Original ResearchFree Access

How Useful Is the Alpha Angle for Discriminating between Symptomatic Patients with Cam-type Femoroacetabular Impingement and Asymptomatic Volunteers?

Published Online:https://doi.org/10.1148/radiol.12112479

Abstract

Purpose

To compare the alpha-angle measurements in volunteers and patients with femoroacetabular impingement (FAI) and to develop potential threshold values.

Materials and Methods

This study was approved by the institutional review board; all individuals signed informed consent. Magnetic resonance (MR) images at 1.5 T in 106 individuals (ages 20–50 years) were analyzed in 53 patients (33 cam- and 20 mixed-type FAI) and 53 age- and sex-matched asymptomatic volunteers. Alpha angles were measured on radially reformatted MR images of the proximal femur by two independent readers. Intraclass correlation coefficient (ICC) and receiver operating characteristic (ROC) were calculated.

Results

Mean alpha angles were highest in the anterosuperior segment: 65.4° ± 11.5 [standard deviation] and 65.2° ± 7.3 for readers 1 and 2 in patients and 53.3° ± 9.6 and 55.0° ± 8.8 in volunteers, respectively (P < .001, patients vs volunteers). Alpha angles greater than 55° were measured in 20 (38%) and 33 (62%) of 53 volunteers for readers 1 and 2, respectively. Maximal alpha angle in any segment was substantially different (P < .001) in patients and volunteers (70.3° ± 11.2 vs 57.9° ± 10.5 for reader 1; 69.4° ± 8.8 vs 58.7° ± 8.9 for reader 2), with a large overlap. Overall interobserver agreement was good (ICC, 0.712). ROC showed the largest area under the curve at the anterosuperior segment: 0.791 and 0.824 for readers 1 and 2, respectively (P < .001). A 55° alpha-angle threshold value gave a sensitivity and specificity of 81% and 65% for reader 1 and of 90% and 47% for reader 2, respectively. A 60° alpha-angle threshold value gave a sensitivity and specificity of 72% and 76% for reader 1 and 80% and 73% for reader 2, respectively.

Conclusion

There is substantial overlap in the alpha-angle measurements between volunteers and patients with cam-type deformities. Discrimination is best at the anterosuperior segment. Increasing the alpha-angle threshold value from 55° to 60° reduces false-positive results while maintaining a reasonable sensitivity.

© RSNA, 2012

Introduction

Femoroacetabular impingement (FAI) is the main cause of early-onset osteoarthritis in nondysplastic hips (1,2). It can arise because of osseous deformities of the femoral head and neck (cam-type FAI), a focal or general overcoverage of the acetabulum (pincer-type FAI), or a combination of the two (mixed-type FAI) (35). Patients presenting with symptoms and clinical signs suggestive of FAI are often initially examined with conventional radiography (6), which is reliable in detecting the osseous anomalies present in pincer-type FAI but lacks sensitivity and specificity for the detection of cam-type FAI (79). While cam-type deformities are most frequent in the anterosuperior aspect of the femoral head-neck junction, it is often not possible to detect these anterosuperior deformities on radiographs (10), and consequently these patients often undergo magnetic resonance (MR) imaging of the hip (3,10).

The alpha angle was initially used to quantify cam-type deformities only at the anterior aspect of the femoral head and neck at MR imaging (11), and later radial plane images were introduced to assess the alpha angle at the whole circumference (3). In the past few years, the alpha angle measurement has been the most common method to quantify the osseous deformities at the femoral head-neck junction, but there has been some controversy about its accuracy for clinical use, because the validation of the alpha angle measurement revealed only moderate agreement after repeated measurements by the same reader and substantial interobserver differences (12,13). Moreover, researchers in several studies have shown that a substantial portion of the healthy young adult population exhibits some degree of cam-type deformities (1417).

A reliable distinction between symptomatic and asymptomatic cam-type deformities would be important to prevent overtreatment of patients who are suspected of having FAI. Thus, the purpose of our study was to compare the alpha-angle measurements in volunteers and patients with FAI and to develop potential threshold values.

Materials and Methods

Study Population

This study was approved by the institutional review board. All individuals signed informed consent. Based on an earlier study population where the role of femoral antetorsion in the development of FAI was assessed with MR imaging (18), all symptomatic patients with FAI who had cam-type deformities of the proximal femur were identified and matched for age and sex with a group of asymptomatic volunteers who underwent MR imaging of the hip.

