Abstract
Background: Knowledge of the location of the weight-bearing portion of the acetabulum would assist orthopaedic surgeons in the treatment of acetabular fractures. With use of controlled experimental transverse, anterior column, and posterior column osteotomies, we investigated the weight-bearing region of the acetabulum.Methods: Twenty-four fresh-frozen cadaveric hip joints were tested. Simulated transverse fractures were evaluated in twelve specimens, and simulated anterior column and posterior column fractures were tested in six specimens each. Each femur and acetabulum was potted and mounted in an aluminum fixture, with the acetabulum positioned in 25 degrees of flexion and 20 degrees of abduction. Each specimen was tested intact and after successive osteotomies. The transverse osteotomies had roof-arc angles of 60, 50, 40, and 30 degrees. The anterior column and posterior column osteotomies were classified as very low, low, intermediate, or high. Compressive loading to 800, 1200, and 1600 newtons was performed four times for each intact specimen and after each osteotomy. A specimen was considered to be stable if no gross dislocation occurred during any of the four loading cycles. Translation of the femur within the acetabulum also was measured during each trial.Results: The number of stable specimens decreased both with higher applied loads and with more superior osteotomies. The stability of the hip was significantly affected by both the location of the fracture and the magnitude of the applied load (p < 0.00005). Translation of the femur within the acetabulum increased with higher applied loads and with more superior osteotomies.Conclusions: Fractures that have a medial roof-arc angle of 45 degrees or less, an anterior roof-arc angle of 25 degrees or less, or a posterior roof-arc angle of 70 degrees or less cross the weight-bearing portion of the acetabulum and necessitate operative treatment.Clinical Relevance: Fractures exiting the posterior column just superior to the ischial spine and those exiting the anterior column through the iliac wing lead to instability. When a patient has a displaced fracture crossing this area, every effort should be made to achieve anatomical reduction in order to prevent posttraumatic arthritis.
Unreduced fractures crossing the weight-bearing dome of the acetabulum lead to arthritis. The etiology of this arthritis is not known. The incongruity may lead to instability or to elevated contact stresses. Thus, the integrity of the weight-bearing dome is considered to be an important prognostic indicator of outcome after an acetabular fracture. However, what actually constitutes the weight-bearing dome remains ill defined3,5,10,11,17. An unreduced fracture that crosses high in the acetabulum predisposes the hip to posttraumatic arthritis, whereas one that crosses low does not3,17,20. Unfortunately, many fractures cross the acetabulum between these two obvious extremes. The level at which an acetabular fracture begins to infringe on the weight-bearing area is poorly defined.
One way to define the weight-bearing dome is as the portion of the acetabulum that is necessary to maintain the stability of the hip. Numerous clinical and laboratory studies have implicated articular instability in the pathogenesis of posttraumatic arthritis1,4,6-9,18,19,24. Most authors have agreed that acetabular fractures resulting in clear clinical or radiographic instability should be treated operatively3,5,10,11,22. However, stability is not always easy to determine. For example, all radiographs of a transverse fracture that crosses the acetabulum just superior to the fovea may show that the femoral head is concentrically reduced even though weight-bearing forces would cause the hip to subluxate. Moreover, the level at which a fracture crossing the acetabulum begins to affect stability is not known.
The purpose of the present study was to examine how displaced transverse, anterior column, and posterior column fractures at various levels of the acetabulum affect the stability of the hip joint. Ultimately, this study should provide a clearer definition of which fractures are important in the pathogenesis of posttraumatic arthritis.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. Funds were received in total or partial support of the research or clinical study presented in this article. The funding source was a research grant from the Orthopaedic Trauma Association.
†Massachusetts General Hospital, 55 Fruit Street, WAC 517K, Boston, Massachusetts 02114. E-mail address: mvraha@partners.org.
‡Bioengineering Laboratory, Department of Orthopaedic Surgery, Louisiana State University Medical Center, 2025 Gravier Street, Suite 400, New Orleans, Louisiana 70112. E-mail address for Dr. Thomas: kthoma@lsumc.edu.
