Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Sep 1.
Published in final edited form as: Magn Reson Imaging. 2013 May 16;31(7):1129–1136. doi: 10.1016/j.mri.2013.01.009

Regional variations in MR relaxation of hip joint cartilage in subjects with and without femoralacetabular impingement☆,☆☆

Karupppasamy Subburaj a,*, Alexander Valentinitsch a, Alexander B Dillon a, Gabby B Joseph a, Xiaojuan Li a, Thomas M Link a, Thomas P Vail b, Sharmila Majumdar a
PMCID: PMC3729918  NIHMSID: NIHMS483938  PMID: 23684960

Abstract

The objective of this study was to analyze regional variations of magnetic resonance (MR) relaxation times (T1ρ and T2) in hip joint cartilage of healthy volunteers and subjects with femoral acetabular impingement (FAI). Morphological and quantitative images of the hip joints of 12 healthy volunteers and 9 FAI patients were obtained using a 3 T MR scanner. Both femoral and acetabular cartilage layers in each joint were semi-automatically segmented on sagittal 3D high-resolution spoiled gradient echo (SPGR) images. These segmented regions of interest (ROIs) were automatically divided radially into twelve equal sub-regions (300 intervals) based on the fitted center of the femur head. The mean value of T1ρ/T2 was calculated in each subregion after superimposing the divided cartilage contours on the MR relaxation (T1ρ/T2) maps to quantify the relaxation times. T1ρ and T2 relaxation times of the femoral cartilage were significantly higher in FAI subjects compared to healthy controls (39.9 ± 3.3 msec in FAI vs. 35.4 ± 2.3 msec in controls for T1ρ (P = 0.0020); 33.9 ± 3.1 msec in FAI vs. 31.1 ± 1.7 msec in controls for T2 (P = 0.0160)). Sub-regional analysis showed significantly different T1ρ and T2 relaxation times in the anterior-superior region (R9) of the hip joint cartilage between subjects with FAI and healthy subjects, suggesting possible regional differences in cartilage matrix composition between these two groups. Receiver operating characteristic (ROC) analysis showed that subregional analysis in femoral cartilage was more sensitive in discriminating FAI joint cartilage from that of healthy joints than global analysis of the whole region (T1ρ: area under the curve (AUC) = 0.981, P = 0.0001 for R9 sub-region; AUC = 0.901, P = 0.002 for whole region; T2: AUC = 0.976, P = 0.0005 for R9 sub-region; AUC = 0.808, P = 0.0124 for whole region). The results of this study demonstrated regional variations in hip cartilage composition using MR relaxation times (T1ρ and T2) and suggested that analysis based on local regions was more sensitive than global measures in subjects with and without FAI.

Keywords: MRI, Hip, Cartilage, T1ρ and T2, Femoral-acetabular impingement

1. Introduction

Over the past decade, conditions such as femoroacetabular impingement (FAI) [1–4] and acetabular dysplasia [5,6] have been identified as pertinent causes of premature osteoarthritis (OA) of the hip joint in young and middle-aged patients [7]. Cam-type FAI, characterized by an anatomic deformity at the femoral head-neck junction, is known to affect the pathogenesis of hip OA [8,9]. Similarly, a range of normal anatomy at the femoral head-neck junction suggests that not all anatomic variations result in joint degeneration. Surgical treatment has been recommended to reduce clinical symptoms and to delay the onset of OA. The outcome of surgical intervention depends on the degree of pre-existing OA with naturally poorer results in patients with advanced changes. Thus, early detection of cartilage degeneration could help identify patients with hip pain who may benefit from early surgical intervention. The biomechanical stresses in the soft tissues depend strongly on the femoral head and acetabular geometry. Computational and imaging studies have shown that, in impinging joints, day-to-day activities involving extensive motion, inducing excessive distortion and shearing of the tissue-bone interface and high contact stress [8,9], lead to morphological and structural damage [9].

Structural anatomy and the extent of hip OA are typically assessed by the plain radiography; however, this imaging modality is known to be insensitive to early stages of OA. The need to noninvasively detect the earliest changes in the degeneration of articular cartilage in order to implement and validate potential early therapeutic interventions has stimulated considerable interest in the development of techniques, like quantitative magnetic resonance imaging (MRI) that can directly probe the macromolecular structure. The utilization of MRI to assess in-vivo cartilage morphology (volume and thickness) [10], anatomical alterations (lesions), composition, and functional properties [11] has received considerable interest in recent years. However, evaluation of the articular hip cartilage with MRI is extremely challenging, because of the thinness of the cartilage and spherical surface geometry of the femoral head and acetabulum. Despite significant improvements in MRI technology over the past decade, a major limitation of currently available sequences is their inability to consistently detect superficial degenerative and posttraumatic cartilage lesions that may progress to more advanced OA. T1ρ and T2 relaxation time mapping have recently emerged as potential markers of early biochemical cartilage degeneration. These measures are highly sensitive to alterations in composition and structural integrity of collagen in the cartilage extracellular matrix in vivo ([12–14]. It is more or less generally agreed in the literature that with increasing degeneration there was an increase in water content and decrease in glycosaminoglycan (GAG) content. It has been shown in literature that the collagen content and its orientation is the major factor in changes of cartilage T2 relaxation times [13,14]. Mlynarik et al. showed that the relaxation mechanisms (T1ρ and T2) in articular cartilage are highly dependent on the static magnetic fields [15]. Keenan et al. reported T1ρ relaxation time inversely correlated with glycosaminoglycan (GAG) content in cartilage regions with normal T2 relaxation time [16]. On the other hand, Menezes et al.[14] and Mlynarik et al. [17], observed focal areas of high and low T1ρ and T2 which were unexplained by GAG concentration or collagen orientation suggesting macromolecular concentration and/or molecular effects. Though in literature conflicting evidence, regarding contributing factors for the variations in T1ρ and T2, are reported, it has been agreed that these measures are sensitive to identify alterations in ECM composition and macromolecular structure and integrity. While these MRI techniques have been investigated extensively in the knee, their application to the hip has been relatively limited despite its importance, in part due to signal-to-noise ratio constraints associated with the deeper position of this joint.

