- Email: [email protected]
Clinical and surgical correlation of hip MR arthrographic findings in adolescents
European Journal of Radiology 85 (2016) 1192–1198
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Clinical and surgical correlation of hip MR arthrographic findings in adolescents Anugayathri Jawahar a,∗ , Aruna Vade a , Laurie Lomasney a , Gokcan Okur a , Douglas Evans b , Perla Subbaiah c a
Department of Radiology, Loyola University Medical Center, 2160, South 1st Avenue, Maywood, Illinois, 60153, United States Department of Orthopaedic Surgery and Rehabilitation, Loyola University Medical Center, 2160, South 1st Avenue, Maywood, Illinois, 60153, United States c Department of Mathematics and Statistics, Oakland University, Rochester, MI, 48309, United States b
a r t i c l e
i n f o
Article history: Received 22 August 2015 Received in revised form 27 March 2016 Accepted 31 March 2016 Keywords: MR hip arthrogram Adolescent Treatment Surgery Symptoms Follow up
a b s t r a c t Objective: The purpose of this study was to analyze the utility of MRA-H in adolescents by comparing the results of imaging with surgical findings and/or clinical outcome. Material and methods: After obtaining appropriate IRB approval, the Radiologic Information System database was queried for all patients 13–18 years of age who underwent MRA-H from 2004 through 2013. The electronic medical record was reviewed for clinical history, clinical examination findings, and operative notes. MRA-H images were reviewed for soft tissue abnormalities (labral tear, paralabral ganglion, articular cartilage loss, synovitis, ligament tears) and bony abnormalities (cam-type femoroacetabular impingement (FAI), pincer-type FAI, hip dysplasia). MRA-H findings were correlated with surgical findings and with clinical outcomes. Results: Twenty-six patients with labral tears by MRA-H were included in study and grouped as follows: Group I) patients who underwent surgical management (n = 10); group II) patients managed non-surgically (medication, intra-articular injection, physical therapy) (n = 9); group III) patients lost to follow up after being advised to have surgery (n = 7). With regard to presenting symptomatology, 87.5% of patients with labral tear had groin pain. Of those patients who were diagnosed with a labral tear, 52% were categorized as idiopathic labral tears, 26% as secondary tears (secondary to abnormal bony morphology), and 22% as traumatic labral tears. The labral tears were found to be anterior in 61% and posterior in 22%. Associated articular cartilage lesions were found in 29% of patients. In group I (surgical patients), MRA-H labral findings were confirmed at surgery in 9/10. Seventy percent of labral tears in our study had some form of abnormal bony morphology. Nine of the 12 patients with bone abnormalities were derived from group I patients. Six out of 7 patients with cam-type FAI had a labral tear. Conclusion: Labral tears diagnosed by MRA-H in the adolescent population correlated well with clinical examination and surgical findings. Also, MRA-H contributed by defining bony morphology that was directly applied to surgical management. Non-surgical management of labral tears diagnosed on MRA-H had a generally favorable outcome. © 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Hip pathologies affecting the labrum in children can be due to congenital causes such as hip dysplasia and femoral acetabular impingement, or acquired causes such as trauma, slipped capital femoral epiphysis (SCFE) and Legg-Calve-Perthes disease (LCP) [1]. Many of these hip pathologies in children can cause labral tears which if left untreated, may lead to premature degenerative arthri-
∗ Corresponding author. E-mail address: [email protected] (A. Jawahar). http://dx.doi.org/10.1016/j.ejrad.2016.03.031 0720-048X/© 2016 Elsevier Ireland Ltd. All rights reserved.
tis over time [2–4]. Magnetic resonance arthrogram of the hip (MRA-H) is used for assessment of labral tears, cartilage lesions and ligamentous injuries that occur with these hip pathologies so that appropriate treatment can be offered [5]. In adults, MRA-H has been shown to have 60%–100% sensitivity range and 44%–100% specificity range for detecting acetabular labral tears [6–8]. Previous literature has also documented a sensitivity range of 50%–79%, and specificity range of 77%–84% for detection of articular cartilage lesions in adults [6–8]. Even though MR arthrography of the hip has been used in the pediatric population in recent years, there are no published articles reviewing the utility of this minimally invasive procedure. The purpose of this study was to analyze the utility
A. Jawahar et al. / European Journal of Radiology 85 (2016) 1192–1198
1193
Fig 1. Axial T1-weighted image with fat saturation shows an anterior labral tear (arrow), in the left hip of a 14-year old male with recent history of trauma while playing baseball game. Patient also had an avulsion fracture of the anterior inferior iliac spine (not shown here).
Fig. 2. 16-Year old male with previous history of surgical treatment for slipped capital femoral epiphysis had multifocal labral tear in the left hip. Coronal T2-weighted image (A) shows an anterosuperior tear (arrow) and coronal oblique PD image (B) shows a sulcus (arrow). The labrum was found to be detached and very degenerative at surgery. Patient underwent labral debridement with repair for the tear.
of hip MR arthrography in adolescents by comparing the MRA-H findings with surgical and/or clinical outcome. 2. Material and methods This HIPPA-compliant retrospective research protocol was reviewed and approved by the Institutional Review Board (IRB number 205599). All delineated IRB requirements for provision of patient safety and privacy of patient information were observed. The Radiology Information System (RIS) was used to identify patients 18 years and younger who had an MR arthrogram of the hip between March 1, 2004 to March 31, 2013, and all were included in the study and were grouped according to their subsequent management plans and patient follow up. The hospital electronic medical records were used to document relevant clinical history and physical examination findings pertaining to the symptomatic hip (Table 1) and clinical outcome. Clinical outcome following surgical or conservative treatment was recorded as resolution, decrease or worsening of hip pain and their ability to resume their prior level of activity (Table 2). The operative
notes of the patients who underwent surgery of the symptomatic hip were reviewed by two radiologists (AJ, AV) and an orthopedic surgeon (DE) and surgical findings were correlated with the MRA-H findings. The operative notes were subjective description of location and nature of labral tear at arthroscopy or surgery. The hip arthrogram technique performed at our hospital is as follows: intra-capsular contrast material is injected under fluoroscopic guidance using an anterior inferior approach to the joint with patient supine and the knee in slight flexion. Capsular distention is achieved with 8–9 ml of dilute gadolinium ((Magnevist-gadopentetate dimeglumine, Bayer, Germany) or (Gadavist-gadobutrol, Bayer, Germany) diluted 1:250 in normal saline) with 4–5 ml of 1% lidocaine. The patient is transferred immediately to MR suite with minimal range of motion on transfer, but without placing weight on the extremity. In this study, MRI was performed in 1.5T magnets (Espree and AERA, Siemens Healthcare, Germany; Echospeed-15, GE Healthcare, Milwaukee) and 3T magnets (Trio and Verio, Siemens Healthcare, Germany). Axial, coronal, coronal oblique, radial PDweighted, and coronal T2-weighted images with fat saturation
1194
A. Jawahar et al. / European Journal of Radiology 85 (2016) 1192–1198
Table 1 Clinical history and physical examination findings pertaining to symptomatic hip. Groups
Hip pain
Groin pain
Buttock pain
Popping of hip
Clicking in hip
Pain related to activity
I II III
2 4 5
7 5 4
1 0 0
3 4 4
0 3 2
4 6 4
Total
11
16
1
11
5
14
Table 2 Clinical diagnosis, corresponding MRA-H labral findings and clinical outcome following treatment. Group
Patient
Clinical diagnosis
MRA-H findings on labrum
Severity of pain on follow up
Group I
1 2 3 4 5 6 7 8 9 10
Bilateral AVN Snapping hip Cam-FAI Multiple exostosis SCFE Cam-FAI DDH Pincer-FAI Perthes disease Perthes disease
Postero-superior tear Antero-superior tear Posterior tear Anterior tear Multifocal tear Incomplete anterior tear Antero-superior labral tear Incomplete antero-superior labral tear Anterior labral tear No labral tear
Resolved Resolved Improved Improved Worsened Resolved Worsened Improved Resolved Lost to follow up
Group II
1 2 3 4 5 6 7 8 9
Cam-FAI Suspected labral tear Cam-FAI Snapping hip Suspected labral tear SCFE Suspected labral tear Suspected labral tear Suspected labral tear
Posteroinferior labral tear Anterior labral tear Posterior labral tear Anterior labral tear Anterior labral tear Intact labrum Anterior labral tear Anterior labral tear Anterior labral tear
Improved Improved Improved Improved Improved Improved Resolved Worsened Worsened
Group III
1 2 3 4 5 6 7
Coxa profunda Cam-FAI Suspected labral tear Suspected labral tear Suspected labral tear Cam-FAI Pincer-FAI
Anterior labral tear Intact labrum Anterior labral tear Anterior labral tear Anterior labral tear Anterior labral tear Posteroinferior labral tear
Lost to follow up Lost to follow up Lost to follow up Lost to follow up Lost to follow up Lost to follow up Lost to follow up
were obtained. Body flex coil was used for imaging the hip. Slice thicknesses of 4 cm was obtained and FOV of ±16 cm was used depending on the physical habitus of the patient. The matrix used for spin echo and gradient echo sequences was 320 × 224 mm. Axial, radial, coronal and coronal oblique PD images had similar parameters including FOV 15 cm, TR 1775 ms, TE 17 ms. Coronal T2-weighted fat saturated images had FOV 15 cm, TR 3400 ms, TE 68 ms. The echo train used for T2W images was 8 ms for GE 3T scanner and 12 ms for GE 1.5T scanner. Radial PD-weighted sequence was oriented orthogonal to the acetabular rim and labrum. The MRA-H images were reviewed by 4 radiologists [AV (pediatric radiologist), AJ, GO, LL (musculoskeletal radiologist)] individually and in consensus for angle measurements and in case of variation in opinion in 2 patient findings who had recess. Kappa inter-observer variability was not calculated in our study since the focus of analysis was reaching a unanimous judgement. The different values of angle measurements were not recorded in two patients where there was varied opinion. However, for these two patients, there was agreement as to whether the angles are within the limits of normal or abnormal alpha angles. The readers were blinded to the surgical data at the time of image interpretation. The images were reviewed for labral tears, articular cartilage lesions and bone abnormalities. Synovial thickening or nodularity, ligament tears and tendon pathologies when identified were documented as well. The acetabular labrum was evaluated for presence of tear, MR signal characteristics and evidence of a perilabral recess. The alpha angle (as described by Pfirrmann [9]: on oblique axial images oriented parallel to the femoral neck, a measured angle between a line parallel to the femoral neck and a line from the
femoral head center to the anterior protruding head/neck cortex from an assumed spherical head; normal value-≤55◦ ) was assessed. MRA-H findings were correlated with surgical findings and clinical management of patients.