The patient group consisted of individuals who were referred for MR arthrography of the hip by the Department Orthopedic Surgery of our hospital during a period of 6 months in 2010. Inclusion criteria for the patient group were typical symptoms (persistent groin pain for more than 3 months in duration) and clinical signs (positive impingement test and internal hip rotation less than 20°) of FAI (11,19), age from 20 to 50 years, and informed consent. Patients were excluded if no cam-type deformity of the femoral head-neck junction was present on MR images. Cam-type deformities at the femoral head-neck junction were defined as an alpha angle greater than 55° at any location around the femoral neck for patient inclusion. To assess this factor, one reader (R.S., with 6 years of experience) evaluated all MR examinations in a separate readout session at least 2 months apart from the main readout to avoid recollection bias. Of 53 patients with FAI, 33 had cam-type FAI, and 20 had mixed-type FAI. Ten patients showed no cam-type deformities and were excluded. Therefore, the inclusion of patients in the study was based on a combination of clinical criteria and imaging criteria.

Volunteers were included if they were asymptomatic (no symptoms, no medical history in regard to abnormalities of the hip joint, negative impingement test, and internal hip rotation of greater than 25° at clinical examination by the same author who performed the assessment mentioned above) and were between the ages of 20 and 50 years. Exclusion criteria for volunteers were pregnancy, prior trauma to the hip, hip problems in infancy or childhood, or symptoms or pain of the hip or groin. All volunteers underwent nonenhanced MR imaging of the hip on the same day they were clinically examined.

Thus, the study population consisted of 106 individuals: 53 FAI patients with cam-type deformities (mean age, 35.6 years) and 53 age- and sex-matched asymptomatic volunteers (mean age, 34.5 years). The patient group contained 33 men (mean age, 34.8 years; range, 20–50 years) and 20 women (mean age, 37.0 years; range, 23–49 years). The volunteer group contained 31 men (mean age, 34.1 years; range, 24–50 years) and 22 women (mean age, 35.2 years; range, 23–50 years). Age was not different between patients and volunteers (P = .5).

MR Imaging

A radiologist injected intraarticular contrast material in a standardized fashion in all patients. One milliliter of a local anesthetic (lidocaine hydrochloride 2%,Rapidocain; Sintetica, Mendrisio, Switzerland), 1 mL of an iodinated contrast agent (iopamidol, Iopamiro 200; Bracco, Milan, Italy) at a dose of 200 mg/mL, and 8–10 mL of a diluted MR contrast agent (gadopentetate dimeglumine, Magnevist; Bayer Healthcare, Berlin, Germany) at a concentration of 2 mmol/L were injected with fluoroscopic guidance. The interval between the contrast agent injection and MR imaging was less than 15 minutes. Volunteers underwent MR imaging of the hip without contrast material administration by using the same MR protocol.

MR imaging was performed with a 1.5-T system (Magnetom Avanto; Siemens Healthcare, Erlangen, Germany). A four-channel body matrix phased-array surface coil (which was placed over the hip of the patient) and a six-channel spine matrix coil (which was integrated in the patient table) were used. Patients were positioned supine on the MR examination table. As part of the routine MR protocol, a three-dimensional data set was obtained with a transverse oblique (parallel to the femoral neck axis) water excitation true fast imaging with steady-state precession (FISP) sequence and the following parameters: repetition time msec/echo time msec, 12.3/5.45; sections, 64; section thickness, 1.25 mm; intersection gap, none; flip angle, 28°; bandwidth, 200 Hz/pixel; field of view, 17 cm; matrix, 384 × 384; number of signals acquired, one). The three-dimensional data set was used to reconstruct radial reformations by using the long axis of the femoral neck as a rotation axis, with a total number of 11 images, corresponding to 22 positions circumferentially around the femoral head-neck junction.

Image Analysis

Two fellowship-trained musculoskeletal radiologists (T.D. and R.S. with 8 and 6 years of experience, respectively) who were blinded to the clinical data of the patients independently analyzed all MR images of the patient group and the volunteer group after having completed a training session, where images in 10 patients who were not included in the study were analyzed in consensus. Radially reformatted images were used to assess the contour of the head-neck junction circumferentially around the femoral head and neck at the following positions: anteroinferior, anterior, anterosuperior, superior, and posterosuperior. These images were used to calculate the alpha angle. The measurements were performed at a picture archiving and communication system workstation (ProVision, release 5.0; Cerner, Kansas City, Mo). The alpha angle was measured between the axis of the femoral neck and the line from the center of the femoral head to the point where the distance from the center of the femoral head to the peripheral contour of the femoral head exceeded the radius of the femoral head (11) (Fig 1). The center of the femoral head was identified by placing a circle over the contour of the femoral head, and automatic visualization of the center of this circle was performed by the picture archiving and communication system software. The axis of the femoral neck was defined as a line that passed through the center of the femoral head and the center of the femoral neck at its narrowest point.

