Comparison of T2 Values in the Lateral and Medial Portions of the Weight-Bearing Cartilage of the Hip for Patients With Symptomatic Femoroacetabular Impingement and Asymptomatic Volunteers Fernando P. Ferro, M.D., Charles P. Ho, M.D., Ph.D., Grant J. Dornan, M.Sc., Rachel K. Surowiec, M.Sc., and Marc J. Philippon, M.D.

Purpose: To develop a simplified method to define a clinically relevant subregion in the course of arthroscopic treatment of femoroacetabular impingement (FAI) using T2 mapping in patients and asymptomatic volunteers. Additionally, we sought to compare the lateral and medial subregion values in asymptomatic volunteers and in patients presenting with FAI. Finally, we wanted to investigate possible associations between patients’ T2 mapping values and demographic variablesdi.e., alpha angle, age, sex, and body mass index (BMI). Methods: Twenty-five asymptomatic volunteers and 23 consecutive symptomatic patients with FAI (cam or mixed type) were prospectively enrolled and evaluated with a sagittal T2 mapping sequence. The weight-bearing region of the acetabular and femoral cartilage was manually segmented and divided into medial and lateral subregions. Median T2 values were determined, and patient characteristics were assessed as potential predictors of T2 values. Results: T2 values in the lateral portion of the acetabulum were lower than in the medial portion for both asymptomatic volunteers (43 v 53 ms; P < .001) and patients with FAI (42 v 49 ms; P ¼ .016). The medial acetabulum (MA) of asymptomatic volunteers had higher T2 values than those of the FAI group (53 v 49 ms; P ¼ .040). The lateralminusemedial difference was significantly larger among asymptomatic volunteers than in patients with FAI (P ¼ .047). Patients with FAI had higher alpha angles than those of the asymptomatic volunteers, but no other associations with patient characteristics were observed. Conclusions: This study’s findings suggest that there are differences in cartilage T2 mapping values between medial and lateral weight-bearing aspects of the hip and may expand the application and usefulness of biochemical magnetic resonance imaging (MRI) techniques, specifically T2 mapping, in the diagnosis of hip cartilage damage with the evaluation of clinically relevant subregions. When comparing asymptomatic volunteers and patients with FAI presenting with cam or mixed type deformity, we observed a significant contrast between the T2 mapping values of the lateral and medial portions of the weight-bearing zone of the acetabular cartilage, whereas such contrast was not observed when zone 3 was analyzed as a whole. Level of Evidence: Level III, development of diagnostic criteria on the basis of consecutive patients with a universally applied reference gold standard.

H

ip osteoarthritis (OA) has been an extremely prevalent problem known for several decades.1 As life expectancy increases worldwide, prevention and early diagnosis will become even more important to

From Steadman Philippon Research Institute (F.P.F, C.P.H., G.J.D., R.K.S.); and The Steadman Clinic (M.J.P.), Vail, Colorado, U.S.A. The authors report the following potential conflict of interest or source of funding: M.J.P. receives support from the Steadman Philippon Research Institute, Smith & Nephew, MIS, Ossur, Siemens, Vail Valley Medical Center, Arthrosurface, DonJoy, Slack, Elsevier, Linvatec, and HIPCO. Received May 8, 2014; accepted February 26, 2015. Address correspondence to Marc J. Philippon, M.D., The Steadman Clinic, Steadman Philippon Research Institute, 181 W Meadow Dr, Ste 1000, Vail, CO 81657, U.S.A. E-mail: [email protected] Ó 2015 by the Arthroscopy Association of North America 0749-8063/14398/$36.00 http://dx.doi.org/10.1016/j.arthro.2015.02.045

reduce the burden of this disease on society.2 In early stages, OA is rather difficult to diagnose through physical examination and plain radiographs alone. Early cartilage changes are thus better identified with magnetic resonance imaging (MRI), which allows noninvasive visualization of soft tissues such as cartilage within the joint.3 Conventional MRI has reasonable accuracy in identifying full-thickness cartilage defects but is limited in diagnostic sensitivity to early changes in cartilage when such changes are still confined to narrow portions of the chondral surface.4 Quantitative (biochemical) MRI techniques have been developed to improve on this lack of diagnostic sensitivity, and recent studies are promising.3,5 Techniques such as delayed gadoliniumenhanced MRI of cartilage (dGEMRIC), T1 relaxation

Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 31, No 8 (August), 2015: pp 1497-1506

1497

1498

F. P. FERRO ET AL.

in the rotating frame (T1rho), sodium MRI, and T2 mapping have different characteristics but have in common the ability to detect early pathologic conditions of the cartilage compared with conventional MRI.4 Among these techniques, T2 mapping is advantageous for several reasons. The technique does not require the use of intravenous gadolinium contrast agent and has a relatively quick acquisition time when compared with the aforementioned techniques, while having widespread availability. T2 mapping, or transverse relaxation time, has shown sensitivity to water content and collagen anisotropy and thus can detect early biochemical changes before the onset of macroscopic structural chondral damage.5,6 After excitation by the radiofrequency MR pulse, T2 relaxation essentially measures the time it takes the protons to relax back to their original state/spin. This results in an exponential decay in which the rate of decay (or T2 value) is largely influenced by the surrounding environment and presence of free water molecules. The T2 value is a biological tissue-specific parameter measuring the ability of proton molecules to move and exchange energy within the cartilaginous matrix. In the past decade, femoroacetabular impingement (FAI) has been recognized as a common cause of hip pain in young active individuals. It has been shown that it can contribute to chondral damage and hip arthritis when left untreated.7 FAI can be classified into 3 basic types: cam, in which the femoral head-neck junction has an abnormal bony convexity; pincer, in which some form of excessive acetabular coverage exists; and mixed type, in which both abnormalities coexist.7,8 During normal hip movement, the spherical head glides smoothly inside an equally spherical acetabular cup. This permits an even distribution of loads across a large surface area. This is coupled with the fact that the labrum creates an effective fluid seal phenomenon that ensures a thin layer of synovial fluid between the head and the acetabulum at all times.9 The result is an effective load distribution across the medial and lateral portions of the hip joint despite its impressive range of motion, which also helps to explain why the hip cartilage is significantly thinner than, e.g., the knee cartilage.10 FAI is a condition of decreased hip congruency and abnormal contact that may arise as a result of abnormal morphologic characteristics in the proximal femur or acetabular rim, or both.11 The most common (cam) impingement is caused by jamming of a protuberant femoral head-neck transition into the acetabular rim during forceful motion, especially flexion/adduction/internal rotation. The resulting shear forces produce outside-in abrasion of the acetabular cartilage or its avulsion from the labrum and the subchondral bone in the anterolateral rim area (or both). In patients with cam FAI, it has been observed that chondral damage

commonly begins at the lateral portion of the weight-bearing area of the hip joint.7 This corresponds predominantly with the lateral portion of zone 3 (middle-superior) which is a main weight-bearing zone according to the geographic zone method as described by Ilizaliturri et al.12 Biochemical MRI studies have been able to identify such damage using techniques such as T2 mapping and dGEMRIC.13-16 To identify chondral damage, it is necessary to manually define regions of interest (ROI), which are drawn, including regions in which most damage typically occurs. Such studies highlight the potential of T2 mapping to provide an earlier diagnosis of chondral pathologic characteristics. However, these studies did not propose a technique that can be standardized and repeated reliably in the clinical setting. To allow for longitudinal follow-up of patients and comparison across patients and centers, the joint surface should be scanned in a predefined automatized manner, without the need for manually drawn ROIs. This study aimed to develop a simplified method to define a clinically relevant subregion in the course of arthroscopic treatment of FAI using T2 mapping in patients and asymptomatic volunteers. Additionally, we sought to compare the lateral and medial subregion values in asymptomatic volunteers and patients presenting with FAI. Finally, we sought to investigate possible associations between participants’ T2 mapping values and demographic variablesdi.e., alpha angle, age, sex, and body mass index (BMI). We hypothesized that the T2 mapping values would be different for the medial and lateral subregions of zone 3 acetabular and femoral cartilage using the proposed simplified method to define clinically subregions because of the inherent biomechanical loading patterns of the hip. Additionally, we hypothesized that the lateral subregion would have higher T2 mapping values than the medial subregion and that this relationship would be present in both asymptomatic volunteers and patients with FAI but would be more pronounced in patients with FAI. Finally, we hypothesized that T2 mapping values would be associated with demographic values, with stronger associations among patients with FAI.