3. Results We identified twenty-six adolescents who underwent MRA-H between 2004 and 2013, and age ranged between 13 and 18 years with an equal number of male and female patients. The patients were grouped based on the treatment offered and performed as follows: Group I) patients who underwent MRA-H followed by surgical management (n = 10); group II) patients who underwent MRA-H followed by non-surgical management (medication, intra-articular steroid injection, physical therapy) (n = 9); group III) patients who underwent MRA-H and were lost to follow up after being advised to have surgery (n = 7). The presenting complaints in patients of the three groups are listed in Table 1. Most patients presented with groin pain or hip pain (location of pain was not specified). Several patients presented with a popping or clicking sensation at the hip or with buttock pain. Three patients presented with a combination of these symptoms. Fifty-four percent of the patients presenting with pain gave a history of pain related to physical activity (40% of patients in group I, 67% patients in group II and 57% of patients in group III). The duration of pain prior to MRA-H was less than 3 months in 9 patients (33.33%), 3 months to 1 year in 9 patients (33.33%), and more than one year in 3 patients (11.11%); details on the duration of pain were not available in 5 patients (18.52%). Recreational activity
A. Jawahar et al. / European Journal of Radiology 85 (2016) 1192–1198 Table 3 MRA-H findings in study patients. Abnormality on MRA-H
Group I
Group II
Group III
Total
Labral tear: Paralabral cysts:
9 2
8 1
6 0
23 3
Cartilage abnormality: Cartilage thinning Full thickness defect Synovial thickening:
2 1 1
1 2 0
0 1 0
3 4 1
Bony abnormality: Cam FAI Pincer FAI SCFE LCP Exostosis AVN Developmental hip dysplasia Coxa Profunda
2 1 1 2 1 1 1 0
3 0 1 0 0 0 0 0
2 1 0 0 0 0 0 1
7 2 2 2 1 1 1 1
was present in 19/26 patients (73%) of our study, of which most of them (10) were soccer, football, basketball or hockey players. 7/26 patients did not have recreational activity, of which 5 were in group I and 2 were in group III. On the contrary, one patient in group I with no tear had no recreational activity. However, the patients in group II and III with no tears did have history of recreational activity like baseball and hockey. 31% had a positive hip impingement test on physical examination. On MR imaging, a total of 23 labral tears were diagnosed, 9 in group I, 8 in group II and 6 in group III patients. Of those 23 tears, 14 were anterior in location (Fig. 1), 3 anterosuperior, 1 posterosuperior, 4 posterior or posteroinferior and 1 multifocal tear (Fig. 2A and B). MRA-H also revealed other associated abnormalities in the hip including paralabral cysts, synovial thickening, cartilage abnormality and bony abnormalities. (Table 3) Individual group analysis are as follows: 3.1. Group I, surgical management (n = 10) Of the 10 patients in group I, 7 presented with groin pain, 2 with hip pain and 1 with buttock pain. Three had been popping in the affected hip joint and 4 had pain with activity. The clinical diagnosis in group I patients were as follows: Slipped capital femoral epiphysis (SCFE) in 1, Legg-Calve-Perthes disease (LCP) in 2, Exostosis in 1, Avascular necrosis in 1, Developmental dysplasia of hip (DDH) in 1, cam-type FAI in 2, pincer-type FAI in 1 and snapping hip in 1. Labral tears (9/10) in group I by location were anterior in 3 patients, anterosuperior in 3, posterior in 1, posterosuperior in 1 and multifocal in location in 1 patient. Etiology of the labral tears were traumatic in 1 (patient with snapping hip who also had history of recreational activity-gym), degenerative in 6 and idiopathic in 2. Imaging findings were correlated with the 10 surgeries completed. The interval between MRA-H examination and performing surgery in group I patients varied from one day to five months. There was agreement between MRA-H and surgery for presence/absence and location of labral tear in 9/10 cases. One false positive and one true negative labral tear was diagnosed. Thus, in this study MRA-H had a sensitivity of 100%, specificity of 50%, positive predictive value of 88.9%, negative predictive value of 100% and accuracy of 90% in detecting labral tears, though with a limitation of small power of surgically correlated results with MRA-H. Abnormal bony morphology was also correlated with surgical findings with confirmation in all 9 patients. Of the 3 patients who had clinical suspicion of FAI, two had abnormal alpha angles (58◦ and 68◦ ) with labral tear. Both underwent labral repair, with one greater trochanteric femoral osteotomy and one femoral neck osteoplasty. The third patient with clinical suspicion of FAI had a
1195
labral tear with osteonecrosis of femoral head and a normal alpha angle (31◦ ) on MRA-H. One patient was diagnosed with SCFE and had prior surgical management for SCFE with an abnormal alpha angle (82◦ ) and multifocal labral tear. Two patients had been finding consistent with LCP on imaging, one of whom with buttock pain had a labral tear and articular cartilage defect. The second patient had an intact labrum with synovial thickening, confirmed. The alpha angle in these last 3 patients could not be evaluated because of the bony morphologic changes. Six patients with bony morphologic abnormalities had secondary labral tears, confirmed at surgery. Three additional patients had labral tears on MRA-H and confirmed at surgery, with diagnoses of hip exostosis, developmental hip dysplasia, and snapping hip. The alpha angle measurement could not be evaluated in the patient with hip exostosis due to morphologic changes, but was normal in the latter two patients. One patient with clinical diagnosis of acetabular retroversion (pincer-FAI) had normal labrum on MRA-H and surgery, and alpha angle (33◦ ) also had identified posteroinferior hyaline cartilage abnormality. Articular cartilage injuries, varying from cartilage thinning to full thickness cartilage defect, were seen in 3 patients who also had labral tears. After surgery, pain resolved in 8/10, and persisted in severity in 1 patient who eventually required a repeat femoral osteoplasty for SCFE. One patient with LCP and an intact labrum who underwent periacetabular osteotomy, surgical hip dislocation with femoral head/neck osteoplasty was lost to follow-up after surgery. 3.2. Group II, medical management (n = 9) Of the 9 patients in group II, 5 presented with groin pain and 4 with hip pain. Four had been popping and 3 patients had been clicking in the affected hip joint and 6 had pain with activity. The clinical diagnosis in group II patients were as follows: cam-type FAI in 2, SCFE in 1, snapping hip in 1, and hip pain for evaluation with suspected labral tear in 5 patients. Labral tears (8/9) in group II by location were anterior in 6 patients, posterior in 1, posteroinferior in 1 patient. Etiology of the labral tears were traumatic in 3 (all of them had recreational physical activity-baseball and football), and idiopathic in 5 patients. In this group of 9 patients, all 3 patients with clinical suspicion of labral tears had a diagnosed labral tear on MRA-H. Two of these 3 patients had a normal alpha angle; neither responded symptomatically to non-surgical management. One of the 3 patients had an abnormal alpha angle (61◦ ); this patient responded favorably to non-surgical management. Though two additional patients had clinical suspicion of FAI, only one had a cam-FAI lesion with abnormal alpha angle (64◦ ). This patient also had an associated articular cartilage abnormality and labral tear on MRA-H. After non-surgical treatment the patient’s symptoms were reduced but persistent. One patient with clinical finding of snapping hip syndrome had a labral tear diagnosed on imaging, and the patient had improved symptomatology with non-surgical management. One patient with a traumatic anterior inferior iliac spine avulsion also had a diagnosed labral tear with abnormal alpha angle (64◦ ) on MRA-H; this patient’s symptoms improved with conservative management. The patient with SCFE had a full thickness articular cartilage defect with normal alpha angle (41◦ ) and an intact labrum. Labral findings in this patient were not confirmed at intertrochanteric osteotomy as the joint space was not entered, and hence this patient was included in group II. The last patient in this group had non-specific hip pain, with labral tear and articular cartilage injury on MRA-H but the pain resolved after conservative management. 3.3. Group III, lost to follow up (n = 7) Of the 7 patients in group III, 4 presented with groin pain and 5 with hip pain. Four patients had been popping and 2 had been
1196
A. Jawahar et al. / European Journal of Radiology 85 (2016) 1192–1198
clicking in the affected hip joint and 4 had pain with activity. The clinical diagnosis in group III patients were as follows: cam-type FAI in 2, pincer-type FAI in 1, coxa profunda in 1 and evaluation for suspected labral tear in 3 patients. Labral tears (6/7) in group III by location were anterior in 5 patients and posteroinferior in 1 patient. Etiology of all the 6 labral tears were idiopathic. Three patients in this group had a clinical suspicion of FAI of whom only one had an abnormal alpha angle (70◦ ), consistent with cam-type FAI. This patient had an intact labrum on MRA-H. The other 2 patients had normal alpha angles, although both had labral tears on MRA-H. The remaining four patients all had labral tears on MRA-H with clinical diagnoses of coxa profunda, greater trochanteric bursitis, sports hernia, and labral tear. The patient with greater trochanteric bursitis also had cam-type FAI morphology on MRA-H with an alpha angle of 70◦ . One patient had a full thickness acetabular articular cartilage injury with an intact labrum. Surgery was recommended for these 6 patients, but not performed and the patients were lost to follow-up.
4. Discussion This study has elucidated several important facts regarding MR hip arthrography in adolescents. In our study population of 26 adolescents aged 13–18 who had MR arthrogram of the hip, labral tears were identified in 23 cases. These patients presented with either a single complaint of hip pain, groin pain, clicking, locking, or pain on flexion/internal rotation, or with a combination of these signs/symptoms. Most notably, labral tears were noted in 14/16 (87.5%) of patients presenting with groin pain, and 14/26 (54%) of patients presenting with pain with activity. A labral tear was noted in 7/26 (27%) of patients with a positive anterior hip impingement test. Findings for location of labral tears and associated cartilage and bone pathology for the study group were not unexpected. Anterior labral tears were most common (14/23), of which 4 patients had FAI. All 5 patients with posterior labral tears in our study had abnormal bony hip morphology (2 pincer-type FAI, 1 cam-type FAI, 1 LCP, and 1 SCFE). The chondral abnormalities detected on MRA-H in 7 patients varied from hyaline articular cartilage thinning and subchondral cyst formation to full thickness cartilage defects. Chondral abnormalities could be confirmed in only 2 group I patients, as the remaining cases were not evaluated surgically. Eighty-seven percent (6/7) of the cam-type FAI patients in our study had a labral tear, of which 4 were located anteriorly. Of the 2 pincer-type FAI patients with acetabular retroversion in our study, one was diagnosed with posteroinferior labral tear without cartilage abnormality and the other had focal posterior articular cartilage abnormality with an intact labrum (Fig. 3). Classification of types of labral tears in this adolescent population yielded interesting results. According to prior descriptions [6,8,10–15], distribution of tears consisted of 4 traumatic tears, and only 6 tears secondary to bony morphologic abnormality, despite the young patient ages. 13/23 patients had labral tears categorized as idiopathic. Also of note, this study reveals that conservative management may be effective in this age group. 10 patients underwent surgical management of labral tears (1 with chronic debilitating pain and 9 after failed conservative management). 8/10 had pain resolution after arthroscopic surgery, and two went on to formal arthrotomy and additional bony reconstruction. In contrast, of the 8 patients with MR diagnosed labral tear managed conservatively in our study, hip pain resolved in 6 and persisted in 2 patients for one year or more. New onset hip pain in pediatric and adolescent patients demands evaluation for acute and evolving pathologies such as
Fig. 3. Radial PD image shows a focal acetabular articular cartilage defect in the right hip (arrow) of a 15-year old female with no history of recent trauma. Patient also had a Pincer-type FAI lesion.
infection, LCP, SCFE, developmental dysplasia of hip (DDH), labral tears, FAI, loose bodies, synovitis, and chondral lesions [16]. Labral tears have been recognized as a common cause of hip pain and mechanical hip symptoms [2–4,17–19]. Although well studied in the adult, there is only recent interest in the pediatric and adolescent populations. Labral tears commonly present with hip pain, clicking, locking, giving way or pain on flexion and internal rotation of the hip [1,3,4,8,10,17,18,20]. These symptoms were paralleled in our study, seen in 23/26 of our patients with MR diagnosed labral tears. Prevalence of labral tear in adults with groin pain has been reported as 22%–55% [6,11,13,18]. In our study, 87.5% with groin pain had labral tear on MRA-H, which may represent a selection bias related to tighter selection parameters for MR arthrography in the adolescent. Even though the most consistent physical examination finding reported with labral tears is the positive anterior hip impingement test [6], only 7/26 (27%) patients in our study presented with this finding. The location of majority of the labral tears described in adult patients are anterosuperior, while the location is more commonly anterior in adolescents and pediatric athletes [4,12,13,17,18,21–24] with detachment of the labrum from articular cartilage documented in 89% of cases [21]. Anterior labral tears were most common (14/23) among the adolescents in our study. Posterior labral tears are more commonly described in patients with posterior hip dislocation, hip dysplasia or discrete traumatic episode [11,19]. All 5 patients with posterior labral tears in our study had abnormal bony morphology. Posterior labral tears have been associated with a concurrent anterior labral lesion [3,4]. Our study did not find this association since only 1/5 posterior tears was associated with a concurrent anterosuperior labral tear, perhaps due to small sample size. Labral tears commonly exist with bony malformations of the hip [16,17]. Cam-type FAI can be congenital or acquired due to DDH, SCFE, avascular necrosis, or trauma [10]. Congenital cam-type FAI is more common in young athletic males and related to reduced concavity at the anterior femoral head-neck junction which comes in contact with the anterosuperior acetabular rim, leading to anterosuperior labral tears and articular cartilage defects [2,7,10,25]. Our findings corroborate this since labral tears were noted in 70% of our patients with bony morphologic abnormality. Further, 87% of
A. Jawahar et al. / European Journal of Radiology 85 (2016) 1192–1198
Fig. 4. Coronal PD image of the left hip shows acetabular and femoral head articular cartilage thinning (open arrows) and a loose body within the superolateral paralabral cyst (straight arrow) in 18-year old male with history of Legg-Calve-Perthes disease.
the cam-type FAI patients in our study had a labral tear, of which 4 were located anteriorly. Labral tears occur frequently in association with chondral defects in FAI [4,14]. Relative risk of chondral damage increases with presence of labral tears and increasing severity of tears [4,6,18]. An isolated labral tear is more frequently found in younger patients, whereas a labral tear in conjunction with chondral lesions is found more often in older patients, suggesting that a labral tear may precede and possibly lead to articular cartilage changes [6,18]. McCarthy et al. found that 73% of adults with labral tear had chondral damage [6,18]. Of the 23 patients with labral tears in our study, only 3 (13%) had an associated cartilage abnormality detected on MR, although cartilage abnormalities were seen in some cases of bony dysmorphology. Paralabral cysts on MRA-H may be associated with labral disorders, though not always [26]. When present, they are an indirect sign of labral tear and are more commonly seen in hip dysplasia [10,18]. Three paralabral cysts were identified in our study. One patient in our study with a paralabral cyst was associated with LCP (although the labrum was characterized as intact on MRA-H, and confirmed to be intact at surgery (Fig. 4)). Besides demonstrating labral tears and articular cartilage lesions, MR arthrography can delineate bony morphologic changes consistent with cam-type FAI such as flattened or convex anterosuperior portion of femoral head-neck junction (Fig. 5) producing an associated abnormal alpha angle >55◦ , subchondral edema, and subchondral cysts in cam-type FAI. This morphology was noted in 7 of our study patients. As the acetabular counterpart for impingement, Pincer-type FAI with a deep acetabulum and excessive acetabular coverage of the femoral is more commonly seen in middle-aged women than in adolescents and is due to morphological abnormality of the acetabular labrum from coxa profunda, acetabular protrusion or retroversion [2,10,25]. There is preferential involvement of anterior labrum due to direct contact of anterior acetabular margin with the femoral head-neck junction. Articular cartilage is relatively preserved in pincer-FAI in the early course of the disease. Later, there may be posteroinferior cartilage damage due to contra coup injury [10]. Of the 2 pincer-type FAI patients with acetabular retroversion in our study, one was diagnosed with posteroinferior labral
1197
Fig. 5. Axial T1-weighted image with fat saturation shows a cam-type FAI lesion with convexity noted at the anterior femoral head-neck junction in a 16-year old male with no history of trauma. Patient also had an anterosuperior labral tear (not shown here).
tear without cartilage abnormality and the other had focal posterior articular cartilage abnormality with an intact labrum (Fig. 3). Treatment of acetabular labral tears often begins with non-surgical management: relative rest, non-steroidal antiinflammatory drugs, and physical therapy. However, physical therapy remains controversial in the treatment of labral tears. Nonsurgical treatment for 10–12 weeks is initially recommended for patients presenting with hip pain and clinical suspicion of labral tear and may be more effective in acute labral tears than in chronic lesions [27]. Of the 8 patients with MR diagnosed labral tear managed conservatively in our study, hip pain resolved in 6. This is an attractive alternative for the pediatric patient. However, there may be an indication for arthroscopy in some patients. Garrison et al. suggest that patients with hip pain persisting for more than 4 weeks in conjunction with MRA-H diagnosis of an isolated labral tear are considered good candidates for arthroscopic management of labral tears as they generally have a good outcome [5]. Schindler et al. found that hip arthroscopy was useful for performing synovial biopsies and removing intra-articular loose bodies in addition to repairing the labral tears, thereby circumventing the need for an open surgery of the hip [28]. 8/10 patients in this series had pain relief following arthroscopy. Comparative analysis of imaging findings to surgical outcomes for 10 patients in our study yielded results similar to those reported in the literature. Prior reports have indicated MRA-H has 60%–100% sensitivity, 44%–100% specificity and accuracy of 91% for detecting acetabular labral lesions in adults [4,9,10,14,15,29]. In our study, MRA-H hip in adolescents had a sensitivity of 100%, specificity of 50%, positive predictive value of 88.9%, negative predictive value of 100% and an accuracy of 90% for detecting labral lesions. The described sensitivity and specificity of MRA-H in detecting chondral lesions in adults are slightly poorer, with range between 62%–79% and 77%–94% respectively [4,8,10,11,22,29]. Of the seven patients with chondral abnormalities on imaging in this study, two went on to surgery with confirmation of lesions, suggesting 100% sensitivity and positive predictive value of 100%. However, this very small sample size makes this statement suspect.
1198
A. Jawahar et al. / European Journal of Radiology 85 (2016) 1192–1198
As a retrospective study, clinical chart review for this study was limited to previously entered data, without rigid and consistent documentation of symptoms and clinical examination findings. Therefore, pertinent information pertaining to the suspected or proven diagnoses was included in the medical records. In addition relatively few adolescent patients in our hospital population had arthrography for evaluation of labral pathology, partly due to relatively recent general acceptance of MRA-H for evaluation of labral pathologies for pediatric patients. However, all study participants were derived from one dedicated sports orthopedic practice, with clinicians well versed in the conservative and surgical techniques for diagnosis and management of hip injuries. However, patterns were identified and diagnoses confirmed in this study allowing for initial conclusions. A prospective study on MRA-H in adolescents with a larger study population and controlled data input would be the next step to confirm the findings of our study. 5. Conclusion Even though labral and/or chondral pathology of the hip are considered risk factors for progressive arthropathy and debilitation, especially for individuals with long life expectancy, utility of hip MR arthrography in the pediatric and adolescent population has been relatively ignored comparative to the adult population. However, our study has shown that MRA-H can accurately identify labral, chondral and bony pathology in youth, with sensitivity and specificity similar to that reported for adults. Further, this early study supports potential benefits of both arthroscopic and conservative management options for patients with labral pathology. Therefore, hip MR arthrography should be considered as a valid imaging resource for pediatric and adolescent patients presenting with acute hip, groin, or buttock pain, clicking, or hip pain exacerbated by athletic activity. Funds Non-funded project. Presented at any scientific meeting or published earlier Presented at the Society of Pediatric Radiology-Annual meeting in May 2014. Source(s) of support In the form of MRI equipment for imaging of patients described in the manuscript was from the Department of Radiology, Loyola University Medical Center, Maywood, IL, USA. Conflict of interests There is no financial or any other form of real or apparent conflict of interest in the context of the subject of this article. Disclosure(s) None References [1] Mininder S. Kocher, Young-Jo Kim, Michael B. Millis, Rahul Mandiga, et al., Hip arthroscopy in children and adolescents, J. Pediatr. Orthop. 25 (September/October (5)) (2005) 680–686.
[2] M. Beck, M. Kalhor, M. Leunig, R. Ganz, 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.] 87-B (2005) 1012–1018. [3] Charlotte Hartig-Andreasen, Kjeld Søballe, Anders Troelsen, The role of the acetabular labrum in hip dysplasia—a literature overview, Acta Orthop. 84 (1) (2013) 60–64. [4] Cheryl Petersilge, Imaging of the acetabular labrum, Magn. Reson. Imaging Clin. N. Am. 13 (2005) 641–652. [5] Craig Garrison, Michael T. Osler, Steven B. Singleton, Rehabilitation after arthroscopy of an acetabular labral tear, N. Am. J. Sports Phys. Ther. 2 (4) (2007) 241–250. [6] Megan M. Groh, Joseph Herrera, A comprehensive review of hip labral tears, Curr. Rev. Musculoskelet. Med. 2 (2009) 105–117. [7] Marius R. Schmid, Hubert P. Nottzli, Marco Nottzli, Tobias F. Wyss, Juerg Hodler, Cartilage lesions in the hip: diagnostic effectiveness of MR arthrography1, Radiology 226 (2003) 382–386. [8] Glen A. Toomayan, W. Russell Holman, Nancy M. Major, Shannon M. Kozlowicz, T. Parker Vail, Sensitivity of MR arthrography in the evaluation of acetabular labral tears, AJR 186 (Febuary) (2006) 449–453. [9] M. Leunig, S. Werlen, A. Ungersbock, et al., Evaluation of the acetabulum labrum by MR arthrography, J. Bone Joint Surg. [Br.] 79 (1997) 230–234. [10] Sébastien Aubry, Danny Bélanger, Caroline Giguère, Martin Lavigne, Magnetic resonance arthrography of the hip: technique and spectrum of findings in younger patients, Insights Imaging 1 (2010) 72–82. [11] F.L. Carty, I. Crosbie, J. Ryan, J.P. Cashman, Imaging of the acetabular labrum: a review, Hard Tissue 2 (Febuary (1)) (2013) 9. [12] L.A. Lage, J.V. Patel, R.N. Villar, The acetabular labral tear: an arthroscopic classification, Arthroscopy 12 (1996) 269–272. [13] A.A. Narvani, E. Tsiridis, C.C. Tai, P. Thomas, Thomas Acetabular labrum and its tears, Br. J. Sports Med. 200337 (2015) 207–211. [14] Alexandros Tzaveas, Richard Villar, Acetabular labral and chondral pathology, Open Sports Med. J. 4 (2010) 64–74. [15] Srino Bharam, Labral tears, extra-articular injuries, and hip arthroscopy in the athlete, Clin. Sports Med. 25 (2006) 279–292. [16] K.R. Berend, T.P. Vail, Hip arthroscopy in the adolescent and pediatric athlete, Clin. Sports Med. 20 (2001) 763–768. [17] Philip A. Dinauer, Kevin P. Murphy, John F. Carroll, Sublabral sulcus at the posteroinferior acetabulum: a potential pitfall in MR arthrography diagnosis of acetabular labral tears, AJR 183 (2004) 1745–1753. [18] J.C. McCarthy, B. Busconi, The role of hip arthroscopy in the diagnosis and treatment of hip disease, Orthopedics 18 (1995) 753–756. [19] G. Neumann, A.D. Mendicuti, K.H. Zou, T. Minas, J. Coblyn, C.S. Winalski, et al., Prevalence of labral tears and cartilage loss in patients with mechanical symptoms of the hip: evaluation using MR arthrography, Osteoarthr. Cartil. 15 (2007) 909–917. [20] K. Klaue, C.W. Durni, R. Ganz, The acetabular rim syndrome: a clinical presentation of dysplasia of the hip, J. Bone Joint Surg. [Br.] 72B (1991) 423–429. [21] R.M. Seldes, V. Tan, J. Hunt, et al., Anatomy, histological features, and vascularity of the adult acetabular labrum, Clin. Orthop. 382 (2001) 232–240. [22] Andrew J. Ziegert, Donna G. Blankenbaker, Arthur A. De Smet, James S. Keene, Kazuhiko Shinki, Jason P. Fine, Comparison of standard hip MR arthrographic imaging planes and sequences for detection of arthroscopically proven labral tear, AJR 192 (2009) 1397–1400. [23] D.G. Blankenbaker, A.A. De Smet, J.S. Keene, J.P. Fine, Classification and localization of acetabular labral tears, Skelet. Radiol. 36 (2007) 391–397. [24] Bryan T. Kelly, Gary S. Shapiro, Christopher W. Digiovanni, Robert L. Buly, Hollis G. Potter, Jo A. Hannafin, Vascularity of the hip labrum: a cadaveric investigation, Arthroscopy 21 (1) (2005) 3–11. [25] Christian W.A. Pfirrmann, Bernard Mengiardi, Claudio Dora, Fabian Kalberer, Marco Zanetti, et al., Cam and pincer femoroacetabular impingement: characteristic MR arthrographic findings in 50 patients1, Radiology 240 (3) (2006) 778–785. [26] Thomas Magee, Gary Hinson, Association of paralabral cysts with acetabular disorders, AJR 174 (2000) 1381–1384. [27] Christopher M. Larson, Jennifer Swaringen, Grant Morrison, A review of hip arthroscopy and its role in the management of adult hip pain, Iowa Orthop. J. 25 (2016) 172–179. [28] A. Schindler, J.J. Lechevallier, N.S. Rao, et al., Diagnostic and therapeutic arthroscopy of the hip in children and adolescents: evaluation of results, J. Pediatr. Orthop. 15 (1995) 317–321. [29] C. Czerny, S. Hofmann, A. Neuhold, C. Tschauner, A. Engel, M.P. Recht, J. Kramer, Lesions of the acetabular labrum: accuracy of MR imaging and MR Arthrography in detection and staging1, Radiology 200 (1996) 225–230.
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Clinical and surgical correlation of hip MR arthrographic findings in adolescents Anugayathri Jawahar a,∗ , Aruna Vade a , Laurie Lomasney a , Gokcan Okur a , Douglas Evans b , Perla Subbaiah c a
Department of Radiology, Loyola University Medical Center, 2160, South 1st Avenue, Maywood, Illinois, 60153, United States Department of Orthopaedic Surgery and Rehabilitation, Loyola University Medical Center, 2160, South 1st Avenue, Maywood, Illinois, 60153, United States c Department of Mathematics and Statistics, Oakland University, Rochester, MI, 48309, United States b
a r t i c l e
i n f o
Article history: Received 22 August 2015 Received in revised form 27 March 2016 Accepted 31 March 2016 Keywords: MR hip arthrogram Adolescent Treatment Surgery Symptoms Follow up
a b s t r a c t Objective: The purpose of this study was to analyze the utility of MRA-H in adolescents by comparing the results of imaging with surgical findings and/or clinical outcome. Material and methods: After obtaining appropriate IRB approval, the Radiologic Information System database was queried for all patients 13–18 years of age who underwent MRA-H from 2004 through 2013. The electronic medical record was reviewed for clinical history, clinical examination findings, and operative notes. MRA-H images were reviewed for soft tissue abnormalities (labral tear, paralabral ganglion, articular cartilage loss, synovitis, ligament tears) and bony abnormalities (cam-type femoroacetabular impingement (FAI), pincer-type FAI, hip dysplasia). MRA-H findings were correlated with surgical findings and with clinical outcomes. Results: Twenty-six patients with labral tears by MRA-H were included in study and grouped as follows: Group I) patients who underwent surgical management (n = 10); group II) patients managed non-surgically (medication, intra-articular injection, physical therapy) (n = 9); group III) patients lost to follow up after being advised to have surgery (n = 7). With regard to presenting symptomatology, 87.5% of patients with labral tear had groin pain. Of those patients who were diagnosed with a labral tear, 52% were categorized as idiopathic labral tears, 26% as secondary tears (secondary to abnormal bony morphology), and 22% as traumatic labral tears. The labral tears were found to be anterior in 61% and posterior in 22%. Associated articular cartilage lesions were found in 29% of patients. In group I (surgical patients), MRA-H labral findings were confirmed at surgery in 9/10. Seventy percent of labral tears in our study had some form of abnormal bony morphology. Nine of the 12 patients with bone abnormalities were derived from group I patients. Six out of 7 patients with cam-type FAI had a labral tear. Conclusion: Labral tears diagnosed by MRA-H in the adolescent population correlated well with clinical examination and surgical findings. Also, MRA-H contributed by defining bony morphology that was directly applied to surgical management. Non-surgical management of labral tears diagnosed on MRA-H had a generally favorable outcome. © 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Hip pathologies affecting the labrum in children can be due to congenital causes such as hip dysplasia and femoral acetabular impingement, or acquired causes such as trauma, slipped capital femoral epiphysis (SCFE) and Legg-Calve-Perthes disease (LCP) [1]. Many of these hip pathologies in children can cause labral tears which if left untreated, may lead to premature degenerative arthri-
∗ Corresponding author. E-mail address: [email protected] (A. Jawahar). http://dx.doi.org/10.1016/j.ejrad.2016.03.031 0720-048X/© 2016 Elsevier Ireland Ltd. All rights reserved.
tis over time [2–4]. Magnetic resonance arthrogram of the hip (MRA-H) is used for assessment of labral tears, cartilage lesions and ligamentous injuries that occur with these hip pathologies so that appropriate treatment can be offered [5]. In adults, MRA-H has been shown to have 60%–100% sensitivity range and 44%–100% specificity range for detecting acetabular labral tears [6–8]. Previous literature has also documented a sensitivity range of 50%–79%, and specificity range of 77%–84% for detection of articular cartilage lesions in adults [6–8]. Even though MR arthrography of the hip has been used in the pediatric population in recent years, there are no published articles reviewing the utility of this minimally invasive procedure. The purpose of this study was to analyze the utility
A. Jawahar et al. / European Journal of Radiology 85 (2016) 1192–1198
1193
Fig 1. Axial T1-weighted image with fat saturation shows an anterior labral tear (arrow), in the left hip of a 14-year old male with recent history of trauma while playing baseball game. Patient also had an avulsion fracture of the anterior inferior iliac spine (not shown here).
Fig. 2. 16-Year old male with previous history of surgical treatment for slipped capital femoral epiphysis had multifocal labral tear in the left hip. Coronal T2-weighted image (A) shows an anterosuperior tear (arrow) and coronal oblique PD image (B) shows a sulcus (arrow). The labrum was found to be detached and very degenerative at surgery. Patient underwent labral debridement with repair for the tear.
of hip MR arthrography in adolescents by comparing the MRA-H findings with surgical and/or clinical outcome. 2. Material and methods This HIPPA-compliant retrospective research protocol was reviewed and approved by the Institutional Review Board (IRB number 205599). All delineated IRB requirements for provision of patient safety and privacy of patient information were observed. The Radiology Information System (RIS) was used to identify patients 18 years and younger who had an MR arthrogram of the hip between March 1, 2004 to March 31, 2013, and all were included in the study and were grouped according to their subsequent management plans and patient follow up. The hospital electronic medical records were used to document relevant clinical history and physical examination findings pertaining to the symptomatic hip (Table 1) and clinical outcome. Clinical outcome following surgical or conservative treatment was recorded as resolution, decrease or worsening of hip pain and their ability to resume their prior level of activity (Table 2). The operative
notes of the patients who underwent surgery of the symptomatic hip were reviewed by two radiologists (AJ, AV) and an orthopedic surgeon (DE) and surgical findings were correlated with the MRA-H findings. The operative notes were subjective description of location and nature of labral tear at arthroscopy or surgery. The hip arthrogram technique performed at our hospital is as follows: intra-capsular contrast material is injected under fluoroscopic guidance using an anterior inferior approach to the joint with patient supine and the knee in slight flexion. Capsular distention is achieved with 8–9 ml of dilute gadolinium ((Magnevist-gadopentetate dimeglumine, Bayer, Germany) or (Gadavist-gadobutrol, Bayer, Germany) diluted 1:250 in normal saline) with 4–5 ml of 1% lidocaine. The patient is transferred immediately to MR suite with minimal range of motion on transfer, but without placing weight on the extremity. In this study, MRI was performed in 1.5T magnets (Espree and AERA, Siemens Healthcare, Germany; Echospeed-15, GE Healthcare, Milwaukee) and 3T magnets (Trio and Verio, Siemens Healthcare, Germany). Axial, coronal, coronal oblique, radial PDweighted, and coronal T2-weighted images with fat saturation
1194
A. Jawahar et al. / European Journal of Radiology 85 (2016) 1192–1198
Table 1 Clinical history and physical examination findings pertaining to symptomatic hip. Groups
Hip pain
Groin pain
Buttock pain
Popping of hip
Clicking in hip
Pain related to activity
I II III
2 4 5
7 5 4
1 0 0
3 4 4
0 3 2
4 6 4
Total
11
16
1
11
5
14
Table 2 Clinical diagnosis, corresponding MRA-H labral findings and clinical outcome following treatment. Group
Patient
Clinical diagnosis
MRA-H findings on labrum
Severity of pain on follow up
Group I
1 2 3 4 5 6 7 8 9 10
Bilateral AVN Snapping hip Cam-FAI Multiple exostosis SCFE Cam-FAI DDH Pincer-FAI Perthes disease Perthes disease
Postero-superior tear Antero-superior tear Posterior tear Anterior tear Multifocal tear Incomplete anterior tear Antero-superior labral tear Incomplete antero-superior labral tear Anterior labral tear No labral tear
Resolved Resolved Improved Improved Worsened Resolved Worsened Improved Resolved Lost to follow up
Group II
1 2 3 4 5 6 7 8 9
Cam-FAI Suspected labral tear Cam-FAI Snapping hip Suspected labral tear SCFE Suspected labral tear Suspected labral tear Suspected labral tear
Posteroinferior labral tear Anterior labral tear Posterior labral tear Anterior labral tear Anterior labral tear Intact labrum Anterior labral tear Anterior labral tear Anterior labral tear
Improved Improved Improved Improved Improved Improved Resolved Worsened Worsened
Group III
1 2 3 4 5 6 7
Coxa profunda Cam-FAI Suspected labral tear Suspected labral tear Suspected labral tear Cam-FAI Pincer-FAI
Anterior labral tear Intact labrum Anterior labral tear Anterior labral tear Anterior labral tear Anterior labral tear Posteroinferior labral tear
Lost to follow up Lost to follow up Lost to follow up Lost to follow up Lost to follow up Lost to follow up Lost to follow up
were obtained. Body flex coil was used for imaging the hip. Slice thicknesses of 4 cm was obtained and FOV of ±16 cm was used depending on the physical habitus of the patient. The matrix used for spin echo and gradient echo sequences was 320 × 224 mm. Axial, radial, coronal and coronal oblique PD images had similar parameters including FOV 15 cm, TR 1775 ms, TE 17 ms. Coronal T2-weighted fat saturated images had FOV 15 cm, TR 3400 ms, TE 68 ms. The echo train used for T2W images was 8 ms for GE 3T scanner and 12 ms for GE 1.5T scanner. Radial PD-weighted sequence was oriented orthogonal to the acetabular rim and labrum. The MRA-H images were reviewed by 4 radiologists [AV (pediatric radiologist), AJ, GO, LL (musculoskeletal radiologist)] individually and in consensus for angle measurements and in case of variation in opinion in 2 patient findings who had recess. Kappa inter-observer variability was not calculated in our study since the focus of analysis was reaching a unanimous judgement. The different values of angle measurements were not recorded in two patients where there was varied opinion. However, for these two patients, there was agreement as to whether the angles are within the limits of normal or abnormal alpha angles. The readers were blinded to the surgical data at the time of image interpretation. The images were reviewed for labral tears, articular cartilage lesions and bone abnormalities. Synovial thickening or nodularity, ligament tears and tendon pathologies when identified were documented as well. The acetabular labrum was evaluated for presence of tear, MR signal characteristics and evidence of a perilabral recess. The alpha angle (as described by Pfirrmann [9]: on oblique axial images oriented parallel to the femoral neck, a measured angle between a line parallel to the femoral neck and a line from the
femoral head center to the anterior protruding head/neck cortex from an assumed spherical head; normal value-≤55◦ ) was assessed. MRA-H findings were correlated with surgical findings and clinical management of patients.
3. Results We identified twenty-six adolescents who underwent MRA-H between 2004 and 2013, and age ranged between 13 and 18 years with an equal number of male and female patients. The patients were grouped based on the treatment offered and performed as follows: Group I) patients who underwent MRA-H followed by surgical management (n = 10); group II) patients who underwent MRA-H followed by non-surgical management (medication, intra-articular steroid injection, physical therapy) (n = 9); group III) patients who underwent MRA-H and were lost to follow up after being advised to have surgery (n = 7). The presenting complaints in patients of the three groups are listed in Table 1. Most patients presented with groin pain or hip pain (location of pain was not specified). Several patients presented with a popping or clicking sensation at the hip or with buttock pain. Three patients presented with a combination of these symptoms. Fifty-four percent of the patients presenting with pain gave a history of pain related to physical activity (40% of patients in group I, 67% patients in group II and 57% of patients in group III). The duration of pain prior to MRA-H was less than 3 months in 9 patients (33.33%), 3 months to 1 year in 9 patients (33.33%), and more than one year in 3 patients (11.11%); details on the duration of pain were not available in 5 patients (18.52%). Recreational activity
A. Jawahar et al. / European Journal of Radiology 85 (2016) 1192–1198 Table 3 MRA-H findings in study patients. Abnormality on MRA-H
Group I
Group II
Group III
Total
Labral tear: Paralabral cysts:
9 2
8 1
6 0
23 3
Cartilage abnormality: Cartilage thinning Full thickness defect Synovial thickening:
2 1 1
1 2 0
0 1 0
3 4 1
Bony abnormality: Cam FAI Pincer FAI SCFE LCP Exostosis AVN Developmental hip dysplasia Coxa Profunda
2 1 1 2 1 1 1 0
3 0 1 0 0 0 0 0
2 1 0 0 0 0 0 1
7 2 2 2 1 1 1 1
was present in 19/26 patients (73%) of our study, of which most of them (10) were soccer, football, basketball or hockey players. 7/26 patients did not have recreational activity, of which 5 were in group I and 2 were in group III. On the contrary, one patient in group I with no tear had no recreational activity. However, the patients in group II and III with no tears did have history of recreational activity like baseball and hockey. 31% had a positive hip impingement test on physical examination. On MR imaging, a total of 23 labral tears were diagnosed, 9 in group I, 8 in group II and 6 in group III patients. Of those 23 tears, 14 were anterior in location (Fig. 1), 3 anterosuperior, 1 posterosuperior, 4 posterior or posteroinferior and 1 multifocal tear (Fig. 2A and B). MRA-H also revealed other associated abnormalities in the hip including paralabral cysts, synovial thickening, cartilage abnormality and bony abnormalities. (Table 3) Individual group analysis are as follows: 3.1. Group I, surgical management (n = 10) Of the 10 patients in group I, 7 presented with groin pain, 2 with hip pain and 1 with buttock pain. Three had been popping in the affected hip joint and 4 had pain with activity. The clinical diagnosis in group I patients were as follows: Slipped capital femoral epiphysis (SCFE) in 1, Legg-Calve-Perthes disease (LCP) in 2, Exostosis in 1, Avascular necrosis in 1, Developmental dysplasia of hip (DDH) in 1, cam-type FAI in 2, pincer-type FAI in 1 and snapping hip in 1. Labral tears (9/10) in group I by location were anterior in 3 patients, anterosuperior in 3, posterior in 1, posterosuperior in 1 and multifocal in location in 1 patient. Etiology of the labral tears were traumatic in 1 (patient with snapping hip who also had history of recreational activity-gym), degenerative in 6 and idiopathic in 2. Imaging findings were correlated with the 10 surgeries completed. The interval between MRA-H examination and performing surgery in group I patients varied from one day to five months. There was agreement between MRA-H and surgery for presence/absence and location of labral tear in 9/10 cases. One false positive and one true negative labral tear was diagnosed. Thus, in this study MRA-H had a sensitivity of 100%, specificity of 50%, positive predictive value of 88.9%, negative predictive value of 100% and accuracy of 90% in detecting labral tears, though with a limitation of small power of surgically correlated results with MRA-H. Abnormal bony morphology was also correlated with surgical findings with confirmation in all 9 patients. Of the 3 patients who had clinical suspicion of FAI, two had abnormal alpha angles (58◦ and 68◦ ) with labral tear. Both underwent labral repair, with one greater trochanteric femoral osteotomy and one femoral neck osteoplasty. The third patient with clinical suspicion of FAI had a
1195
labral tear with osteonecrosis of femoral head and a normal alpha angle (31◦ ) on MRA-H. One patient was diagnosed with SCFE and had prior surgical management for SCFE with an abnormal alpha angle (82◦ ) and multifocal labral tear. Two patients had been finding consistent with LCP on imaging, one of whom with buttock pain had a labral tear and articular cartilage defect. The second patient had an intact labrum with synovial thickening, confirmed. The alpha angle in these last 3 patients could not be evaluated because of the bony morphologic changes. Six patients with bony morphologic abnormalities had secondary labral tears, confirmed at surgery. Three additional patients had labral tears on MRA-H and confirmed at surgery, with diagnoses of hip exostosis, developmental hip dysplasia, and snapping hip. The alpha angle measurement could not be evaluated in the patient with hip exostosis due to morphologic changes, but was normal in the latter two patients. One patient with clinical diagnosis of acetabular retroversion (pincer-FAI) had normal labrum on MRA-H and surgery, and alpha angle (33◦ ) also had identified posteroinferior hyaline cartilage abnormality. Articular cartilage injuries, varying from cartilage thinning to full thickness cartilage defect, were seen in 3 patients who also had labral tears. After surgery, pain resolved in 8/10, and persisted in severity in 1 patient who eventually required a repeat femoral osteoplasty for SCFE. One patient with LCP and an intact labrum who underwent periacetabular osteotomy, surgical hip dislocation with femoral head/neck osteoplasty was lost to follow-up after surgery. 3.2. Group II, medical management (n = 9) Of the 9 patients in group II, 5 presented with groin pain and 4 with hip pain. Four had been popping and 3 patients had been clicking in the affected hip joint and 6 had pain with activity. The clinical diagnosis in group II patients were as follows: cam-type FAI in 2, SCFE in 1, snapping hip in 1, and hip pain for evaluation with suspected labral tear in 5 patients. Labral tears (8/9) in group II by location were anterior in 6 patients, posterior in 1, posteroinferior in 1 patient. Etiology of the labral tears were traumatic in 3 (all of them had recreational physical activity-baseball and football), and idiopathic in 5 patients. In this group of 9 patients, all 3 patients with clinical suspicion of labral tears had a diagnosed labral tear on MRA-H. Two of these 3 patients had a normal alpha angle; neither responded symptomatically to non-surgical management. One of the 3 patients had an abnormal alpha angle (61◦ ); this patient responded favorably to non-surgical management. Though two additional patients had clinical suspicion of FAI, only one had a cam-FAI lesion with abnormal alpha angle (64◦ ). This patient also had an associated articular cartilage abnormality and labral tear on MRA-H. After non-surgical treatment the patient’s symptoms were reduced but persistent. One patient with clinical finding of snapping hip syndrome had a labral tear diagnosed on imaging, and the patient had improved symptomatology with non-surgical management. One patient with a traumatic anterior inferior iliac spine avulsion also had a diagnosed labral tear with abnormal alpha angle (64◦ ) on MRA-H; this patient’s symptoms improved with conservative management. The patient with SCFE had a full thickness articular cartilage defect with normal alpha angle (41◦ ) and an intact labrum. Labral findings in this patient were not confirmed at intertrochanteric osteotomy as the joint space was not entered, and hence this patient was included in group II. The last patient in this group had non-specific hip pain, with labral tear and articular cartilage injury on MRA-H but the pain resolved after conservative management. 3.3. Group III, lost to follow up (n = 7) Of the 7 patients in group III, 4 presented with groin pain and 5 with hip pain. Four patients had been popping and 2 had been
1196
A. Jawahar et al. / European Journal of Radiology 85 (2016) 1192–1198
clicking in the affected hip joint and 4 had pain with activity. The clinical diagnosis in group III patients were as follows: cam-type FAI in 2, pincer-type FAI in 1, coxa profunda in 1 and evaluation for suspected labral tear in 3 patients. Labral tears (6/7) in group III by location were anterior in 5 patients and posteroinferior in 1 patient. Etiology of all the 6 labral tears were idiopathic. Three patients in this group had a clinical suspicion of FAI of whom only one had an abnormal alpha angle (70◦ ), consistent with cam-type FAI. This patient had an intact labrum on MRA-H. The other 2 patients had normal alpha angles, although both had labral tears on MRA-H. The remaining four patients all had labral tears on MRA-H with clinical diagnoses of coxa profunda, greater trochanteric bursitis, sports hernia, and labral tear. The patient with greater trochanteric bursitis also had cam-type FAI morphology on MRA-H with an alpha angle of 70◦ . One patient had a full thickness acetabular articular cartilage injury with an intact labrum. Surgery was recommended for these 6 patients, but not performed and the patients were lost to follow-up.
4. Discussion This study has elucidated several important facts regarding MR hip arthrography in adolescents. In our study population of 26 adolescents aged 13–18 who had MR arthrogram of the hip, labral tears were identified in 23 cases. These patients presented with either a single complaint of hip pain, groin pain, clicking, locking, or pain on flexion/internal rotation, or with a combination of these signs/symptoms. Most notably, labral tears were noted in 14/16 (87.5%) of patients presenting with groin pain, and 14/26 (54%) of patients presenting with pain with activity. A labral tear was noted in 7/26 (27%) of patients with a positive anterior hip impingement test. Findings for location of labral tears and associated cartilage and bone pathology for the study group were not unexpected. Anterior labral tears were most common (14/23), of which 4 patients had FAI. All 5 patients with posterior labral tears in our study had abnormal bony hip morphology (2 pincer-type FAI, 1 cam-type FAI, 1 LCP, and 1 SCFE). The chondral abnormalities detected on MRA-H in 7 patients varied from hyaline articular cartilage thinning and subchondral cyst formation to full thickness cartilage defects. Chondral abnormalities could be confirmed in only 2 group I patients, as the remaining cases were not evaluated surgically. Eighty-seven percent (6/7) of the cam-type FAI patients in our study had a labral tear, of which 4 were located anteriorly. Of the 2 pincer-type FAI patients with acetabular retroversion in our study, one was diagnosed with posteroinferior labral tear without cartilage abnormality and the other had focal posterior articular cartilage abnormality with an intact labrum (Fig. 3). Classification of types of labral tears in this adolescent population yielded interesting results. According to prior descriptions [6,8,10–15], distribution of tears consisted of 4 traumatic tears, and only 6 tears secondary to bony morphologic abnormality, despite the young patient ages. 13/23 patients had labral tears categorized as idiopathic. Also of note, this study reveals that conservative management may be effective in this age group. 10 patients underwent surgical management of labral tears (1 with chronic debilitating pain and 9 after failed conservative management). 8/10 had pain resolution after arthroscopic surgery, and two went on to formal arthrotomy and additional bony reconstruction. In contrast, of the 8 patients with MR diagnosed labral tear managed conservatively in our study, hip pain resolved in 6 and persisted in 2 patients for one year or more. New onset hip pain in pediatric and adolescent patients demands evaluation for acute and evolving pathologies such as
Fig. 3. Radial PD image shows a focal acetabular articular cartilage defect in the right hip (arrow) of a 15-year old female with no history of recent trauma. Patient also had a Pincer-type FAI lesion.
infection, LCP, SCFE, developmental dysplasia of hip (DDH), labral tears, FAI, loose bodies, synovitis, and chondral lesions [16]. Labral tears have been recognized as a common cause of hip pain and mechanical hip symptoms [2–4,17–19]. Although well studied in the adult, there is only recent interest in the pediatric and adolescent populations. Labral tears commonly present with hip pain, clicking, locking, giving way or pain on flexion and internal rotation of the hip [1,3,4,8,10,17,18,20]. These symptoms were paralleled in our study, seen in 23/26 of our patients with MR diagnosed labral tears. Prevalence of labral tear in adults with groin pain has been reported as 22%–55% [6,11,13,18]. In our study, 87.5% with groin pain had labral tear on MRA-H, which may represent a selection bias related to tighter selection parameters for MR arthrography in the adolescent. Even though the most consistent physical examination finding reported with labral tears is the positive anterior hip impingement test [6], only 7/26 (27%) patients in our study presented with this finding. The location of majority of the labral tears described in adult patients are anterosuperior, while the location is more commonly anterior in adolescents and pediatric athletes [4,12,13,17,18,21–24] with detachment of the labrum from articular cartilage documented in 89% of cases [21]. Anterior labral tears were most common (14/23) among the adolescents in our study. Posterior labral tears are more commonly described in patients with posterior hip dislocation, hip dysplasia or discrete traumatic episode [11,19]. All 5 patients with posterior labral tears in our study had abnormal bony morphology. Posterior labral tears have been associated with a concurrent anterior labral lesion [3,4]. Our study did not find this association since only 1/5 posterior tears was associated with a concurrent anterosuperior labral tear, perhaps due to small sample size. Labral tears commonly exist with bony malformations of the hip [16,17]. Cam-type FAI can be congenital or acquired due to DDH, SCFE, avascular necrosis, or trauma [10]. Congenital cam-type FAI is more common in young athletic males and related to reduced concavity at the anterior femoral head-neck junction which comes in contact with the anterosuperior acetabular rim, leading to anterosuperior labral tears and articular cartilage defects [2,7,10,25]. Our findings corroborate this since labral tears were noted in 70% of our patients with bony morphologic abnormality. Further, 87% of
A. Jawahar et al. / European Journal of Radiology 85 (2016) 1192–1198
Fig. 4. Coronal PD image of the left hip shows acetabular and femoral head articular cartilage thinning (open arrows) and a loose body within the superolateral paralabral cyst (straight arrow) in 18-year old male with history of Legg-Calve-Perthes disease.
the cam-type FAI patients in our study had a labral tear, of which 4 were located anteriorly. Labral tears occur frequently in association with chondral defects in FAI [4,14]. Relative risk of chondral damage increases with presence of labral tears and increasing severity of tears [4,6,18]. An isolated labral tear is more frequently found in younger patients, whereas a labral tear in conjunction with chondral lesions is found more often in older patients, suggesting that a labral tear may precede and possibly lead to articular cartilage changes [6,18]. McCarthy et al. found that 73% of adults with labral tear had chondral damage [6,18]. Of the 23 patients with labral tears in our study, only 3 (13%) had an associated cartilage abnormality detected on MR, although cartilage abnormalities were seen in some cases of bony dysmorphology. Paralabral cysts on MRA-H may be associated with labral disorders, though not always [26]. When present, they are an indirect sign of labral tear and are more commonly seen in hip dysplasia [10,18]. Three paralabral cysts were identified in our study. One patient in our study with a paralabral cyst was associated with LCP (although the labrum was characterized as intact on MRA-H, and confirmed to be intact at surgery (Fig. 4)). Besides demonstrating labral tears and articular cartilage lesions, MR arthrography can delineate bony morphologic changes consistent with cam-type FAI such as flattened or convex anterosuperior portion of femoral head-neck junction (Fig. 5) producing an associated abnormal alpha angle >55◦ , subchondral edema, and subchondral cysts in cam-type FAI. This morphology was noted in 7 of our study patients. As the acetabular counterpart for impingement, Pincer-type FAI with a deep acetabulum and excessive acetabular coverage of the femoral is more commonly seen in middle-aged women than in adolescents and is due to morphological abnormality of the acetabular labrum from coxa profunda, acetabular protrusion or retroversion [2,10,25]. There is preferential involvement of anterior labrum due to direct contact of anterior acetabular margin with the femoral head-neck junction. Articular cartilage is relatively preserved in pincer-FAI in the early course of the disease. Later, there may be posteroinferior cartilage damage due to contra coup injury [10]. Of the 2 pincer-type FAI patients with acetabular retroversion in our study, one was diagnosed with posteroinferior labral
1197
Fig. 5. Axial T1-weighted image with fat saturation shows a cam-type FAI lesion with convexity noted at the anterior femoral head-neck junction in a 16-year old male with no history of trauma. Patient also had an anterosuperior labral tear (not shown here).
tear without cartilage abnormality and the other had focal posterior articular cartilage abnormality with an intact labrum (Fig. 3). Treatment of acetabular labral tears often begins with non-surgical management: relative rest, non-steroidal antiinflammatory drugs, and physical therapy. However, physical therapy remains controversial in the treatment of labral tears. Nonsurgical treatment for 10–12 weeks is initially recommended for patients presenting with hip pain and clinical suspicion of labral tear and may be more effective in acute labral tears than in chronic lesions [27]. Of the 8 patients with MR diagnosed labral tear managed conservatively in our study, hip pain resolved in 6. This is an attractive alternative for the pediatric patient. However, there may be an indication for arthroscopy in some patients. Garrison et al. suggest that patients with hip pain persisting for more than 4 weeks in conjunction with MRA-H diagnosis of an isolated labral tear are considered good candidates for arthroscopic management of labral tears as they generally have a good outcome [5]. Schindler et al. found that hip arthroscopy was useful for performing synovial biopsies and removing intra-articular loose bodies in addition to repairing the labral tears, thereby circumventing the need for an open surgery of the hip [28]. 8/10 patients in this series had pain relief following arthroscopy. Comparative analysis of imaging findings to surgical outcomes for 10 patients in our study yielded results similar to those reported in the literature. Prior reports have indicated MRA-H has 60%–100% sensitivity, 44%–100% specificity and accuracy of 91% for detecting acetabular labral lesions in adults [4,9,10,14,15,29]. In our study, MRA-H hip in adolescents had a sensitivity of 100%, specificity of 50%, positive predictive value of 88.9%, negative predictive value of 100% and an accuracy of 90% for detecting labral lesions. The described sensitivity and specificity of MRA-H in detecting chondral lesions in adults are slightly poorer, with range between 62%–79% and 77%–94% respectively [4,8,10,11,22,29]. Of the seven patients with chondral abnormalities on imaging in this study, two went on to surgery with confirmation of lesions, suggesting 100% sensitivity and positive predictive value of 100%. However, this very small sample size makes this statement suspect.
1198
A. Jawahar et al. / European Journal of Radiology 85 (2016) 1192–1198
As a retrospective study, clinical chart review for this study was limited to previously entered data, without rigid and consistent documentation of symptoms and clinical examination findings. Therefore, pertinent information pertaining to the suspected or proven diagnoses was included in the medical records. In addition relatively few adolescent patients in our hospital population had arthrography for evaluation of labral pathology, partly due to relatively recent general acceptance of MRA-H for evaluation of labral pathologies for pediatric patients. However, all study participants were derived from one dedicated sports orthopedic practice, with clinicians well versed in the conservative and surgical techniques for diagnosis and management of hip injuries. However, patterns were identified and diagnoses confirmed in this study allowing for initial conclusions. A prospective study on MRA-H in adolescents with a larger study population and controlled data input would be the next step to confirm the findings of our study. 5. Conclusion Even though labral and/or chondral pathology of the hip are considered risk factors for progressive arthropathy and debilitation, especially for individuals with long life expectancy, utility of hip MR arthrography in the pediatric and adolescent population has been relatively ignored comparative to the adult population. However, our study has shown that MRA-H can accurately identify labral, chondral and bony pathology in youth, with sensitivity and specificity similar to that reported for adults. Further, this early study supports potential benefits of both arthroscopic and conservative management options for patients with labral pathology. Therefore, hip MR arthrography should be considered as a valid imaging resource for pediatric and adolescent patients presenting with acute hip, groin, or buttock pain, clicking, or hip pain exacerbated by athletic activity. Funds Non-funded project. Presented at any scientific meeting or published earlier Presented at the Society of Pediatric Radiology-Annual meeting in May 2014. Source(s) of support In the form of MRI equipment for imaging of patients described in the manuscript was from the Department of Radiology, Loyola University Medical Center, Maywood, IL, USA. Conflict of interests There is no financial or any other form of real or apparent conflict of interest in the context of the subject of this article. Disclosure(s) None References [1] Mininder S. Kocher, Young-Jo Kim, Michael B. Millis, Rahul Mandiga, et al., Hip arthroscopy in children and adolescents, J. Pediatr. Orthop. 25 (September/October (5)) (2005) 680–686.
[2] M. Beck, M. Kalhor, M. Leunig, R. Ganz, 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.] 87-B (2005) 1012–1018. [3] Charlotte Hartig-Andreasen, Kjeld Søballe, Anders Troelsen, The role of the acetabular labrum in hip dysplasia—a literature overview, Acta Orthop. 84 (1) (2013) 60–64. [4] Cheryl Petersilge, Imaging of the acetabular labrum, Magn. Reson. Imaging Clin. N. Am. 13 (2005) 641–652. [5] Craig Garrison, Michael T. Osler, Steven B. Singleton, Rehabilitation after arthroscopy of an acetabular labral tear, N. Am. J. Sports Phys. Ther. 2 (4) (2007) 241–250. [6] Megan M. Groh, Joseph Herrera, A comprehensive review of hip labral tears, Curr. Rev. Musculoskelet. Med. 2 (2009) 105–117. [7] Marius R. Schmid, Hubert P. Nottzli, Marco Nottzli, Tobias F. Wyss, Juerg Hodler, Cartilage lesions in the hip: diagnostic effectiveness of MR arthrography1, Radiology 226 (2003) 382–386. [8] Glen A. Toomayan, W. Russell Holman, Nancy M. Major, Shannon M. Kozlowicz, T. Parker Vail, Sensitivity of MR arthrography in the evaluation of acetabular labral tears, AJR 186 (Febuary) (2006) 449–453. [9] M. Leunig, S. Werlen, A. Ungersbock, et al., Evaluation of the acetabulum labrum by MR arthrography, J. Bone Joint Surg. [Br.] 79 (1997) 230–234. [10] Sébastien Aubry, Danny Bélanger, Caroline Giguère, Martin Lavigne, Magnetic resonance arthrography of the hip: technique and spectrum of findings in younger patients, Insights Imaging 1 (2010) 72–82. [11] F.L. Carty, I. Crosbie, J. Ryan, J.P. Cashman, Imaging of the acetabular labrum: a review, Hard Tissue 2 (Febuary (1)) (2013) 9. [12] L.A. Lage, J.V. Patel, R.N. Villar, The acetabular labral tear: an arthroscopic classification, Arthroscopy 12 (1996) 269–272. [13] A.A. Narvani, E. Tsiridis, C.C. Tai, P. Thomas, Thomas Acetabular labrum and its tears, Br. J. Sports Med. 200337 (2015) 207–211. [14] Alexandros Tzaveas, Richard Villar, Acetabular labral and chondral pathology, Open Sports Med. J. 4 (2010) 64–74. [15] Srino Bharam, Labral tears, extra-articular injuries, and hip arthroscopy in the athlete, Clin. Sports Med. 25 (2006) 279–292. [16] K.R. Berend, T.P. Vail, Hip arthroscopy in the adolescent and pediatric athlete, Clin. Sports Med. 20 (2001) 763–768. [17] Philip A. Dinauer, Kevin P. Murphy, John F. Carroll, Sublabral sulcus at the posteroinferior acetabulum: a potential pitfall in MR arthrography diagnosis of acetabular labral tears, AJR 183 (2004) 1745–1753. [18] J.C. McCarthy, B. Busconi, The role of hip arthroscopy in the diagnosis and treatment of hip disease, Orthopedics 18 (1995) 753–756. [19] G. Neumann, A.D. Mendicuti, K.H. Zou, T. Minas, J. Coblyn, C.S. Winalski, et al., Prevalence of labral tears and cartilage loss in patients with mechanical symptoms of the hip: evaluation using MR arthrography, Osteoarthr. Cartil. 15 (2007) 909–917. [20] K. Klaue, C.W. Durni, R. Ganz, The acetabular rim syndrome: a clinical presentation of dysplasia of the hip, J. Bone Joint Surg. [Br.] 72B (1991) 423–429. [21] R.M. Seldes, V. Tan, J. Hunt, et al., Anatomy, histological features, and vascularity of the adult acetabular labrum, Clin. Orthop. 382 (2001) 232–240. [22] Andrew J. Ziegert, Donna G. Blankenbaker, Arthur A. De Smet, James S. Keene, Kazuhiko Shinki, Jason P. Fine, Comparison of standard hip MR arthrographic imaging planes and sequences for detection of arthroscopically proven labral tear, AJR 192 (2009) 1397–1400. [23] D.G. Blankenbaker, A.A. De Smet, J.S. Keene, J.P. Fine, Classification and localization of acetabular labral tears, Skelet. Radiol. 36 (2007) 391–397. [24] Bryan T. Kelly, Gary S. Shapiro, Christopher W. Digiovanni, Robert L. Buly, Hollis G. Potter, Jo A. Hannafin, Vascularity of the hip labrum: a cadaveric investigation, Arthroscopy 21 (1) (2005) 3–11. [25] Christian W.A. Pfirrmann, Bernard Mengiardi, Claudio Dora, Fabian Kalberer, Marco Zanetti, et al., Cam and pincer femoroacetabular impingement: characteristic MR arthrographic findings in 50 patients1, Radiology 240 (3) (2006) 778–785. [26] Thomas Magee, Gary Hinson, Association of paralabral cysts with acetabular disorders, AJR 174 (2000) 1381–1384. [27] Christopher M. Larson, Jennifer Swaringen, Grant Morrison, A review of hip arthroscopy and its role in the management of adult hip pain, Iowa Orthop. J. 25 (2016) 172–179. [28] A. Schindler, J.J. Lechevallier, N.S. Rao, et al., Diagnostic and therapeutic arthroscopy of the hip in children and adolescents: evaluation of results, J. Pediatr. Orthop. 15 (1995) 317–321. [29] C. Czerny, S. Hofmann, A. Neuhold, C. Tschauner, A. Engel, M.P. Recht, J. Kramer, Lesions of the acetabular labrum: accuracy of MR imaging and MR Arthrography in detection and staging1, Radiology 200 (1996) 225–230.