Due to the thin, adhesive nature of the two articular hip cartilage layers, morphological and relaxometry analysis of this tissue has generally considered the two layers (femoral and acetabular) as a single unit [10]. However, the biomechanical loading of the hip joint and thus the degenerative changes in Osteoarthritis (OA) may vary locally, depending on the physical activity being performed, anatomy of the joint, and the composition of the cartilage [18,9], thus emphasizing the need for regional analysis of healthy and degenerated hip cartilage. We tested the hypothesis that an analysis based on local regions of the hip cartilage is more sensitive than global measures of hip cartilage morphometry and composition.

We devised a method to automatically divide the femoral and acetabular cartilage into regional segments using anatomical and image coordinates and demonstrate regional variations in healthy volunteers and subjects with femoral acetabular impingement. We specifically aimed to demonstrate that [1] MR relaxation times (T1ρ/T2) in these sub-regions are significantly different from those of the full region of interest (complete femoral and acetabular cartilage), and that [2] sub-regional analysis of T1ρ/T2 relaxation times better discriminates subjects with FAI from healthy controls.

2. Materials and methods

2.1. Subjects

Twelve healthy volunteers (Age = 29.9 ± 10.9 years, range = 22–60 years; BMI = 23.5 ± 2.8 kg/m2, range = 17.9–28.1 kg/m2) and 9 subjects who presented with symptomatic FAI based on clinical examination and plain radiographic findings [19] (Age = 36.6 ± 9.7 years, range = 23–52 years; BMI = 26.2 ± 6.9 kg/m2, range = 18.8–38.4 kg/m2) were recruited as a part of a large cohort study approved by the institutional review board of our institution. All subjects signed informed consent and Health Insurance Portability and Accountability Act (HIPAA) forms prior to data collection.

2.2. MR image acquisition

All imaging was performed with a 3-Tesla MR scanner (GE MR750, GE Healthcare, Waukesha, WI) and a cardiac coil (GE Healthcare, Waukesha, WI). Patient positioning aids were used to immobilize and support patients, the coil was positioned reproducibly lateral and anterior to the joint of interest, and there was a consistent, reproducible, and comfortable hip positioning during scanning. Participants were positioned supine, with the knees extended and the feet held together by adhesive tape for standardizing the hip and knee angles. The imaging protocol included sagittal and coronal T2-weighted fat-saturated fast spin-echo (FSE) images (TR/TE = 3678/60 msec, field of view (FOV) = 14 cm, matrix = 288×224, slice thickness = 4 mm, no gap, echo train length [ETL] = 16, band-width = 50 kHz, NEX = 4, scan time = 3.5 minutes each) for clinical grading, oblique axial T2-weighted FSE images (slice thickness = 3 mm, no gap, matrix = 288×224, FOV = 18 cm, scan time = 4 minutes) for alpha angle (α) measurement, high resolution, fat-suppressed, sagittal 3D fat-suppressed spoiled gradient echo (SPGR) images (TR/TE = 30.4/12.6 msec, slice thickness = 1.5 mm, in-plane resolution = 0.27 mm, matrix = 512×512, FOV = 14 cm, scan time = 9.20 minutes) for cartilage morphology assessment, and a combined T1ρ/T2 quantification sequence [20], allowing post-processing creation of T1ρ and T2 maps. The combined sequence is composed of two parts: magnetization preparation for either T1ρ or T2 weighting, followed by a 3D SPGR acquisition during transient signal evolution immediately after magnetization preparation. A flag is defined in the pulse sequence to switch between T1ρ and T2 preparation according to numbers of time of spin-lock (TSL) and time of echo (TE) defined (for T1ρ preparation: time of spin lock [TSL] = 0/15/30/45 msec, spin-lock frequency = 500 Hz, views per segment (VPS) = 64, time of recovery = 1.2 sec; for T2 preparation: TE = 0/10.4/20.8/41.7 msec; for both for T1ρ and T2: FOV = 14 cm, matrix = 256×128, views per segment (VPS) = 64, bandwidth = 62.5 kHz, time of recovery = 1.2 sec, slice thickness = 4 mm, no gap, in-plane resolution = 0.5 mm, scan time = 13.40 minutes). The raw data of the MR images were acquired with the matrix 256×128, and then interpolated on the k-space in order to obtain the MR images with the matrix of 256×256. To evaluate reproducibility, all images for a subset of healthy subjects (all males, mean age = 23 ± 3.6 years, N = 4) were acquired twice in the same session, and subjects were repositioned in between scans.