Figure 1a:

Figure 1a: Interobserver differences for alpha-angle measurements at the anterosuperior position in a 26-year-old asymptomatic male volunteer. (a) Schematic diagram of the proximal femur with a medial view onto the femoral head illustrating the different imaging planes where the alpha-angle measurements were performed. The image containing the anterosuperior position is highlighted with a thick green line. (b) Reconstructed MR image from a three-dimensional true FISP sequence at the anterosuperior position over the proximal femur, at the level indicated in a. The dotted circle, which indicates the position of images in b–d on the schematic in a, is used as a reference to draw the alpha angle. (c) Same reconstructed MR image as in b shows alpha angle measured by reader 1. The alpha angle was defined by a line between the center of the femoral head and the point where the distance from the center of the femoral head to the peripheral contour of the femoral head exceeds the radius of the femoral head and by a second line in the axis of the femoral neck. (d) Same reconstructed MR image as in b shows alpha angle measured by reader 2. Because a slightly different point was taken on the peripheral contour of the femoral head by reader 2, a difference of 8° results to the measurement of reader 1.

Figure 1b:

Figure 1b: Interobserver differences for alpha-angle measurements at the anterosuperior position in a 26-year-old asymptomatic male volunteer. (a) Schematic diagram of the proximal femur with a medial view onto the femoral head illustrating the different imaging planes where the alpha-angle measurements were performed. The image containing the anterosuperior position is highlighted with a thick green line. (b) Reconstructed MR image from a three-dimensional true FISP sequence at the anterosuperior position over the proximal femur, at the level indicated in a. The dotted circle, which indicates the position of images in b–d on the schematic in a, is used as a reference to draw the alpha angle. (c) Same reconstructed MR image as in b shows alpha angle measured by reader 1. The alpha angle was defined by a line between the center of the femoral head and the point where the distance from the center of the femoral head to the peripheral contour of the femoral head exceeds the radius of the femoral head and by a second line in the axis of the femoral neck. (d) Same reconstructed MR image as in b shows alpha angle measured by reader 2. Because a slightly different point was taken on the peripheral contour of the femoral head by reader 2, a difference of 8° results to the measurement of reader 1.

Figure 1c:

Figure 1c: Interobserver differences for alpha-angle measurements at the anterosuperior position in a 26-year-old asymptomatic male volunteer. (a) Schematic diagram of the proximal femur with a medial view onto the femoral head illustrating the different imaging planes where the alpha-angle measurements were performed. The image containing the anterosuperior position is highlighted with a thick green line. (b) Reconstructed MR image from a three-dimensional true FISP sequence at the anterosuperior position over the proximal femur, at the level indicated in a. The dotted circle, which indicates the position of images in b–d on the schematic in a, is used as a reference to draw the alpha angle. (c) Same reconstructed MR image as in b shows alpha angle measured by reader 1. The alpha angle was defined by a line between the center of the femoral head and the point where the distance from the center of the femoral head to the peripheral contour of the femoral head exceeds the radius of the femoral head and by a second line in the axis of the femoral neck. (d) Same reconstructed MR image as in b shows alpha angle measured by reader 2. Because a slightly different point was taken on the peripheral contour of the femoral head by reader 2, a difference of 8° results to the measurement of reader 1.

Figure 1d:

Figure 1d: Interobserver differences for alpha-angle measurements at the anterosuperior position in a 26-year-old asymptomatic male volunteer. (a) Schematic diagram of the proximal femur with a medial view onto the femoral head illustrating the different imaging planes where the alpha-angle measurements were performed. The image containing the anterosuperior position is highlighted with a thick green line. (b) Reconstructed MR image from a three-dimensional true FISP sequence at the anterosuperior position over the proximal femur, at the level indicated in a. The dotted circle, which indicates the position of images in b–d on the schematic in a, is used as a reference to draw the alpha angle. (c) Same reconstructed MR image as in b shows alpha angle measured by reader 1. The alpha angle was defined by a line between the center of the femoral head and the point where the distance from the center of the femoral head to the peripheral contour of the femoral head exceeds the radius of the femoral head and by a second line in the axis of the femoral neck. (d) Same reconstructed MR image as in b shows alpha angle measured by reader 2. Because a slightly different point was taken on the peripheral contour of the femoral head by reader 2, a difference of 8° results to the measurement of reader 1.

To make sure that the alpha angle was measured on exactly the same image by both readers, the first reader recorded the image number of the position where the alpha-angle measurement was performed after each of his measurements. The second reader later independently performed the measurement at the same position (as indicated by the image number given by reader 1). Both reader 1 and reader 2 independently measured all alpha angles in every patient and volunteer. If two reconstructed images were available for alpha-angle measurements at a given location (eg, anterosuperior position), the image with the larger alpha angle was selected for measurement by the first reader.

Statistical Analysis

Descriptive statistics were used to characterize the alpha angle for the different study subgroups, and mean values and standard deviations were calculated. Differences in age between the different study groups were assessed by using the two-tailed Mann-Whitney U test, and the alpha-angle differences were assessed with a Student t test, and a difference with P < .05 was considered significant. The relationship between age and the alpha angle was analyzed by using a Pearson correlation. Interobserver agreement was quantified by using the intraclass correlation coefficient (ICC). Receiver operating characteristics (ROCs) were calculated to assess diagnostic characteristics of different alpha-angle threshold values. All analyses were performed with statistical software (SPSS for Windows, release 17.0; SPSS, Chicago, Ill).

