Sage Journals: Discover world-class research

Abstract

Background:

Acetabular morphology in hip dysplasia is typically assessed using computed tomography (CT) for bone coverage and magnetic resonance imaging (MRI) for soft tissues. However, agreement between CT and 3.0-T fast field echo (FFE) MRI for anterior and posterior acetabular sector angles (AASA, PASA) remains insufficiently defined.

Hypothesis:

FFE T1-weighted MRI measurements would strongly correlate with CT-based acetabular sector angles (ASAs).

Study Design:

Cohort study (diagnosis); Level of evidence, 2.

Methods:

A total of 65 symptomatic dysplastic hips were evaluated. Two independent observers measured on CT and 3.0-T MRI scans the lateral center-edge angle (LCEA), anterior and posterior acetabular wall indices (AWI, PWI), Tönnis angle, and pelvic signs. Equatorial, intermediate, and proximal AASA and PASA were obtained on each modality. CT-MRI agreement was assessed using Spearman rho (ρ).

Results:

Intermediate AASA demonstrated strong CT-MRI correlation (ρ = 0.807), whereas equatorial and proximal AASA correlations were moderate (ρ = 0.408 and 0.398, respectively). All PASA measurements showed good MRI-CT agreement (equatorial ρ = 0.766; intermediate ρ = 0.747; proximal ρ = 0.739). AWI correlated well with CT-AASAs but weakly with MRI-derived AASAs (equatorial ρ = 0.345; intermediate ρ = 0.325; proximal ρ = 0.255).

Conclusion:

T1-weighted FFE MRI cannot currently replace CT for measuring acetabular coverage of the femoral head. Although MRI and CT seemed to correlate better at the posterior level, they did not correlate accurately anteriorly; thus, the estimation of AWI by MRI with the current echo sequence alone may lead to misinterpretation. Given that the most common type of dysplasia in the setting of normal LCEA is that with an anterior wall defect only, the authors strongly recommend using CT (or other MRI sequences) to assess patients who are potential candidates for hip joint preservation surgery.

Keywords

hip dysplasia acetabular sector angles MRI CT acetabular wall coverage

Acetabular dysplasia is characterized by an insufficient acetabular bone coverage of the femoral head leading to an overloading of the lateral-most end with consequent chondrolabral damage and development of secondary osteoarthritis. In the prearthritic stage, symptomatic dysplasia often requires a surgical realignment osteotomy (ie, periacetabular osteotomy) to relieve pain and prevent early onset of secondary hip arthritis.^5,40,41

Acetabular dysplasia does not always result from a deficient lateral (ie, global) coverage of the femoral head, as variants of dysplasia have been described in the presence of so-called borderline or even normal lateral center-edge angle (LCEA). That is, there may be cases of dysplasia with exclusively anterior or posterior wall defect.^3,40 The most frequent radiographic parameters used to estimate the volume of acetabular coverage are LCEA, Tönnis angle (or acetabular index), Sharp angle, Lequesne or anterior center-edge angle, femoral head coverage index, crossover sign, and posterior wall sign.^9,10 Recently, novel measurements such as anterior wall index (AWI) and posterior wall index (PWI) have been described to assess anteroposterior coverage of the acetabulum.³⁰ However, radiographic measurements do not provide 3D or exact volumetric information on the acetabular coverage. Because of this, more complex cross-sectional studies are essential to better characterize the acetabular morphology.

For this matter, computed tomography (CT) has been useful to define patterns of acetabular coverage deficiency through different methods. Anda et al¹ in 1986 popularized a method to determine anterior and posterior femoral head coverage with 3 measurements in the axial plane: anterior acetabular sector angle (AASA), posterior acetabular sector angle (PASA), and the sum of both, horizontal acetabular sector angle (HASA). However, these measurements were performed only at the equatorial level (ie, center of rotation), without considering more proximal or distal levels of coverage. Subsequently, Nahal et al²⁴ and Verhaegen et al³⁹ further validated sector angle measurements in a cohort of dysplastic versus nondysplastic patients at different axial positions (equatorial, intermediate, and proximal) in CT scans, discriminating between normal and dysplastic hips, with further characterization of specific areas of undercoverage more proximal to the center of rotation.

Magnetic resonance imaging (MRI), unlike CT, is a useful tool to assess the soft tissues as well as detect labral and osteochondral lesions.²² Advances in image acquisition and processing have also allowed better visualization of bony contrast, multiplanar reconstructions, and spatial volume rendering similar to CT.¹² Currently, MRI-based measurements of femoral deformities have been validated,^20,37 but it is not yet clear whether acetabular volume can be calculated accurately with MRI as with CT. Recently, Goronzy et al¹³ compared MRI and CT measurements of acetabular bony morphology by measuring sector angles with the “clock method,” showing excellent results of intra- and interrater reliability between the 2 imaging techniques. However, no sample calculation was performed for that study, and the cohort included a small number of patients. In the absence of further studies validating acetabular volume measurements on MRI with CT measurements, and given the difficulty in reproducibility using the clock method,¹⁷ the primary objective of the present study was to assess the correlation of acetabular sector angle measurements between MRI and CT, with the latter being considered as the current gold standard.²⁴ The secondary objective was to correlate the measurement of acetabular sector angles with common radiological measurements of acetabular coverage, including LCEA, Tönnis angle, Sharp angle, extrusion index, AWI, PWI, crossover sign, posterior wall index, and ischial spine sign.