Hips were obtained from donors who had no history of osteoarthritis. To provide a standard reference for alignment in the sagittal, coronal, and transverse planes, two Steinmann pins were inserted into each hip before the pelvis was separated along the midsacral axis, forming a left-right pair. Each specimen was examined visually and radiographically for evidence of any abnormalities of the hip and then was frozen. Before testing, the specimen was thawed overnight at room temperature. The femur then was separated from the acetabulum, and all muscles and connective tissues, including the hip capsule, were removed. The acetabular labrum was left intact.
The method of mechanical testing was similar to that described previously21. Each femur was potted in a cylindrical fixture with use of casting plaster. The femur was oriented vertically in slight adduction so that the load that was applied through the femoral head was directed down the femoral shaft. The fixture containing the femur was mounted onto an instrumented x-y displacement table (model LS 403; Micro Slide, Haupauge, New York) and secured to the load cell of a mechanical testing machine (858 Bionix TestStar System; MTS Systems, Minneapolis, Minnesota). The position of the x-y table was set to zero at the start of each loading cycle, and therefore the table gave a continuous readout of the translation of the femur during testing.
Each acetabulum was potted in a rectangular fixture with use of casting plaster and was mounted onto the testing apparatus in 25 degrees of flexion and 20 degrees of abduction. This position was chosen on the basis of the findings of Pedersen et al.16, who demonstrated that the magnitude and direction of the resultant force across the hip varies during the gait cycle. With the acetabulum mounted in the position just described, the orientation of the applied load corresponded to that of the peak resultant vector across the hip, which occurs just before toe-off.
A compressive preload of fifty newtons was applied manually and then additional compression was applied at a rate of 200 newtons per second in load control until a limit of 800, 1200, or 1600 newtons was reached. Each specimen was loaded four times to each of these three limits. Loads of 800 and 1200 newtons typically are seen when a patient is lying supine during bed rest, whereas 1600 newtons approaches the peak load typically seen during gait. Simultaneous recordings of load, actuator displacement, and translation of the femur were made during each test. A specimen was considered to be stable if no gross dislocation (that is, complete dislocation of the femoral head from the acetabulum) occurred during any of the four loading cycles. Gross dislocation was visually obvious and also was characterized by extreme translation of the femur on the x-y table and an immediate decrease in the measured load.
Medial, posterior, and anterior roof-arc angles, as defined by Matta et al.10, were measured to characterize the various osteotomies. The medial roof-arc angle was measured on an anteroposterior radiograph, the anterior roof-arc angle was measured on an obturator oblique radiograph, and the posterior roof-arc angle was measured on an iliac oblique radiograph (Figs. 1-A, 1-B, and 1-C).
The simulated transverse fractures were created with use of custom-designed guides that produced cuts with various roof-arc angles (60, 50, 40, and 30 degrees)21. The simulated anterior column and posterior column fractures were created in a freehand fashion by making cuts that corresponded to clinically observed fracture patterns. The anterior column fractures were classified as very low (at the level of the iliopubic eminence), low, intermediate (between the anterior inferior iliac spine and the anterior superior iliac spine), and high (Fig. 2-A). Similarly, the posterior column fractures ranged from very low (just superior to the ischial spine) to high (at the level of the greater sciatic notch) (Fig. 2-B). Each specimen was tested intact and then after creation of the osteotomy.
For each type of simulated fracture (transverse, anterior column, and posterior column), relationships among the percentage of stable specimens, the location of the osteotomy, and the magnitude of the applied load were examined with use of frequency-table analysis (Pearson's chi-square or Fisher's exact test).
The data on femoral translation, recorded by the instrumented x-y table, were analyzed by determining the maximum distance traveled during each of the loading cycles, independent of the path (calculated as the straight-line distance from the origin to the point of maximum translation)21. The translation data then were analyzed with use of a two-way analysis of variance, with the applied load and the location of the simulated fracture as direct effects. Specific differences among the data then were determined post hoc with use of the Student t test.
For all statistical tests, a p value of less than 0.05 was considered significant. When necessary, p values were adjusted for multiple comparisons. Statistical analysis was performed with use of BMDP statistical software (BMDP Dynamic Version 7.0; BMDP Statistical Software, Los Angeles, California).