2.3. MR image analysis

All MR images were reviewed by an experienced musculoskeletal radiologist (TML) to evaluate focal cartilage abnormalities, labral tears, bone marrow edema pattern as well as other morphological abnormalities; based on femoral-head deformities with concurrent cartilage and labral abnormalities as well as increased alpha angles [21] and typical clinical findings subjects were categorized into the FAI group. All but one of the FAI patients had cam-type deformity. MR images were transferred to a HP workstation (Hewlett-Packard, Palo Alto, CA) for off-line quantification of MR relaxation times (T1ρ and T2). Fig. 1 shows the operational flow of data processing. Cartilage regions of interest were defined on sagittal 3D high-resolution SPGR images using an in-house software program based on a spline-based semi-automated (automated Gaussian-Laplacian filtering-based noise reduction and edge detection and manual correction) segmentation algorithm in MATLAB (Mathworks Inc, El Segundo, CA) [22]. Cartilage segmentation was defined into three regions of interest: acetabulum, femur, and the combination of the two (called bi-layered) (Fig. 2). Quality control readings of all segmentations were performed an experienced musculoskeletal radiologist (TML) before quantification. A sphere was fitted to the contours of the bone-cartilage of the femur head and the center of the sphere was taken as the fitted center of the femoral head (CFH) for further computation. Starting from CFH, the sphere was divided was divided into 4 quadrants (anterior-superior, anterior-inferior, posterior-superior, and posterior-inferior sub-regions) by vertical (Pv) and horizontal (Ph) planes, parallel to the coronal and axial imaging planes respectively, passing through CFH. Then each quadrant (e.g. anterior-superior) was further divided into three equal sub-regions (30° intervals), for a total of 12 regions, thereof only 9 sub-regions (R2–R10) contained cartilage. For acetabular cartilage analysis, only sub-regions R2 to R8 contained cartilage.

Fig. 1.

Fig. 1

Open in a new tab

Data-flow-diagram of the methodology.

Fig. 2.

Fig. 2

Open in a new tab

3D model of the segmented cartilage superimposed on the femur. A sphere approximating the femoral head was computed from the bone-cartilage interface contours.

The T1ρ and T2 maps were reconstructed by fitting the image intensity (voxel-by-voxel) to the equation below using a Levenberg-Marquardt mono-exponential fitting algorithm developed in-house: STSL = S0 e−TSL/T1ρ for T1ρ fitting, where S0 is the signal intensity when TSL = 0 ms; STE = S0 e−TE/T2 for T2 fitting, where S0 is signal intensity when preparation TE = 0. To minimize the error due to motion between the scans, T1ρ- and T2-weighted images with the shortest TSL = 0 and TE = 0 (therefore with highest SNR) were rigidly registered to high-resolution SPGR images acquired in the same examination using the VTK CISG Registration Toolkit (Kitware Inc., Clifton Park, NY). The transformation matrix was applied to the reconstructed T1ρ and T2 maps. The original splines of segmented cartilage contours from the high-resolution SPGR images were superimposed on the corresponding reconstructed T1ρ and T2 maps to define the regions of interest (ROIs) for T1ρ and T2 assessment. To reduce artifacts caused by partial volume effects with synovial fluid, regions were corrected to exclude fluid pixels on the relaxation-time map before quantification.

2.4. Statistical analysis

Reproducibility T1ρ and T2 measurements were determined by calculating the coefficients of variation (CVRMS). To assess the differences in T1ρ and T2 in the sub-divided regions compared to that of the global region, a student’s t-test was performed. Analysis of variance (ANOVA) was used to identify whether the differences in mean MR relaxation times between control group and FAI group differed between the sub-regions of the cartilage. Post hoc tests were then used to identify which sub-regions demonstrated significant differences between groups. The ability of mean values of global ROIs and different sub-regions of the cartilage to discriminate between healthy controls and subjects with FAI was assessed using the area under the curve (AUC) of the receiver operating characteristic (ROC) curve analysis. The significance of the discriminations was evaluated with a two-tailed t-test. All statistical analysis was performed using JMP 7.0 (SAS Institute, Cary, NC). The significance level was set to 5% and p-values were adjusted for the 9 sub-regions, considered for analysis, based on the Bonferroni method.

3. Results

3.1. Reproducibility

Coefficient of variation values for measuring reproducibility of relaxation time measurement ranged from 1.2% to 3.3% for T1ρ and 3.1% to 4.7% for T2. The reproducibility of T1ρ and T2 measurements in subregions of the cartilage was as follows: for T1ρ (range = 1.2–2.8%), R3 = 2.1%; R4 = 2.8%; R5 = 2.4%; R6 = 1.8%; R7 = 1.6%; R8 = 1.2%; R9 = 1.4%; R10 = 1.9% and for T2 (range = 1.5–3.1%), R3 = 2.4%; R4 = 2.6%; R5 = 3.1%; R6 = 1.9%; R7 = 1.7%; R8 = 1.5%; R9 = 1.8%; R10 = 2.1%.

3.2. Global T1ρ and T2 measurements

Representative Color-coded T1ρ and T2 maps overlaid on first-echo MR images are shown in Fig. 3. The mapping of the cartilage T1ρ and T2 maps onto the 3-D representation of the segmented femoral head are shown in Fig. 4. The T1ρ and T2 times in the bi-layered (i.e. combined femoral and acetabular cartilage), femoral, and acetabular cartilage regions of interest of the controls and FAI subjects are given in Table 1.

Fig. 3.

Fig. 3

Open in a new tab

Color-coded T1ρ and T2 relaxation time maps of hip cartilage overlaid on the first echo MR images. Sub-regions for hip joint cartilage assessment (A – Anterior; P – Posterior; S – Superior; I – Inferior; AS – Anterior-Superior; PS – Posterior-Superior; AI – Anterior-Inferior; PI – Posterior-Inferior; Ph – Horizontal plane; Pv – Vertical plane.) The bar represents the color specific T1ρ and T2 values in msec.

Fig. 4.

Fig. 4

Open in a new tab

Representative three-dimensional T1ρ and T2 maps of femoral cartilage overlaid on the femoral head. The color bar represents the color specific T1ρ and T2 values in msec.

Table 1.

Global T1ρ and T2 relaxation time measurements (in msec, mean ± SD) in healthy controls and FAI subjects.