Methods

This was a prospective cohort study approved by the institution's research ethics board (IRB00010193; protocol #3985), carried out between December 2022 and August 2023. Initially, 80 hips (78 patients) were referred to the institutional hip preservation surgery clinic for which pelvic and lateral hip radiographs, 3D-CT, and 3D-MRI were indicated as a standard practice. This condition included both patients with frank dysplasia, defined by LCEA ≤20°, and those with so-called borderline dysplasia, defined by LCEA >20°, plus an AWI ≤0.30 and/or a PWI ≤0.85, as proposed by Bali et al.³ None of the included patients had a diagnosis of femoroacetabular impingement. Hip instability (ie, the main clinical finding of hip dysplasia) was diagnosed by the treating physician based not only on radiologic parameters but also on a comprehensive evaluation of patient symptoms (ie, pain or positive apprehension test) and clinical findings, including flexion >100°, extension >20°, internal rotation in 90° of flexion >30°, external rotation in 0° of flexion >25°, adduction >40°, and/or abduction >55°.³³ Fourteen hips (in 14 patients) were excluded: 7 of them due to inability to perform any of the cross-sectional studies adequately (either CT or MRI), 3 with deformities of the proximal femur that made angular measurements difficult (including cases with Perthes disease and slipped capital femoral epiphysis), and 4 with previous hip surgeries with osteosynthesis or implants that made measurements of various angles not feasible. Thus, 66 hips in 65 patients were finally included with adequate imaging studies. Patient data were obtained from the hospital electronic database, digitized since 2008.

All patients with adequate radiographs, CT scans, and MRI scans¹⁰ were included in the analysis. All anteroposterior pelvic radiographs were taken in supine position, with the x-ray tube placed at a distance of 1.5 to 2 m. CT scans were performed with low-dose radiation, with a slice thickness of 2 mm, from the upper plate of L5 to 1 cm below the lesser trochanter, and MRI scans were performed on a 3.0-T device from the upper plate of L5 to 1 cm below the lesser trochanter as well. MRI bony landmark measurements were performed on T1-weighted sequences consisting of T1-weighted fast field echo (FFE), which is a type of T1-weighted gradient-recalled echo (GRE),¹² as well as in fat suppression (T1 fat-saturated) and fluid-sensitive sequences, without using other specialized techniques such as volumetric interpolated breath-hold examination (VIBE). Both cross-sectional methods included an additional axial slice at the level of the femoral condyles to identify the posterior condylar line in order to calculate femoral version. Both sectional studies included axial oblique slices in the femoral neck axis for radial reconstruction. All reconstructions on both CT and MRI were obtained after multiplanar reorientation for pelvic alignment. First, the axial plane was corrected to the true anatomic axial orientation using the line connecting both ischial tuberosities, parallel to the imaging table. Second, the coronal and sagittal planes were realigned to the anterior pelvic plane by using both anterior superior iliac spines and the pubic symphysis as references. This ensured that pelvic tilt and/or rotation did not influence acetabular sector angle measurements.

For the radiological evaluation, 2 independent observers, a specialized radiologist expert in musculoskeletal pathologies and a senior fellow in hip preservation surgery, evaluated all radiographs and cross-sectional imaging studies. In case of disagreement, a hip preservation surgeon contributed with the radiological measurements. Radiographic analyses were performed with RAIM-Java Dicom viewer Version 3.2, whereas Alma Workstation Version 6.0 software was used for CT and MRI analyses. Radiographic metrics included the LCEA; acetabular index; acetabular wall indices (AWI, PWI); presence of retroversion signs (crossover sign, posterior wall sign, and ischial spine sign); and pelvic tilt, using the sacro-femoral-pubic (SFP) angle and the pubic symphysis–sacroiliac joint index (PS-SI)²⁹ (Figure 1).

Figure 1.

Radiographic measurement of sacro-femoral-pubic (SFP) angle, pubic symphysis–sacroiliac joint index (PS-SI), lateral center-edge angle (LCEA), and acetabular index (AI).