Methods This study was approved by our institutional review board, and all asymptomatic volunteers and patients with FAI provided informed consent. Twenty-five asymptomatic volunteers were prospectively enrolled. Volunteers were deemed asymptomatic by a self-administered subjective scoring form (pain/swelling/stiffness visual analogue score, Tegner score, and a modified hip score), an objective clinical examination performed by a sports medicine orthopaedic surgeon (M.J.P.), and morphologic MRI examination by a musculoskeletal radiologist (C.P.H.). Clinical examination included evaluation of

1499

T2 VALUES OF WEIGHT BEARING HIP CARTILAGE

limb alignment, pain/tenderness, and the following tests: FABER test, hip dial test, anterior impingement sign, and evaluation of range of motion in the supine (flexion, abduction, and adduction) and prone (internal and external) position. Exclusion criteria included symptoms (e.g., pain, stiffness, and swelling) in the hip or knee, or both, of the imaged side; previous injury or surgery in the hip and knee; history of inflammatory arthritis or infection in the joint of interest; and evidence of cartilage lesions from a conventional morphologic MRI examination using the Hip Osteoarthritis MRI Scoring System score. A second group of 23 symptomatic patients with FAI was prospectively enrolled, consecutively, from the clinical practice of one hip arthroscopy specialist (M.J.P.). For this group, patients were enrolled consecutively to avoid biases. Therefore, inclusion criteria followed the clinical surgical indications. FAI is a clinical diagnosis based on symptoms, clinical examination, and imaging. The senior orthopaedic surgeon (M.J.P.) diagnosed these patients with FAI. Inclusion criteria included groin pain associated with athletic activity and positive physical examination findings, including positive anterior impingement sign and positive flexion, abduction, and external rotation (FABER test). Cam impingement was defined by an alpha angle greater than 55 , and pincer impingement was defined by the presence of the crossover sign or center edge angle greater than 40 , coxa profunda, or coxa protrusio. MRI criteria included the confirmation of cam or pincer deformities, or both, coupled with focal labral tearing. Exclusion criteria included age younger than 18 years, diagnosis of hip instability, previous hip surgery, previous hip trauma such as fracture or dislocation, and radiographic signs of severe or progressive OA. Patients with less than 2 mm of joint space associated with age beyond 50 years were not considered for surgery.

Additionally, patients were excluded from this study if they had undergone previous hip surgery. All patients received a preoperative standard clinical hip MRI, which included a T2 mapping sequence, which was acquired after all clinical sequences. Preoperative MRI scans were acquired at our institution within 48 hours before surgery in all cases. Image Acquisition and Morphologic Analysis Magnetic resonance imaging of the hip joint was performed at 3.0 T (MAGNETOM Verio, Siemens Medical Solutions, Erlangen, Germany) using a 4-channel flex coil (Siemens Medical Solutions) with the patient supine. The time delay between the patient lying on the scanner table for the MRI session and the acquisition of the first sequences was limited to less than 5 minutes in all volunteers and patients. The scanning protocol consisted of (1) a 3-dimensional (3D) fat-suppressed sampling perfection with application optimized contrasts using different flip angle evolution (FS SPACE) scan; (2) a multiecho spinecho T2 mapping scan in the sagittal plane; (3) a T2 weighted turbo-spin echo sequence in the axial plane; (4) a proton density turbo-spin echo scan in the coronal plane; and (5) a limited axial T1 scan distally at the level of the knee and femoral condyles for evaluation of femoral version. The FS SPACE scan was reformatted in all 3 planes, including oblique axial images along the femoral neck/head axis for evaluation of the alpha angle. Detailed scan parameters are depicted in Table 1. The T2 mapping sequence was performed at the end of the examination after all morphologic scans, approximately 17 minutes after the patient entered the scanner to allow for unloading of the cartilage within a clinical scan time slot.5 The alpha angle was determined by one orthopaedic surgeon (M.J.P.) in both the asymptomatic volunteers