Regions of Interest T1ρ (msec)
T2 (msec)
Controls FAI P-Value Controls FAI P-Value
Bi-Layered Cartilage 35.5 ± 2.4 38.8 ± 2.8 0.0133 31.1 ± 1.9 33.1 ± 1.7 0.0511
Femoral Cartilage 35.4 ± 2.3 39.9 ± 3.3 0.0020 31.1 ± 1.7 33.9 ± 3.1 0.0160
Acetabular Cartilage 35.4 ± 2.8 36.8 ± 3.0 0.3188 31.0 ± 2.3 31.7 ± 2.5 0.5622
Open in a new tab

3.3. Sub-regional analysis of T1ρ and T2 measurements

With respect to the first objective, this analysis determined whether MR relaxation times (T1ρ and T2) in sub-regions (Fig. 3) are significantly different from the whole ROI in healthy subjects. The detailed results of T1ρ and T2 measurements in these sub-regions and whole ROIs are shown in Fig. 5. The T1ρ relaxation times of sub-regions R9 (30.4 ± 2.5 msec, P = 0.0025) and R10 (30.5 ± 2.1 msec, P = 0.0002) in femoral cartilage were significantly lower than that of the whole cartilage ROI (35.5 ± 2.4 ms). In the acetabular layer, R7 (30.7 ± 3.7 msec, P = 0.0020) and R8 (23.9 ± 3.3 msec, P = <0.0001) had significantly lower T1ρ values than the whole ROI (35.4 ± 2.8 msec). When we investigated T2 relaxation times separately and compared them with those in the bi-layered cartilage ROIs, only subregion R9 (26.5 ± 2.7 msec, P = 0.0003) was significantly different from the whole ROI (31.1 ± 1.9 ms). In the femoral cartilage layer, sub-regions R9 (26.3 ± 2.7 msec, P = 0.004) and R10 (27.2 ± 3.0 msec, P = 0.0003) had significantly lower T2 times than the whole ROI (31.0 ± 1.7 msec). In the acetabular layer, sub-region R8 (21.2 ± 4.7 msec, P = <0.0001) had a significantly lower mean T2 value than the whole ROI (31.0 ± 2.3 msec).

Fig. 5.

Fig. 5

Open in a new tab

Relaxation times in sub-regions of the hip cartilage of controls and FAI patients (A–C) T1ρ relaxation times in bi-layered, femoral and acetabular cartilage regions, (D–F) T2 relaxation times in bi-layered, femoral and acetabular cartilage regions. Error bars represent standard deviation; (*) represents significant difference between controls and FAI subjects relaxation times.

The second objective of this study was to demonstrate that the sub-regional analysis better differentiates subjects with FAI from the healthy controls. When we analyzed the sub-regional T1ρ relaxation times using ROC analysis (Fig. 6), the sub-region R9 better accomplished this task than the whole ROI in both bi-layered (AUC = 0.981, P = 0.0001 for R9 sub-region; AUC = 0.835, P = 0.0127 for whole region) and femoral cartilage (AUC = 0.981, P = 0.0001 for R9; AUC = 0.901, P = 0.002 for whole region). Also, higher AUC values were observed in anterior-superior sub-regions R7 and R8 of all (bi-layered, femoral and acetabular) cartilages (Table 2). Sub-regional analysis of T2 relaxation times showed that the R9 sub-region better discriminated both the groups than the whole region of interest in both the bi-layered ROIs (AUC = 0.953, P = 0.0026 for R9 subregion; AUC = 0.795, P = 0.0417 for whole region) as well as in femoral cartilage (AUC = 0.976, P = 0.0005 for R9 sub-region; AUC = 0.808, P = 0.0124 for whole region).

Fig. 6.

Fig. 6

Open in a new tab

Representative ROC curve plots of T1ρ relaxation times in R7, R8, and R9 sub-regions of the femoral cartilage.

Table 2.

AUCs of the ROC analysis of mean T1ρ and T2 values to discriminate between subjects with and without FAI (boldface indicates AUC> = 0.75).

ROI Acetabular Cartilage
Bi-Layered Cartilage
Femoral Cartilage
T1ρ T2 T1ρ T2 T1ρ T2
Whole 0.692 0.6025 0.8352† 0.7948† 0.9011† 0.8076†
R2 0.5083 0.6153 0.4546 0.6346 0.600 0.6923
R3 0.633 0.4821 0.60 0.4821 0.5846 0.4732
R4 0.5167 0.5714 0.6154 0.5000 0.7231 0.6428
R5 0.6833 0.6785 0.6154 0.5536 0.6885 0.6901
R6 0.4179 0.5000 0.4615 0.5892 0.5231 0.5714
R7 0.7709† 0.5952 0.7684† 0.7857† 0.7685† 0.8392†
R8 0.7508† 0.6947 0.7538† 0.7642† 0.8000† 0.7904†
R9 0.9808‡ 0.9523† 0.9808‡ 0.9761‡
R10 0.7185 0.5476 0.7185 0.5476
Open in a new tab
†

Significant (p<0.05) subject discrimination based on ROC analysis.

‡

Significant (p<0.001) subject discrimination based on ROC analysis.

Our results also showed that sub-region R9 in both the bi-layered and femoral cartilage ROIs showed significantly different MR relaxation times (T1ρ and T2) between FAI and controls (T1ρ : 39.0 ± 4.3 msec vs. 30.4 ± 2.5 msec, P = 0.0002; T2: 32.6 ± 2.9 msec vs. 26.5 ± 2.7 msec, P = 0.0035).

4. Discussion

In this study, we developed an automatic method for sub-regional analysis of variations in MR relaxation times of hip cartilage, and demonstrated that analysis based on local regions was more sensitive than global measures of the same parameters in hip cartilage for discriminating subjects with and without FAI. Our results showed that (a) T1ρ and T2 relaxation times in the anterior-superior cartilage sub-region were significantly different from those of the global cartilage ROI, and (b) the anterior-superior cartilage region was more sensitive in discriminating subjects with FAI from those without than the global regions of interest.