To standardize measurements, LCEA was calculated to the most lateral aspect of the anterior acetabular rim, defined as the most lateral whitish sclerotic line along the acetabular roof.²³ The AWI and PWI were measured as described by Siebenrock et al³⁰ by placing a circle over the femoral head that best approximates the shape of the femoral head and the center of rotation, after which the radius of the circle was determined. Then, a line was drawn along the axis of the femoral neck, cutting the circle through its center of rotation. The distances along this line between the medial intersection of the circle (line a, anterior wall) and the lateral intersection of the circle (line p, posterior wall) were recorded, and the AWI and PWI were then calculated by dividing lines a and p, respectively, by the radius of the circle (Figure 2). The SFP angle was calculated by the angle formed between a line from the midpoint of the upper endplate of S1 (found by determining the midpoint of a line between the lateral bodies of the L5-S1 facet joints), the center of rotation of the femoral head, and the upper midpoint of the pubic symphysis. Pelvic tilt was considered to be equal to 75 SFP angle.²⁹ The left and right SFP angles were measured, and when a difference >1° was obtained, the mean of the 2 measurements was the one used.⁶ The PS-SI was considered as the ratio of the length of a line drawn from the upper edge of the center of the pubic symphysis to its intersection with the midpoint of an intersacroiliac line (a line drawn between both undersides of the sacroiliac joints).²⁹

Figure 2.

A circle is drawn to approximate the femoral head, and the radius of the head (r) is determined. A line from the medial edge of the circle to the anterior (a) and posterior (p) wall is drawn and measured along the femoral neck axis. The anterior wall index and posterior wall index are calculated as a/r and p/r, respectively.

The same 2 independent observers performed all CT- and MRI-based measurements, which included acetabular version at the equatorial level, AASA, and PASA. Acetabular version was calculated in the axial plane using a line joining the anterior and posterior margins of the bony acetabular rim, correcting for rotation by the position of the ischial tuberosities.³² With regard to MRI characterization, labral thickness and width were not routinely measured or characterized, other than describing the labrum as torn or not and with or without associated paralabral cysts.

Measurements of acetabular sector angles were performed on axial slices, both on CT and MRI, following the guidelines established by Nahal et al,²⁴ who recently validated the original measurement method described by Anda¹ in 1986. These measurements were performed at 3 different levels: equatorial (ie, at the level of the center of rotation), proximal, and intermediate. The equatorial level was defined on a coronal image of both CT and MRI, located at the level of the center of rotation of both femoral heads. At this level, measurements were made of the AASA, which was determined as the angle between a line connecting the center of each femoral head and a line from the center of the femoral head to the anterior margin of the acetabulum, as well as the PASA, which was measured as the angle between a line connecting the center of each femoral head and a line from the center of the femoral head to the posterior margin of the acetabulum (eq-AASA and eq-PASA, respectively). The proximal level was identified as 2.5 mm below the superior surface of both femoral heads in a coronal image. At this level, proximal AASA and PASA (pro-AASA and pro-PASA) measurements were performed as described above. Finally, the intermediate level (int-AASA and int-PASA) was defined as the midpoint between the proximal and equatorial levels. At this level, AASA and PASA measurements were made using the same method³⁹ (Figure 3).

Figure 3.

(A) Coronal section of computed tomography (CT) scan, where the equatorial, proximal, and intermediate levels are determined. Axial sections of (B) magnetic resonance image and (C) CT scan at the 3 levels where anterior acetabular sector angle (AASA) and posterior acetabular sector angle (PASA) measurements were performed.

Reliability

Before measurements were performed on radiographs, CT scans, and MRI scans, a training session was carried out involving the reading of 10 radiographs, CT scans, and MRI scans from cases not belonging to the included cohort. The mean values of the 2 observers’ measurements were then compared. Agreement between the measurements was assessed by Bland-Altman limits of agreement analysis. The difference in measurements between the 2 independent observers was not statistically significant based on paired t tests (all P values >.05), with distribution of measurements within 95% limits of agreement when analyzing the difference of measurement/mean measurement. Mean interobserver difference was −0.45° (95% CI, −1.32° to −0.39°; P = .07) for radiographic LCEA; −0.27° (95% CI, −0.58° to 0.52°; P = .92) for radiographic anterior center-edge angle; −0.015 (95% CI, −0.03 to 0.002; P = .096) for radiographic AWI; −0.018 (95% CI, −0.08 to 0.01; P = .086) for radiographic PWI; −0.008 (95% CI, −0.38 to 0.36; P = .96) for PS-SI; 0.006° (95% CI, −0.44° to 0.44°; P = .92) for SFP angle; −0.24 mm (95% CI, −0.99 to 0.5 mm; P = .51) for tomographic eq-AASA; 1.36° (95% CI, −0.15° to 2.87°; P = .08) for tomographic pelvic tilt; and −0.33° (95% CI, −0.83° to 0.16°, P = .18) for CT-acetabular version.

Statistical Analysis

Statistical analysis was performed with Statistical Package for the Social Sciences (SPSS) Version 16 (IBM). Nonparametric tests were used for the analysis. The independent-samples t test was used for scale data and the chi-square test for categorical data. Spearman correlation (rho coefficient) was used to determine whether correlations existed for scale data including the different radiographic, CT, and MRI measurements. Spearman coefficients between 0.10 and 0.39 were considered as weak correlation, between 0.4 and 0.69 as moderate correlation, between 0.7 and 0.89 as strong correlation, and between 0.9 and 1 as very strong correlation.³² Statistical significance was set at P < .05.

