CORR Insights®: Which Two-dimensional Radiographic Measurements of Cam Femoroacetabular Impingement Best Describe the Three-dimensional Shape of the Proximal Femur?

Where Are We Now?

How orthopaedic surgeons diagnose and treat a patient with hip pain from cam femoroacetabular impingement (FAI) remains controversial [3-5, 11]. Orthopaedic surgeons generally use two-dimensional (2-D) radiographs or single-plane slices from CT or MRI to diagnose cam FAI [4, 11]. But the reliability of these images, as well as the reference values for standard reported variables, like the alpha angle, are debatable [3, 5]. Two-dimensional measures provide some information about the underlying bone morphology but are prone to errors stemming from the positioning of the patient, the orientation of the reconstructed image plane, and the location of the abnormal morphology, all of which contribute to a high rate of misdiagnosis [9]. Although imaging alone without considering the patient’s symptoms and movement limitations is not sufficient to identify FAI, imaging remains an important part of diagnosing this condition and can be used to help inform the treatment plan.

Surgeons generally treat cam FAI with proximal femoral osteoplasty to improve ROM and eliminate bone and soft-tissue impingement. Removing too much bone can result in a weakened femoral neck (with a risk of subsequent fracture) and disrupt the vascular supply, whereas removing too little bone can lead to continued pain and progressive arthritis [10]. As a result, researchers have championed standardized three-dimensional (3-D) imaging that highlights abnormal morphology as a better way to diagnosis and to assist surgeons with presurgical decisions [6-8].

Previous studies [1, 6, 7] found that surgeons can use statistical shape matching to compare 3-D reconstructions of the hip to normal hip morphology rather than the idealized (and unrealistic) spherical shape comparison. By building off of this research, the current study Atkins and colleagues [2] set out to: (1) Determine the correlation between 2-D radiographic measurements of cam FAI and 3-D metrics of proximal femoral shape, (2) identify the combination of radiographic measurements from plain film projections that are most effective at predicting the 3-D shape of the proximal femur, and (3) consistently quantify the bony defects that are associated with FAI. The authors of the current study also used sophisticated 3-D measurement techniques to find a combination of simple, standard 2-D imaging approaches that could bridge the gap between current practice (2-D radiograph or reconstructed slices from CT/MRI) and what might eventually become standard practice (3-D shape measurements) [2].

Where Do We Need To Go?

The current study by Atkins and colleagues [2] represents a step toward validating a new standard combination of medical images to predict 3-D bone morphology. The authors found that the use of a 3-D statistical shape model to obtain a shape score can more easily describe the overall shape of the entire proximal femur compared to previous reconstructed 3-D images where the lesions are not easily described in quantitative terms [2]. This shape score served as the reference standard against which standard 2-D image measurements for cam deformity could be compared. Using this standard, the authors found that two imaging views (Meyer lateral and 90^o Dunn) used together were the most predictive 3-D shape of the proximal femur in the region of the cam deformity.

Although suggested as a new clinical standard, Atkins and colleagues [2] also acknowledged that these images described less than half of the 3-D variations in the shape of the cam deformity. Therefore, researchers need to fully develop automated 3-D visualizations of the proximal femoral shape as well as to identify simple scored values, such as the shape index value, that can be validated and implemented into standard clinical practice. Using 3-D imaging and having standard 3-D values would eliminate some of the variability from patient orientation that occurs while obtaining radiographs and could reduce the likelihood that critical bony abnormalities are missed within the symptomatic patients. Three-dimensional imaging would also help doctors better prepare for surgeries that involve reshaping the bone.

Ultimately, researchers still need to correlate the methods used to determine long-term patient outcomes. Advanced imaging methods that utilize 3-D imaging of the bone need to be made readily available to doctors and these advanced methods should be validated against patient outcomes. Future research should educate clinicians on how to use these advanced imaging methods for diagnosis and treatment planning.

How Do We Get There?

An incremental approach is needed to improve the imaging that surgeons use for FAI diagnosis and treatment planning. Regarding the current study, Atkins and colleagues presented results that used reconstructed 2-D slices from 3-D imaging as a basis of comparison to the statistical 3-D shape models. A next step would be to compare the 3-D statistical shape model against actual patient radiographs. Similar incremental steps could be taken for most of the current FAI imaging studies. Retrospective analyses that utilize large patient databases where multiple imaging modalities are available could be a powerful and achievable next step. Convergent validity analyses should be taken to compare traditional radiographs with reconstructed 2-D images from CT and MRI with different 3-D shape scores. Additionally, these images should be compared against the diagnosis and treatment approach selected by the doctor to better understand the relationship between the imaging technique used and the eventual treatment decision. Finally, these treatment decisions should be compared against any available patient outcomes that were recorded so that the complete path from imaging to diagnosis, diagnosis to treatment, and treatment to outcome could be analyzed. A retrospective analysis with already existing data would be the next logical step, particularly if multiple treating institutions were willing to pool together their available data for comparison.

Before committing to a multicenter, prospective study that follows patients for several years after treatment, researchers should first establish an evidence-based guideline regarding the best surgical approaches for patients with cam FAI. While this evidence-based guideline will be debated and continuously refined, a guideline similar to those adopted by professional societies, such as the American Academy of Orthopaedic Surgeons, should be developed prior to the launch of a large, prospective study. Smaller studies, such as the one by Atkins and colleagues [2], could validate 3-D imaging techniques and larger retrospective studies could identify the parameters to be explored in a large prospective study.

Orthopaedic companies will need to take a role in the process of making automated 3-D quantification techniques more available, and should leverage the wealth of information that is now available when developing these new imaging products. Large questionnaire studies could identify orthopaedic surgeons’ preferences and a plan should be put in place that uses existing avenues, such as professional societies, to educate doctors on how to use these new models and images.