Table 1. Parameters of the Imaging Sequences Used in the Study Scans for Patients and Asymptomatic Volunteers Sequence Repetition time, ms Echo time, ms Field of view, mm Matrix Voxel size, mm Slice thickness, mm Distance factor, % No. of slices Echo trains/slice Turbo factor Examination time, hr:min

T2 Mapping Sagittal 2,080 18.0-90.0 200 256  256 0.8  0.8  2.0 2 100% 20 e e 6:45

PD-TSE SPACE (Reformatted in All 3 Planes) 1,500 44 192 256  256 0.8  0.8  0.9 0.9 e 96 e 84 8:00

T2w-TSE Axial 3,990 91 165 256  192 0.9  0.6  3.0 3 10% 38 6 20 1:45

T2w-PD TSE Coronal 3,130 30 175 320  256 0.9  0.7  3.0 3 10% 30 24 8 2:35

T1-TSE Axial 700 33 280 256 192 1.6  1.1  5.0 5 10% 15 9 20 0:37

MR parameters for quantitative and morphologic imaging. Ax, axial, Cor, coronal; PD, proton density; T2w, T2 weighted; TSE, turbo spin echo; SPACE, single slab 3-dimensional TSE sequence (sampling perfection with application optimized contrasts using different flip angle evolution).

1500

F. P. FERRO ET AL.

and patients for a more thorough evaluation of both groups. The alpha angles for each volunteer and patient were obtained from an oblique axial reformation of the FS SPACE scan using the measurement technique described by Nötzli et al. (Fig 1).17 Using this technique, the angle is smaller if the femoral head is spherical and has a slim head-neck junction, yet larger if a decreased head-neck offset and a cam deformity are present. Image Analysis Acetabular and femoral cartilages were manually segmented using Mimics software (Materialise, Plymouth, MI) directly on the second echo of the T2 mapping sequence on a slice-by-slice basis by one rater (orthopaedic surgeon). To facilitate the exclusion of areas of synovial fluid and chemical shift artifact, the rater simultaneously examined the corresponding sagittal sampling perfection with application-optimized contrasts using a different flip angle evolution sequence on a neighboring monitor. Segmentations were evaluated by the senior radiologist (C.P.H.), who has 27 years of experience in musculoskeletal imaging, as a quality control measure. To simplify the method in a reproducible manner, we modified Ilizaliturri et al.’s original geographic zone method to divide both the femoral and acetabular articular surfaces into 6 zones (Fig 2).12 In the sagittal plane of the center of the femoral head, a horizontal line was drawn on its equator. The equator line was divided into 3 equal segments by 2 vertical parallel planes (Fig 3). These planes divided the femoral and acetabular articular surfaces into 3 regions. The authors chose to manually segment only the central region, which corresponds to zone 3 (middle-superior)

Fig 1. Magnetic resonance image depicting the alpha angle measurement. Point A is the anterior point at which the distance from the center of the head (hc) exceeds the radius (r) of the subchondral surface of the femoral head. The alpha angle is measured as the angle between A-hc and hc-nc, where nc is the center of the neck at the narrowest point.

Fig 2. Sagittal view of the left hip. The horizontal red line crosses the center of the femoral head. The vertical green lines are the vertical planes dividing the femoral head into 3 zones. The combination of these 3 lines divides the femoral head into 6 zones, as described in the geographic zone method.

in Ilizaliturri et al.’s method, because zones 2 (anterosuperior) and 4 (posterosuperior) have a more variable chondral layer that is often too thin when viewed from sagittal slices.12 Zones 1 (anteroinferior) and 5 (posteroinferior) are even thinner and noneweight bearing and therefore were not evaluated. Furthermore, chondral damage resulting from FAI typically occurs between the superolateral and anterolateral portions of the

Fig 3. Three-dimensional representation of the femoral head and acetabulum after the 2 vertical planes are drawn. These planes were used in all image slices to aid in precise definition of zone 3 as the femoral and acetabular cartilages were manually segmented.

1501

T2 VALUES OF WEIGHT BEARING HIP CARTILAGE

acetabulum, around the 1:00 o’ clock position, which corresponds predominantly to zone 3 (Fig 4).7 To divide zone 3 into 2 equal parts (medial and lateral portions within the joint), the coronal slice at the center of the femoral head was used. The highest (most proximal) point of the femoral head was marked and served as a border between the medial and lateral regions (Fig 5). The medial and lateral regions were segmented with different masks, allowing the data to be analyzed separately (Fig 6). T2 values were calculated using a Siemens WIP (work in progress algorithm), modified from the Siemens MapIt software algorithm (Siemens Medical Solutions). The cartilage segmentations (medial and lateral segmentations of the femur and acetabulum) were exported from Mimics as binary image files and imported into a custom Matlab (Mathworks, Natick, MA) program. The program overlaid the T2 segmentations onto the anatomic sequence (Fig 7), and T2 parameters (mean, standard deviation, median, minimum, maximum, and number of pixels) were populated and exported to an Excel (Microsoft, Redmond WA) document. Only T2 values between 10 ms and 110 ms were included in the analysis to exclude outliers such as subchondral bone and synovial fluid (T2 values > 110 ms) inadvertently included in the cartilage segmentation margins, as well as T2 values rejected because of poor fit (T2 values of 0 ms). Statistical Analysis Median T2 values and standard deviation of T2 values for each participant and each of the 4 segmented subregionsdlateral acetabulum (LA), medial acetabulum (MA), lateral femur (LF) and medial femur (MF)dwere compiled for further analysis. The median T2 value was chosen over the mean T2 value because it is less susceptible to the skewing effect of outliers from the segmentation margins and partial volume artifact. Age, sex, BMI, group (asymptomatic or FAI), and alpha angle were investigated as potential predictors of

Fig 4. Representation of the geographic zone method for the right acetabulum and femur. Zone 2 is anterior-superior. Zone 1 is anterior-inferior. The red zone marks the lateral portion of zone 3, and yellow marks the medial portion of zone 3.

T2 relaxation time. Relationships among possible predictors were investigated before testing for association with T2 value patterns. Nonparametric statistical testing was performed including Mann-Whitney U tests, Wilcoxon signed-rank test, Fisher exact test, and Spearman correlation (rho). P < .05 were deemed significant, and all statistical analyses were performed using IBM SPSS Statistics, version 20 (SPSS, Chicago, IL).

Results Participant Characteristics in Asymptomatic and FAI Groups Age, sex, BMI, and alpha angle information was stratified by participant group and is presented in Table 2. All patients presented with cam deformity or mixed-type deformity (cam plus pincer). No significant associations were found between group and age, sex, or BMI, but patients with FAI exhibited significantly higher alpha angles than those of asymptomatic volunteers. Men exhibited significantly higher BMI (median, 25.6; range, 20.9 to 45.2) than did women (median, 21.6; range, 19 to 30.6; Mann-Whitney U test, P < .001). T2 Values: Quantifying and Comparing T2 Values between Subregions We used the median T2 values of each of the 4 segmented subregions: LA, MA, LF, and MF. A separate analysis was made regarding the difference between a given participant’s lateral and medial subregions (i.e., LA minus MA; LF minus MF). The median and range of T2 values of all subregions are summarized in Table 3. For comparison, median and range of T2 values of the entire weight-bearing region (zone 3, which includes both the medial and lateral subregions) are presented. The most relevant findings are outlined. T2 values in the lateral portion of the acetabulum were lower than in the medial portion, and this was true for both asymptomatic participants (43 ms v 53 ms; P < .001) and those with FAI (42 ms v 49 ms; P ¼ .016).