A sample size calculation was performed to evaluate the null hypothesis of correlation between the acetabular sector angle (ASA) on CT versus MRI considering a coefficient ≤0.6. As an alternative hypothesis, a correlation of 0.8 was expected, with a power of 80% and an alpha of 5%, and using a 2-tailed test, therefore needing at least 51 patients. The expected correlation was based on estimates obtained in previous studies.¹³ The power analysis and all statistical analyses were performed with STATA Version 16.

Results

Demographic Characteristics

The final study population included 66 hips (65 patients); mean age of the patients was 29.98 ± 7.7 years. Mean LCEA, Tönnis angle, and SFP angle were 27.04°± 6°, 7°± 3.8°, and 68.2°± 5.8°, respectively. Mean AWI, PWI, and pelvic incidence were 0.24 ± 0.1, 0.96 ± 0.18, and 6.7 ± 5.7, respectively. We found that 46% of the hips had crossover sign, whereas the ischial spine sign was present in 21.5% of cases (Table 1). The mean CT-equatorial acetabular version was 19.14°± 4.9°, whereas the mean MRI-equatorial acetabular version was 17.5°± 5.5°.

Table 1

Patient Characteristics ^a

Variable	Value
No. of hips (patients)	66 (65)
Age, y	29.98 ± 7.7
Body mass index, kg/m²	24.5 ± 10
Lateral center-edge angle, deg	27.04 ± 6
Tönnis angle, deg	7 ± 3.8
Crossover sign, % (n/N)	46.2 (30/65)
Ischiatic spine sign, % (n/N)	21.5 (14/65)
Anterior wall index	0.24 ± 0.1
Posterior wall index	0.96 ± 0.18
Sacro-femoral-pubic angle, deg	68.2 ± 5.8
Pubic symphysis–sacroiliac joint index	95.3 ± 13.3
Pelvic tilt, deg	6.7 ± 5.7

Values are expressed as mean ± SD unless otherwise noted.

Mean eq-AASA and eq-PASA were 52.7°± 5.8° and 95.5°± 10° for CT and 61.9°± 14.8° and 95.7°± 7.5° for MRI, respectively. Mean int-AASA and int-PASA were 73.74°± 10.5° and 105°± 12° for CT and 83.7°± 82.7° and 102.8°± 16° for MRI, respectively. Mean pro-AASA and pro-PASA were 112.5°± 14.1° and 127.2°± 28.2° for CT and 107°± 23.7° and 121.6°± 26.4° for MRI, respectively (Table 2).

Table 2

Mean Acetabular Sector Angles at Equatorial, Intermediate, and Proximal Levels ^a

Level	Imaging Modality	AASA, deg	PASA, deg
Proximal	CT	112.5 ± 14.1	127.2 ± 28.2
	MRI	107 ± 23.7	121.6 ± 26.4
Intermediate	CT	73.74 ± 10.5	105 ± 12
	MRI	83.7 ± 82.7	102.8 ± 16
Equatorial	CT	52.7 ± 5.8	95.5 ± 10
	MRI	61.9 ± 14.8	95.7 ± 7.5

Values are expressed as mean ± SD. AASA, anterior acetabular sector angle; CT, computed tomography; MRI, magnetic resonance imaging; PASA, posterior acetabular sector angle.

Correlation Between Acetabular Sector Angle Measurements on MRI and CT Scans

Spearman rho coefficients were 0.408 for eq-AASA, 0.807 for int-AASA, 0.398 for pro-AASA, 0.766 for eq-PASA, 0.747 for int-PASA, 0.739 for pro-PASA, and 0.846 for acetabular version (Table 3).

Table 3

Correlation Between Acetabular Sector Angle Measurements on Magnetic Resonance Imaging and Computed Tomography (Spearman Rho) ^a

Level	AASA	PASA
Proximal	0.398 ^b	0.739 ^b
Intermediate	0.807 ^b	0.747 ^b
Equatorial	0.408 ^b	0.766 ^b

AASA, anterior acetabular sector angle; PASA, posterior acetabular sector angle.

P < .001.

Linear regression demonstrated markedly different concordance patterns for AASA and PASA values. For AASA, coefficients of determination were extremely low at all levels—equatorial R² = 0.029, intermediate R² = 0.052, and proximal R² = 0.005—indicating that MRI measurements of anterior coverage did not align with the CT reference standards (Figures 4 -6). For PASA, we found variability but overall better agreement at all 3 levels: equatorial R² = 0.328, intermediate R² = 0.151, and proximal R² = 0.875 (Figures 7 -9).

Figure 4.

Simple scatter plot with a linear fit of the correlation between 2 angles: the computed tomography (CT) anterior acetabular sector angle (AASA) at the equatorial level and the magnetic resonance imaging (MRI) AASA at the equatorial level (linear R² = 0.029).

Figure 5.

Simple scatter plot with a linear fit of the correlation between 2 angles: the computed tomography (CT) anterior acetabular sector angle (AASA) at intermediate level and the magnetic resonance imaging (MRI) AASA at intermediate level (linear R² = 0.637).

Figure 6.