The reported T1ρ (31.8–41.4 msec) and T2 (28.4–33.6 msec) values in this study are in the expected range compared with previously reported T1ρ (33.79–45.47 msec) [10] and T2 (29.4–33.4 msec) [10,23,24] relaxation times of normal hip joint cartilage. Short-term reproducibility indicates that in-vivo hip cartilage T1ρ and T2 measurements are highly reproducible (CV < 5%).

Longer cartilage relaxation times (T1ρ and T2) were observed in subjects with FAI than healthy controls. Similarly, Carballido-Gamio et al. [10] reported longer a T1ρ and T2 relaxation times in a subject with mild hip OA compared to controls. Nishii et al. [24] also observed longer T2 values in a group with early OA compared to a control group of volunteers, though his group analyzed only mid-coronal slices of the hip cartilage in subjects with hip dysplasia. Several studies have reported that, T2 is sensitive to changes in hydration, collagen concentration [25], and orientation of the highly organized anisotropic arrangement of collagen fibrils in the extracellular cartilage matrix [13,26]. It has been observed that with progression of knee OA, morphological parameters such as cartilage thickness and volume decrease [27], and that molecular and biochemical cartilage properties such as T2 relaxation time [13], T1ρ relaxation time [28] increase. T2 measurement is widely used to assess the macromolecular constitution of cartilage, but collagen concentration and orientation alters the relaxation rate. On the other hand, changes in collagen orientation may be among the first changes to occur and can be visualized by T2 mapping. Cardenas-Blanco et al. also observed elevated T1ρ relaxation times in symptomatic FAI subjects than in controls [29]. Regatte et al., found that T1ρ-weighted MR imaging is more sensitive to proteoglycan depletion than T2-weighted MR imaging [30,31]. The primary mechanisms that contribute to cartilage T1ρ are from dipolar relaxation and chemical exchange, which are affected by the changes in macromolecular content and structure. Studies have shown that the T1ρ dispersion is primarily due to residual dipolar interaction and contribution due to chemical exchange between –OH and –NH groups of proteoglycan (PG) with bulk water is only ~6% at 3 T [12,15,32]. Keenan et al. observed higher T1ρ relaxation time in cartilage regions with lower GAG content [16]. On the other hand, Menezes et al. reported that collagen concentration may be the dominant factor in determining T1ρ and T2 relaxation, but cautioned against concluding observed differences in T1ρ and T2 to concentration differences or orientation differences [14]. They have observed focal regions of high and low T1ρ and T2 which were unexplained by GAG concentration or collagen orientation suggesting macromolecular concentration and/or molecular effects. Similar results were reported by Mlynarik et al. [17] also. We also observed a larger dynamic range for T1ρ compared to T2 in both subjects with FAI and healthy controls, which agrees with earlier studies on subjects with knee OA [10], FAI [29], and healthy controls [24,27]. The understanding of the behavior of these biomarkers derived from MRI in characterizing hip cartilage and degeneration is still in the early stages. This study demonstrates that T1ρ and T2 may be a valuable biomarker in identifying subjects with pathological risks to progression to OA, and could allow one to quantifying and potentially monitor structural and biochemical changes that occur in early degeneration.

Significantly different T1ρ and T2 relaxation times in the anterior-superior regions (R7, R8, and R9) of the hip joint cartilage suggest possible regional differences in cartilage matrix composition between subjects with FAI and healthy subjects. ROC analysis showed that subregional analysis especially in anterior-superior regions in femoral and acetabular cartilage better discriminate FAI joints from the healthy joints. It has been recognized that FAI can cause premature OA if appropriate intervention is not performed early [3]. Bittersohl et al. [33] observed that in subjects with cam-type FAI, focal cartilage damage was most frequent in the superior-anterior regions (similar to R7, R8 and R9 sub-regions in this study) and the damage pattern was well localized (reflected by the T1 value distribution). Cardenas-Blanco et al. also observed significantly elevated T1ρ relaxation times in anterior (R9 and R8 sub-regions in this study) and anterior-superior regions (R7 and R8 sub-regions in this study) of the hip cartilage in symptomatic FAI subjects than in controls [29]. The collagen fibers have an ordered structure, making the water associated with them exhibit both magnetization transfer and magic angle (collagen fibers orientation to the applied static magnetic field B0) effects [13,34]. Because of its curved surface, the hip joint cartilage is susceptible to magic angle effect. Cova et al. observed regional differences in signal intensity of hip cartilage and suggested this behavior may be due to the anisotropic arrangement of collagen fibers within a specific zone [35]. These regions overestimate the mean T1ρ and T2 relaxation times, which makes the relaxation times in anterior and anterior-superior sub-regions (R9 and R10) with perpendicular orientation of collagen fibers relative to the B0 field significantly lower than the global mean. But the influence of regional difference in composition, biological activity, and joint biomechanics on T1ρ and T2 relaxation times must be taken into consideration while studying the magic angle effect. Stelzeneder et al. [9] also reported that cartilage defects in FAI hips might be of a true focal nature and may not affect the surrounding cartilage, based on relative dGEMRIC indices. Ilizaliturri et al. [36] and Sampson [37] proposed the use of regional zones for precise description of a lesion’s position and found that most cartilage lesions from direct damage due to morphologic conflicts resulting from FAI are in zones 2 and 3 (superior and anterior-superior regions) near the fovea (similar to R6–R9 sub-regions in this study). This clearly suggests that the damage due to FAI pathologic deformities appears to be region-specific and the surrounding cartilage regions are less affected (showing less or no change in biochemical composition). In contrast, Kim et al. [38] showed that subjects with hip dysplasia have a more generalized pattern of cartilage damage. Literature suggests that cartilage loss in OA and physiological loading may not be uniform throughout the cartilage plates [23], but may affect certain regions more frequently and more strongly than others, and may thus exhibit a different biochemical composition and degeneration [9,39]. In our cases with cam-type FAI, significantly elevated T1ρ and T2 relaxation times were observed in the anterior to superior regions of the cartilage that reflects the change in biochemical composition in those regions. These findings are consistent with previous studies in literature on location of hip joint cartilage lesions in FAI patients [29,33,37]. The damage in those regions may be caused from impaction of the head, demarcation zones from FAI, and subchondral cysts secondary to FAI. In this study, we successfully developed a method for measuring regional variations in cartilage composition in anatomically different sub-regions of articular hip cartilage. We also demonstrated that sub-regional analysis was more sensitive in discriminating subjects with FAI from healthy controls.