1502

F. P. FERRO ET AL.

comparing the all of zone 3 for asymptomatic volunteers and patients with FAI. The median T2 value for all of zone 3 was very similar in both the acetabular (46 ms v 43 ms; P ¼ .469) and femoral sides (54 ms v 51 ms; P ¼ .515). We did not observe significant distinctions in T2 values of the femoral cartilage based on participant group or lateral-medial comparisons.

Fig 5. Using a coronal slice, the highest (most proximal) point of the femoral head was determined. This point divided the femoral head into 2 subregions: medial and lateral. The same procedure was done for the acetabular cartilage. (H, highest [most proximal] point of the femoral head.)

Likewise, the standard deviation of T2 values was significantly lower in the lateral portion of the acetabulum for both asymptomatic participants (14.3 ms v 20.3 ms; P ¼ .003) and those with FAI (16.8 ms v 21.3 ms; P < .001). In the medial acetabulum, asymptomatic volunteers had significantly higher T2 values than did patients with FAI (53 ms v 49 ms; P ¼ .040). Furthermore, we observed that the lateral-minusemedial difference was significantly larger among asymptomatic volunteers than in patients with FAI (P ¼ .047). When a lateral versus medial distinction was not made, we did not observe such a contrast when

T2 Values, Age, and BMI Given the tertiary evaluation of participant characteristics in the present study, analysis of T2 patterns proceeded with age and BMI as a relevant and independent set of potential predictors. We looked for associations between the T2 values of the different segmented subregions and age and BMI. Median T2 values in the MA were significantly correlated with age (rho, 0.305; P ¼ .037) and BMI (rho, 0.384; P ¼ .007). Similarly, standard deviation of T2 values in the MA were significantly correlated with both age (rho, e0.349; P ¼ .016) and BMI (rho ¼ .352; P ¼ .014). BMI was significantly correlated with standard deviation of T2 values in the lateral (rho, 0.355; P ¼ .013) and medial (rho , 0.492; P < .001) femur. Age was significantly correlated with lateral-minusemedial difference of T2 values in the acetabulum (rho, 0.389; P ¼ .007), meaning that the LAMA difference tends to increase with age.

Discussion A significant difference in T2 values between patients with FAI and asymptomatic volunteers was observed when comparing the lateral to the medial portions of the weight-bearing acetabular cartilage. In this study, we propose a novel and simplified method to divide the weight-bearing region of the hip joint into 2 parts: lateral and medial. We hypothesized that T2 mapping analysis of the hip cartilage as a whole would not be

Fig 6. Three-dimensional representation of the masks from one asymptomatic volunteer obtained after manual segmentation of the femoral and acetabular cartilages. Red is the lateral acetabulum. Yellow is the medial acetabulum. Blue is the lateral femur. Green is the medial femur.

T2 VALUES OF WEIGHT BEARING HIP CARTILAGE

Fig 7. T2 overlay example of medial femoral and acetabular cartilage segmentation. (A) T2 anatomic scan with T2 map overlay of segmentation. (B) Three-dimensional (3D) acetabular T2 segmentation created from manual segmentation. (C) 3D femoral T2 segmentation created from manual segmentation.