Simple scatter plot with a linear fit of the correlation between 2 angles: the computed tomography (CT) anterior acetabular sector angle (AASA) at the proximal level and the magnetic resonance imaging (MRI) AASA at the proximal level (linear R² = 0.005).

Figure 7.

Simple scatter plot with a linear fit of the correlation between 2 angles: the computed tomography (CT) posterior acetabular sector angle (PASA) at the equatorial level and the magnetic resonance imaging (MRI) PASA at the equatorial level (linear R² = 0.328).

Figure 8.

Simple scatter plot with a linear fit of the correlation between 2 angles: the computed tomography (CT) posterior acetabular sector angle (PASA) at the intermediate level and the magnetic resonance imaging (MRI) PASA at the intermediate level (linear R² = 0.151).

Figure 9.

Simple scatter plot with a linear fit of the correlation between 2 angles: the computed tomography (CT) posterior acetabular sector angle (PASA) at the proximal level and the magnetic resonance imaging (MRI) PASA at the proximal level (linear R² = 0.875).

Correlation Between Radiographic AWI and AASA Measured on MRI and CT Scans

The correlation coefficients of AWI with equatorial, intermediate, and proximal tomographic AASAs were 0.779, 0.531, and 0.613, respectively (ie, good to strong). In contrast, the correlation coefficients between AWI with equatorial, intermediate, and proximal MRI AASAs were 0.345, 0.325, and 0.255, respectively (ie, weak) (Table 4).

Table 4

CT and MRI Correlation Coefficients With AWI and PWI (Spearman Rho) ^a

	CT (Proximal)	CT (Intermediate)	CT (Equatorial)	MRI (Proximal)	MRI (Intermediate)	MRI (Equatorial)
AASA (AWI)	0.613 ^b	0.531 ^c	0.779 ^c	0.255 ^b	0.325 ^c	0.345 ^c
PASA (PWI)	–0.200	–0.157	–0.207	–0.146	–0.029	–0.224
AASA (AWI)	–0.079	–0.204	–0.073	–0.030	0.358 ^c	0.565 ^c
PASA (PWI)	0.385 ^c	0.664 ^c	0.673 ^c	0.226	0.065	0.720 ^c

AASA, anterior acetabular sector angle; AWI, anterior wall index; CT, computed tomography; MRI, magnetic resonance imaging; PASA, posterior acetabular sector angle; PWI, posterior wall index.

P < .001.

P < .01.

Correlation Between Radiographic PWI and PASA Measured on MRI and CT Scans

The correlation coefficients of PWI with equatorial, intermediate, and proximal CT PASAs were 0.673, 0.664, and 0.385, respectively (moderate to strong). In contrast, the correlation coefficients between PWI and equatorial, intermediate, and proximal MRI PASAs were 0.720, 0.065, and 0.226, respectively (strong for equatorial but weak for intermediate and proximal PASAs) (Table 4).

Discussion

Currently, very few studies have focused on assessing the reliability of MRI for the characterization of acetabular bone morphology.^13,34 Rather, the use of MRI has focused more on the finding (or not) of chondral and acetabular labral lesions, in addition to ruling out associated pathologies (eg, oncological).^15,27 In the present study, our primary objective was to evaluate the correlation between ASA measurements (AASA and PASA measurements at 3 levels) between MRI and CT, with the latter being the current gold standard for the evaluation of acetabular morphology in candidates for hip preservation surgery, in order to elucidate whether MRI could replace CT and thus avoid additional irradiation.

CT is currently a validated imaging modality for delineating acetabular morphology in hip preservation patients, with several methods available for measuring acetabular volume. Among them are the ASAs implemented by Anda et al¹ in the 1980s. Since then, sector angles have been not only measured at the equatorial level but also validated at more proximal levels as we used in the current study, where more specific anterior or posterior wall deficiencies may help distinguish cases with mild or so-called borderline dysplasia.^24,39 Other, similar 3D-CT-based measurements¹⁷ assess the percentage of femoral head coverage at each clock position using 3D reconstruction. Based on the findings by Larson et al,¹⁷ normative data were established for acetabular coverage both globally and at specific locations, including the anterior zone (corresponding to 3 o’clock), the lateral zone (12 o’clock), and the posterior zone (9 o’clock). Likewise, Nepple et al²⁶ applied the same method in a 3D-CT study including a cohort of patients with acetabular dysplasia and characterized 3 patterns of acetabular deficiency: global deficiency, anterosuperior deficiency, and posterosuperior deficiency. So far, 3D-CT measurements of acetabular wall coverage help hip preservation surgeons better distinguish among dysplasia and femoroacetabular impingement, being the current cross-sectional study for the painful young adult hip.³¹

MRI, initially intended to evaluate the soft tissues, has been also implemented in the evaluation of bony acetabular morphology in the painful young adult hip.³⁴ In a retrospective analysis of 50 painful hips in young adults, Subramanian et al³⁴ compared center-edge angle, Tönnis angle, AASA, PASA, and acetabular version between MRI-T1-VIBE inversion and CT images and found that T1-VIBE inversion sequence was an effective alternative to CT with the added advantage of alleviating radiation exposure. Similarly, Breighner et al⁸ used zero-echo-time T1-weighted sequences to measure coronal and sagittal center-edge angles, femoral neck-shaft angle, acetabular version (at 1-, 2-, and 3-o’clock positions), Tönnis angle, alpha angle, and modified beta angle and reported intraclass correlation coefficient values ranging from 0.636 to 0.990 for zero-echo-time and 0.747 to 0.983 for CT, indicating good to excellent agreement. Comparably, Goronzy et al¹³ used the clockface technique in a cohort of 20 symptomatic patients and found that VIBE-sequenced MRI had a high correlation with tomography, with good intra- and interobserver reliability and correlation index values ranging from 0.969 to 0.999. The same study group showed similar results with excellent agreement (intraclass correlation coefficient 0.98-0.99) for the ASAs measured at 1, 12, and 11 o'clock as well as for the equatorial acetabular version (intraclass correlation coefficient 0.95).⁷ Like the former authors, Goronzy et al¹³ used a different MRI sequence than the one used in our study to measure the sector angles, consisting of a 3D-isotropic VIBE sequence with a slice thickness of 0.8 mm in a 3.0-T Siemens Magnetom Verio device.⁵ Although the FFE sequence used in our cohort is a basic 3D gradient echo sequence, the VIBE sequence used in other studies is an advanced, fast 3D-gradient echo sequence that uses breath-holding to reduce motion artifacts and improve image quality, being also applicable with fat suppression to distinguish contrast between bone and soft tissue.³⁸ Additionally, at our institution, it was not possible to standardize the use of the clockface method because it requires additional radial reconstruction angles to be plotted by hand since there is no specific software, with the potential bias of error in measurement. However, the measurement method of 3-level ASAs used in the present study has been extensively validated.^24,39

The most important finding of our study was that using ASA measurements at 3 different levels (proximal, intermediate, and equatorial) had a strong correlation between MRI and CT for the assessment of posterior coverage, whereas they had a worse correlation (ie, moderate to weak) for delineating anterior wall coverage. This is an interesting finding that could be attributed to the difficulty in differentiating between bony and labral tissue on MRI scans with the FFE sequence. However, measurements of PASAs were similar in both imaging modalities. That is, it seems to be easier to detect bony margins posteriorly than anteriorly, especially with the FFE and fat-suppressed sequences. As reported in a recent systematic review, spoiled GRE sequences such as VIBE provide enhanced bone-to-soft tissue contrast when compared with the FFE sequence, particularly when used with Dixon reconstruction.¹⁶ Also, fat saturation is usually applied to the VIBE sequence in order to better differentiate cortical bone from fat containing bone marrow and avoid chemical shift artifacts.¹⁸ Therefore, because the hip labrum, which is more robust anteriorly than posteriorly (especially in dysplastic hips), is mainly composed of collagen fibers, fibrocartilage, and a varying amount of water,^4,11 the VIBE sequence may be more appropriate than the FFE to delineate bony margins anteriorly.¹⁹

Several classic radiographic measurements are widely used in clinical practice to assess acetabular coverage.³⁶ These metrics are relatively easy to measure and of low complexity; however, they require both proper patient positioning and accurate acquisition techniques.²¹ Several studies have shown that radiographic measurements are subject to change according to pelvic tilt.^14,35 Within the radiographic armamentarium, AWI is one of the most widely used measurements, most recently validated by Slullitel et al,³² who found a strong correlation between this metric and equatorial AASA on tomography. This same result was found in our study with both AASA and PASA on CT, but not on MRI, where the correlation between the 3 MRI-AASAs and AWI was generally weak. For PWI, in contrast, a better correlation was seen with the PASAs, being moderate to strong in both CT and MRI. This implies that radiographs correlated better with CT than with MRI, especially for assessing anterior acetabular coverage. Nonetheless, others have shown poor correlations between radiographic measurement of AWI and PWI with that of axial-oblique CT projections, even for posterior coverage (correlation coefficient 0.37-0.45),²⁵ although ASAs were not measured. Thus, it is currently recommended that the assessment of acetabular morphology should not be performed with a single radiographic parameter but rather should entail more complex sectional images.

Our study was not without limitations. First, the selected cohort, although prospective, included only patients with hip joint pathology and did not include an asymptomatic cohort of patients, which is usually necessary to validate measurements in hip joint preservation surgery.^2,28 Second, the cohort included a group of hips that appeared to have a diagnosis of dysplasia with some degree of anterior coverage defect, considering the overall low AWI of the series.³ Perhaps including a cohort of patients with more homogeneous measurements would have been useful to compare measurement of AASA and PASA among different dysplasia subtypes.^3,40 Third, the CT and MRI slices in this series, used at the institution, were performed every 2 mm, and a more accurate spatial position in the axial plane would require slices performed every 0.5 to 0.8 mm.

Conclusion

T1-weighted FFE MRI cannot currently replace CT for measuring acetabular coverage of the femoral head. Although MRI and CT seemed to correlate better at the posterior level, they did not correlate accurately anteriorly; thus, the estimation of AWI by MRI with the current echo sequence solely may lead to misinterpretation. Given that the most common type of dysplasia in the setting of normal LCEA is that with an anterior wall defect only,^2,3 we strongly recommend using CT (or other MRI sequences) to assess cases that are potential candidates for hip joint preservation surgery.

Footnotes

Final revision submitted February 2, 2026; accepted February 10, 2026.

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical approval for this study was obtained from Hospital Italiano de Buenos Aires IRB00010193; protocol #3985.

ORCID iD

Agustin Albani-Forneris

References

Anda

Svenningsen

Dale

Benum

The acetabular sector angle of the adult hip determined by computed tomography. Acta Radiol Diagn. 1986;27(4):443-447.

Anderson

Erickson

Chrastil

Peters

CL.

Acetabular wall indices help to distinguish acetabular coverage in asymptomatic adults with varying morphologies. Clin Orthop Relat Res. 2017;475(4):1027-1033.

Bali

Smit

Ibrahim

, et al. Ottawa classification for symptomatic acetabular dysplasia assessment of interobserver and intraobserver reliability. Bone Joint Res. 2020;9(5):242-249.

Beaulé

O’Neill

Rakhra

Acetabular labral tears. J Bone Joint Surg Am. 2009;91(3):701-710.

Bellova

Blum

Hartmann

Thielemann

Günther

Goronzy

MRI-based assessment of acetabular version and coverage after previous Pemberton osteotomy in skeletally mature patients. J Child Orthop. 2021;15(3):223-231.

Blondel

Schwab

Patel

, et al. Sacro-femoral-pubic angle: a coronal parameter to estimate pelvic tilt. Eur Spine J. 2012;21(4):719-724.

Blum

Franken

Hartmann

, et al. MRI-based static and functional assessment of complex hip deformities in comparison with CT: a validation study. MAGMA. 2020;33(6):829-838.

Breighner

Bogner

Lee

Koff

Potter

HG.

Evaluation of osseous morphology of the hip using zero echo time magnetic resonance imaging. Am J Sports Med. 2019;47(14):3460-3468.

Clohisy

Carlisle

Beaulé

, et al. A systematic approach to the plain radiographic evaluation of the young adult hip. J Bone Joint Surg Am. 2008;90(suppl 4):47-66.

10.

Clohisy

Carlisle

Trousdale

, et al. Radiographic evaluation of the hip has limited reliability. Clin Orthop Relat Res. 2009;467(3):666-675.

11.

Dwyer

Jones

Hogan

Field

McCarthy

Noble

PC.

The acetabular labrum regulates fluid circulation of the hip joint during functional activities. Am J Sports Med. 2014;42(4):812-819.

12.

Florkow

Willemsen

Mascarenhas

Oei

EHG

van Stralen

Seevinck

PR.

Magnetic resonance imaging versus computed tomography for three-dimensional bone imaging of musculoskeletal pathologies: a review. J Magn Reson Imaging. 2022;56(1):11-34.

13.

Goronzy

Blum

Hartmann

, et al. Is MRI an adequate replacement for CT scans in the three-dimensional assessment of acetabular morphology? Acta Radiol. 2019;60(6):726-734.

14.

Henebry

Gaskill

The effect of pelvic tilt on radiographic markers of acetabular coverage. Am J Sports Med. 2013;41(11):2599-2603.

15.

Higashihira

Kobayashi

Oishi

, et al. Comparison between 3-dimensional multiple-echo recombined gradient echo magnetic resonance imaging and arthroscopic findings for the evaluation of acetabular labrum tear. Arthroscopy. 2019;35(10):2857-2865.

16.

Kavrakova

Haage

Stueckle

CA.

Current state-of-the-art 3D MRI sequences for assessing bone morphology with emphasis on cranial and spinal imaging: a narrative review. Rofo. 2026;198(3):301-324.

17.

Larson

Moreau-Gaudry

Kelly

, et al. Are normal hips being labeled as pathologic? A CT-based method for defining normal acetabular coverage. Clin Orthop Relat Res. 2015;473(4):1247-1254.

18.

Dixon techniques for water and fat imaging. J Magn Reson Imaging. 2008;28(3):543-558.

19.

Mak

Teh

Magnetic resonance imaging of the hip: anatomy and pathology. Pol J Radiol. 2020;85(1):e489-e508.

20.

Mao

Liang

Ding

, et al. The consistency between measurements of the femoral neck anteversion angle in DDH on three-dimensional CT and MRI. Acta Radiol. 2016;57(6):716-720.

21.

Mascarenhas

Ayeni

Egund

, et al. Imaging methodology for hip preservation: techniques, parameters, and thresholds. Semin Musculoskelet Radiol. 2019;23(3):197-226.

22.

Mintz

Hooper

Connell

Buly

Padgett

Potter

HG.

