- Email: [email protected]
Greater trochanter pain syndrome: A descriptive MR imaging study
G Model
ARTICLE IN PRESS
EURR-6820; No. of Pages 6
European Journal of Radiology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Greater trochanter pain syndrome: A descriptive MR imaging study Michail E. Klontzas 1 , Apostolos H. Karantanas ∗ Department of Medical Imaging, University Hospital of Heraklion and Radiology Section, Medical School University of Crete, Greece
a r t i c l e
i n f o
Article history: Received 15 January 2014 Received in revised form 8 April 2014 Accepted 16 June 2014 Keywords: Greater trochanter pain syndrome MR imaging CE angle Gluteal tendinopathy Bursitis
a b s t r a c t Objective: Greater trochanter pain syndrome (GTPS) is a diverse clinical entity caused by a variety of underlying conditions. We sought to explore the impact of (1) hip morphology, namely the center-edge angle (CEa) and femoral neck-shaft (NSa) angle, (2) hip abductor tendon degeneration, (3) the dimensions of peritrochanteric edema and (4) bursitis, on the presence of GTPS, using MR imaging. Materials and methods: The presence of pain was prospectively assessed blindly by the senior author. CEa and NSa were blindly measured in 174 hip MR examinations, after completion of the clinical evaluation by another evaluator. The existence and dimensions of T2 hyperintensity of the peritrochanteric soft tissues, the existence and dimensions of bursae, as well as degeneration and tearing of gluteus tendons were also recorded. Results: Out of 174 examinations, 91 displayed peritrochanteric edema (group A) and 34 bursitis, all with peritrochanteric edema (group B). A number of 78 patients from both A and B groups, showed gluteus medius tendon degeneration and one tendon tear. CEa of groups A and B were 6◦ higher than those of normals (group C, P = 0.0038). The mean age of normals was 16.6 years less than in group A and 19.8 years less than in group B (P < 0.0001). Bursitis was associated with pain with a negative predictive value of 97% (P = 0.0003). Conclusion: Acetabular morphology is associated with GTPS and the absence of bursitis was proved to be clinically relevant. Peritrochanteric edema alone was not associated with local pain. © 2014 Published by Elsevier Ireland Ltd.
1. Introduction Greater trochanter pain syndrome (GTPS) is characterized by pain and tenderness over the great trochanter. Its diagnosis is based on a combination of data from medical history, physical examination and imaging findings. Female gender in the sixth decade, a femoral neck-shaft angle (NSa) less than 134◦ , and leg length discrepancies, have been described as risk factors for GTPS [1–3]. A variety of conditions, such as degenerative hip disease, femoroacetabular impingement, femoral head avascular necrosis, infection and conditions that can modify hip biomechanics, such as knee osteoarthritis, iliotibial band syndrome and lumbar spine degenerative disease, can clinically mimic GTPS, making the clinical differential diagnosis extremely complicated [4,5].
∗ Corresponding author at: University of Crete, Department of Medical Imaging, University Hospital, Voutes 71110, Heraklion-Crete, Greece. Tel.: +30 2813392541; fax: +30 2813542095. E-mail addresses: [email protected] (M.E. Klontzas), [email protected], [email protected] (A.H. Karantanas). 1 Address: Souliou 15 Poros, Heraklion, 71307 Crete, Greece. Tel.: +30 6977795314.
Unlike the original belief that the term “GTPS” is synonymous to “trochanteric bursitis”, we nowadays accept that this syndrome may result not only by an inflamed bursa, but also by gluteus tendinopathy/tear and external coxa saltans, which refers to iliotibial band snapping [6–8]. Other rare disorders may also result in GTPS [9–12]. Kong et al. described various findings associated with GTPS, as seen in hip MR imaging examinations and underlined the need of a specific diagnosis in order to optimize the treatment strategies [13]. Haliloglu et al. found that T2 peritrochanteric hyperintensity representing edema is by far the most common finding but is rarely related to clinical symptoms [14]. Various studies have reported association between MR imaging findings and the presence of pain [1,15–17]. In the absence of peritrochanteric hyperintensity on fluid sensitive sequences, the GTPS is an unlike diagnosis [15]. On the other hand, large amounts of fluid within bursae, may correlate with clinical presentation [15,18]. No study, to the best of our knowledge, has compared patients with isolated peritrochanteric T2 hyperintensity in the absence of tendinous tears, with patients demonstrating bursitis. Moreover, no study has evaluated the role of acetabular morphology in patients with GTPS. We sought to assess the relation of acetabular morphology and NSa with the presence of GTPS and to evaluate the association
http://dx.doi.org/10.1016/j.ejrad.2014.06.009 0720-048X/© 2014 Published by Elsevier Ireland Ltd.
Please cite this article in press as: Klontzas ME, Karantanas AH. Greater trochanter pain syndrome: A descriptive MR imaging study. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.06.009
G Model
ARTICLE IN PRESS
EURR-6820; No. of Pages 6
M.E. Klontzas, A.H. Karantanas / European Journal of Radiology xxx (2014) xxx–xxx
2
between peritrochanteric edema and bursitis on MR imaging and the existence of pain. Secondly, we evaluated the impact of hip abductor tendon degeneration on the presence of pain. Finally, we explored whether the dimensions of peritrochanteric edema and bursitis, could imply the presence of a painful hip. 2. Materials and methods 2.1. Patients A total of 224 consecutive MR imaging examinations of hip joints from 141 patients, referred to our department for various clinical indications from June 2008 to April 2013, were prospectively evaluated. The most common indications included early and radiologically occult osteoarthritis, trochanteritis, early osteonecrosis, transient bone marrow edema, and labral abnormalities. Our study has been approved by our hospital’s ethics committee, was performed in the context of the principles of Helsinki Declaration and all patients have signed an informed consent. Patients with severe degenerative osteoarthritis of the hip and/or the knee joints (by means of presence of marginal osteophytes and subarticular geodes), previous septic or inflammatory arthritis, tumors around hip joints, previous hip surgery or radiotherapy, advanced femoral head osteonecrosis (ARCO grade III or more) as well as patients who had local corticosteroid injections, were excluded from our study. Exclusion was made both by imaging (plain radiographs and/or MR imaging) and clinical means. The 1.5 mm joint space was considered to represent advanced osteoarthritis. In total, 174 hip joint examinations from 92 patients (mean age 53.99 ± 15.95 years) were finally included. The population of our study consisted of 26 male and 66 female patients.
Fig. 1. A 55-year-old, asymptomatic for trochanteric pain, female patient. The coronal STIR image shows peritrochanteric edema (arrows).
on fluid sensitive sequences within the tendon together with attenuation or thinning of the tendon. Discontinuity of the tendon, with or without osseous avulsion, was diagnostic of full thickness tear [19]. 2.4. Measurements In order to assess the acetabular morphology and the morphology of the upper third of the femur, we performed two measurements (Fig. 5). The center-edge angle (CEa) was used in order to evaluate the coverage of the femoral head by the acetabulum. This was done by drawing a circle around both the femoral heads in order to find their center. Then, the center of each femoral head was used as the tip of an angle, with one side perpendicular to the line that crosses the centers of the two
2.2. Imaging All examinations were performed at 1.5T (Vision Hybrid, Siemens, Erlangen, Germany) using a phased-array flexible torso coil. From the standard hip MR imaging protocol utilized in our department, the following 4 sequences were used in the study: Coronal T1 weighted (w) turbo spin-echo (TSE) images (TR/TE: 532/13; slice thickness: 4 mm; matrix: 512 × 256; FOV: 37.5 cm; ETL: 3), axial T1-w TSE images (TR/TE: 511/12; slice thickness: 4 mm; matrix: 320 × 320; FOV: 40 cm; ETL: 3), coronal T1-w turbo inversion recovery (STIR) images (TR/TE/TI: 4960/53/170; slice thickness: 4 mm; matrix: 320 × 288; FOV: 40 cm; ETL: 5), and axial fat suppressed T2-w TSE images (TR/TE: 3250/80; slice thickness: 4 mm; matrix: 280 × 320; FOV: 40 cm × 35 cm; ETL: 5). Fat suppression was achieved with spectral presaturation. 2.3. MR imaging findings classification All MR imaging examinations were blindly evaluated by the first author, after completion of the clinical examination. The study population was categorized in three distinct groups: group A presenting with peritrochanteric edema, assessed as a poorly defined area of high signal intensity on fluid sensitive sequences, seen on both axial and coronal images (Fig. 1); group B showing distention of at least one of the region’s bursae, namely trochanteric or subgluteus maximus, subgluteus medius and subgluteus minimus (Figs. 2–4) assessed on two planes of fluid sensitive sequences; group C including examinees without peritrochanteric edema or bursal distention. Tendon degeneration and/or partial or complete tear was recorded in all MR imaging examinations. Tendinopathy – tendon degeneration was diagnosed when there was abnormally increased signal within the tendon on T1-w sequences, with or without thickening of the tendon (Fig. 3). Partial tearing of the tendon was diagnosed when increased signal intensity was depicted
Fig. 2. A 70-year-old female patient with greater trochanter pain syndrome on the left. (A) The coronal STIR MR image shows a large trochanteric bursa (arrows). (B) The axial fat suppressed T2-w MR image shows the trochanteric bursa (open arrow) and in addition bilateral iliopsoas bursae (arrows).
Please cite this article in press as: Klontzas ME, Karantanas AH. Greater trochanter pain syndrome: A descriptive MR imaging study. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.06.009
G Model EURR-6820; No. of Pages 6
ARTICLE IN PRESS M.E. Klontzas, A.H. Karantanas / European Journal of Radiology xxx (2014) xxx–xxx
3
Fig. 3. A 56-year-old female patient with right sided greater trochanter pain syndrome. (A) The axial T1-w MR image shows thickening and abnormally high signal of the right gluteus minimus tendon (arrow). The contralateral tendon on the left is normal (open arrow). (B) The axial fat suppressed T2-w MR image shows subgluteus minimus bursa (arrow), subgluteus medius bursa (thick arrow) and abnormal intratendinous signal within the thickened gluteus minimus tendon (open arrow). The coronal STIR MR images show subgluteus minimus bursa (arrow in C) and subgluteus medius bursa (arrow in D). Peritrochanteric edema is shown on the left side (small arrows in D).
femoral heads and the other side crossing the lateral edge of the acetabulum. In addition, the NSa was measured in an attempt to confirm the association between the morphology of the proximal femur and the development of GTPS. CEa and NSa were measured on mid-coronal T1-w TSE MR images. NSa and CEa of 20 randomly chosen hips, were measured twice in order to assess the intra-rater reliability by means of intra-class correlation coefficient (ICC) and coefficient of variation (CV) for duplicate measurements. The dimensions (anterior–posterior, superior–inferior, left–right) of the peritrochanteric edema and of the distended bursae, were also measured on the fluid sensitive sequences. The highest value of those dimensions was used for our analysis.
2.5. Clinical data The senior author of the study, a radiologist with a 27-year experience on musculoskeletal disorders, prospectively and without knowing the clinical indication for the MR imaging scan, performed a thorough clinical examination on site, just before undergoing the MR imaging examination. The clinical tests included: (a) pain during active abduction and internal rotation, (b) a positive Trendelenburg’s sign and (c) pain on pressure over the greater trochanter with the patient in the side-lying position. The presence of lateral hip pain together with at least one positive clinical test, were the prerequisites for diagnosing the presence of GTPS [1,20]. The pain was recorded as either present or absent. The evaluator of the MR imaging scans was blinded to all data recorded by the senior author.
2.6. Statistical analysis Data were analyzed using MedCalc 12.7.1 statistical software (MedCalc Software, Belgium). Our data were non-Gaussian and our three unmatched groups included outliers. Thus, comparison between CEa and NSa as well as patients’ age was performed using the non-parametric Kruskal–Wallis test. Normality was tested graphically by histogram inspection and numerically using the Shapiro–Wilk test. Fisher’s exact test was used to determine the association between MR imaging findings (peritrochanteric edema or bursitis) and clinically overt pain, as well as the association between the existence of tendon degeneration and pain. In addition, Mann–Whitney U test was used to analyze the association between the dimensions of peritendinous edema and bursitis with pain and those without pain. Positive predictive value (PPV) and negative predictive value (NPV) of bursitis with regard to pain were calculated. True negative cases were patients without bursitis and GTPS, false negative cases were patients who presented with GTPS but no overt bursitis on MR imaging, true positive cases were patients with both bursitis and GTPS and false positive were patients with bursitis but no clinical signs of GTPS. Results were regarded as statistically significant with a P-value of less than 0.05.
3. Results Data analysis revealed that 91 (52.3%) hips displayed peritrochanteric T2 hyperintensity corresponding to soft tissue edema. Thirty-four (19.5%) hips had a distended bursa (18 trochanteric,
Please cite this article in press as: Klontzas ME, Karantanas AH. Greater trochanter pain syndrome: A descriptive MR imaging study. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.06.009
G Model EURR-6820; No. of Pages 6 4
ARTICLE IN PRESS M.E. Klontzas, A.H. Karantanas / European Journal of Radiology xxx (2014) xxx–xxx
Fig. 6. Age distribution in all 3 groups of examined patients.
degeneration and 1 gluteus medius tendon full tear with atrophy and fatty infiltration of the muscle. There were 44 degenerated tendons in group A, 15 in group B and 19 in group C.
3.1. Patient’s age
Fig. 4. A 25-year-old male patient with ankylosing spondylitis and no indication of greater trochanter pain syndrome. The axial fat suppressed T2-w (A) and coronal STIR (B) MR images, show bilateral trochanteric bursae (arrows). Note the subarticular edema in the right sacroiliac joint (open arrow in B).
A statistically significant difference was found between the ages of the three groups (P < 0.0001). Subjects in group C were 16.6 years younger (mean age 41.4 years old – 95%CI from 37.3 to 45.5 years) compared to group A (mean age 58.0 years old – 95%CI from 55.4 to 60.7 years) and 19.8 years younger than group B (mean age 61.2 years old – 95%CI from 55.6 to 66.8 years) (Fig. 6).
3.2. Acetabular morphology 9 subgluteus medius, 6 trochanteric and subgluteus medius and 1 subgluteus minimus and medius). All the 34 hips of group B demonstrated associated peritrochanteric edema. In 49 hips, there were neither peritrochanteric edema nor bursitis. In addition, 78 out of 174 (44.8%) hips displayed gluteus medius tendon
The CEa of normal subjects were significantly different compared to those measured in groups B and A (P = 0.0038). Hips from patients in group C had a mean CEa of 36.7◦ (95%CI from 34.2◦ to 39.2◦ ), compared to 41.0◦ of group A (95%CI from 39.0◦ to 43.0◦ ) and 42.6◦ of group B (95%CI from 39.6◦ to 45.5◦ ) (Fig. 7).
Fig. 5. Measurements of the center-edge angle of Wiberg (CEa) and neck-shaft angle (NSa) on a mid-coronal T1-w MR image.
Please cite this article in press as: Klontzas ME, Karantanas AH. Greater trochanter pain syndrome: A descriptive MR imaging study. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.06.009
G Model
ARTICLE IN PRESS
EURR-6820; No. of Pages 6
M.E. Klontzas, A.H. Karantanas / European Journal of Radiology xxx (2014) xxx–xxx
5
subjects with GTPS (mean 2.4 cm, 95%CI from 2.0 to 2.9 cm) and those without (mean 2.9 cm, 95%CI from 2.75 to 3.1) (P = 0.1545). 4. Discussion
Fig. 7. Association of the center to edge angles in all 3 groups of patients.
Table 1 Association between bursitis and GPTS.a P = 0.0003
GTPS
No GTPS
Total
8 4
26 136
34 140
12
162
174
Bursitis No bursitis Total a
GTPS, greater trochanter pain syndrome.
3.3. Femoral morphology No statistically significant difference (P = 0.4260) was found between the NSa of hips in group C (mean NSa 125.8◦ , 95%CI from 123.1◦ to 128.5◦ ) when compared to group A (mean NSa 124.9◦ , 95% from 123.4◦ to 126.4◦ ) and group B (mean NSa 123.4◦ , 95%CI from 120.1◦ to 126.6◦ ). Moreover, the intra-rater reliability was high, with a CV for duplicate measurements of 3.2% and 6.5% for NSa and CEa measurements respectively. In addition, the ICC was 0.81 (95%CI from 0.58 to 0.92) and 0.9 (95%CI from 0.77 to 0.96) for NSa and CEa measurements respectively. 3.4. Positive tests for GTPS A statistically significant association between the existence of bursitis and the presence of GTPS (P = 0.0003) was found. Eight out of 34 patients with bursitis experienced pain, whereas only 4 of those without bursitis reported clinical symptoms, which yields a negative predictive value of 97.14% (95%CI from 92.85% to 99.22%) and a positive predicted value of 23.53% (95%CI from 10.75% to 41.17%) (Table 1). The presence of gluteus medius degeneration was not associated with pain, as only 8 out of 79 patients with degeneration (including the one with tear) experienced pain when clinically examined and 4 out of 95 patients without degeneration had painful hips (P = 0.1437). Analysis of association between the presence of pain and either the CEa or NSa did not show any statistical significance. 3.5. Dimensions No statistically significant difference was found between the maximum bursal size in symptomatic (mean 2.5 cm, 95%CI from 1.3 to 3.7) and asymptomatic (mean 1.9 cm, 95%CI from 1.4 to 2.4) subjects (P = 0.2496). Likewise, the maximum dimensions of peritrochanteric T2 hyperintensity did not differ significantly between
Our study showed that a relationship exists between acetabular morphology and the presence of peritrochanteric bursitis. In addition, the NPV of bursitis on MR imaging, is a clinically important finding. GTPS is a common cause of lateral hip pain with a broad differential diagnosis and is defined as pain and tenderness to palpation over the greater trochanter [21,22]. The anatomy of the trochanteric region is extremely complex, including the tendons of gluteus muscles, the piriformis and obturator tendons and several bursae, the most important of which are the subgluteus medius and subgluteus minimus bursae, as well as the trochanteric bursa [19,23–26]. Peritrochanteric pain may result from impingement of the iliotibial band, tendinopathy/tendon tearing and inflammation of bursae, either primary or secondary [13,15,21]. No study, to the best of our knowledge, has proposed a possible association between the acetabular morphology and GTPS. Our hypothesis was that an altered acetabular morphology by means of increased or deficient acetabular coverage, might induce asymmetric loading. In our study, patients with bursitis show on average 6◦ higher CEa compared to patients without peritrochanteric edema and distended bursae (group C). Thus, overcoverage of the femoral head may be associated with bursitis and subsequently with GTPS. This could be possibly explained by the assumption that a modified biomechanical balance of the hip joint, imposes excessive stress to the gluteus muscles and tendons, thus resulting to bursal inflammation. In addition we attempted, using MR imaging, to reproduce the results of Fearon et al. [3] who have proposed a relationship between low NSa and GTPS. We were not able to confirm that there is a statistically significant difference between NSa of both groups A and B, when compared to group C. MR imaging measurement of NSa may yield different angle values compared to conventional measurements on plain radiographs. A possible explanation of this, could be that feet are fixed during plain hip radiographs (15◦ internal rotation) whereas during standard hip MR imaging examinations they are randomly positioned. Another explanation for the discrepancy mentioned above could be patients’ selection. In the paper by Fearon et al., women with gynoid distribution of peripheral adiposity were mainly examined, whereas in the present study patients were sequentially examined. The subjects in group C were significantly younger than those in groups A and B. In addition, patients of group A, were on the average only 3 years younger than those of the bursitis group (58.0 vs. 61.2 years) and it seems that the two groups share a similar acetabular morphology. Bursitis possibly represents just a progression of tendinopathy, a hypothesis that is also supported by the fact that we found peritrochanteric edema in all patients with bursitis. Although T2 hyperintensity has been found to be associated with tears [7], our findings are in agreement with the ones by Haliloglu et al. who found poor association of T2 hyperintensity with the clinical symptoms [14]. Moreover, although peritendinous edema has been proposed to suggest early degenerative tendinopathy, edema may also present in-between the greater trochanter and the iliotibial band secondary to external coxa saltans. Thus, edema may be present in the absence of tendinopathy. Peritrochanteric bursitis has been found to be the most common etiology of GTPS in agreement with our findings [2]. Others reported that bursal fluid results from tendon injury and is not the primary source of GTPS [27] or according to others is quite rare [8]. In the present study, a large number of patients with bursitis, experienced pain compared to patients of groups A and
Please cite this article in press as: Klontzas ME, Karantanas AH. Greater trochanter pain syndrome: A descriptive MR imaging study. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.06.009
G Model EURR-6820; No. of Pages 6 6
ARTICLE IN PRESS M.E. Klontzas, A.H. Karantanas / European Journal of Radiology xxx (2014) xxx–xxx
C. However, not all patients of group B had a painful trochanteric region, which highlights the importance of the negative predictive value of 97% of this imaging finding. In other words, absence of bursitis and presence of simple peritrochanteric edema and/or tendinosis suggests that GTPS is unlike. The findings of the present study are in agreement with those of Blankenbaker et al., who found that peritrochanteric edema and tendinosis are present in many patients without GTPS [15]. Our findings are also in agreement with those reported by Haliloglu et al. [14] who showed that T2 hyperintensity is a much more common MR imaging finding than tears. The depiction of gluteus medius tear in only one patient of group B, is not in agreement compared to other reports which show a wide spectrum of prevalence, some suggesting that this is the most common cause of GTPS. Our results differ from those of Bird et al., who found common the tearing of the gluteus medius and rare the presence of bursitis [1]. We agree though with Bird et al, that gluteus medius tendinosis is the most commonly found tendinopathy. In the study by Kingzett-Taylor et al., the partial and complete tears were more common than tendinosis with surgical confirmation in 6 patients [16]. In the same study, no bursitis was found in patients without tendinous tear. However, others found that the gluteal tendon lesions predomimate in the elderly (mean age 83.4 years) females [27]. We did not include patients of this age in our study. Another explanation of this discrepancy is related to the fact that the patients included in the present study were examined with an indication for a painful hip joint and groin and not specifically for exploring GTPS. Finally, we found no association between gluteus medius tendinosis and a positive clinical examination. Indeed, degeneration of gluteus tendons existed in all three groups, both in painful and non painful hips. Our attempt to check whether the presence of GTPS was associated to the size of bursitis or peritrochanteric edema, did not yield statistically significant results. This finding agrees with the results of Blankenbaker et al. [15] who concluded that the size of T2 hyperintensity does not correlate with pain, although they did not actually measure absolute values but rather they applied a 4 scale semiquantitative classification of lesions. A limitation of our study is related to the absence of control group with completely asymptomatic subjects. Inclusion of such a control group might add important information regarding the MR imaging findings and the prevalence of false positive clinical examination. Although the reviewer of the MR imaging scans was unaware of the physical examination findings, the interpretation was not blinded as all patients were symptomatic with various indications regarding the hip joint area. Another limitation of the study is the lack of sagittal images in the routine protocol of the hip, which might have resulted to downgrading of findings, particularly regarding partial tears of the tendons. The size of the lesions was measured once by a single observer and thus interobserver variation has not been evaluated. Finally, the lack of plain radiographic analysis might have missed peri and/or intratendinous calcifications in the context of hydroxyapatite crystal deposition, which might explain the presence of GTPS in the absence of any other findings. In the light of newer surgical techniques for treating the GTPS [28], future longitudinal studies could define the natural course of hip rotator cuff pathology and thus estimate which is the point that marks appearance of pain. This point could be a possible landmark for the onset of treatment. In conclusion, the results of the present study suggest an association between acetabular morphology and the existence of peritrochanteric bursitis, and underline the significance of MR imaging findings of bursitis.
Conflict of interest No conflict of interest to disclose. References [1] Bird PA, Oakley SP, Shnier R, Kirkham BW. Prospective evaluation of magnetic resonance imaging and physical examination findings in patients with greater trochanteric pain syndrome. Arthritis Rheum 2001;44(9):2138–45. [2] Tortolani PJ, Carbone JJ, Quartararo LG. Greater trochanteric pain syndrome in patients referred to orthopedic spine specialists. Spine J 2002;2(4):251–4. [3] Fearon AM, Stephens S, Cook JL, Smith PN, Neeman T, Cormick W, et al. The relationship of femoral neck shaft angle and adiposity to greater trochanteric pain syndrome in women. A case control morphology and anthropometric study. Br J Sports Med 2012;46(12):888–92. [4] Segal NA, Felson DT, Torner JC, Zhu Y, Curtis JR, Niu J, et al. Greater trochanteric pain syndrome: epidemiology and associated factors. Arch Phys Med Rehabil 2007;88(8):988–92. [5] Tibor LM, Sekiya JK. Differential diagnosis of pain around the hip joint. Arthroscopy 2008;24(12):1407–21. [6] Lequesne M, Djian P, Vuillemin V, Mathieu P. Prospective study of refractory greater trochanter pain syndrome. MRI findings of gluteal tendon tears seen at surgery. Clinical and MRI results of tendon repair. Joint Bone Spine 2008;75(4):458–64. [7] Cvitanic O, Henzie G, Skezas N, Lyons J, Minter J. MRI diagnosis of tears of the hip abductor tendons (gluteus medius and gluteus minimus). AJR: Am J Roentgenol 2004;182(1):137–43. [8] Karpinski MR, Piggott H. Greater trochanteric pain syndrome. A report of 15 cases. J Bone Joint Surg Br 1985;67(5):762–3. [9] Yang I, Hayes CW, Biermann JS. Calcific tendinitis of the gluteus medius tendon with bone marrow edema minicking metastatic disease. Skeletal Radiol 2002;31(6):359–61. [10] Sakai T, Shimaoka Y, Sugimoto M, Koizumi T. Acute calcific tendinitis of the gluteus medius: a case report with serial magnetic resonance imaging findings. J Orthop Sci 2004;9(4):404–7. [11] Boutry N, Khalil C, Jaspart M, Marie-Helene V, Demondion X, Cotten A. Imaging of the hip in patients with rheumatic disorders. Eur J Radiol 2007;63(1): 49–58. [12] Mellado JM, Perez del Palomar L, Diaz L, Ramos A, Sauri A. Long-standing Morel–Lavallee lesions of the trochanteric region and proximal thigh. MRI features in five patients. AJR: Am J Roentgenol 2004;182(5):1289–94. [13] Kong A, Van der Vliet A, Zadow S. MRI and US of gluteal tendinopathy in greater trochanteric pain syndrome. Eur Radiol 2007;17(7):1772–83. [14] Haliloglu N, Inceoglu D, Sahin G. Assessment of peritrochanteric high T2 signal depending on the age and gender of the patients. Eur J Radiol 2010;75(1): 64–6. [15] Blankenbaker DG, Ullrick SR, Davis KW, De Smet AA, Haaland B, Fine JP. Correlation of MRI findings with clinical findings of trochanteric pain syndrome. Skeletal Radiol 2008;37(10):903–9. [16] Kingzett-Taylor A, Tirman PF, Feller J, McGann W, Prieto V, Wischer T, et al. Tendinosis and tears of gluteus medius and minimus muscles as a cause of hip pain: MR imaging findings. AJR: Am J Roentgenol 1999;173(4):1123–6. [17] Petchprapa CN, Bencardino JT. Tendon injuries of the hip. Magn Reson Imaging Clin N Am 2013;21(1):75–96. [18] Pfirrmann CW, Notzli HP, Dora C, Hodler J, Zanetti M. Abductor tendons and muscles assessed at MR imaging after total hip arthroplasty in asymptomatic and symptomatic patients. Radiology 2005;235(3):969–76. [19] Hoffmann A, Pfirrmann CWA. The hip abductors at MR imaging. Eur J Radiol 2012;81(12):3755–62. [20] Shbeeb MI, Matteson EL. Trochanteric bursitis (greater trochanter pain syndrome). Mayo Clin Proc 1996;71(6):565–9. [21] Ilizaliturri VM, Camacho-Galindo J, Evia Ramirez AN, Gonzalez Ibarra YL, McMillan S, Busconi BD. Soft tissue pathology around the hip. Clin Sports Med 2011;30(2):391–415. [22] Rasmussen KJ, Fano N. Trochanteric bursitis. Treatment by corticosteroid injection. Scand J Rheumatol 1985;14(4):417–20. [23] Pfirrmann CW, Chung CB, Theumann NH, Trudell DJ, Resnick D. Greater trochanter of the hip: attachment of the abductor mechanism and a complex of three Bursae-MR imaging and MR bursography in cadavers and MR imaging in asymptomatic volunteers. Radiology 2001;221(2):469–77. [24] Dwek J, Pfirrmann C, Stanley A, Pathria M, Chung CB. MR imaging of the hip abductors: normal anatomy and commonly encountered pathology at the greater trochanter. Magn Reson Imaging Clin N Am 2005;13(4):691–704. [25] Chatha DS, Arora R. MR imaging of the normal hip. Magn Reson Imaging Clin N Am 2005;13(4):605–15. [26] Blankenbaker DG, Tuite MJ. The painful hip: new concepts. Skeletal Radiol 2006;35(6):352–70. [27] Bunker TD, Esler CNA, Leach WJ. Rotator-cuff tear of the hip. J Bone Joint Surg Br 1997;79(4):618–20. [28] Barnthouse NC, Wente TM, Voos JE. Greater trochanteric pain syndrome: endoscopic treatment options. Oper Tech Sports Med 2012;20:320–4.
Please cite this article in press as: Klontzas ME, Karantanas AH. Greater trochanter pain syndrome: A descriptive MR imaging study. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.06.009
ARTICLE IN PRESS
EURR-6820; No. of Pages 6
European Journal of Radiology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Greater trochanter pain syndrome: A descriptive MR imaging study Michail E. Klontzas 1 , Apostolos H. Karantanas ∗ Department of Medical Imaging, University Hospital of Heraklion and Radiology Section, Medical School University of Crete, Greece
a r t i c l e
i n f o
Article history: Received 15 January 2014 Received in revised form 8 April 2014 Accepted 16 June 2014 Keywords: Greater trochanter pain syndrome MR imaging CE angle Gluteal tendinopathy Bursitis
a b s t r a c t Objective: Greater trochanter pain syndrome (GTPS) is a diverse clinical entity caused by a variety of underlying conditions. We sought to explore the impact of (1) hip morphology, namely the center-edge angle (CEa) and femoral neck-shaft (NSa) angle, (2) hip abductor tendon degeneration, (3) the dimensions of peritrochanteric edema and (4) bursitis, on the presence of GTPS, using MR imaging. Materials and methods: The presence of pain was prospectively assessed blindly by the senior author. CEa and NSa were blindly measured in 174 hip MR examinations, after completion of the clinical evaluation by another evaluator. The existence and dimensions of T2 hyperintensity of the peritrochanteric soft tissues, the existence and dimensions of bursae, as well as degeneration and tearing of gluteus tendons were also recorded. Results: Out of 174 examinations, 91 displayed peritrochanteric edema (group A) and 34 bursitis, all with peritrochanteric edema (group B). A number of 78 patients from both A and B groups, showed gluteus medius tendon degeneration and one tendon tear. CEa of groups A and B were 6◦ higher than those of normals (group C, P = 0.0038). The mean age of normals was 16.6 years less than in group A and 19.8 years less than in group B (P < 0.0001). Bursitis was associated with pain with a negative predictive value of 97% (P = 0.0003). Conclusion: Acetabular morphology is associated with GTPS and the absence of bursitis was proved to be clinically relevant. Peritrochanteric edema alone was not associated with local pain. © 2014 Published by Elsevier Ireland Ltd.
1. Introduction Greater trochanter pain syndrome (GTPS) is characterized by pain and tenderness over the great trochanter. Its diagnosis is based on a combination of data from medical history, physical examination and imaging findings. Female gender in the sixth decade, a femoral neck-shaft angle (NSa) less than 134◦ , and leg length discrepancies, have been described as risk factors for GTPS [1–3]. A variety of conditions, such as degenerative hip disease, femoroacetabular impingement, femoral head avascular necrosis, infection and conditions that can modify hip biomechanics, such as knee osteoarthritis, iliotibial band syndrome and lumbar spine degenerative disease, can clinically mimic GTPS, making the clinical differential diagnosis extremely complicated [4,5].
∗ Corresponding author at: University of Crete, Department of Medical Imaging, University Hospital, Voutes 71110, Heraklion-Crete, Greece. Tel.: +30 2813392541; fax: +30 2813542095. E-mail addresses: [email protected] (M.E. Klontzas), [email protected], [email protected] (A.H. Karantanas). 1 Address: Souliou 15 Poros, Heraklion, 71307 Crete, Greece. Tel.: +30 6977795314.
Unlike the original belief that the term “GTPS” is synonymous to “trochanteric bursitis”, we nowadays accept that this syndrome may result not only by an inflamed bursa, but also by gluteus tendinopathy/tear and external coxa saltans, which refers to iliotibial band snapping [6–8]. Other rare disorders may also result in GTPS [9–12]. Kong et al. described various findings associated with GTPS, as seen in hip MR imaging examinations and underlined the need of a specific diagnosis in order to optimize the treatment strategies [13]. Haliloglu et al. found that T2 peritrochanteric hyperintensity representing edema is by far the most common finding but is rarely related to clinical symptoms [14]. Various studies have reported association between MR imaging findings and the presence of pain [1,15–17]. In the absence of peritrochanteric hyperintensity on fluid sensitive sequences, the GTPS is an unlike diagnosis [15]. On the other hand, large amounts of fluid within bursae, may correlate with clinical presentation [15,18]. No study, to the best of our knowledge, has compared patients with isolated peritrochanteric T2 hyperintensity in the absence of tendinous tears, with patients demonstrating bursitis. Moreover, no study has evaluated the role of acetabular morphology in patients with GTPS. We sought to assess the relation of acetabular morphology and NSa with the presence of GTPS and to evaluate the association
http://dx.doi.org/10.1016/j.ejrad.2014.06.009 0720-048X/© 2014 Published by Elsevier Ireland Ltd.
Please cite this article in press as: Klontzas ME, Karantanas AH. Greater trochanter pain syndrome: A descriptive MR imaging study. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.06.009
G Model
ARTICLE IN PRESS
EURR-6820; No. of Pages 6
M.E. Klontzas, A.H. Karantanas / European Journal of Radiology xxx (2014) xxx–xxx
2
between peritrochanteric edema and bursitis on MR imaging and the existence of pain. Secondly, we evaluated the impact of hip abductor tendon degeneration on the presence of pain. Finally, we explored whether the dimensions of peritrochanteric edema and bursitis, could imply the presence of a painful hip. 2. Materials and methods 2.1. Patients A total of 224 consecutive MR imaging examinations of hip joints from 141 patients, referred to our department for various clinical indications from June 2008 to April 2013, were prospectively evaluated. The most common indications included early and radiologically occult osteoarthritis, trochanteritis, early osteonecrosis, transient bone marrow edema, and labral abnormalities. Our study has been approved by our hospital’s ethics committee, was performed in the context of the principles of Helsinki Declaration and all patients have signed an informed consent. Patients with severe degenerative osteoarthritis of the hip and/or the knee joints (by means of presence of marginal osteophytes and subarticular geodes), previous septic or inflammatory arthritis, tumors around hip joints, previous hip surgery or radiotherapy, advanced femoral head osteonecrosis (ARCO grade III or more) as well as patients who had local corticosteroid injections, were excluded from our study. Exclusion was made both by imaging (plain radiographs and/or MR imaging) and clinical means. The 1.5 mm joint space was considered to represent advanced osteoarthritis. In total, 174 hip joint examinations from 92 patients (mean age 53.99 ± 15.95 years) were finally included. The population of our study consisted of 26 male and 66 female patients.
Fig. 1. A 55-year-old, asymptomatic for trochanteric pain, female patient. The coronal STIR image shows peritrochanteric edema (arrows).
on fluid sensitive sequences within the tendon together with attenuation or thinning of the tendon. Discontinuity of the tendon, with or without osseous avulsion, was diagnostic of full thickness tear [19]. 2.4. Measurements In order to assess the acetabular morphology and the morphology of the upper third of the femur, we performed two measurements (Fig. 5). The center-edge angle (CEa) was used in order to evaluate the coverage of the femoral head by the acetabulum. This was done by drawing a circle around both the femoral heads in order to find their center. Then, the center of each femoral head was used as the tip of an angle, with one side perpendicular to the line that crosses the centers of the two
2.2. Imaging All examinations were performed at 1.5T (Vision Hybrid, Siemens, Erlangen, Germany) using a phased-array flexible torso coil. From the standard hip MR imaging protocol utilized in our department, the following 4 sequences were used in the study: Coronal T1 weighted (w) turbo spin-echo (TSE) images (TR/TE: 532/13; slice thickness: 4 mm; matrix: 512 × 256; FOV: 37.5 cm; ETL: 3), axial T1-w TSE images (TR/TE: 511/12; slice thickness: 4 mm; matrix: 320 × 320; FOV: 40 cm; ETL: 3), coronal T1-w turbo inversion recovery (STIR) images (TR/TE/TI: 4960/53/170; slice thickness: 4 mm; matrix: 320 × 288; FOV: 40 cm; ETL: 5), and axial fat suppressed T2-w TSE images (TR/TE: 3250/80; slice thickness: 4 mm; matrix: 280 × 320; FOV: 40 cm × 35 cm; ETL: 5). Fat suppression was achieved with spectral presaturation. 2.3. MR imaging findings classification All MR imaging examinations were blindly evaluated by the first author, after completion of the clinical examination. The study population was categorized in three distinct groups: group A presenting with peritrochanteric edema, assessed as a poorly defined area of high signal intensity on fluid sensitive sequences, seen on both axial and coronal images (Fig. 1); group B showing distention of at least one of the region’s bursae, namely trochanteric or subgluteus maximus, subgluteus medius and subgluteus minimus (Figs. 2–4) assessed on two planes of fluid sensitive sequences; group C including examinees without peritrochanteric edema or bursal distention. Tendon degeneration and/or partial or complete tear was recorded in all MR imaging examinations. Tendinopathy – tendon degeneration was diagnosed when there was abnormally increased signal within the tendon on T1-w sequences, with or without thickening of the tendon (Fig. 3). Partial tearing of the tendon was diagnosed when increased signal intensity was depicted
Fig. 2. A 70-year-old female patient with greater trochanter pain syndrome on the left. (A) The coronal STIR MR image shows a large trochanteric bursa (arrows). (B) The axial fat suppressed T2-w MR image shows the trochanteric bursa (open arrow) and in addition bilateral iliopsoas bursae (arrows).
Please cite this article in press as: Klontzas ME, Karantanas AH. Greater trochanter pain syndrome: A descriptive MR imaging study. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.06.009
G Model EURR-6820; No. of Pages 6
ARTICLE IN PRESS M.E. Klontzas, A.H. Karantanas / European Journal of Radiology xxx (2014) xxx–xxx
3
Fig. 3. A 56-year-old female patient with right sided greater trochanter pain syndrome. (A) The axial T1-w MR image shows thickening and abnormally high signal of the right gluteus minimus tendon (arrow). The contralateral tendon on the left is normal (open arrow). (B) The axial fat suppressed T2-w MR image shows subgluteus minimus bursa (arrow), subgluteus medius bursa (thick arrow) and abnormal intratendinous signal within the thickened gluteus minimus tendon (open arrow). The coronal STIR MR images show subgluteus minimus bursa (arrow in C) and subgluteus medius bursa (arrow in D). Peritrochanteric edema is shown on the left side (small arrows in D).
femoral heads and the other side crossing the lateral edge of the acetabulum. In addition, the NSa was measured in an attempt to confirm the association between the morphology of the proximal femur and the development of GTPS. CEa and NSa were measured on mid-coronal T1-w TSE MR images. NSa and CEa of 20 randomly chosen hips, were measured twice in order to assess the intra-rater reliability by means of intra-class correlation coefficient (ICC) and coefficient of variation (CV) for duplicate measurements. The dimensions (anterior–posterior, superior–inferior, left–right) of the peritrochanteric edema and of the distended bursae, were also measured on the fluid sensitive sequences. The highest value of those dimensions was used for our analysis.
2.5. Clinical data The senior author of the study, a radiologist with a 27-year experience on musculoskeletal disorders, prospectively and without knowing the clinical indication for the MR imaging scan, performed a thorough clinical examination on site, just before undergoing the MR imaging examination. The clinical tests included: (a) pain during active abduction and internal rotation, (b) a positive Trendelenburg’s sign and (c) pain on pressure over the greater trochanter with the patient in the side-lying position. The presence of lateral hip pain together with at least one positive clinical test, were the prerequisites for diagnosing the presence of GTPS [1,20]. The pain was recorded as either present or absent. The evaluator of the MR imaging scans was blinded to all data recorded by the senior author.
2.6. Statistical analysis Data were analyzed using MedCalc 12.7.1 statistical software (MedCalc Software, Belgium). Our data were non-Gaussian and our three unmatched groups included outliers. Thus, comparison between CEa and NSa as well as patients’ age was performed using the non-parametric Kruskal–Wallis test. Normality was tested graphically by histogram inspection and numerically using the Shapiro–Wilk test. Fisher’s exact test was used to determine the association between MR imaging findings (peritrochanteric edema or bursitis) and clinically overt pain, as well as the association between the existence of tendon degeneration and pain. In addition, Mann–Whitney U test was used to analyze the association between the dimensions of peritendinous edema and bursitis with pain and those without pain. Positive predictive value (PPV) and negative predictive value (NPV) of bursitis with regard to pain were calculated. True negative cases were patients without bursitis and GTPS, false negative cases were patients who presented with GTPS but no overt bursitis on MR imaging, true positive cases were patients with both bursitis and GTPS and false positive were patients with bursitis but no clinical signs of GTPS. Results were regarded as statistically significant with a P-value of less than 0.05.
3. Results Data analysis revealed that 91 (52.3%) hips displayed peritrochanteric T2 hyperintensity corresponding to soft tissue edema. Thirty-four (19.5%) hips had a distended bursa (18 trochanteric,
Please cite this article in press as: Klontzas ME, Karantanas AH. Greater trochanter pain syndrome: A descriptive MR imaging study. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.06.009
G Model EURR-6820; No. of Pages 6 4
ARTICLE IN PRESS M.E. Klontzas, A.H. Karantanas / European Journal of Radiology xxx (2014) xxx–xxx
Fig. 6. Age distribution in all 3 groups of examined patients.
degeneration and 1 gluteus medius tendon full tear with atrophy and fatty infiltration of the muscle. There were 44 degenerated tendons in group A, 15 in group B and 19 in group C.
3.1. Patient’s age
Fig. 4. A 25-year-old male patient with ankylosing spondylitis and no indication of greater trochanter pain syndrome. The axial fat suppressed T2-w (A) and coronal STIR (B) MR images, show bilateral trochanteric bursae (arrows). Note the subarticular edema in the right sacroiliac joint (open arrow in B).
A statistically significant difference was found between the ages of the three groups (P < 0.0001). Subjects in group C were 16.6 years younger (mean age 41.4 years old – 95%CI from 37.3 to 45.5 years) compared to group A (mean age 58.0 years old – 95%CI from 55.4 to 60.7 years) and 19.8 years younger than group B (mean age 61.2 years old – 95%CI from 55.6 to 66.8 years) (Fig. 6).
3.2. Acetabular morphology 9 subgluteus medius, 6 trochanteric and subgluteus medius and 1 subgluteus minimus and medius). All the 34 hips of group B demonstrated associated peritrochanteric edema. In 49 hips, there were neither peritrochanteric edema nor bursitis. In addition, 78 out of 174 (44.8%) hips displayed gluteus medius tendon
The CEa of normal subjects were significantly different compared to those measured in groups B and A (P = 0.0038). Hips from patients in group C had a mean CEa of 36.7◦ (95%CI from 34.2◦ to 39.2◦ ), compared to 41.0◦ of group A (95%CI from 39.0◦ to 43.0◦ ) and 42.6◦ of group B (95%CI from 39.6◦ to 45.5◦ ) (Fig. 7).
Fig. 5. Measurements of the center-edge angle of Wiberg (CEa) and neck-shaft angle (NSa) on a mid-coronal T1-w MR image.
Please cite this article in press as: Klontzas ME, Karantanas AH. Greater trochanter pain syndrome: A descriptive MR imaging study. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.06.009
G Model
ARTICLE IN PRESS
EURR-6820; No. of Pages 6
M.E. Klontzas, A.H. Karantanas / European Journal of Radiology xxx (2014) xxx–xxx
5
subjects with GTPS (mean 2.4 cm, 95%CI from 2.0 to 2.9 cm) and those without (mean 2.9 cm, 95%CI from 2.75 to 3.1) (P = 0.1545). 4. Discussion
Fig. 7. Association of the center to edge angles in all 3 groups of patients.
Table 1 Association between bursitis and GPTS.a P = 0.0003
GTPS
No GTPS
Total
8 4
26 136
34 140
12
162
174
Bursitis No bursitis Total a
GTPS, greater trochanter pain syndrome.
3.3. Femoral morphology No statistically significant difference (P = 0.4260) was found between the NSa of hips in group C (mean NSa 125.8◦ , 95%CI from 123.1◦ to 128.5◦ ) when compared to group A (mean NSa 124.9◦ , 95% from 123.4◦ to 126.4◦ ) and group B (mean NSa 123.4◦ , 95%CI from 120.1◦ to 126.6◦ ). Moreover, the intra-rater reliability was high, with a CV for duplicate measurements of 3.2% and 6.5% for NSa and CEa measurements respectively. In addition, the ICC was 0.81 (95%CI from 0.58 to 0.92) and 0.9 (95%CI from 0.77 to 0.96) for NSa and CEa measurements respectively. 3.4. Positive tests for GTPS A statistically significant association between the existence of bursitis and the presence of GTPS (P = 0.0003) was found. Eight out of 34 patients with bursitis experienced pain, whereas only 4 of those without bursitis reported clinical symptoms, which yields a negative predictive value of 97.14% (95%CI from 92.85% to 99.22%) and a positive predicted value of 23.53% (95%CI from 10.75% to 41.17%) (Table 1). The presence of gluteus medius degeneration was not associated with pain, as only 8 out of 79 patients with degeneration (including the one with tear) experienced pain when clinically examined and 4 out of 95 patients without degeneration had painful hips (P = 0.1437). Analysis of association between the presence of pain and either the CEa or NSa did not show any statistical significance. 3.5. Dimensions No statistically significant difference was found between the maximum bursal size in symptomatic (mean 2.5 cm, 95%CI from 1.3 to 3.7) and asymptomatic (mean 1.9 cm, 95%CI from 1.4 to 2.4) subjects (P = 0.2496). Likewise, the maximum dimensions of peritrochanteric T2 hyperintensity did not differ significantly between
Our study showed that a relationship exists between acetabular morphology and the presence of peritrochanteric bursitis. In addition, the NPV of bursitis on MR imaging, is a clinically important finding. GTPS is a common cause of lateral hip pain with a broad differential diagnosis and is defined as pain and tenderness to palpation over the greater trochanter [21,22]. The anatomy of the trochanteric region is extremely complex, including the tendons of gluteus muscles, the piriformis and obturator tendons and several bursae, the most important of which are the subgluteus medius and subgluteus minimus bursae, as well as the trochanteric bursa [19,23–26]. Peritrochanteric pain may result from impingement of the iliotibial band, tendinopathy/tendon tearing and inflammation of bursae, either primary or secondary [13,15,21]. No study, to the best of our knowledge, has proposed a possible association between the acetabular morphology and GTPS. Our hypothesis was that an altered acetabular morphology by means of increased or deficient acetabular coverage, might induce asymmetric loading. In our study, patients with bursitis show on average 6◦ higher CEa compared to patients without peritrochanteric edema and distended bursae (group C). Thus, overcoverage of the femoral head may be associated with bursitis and subsequently with GTPS. This could be possibly explained by the assumption that a modified biomechanical balance of the hip joint, imposes excessive stress to the gluteus muscles and tendons, thus resulting to bursal inflammation. In addition we attempted, using MR imaging, to reproduce the results of Fearon et al. [3] who have proposed a relationship between low NSa and GTPS. We were not able to confirm that there is a statistically significant difference between NSa of both groups A and B, when compared to group C. MR imaging measurement of NSa may yield different angle values compared to conventional measurements on plain radiographs. A possible explanation of this, could be that feet are fixed during plain hip radiographs (15◦ internal rotation) whereas during standard hip MR imaging examinations they are randomly positioned. Another explanation for the discrepancy mentioned above could be patients’ selection. In the paper by Fearon et al., women with gynoid distribution of peripheral adiposity were mainly examined, whereas in the present study patients were sequentially examined. The subjects in group C were significantly younger than those in groups A and B. In addition, patients of group A, were on the average only 3 years younger than those of the bursitis group (58.0 vs. 61.2 years) and it seems that the two groups share a similar acetabular morphology. Bursitis possibly represents just a progression of tendinopathy, a hypothesis that is also supported by the fact that we found peritrochanteric edema in all patients with bursitis. Although T2 hyperintensity has been found to be associated with tears [7], our findings are in agreement with the ones by Haliloglu et al. who found poor association of T2 hyperintensity with the clinical symptoms [14]. Moreover, although peritendinous edema has been proposed to suggest early degenerative tendinopathy, edema may also present in-between the greater trochanter and the iliotibial band secondary to external coxa saltans. Thus, edema may be present in the absence of tendinopathy. Peritrochanteric bursitis has been found to be the most common etiology of GTPS in agreement with our findings [2]. Others reported that bursal fluid results from tendon injury and is not the primary source of GTPS [27] or according to others is quite rare [8]. In the present study, a large number of patients with bursitis, experienced pain compared to patients of groups A and
Please cite this article in press as: Klontzas ME, Karantanas AH. Greater trochanter pain syndrome: A descriptive MR imaging study. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.06.009
G Model EURR-6820; No. of Pages 6 6
ARTICLE IN PRESS M.E. Klontzas, A.H. Karantanas / European Journal of Radiology xxx (2014) xxx–xxx
C. However, not all patients of group B had a painful trochanteric region, which highlights the importance of the negative predictive value of 97% of this imaging finding. In other words, absence of bursitis and presence of simple peritrochanteric edema and/or tendinosis suggests that GTPS is unlike. The findings of the present study are in agreement with those of Blankenbaker et al., who found that peritrochanteric edema and tendinosis are present in many patients without GTPS [15]. Our findings are also in agreement with those reported by Haliloglu et al. [14] who showed that T2 hyperintensity is a much more common MR imaging finding than tears. The depiction of gluteus medius tear in only one patient of group B, is not in agreement compared to other reports which show a wide spectrum of prevalence, some suggesting that this is the most common cause of GTPS. Our results differ from those of Bird et al., who found common the tearing of the gluteus medius and rare the presence of bursitis [1]. We agree though with Bird et al, that gluteus medius tendinosis is the most commonly found tendinopathy. In the study by Kingzett-Taylor et al., the partial and complete tears were more common than tendinosis with surgical confirmation in 6 patients [16]. In the same study, no bursitis was found in patients without tendinous tear. However, others found that the gluteal tendon lesions predomimate in the elderly (mean age 83.4 years) females [27]. We did not include patients of this age in our study. Another explanation of this discrepancy is related to the fact that the patients included in the present study were examined with an indication for a painful hip joint and groin and not specifically for exploring GTPS. Finally, we found no association between gluteus medius tendinosis and a positive clinical examination. Indeed, degeneration of gluteus tendons existed in all three groups, both in painful and non painful hips. Our attempt to check whether the presence of GTPS was associated to the size of bursitis or peritrochanteric edema, did not yield statistically significant results. This finding agrees with the results of Blankenbaker et al. [15] who concluded that the size of T2 hyperintensity does not correlate with pain, although they did not actually measure absolute values but rather they applied a 4 scale semiquantitative classification of lesions. A limitation of our study is related to the absence of control group with completely asymptomatic subjects. Inclusion of such a control group might add important information regarding the MR imaging findings and the prevalence of false positive clinical examination. Although the reviewer of the MR imaging scans was unaware of the physical examination findings, the interpretation was not blinded as all patients were symptomatic with various indications regarding the hip joint area. Another limitation of the study is the lack of sagittal images in the routine protocol of the hip, which might have resulted to downgrading of findings, particularly regarding partial tears of the tendons. The size of the lesions was measured once by a single observer and thus interobserver variation has not been evaluated. Finally, the lack of plain radiographic analysis might have missed peri and/or intratendinous calcifications in the context of hydroxyapatite crystal deposition, which might explain the presence of GTPS in the absence of any other findings. In the light of newer surgical techniques for treating the GTPS [28], future longitudinal studies could define the natural course of hip rotator cuff pathology and thus estimate which is the point that marks appearance of pain. This point could be a possible landmark for the onset of treatment. In conclusion, the results of the present study suggest an association between acetabular morphology and the existence of peritrochanteric bursitis, and underline the significance of MR imaging findings of bursitis.
Conflict of interest No conflict of interest to disclose. References [1] Bird PA, Oakley SP, Shnier R, Kirkham BW. Prospective evaluation of magnetic resonance imaging and physical examination findings in patients with greater trochanteric pain syndrome. Arthritis Rheum 2001;44(9):2138–45. [2] Tortolani PJ, Carbone JJ, Quartararo LG. Greater trochanteric pain syndrome in patients referred to orthopedic spine specialists. Spine J 2002;2(4):251–4. [3] Fearon AM, Stephens S, Cook JL, Smith PN, Neeman T, Cormick W, et al. The relationship of femoral neck shaft angle and adiposity to greater trochanteric pain syndrome in women. A case control morphology and anthropometric study. Br J Sports Med 2012;46(12):888–92. [4] Segal NA, Felson DT, Torner JC, Zhu Y, Curtis JR, Niu J, et al. Greater trochanteric pain syndrome: epidemiology and associated factors. Arch Phys Med Rehabil 2007;88(8):988–92. [5] Tibor LM, Sekiya JK. Differential diagnosis of pain around the hip joint. Arthroscopy 2008;24(12):1407–21. [6] Lequesne M, Djian P, Vuillemin V, Mathieu P. Prospective study of refractory greater trochanter pain syndrome. MRI findings of gluteal tendon tears seen at surgery. Clinical and MRI results of tendon repair. Joint Bone Spine 2008;75(4):458–64. [7] Cvitanic O, Henzie G, Skezas N, Lyons J, Minter J. MRI diagnosis of tears of the hip abductor tendons (gluteus medius and gluteus minimus). AJR: Am J Roentgenol 2004;182(1):137–43. [8] Karpinski MR, Piggott H. Greater trochanteric pain syndrome. A report of 15 cases. J Bone Joint Surg Br 1985;67(5):762–3. [9] Yang I, Hayes CW, Biermann JS. Calcific tendinitis of the gluteus medius tendon with bone marrow edema minicking metastatic disease. Skeletal Radiol 2002;31(6):359–61. [10] Sakai T, Shimaoka Y, Sugimoto M, Koizumi T. Acute calcific tendinitis of the gluteus medius: a case report with serial magnetic resonance imaging findings. J Orthop Sci 2004;9(4):404–7. [11] Boutry N, Khalil C, Jaspart M, Marie-Helene V, Demondion X, Cotten A. Imaging of the hip in patients with rheumatic disorders. Eur J Radiol 2007;63(1): 49–58. [12] Mellado JM, Perez del Palomar L, Diaz L, Ramos A, Sauri A. Long-standing Morel–Lavallee lesions of the trochanteric region and proximal thigh. MRI features in five patients. AJR: Am J Roentgenol 2004;182(5):1289–94. [13] Kong A, Van der Vliet A, Zadow S. MRI and US of gluteal tendinopathy in greater trochanteric pain syndrome. Eur Radiol 2007;17(7):1772–83. [14] Haliloglu N, Inceoglu D, Sahin G. Assessment of peritrochanteric high T2 signal depending on the age and gender of the patients. Eur J Radiol 2010;75(1): 64–6. [15] Blankenbaker DG, Ullrick SR, Davis KW, De Smet AA, Haaland B, Fine JP. Correlation of MRI findings with clinical findings of trochanteric pain syndrome. Skeletal Radiol 2008;37(10):903–9. [16] Kingzett-Taylor A, Tirman PF, Feller J, McGann W, Prieto V, Wischer T, et al. Tendinosis and tears of gluteus medius and minimus muscles as a cause of hip pain: MR imaging findings. AJR: Am J Roentgenol 1999;173(4):1123–6. [17] Petchprapa CN, Bencardino JT. Tendon injuries of the hip. Magn Reson Imaging Clin N Am 2013;21(1):75–96. [18] Pfirrmann CW, Notzli HP, Dora C, Hodler J, Zanetti M. Abductor tendons and muscles assessed at MR imaging after total hip arthroplasty in asymptomatic and symptomatic patients. Radiology 2005;235(3):969–76. [19] Hoffmann A, Pfirrmann CWA. The hip abductors at MR imaging. Eur J Radiol 2012;81(12):3755–62. [20] Shbeeb MI, Matteson EL. Trochanteric bursitis (greater trochanter pain syndrome). Mayo Clin Proc 1996;71(6):565–9. [21] Ilizaliturri VM, Camacho-Galindo J, Evia Ramirez AN, Gonzalez Ibarra YL, McMillan S, Busconi BD. Soft tissue pathology around the hip. Clin Sports Med 2011;30(2):391–415. [22] Rasmussen KJ, Fano N. Trochanteric bursitis. Treatment by corticosteroid injection. Scand J Rheumatol 1985;14(4):417–20. [23] Pfirrmann CW, Chung CB, Theumann NH, Trudell DJ, Resnick D. Greater trochanter of the hip: attachment of the abductor mechanism and a complex of three Bursae-MR imaging and MR bursography in cadavers and MR imaging in asymptomatic volunteers. Radiology 2001;221(2):469–77. [24] Dwek J, Pfirrmann C, Stanley A, Pathria M, Chung CB. MR imaging of the hip abductors: normal anatomy and commonly encountered pathology at the greater trochanter. Magn Reson Imaging Clin N Am 2005;13(4):691–704. [25] Chatha DS, Arora R. MR imaging of the normal hip. Magn Reson Imaging Clin N Am 2005;13(4):605–15. [26] Blankenbaker DG, Tuite MJ. The painful hip: new concepts. Skeletal Radiol 2006;35(6):352–70. [27] Bunker TD, Esler CNA, Leach WJ. Rotator-cuff tear of the hip. J Bone Joint Surg Br 1997;79(4):618–20. [28] Barnthouse NC, Wente TM, Voos JE. Greater trochanteric pain syndrome: endoscopic treatment options. Oper Tech Sports Med 2012;20:320–4.
Please cite this article in press as: Klontzas ME, Karantanas AH. Greater trochanter pain syndrome: A descriptive MR imaging study. Eur J Radiol (2014), http://dx.doi.org/10.1016/j.ejrad.2014.06.009