A total of twenty-four fresh-frozen cadaveric hip joints were tested. Simulated transverse fractures were evaluated in twelve specimens, and simulated anterior column and posterior column fractures were tested in six specimens each. For the evaluation of the transverse fractures, all specimens were tested intact and then after each of four simulated fractures (with medial roof-arc angles of 60, 50, 40, and 30 degrees). Similarly, for the evaluations of both the anterior column and the posterior column fractures, each specimen was tested intact and then after each of four simulated fractures (very low, low, intermediate, and high). All twenty-four specimens were subjected to three magnitudes of load (800, 1200, and 1600 newtons) and four cycles of each load.
During the mechanical testing, each specimen was classified as either stable (if the femoral head remained concentrically reduced throughout all four loading cycles) or unstable (if there was any gross dislocation or if the femoral head was not in contact with the acetabulum throughout all four cycles). Gross dislocation was visually obvious and also was reflected by changes in the measured load as well as in femoral translation as recorded by the instrumented x-y table. When the femoral head dislocated, the sustained load immediately decreased to zero newtons and the measured translation was extreme. Although femoral translations occurred in stable specimens, these translations were relatively small and increased gradually with applied load. The average translation for the intact specimens was 3.69 millimeters at 800 newtons, 5.65 millimeters at 1200 newtons, and 7.92 millimeters at 1600 newtons. Despite these minor translations, the femoral head remained concentrically reduced beneath the intact acetabulum. Observations suggested that these minor translations resulted from deformations in the loading frame.
The stability of the specimens decreased with progressively superior fractures and with higher loads. The decrease in stability seen with more superior fractures was significant (p < 0.0001), as will be described. However, the decrease in stability seen with higher loads was not found to be significant, with the numbers available. Most likely, slight deformations in the loading frame resulted in slight rotations of the acetabulum, causing some specimens to be stable at lower loads and unstable at higher loads. Specimens that were unstable at a given load were always unstable at higher loads. For each fracture type and each fracture location, all specimens were either consistently stable or consistently unstable throughout all trials.
The radiographic measurements of the medial roof-arc angle confirmed the accuracy of the transverse osteotomies as well as the consistency of the anterior column and posterior column osteotomies (Table I).
Simulated Transverse Fractures
Testing of the transverse fractures showed that no specimen with a roof-arc angle of 30 degrees was stable and that the number of stable specimens decreased with progressively higher fractures. At 800 newtons, ten of the twelve specimens with a medial roof-arc angle of 60 degrees, six with an angle of 50 degrees, and only one with an angle of 40 degrees were stable (Fig. 3-A). The same trends were observed at 1200 newtons: nine of the twelve specimens with a roof-arc angle of 60 degrees, four with an angle of 50 degrees, and only one with an angle of 40 degrees were stable (Fig. 3-B). Similar results were observed at 1600 newtons, but with a lower proportion of stable specimens (seven of twelve) when the roof-arc angle was 60 degrees (Fig. 3-C).
When all of the data were considered as a single group, stability was significantly affected by both the magnitude of the applied load and the location of the simulated fracture (Pearson's chi-square test, p < 0.00005). When the relationship between the stability of the specimen and the roof-arc angle was considered separately, stability was found to be significantly affected by the roof-arc angle at each applied load (Pearson's chi-square test, p < 0.0001 for all comparisons). When the relationship between stability and the applied load was considered separately, stability was not found to be significantly affected by the applied load at roof-arc angles of 60, 50, or 40 degrees (Pearson's chi-square test, p > 0.3 for all comparisons).
Pairwise comparisons of the percentages of stable specimens were performed with use of Fisher's exact test (Table II). As there were six pairwise comparisons at each load level, an individual test had to have a p value of less than 0.0083 to be considered significant (to maintain an overall p value of less than 0.05). At 800 newtons, significant differences in stability were observed between the intact specimens and the specimens with a roof-arc angle of 50 degrees, between the intact specimens and the specimens with a roof-arc angle of 40 degrees, and between the specimens with a roof-arc angle of 60 degrees and those with a roof-arc angle of 40 degrees (Table II). With the number of specimens tested, a trend toward a significant difference in stability was noted between the specimens with a roof-arc angle of 50 degrees and those with a roof-arc angle of 40 degrees.
Similarly, at 1200 newtons, significant differences in stability were found between the intact specimens and the specimens with a roof-arc angle of 50 degrees, between the intact specimens and those with a roof-arc angle of 40 degrees, and between the specimens with a roof-arc angle of 60 degrees and those with a roof-arc angle of 40 degrees (Table II). With the number of specimens tested, there was a trend toward a significant difference in stability between the specimens with a roof-arc angle of 60 degrees and those with a roof-arc angle of 50 degrees. A similar trend was noted when the specimens were loaded to 1600 newtons. Specifically, significant differences in stability were observed between the intact specimens and those with a roof-arc angle of 50 degrees and between the intact specimens and the specimens with a roof-arc angle of 40 degrees (Table II). A trend toward a significant difference also was observed between the intact specimens and the specimens with a roof-arc angle of 60 degrees and between the specimens with a roof-arc angle of 60 degrees and those with a roof-arc angle of 40 degrees.
Simulated Anterior Column Fractures
All specimens with very low, low, and intermediate anterior column fractures remained stable throughout all tests (Fig. 4). Testing of the high fractures revealed that four of the six specimens were stable at 800 newtons, only two of the six were stable at 1200 newtons, and no specimen was stable at 1600 newtons. The average anterior roof-arc angle was 87 degrees for the very low fractures, 64 degrees for the low fractures, 43 degrees for the intermediate fractures, and 21 degrees for the high fractures (Table I). As the weight-bearing dome was intact on the anteroposterior and iliac oblique radiographs of all specimens, the medial and posterior roof-arc angles could not be measured.
The stability of the specimens with an intermediate fracture was significantly different from that of the specimens with a high fracture when all loads were considered together (Pearson's chi-square test, p < 0.0002); therefore, an analysis was performed to determine if there were differences in stability between these two groups at each applied load. No significant difference was detected at 800 newtons (Fisher's exact test, p = 0.22), but significant differences were detected at both 1200 and 1600 newtons (Fisher's exact test, p < 0.03 for both comparisons). Finally, when only high fractures were considered, stability was found to be significantly affected by the applied load (Pearson's chi-square test, p < 0.05).
Simulated Posterior Column Fractures
Overall, the posterior column fractures were the least stable (Fig. 5). All six specimens with a low fracture were unstable. At 800 newtons, five of the six specimens with a very low fracture were stable. The posterior roof-arc angle averaged 68 degrees for the very low fractures and 48 degrees for the low fractures (Table I). As the weight-bearing dome was intact on the anteroposterior and obturator oblique radiographs of all specimens, the medial and anterior roof-arc angles could not be measured.
The stability of the intact specimens was significantly different from that of the specimens with a very low fracture when all loads were considered together (Pearson's chi-square test, p < 0.02); therefore, an analysis was carried out to determine if there were differences in stability between these two groups at each applied load. No significant difference was detected at either 800 or 1200 newtons (Fisher's exact test, p = 0.5 and p = 0.1, respectively), but a significant difference was detected at 1600 newtons (Fisher's exact test, p < 0.03). Finally, when only the very low fractures were considered, stability was not found to be significantly affected by the applied load (Pearson's chi-square test, p = 0.2).
The present study was based on the assumption that the critical weight-bearing dome can be defined as the portion of the acetabulum that is necessary to maintain the stability of the hip. This definition seems reasonable as joint stability appears to play an important role in the pathogenesis of posttraumatic arthritis. Several investigators have induced arthritis in animals by generating instability of the joint1,6,7,9,18. Clinically, a high prevalence of arthritis is seen in hips rendered unstable as the result of posterior wall acetabular fractures. Epstein found that arthritis was likely to develop in hips that redislocated after the initial reduction2. Redislocation could be considered gross instability. However, lesser degrees of instability also may be detrimental. Acetabular fractures have a poorer prognosis when the femoral head is not concentrically reduced beneath the dome (that is, when the femoral head is slightly subluxated) than when the head is concentrically reduced3,13.
The instabilities created in our model differ from those that would result in association with displaced acetabular fractures in vivo. In our experiments, the portion of the acetabulum medial to the simulated fracture was removed. This allowed the hip to dislocate grossly when it became unstable. In acetabular fractures in vivo, the portion of the acetabulum medial to the fracture is displaced but is not completely removed. The displacement changes the shape of the acetabulum and allows the hip to subluxate but not to dislocate grossly. Nevertheless, the portion of the acetabulum that is necessary to maintain normal stability of the femoral head should be the same regardless of whether the portion of the acetabulum medial to the fracture is present or absent. Similarly, the fact that the capsule was removed from our specimens should not have affected stability. When the hip subluxates medially in vivo, the capsule collapses. Thus, stability is provided by the osseous acetabulum and not by the capsule.
In our study, load was applied to the hip in only one direction. However, this direction mimics the loading of the hip at a critical stage of the gait cycle. The maximum resultant force across the acetabulum occurs just before toe-off during the gait cycle and is oriented 25 degrees posterior and 20 degrees medial to the zenith of the acetabulum16. Our experiments reproduced this load vector. In fact, during the stance phase of gait, forces on the acetabulum are always directed posteriorly and medially. During the swing phase and the early stage of stance phase (that is, heel-strike), the hip is in greater flexion, so loads are directed slightly more posteriorly but are of a much smaller magnitude. In our study, directing loads more posteriorly probably would have increased the instability of the specimens with transverse and posterior column fractures but would have increased the stability of those with anterior column fractures.
In our study, the instability of each type of simulated fracture (transverse, anterior column, and posterior column) was greater at higher loads than at lower loads. We could not demonstrate that instability increased with higher loads for all fractures, but there was certainly a trend in that direction. There are two possible explanations for this observation. Either there truly was increased instability for a given fracture at higher loads or the observation was an artifact of our model. In any case, our data for loads of 800 newtons should provide a conservative estimate of instability—that is, if a fracture with a given roof-arc angle is unstable at 1600 newtons, it may or may not be unstable at 800 newtons. However, if a fracture is unstable at 800 newtons, it is clearly unstable at higher loads.
Currently, instability of the hip is considered to be an indication for operative repair of a displaced acetabular fracture. Most surgeons recommend open reduction and internal fixation of the acetabulum if the femoral head does not stay under the dome of the acetabulum5,13-15,22. However, this decision has to be made on the basis of the limited information provided by initial radiographs. A hip that appears to be concentrically located with the patient supine in bed could easily dislocate when loads are directed more posteriorly21. In a long-term clinical evaluation, Matta divided operatively treated acetabular fractures into three groups according to the type of reduction: anatomical reduction, imperfect reduction with the femoral head placed concentrically beneath the intact acetabulum, and imperfect reduction with the femoral head clearly subluxated13. An excellent or good result was reported for 81 percent (150) of 185 hips that had an anatomical reduction, for 67 percent (thirty-five) of fifty-two hips that had an imperfect concentric reduction, and for only nine of eighteen hips that had clear subluxation. If concentric reduction on plain radiographs correctly predicted an intact weight-bearing acetabulum, patients in the second group would have had results similar to those in the group that had an anatomical reduction. A fluoroscopic examination may provide a better indication of stability, but anesthesia is necessary and physiological forces are difficult to generate23.
Radiographic roof-arc angles are frequently used to determine whether or not a fracture crosses the weight-bearing dome. Matta et al. originally defined roof-arc angles and noted that fractures with a medial or posterior roof-arc angle of 30 degrees or less or an anterior roof-arc angle of 20 degrees or less usually were associated with a poor result10. On the basis of this finding and the available biomechanical data, they suggested that fractures with medial, anterior, and posterior roof-arc angles of less than 30, 40, or 50 degrees, respectively, infringed on the weight-bearing area. However, their recommendations changed over the years and were based primarily on clinical impressions rather than on objective data11,12. Most recently, Olson and Matta suggested that an acetabular fracture can be treated nonoperatively if the roof-arc angle is greater than 45 degrees on all three radiographs and if the hip remains concentrically reduced with the patient out of traction13,15.
Our study provides objective data for determining the roof-arc angles that define the weight-bearing dome. These data suggest that a posterior roof-arc angle of 45 degrees would underestimate the critical weight-bearing area, whereas an anterior roof-arc angle of 45 degrees would overestimate the area. In the present study, eleven of the twelve transverse fractures were unstable at 800 newtons when the average medial roof-arc angle measured 40 degrees (Fig. 3-A). Indeed, significant instability was observed in association with the transverse fractures when the average medial roof-arc angle measured 50 degrees (p = 0.0069), and two of the twelve transverse fractures were unstable when the average medial roof-arc angle measured 60 degrees (Fig. 3-A). In other words, a medial roof-arc angle of 40 degrees or less virtually ensured instability but a medial roof-arc angle that was greater than 40 degrees did not guarantee stability.
The fact that transverse fractures with a roof-arc angle of greater than 40 degrees were unstable can be explained by the level of the osteotomy on the posterior column. We found that even very low posterior column fractures generated instability and that all specimens with a low posterior column fracture were unstable (Fig. 5). The average posterior roof-arc angle (and standard deviation) for the very low posterior column fractures measured 68 ± 4.2 degrees (Table I). Interestingly, all unstable transverse fractures had a posterior roof-arc angle of less than 60 degrees, regardless of whether the medial roof-arc angle measured 30, 40, 50, or 60 degrees. Anterior column fractures proved much more stable (Fig. 4). Only two of the six specimens with a high anterior column fracture were unstable at 800 newtons, with an average anterior roof-arc angle of 21 ± 1.3 degrees (Table I). On the basis of our findings, it seems reasonable to suggest that anterior, medial, and posterior roof-arc angles of 25, 45, and 70 degrees, respectively, safely define the limits of the weight-bearing area.
However, few clinical data are available with which to evaluate the accuracy of these limits. Olson and Matta evaluated twenty-three patients who had an acetabular fracture that was treated nonoperatively15. In eleven of these patients, the roof-arc angle was greater than 45 degrees on all three radiographs and the weight-bearing dome therefore was thought to be intact. Nine of these eleven patients had a good or excellent result, one had a fair result, and one had a poor result. The patients who had the fair and poor results had posterior roof-arc angles of 50 and 55 degrees, respectively. According to our criteria, both of these patients had involvement of the weight-bearing area. Heeg et al. evaluated the effects of congruency and involvement of the weight-bearing dome on the results of nonoperative treatment of acetabular fractures3. Of the forty-six hips that had a congruent reduction, five (11 percent) had a fair or poor result. This finding suggests that congruency alone cannot be used to judge involvement of the weight-bearing dome. Anterior, medial, and posterior roof-arc angles of at least 40, 30, and 50 degrees, respectively, were used to define the weight-bearing area. Of the twenty-seven hips in which the fracture did not involve this area, three (11 percent) had a fair or poor result. Although the exact location of the fractures was not given, it is possible that our criteria would have corresponded better with the prognosis.
In our study, we used roof-arc angles to characterize the location of a fracture and thereby predict whether it would affect the stability of the hip. However, caution is necessary when these measurements are used to influence the clinical decision-making process. Because this was an in vitro study, it was possible to obtain perfect oblique radiographs of every hip. In the clinical situation, oblique radiographs often are made with the hip underrotated. The position of the hip could affect measurements of the roof-arc angle and could lead to erroneous conclusions regarding the stability of the hip. In addition, it is sometimes difficult to make accurate measurements, especially for anterior fractures.
Given the inherent difficulties involved in the measurement of the roof-arc angle, it seems appropriate to provide a second means of defining the weight-bearing area. In the present study, fractures exiting the posterior column just superior to the ischial spine and those exiting the anterior column through the iliac wing led to instability. The articular surface between these two extremes could be considered the weight-bearing area. This finding suggests that, when a displaced fracture crosses this area, every effort should be made to reduce the fracture anatomically in order to prevent posttraumatic arthritis.
NOTE: The authors thank William D. Johnson, M.S., Associate Professor, Department of Biometry and Genetics, Louisiana State University Medical Center, New Orleans, for assisting with the statistical analysis.
Davis, W., and Maskowitz, R. W.: Degenerative joint changes following posterior cruciate ligament section in the rabbit. Clin. Orthop.,93: 307-312, 1973.93307
1973
[PubMed]
Epstein, H. C.: Posterior fracture-dislocations of the hip. Long-term follow-up. J. Bone and Joint Surg.,56-A: 1103-1127, Sept. 1974.56-A1103
1974
Heeg, M.; Oostvogel, H. J.; and Klasen, H. J.: Conservative treatment of acetabular fractures: the role of the weight-bearing dome and anatomic reduction in the ultimate results. J. Trauma,27: 555-559, 1987.27555
1987
[PubMed]
Kannus, P., and Järvinen, M.: Conservatively treated tears of the anterior cruciate ligament. Long-term results. J. Bone and Joint Surg.,69-A: 1007-1012, Sept. 1987.69-A1007
1987
Letournel, E., and Judet, R.: Fractures of the Acetabulum, edited by R. A. Elson. Ed. 2. New York, Springer, 1993.
McDevitt, C. A., and Muir, H.: Biochemical changes in the cartilage of the knee in experimental and natural osteoarthritis in the dog. J. Bone and Joint Surg.,58-B(1): 94-101, 1976.58-B(1)94
1976
McDevitt, C.; Gilbertson, E.; and Muir, H.: An experimental model of osteoarthritis; early morphological and biochemical changes. J. Bone and Joint Surg.,59-B(1): 24-35, 1977.59-B(1)24
1977
Mankin, H. J.; Mow, V. C.; and Buckwalter, J. A.: Form and function of articular cartilage. In Orthopaedic Basic Science, pp. 27-28. Edited by S. R. Simon. Rosemont, Illinois, American Academy of Orthopaedic Surgeons, 1994.
Marshall, J. L.: Periarticular osteophytes. Initiation and formation in the knee of the dog. Clin. Orthop.,62: 37-47, 1969.6237
1969
[PubMed]
Matta, J. M.; Anderson, L. M.; Epstein, H. C.; and Hendricks, P.: Fractures of the acetabulum. A retrospective analysis. Clin. Orthop.,205: 230-240, 1986.205230
1986
[PubMed]
Matta, J. M.; Mehne, D. K.; and Roffi, M.: Fractures of the acetabulum. Early results of a prospective study. Clin. Orthop.,205: 241-250, 1986.205241
1986
[PubMed]
Matta, J. M., and Merritt, P. O.: Displaced acetabular fractures. Clin. Orthop.,230: 83-97, 1988.23083
1988
[PubMed]
Matta, J. M.: Fractures of the acetabulum: accuracy of reduction and clinical results in patients managed operatively within three weeks after the injury. J. Bone and Joint Surg.,78-A: 1632-1645, Nov. 1996.78-A1632
1996
Mayo, K. A.: Open reduction and internal fixation of fractures of the acetabulum. Results in 163 fractures. Clin. Orthop.,305: 31-37, 1994.30531
1994
[PubMed]
Olson, S. A., and Matta, J. M.: The computerized tomography subchondral arc: a new method of assessing acetabular articular continuity after fracture (a preliminary report). J. Orthop. Trauma,7: 402-413, 1993.7402
1993
[PubMed]
Pedersen, D. R.; Brand, R. A.; and Davy, D. T.: Pelvic muscle and acetabular contact forces during gait. J. Biomech.,30: 959-965, 1997.30959
1997
[PubMed]
Pennal, G. F.; Davidson, J.; Garside, H.; and Plewes, J.: Results of treatment of acetabular fractures. Clin. Orthop.,151: 115-123, 1980.151115
1980
[PubMed]
Pond, M. J., and Nuki, G.: Experimentally-induced osteoarthritis in the dog. Ann. Rheumat. Dis.,32: 387-388, 1973.32387
1973
[PubMed]
Rasmussen, P. S.: Tibial condylar fractures. Impairment of knee joint stability as an indication for surgical treatment. J. Bone and Joint Surg.,55-A: 1331-1350, Oct. 1973.55-A1331
1973
Rowe, C. R., and Lowell, J. D.: Prognosis of fractures of the acetabulum. J. Bone and Joint Surg.,43-A: 30-59, Jan. 1961.43-A30
1961
Thomas, K. A.; Vrahas, M. S.; Noble, J. W., Jr.; Bearden, C. M.; and Reid, J. S.: Evaluation of hip stability after simulated transverse acetabular fractures. Clin. Orthop.,340: 244-256, 1997.340244
1997
[PubMed]
Tile, M.: Fractures of the Pelvis and Acetabulum. Ed. 2. Baltimore, Williams and Wilkins, 1995.
Tornetta, P., III: Non-operative management of acetabular fractures. The use of dynamic stress views. J. Bone and Joint Surg.,81-B(1): 67-70, 1999.81-B(1)67
1999
Watson, H. K., and Ballet, F. L.: The SLAC wrist: scapholunate advanced collapse pattern of degenerative arthritis. J. Hand Surg.,9A: 358-365, 1984.9A358
1984