Results

Alpha Angles

Mean alpha angles were largest at the anterosuperior position (Table 1). The mean alpha angles in patients were 65.4° ± 11.5 and 65.2° ± 7.3 for readers 1 and 2, respectively. The mean alpha angles in volunteers were 53.3° ± 9.6 and 55.0° ± 8.8 for readers 1 and 2, respectively (P < .001, patients vs volunteers for both readers 1 and 2). At the anterior, the anterosuperior, and the superior positions, alpha angles were significantly larger in patients than in volunteers (Table 1).

Table 1 Mean Alpha Angles in Patients with FAI and Asymptomatic Volunteers

Table 1

Note.—Data are mean alpha angles in degrees ± standard deviations. P values < .05 denote significant differences.

Alpha angles greater than 55° were measured in 20 (38%) of 53 volunteers by reader 1 and in 33 (62%) of 53 volunteers by reader 2 (Table 2). Most volunteers with alpha angles greater than 55° were male (85% for reader 1 and 76% for reader 2). Most increased alpha angles in volunteers were found anterosuperiorly (32% and 51% for readers 1 and 2, respectively) and superiorly (19% and 32% for readers 1 and 2, respectively) (Fig 2).

Table 2 Number of Asymptomatic Volunteers with Alpha Angles Greater than 55 Degrees

Table 2

Note.—Data are numbers of volunteers, and numbers in parentheses are percentages.

Figure 2a:

Figure 2a: Alpha angle is substantially higher at anterosuperior position compared with anterior position in a 24-year-old asymptomatic male volunteer. (a) Reconstructed MR image from a three-dimensional true FISP sequence with measurement at the anterior position over the proximal femur. (b) Reconstructed MR image with measurement in the same patient at the anterosuperior position. The dotted circle is described in Figure 1. Insets = position of the reconstructed MR image on the schematic of the proximal femur.

Figure 2b:

Figure 2b: Alpha angle is substantially higher at anterosuperior position compared with anterior position in a 24-year-old asymptomatic male volunteer. (a) Reconstructed MR image from a three-dimensional true FISP sequence with measurement at the anterior position over the proximal femur. (b) Reconstructed MR image with measurement in the same patient at the anterosuperior position. The dotted circle is described in Figure 1. Insets = position of the reconstructed MR image on the schematic of the proximal femur.

The maximal alpha angle in any segment was substantially different (P < .001 for both readers) in patients and volunteers (70.3° ± 11.2 for patients vs 57.9° ± 10.5 for volunteers for reader 1; 69.4° ± 8.8 for patients vs 58.7° ± 8.9 for volunteers for reader 2), albeit with a large overlap (Table 3). Both in patients and volunteers, most maximal alpha-angle values were found at the anterosuperior radial position: In 22 (42%) of 53 patients, reader 1 detected the maximal alpha angle at the anterosuperior position, whereas this was the case in 26 (49%) of 53 patients for reader 2. In volunteers, the corresponding values for the anterosuperior position were identical (42% and 49% for readers 1 and 2, respectively).

Table 3 Maximal Alpha Angles in Patients with FAI and Asymptomatic Volunteers

Table 3

Note.—Data are mean alpha angles in degrees ± standard deviations. Ranges are in degrees. P values < .05 denote significant differences. NA = not applicable.

There was no correlation between age and the maximal alpha angle (R = 0.119, P = .4 [reader 1]; R = 0.060, P = .6 [reader 2]) or between age and the presence of alpha angles greater than 55° in volunteers (R = −0.019, P = .9 [reader 1]; R = −0.089, P = .5 [reader 2]).

Overall interobserver agreement between readers 1 and 2 was good, with an ICC of 0.712, but varied substantially for the different positions (Table 4).

Table 4 Interreader Agreement between Reader 1 and 2 for Alpha-Angle Measurements

Table 4

ROC Analysis

ROC analysis showed that the largest area under the curve was at the anterosuperior position, with 0.791 and 0.824 for readers 1 and 2, respectively (P < .001 for both readers), for discrimination between patients with FAI and asymptomatic volunteers (Fig 3). A 55° alpha-angle threshold value resulted in a sensitivity of 81% and a specificity of 65% for reader 1 and in a sensitivity of 90% and a specificity of 47% for reader 2 (Fig 4). When the alpha-angle threshold value was increased to 60°, a sensitivity of 72% and specificity of 76% resulted for reader 1, and a sensitivity of 80% and specificity of 73% resulted for reader 2. The number of volunteers with abnormal alpha angles at the anterosuperior position was 17 (32%) and 27 (51%) for readers 1 and 2, respectively, with a 55° threshold value. When the threshold value was increased to 60°, the number of volunteers with abnormal alpha angles was reduced to 12 (23%) and 14 (26%) for readers 1 and 2, respectively.

Figure 3:

Figure 3: ROC curves of alpha angles as measured by reader 1 for discrimination between patients with FAI and asymptomatic volunteers showed the largest area under the curve at the anterosuperior position (black line). Discrimination between patients and volunteers was smaller for the other positions, with decreasing areas under the curve for the superior position (green line), the anterior position (blue line), the posterosuperior position (purple line), and the anteroinferior position (orange line). Red dashed line = ROC curve for the maximal alpha angle at any location. * = point on the curve for the anterosuperior position that is farthest from the line of nondiscrimination (straight gray line) and is the point of maximal discrimination: This corresponds to an alpha angle of 55° (sensitivity, 81%; specificity, 65%).

Figure 4:

Figure 4: Graph shows sensitivity and specificity values for discrimination between patients with FAI and asymptomatic volunteers with alpha angles at the anterosuperior position, as obtained by using ROC curves. The graph shows a substantially higher specificity if the alpha-angle threshold value is set at 60° compared with 55° but an associated reduction in sensitivity. R1 = reader 1, R2 = reader 2.

ROC analysis further showed that discrimination between patients and volunteers was also possible at the anterior and superior position, but with smaller areas under the curve: At the anterior position, the area under the curve was 0.601 for reader 1 and 0.637 for reader 2. The discrimination between patients and volunteers was significant only for reader 2 (P = .015) but not for reader 1 (P = .73) at the anterior position. At the superior position, the area under the curve was 0.653 for reader 1 and 0.660 for reader 2, which was significant (P = .007 and .005, for readers 1 and 2, respectively).

As a secondary means of distinguishing patients with FAI and asymptomatic volunteers, ROC analysis was used to identify the maximal alpha angle from any position, with an area under the curve of 0.795 for reader 1 and of 0.807 for reader 2 (P < .001 for both readers). For the maximal alpha angle at any position, a 55° threshold value resulted in a sensitivity of 88% and a specificity of 56% for reader 1 and in a sensitivity of 97% and a specificity of 36% for reader 2. With a maximal alpha-angle threshold value of 60°, a sensitivity of 82% and a specificity of 66% were obtained for reader 1 and a sensitivity of 93% and a specificity of 66% were obtained for reader 2.

Discussion

In this study, we assessed the usefulness of alpha-angle measurements on radial images in distinguishing between patients with FAI and asymptomatic volunteers. In most prior studies, investigators used either a 50° or a 55° threshold value for distinguishing normal and abnormal alpha angles: While the 50° threshold value is occasionally cited in the orthopedic literature (15,20,21), the 55° threshold value is used more commonly both in the radiologic and the orthopedic literature (3,12,13,17,22,23). However, the use of the alpha angle for assessment of cam-type deformities is disputed, and so far no conclusive data are available for determining an ideal alpha-angle threshold value (16,24).

In our study, the largest alpha angles were measured at the anterosuperior position in both patients and volunteers: Only in every fourth individual with increased alpha angles at any location was there also an increased alpha angle at the anterior position. This is in accordance with the findings of Rakhra and colleagues (22) who reported that, in 54% of 41 patients who were suspected of having FAI, the alpha angle was less than 55° on the axial oblique MR images, but it was 55° or larger in the same patient when the radial plane images were assessed.

A large number of volunteers in our study had alpha angles that might be considered abnormal by these prior criteria—this was especially evident at the anterosuperior position. The large number of cam-type deformities in the asymptomatic population is in accordance with the literature (1417), where cam-type deformities of the proximal femur were commonly detected in patients undergoing CT of the abdomen and pelvis who had asymptomatic hip joints: In a study with 50 patients, some degree of cam-type deformity was found in 74% of asymptomatic patients in at least one plane of the reconstructed CT data set with a nonquantitative assessment (17). In a similar CT analysis of 50 hip joints, Narayanaswamy et al (25) found alpha angles greater than 55° in 56% of asymptomatic patients. Alpha angles of greater than 55° were found in 34% of patients in a study that used MR imaging with radial reconstructions (15). In a population of 244 young asymptomatic male individuals who underwent MR imaging with radial reconstructions, definite cam-type deformities were detected in 24%, and a mild decrease of the femoral head-neck offset was even seen in 74%, mostly at the anterosuperior position (16).

While the overall interreader agreement for alpha-angle measurements was good in our study, the two readers often did not agree whether the alpha angle was just above or just below the 55° threshold value in a specific individual. This is in accordance with recent reports: Validation of alpha-angle measurements in 50 consecutive patients who underwent MR imaging of the hip showed only moderate reproducibility after repeated assessment by the same reader (12). Other studies showed interobserver differences of up to 30% (13) and an ICC of only 0.50 for assessment of the alpha angle (16).

Discrimination between patients with FAI and asymptomatic volunteers was best at the anterosuperior position, as shown by the ROC analysis results. Discrimination between patients and volunteers was also possible at the anterior position and the superior position, but the area under the curve was substantially smaller at these positions. These results underline the importance of analyzing deformities of the proximal femur with radial images rather than only measuring the alpha angle at the anterior position, which is in accordance with other alpha-angle measurements in patients with FAI (3,22).

ROC analysis results also showed that the maximal alpha angle at any position can be used for discrimination between patients with FAI and asymptomatic volunteers, with an area under the curve that was similar to that at the anterosuperior radial position.

The appropriate threshold value for the alpha angle is an important issue. Our study has demonstrated a substantial overlap in alpha angles between patients and volunteers. An alpha-angle threshold value of 55° at the anterosuperior position showed a moderate to good sensitivity, but the associated low specificity is detrimental for distinguishing patients from volunteers, with a substantial number of false-positive results and the associated risk of overdiagnosis.

Increasing the alpha-angle threshold value from 55° to 60° resulted in a substantial gain in specificity, with a moderate loss in sensitivity, but was not an accurate measure for discriminating between patients with FAI and volunteers either. Pollard and colleagues (24) measured alpha angles on cross-table lateral radiographs in 83 healthy individuals and suggested that the alpha-angle threshold value should be increased to 63° on the basis of the 95% reference interval in their study population. In our MR data, we had an increased specificity, but a substantial loss of sensitivity, when we increased the alpha-angle threshold value to more than 60°. An alpha-angle threshold value of more than 60°, therefore, does not seem suitable for a reliable detection of patients with FAI.

Some authors have raised concerns about a possible overdiagnosis and overtreatment of patients who are suspected of having FAI (26,27). Further, the surgical treatment of FAI is associated with possible complications, such as fracture, nerve damage, prolonged pain, and others (28,29), and with the need for conversion to total hip arthroplasty in some patients (30,31). This factor highlights the need for a precise indication for FAI surgery before a patient undergoes surgery.

Because of the large overlap between alpha angles in patients with cam-type deformities and volunteers, it is difficult to describe an optimal alpha-angle threshold value that is both highly specific and sensitive. While it is important to recognize the features of FAI when analyzing a radiograph or MR image of the hip so that patients suffering from FAI can get the appropriate treatment, both radiologists and orthopedic surgeons should be cautious not to overcall this diagnosis, and the imaging findings need to be correlated with the clinical findings in every patient who is suspected of having FAI.

Our study has several limitations: All patients with FAI underwent MR arthrography of the hip, and all volunteers underwent nonenhanced MR of the hip; therefore, the two readers were not blinded to the fact whether an individual belonged to the patient group or the volunteer group. However, both readers performed the measurements in the same way in the patient group and in the volunteer group to minimize a possible bias. Apart from the MR sequences used for measuring the alpha angles, the two readers were allowed to see the images from the other MR sequences from the same examination for the purpose of orientation when evaluating the radial images, which is a potential source of bias. The clinical diagnosis of FAI is dependent on the experience of the orthopedic surgeon and can be affected by the subjectivity of the clinical examination, which might have resulted in an inclusion bias. Further, the inclusion of the patients with FAI in the study was based on a combination of clinical and imaging findings and not on surgical findings, which is also a limitation of our study. Alpha-angle measurements were performed on the same image by reader 1 and reader 2; if each reader had selected the image for the measurements individually, the interobserver agreement in our study might have been lower. Finally, our study only focused on the alpha angle when we assessed cam-type deformities of the femoral head-neck junction, and other methods of assessing cam-type deformities, such as the less commonly used femoral head-neck offset (20), the offset ratio (13), and the anterior femoral distance (32), were not evaluated.

In conclusion, the alpha angle does not help to accurately discriminate between volunteers and patients with FAI who have cam-type deformities. Discrimination is best at the anterosuperior segment. Increasing the alpha-angle threshold value from 55° to 60° reduces false-positive results while maintaining a reasonable sensitivity.

Advances in Knowledge
  • • Due to the large overlap in alpha-angle values between volunteers (mean, 53.3° ± 9.6 [standard deviation]) and patients (65.4° ± 11.5) (anterosuperior segment for reader 1), the alpha angle is not an accurate method for distinguishing asymptomatic volunteers from patients with cam-type deformities of the femoral head-neck junction.

  • • While a 55° alpha-angle threshold value has a good sensitivity for detection of cam-type deformities (81% [reader 1] and 90% [reader 2]), the low specificity (65% [reader 1] and 47% [reader 2]) yields many false-positive results.

Implication for Patient Care
  • • When utilizing the alpha angle to assess the contour of the femoral head-neck junction, radiologists and orthopedic surgeons should be cautious not to overcall the diagnosis of femoroacetabular impingement, because of the large overlap of measurements.

Disclosures of Potential Conflicts of Interest: R.S. No potential conflicts of interest to disclose. T.J.D. No potential conflicts of interest to disclose. P.O.Z. No potential conflicts of interest to disclose. C.W.A.P. No potential conflicts of interest to disclose.

Author Contributions

Author contributions: Guarantor of integrity of entire study, R.S.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, all authors; clinical studies, R.S., T.J.D., P.O.Z.; statistical analysis, R.S., C.W.A.P.; and manuscript editing, all authors

References

  • 1 Ganz R, Parvizi J, Beck M, Leunig M, Nötzli H, Siebenrock KA. Femoroacetabular impingement: a cause for osteoarthritis of the hip. Clin Orthop Relat Res 2003;(417):112–120. MedlineGoogle Scholar
  • 2 Wagner S, Hofstetter W, Chiquet Met al.. Early osteoarthritic changes of human femoral head cartilage subsequent to femoro-acetabular impingement. Osteoarthritis Cartilage 2003;11(7):508–518. Crossref, MedlineGoogle Scholar
  • 3 Pfirrmann CW, Mengiardi B, Dora C, Kalberer F, Zanetti M, Hodler J. Cam and pincer femoroacetabular impingement: characteristic MR arthrographic findings in 50 patients. Radiology 2006;240(3):778–785. LinkGoogle Scholar
  • 4 Allen D, Beaulé PE, Ramadan O, Doucette S. Prevalence of associated deformities and hip pain in patients with cam-type femoroacetabular impingement. J Bone Joint Surg Br 2009;91(5):589–594. Crossref, MedlineGoogle Scholar
  • 5 Beck M, Kalhor M, Leunig M, Ganz R. Hip morphology influences the pattern of damage to the acetabular cartilage: femoroacetabular impingement as a cause of early osteoarthritis of the hip. J Bone Joint Surg Br 2005;87(7):1012–1018. Crossref, MedlineGoogle Scholar
  • 6 Tannast M, Zheng G, Anderegg Cet al.. Tilt and rotation correction of acetabular version on pelvic radiographs. Clin Orthop Relat Res 2005;438:182–190. Crossref, MedlineGoogle Scholar
  • 7 Werner CM, Copeland CE, Ruckstuhl Tet al.. Radiographic markers of acetabular retroversion: correlation of the cross-over sign, ischial spine sign and posterior wall sign. Acta Orthop Belg 2010;76(2):166–173. MedlineGoogle Scholar
  • 8 Jamali AA, Mladenov K, Meyer DCet al.. Anteroposterior pelvic radiographs to assess acetabular retroversion: high validity of the “cross-over-sign”. J Orthop Res 2007;25(6):758–765. Crossref, MedlineGoogle Scholar
  • 9 Clohisy JC, Carlisle JC, Trousdale Ret al.. Radiographic evaluation of the hip has limited reliability. Clin Orthop Relat Res 2009;467(3):666–675. Crossref, MedlineGoogle Scholar
  • 10 Dudda M, Albers C, Mamisch TC, Werlen S, Beck M. Do normal radiographs exclude asphericity of the femoral head-neck junction? Clin Orthop Relat Res 2009;467(3):651–659. Crossref, MedlineGoogle Scholar
  • 11 Nötzli HP, Wyss TF, Stoecklin CH, Schmid MR, Treiber K, Hodler J. The contour of the femoral head-neck junction as a predictor for the risk of anterior impingement. J Bone Joint Surg Br 2002;84(4):556–560. Crossref, MedlineGoogle Scholar
  • 12 Nouh MR, Schweitzer ME, Rybak L, Cohen J. Femoroacetabular impingement: can the alpha angle be estimated? AJR Am J Roentgenol 2008;190(5):1260–1262. Crossref, MedlineGoogle Scholar
  • 13 Lohan DG, Seeger LL, Motamedi K, Hame S, Sayre J. Cam-type femoral-acetabular impingement: is the alpha angle the best MR arthrography has to offer? Skeletal Radiol 2009;38(9):855–862. Crossref, MedlineGoogle Scholar
  • 14 Panzer S, Augat P, Esch U. CT assessment of herniation pits: prevalence, characteristics, and potential association with morphological predictors of femoroacetabular impingement. Eur Radiol 2008;18(9):1869–1875. Crossref, MedlineGoogle Scholar
  • 15 Hack K, Di Primio G, Rakhra K, Beaulé PE. Prevalence of cam-type femoroacetabular impingement morphology in asymptomatic volunteers. J Bone Joint Surg Am 2010;92(14):2436–2444. Crossref, MedlineGoogle Scholar
  • 16 Reichenbach S, Jüni P, Werlen Set al.. Prevalence of cam-type deformity on hip magnetic resonance imaging in young males: a cross-sectional study. Arthritis Care Res (Hoboken) 2010;62(9):1319–1327. Crossref, MedlineGoogle Scholar