able to identify initial and localized cartilage damage in patients undergoing surgery for FAI and asymptomatic volunteers, because the increased T2 reading of the lateral damaged areas would be “washed out” by the normal cartilage of the rest of the joint. We observed a significant difference between patients with FAI and asymptomatic controls. When comparing the lateral to the medial portions of the weight-bearing acetabular cartilage and when the lateral versus medial distinction was not made and the entire regions were compared, we were not able to statistically observe this difference, corroborating our hypothesis. Previous research has found that the “baseline” values of the medial and lateral cartilages of the acetabulum are not the same,6,15 and our results corroborated this. There is indeed a topographic variation that has been observed using T2, T2*, and dGEMRIC, and the reason for this is not completely understood. The explanation is probably related to the fact that the hip joint is a weight-bearing joint that should concentrate loads in the most superior portions of the joint. In addition, range of motion is larger for flexion and abduction than for extension and adduction, which probably contributes to different loading patterns in both the sagittal and coronal planes. This fact must be taken into account before interpreting isolated T2 mapping values in any given region of the hip. Therefore, interpretation of isolated absolute T2 values does not seem to be the answer. The novelty of our study is an attempt to interpret T2 values with an internal comparison (i.e., comparing the lateral and medial values from the same individual), and we believe that our findings shed some light on a research pathway that could yield a clinical application in the future. The median T2 values of the LA were similar between the 2 groups (43 ms v 42 ms for asymptomatic

1503

volunteers and patients with FAI, respectively), but there was a larger difference for the MA (53 ms v 49 ms for asymptomatic volunteers and patients with FAI, respectively). Considering this, we decided to analyze not only the T2 values of the entire zone but also the difference between lateral and medial T2 values in the same patients and volunteers. The main finding of our study was that there is a significant difference between patients with FAI and asymptomatic volunteers when comparing the lateral to the medial portions of the weight-bearing acetabular cartilage. The median difference between the lateral and medial T2 values in asymptomatic volunteers was 8 ms, whereas this median was 3 ms in patients with FAI, which was statistically significant. When a lateral versus medial distinction was not made, we did not observe such a contrast when comparing all of zone 3 in asymptomatic participants and those with FAI. The median T2 value for all of zone 3 was very similar for both the acetabular (46 ms v 43 ms; P ¼ .469) and femoral sides (54 ms v 51 ms; P ¼ .515) comparing the asymptomatic volunteers and patients with FAI, respectively. These data suggest that there was a “washout” of values when not fully taking into account the known anatomic loading patterns of the hip and should be of consideration for future quantitative mapping studies using subregions to analyze values. In the present study, statistical differences were observed only within the acetabular cartilage. Because FAI is a condition of decreased hip congruency, abnormal contact may arise as a result of abnormal morphologic conditions in the proximal femur or acetabular rim, or both. The most common (cam) impingement is caused by jamming of a protuberant femoral head-neck transition into the acetabular rim during forceful motion, especially flexion/adduction/ internal rotation. The resulting shear forces produce outside-in abrasion of the acetabular cartilage or its avulsion from the labrum and the subchondral bone in the anterolateral rim area, or both. All patients presented with a degree of cam deformity, and therefore we hypothesize that this is why we did not observe a difference in the femoral cartilage but instead only in the cartilage of the acetabulum. Before data acquisition, we did expect to observe a significant difference when comparing the T2 values of the lateral acetabular cartilage alone. However, this was not confirmed statistically. Our data showed that a difference between the 2 groups can only be appreciated when also considering the medial acetabulum T2 values, i.e., comparing the difference between these 2 adjacent subregions. In the asymptomatic group, the MA subregion showed higher T2 values. We hypothesize that normal hip biomechanics will distribute load more evenly around the center of the hip joint, thus increasing load on the medial acetabular cartilage,

1504

F. P. FERRO ET AL.

Table 2. Participant Characteristics Stratified by Group Asymptomatic volunteers Patients with FAI P Value

Age 28 (22, 32) 28 (18, 35) .797 (Mann-Whitney U test)

Male/Female 13/12 14/9 .573 (Fisher exact test)

Alpha Angle,  49 (32, 71) 66 (44, 89) < .001 (Mann-Whitney U test)

BMI 25.5 (20.2, 45.2) 22.5 (19, 30) .076 (Mann-Whitney U test)

Data presented as median (minimum, maximum) unless otherwise indicated. BMI, body mass index; FAI, femoroacetabular impingement.

which is in line with the femoral neck axis. In patients with FAI, we observed a decrease in MA T2 values, which could be caused by an increased load transfer to the anterolateral border of the joint, which causes excessive stress in this specific subregion. Notably, a previous study found higher T2 values in the medial acetabular cartilage of asymptomatic participants.15 Interestingly, it has been widely reported that cartilage damage in FAI begins in the lateral portion of the acetabular cartilage.7,18,19 Cam morphologic features result in abnormal shear forces and abrasion of the cartilage, especially in the lateral portion of the anterosuperior rim area. Chondral delamination may occur, as may labral tearing and detachment. In earlier stages of chondral damage, the femoral weight-bearing cartilage seems to be spared because the area that conflicts with the acetabular cartilage is the head-neck transition and not the weight-bearing femoral cartilage. As the joint mechanics deteriorate further, the chondral damage tends to become more widespread, leading to OA. Pincer morphologic characteristics will typically cause primary labral crushing between the femoral neck and the excessive acetabular rim, with little primary chondral damage.4 Because the method we used to separate the lateral portion from the medial portion is easily reproducible,

we believe that this method can be used in the clinical setting for a quick assessment of differences between the medial and lateral cartilage before hip arthroscopy or surgical intervention. That could be a very useful tool to identify cartilage damage in early stages without requiring manual segmentation of suspicious areas, potentially modifying treatment strategy and yielding a more precise prognosis. Patients with FAI exhibited significantly higher alpha angles than did asymptomatic volunteers. Previous research has shown the same phenomenon, citing an average “normal” alpha angle of about 45 19,20 and stating that a cam deformity is typically present if this angle is greater than 60 .21,22 We observed a significant association between the LA-MA difference and the alpha angle, and an association trend between the alpha angle and increased T2 values in the lateral subregion of the acetabulum. This finding confirms the possible mechanism of chondral damage that was proposed by Ganz et al.,7 according to which the cam deformity would precipitate damage to the lateral portion of the acetabular cartilage alone, especially in early stages. In a previous study, it was shown that an increased alpha angle was significantly associated with macroscopic cartilage damage during hip arthroscopy.23 Patients with no chondral defects

Table 3. Median and Range of T2 Values for All Subregions Median T2 (ms) Subregion All zone 3 acetabulum LA MA Wilcoxon signed-rank test P value All zone 3 femur LF MF Wilcoxon signed-rank test P value L e M acetabulum L e M femur

Asymptomatic Participants 46 (34, 57) 43 (32, 60) 53 (38, 64.5) < .001 54 (45, 66) 54 (40, 74) 55 (27, 73) .882 8 (24.5, þ6) þ2 (30, þ30)

Patients With FAI 43 (40, 57) 42 (38, 57) 49 (36, 69.5) .016 51 (39, 77) 51 (35, 86) 51 (41, 70) .592 3 (28.5, þ10) 3.5 (30, þ31)

Standard Deviation of T2 (ms) Mann-Whitney U Test P Value .469 .606 .040

Asymptomatic Participants 17.5 (11.4, 23.5) 14.3 (8, 30.5) 20.3 (10.8, 26.3) .003

Patients With FAI 19.8 (14.4, 25.0) 16.8 (8.7, 24.6) 21.3 (17, 26) P

Comparison of T2 Values in the Lateral and Medial Portions of the Weight-Bearing Cartilage of the Hip for Patients With Symptomatic Femoroacetabular Impingement and Asymptomatic Volunteers.

To develop a simplified method to define a clinically relevant subregion in the course of arthroscopic treatment of femoroacetabular impingement (FAI)...
1MB Sizes 4 Downloads 11 Views