ROC analysis showed that sub-regions in femoral cartilage better discriminated FAI joints from the healthy ones. Though the AUC values in ROC analysis were similar in both ROIs (femoral cartilage and bi-layered cartilage), femoral cartilage sub-regions ROI showed better performance in differentiating the two groups. Acetabular cartilage sub-regions on the other hand show significance in only T1ρ values in differentiating the two groups. These results suggest that analyzing hip cartilage as a single ROI (i.e. a bi-layered cartilage unit) may not be as efficacious as examining separate femoral and acetabular cartilaginous ROIs in the classification of subjects with hip abnormalities. The global mean T1ρ and T2 values in bi-layered and acetabular cartilage regions were not significantly different between subjects with and without FAI, as they were in the femoral cartilage. This suggests that the analysis of individualized (femoral and acetabular) regions of interest gives better insight on region-specific degeneration and allow us to exclude joint fluid with high signal intensity between the acetabular and femoral cartilage layers during quantification [23,33].

The limitations of our study have to be considered while interpreting the findings. The number of subjects was small (12 healthy subjects and 9 FAI patients). Cartilage T1ρ and T2 relaxation times were assessed only in central regions of sagittal hip joint images, and medial and lateral areas were not evaluated because the curved joint surface that is susceptible to partial volume effect [10]. Although the magic angle effect has been discussed in the literature for T2 [34,40] and T1ρ [32,34,41] in knee cartilage or cartilage specimens, no studies, to our knowledge, have documented an orientation dependence of T2 and T1ρ in cartilage of the hip joint in vivo [35] and its effect on measures within sub-regions, which needs to be further investigated. Furthermore, delineation of the cartilage surface and defining regions of interest for quantification of femur and acetabulum separately was difficult and may be improved using a device to induce continuous leg traction during the image acquisition [24] or better pulse sequences that provide brighter signal intensity from fluid than the conventional sequences, such as multi-echo recombined gradient echo sequence [42]. The latter would also reduce the partial volume artifact on the superficial layer (voxels closer to the articular surface) of the cartilage, which would be useful for laminar analysis of MR relaxation times [18,24].

In summary, this study demonstrated (a) regional analysis of cartilage relaxation times (T1ρ and T2) in the hip joint is highly reproducible, (b) variations in regional composition at the hip cartilage using MR relaxation times (T1ρ and T2), and (c) that analysis based on local regions is more sensitive than global measures of hip cartilage composition and that cartilage degeneration in subjects with and without femoral-acetabular impingement may be region-specific.

Footnotes

☆

Institution at which the study performed: University of California San Francisco.

☆☆

Funding Source: NIH/NIAMS P50 AR060752 (SM), NIAMS U01 AR055079 (SM).