  • 17 Kang AC, Gooding AJ, Coates MH, Goh TD, Armour P, Rietveld J. Computed tomography assessment of hip joints in asymptomatic individuals in relation to femoroacetabular impingement. Am J Sports Med 2010;38(6):1160–1165. Crossref, MedlineGoogle Scholar
  • 18 Sutter R, Dietrich TJ, Zingg PO, Pfirrmann CWA. Femoral antetorsion: comparing asymptomatic volunteers and patients with femoroacetabular impingement. Radiology 2012;263(2):475–483. LinkGoogle Scholar
  • 19 Klaue K, Durnin CW, Ganz R. The acetabular rim syndrome: a clinical presentation of dysplasia of the hip. J Bone Joint Surg Br 1991;73(3):423–429. Crossref, MedlineGoogle Scholar
  • 20 Barton C, Salineros MJ, Rakhra KS, Beaulé PE. Validity of the alpha angle measurement on plain radiographs in the evaluation of cam-type femoroacetabular impingement. Clin Orthop Relat Res 2011;469(2):464–469. Crossref, MedlineGoogle Scholar
  • 21 Barros HJ, Camanho GL, Bernabé AC, Rodrigues MB, Leme LE. Femoral head-neck junction deformity is related to osteoarthritis of the hip. Clin Orthop Relat Res 2010;468(7):1920–1925. Crossref, MedlineGoogle Scholar
  • 22 Rakhra KS, Sheikh AM, Allen D, Beaulé PE. Comparison of MRI alpha angle measurement planes in femoroacetabular impingement. Clin Orthop Relat Res 2009;467(3):660–665. Crossref, MedlineGoogle Scholar
  • 23 Siebenrock KA, Ferner F, Noble PC, Santore RF, Werlen S, Mamisch TC. The cam-type deformity of the proximal femur arises in childhood in response to vigorous sporting activity. Clin Orthop Relat Res 2011;469(11):3229–3240. Crossref, MedlineGoogle Scholar
  • 24 Pollard TC, Villar 0RN, Norton MRet al.. Femoroacetabular impingement and classification of the cam deformity: the reference interval in normal hips. Acta Orthop 2010;81(1):134–141. Crossref, MedlineGoogle Scholar
  • 25 Narayanaswamy SM, Chakraverty JK, Sullivan C, Kamath S. Prevalence of morphological abnormalities predisposing to femoroacetabular impingement in aymptomatic young population on computed tomography [abstr]. In: Radiological Society of North America scientific assembly and annual meeting program. Oak Brook, Ill: Radiological Society of North American, 2010. http://rsna2010.rsna.org/search/search.cfm?action=add&filter=Author&value=90005. Accessed October 9, 2011. Google Scholar
  • 26 Palmer WE. Femoroacetabular impingement: caution is warranted in making imaging-based assumptions and diagnoses. Radiology 2010;257(1):4–7. LinkGoogle Scholar
  • 27 Hartofilakidis G, Bardakos NV, Babis GC, Georgiades G. An examination of the association between different morphotypes of femoroacetabular impingement in asymptomatic subjects and the development of osteoarthritis of the hip. J Bone Joint Surg Br 2011;93(5):580–586. Crossref, MedlineGoogle Scholar
  • 28 Ganz R, Gill TJ, Gautier E, Ganz K, Krügel N, Berlemann U. Surgical dislocation of the adult hip a technique with full access to the femoral head and acetabulum without the risk of avascular necrosis. J Bone Joint Surg Br 2001;83(8):1119–1124. Crossref, MedlineGoogle Scholar
  • 29 Matsuda DK, Carlisle JC, Arthurs SC, Wierks CH, Philippon MJ. Comparative systematic review of the open dislocation, mini-open, and arthroscopic surgeries for femoroacetabular impingement. Arthroscopy 2011;27(2):252–269. Crossref, MedlineGoogle Scholar
  • 30 Philippon MJ, Briggs KK, Yen YM, Kuppersmith DA. Outcomes following hip arthroscopy for femoroacetabular impingement with associated chondrolabral dysfunction: minimum two-year follow-up. J Bone Joint Surg Br 2009;91(1):16–23. Crossref, MedlineGoogle Scholar
  • 31 Horisberger M, Brunner A, Herzog RF. Arthroscopic treatment of femoroacetabular impingement of the hip: a new technique to access the joint. Clin Orthop Relat Res 2010;468(1):182–190. Crossref, MedlineGoogle Scholar
  • 32 Ito K, Minka MA, Leunig M, Werlen S, Ganz R. Femoroacetabular impingement and the cam-effect: a MRI-based quantitative anatomical study of the femoral head-neck offset. J Bone Joint Surg Br 2001;83(2):171–176. Crossref, MedlineGoogle Scholar

Article History

Received December 1, 2011; revision requested January 19, 2012; final revision received February 6; accepted February 23; final version accepted February 27.
Published online: Aug 2012
Published in print: Aug 2012