable osseous construct, with an approximate realignment of the major fracture fragments to within 1 to 5 mm of their anatomical origin. Residual gaps were obliterated with use of autologous bone graft from the femoral head. Intraoperative image intensification was used to confirm this realignment.
While our series included both elderly patients with osteopenic bone who sustained low-energy trauma and younger patients who had had major traumatic insults, as we emphasized in our paper, an acute total hip arthroplasty is much more readily advised for an elderly patient with a limited activity level and life expectancy than for a young individual. Nevertheless, for a responsible young adult who sustains an acetabular fracture with extensive impaction of the femoral head and acetabulum, an acute total hip arthroplasty may be the only therapeutic option available that will permit a return to most activities. Certainly, in the presence of a long life expectancy, the anticipation of arthroplastic failure and the need to undertake one or more exchange arthroplasties is a valid limitation of the method.