References

  • 1.Clohisy JC, Beaule PE, O’Malley A, Safran MR, Schoenecker P. AOA symposium. Hip disease in the young adult: current concepts of etiology and surgical treatment. J Bone Joint Surg Am. 2008;90(10):2267–81. doi: 10.2106/JBJS.G.01267. [DOI] [PubMed] [Google Scholar]
  • 2.Leunig M, Beaule PE, Ganz R. The concept of femoroacetabular impingement: current status and future perspectives. Clin Orthop Relat Res. 2009;467(3):616–22. doi: 10.1007/s11999-008-0646-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.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):1012–8. doi: 10.1302/0301-620X.87B7.15203. [DOI] [PubMed] [Google Scholar]
  • 4.Ganz R, Parvizi J, Beck M, Leunig M, Notzli H, Siebenrock KA. Femoroacetabular impingement: a cause for osteoarthritis of the hip. Clin Orthop Relat Res. 2003;417:112–20. doi: 10.1097/01.blo.0000096804.78689.c2. [DOI] [PubMed] [Google Scholar]
  • 5.Jacobsen S. Adult hip dysplasia and osteoarthritis. Studies in radiology and clinical epidemiology. Acta Orthop Suppl. 2006;77(324):1–37. [PubMed] [Google Scholar]
  • 6.Lievense AM, Bierma-Zeinstra SM, Verhagen AP, Verhaar JA, Koes BW. Influence of hip dysplasia on the development of osteoarthritis of the hip. Ann Rheum Dis. 2004;63(6):621–6. doi: 10.1136/ard.2003.009860. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Anderson LA, Peters CL, Park BB, Stoddard GJ, Erickson JA, Crim JR. Acetabular cartilage delamination in femoroacetabular impingement. Risk factors and magnetic resonance imaging diagnosis. J Bone Joint Surg Am. 2009;91(2):305–13. doi: 10.2106/JBJS.G.01198. [DOI] [PubMed] [Google Scholar]
  • 8.Chegini S, Beck M, Ferguson SJ. The effects of impingement and dysplasia on stress distributions in the hip joint during sitting and walking: a finite element analysis. J Orthop Res. 2009;27(2):195–201. doi: 10.1002/jor.20747. [DOI] [PubMed] [Google Scholar]
  • 9.Stelzeneder D, Mamisch TC, Kress I, Domayer SE, Werlen S, Bixby SD, et al. Patterns of joint damage seen on MRI in early hip osteoarthritis due to structural hip deformities. Osteoarthritis Cartilage. 2012;20(7):661–9. doi: 10.1016/j.joca.2012.03.014. [DOI] [PubMed] [Google Scholar]
  • 10.Carballido-Gamio J, Link TM, Li X, Han ET, Krug R, Ries MD, et al. Feasibility and reproducibility of relaxometry, morphometric, and geometrical measurements of the hip joint with magnetic resonance imaging at 3 T. J Magn Reson Imaging. 2008;28(1):227–35. doi: 10.1002/jmri.21411. [DOI] [PubMed] [Google Scholar]
  • 11.Subburaj K, Souza RB, Stehling C, Wyman BT, Le Graverand-Gastineau MP, Link TM, et al. Association of MR relaxation and cartilage deformation in knee osteoarthritis. J Orthop Res. 2012;30(6):919–26. doi: 10.1002/jor.22031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Duvvuri U, Goldberg AD, Kranz JK, Hoang L, Reddy R, Wehrli FW, et al. Water magnetic relaxation dispersion in biological systems: the contribution of proton exchange and implications for the noninvasive detection of cartilage degradation. Proc Natl Acad Sci U S A. 2001;98(22):12479–84. doi: 10.1073/pnas.221471898. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Mosher TJ, Dardzinski BJ. Cartilage MRI T2 relaxation time mapping: overview and applications. Semin Musculoskelet Radiol. 2004;8(4):355–68. doi: 10.1055/s-2004-861764. [DOI] [PubMed] [Google Scholar]
  • 14.Menezes NM, Gray ML, Hartke JR, Burstein D. T2 and T1rho MRI in articular cartilage systems. Magn Reson Med. 2004;51(3):503–9. doi: 10.1002/mrm.10710. [DOI] [PubMed] [Google Scholar]
  • 15.Mlynarik V, Szomolanyi P, Toffanin R, Vittur F, Trattnig S. Transverse relaxation mechanisms in articular cartilage. J Magn Reson. 2004;169(2):300–7. doi: 10.1016/j.jmr.2004.05.003. [DOI] [PubMed] [Google Scholar]
  • 16.Keenan KE, Besier TF, Pauly JM, Han E, Rosenberg J, Smith RL, et al. Prediction of glycosaminoglycan content in human cartilage by age, T1rho and T2 MRI. Osteoarthritis Cartilage. 2011;19(2):171–9. doi: 10.1016/j.joca.2010.11.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Mlynarik V, Trattnig S, Huber M, Zembsch A, Imhof H. The role of relaxation times in monitoring proteoglycan depletion in articular cartilage. J Magn Reson Imaging. 1999;10(4):497–502. doi: 10.1002/(sici)1522-2586(199910)10:4<497::aid-jmri1>3.0.co;2-t. [DOI] [PubMed] [Google Scholar]
  • 18.Watanabe A, Boesch C, Siebenrock K, Obata T, Anderson SE. T2 mapping of hip articular cartilage in healthy volunteers at 3 T: a study of topographic variation. J Magn Reson Imaging. 2007;26(1):165–71. doi: 10.1002/jmri.21014. [DOI] [PubMed] [Google Scholar]
  • 19.Tannast M, Siebenrock KA, Anderson SE. Femoroacetabular impingement: radiographic diagnosis – what the radiologist should know. AJR Am J Roentgenol. 2007;188(6):1540–52. doi: 10.2214/AJR.06.0921. [DOI] [PubMed] [Google Scholar]
  • 20.Li JSX, Liang F, Vishnudas KS, Chen W, Banerjee S, Majumdar S. Simultaneous acquisition of T1rho and T2 quantification in cartilage – reproducibility and diurnal variation. Proc Intl Soc Mag Reson Med. 2011:19. [Google Scholar]
  • 21.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):778–85. doi: 10.1148/radiol.2403050767. [DOI] [PubMed] [Google Scholar]
  • 22.Carballido-Gamio J, Bauer J, Lee KY, Krause S, Majumdar S. Combined image processing techniques for characterization of MRI cartilage of the knee. Conf Proc IEEE Eng Med Biol Soc. 2005;3:3043–6. doi: 10.1109/IEMBS.2005.1617116. [DOI] [PubMed] [Google Scholar]