Magnetic resonance imaging of the hip: detection of labral and chondral abnormalities using noncontrast imaging. Arthroscopy. 2005;21(4):385-393.

23.

Monazzam

Bomar

Cidambi

Kruk

Hosalkar

Lateral center-edge angle on conventional radiography and computed tomography. Clin Orthop Relat Res. 2013;471(7):2233-2237.

24.

Nahal

Slullitel

Kamenaga

, et al. Acetabular coverage analysis of the proximal femoral head accurately characterizes dysplastic acetabular morphology. J Orthop Res. 2023;41(6):1273-1282.

25.

Nazaroff

Mark

Learned

Wang

Measurement of acetabular wall indices: comparison between CT and plain radiography. J Hip Preserv Surg. 2021;8(1):51-57.

26.

Nepple

Wells

Ross

Bedi

Schoenecker

Clohisy

JC.

Three patterns of acetabular deficiency are common in young adult patients with acetabular dysplasia. Clin Orthop Relat Res. 2017;475(4):1037-1044.

27.

Nishii

Tanaka

Nakanishi

Sugano

Miki

Yoshikawa

Fat-suppressed 3D spoiled gradient-echo MRI and MDCT arthrography of articular cartilage in patients with hip dysplasia. AJR Am J Roentgenol. 2005;185(2):379-385.

28.

Piuzzi

NS.

CORR insights: acetabular wall indices help to distinguish acetabular coverage in asymptomatic adults with varying morphologies. Clin Orthop Relat Res. 2017;475(4):1034-1036.

29.

Roussot

Salih

Grammatopoulos

Witt

JD.

What is the pelvic tilt in acetabular dysplasia and does it change following peri-acetabular osteotomy?

J Hip Preserv Surg. 2020;7(4):777-785.

30.

Siebenrock

Kistler

Schwab

Büchler

Tannast

The acetabular wall index for assessing anteroposterior femoral head coverage in symptomatic patients. Clin Orthop Relat Res. 2012;470(12):3355-3360.

31.

Slullitel

Coutu

Buttaro

Beaule

Grammatopoulos

Hip preservation surgery and the acetabular fossa: a canary in a coal mine?

Bone Joint Res. 2020;9(12):857-869.

32.

Slullitel

Holc

Buljubasich

, et al. Anterior center-edge angle is less reliable than anterior wall index to predict anterior coverage of the femoral head. Am J Sports Med. 2023;51(8):2151-2160.

33.

Steppacher

Zurmühle

Puls

, et al. Periacetabular osteotomy restores the typically excessive range of motion in dysplastic hips with a spherical head. Clin Orthop Relat Res. 2015;473(4):1404-1416.

34.

Subramanian

Hegde

Azzopardi

, et al. TI VIBE inversion MRI—an alternative to CT for imaging of hip pain. J Clin Orthop Trauma. 2021;19:196-199.

35.

Tannast

Fritsch

Zheng

Siebenrock

Steppacher

SD.

Which radiographic hip parameters do not have to be corrected for pelvic rotation and tilt?

Clin Orthop Relat Res. 2015;473(4):1255-1266.

36.

Tannast

Hanke

Zheng

Steppacher

Siebenrock

KA.

What are the radiographic reference values for acetabular under- and overcoverage?

Clin Orthop Relat Res. 2015;473(4):1234-1246.

37.

Tomczak

Guenther

Rieber

Mergo

Ros

Brambs

HJ.

MR imaging measurement of the femoral antetorsional angle as a new technique: comparison with CT in children and adults. AJR Am J Roentgenol. 1997;168(3):791-794.

38.

Tsuchiya

Gomyo

Katase

Hiraoka

Tateishi

Magnetic resonance bone imaging: applications to vertebral lesions. Jpn J Radiol. 2023;41(11):1173-1185.

39.

Verhaegen

JCF

DeVries

Horton

, et al. Acetabular sector angles in asymptomatic and dysplastic hips: defining dysplasia and thresholds to guide management. J Bone Joint Surg Am. 2023;105(21):1709-1720.

40.

Wilkin

Ibrahim

Smit

Beaulé

PE.

A contemporary definition of hip dysplasia and structural instability: toward a comprehensive classification for acetabular dysplasia. J Arthroplasty. 2017;32(9S):S20-S27.

41.

Wylie

Peters

Aoki

SK.

Natural history of structural hip abnormalities and the potential for hip preservation. J Am Acad Orthop Surg. 2018;26(15):515-525.

Fast Field Echo Magnetic Resonance Imaging For Quantifying Acetabular Wall Coverage: A Validation Study With Computed Tomography

Abstract

Background:

Hypothesis:

Study Design:

Methods:

Results:

Conclusion:

Keywords

Methods

Reliability

Statistical Analysis

Results

Demographic Characteristics

Correlation Between Acetabular Sector Angle Measurements on MRI and CT Scans

Correlation Between Radiographic AWI and AASA Measured on MRI and CT Scans

Correlation Between Radiographic PWI and PASA Measured on MRI and CT Scans

Discussion

Conclusion

Footnotes

ORCID iD

References