  • 23.Nishii T, Shiomi T, Tanaka H, Yamazaki Y, Murase K, Sugano N. Loaded cartilage T2 mapping in patients with hip dysplasia. Radiology. 2010;256(3):955–65. doi: 10.1148/radiol.10091928. [DOI] [PubMed] [Google Scholar]
  • 24.Nishii T, Tanaka H, Sugano N, Sakai T, Hananouchi T, Yoshikawa H. Evaluation of cartilage matrix disorders by T2 relaxation time in patients with hip dysplasia. Osteoarthritis Cartilage. 2008;16(2):227–33. doi: 10.1016/j.joca.2007.06.003. [DOI] [PubMed] [Google Scholar]
  • 25.Liess C, Lusse S, Karger N, Heller M, Gluer CC. Detection of changes in cartilage water content using MRI T2-mapping in vivo. Osteoarthritis Cartilage. 2002;10(12):907–13. doi: 10.1053/joca.2002.0847. [DOI] [PubMed] [Google Scholar]
  • 26.Burstein D, Gray M, Mosher T, Dardzinski B. Measures of molecular composition and structure in osteoarthritis. Radiol Clin North Am. 2009;47(4):675–86. doi: 10.1016/j.rcl.2009.04.003. [DOI] [PubMed] [Google Scholar]
  • 27.Li X, Benjamin Ma C, Link TM, Castillo DD, Blumenkrantz G, Lozano J, et al. In vivo T(1rho) and T(2) mapping of articular cartilage in osteoarthritis of the knee using 3 T MRI. Osteoarthritis Cartilage. 2007;15(7):789–97. doi: 10.1016/j.joca.2007.01.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Li X, Pai A, Blumenkrantz G, Carballido-Gamio J, Link T, Ma B, et al. Spatial distribution and relationship of T1rho and T2 relaxation times in knee cartilage with osteoarthritis. Magn Reson Med. 2009;61(6):1310–8. doi: 10.1002/mrm.21877. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Arturo Cardenas-Blanco KR, Andrew Speirs, Ian Cameron, Mark Schweitzer, Paul Beaule. Noncontrast cartilage assessment (T1ρ) of the hip in femoroacetabular impingment: Can we predict early changes? Proc Intl Soc Mag Reson Med. 2012;20:3307. [Google Scholar]
  • 30.Regatte RR, Akella SV, Borthakur A, Kneeland JB, Reddy R. Proteoglycan depletion-induced changes in transverse relaxation maps of cartilage: comparison of T2 and T1rho. Acad Radiol. 2002;9(12):1388–94. doi: 10.1016/s1076-6332(03)80666-9. [DOI] [PubMed] [Google Scholar]
  • 31.Regatte RR, Akella SV, Lonner JH, Kneeland JB, Reddy R. T1rho relaxation mapping in human osteoarthritis (OA) cartilage: comparison of T1rho with T2. J Magn Reson Imaging. 2006;23(4):547–53. doi: 10.1002/jmri.20536. [DOI] [PubMed] [Google Scholar]
  • 32.Akella SV, Regatte RR, Wheaton AJ, Borthakur A, Reddy R. Reduction of residual dipolar interaction in cartilage by spin-lock technique. Magn Reson Med. 2004;52(5):1103–9. doi: 10.1002/mrm.20241. [DOI] [PubMed] [Google Scholar]
  • 33.Bittersohl B, Steppacher S, Haamberg T, Kim YJ, Werlen S, Beck M, et al. Cartilage damage in femoroacetabular impingement (FAI): preliminary results on comparison of standard diagnostic vs delayed gadolinium-enhanced magnetic resonance imaging of cartilage (dGEMRIC) Osteoarthritis Cartilage. 2009;17(10):1297–306. doi: 10.1016/j.joca.2009.04.016. [DOI] [PubMed] [Google Scholar]
  • 34.Mosher TJ, Smith H, Dardzinski BJ, Schmithorst VJ, Smith MB. MR imaging and T2 mapping of femoral cartilage: in vivo determination of the magic angle effect. AJR Am J Roentgenol. 2001;177(3):665–9. doi: 10.2214/ajr.177.3.1770665. [DOI] [PubMed] [Google Scholar]
  • 35.Cova M, Toffanin R, Frezza F, Pozzi-Mucelli M, Mlynarik V, Pozzi-Mucelli RS, et al. Magnetic resonance imaging of articular cartilage: ex vivo study on normal cartilage correlated with magnetic resonance microscopy. Eur Radiol. 1998;8(7):1130–6. doi: 10.1007/s003300050520. [DOI] [PubMed] [Google Scholar]
  • 36.Ilizaliturri VM, Jr, Byrd JW, Sampson TG, Guanche CA, Philippon MJ, Kelly BT, et al. A geographic zone method to describe intra-articular pathology in hip arthroscopy: cadaveric study and preliminary report. Arthroscopy. 2008;24(5):534–9. doi: 10.1016/j.arthro.2007.11.019. [DOI] [PubMed] [Google Scholar]
  • 37.Sampson TG. Arthroscopic treatment for chondral lesions of the hip. Clin Sports Med. 2011;30(2):331–48. doi: 10.1016/j.csm.2010.12.012. [DOI] [PubMed] [Google Scholar]
  • 38.Kim YJ, Jaramillo D, Millis MB, Gray ML, Burstein D. Assessment of early osteoarthritis in hip dysplasia with delayed gadolinium-enhanced magnetic resonance imaging of cartilage. J Bone Joint Surg Am. 2003;85-A(10):1987–92. doi: 10.2106/00004623-200310000-00019. [DOI] [PubMed] [Google Scholar]
  • 39.Wirth W, Eckstein F. A technique for regional analysis of femorotibial cartilage thickness based on quantitative magnetic resonance imaging. IEEE Trans Med Imaging. 2008;27(6):737–44. doi: 10.1109/TMI.2007.907323. [DOI] [PubMed] [Google Scholar]
  • 40.Xia Y, Zheng S. Reversed laminar appearance of articular cartilage by T1-weighting in 3D fat-suppressed spoiled gradient recalled echo (SPGR) imaging. J Magn Reson Imaging. 2010;32(3):733–7. doi: 10.1002/jmri.22278. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Li X, Cheng J, Lin K, Saadat E, Bolbos RI, Jobke B, et al. Quantitative MRI using T1rho and T2 in human osteoarthritic cartilage specimens: correlation with biochemical measurements and histology. Magn Reson Imaging. 2011;29(3):324–34. doi: 10.1016/j.mri.2010.09.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.White ML, Zhang Y, Healey K. Cervical spinal cord multiple sclerosis: evaluation with 2D multi-echo recombined gradient echo MR imaging. J Spinal Cord Med. 2011;34(1):93–8. doi: 10.1179/107902610X12911165975025. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES