Fusion of positron emission tomography/computed tomography with magnetic resonance imaging reveals hamstring peritendonitis in polymyalgia rheumatica

Fusion of positron emission tomography/computed tomography with magnetic resonance imaging... Abstract Objectives To characterize 18F-fluorodeoxyglucose (18F-FDG) uptake on whole-body PET/CT in PMR, and identify its precise anatomic correlate using MRI. Methods Patients with newly diagnosed PMR according to the 2012 EULAR/ACR classification criteria were prospectively recruited. Participants with GCA were excluded. A whole-body 18F-FDG PET/CT scan was performed in all untreated patients. Qualitative and semiquantitative [standardized uptake value maximum (SUVmax)] scoring of abnormal 18F-FDG uptake was undertaken. MRI of the pelvis, knee and wrist and hand was performed in three representative patients with anatomical correlation of FDG-avid sites carried out using Medview fusion software. Results Twenty-two patients with PMR were recruited. Their mean age was 68.3 years (s.d. 6.3) and 13/22 were male. On whole-body PET/CT, 18F-FDG uptake adjacent to the ischial tuberosities was observed in 21 participants (95.4%) and recorded the highest mean SUVmax value [3.6 (s.d. 1.7)]. A high frequency of posteromedial knee (61.9%) and wrist and/or hand involvement (66.7%) was also appreciated. MRI of the pelvis revealed high T2 signal surrounding the proximal hamstring tendon origins of both semimembranosus and the conjoint tendon of the semitendinosus and biceps femoris. At the knee, peritendonitis at the distal insertion of the semimembranosus was observed. PET/MRI fusion at the pelvis and knee confirmed semimembranosus peritendonitis as the anatomical correlate of 18F-FDG uptake adjacent to the ischial tuberosities and of posteromedial knee structures. Conclusion Hamstring peritendonitis is a common and distinctive manifestation of PMR on whole-body PET/CT. Trial registration Australian New Zealand Clinical Trials Registry, http://www.anzctr.org.au, ACTRN1261400696695. polymyalgia rheumatica, whole-body positron emission tomography/computed tomography, magnetic resonance imaging, hamstring, peritendonitis Rheumatology key messages PMR exhibits a distinctive pattern of abnormal 18F-FDG uptake on whole-body PET/CT. Hamstring peritendonitis correlates with 18F-FDG uptake adjacent to the ischial tuberosities and of posteromedial knee structures. Frequent wrist and/or hand involvement is observed on whole-body PET/CT in PMR patients. Introduction PMR is an inflammatory condition predominantly characterized by subacute-onset shoulder and hip pain and stiffness in patients ⩾50 years of age. Some controversy has existed as to whether PMR represents a distinct disease entity due to its close relationship with both GCA and late-onset RA, however, imaging modalities such as PET have recently uncovered its distinctive pathology [1–3]. Blockmans et al. [4] first reported the ability of PET to detect articular inflammation in PMR, finding abnormal 18F-fluorodeoxyglucose (18F-FDG) uptake in the shoulders and hips in 94 and 89% of patients, respectively. They additionally described a novel finding involving the processi spinosi of the vertebrae, which has since been localized to the interspinous bursa on MRI [5, 6]. The pairing of whole-body PET with low-dose CT has facilitated further study of the anatomic distribution of PMR. 18F-FDG uptake in the trochanteric region and adjacent to the ischial tuberosities has been attributed to trochanteric and ischiogluteal bursitis, respectively [7]. A high frequency of knee involvement due to capsulitis has similarly been reported [8]. However, the low resolution of whole-body PET/CT images remains a limitation to precise identification of anatomic correlates of abnormal 18F-FDG uptake. Peripheral joint involvement is appreciated clinically in almost half of PMR patients [9]. Distinguishing this group from late-onset RA at presentation is especially difficult. However, it is increasingly apparent from imaging studies that the prevailing pathology in PMR is extra-articular rather than intra-articular [10]. Unlike RA, a predilection for tendinous and bursal structures combined with capsular joint involvement is typically observed. In this study we sought to further characterize the pattern of 18F-FDG uptake on whole-body PET/CT in untreated patients with newly diagnosed PMR. Utilizing dedicated hand views, additional emphasis was also placed upon the distribution and frequency of peripheral joint involvement. MRI was subsequently performed in representative patients before fusion technology was employed to precisely identify anatomic correlates of abnormal 18F-FDG uptake. Methods Patients Patients with newly diagnosed PMR according to the 2012 EULAR/ACR classification criteria (utilizing the clinical scoring algorithm) [9] were prospectively recruited from primary care, community and hospital rheumatology practices. Participants who had received prior prednisolone treatment and those with symptoms suggestive of GCA (headache, scalp tenderness, jaw claudication or visual disturbance) were excluded, as were patients with a history of cancer within the past 5 years, neuromuscular disease, active infection, other inflammatory conditions and chronic pain syndromes. The study was approved by the Austin Health Human Research Ethics Committee prior to commencement and registered with the Australian New Zealand Clinical Trials Registry (trial identification ACTRN1261400696695). Clinical and laboratory variables Following appropriate written consent, demographic and clinical data were collected. Participants also completed the HAQ disability index (HAQ-DI). A standard physical examination was carried out with specific focus on the presence of peripheral synovitis. Baseline laboratory testing comprised full blood count, CRP, ESR, creatinine kinase, thyroid function testing, RF and ACPAs. Bilateral wrist and hand X-rays were performed to assess for erosive change. B-mode and power Doppler US was undertaken at both wrists by a rheumatologist experienced in US (C.E.O.) using a LOGIQ e machine (GE Healthcare, Chicago, IL, USA) with linear 5–13 MHz transducer. Whole-body PET/CT A whole body 18F-FDG PET/CT scan (Gemini-TF64 or Ingenuity-TF128 PET/CT system; Phillips, Cleveland, OH, USA) from skull vertex to toes (including dedicated hand views) was performed within 7 days of the screening visit and prior to the commencement of prednisolone treatment. Following a minimum 6 h fast, participants received a weight-based i.v. injection of 18F-FDG. The emission study and low-dose CT was acquired after a 1 h rest period, with CT attenuation correction performed on all whole-body and dedicated hand images prior to independent qualitative and semiquantitative analysis of abnormal 18F-FDG uptake by two experienced nuclear medicine physicians (A.P. and S.T.L.). Any significant differences in results were resolved by consensus. Qualitative analysis involved visual evaluation of 18F-FDG uptake using the following scoring system: 0, no uptake; 1, lower than normal liver uptake; 2, similar to normal liver uptake; 3, higher than normal liver uptake. A score ⩾1 was considered a positive result. For semiquantitative analysis, a region of interest was drawn around each site and the maximum standardized uptake value (SUVmax) calculated. The following sites were specifically assessed: two shoulder joints, two hip joints, two trochanteric regions, regions adjacent to the two ischial tuberosities, most FDG-avid interspinous bursa, two knee joints, two wrist joints and two hands. MRI correlation and fusion of PET/CT with MRI To precisely identify anatomic correlates of abnormal 18F-FDG uptake, MRI was performed in three representative patients at the pelvis, knee and wrist and hand using a 1.5 T magnet (Siemens Avanto, Erlangen, Germany) with a Siemens i-PAT compatible six-element design phased array body matrix, Siemens 12 channel transmit/receive configuration knee coil and Invivo 8 channel wrist coil, respectively. For imaging of the pelvis, non-contrast T1-weighted fast spin-echo (FSE) sequences in the axial plane, proton density-weighted FSE sequences in the axial, sagittal and coronal planes with fat saturation in the coronal plane and T2-weighted FSE sequences with fat saturation in the axial plane centred on the ischial tuberosities were obtained. For imaging of the knee, proton density-weighted FSE and T2-weighted FSE sequences with fat saturation in the coronal plane, proton density-weighted FSE with fat saturation in the sagittal plane and proton density-weighted FSE and T2-weighted FSE images with fat saturation in the axial plane were obtained. For imaging of the wrist and hand, the following sequences were performed before i.v. injection: proton density-weighted FSE in the axial, coronal and sagittal planes with fat saturation in the coronal plane; T2-weighted FSE in the axial plane with fat suppression and 3D T2*-weighted GRE in the coronal plane. Following i.v. injection of 10 ml of gadolinium-based contrast agent (Dotarem, gadoterate meglumine, Guerbet USA, Princeton, NJ, USA), T1-weighted FSE fat suppression images in the axial and coronal planes were obtained. Interpretation of MRIs was performed by two experienced musculoskeletal radiologists (L.Y., R.Z.). Medview fusion software (Ann Arbor, MI, USA) was utilized to align and superimpose PET and MRI datasets and to anatomically correlate sites of abnormal 18F-FDG uptake adjacent to the ischial tuberosities and of posteromedial knee structures. Statistical analysis Results for continuous variables are reported as mean (s.d.) or median [interquartile range (IQR)]. Categorical variables are represented as proportions (percentages). All statistical analyses were undertaken using Stata 13.0 (StataCorp, College Station, TX, USA). Results Patient demographics, clinical and laboratory features Twenty-two patients met the inclusion criteria for a diagnosis of PMR [mean clinical algorithm score 5.1 (s.d. 0.7)]. Their demographic data are summarized in Table 1. The mean age at diagnosis was 68.3 years (s.d. 6.3). Thirteen patients (59.1%) were male and all participants were Caucasian. Table 1 Baseline demographic data and clinical features Demographic or clinical features  Value  Age, mean (s.d.), years  68.3 (6.3)  Male, n (%)  13 (59.1)  Caucasian, n (%)  22 (100)  BMI, mean (s.d.), kg/m2  28.6 (5.1)  Ever smoked, n (%)  10 (45.4)  Shoulder pain, n (%)  22 (100)  Hip pain, n (%)  18 (81.8)  Back pain, n (%)  13 (59.1)  Knee pain, n (%)  14 (63.6)  Wrist/hand pain, n (%)  12 (54.5)  Morning stiffness >45 min, n (%)  18 (81.8)  Median CRP, median (IQR), mg/l  40.8 (24.6–67.1)  Median ESR, median (IQR), mm  45.5 (29–64)  PMR-activity score, mean (s.d.)  71.3 (38.8)  Median HAQ-DI score, median (IQR)  2 (1.4–2.2)  Demographic or clinical features  Value  Age, mean (s.d.), years  68.3 (6.3)  Male, n (%)  13 (59.1)  Caucasian, n (%)  22 (100)  BMI, mean (s.d.), kg/m2  28.6 (5.1)  Ever smoked, n (%)  10 (45.4)  Shoulder pain, n (%)  22 (100)  Hip pain, n (%)  18 (81.8)  Back pain, n (%)  13 (59.1)  Knee pain, n (%)  14 (63.6)  Wrist/hand pain, n (%)  12 (54.5)  Morning stiffness >45 min, n (%)  18 (81.8)  Median CRP, median (IQR), mg/l  40.8 (24.6–67.1)  Median ESR, median (IQR), mm  45.5 (29–64)  PMR-activity score, mean (s.d.)  71.3 (38.8)  Median HAQ-DI score, median (IQR)  2 (1.4–2.2)  Consistent with the 2012 EULAR/ACR classification criteria, all patients possessed bilateral shoulder pain, with 18/22 (81.8%) also reporting concomitant hip pain. Fourteen participants (63.6%) indicated involvement of their knees, while 12 (54.5%) reported wrist and/or hand symptoms. The median duration of early morning stiffness was 120 min (IQR 60–120). On clinical examination, synovitis at the wrist and/or hand was detected in nine patients (40.9%) and six (27.3%) exhibited swelling and/or tenderness of the knee. All participants were negative for ACPAs. Three patients returned a low positive RF result [14 IU/ml in two participants and 17 IU/ml in the third (n < 14 IU/ml)]. A single patient was found to possess a subtle marginal erosion of the right fourth metacarpal head on bilateral wrist and hand X-rays. Using US, greyscale synovitis was detected at the wrist in eight patients (36.4%), five (22.7%) of whom had associated power Doppler signal. Using the PMR activity score, disease activity was judged to be high [mean 71.3 (s.d. 38.8)]. Participants reported a severe level of disability because of their symptoms on the HAQ-DI [median 2 (IQR 1.4–2.2)]. Whole-body PET/CT A mean dose of 288.6 MBq (s.d. 29.6) of 18F-FDG was administered intravenously prior to whole-body PET/CT. The average time from injection to scan acquisition was 66.6 min (s.d. 9.3). In one patient, imaging of the knees could not be undertaken because the study was abandoned due to severe pain. Hand views obtained in another individual were deemed insufficient upon consensus review. Abnormal 18F-FDG uptake in a peri-articular distribution was observed at the shoulder and hip joints in 22 (100%) and 19 patients (86.4%), respectively (Table 2). FDG avidity was present in all cases in the trochanteric regions. Table 2 Analysis of 18 F-FDG uptake on whole-body PET/CT Site  Frequency of 18F-FDG uptake ≥1, n (%)  Frequency of 18F-FDG uptake ≥2, n (%)  Mean reference score  Mean SUVmax  Shoulder  22 (100)  17 (77.3)  2.4 (0.9)  3.4 (1.4)  Hip  19 (86.4)  17 (77.3)  2 (1.1)  3.0 (2.0)  Trochanteric  22 (100)  13 (59.1)  1.9 (0.8)  2.5 (1.1)  Ischial  21 (95.4)  20 (90.9)  2.4 (0.6)  3.6 (1.7)  Interspinous  18 (81.8)  12 (54.5)  1.8 (1.2)  2.9 (1.9)  Knee (n = 21)  16 (76.2)            Posteromedial  13 (61.9)  11 (52.4)  1.5 (1.3)  3.2 (1.0)      Capsule  10 (47.6)  6 (28.6)  1.0 (1.2)  2.9 (0.9)  Wrist/hand (n = 21)  14 (66.7)            Wrist joint  12 (57.1)  6 (28.6)  1.0 (1.1)  2.8 (1.1)      Volar hand  7 (33.3)  3 (14.3)  0.6 (1.0)  2.4 (0.9)  Site  Frequency of 18F-FDG uptake ≥1, n (%)  Frequency of 18F-FDG uptake ≥2, n (%)  Mean reference score  Mean SUVmax  Shoulder  22 (100)  17 (77.3)  2.4 (0.9)  3.4 (1.4)  Hip  19 (86.4)  17 (77.3)  2 (1.1)  3.0 (2.0)  Trochanteric  22 (100)  13 (59.1)  1.9 (0.8)  2.5 (1.1)  Ischial  21 (95.4)  20 (90.9)  2.4 (0.6)  3.6 (1.7)  Interspinous  18 (81.8)  12 (54.5)  1.8 (1.2)  2.9 (1.9)  Knee (n = 21)  16 (76.2)            Posteromedial  13 (61.9)  11 (52.4)  1.5 (1.3)  3.2 (1.0)      Capsule  10 (47.6)  6 (28.6)  1.0 (1.2)  2.9 (0.9)  Wrist/hand (n = 21)  14 (66.7)            Wrist joint  12 (57.1)  6 (28.6)  1.0 (1.1)  2.8 (1.1)      Volar hand  7 (33.3)  3 (14.3)  0.6 (1.0)  2.4 (0.9)  mean reference score: mean reference to normal liver score. Adjacent to the ischial tuberosities (Fig. 1A and B) represented another site of frequent 18F-FDG uptake (95.4%) and recorded the highest value for mean reference score compared with normal liver [2.4 (s.d. 0.6)] and highest mean SUVmax [3.6 (s.d. 1.7)]. Fig. 1 View largeDownload slide Whole-body PET/CT findings (A and B) Characteristic intense 18F-FDG uptake adjacent to the ischial tuberosities (white arrows). (C and D) FDG avidity of posteromedial knee structures. (E and F) 18F-FDG uptake in the hands in a volar distribution. Fig. 1 View largeDownload slide Whole-body PET/CT findings (A and B) Characteristic intense 18F-FDG uptake adjacent to the ischial tuberosities (white arrows). (C and D) FDG avidity of posteromedial knee structures. (E and F) 18F-FDG uptake in the hands in a volar distribution. At the knees, abnormal 18F-FDG uptake was observed in 16 patients (76.2%). In the majority of these, focal areas of intense FDG avidity in a posteromedial distribution were appreciated (Fig. 1C and D). A high frequency of wrist and/or hand involvement was also detected (66.7%). At the hand, a volar distribution of abnormal 18F-FDG uptake was typically observed (Fig. 1E and F). Vascular 18F-FDG uptake in the large vessels consistent with a diagnosis of GCA (qualitative score ⩾2) was not present in any participant. MRI correlation and fusion of PET/CT with MRI MRI of the pelvis in a patient with characteristic, intense 18F-FDG uptake adjacent to the ischial tuberosities revealed bilateral and symmetric high T2 signal surrounding the proximal hamstring tendon origins of the semimembranosus and the conjoint tendon of the semitendinosus and biceps femoris (Fig. 2D). These changes were confined to the paratenon, with only minimally increased signal noted for the tendons themselves. There was similarly no abnormality observed at the musculotendinous junctions, nor any evidence of bone oedema adjacent to the tendon entheses. Fig. 2 View largeDownload slide Fusion of PET/CT with MRI adjacent to the ischial tuberosities (A) 3D fused 18F-FDG PET/CT. (B) Low-dose CT. (C) Fused 18F-FDG PET/CT. (D) T2-weighted FS MRI reveals high signal surrounding the proximal hamstring tendon origins of the semimembranosus and biceps femoris (yellow arrows). (E) Fused 18F-FDG PET/MRI confirms peritendonitis of the proximal hamstring tendon origins as the anatomical correlate of 18F-FDG uptake at this site. FS: fat suppressed. Fig. 2 View largeDownload slide Fusion of PET/CT with MRI adjacent to the ischial tuberosities (A) 3D fused 18F-FDG PET/CT. (B) Low-dose CT. (C) Fused 18F-FDG PET/CT. (D) T2-weighted FS MRI reveals high signal surrounding the proximal hamstring tendon origins of the semimembranosus and biceps femoris (yellow arrows). (E) Fused 18F-FDG PET/MRI confirms peritendonitis of the proximal hamstring tendon origins as the anatomical correlate of 18F-FDG uptake at this site. FS: fat suppressed. At the knee, MRI in a representative patient with posteromedial 18F-FDG uptake on whole-body PET/CT revealed peritendonitis of the semimembranosus and to a lesser extent the gracilis tendons at their distal insertion (Fig. 3D). No bone oedema was appreciated at the medial tibial plateau. The distal semitendinosus tendon was normal in this instance. Fig. 3 View largeDownload slide Fusion of PET/CT with MRI of posteromedial knee structures (A) 3D fused 18F-FDG PET/CT. (B) Low-dose CT. (C) Fused 18F-FDG PET/CT. (D) T2-weighted FS MRI reveals high signal surrounding the distal hamstring tendon insertions of the semimembranosus and gracilis (yellow arrows). (E) Fused 18F-FDG PET/MRI confirms peritendonitis of the distal semimembranosus tendon as the anatomical correlate of 18F-FDG uptake at this site. FS: fat suppressed. Fig. 3 View largeDownload slide Fusion of PET/CT with MRI of posteromedial knee structures (A) 3D fused 18F-FDG PET/CT. (B) Low-dose CT. (C) Fused 18F-FDG PET/CT. (D) T2-weighted FS MRI reveals high signal surrounding the distal hamstring tendon insertions of the semimembranosus and gracilis (yellow arrows). (E) Fused 18F-FDG PET/MRI confirms peritendonitis of the distal semimembranosus tendon as the anatomical correlate of 18F-FDG uptake at this site. FS: fat suppressed. MRI of the wrist and hand in a patient with a volar distribution of 18F-FDG uptake revealed extensive synovial thickening and enhancement within the carpal tunnel with extension to involve the common flexor tendon sheaths of the digits, in keeping with tenosynovitis (Fig. 4B). Mild extensor tenosynovitis was also appreciated, along with moderate mid-carpal joint synovitis (Fig. 4D). Fig. 4 View largeDownload slide MRI findings of the wrist and hand (A) Volar 18F-FDG uptake on whole-body PET/CT in a representative patient (white arrow). (B) T2-weighted FS MRI of the wrist and hand reveals corresponding flexor tenosynovitis. (C) Fused 18F-FDG PET/CT volume rendered at the hands in the same patient. (D) Mid-carpal joint synovitis seen on T1-weighted FS MRI. FS: fat suppressed. Fig. 4 View largeDownload slide MRI findings of the wrist and hand (A) Volar 18F-FDG uptake on whole-body PET/CT in a representative patient (white arrow). (B) T2-weighted FS MRI of the wrist and hand reveals corresponding flexor tenosynovitis. (C) Fused 18F-FDG PET/CT volume rendered at the hands in the same patient. (D) Mid-carpal joint synovitis seen on T1-weighted FS MRI. FS: fat suppressed. PET/MRI fusion at the pelvis subsequently confirmed peritendonitis of the proximal hamstring tendon origins as the anatomical correlate of abnormal 18F-FDG uptake adjacent to the ischial tuberosities on whole-body PET/CT (Fig. 2E), while abnormal 18F-FDG uptake of posteromedial knee structures correlated with peritendonitis of the distal semimembranosus tendon (Fig. 3E). The same technique could not be utilized at the wrist and hand due to discrepancies in positioning between the two imaging modalities at this site. Discussion On whole-body PET/CT, 22 untreated patients with newly diagnosed PMR exhibited a distinctive pattern of abnormal 18F-FDG uptake around the shoulder and hip joints, in the trochanteric regions and interspinous bursae and adjacent to the ischial tuberosities. In addition, a high frequency of knee and wrist and/or hand involvement was appreciated. Using PET/MRI fusion, hamstring peritendonitis was identified as the anatomical correlate of 18F-FDG uptake adjacent to the ischial tuberosities and of posteromedial knee structures. Uptake adjacent to the ischial tuberosities has been previously reported in whole-body PET/CT studies of PMR patients at rates ranging from 52 to 94% [1–3, 7, 11–14]. Our results reinforce a high incidence of involvement at this site, with abnormal uptake present in all but one patient. Furthermore, these focal areas of FDG avidity within the pelvis recorded the highest value for mean reference score compared with normal liver and highest mean SUVmax. Intense 18F-FDG uptake adjacent to the ischial tuberosities should therefore be considered an imaging hallmark of PMR on whole-body PET/CT. The ischiogluteal bursa is situated over the ischial tuberosity, hence abnormal 18F-FDG uptake in this region on whole-body PET/CT has been previously attributed to ischiogluteal bursitis [15]. Cantini et al. [16] studied hip inflammatory changes on US and MRI in 20 consecutive PMR patients but found no significant difference in the incidence of ischiogluteal bursitis compared with a group of controls (20% vs 25%). Conversely, a recent retrospective study of whole-body PET/CT proposed that praepubic 18F-FDG uptake in PMR represented a correlate of pectineus and adductor longus enthesitis and tenosynovitis [13]. The superior ramus of the ischium is also the tendinous origin for several hamstring muscles [17]. Given that our MRI correlation at the pelvis revealed high T2 signal confined to the paratenon of semimembranosus and the conjoint tendon of semitendinosus and biceps femoris without corresponding abnormality of the ischiogluteal bursa, we conclude that hamstring peritendonitis (as opposed to ischiogluteal bursitis) represents the anatomical correlate of 18F-FDG uptake adjacent to the ischial tuberosities in PMR. Cimmino et al. [8] first reported a high frequency of knee involvement in PMR on whole-body PET/CT. In their study, 24/25 patients (96%) exhibited 18F-FDG uptake in a distribution that highlighted the contour of the fibrous joint capsule. Focal areas of FDG avidity were also appreciated in 13 participants (52%), predominantly in a posteromedial distribution. We have similarly detected 18F-FDG uptake at the knee capsule in 10/21 patients (47.6%) and of posteromedial knee structures in 13/21 patients (61.9%). Based on results from the MRI correlation undertaken at the knee, this pattern of posteromedial knee involvement on whole-body PET/CT is also attributable to peritendonitis of the semimembranosus tendon at its insertion on the medial condyle of the tibia. Pelvic girdle symptoms are said to occur in ∼60% of PMR cases, although a higher incidence of hip pain was recorded in this study (81.8%). Patients with PMR commonly report gluteal and posterior thigh pain and stiffness that are aggravated by prolonged sitting. It has been previously acknowledged that hip joint synovitis provides an inadequate explanation for this distribution of symptoms, given such pathology is typically felt in the groin [16]. Our documentation of hamstring peritendonitis on PET/MRI fusion would therefore appear to provide a better explanation for this symptomatology. To our knowledge, this is the first study to evaluate the distribution and frequency of peripheral joint involvement on whole-body PET/CT in PMR. Dedicated hand views were utilized for this purpose, detecting abnormal 18F-FDG uptake at the wrist and/or hand in 14 patients (66.7%). Twelve participants (54.5%) reported wrist and/or hand pain at study enrolment, with nine of these individuals (75.0%) possessing corresponding findings on whole-body PET/CT. Upon further analysis of this subgroup, reasonable concordance [7/8 patients (87.5%)] was seen between the presence of abnormal 18F-FDG uptake at the wrist and/or hand and greyscale synovitis on US. None of the participants with wrist and/or hand involvement on whole-body PET/CT possessed erosive changes on plain X-ray. There were similar frequencies of abnormal 18F-FDG uptake observed at the shoulder (100% vs 100%), hip (92.9% vs 71.4%), trochanteric regions (100% vs 100%) and adjacent to the ischial tuberosities (100% vs 85.7%) between patients with wrist and/or hand involvement and those without. Values for the mean reference score and SUVmax at the wrist/hand were lower than most other sites throughout the body, with only six (28.6%) and three (14.3%) patients exhibiting 18F-FDG uptake ⩾2 at the wrist joint and volar hand, respectively. While this may infer an overestimate of abnormal findings at the wrist and/or hand in this study, significant FDG avidity was consistently appreciated at the knee. It therefore seems clear that both clinical and subclinical peripheral joint involvement occurs frequently in PMR, firmly contradicting the notion that PMR is a condition limited to the shoulder and hip girdle but also clouding its relationship with late-onset RA. In the 2012 EULAR/ACR classification criteria, peripheral synovitis, presence of RF and/or ACPAs and hip pain/limited range of motion best discriminated RA from PMR, yet distal swelling, tenosynovitis or arthritis was appreciated in 39% of the PMR cohort studied and 51% reported other joint pain [9]. This, together with our findings, demonstrates the difficulty of using peripheral joint involvement as a means of distinguishing PMR from late-onset RA. Further studies of abnormal 18F-FDG uptake in the peripheries on whole-body PET/CT and the longitudinal outcomes of PMR patients with this pattern of involvement are needed. Correlation of volar 18F-FDG uptake at the hand with flexor tenosynovitis on MRI is also noteworthy. A previous study of 13 consecutive patients with PMR (who were not selected on the basis of peripheral arthritis) and 13 controls who underwent MRI of the wrist and hand demonstrated an increased occurrence of tenosynovitis but not synovitis [18]. Extensor tendon tenosynovitis was seen in 9/26 PMR hands (34.6%) compared with 1/26 controls (3.8%) (P = 0.002), with flexor tenosynovitis appreciated in 12/26 PMR hands (46.1%) compared with 4/26 (15.4%) controls (P = 0.03). Since seven patients (33.3%) in our study exhibited a volar distribution of abnormal 18F-FDG uptake at the hand, the incidence of flexor tenosynovitis appears to be similar in this cohort. Together, these results highlight a predilection for tendon involvement in PMR and further substantiate the hypothesis that intra-articular inflammation is not the prevailing pathology in this condition. McGonagle et al. [19] initially proposed that a capsular and extracapsular pattern of joint involvement predominated following their MRI study comparing the shoulders of PMR patients with those affected by RA. More recently, this distribution of inflammation on whole-body MRI has been associated with complete patient-reported glucocorticoid response [20]. Recognition of a distinct imaging phenotype on both MRI and whole-body PET/CT has significant implications for ongoing research in the PMR field. Utilization of these modalities in future studies will be vital to ensure population homogeneity beyond that which can be achieved with conventional classification criteria. There are several limitations to our study. First, small numbers may have resulted in an overrepresentation of participants with peripheral joint involvement by chance alone. Furthermore, the 2012 EULAR/ACR classification criteria have been utilized to determine a diagnosis of PMR in this instance. While these criteria have been established as highly sensitive, their ability to discriminate PMR from other inflammatory conditions with shoulder pain has been questioned [21, 22]. Misclassification therefore remains an inherent difficulty when researching PMR, with ∼8% of patients according to the 2012 criteria having a change in diagnosis after long-term follow-up [9, 22]. Three patients returned low positive RF results that were deemed clinically insignificant. All three of these were notably ACPA negative. Only one of these individuals [RF 14 IU/l (normal <14 IU/ml)] reported wrist and/or hand pain, however, synovitis could not be appreciated on physical examination. Importantly, the exclusion of these three patients from the statistical analyses does not significantly alter the study results. In particular, knee and wrist and/or hand involvement is still appreciated in the remaining 13/18 participants (72.2%) and 11/18 participants (61.1%), respectively. Finally, neither healthy controls nor controls with other rheumatic conditions (e.g. late-onset RA, OA) were enrolled as comparator groups in this study. In the former case, this was due to ethical concerns regarding the radiation exposure associated with whole-body PET/CT. It is well established, however, that 18F-FDG uptake in a distribution like that observed in our patients does not routinely occur in the absence of underlying pathology. Furthermore, a recent retrospective study comparing PET/CT findings in patients with PMR and cancer controls without rheumatologic disease demonstrated highly statistically significant differences in 18F-FDG uptake at multiple sites including the shoulders, interspinous bursae and adjacent to the ischial tuberosities (P < 0.00001 for FDG uptake score) [11]. Whether the involvement of posteromedial knee structures that we have documented is highly specific to PMR remains unknown and will, in the future, need to be compared with patterns of peripheral joint 18F-FDG uptake in other rheumatic conditions. Conclusion Intense 18F-FDG uptake adjacent to the ischial tuberosities and of posteromedial knee structures is common among untreated patients with PMR on whole-body PET/CT. Using fusion technology, we have provided the first documentation of a hamstring peritendonitis as the anatomical correlate of abnormalities detected at both of these sites. In doing so, an alternate explanation for the aetiology of the gluteal and posterior thigh pain that typically characterizes PMR is proposed. A high frequency of wrist and/or hand involvement has also been observed for the first time in this study by performing dedicated views on whole-body PET/CT. Abnormal 18F-FDG uptake in a volar distribution at the hand was found to correlate with flexor tenosynovitis on MRI. These findings add to our growing understanding of the imaging phenotype of PMR and further highlight its distinctive pathology, including a predilection for extra-articular structures distal to the shoulder and pelvic girdles. Further studies are now needed to determine the relationship between peripheral abnormalities detected on whole-body PET/CT and the longitudinal outcomes in this subset of PMR patients. Acknowledgements C.E.O. has been supported by research grants received from Arthritis Australia and Austin Medical Research Foundation. Funding: This work was supported by an unrestricted educational grant from AbbVie. Disclosure statement: The authors have declared no conflicts of interest. References 1 Takahashi H, Yamashita H, Kubota K et al.   Differences in fluorodeoxyglucose positron emission tomography/computed tomography findings between elderly onset rheumatoid arthritis and polymyalgia rheumatica. Mod Rheumatol  2015; 25: 546– 51. Google Scholar CrossRef Search ADS PubMed  2 Yamashita H, Kubota K, Takahashi Y et al.   Similarities and differences in fluorodeoxyglucose positron emission tomography/computed tomography findings in spondyloarthropathy, polymyalgia rheumatica and rheumatoid arthritis. Joint Bone Spine  2013; 80: 171– 7. Google Scholar CrossRef Search ADS PubMed  3 Wakura D, Kotani T, Takeuchi T et al.   Differentiation between polymyalgia rheumatica (PMR) and elderly-onset rheumatoid arthritis using 18F-fluorodeoxyglucose positron emission tomography/computed tomography: is enthesitis a new pathological lesion in PMR? PLoS One  [Electronic Resource] 2016; 11: e0158509. Google Scholar CrossRef Search ADS PubMed  4 Blockmans D, De Ceuninck L, Vanderschueren S et al.   Repetitive 18-fluorodeoxyglucose positron emission tomography in isolated polymyalgia rheumatica: a prospective study in 35 patients. Rheumatology  2006; 46: 672– 7. Google Scholar CrossRef Search ADS PubMed  5 Salvarani C, Barozzi L, Boiardi L et al.   Lumbar interspinous bursitis in active polymyalgia rheumatica. Clin Exp Rheumatol  2013; 31: 526– 31. Google Scholar PubMed  6 Salvarani C, Barozzi L, Cantini F et al.   Cervical interspinous bursitis in active polymyalgia rheumatica. Ann Rheum Dis  2008; 67: 758– 61. Google Scholar CrossRef Search ADS PubMed  7 Yamashita H, Kubota K, Takahashi Y et al.   Whole-body fluorodeoxyglucose positron emission tomography/computed tomography in patients with active polymyalgia rheumatica: evidence for distinctive bursitis and large-vessel vasculitis. Mod Rheumatol  2012; 22: 705– 11. Google Scholar CrossRef Search ADS PubMed  8 Cimmino MA, Camellino D, Paparo F et al.   High frequency of capsular knee involvement in polymyalgia rheumatica/giant cell arteritis patients studied by positron emission tomography. Rheumatology  2013; 52: 1865– 72. Google Scholar CrossRef Search ADS PubMed  9 Dasgupta B, Cimmino MA, Kremers HM et al.   2012 Provisional classification criteria for polymyalgia rheumatica: a European League Against Rheumatism/American College of Rheumatology collaborative initiative. Arthritis Rheum  2012; 64: 943– 54. Google Scholar CrossRef Search ADS PubMed  10 Mackie SL, Koduri G, Hill CL et al.   Accuracy of musculoskeletal imaging for the diagnosis of polymyalgia rheumatica: systematic review. RMD Open  2015; 1: e000100. Google Scholar CrossRef Search ADS PubMed  11 Sondag M, Guillot X, Verhoeven F et al.   Utility of 18F-fluoro-dexoxyglucose positron emission tomography for the diagnosis of polymyalgia rheumatica: a controlled study. Rheumatology  2016; 55: 1452– 7. Google Scholar CrossRef Search ADS PubMed  12 Palard-Novello X, Querellou S, Gouillou M et al.   Value of 18F-FDG PET/CT for therapeutic assessment of patients with polymyalgia rheumatica receiving tocilizumab as first-line treatment. Eur J Nucl Med Mol Imaging  2016; 43: 773– 9. Google Scholar CrossRef Search ADS PubMed  13 Rehak Z, Sprlakova-Pukova A, Bortlicek Z et al.   PET/CT imaging in polymyalgia rheumatica: praepubic 18F-FDG uptake correlates with pectineus and adductor longus muscles enthesitis and with tenosynovitis. Radiol Oncol  2017; 51: 8– 14. Google Scholar CrossRef Search ADS PubMed  14 Rehak Z, Vasina J, Nemec P et al.   Various forms of 18F-FDG PET and PET/CT findings in patients with polymyalgia rheumatica. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub  2015; 159: 629– 36. Google Scholar PubMed  15 Hitora T, Kawaguchi Y, Mori M et al.   Ischiogluteal bursitis: a report of three cases with MR findings. Rheumatol Int  2009; 29: 455– 8. Google Scholar CrossRef Search ADS PubMed  16 Cantini F, Niccoli L, Nannini C et al.   Inflammatory changes of hip synovial structures in polymyalgia rheumatica. Clin Exp Rheumatol  2005; 23: 462– 8. Google Scholar PubMed  17 Moore KL., Dally AF. Clinically Oriented Anatomy , 4th edn. Baltimore, MD: Lippincott Williams & Wilkins, 1999. 18 Cimmino MA, Parodi M, Zampogna G, Barbieri F, Garlaschi G. Polymyalgia rheumatica is associated with extensor tendon tenosynovitis but not with synovitis of the hands: a magnetic resonance imaging study. Rheumatology  2011; 50: 494– 9. Google Scholar CrossRef Search ADS PubMed  19 McGonagle D, Pease C, Marzo-Ortega H et al.   Comparison of extracapsular changes by magnetic resonance imaging in patients with rheumatoid arthritis and polymyalgia rheumatica. J Rheumatol  2001; 28: 1837– 41. Google Scholar PubMed  20 Mackie SL, Pease CT, Fukuba E et al.   Whole-body MRI of patients with polymyalgia rheumatica identifies a distinct subset with complete patient-reported response to glucocorticoids. Ann Rheum Dis  2015; 74: 2188– 92. Google Scholar CrossRef Search ADS PubMed  21 Macchioni P, Boiardi L, Catanoso M, Pazzola G, Salvarani C. Performance of the new 2012 EULAR/ACR classification criteria for polymyalgia rheumatica: comparison with the previous criteria in a single-centre study. Ann Rheum Dis  2014; 73: 1190– 3. Google Scholar CrossRef Search ADS PubMed  22 Ozen G, Inanc N, Unal AU et al.   Assessment of the new 2012 EULAR/ACR clinical classification criteria for polymyalgia rheumatica: a prospective multicenter study. J Rheumatol  2016; 43: 893– 900. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Rheumatology Oxford University Press

Fusion of positron emission tomography/computed tomography with magnetic resonance imaging reveals hamstring peritendonitis in polymyalgia rheumatica

Loading next page...
 
/lp/ou_press/fusion-of-positron-emission-tomography-computed-tomography-with-PbOWNaw8D3
Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oup.com
ISSN
1462-0324
eISSN
1462-0332
D.O.I.
10.1093/rheumatology/kex411
Publisher site
See Article on Publisher Site

Abstract

Abstract Objectives To characterize 18F-fluorodeoxyglucose (18F-FDG) uptake on whole-body PET/CT in PMR, and identify its precise anatomic correlate using MRI. Methods Patients with newly diagnosed PMR according to the 2012 EULAR/ACR classification criteria were prospectively recruited. Participants with GCA were excluded. A whole-body 18F-FDG PET/CT scan was performed in all untreated patients. Qualitative and semiquantitative [standardized uptake value maximum (SUVmax)] scoring of abnormal 18F-FDG uptake was undertaken. MRI of the pelvis, knee and wrist and hand was performed in three representative patients with anatomical correlation of FDG-avid sites carried out using Medview fusion software. Results Twenty-two patients with PMR were recruited. Their mean age was 68.3 years (s.d. 6.3) and 13/22 were male. On whole-body PET/CT, 18F-FDG uptake adjacent to the ischial tuberosities was observed in 21 participants (95.4%) and recorded the highest mean SUVmax value [3.6 (s.d. 1.7)]. A high frequency of posteromedial knee (61.9%) and wrist and/or hand involvement (66.7%) was also appreciated. MRI of the pelvis revealed high T2 signal surrounding the proximal hamstring tendon origins of both semimembranosus and the conjoint tendon of the semitendinosus and biceps femoris. At the knee, peritendonitis at the distal insertion of the semimembranosus was observed. PET/MRI fusion at the pelvis and knee confirmed semimembranosus peritendonitis as the anatomical correlate of 18F-FDG uptake adjacent to the ischial tuberosities and of posteromedial knee structures. Conclusion Hamstring peritendonitis is a common and distinctive manifestation of PMR on whole-body PET/CT. Trial registration Australian New Zealand Clinical Trials Registry, http://www.anzctr.org.au, ACTRN1261400696695. polymyalgia rheumatica, whole-body positron emission tomography/computed tomography, magnetic resonance imaging, hamstring, peritendonitis Rheumatology key messages PMR exhibits a distinctive pattern of abnormal 18F-FDG uptake on whole-body PET/CT. Hamstring peritendonitis correlates with 18F-FDG uptake adjacent to the ischial tuberosities and of posteromedial knee structures. Frequent wrist and/or hand involvement is observed on whole-body PET/CT in PMR patients. Introduction PMR is an inflammatory condition predominantly characterized by subacute-onset shoulder and hip pain and stiffness in patients ⩾50 years of age. Some controversy has existed as to whether PMR represents a distinct disease entity due to its close relationship with both GCA and late-onset RA, however, imaging modalities such as PET have recently uncovered its distinctive pathology [1–3]. Blockmans et al. [4] first reported the ability of PET to detect articular inflammation in PMR, finding abnormal 18F-fluorodeoxyglucose (18F-FDG) uptake in the shoulders and hips in 94 and 89% of patients, respectively. They additionally described a novel finding involving the processi spinosi of the vertebrae, which has since been localized to the interspinous bursa on MRI [5, 6]. The pairing of whole-body PET with low-dose CT has facilitated further study of the anatomic distribution of PMR. 18F-FDG uptake in the trochanteric region and adjacent to the ischial tuberosities has been attributed to trochanteric and ischiogluteal bursitis, respectively [7]. A high frequency of knee involvement due to capsulitis has similarly been reported [8]. However, the low resolution of whole-body PET/CT images remains a limitation to precise identification of anatomic correlates of abnormal 18F-FDG uptake. Peripheral joint involvement is appreciated clinically in almost half of PMR patients [9]. Distinguishing this group from late-onset RA at presentation is especially difficult. However, it is increasingly apparent from imaging studies that the prevailing pathology in PMR is extra-articular rather than intra-articular [10]. Unlike RA, a predilection for tendinous and bursal structures combined with capsular joint involvement is typically observed. In this study we sought to further characterize the pattern of 18F-FDG uptake on whole-body PET/CT in untreated patients with newly diagnosed PMR. Utilizing dedicated hand views, additional emphasis was also placed upon the distribution and frequency of peripheral joint involvement. MRI was subsequently performed in representative patients before fusion technology was employed to precisely identify anatomic correlates of abnormal 18F-FDG uptake. Methods Patients Patients with newly diagnosed PMR according to the 2012 EULAR/ACR classification criteria (utilizing the clinical scoring algorithm) [9] were prospectively recruited from primary care, community and hospital rheumatology practices. Participants who had received prior prednisolone treatment and those with symptoms suggestive of GCA (headache, scalp tenderness, jaw claudication or visual disturbance) were excluded, as were patients with a history of cancer within the past 5 years, neuromuscular disease, active infection, other inflammatory conditions and chronic pain syndromes. The study was approved by the Austin Health Human Research Ethics Committee prior to commencement and registered with the Australian New Zealand Clinical Trials Registry (trial identification ACTRN1261400696695). Clinical and laboratory variables Following appropriate written consent, demographic and clinical data were collected. Participants also completed the HAQ disability index (HAQ-DI). A standard physical examination was carried out with specific focus on the presence of peripheral synovitis. Baseline laboratory testing comprised full blood count, CRP, ESR, creatinine kinase, thyroid function testing, RF and ACPAs. Bilateral wrist and hand X-rays were performed to assess for erosive change. B-mode and power Doppler US was undertaken at both wrists by a rheumatologist experienced in US (C.E.O.) using a LOGIQ e machine (GE Healthcare, Chicago, IL, USA) with linear 5–13 MHz transducer. Whole-body PET/CT A whole body 18F-FDG PET/CT scan (Gemini-TF64 or Ingenuity-TF128 PET/CT system; Phillips, Cleveland, OH, USA) from skull vertex to toes (including dedicated hand views) was performed within 7 days of the screening visit and prior to the commencement of prednisolone treatment. Following a minimum 6 h fast, participants received a weight-based i.v. injection of 18F-FDG. The emission study and low-dose CT was acquired after a 1 h rest period, with CT attenuation correction performed on all whole-body and dedicated hand images prior to independent qualitative and semiquantitative analysis of abnormal 18F-FDG uptake by two experienced nuclear medicine physicians (A.P. and S.T.L.). Any significant differences in results were resolved by consensus. Qualitative analysis involved visual evaluation of 18F-FDG uptake using the following scoring system: 0, no uptake; 1, lower than normal liver uptake; 2, similar to normal liver uptake; 3, higher than normal liver uptake. A score ⩾1 was considered a positive result. For semiquantitative analysis, a region of interest was drawn around each site and the maximum standardized uptake value (SUVmax) calculated. The following sites were specifically assessed: two shoulder joints, two hip joints, two trochanteric regions, regions adjacent to the two ischial tuberosities, most FDG-avid interspinous bursa, two knee joints, two wrist joints and two hands. MRI correlation and fusion of PET/CT with MRI To precisely identify anatomic correlates of abnormal 18F-FDG uptake, MRI was performed in three representative patients at the pelvis, knee and wrist and hand using a 1.5 T magnet (Siemens Avanto, Erlangen, Germany) with a Siemens i-PAT compatible six-element design phased array body matrix, Siemens 12 channel transmit/receive configuration knee coil and Invivo 8 channel wrist coil, respectively. For imaging of the pelvis, non-contrast T1-weighted fast spin-echo (FSE) sequences in the axial plane, proton density-weighted FSE sequences in the axial, sagittal and coronal planes with fat saturation in the coronal plane and T2-weighted FSE sequences with fat saturation in the axial plane centred on the ischial tuberosities were obtained. For imaging of the knee, proton density-weighted FSE and T2-weighted FSE sequences with fat saturation in the coronal plane, proton density-weighted FSE with fat saturation in the sagittal plane and proton density-weighted FSE and T2-weighted FSE images with fat saturation in the axial plane were obtained. For imaging of the wrist and hand, the following sequences were performed before i.v. injection: proton density-weighted FSE in the axial, coronal and sagittal planes with fat saturation in the coronal plane; T2-weighted FSE in the axial plane with fat suppression and 3D T2*-weighted GRE in the coronal plane. Following i.v. injection of 10 ml of gadolinium-based contrast agent (Dotarem, gadoterate meglumine, Guerbet USA, Princeton, NJ, USA), T1-weighted FSE fat suppression images in the axial and coronal planes were obtained. Interpretation of MRIs was performed by two experienced musculoskeletal radiologists (L.Y., R.Z.). Medview fusion software (Ann Arbor, MI, USA) was utilized to align and superimpose PET and MRI datasets and to anatomically correlate sites of abnormal 18F-FDG uptake adjacent to the ischial tuberosities and of posteromedial knee structures. Statistical analysis Results for continuous variables are reported as mean (s.d.) or median [interquartile range (IQR)]. Categorical variables are represented as proportions (percentages). All statistical analyses were undertaken using Stata 13.0 (StataCorp, College Station, TX, USA). Results Patient demographics, clinical and laboratory features Twenty-two patients met the inclusion criteria for a diagnosis of PMR [mean clinical algorithm score 5.1 (s.d. 0.7)]. Their demographic data are summarized in Table 1. The mean age at diagnosis was 68.3 years (s.d. 6.3). Thirteen patients (59.1%) were male and all participants were Caucasian. Table 1 Baseline demographic data and clinical features Demographic or clinical features  Value  Age, mean (s.d.), years  68.3 (6.3)  Male, n (%)  13 (59.1)  Caucasian, n (%)  22 (100)  BMI, mean (s.d.), kg/m2  28.6 (5.1)  Ever smoked, n (%)  10 (45.4)  Shoulder pain, n (%)  22 (100)  Hip pain, n (%)  18 (81.8)  Back pain, n (%)  13 (59.1)  Knee pain, n (%)  14 (63.6)  Wrist/hand pain, n (%)  12 (54.5)  Morning stiffness >45 min, n (%)  18 (81.8)  Median CRP, median (IQR), mg/l  40.8 (24.6–67.1)  Median ESR, median (IQR), mm  45.5 (29–64)  PMR-activity score, mean (s.d.)  71.3 (38.8)  Median HAQ-DI score, median (IQR)  2 (1.4–2.2)  Demographic or clinical features  Value  Age, mean (s.d.), years  68.3 (6.3)  Male, n (%)  13 (59.1)  Caucasian, n (%)  22 (100)  BMI, mean (s.d.), kg/m2  28.6 (5.1)  Ever smoked, n (%)  10 (45.4)  Shoulder pain, n (%)  22 (100)  Hip pain, n (%)  18 (81.8)  Back pain, n (%)  13 (59.1)  Knee pain, n (%)  14 (63.6)  Wrist/hand pain, n (%)  12 (54.5)  Morning stiffness >45 min, n (%)  18 (81.8)  Median CRP, median (IQR), mg/l  40.8 (24.6–67.1)  Median ESR, median (IQR), mm  45.5 (29–64)  PMR-activity score, mean (s.d.)  71.3 (38.8)  Median HAQ-DI score, median (IQR)  2 (1.4–2.2)  Consistent with the 2012 EULAR/ACR classification criteria, all patients possessed bilateral shoulder pain, with 18/22 (81.8%) also reporting concomitant hip pain. Fourteen participants (63.6%) indicated involvement of their knees, while 12 (54.5%) reported wrist and/or hand symptoms. The median duration of early morning stiffness was 120 min (IQR 60–120). On clinical examination, synovitis at the wrist and/or hand was detected in nine patients (40.9%) and six (27.3%) exhibited swelling and/or tenderness of the knee. All participants were negative for ACPAs. Three patients returned a low positive RF result [14 IU/ml in two participants and 17 IU/ml in the third (n < 14 IU/ml)]. A single patient was found to possess a subtle marginal erosion of the right fourth metacarpal head on bilateral wrist and hand X-rays. Using US, greyscale synovitis was detected at the wrist in eight patients (36.4%), five (22.7%) of whom had associated power Doppler signal. Using the PMR activity score, disease activity was judged to be high [mean 71.3 (s.d. 38.8)]. Participants reported a severe level of disability because of their symptoms on the HAQ-DI [median 2 (IQR 1.4–2.2)]. Whole-body PET/CT A mean dose of 288.6 MBq (s.d. 29.6) of 18F-FDG was administered intravenously prior to whole-body PET/CT. The average time from injection to scan acquisition was 66.6 min (s.d. 9.3). In one patient, imaging of the knees could not be undertaken because the study was abandoned due to severe pain. Hand views obtained in another individual were deemed insufficient upon consensus review. Abnormal 18F-FDG uptake in a peri-articular distribution was observed at the shoulder and hip joints in 22 (100%) and 19 patients (86.4%), respectively (Table 2). FDG avidity was present in all cases in the trochanteric regions. Table 2 Analysis of 18 F-FDG uptake on whole-body PET/CT Site  Frequency of 18F-FDG uptake ≥1, n (%)  Frequency of 18F-FDG uptake ≥2, n (%)  Mean reference score  Mean SUVmax  Shoulder  22 (100)  17 (77.3)  2.4 (0.9)  3.4 (1.4)  Hip  19 (86.4)  17 (77.3)  2 (1.1)  3.0 (2.0)  Trochanteric  22 (100)  13 (59.1)  1.9 (0.8)  2.5 (1.1)  Ischial  21 (95.4)  20 (90.9)  2.4 (0.6)  3.6 (1.7)  Interspinous  18 (81.8)  12 (54.5)  1.8 (1.2)  2.9 (1.9)  Knee (n = 21)  16 (76.2)            Posteromedial  13 (61.9)  11 (52.4)  1.5 (1.3)  3.2 (1.0)      Capsule  10 (47.6)  6 (28.6)  1.0 (1.2)  2.9 (0.9)  Wrist/hand (n = 21)  14 (66.7)            Wrist joint  12 (57.1)  6 (28.6)  1.0 (1.1)  2.8 (1.1)      Volar hand  7 (33.3)  3 (14.3)  0.6 (1.0)  2.4 (0.9)  Site  Frequency of 18F-FDG uptake ≥1, n (%)  Frequency of 18F-FDG uptake ≥2, n (%)  Mean reference score  Mean SUVmax  Shoulder  22 (100)  17 (77.3)  2.4 (0.9)  3.4 (1.4)  Hip  19 (86.4)  17 (77.3)  2 (1.1)  3.0 (2.0)  Trochanteric  22 (100)  13 (59.1)  1.9 (0.8)  2.5 (1.1)  Ischial  21 (95.4)  20 (90.9)  2.4 (0.6)  3.6 (1.7)  Interspinous  18 (81.8)  12 (54.5)  1.8 (1.2)  2.9 (1.9)  Knee (n = 21)  16 (76.2)            Posteromedial  13 (61.9)  11 (52.4)  1.5 (1.3)  3.2 (1.0)      Capsule  10 (47.6)  6 (28.6)  1.0 (1.2)  2.9 (0.9)  Wrist/hand (n = 21)  14 (66.7)            Wrist joint  12 (57.1)  6 (28.6)  1.0 (1.1)  2.8 (1.1)      Volar hand  7 (33.3)  3 (14.3)  0.6 (1.0)  2.4 (0.9)  mean reference score: mean reference to normal liver score. Adjacent to the ischial tuberosities (Fig. 1A and B) represented another site of frequent 18F-FDG uptake (95.4%) and recorded the highest value for mean reference score compared with normal liver [2.4 (s.d. 0.6)] and highest mean SUVmax [3.6 (s.d. 1.7)]. Fig. 1 View largeDownload slide Whole-body PET/CT findings (A and B) Characteristic intense 18F-FDG uptake adjacent to the ischial tuberosities (white arrows). (C and D) FDG avidity of posteromedial knee structures. (E and F) 18F-FDG uptake in the hands in a volar distribution. Fig. 1 View largeDownload slide Whole-body PET/CT findings (A and B) Characteristic intense 18F-FDG uptake adjacent to the ischial tuberosities (white arrows). (C and D) FDG avidity of posteromedial knee structures. (E and F) 18F-FDG uptake in the hands in a volar distribution. At the knees, abnormal 18F-FDG uptake was observed in 16 patients (76.2%). In the majority of these, focal areas of intense FDG avidity in a posteromedial distribution were appreciated (Fig. 1C and D). A high frequency of wrist and/or hand involvement was also detected (66.7%). At the hand, a volar distribution of abnormal 18F-FDG uptake was typically observed (Fig. 1E and F). Vascular 18F-FDG uptake in the large vessels consistent with a diagnosis of GCA (qualitative score ⩾2) was not present in any participant. MRI correlation and fusion of PET/CT with MRI MRI of the pelvis in a patient with characteristic, intense 18F-FDG uptake adjacent to the ischial tuberosities revealed bilateral and symmetric high T2 signal surrounding the proximal hamstring tendon origins of the semimembranosus and the conjoint tendon of the semitendinosus and biceps femoris (Fig. 2D). These changes were confined to the paratenon, with only minimally increased signal noted for the tendons themselves. There was similarly no abnormality observed at the musculotendinous junctions, nor any evidence of bone oedema adjacent to the tendon entheses. Fig. 2 View largeDownload slide Fusion of PET/CT with MRI adjacent to the ischial tuberosities (A) 3D fused 18F-FDG PET/CT. (B) Low-dose CT. (C) Fused 18F-FDG PET/CT. (D) T2-weighted FS MRI reveals high signal surrounding the proximal hamstring tendon origins of the semimembranosus and biceps femoris (yellow arrows). (E) Fused 18F-FDG PET/MRI confirms peritendonitis of the proximal hamstring tendon origins as the anatomical correlate of 18F-FDG uptake at this site. FS: fat suppressed. Fig. 2 View largeDownload slide Fusion of PET/CT with MRI adjacent to the ischial tuberosities (A) 3D fused 18F-FDG PET/CT. (B) Low-dose CT. (C) Fused 18F-FDG PET/CT. (D) T2-weighted FS MRI reveals high signal surrounding the proximal hamstring tendon origins of the semimembranosus and biceps femoris (yellow arrows). (E) Fused 18F-FDG PET/MRI confirms peritendonitis of the proximal hamstring tendon origins as the anatomical correlate of 18F-FDG uptake at this site. FS: fat suppressed. At the knee, MRI in a representative patient with posteromedial 18F-FDG uptake on whole-body PET/CT revealed peritendonitis of the semimembranosus and to a lesser extent the gracilis tendons at their distal insertion (Fig. 3D). No bone oedema was appreciated at the medial tibial plateau. The distal semitendinosus tendon was normal in this instance. Fig. 3 View largeDownload slide Fusion of PET/CT with MRI of posteromedial knee structures (A) 3D fused 18F-FDG PET/CT. (B) Low-dose CT. (C) Fused 18F-FDG PET/CT. (D) T2-weighted FS MRI reveals high signal surrounding the distal hamstring tendon insertions of the semimembranosus and gracilis (yellow arrows). (E) Fused 18F-FDG PET/MRI confirms peritendonitis of the distal semimembranosus tendon as the anatomical correlate of 18F-FDG uptake at this site. FS: fat suppressed. Fig. 3 View largeDownload slide Fusion of PET/CT with MRI of posteromedial knee structures (A) 3D fused 18F-FDG PET/CT. (B) Low-dose CT. (C) Fused 18F-FDG PET/CT. (D) T2-weighted FS MRI reveals high signal surrounding the distal hamstring tendon insertions of the semimembranosus and gracilis (yellow arrows). (E) Fused 18F-FDG PET/MRI confirms peritendonitis of the distal semimembranosus tendon as the anatomical correlate of 18F-FDG uptake at this site. FS: fat suppressed. MRI of the wrist and hand in a patient with a volar distribution of 18F-FDG uptake revealed extensive synovial thickening and enhancement within the carpal tunnel with extension to involve the common flexor tendon sheaths of the digits, in keeping with tenosynovitis (Fig. 4B). Mild extensor tenosynovitis was also appreciated, along with moderate mid-carpal joint synovitis (Fig. 4D). Fig. 4 View largeDownload slide MRI findings of the wrist and hand (A) Volar 18F-FDG uptake on whole-body PET/CT in a representative patient (white arrow). (B) T2-weighted FS MRI of the wrist and hand reveals corresponding flexor tenosynovitis. (C) Fused 18F-FDG PET/CT volume rendered at the hands in the same patient. (D) Mid-carpal joint synovitis seen on T1-weighted FS MRI. FS: fat suppressed. Fig. 4 View largeDownload slide MRI findings of the wrist and hand (A) Volar 18F-FDG uptake on whole-body PET/CT in a representative patient (white arrow). (B) T2-weighted FS MRI of the wrist and hand reveals corresponding flexor tenosynovitis. (C) Fused 18F-FDG PET/CT volume rendered at the hands in the same patient. (D) Mid-carpal joint synovitis seen on T1-weighted FS MRI. FS: fat suppressed. PET/MRI fusion at the pelvis subsequently confirmed peritendonitis of the proximal hamstring tendon origins as the anatomical correlate of abnormal 18F-FDG uptake adjacent to the ischial tuberosities on whole-body PET/CT (Fig. 2E), while abnormal 18F-FDG uptake of posteromedial knee structures correlated with peritendonitis of the distal semimembranosus tendon (Fig. 3E). The same technique could not be utilized at the wrist and hand due to discrepancies in positioning between the two imaging modalities at this site. Discussion On whole-body PET/CT, 22 untreated patients with newly diagnosed PMR exhibited a distinctive pattern of abnormal 18F-FDG uptake around the shoulder and hip joints, in the trochanteric regions and interspinous bursae and adjacent to the ischial tuberosities. In addition, a high frequency of knee and wrist and/or hand involvement was appreciated. Using PET/MRI fusion, hamstring peritendonitis was identified as the anatomical correlate of 18F-FDG uptake adjacent to the ischial tuberosities and of posteromedial knee structures. Uptake adjacent to the ischial tuberosities has been previously reported in whole-body PET/CT studies of PMR patients at rates ranging from 52 to 94% [1–3, 7, 11–14]. Our results reinforce a high incidence of involvement at this site, with abnormal uptake present in all but one patient. Furthermore, these focal areas of FDG avidity within the pelvis recorded the highest value for mean reference score compared with normal liver and highest mean SUVmax. Intense 18F-FDG uptake adjacent to the ischial tuberosities should therefore be considered an imaging hallmark of PMR on whole-body PET/CT. The ischiogluteal bursa is situated over the ischial tuberosity, hence abnormal 18F-FDG uptake in this region on whole-body PET/CT has been previously attributed to ischiogluteal bursitis [15]. Cantini et al. [16] studied hip inflammatory changes on US and MRI in 20 consecutive PMR patients but found no significant difference in the incidence of ischiogluteal bursitis compared with a group of controls (20% vs 25%). Conversely, a recent retrospective study of whole-body PET/CT proposed that praepubic 18F-FDG uptake in PMR represented a correlate of pectineus and adductor longus enthesitis and tenosynovitis [13]. The superior ramus of the ischium is also the tendinous origin for several hamstring muscles [17]. Given that our MRI correlation at the pelvis revealed high T2 signal confined to the paratenon of semimembranosus and the conjoint tendon of semitendinosus and biceps femoris without corresponding abnormality of the ischiogluteal bursa, we conclude that hamstring peritendonitis (as opposed to ischiogluteal bursitis) represents the anatomical correlate of 18F-FDG uptake adjacent to the ischial tuberosities in PMR. Cimmino et al. [8] first reported a high frequency of knee involvement in PMR on whole-body PET/CT. In their study, 24/25 patients (96%) exhibited 18F-FDG uptake in a distribution that highlighted the contour of the fibrous joint capsule. Focal areas of FDG avidity were also appreciated in 13 participants (52%), predominantly in a posteromedial distribution. We have similarly detected 18F-FDG uptake at the knee capsule in 10/21 patients (47.6%) and of posteromedial knee structures in 13/21 patients (61.9%). Based on results from the MRI correlation undertaken at the knee, this pattern of posteromedial knee involvement on whole-body PET/CT is also attributable to peritendonitis of the semimembranosus tendon at its insertion on the medial condyle of the tibia. Pelvic girdle symptoms are said to occur in ∼60% of PMR cases, although a higher incidence of hip pain was recorded in this study (81.8%). Patients with PMR commonly report gluteal and posterior thigh pain and stiffness that are aggravated by prolonged sitting. It has been previously acknowledged that hip joint synovitis provides an inadequate explanation for this distribution of symptoms, given such pathology is typically felt in the groin [16]. Our documentation of hamstring peritendonitis on PET/MRI fusion would therefore appear to provide a better explanation for this symptomatology. To our knowledge, this is the first study to evaluate the distribution and frequency of peripheral joint involvement on whole-body PET/CT in PMR. Dedicated hand views were utilized for this purpose, detecting abnormal 18F-FDG uptake at the wrist and/or hand in 14 patients (66.7%). Twelve participants (54.5%) reported wrist and/or hand pain at study enrolment, with nine of these individuals (75.0%) possessing corresponding findings on whole-body PET/CT. Upon further analysis of this subgroup, reasonable concordance [7/8 patients (87.5%)] was seen between the presence of abnormal 18F-FDG uptake at the wrist and/or hand and greyscale synovitis on US. None of the participants with wrist and/or hand involvement on whole-body PET/CT possessed erosive changes on plain X-ray. There were similar frequencies of abnormal 18F-FDG uptake observed at the shoulder (100% vs 100%), hip (92.9% vs 71.4%), trochanteric regions (100% vs 100%) and adjacent to the ischial tuberosities (100% vs 85.7%) between patients with wrist and/or hand involvement and those without. Values for the mean reference score and SUVmax at the wrist/hand were lower than most other sites throughout the body, with only six (28.6%) and three (14.3%) patients exhibiting 18F-FDG uptake ⩾2 at the wrist joint and volar hand, respectively. While this may infer an overestimate of abnormal findings at the wrist and/or hand in this study, significant FDG avidity was consistently appreciated at the knee. It therefore seems clear that both clinical and subclinical peripheral joint involvement occurs frequently in PMR, firmly contradicting the notion that PMR is a condition limited to the shoulder and hip girdle but also clouding its relationship with late-onset RA. In the 2012 EULAR/ACR classification criteria, peripheral synovitis, presence of RF and/or ACPAs and hip pain/limited range of motion best discriminated RA from PMR, yet distal swelling, tenosynovitis or arthritis was appreciated in 39% of the PMR cohort studied and 51% reported other joint pain [9]. This, together with our findings, demonstrates the difficulty of using peripheral joint involvement as a means of distinguishing PMR from late-onset RA. Further studies of abnormal 18F-FDG uptake in the peripheries on whole-body PET/CT and the longitudinal outcomes of PMR patients with this pattern of involvement are needed. Correlation of volar 18F-FDG uptake at the hand with flexor tenosynovitis on MRI is also noteworthy. A previous study of 13 consecutive patients with PMR (who were not selected on the basis of peripheral arthritis) and 13 controls who underwent MRI of the wrist and hand demonstrated an increased occurrence of tenosynovitis but not synovitis [18]. Extensor tendon tenosynovitis was seen in 9/26 PMR hands (34.6%) compared with 1/26 controls (3.8%) (P = 0.002), with flexor tenosynovitis appreciated in 12/26 PMR hands (46.1%) compared with 4/26 (15.4%) controls (P = 0.03). Since seven patients (33.3%) in our study exhibited a volar distribution of abnormal 18F-FDG uptake at the hand, the incidence of flexor tenosynovitis appears to be similar in this cohort. Together, these results highlight a predilection for tendon involvement in PMR and further substantiate the hypothesis that intra-articular inflammation is not the prevailing pathology in this condition. McGonagle et al. [19] initially proposed that a capsular and extracapsular pattern of joint involvement predominated following their MRI study comparing the shoulders of PMR patients with those affected by RA. More recently, this distribution of inflammation on whole-body MRI has been associated with complete patient-reported glucocorticoid response [20]. Recognition of a distinct imaging phenotype on both MRI and whole-body PET/CT has significant implications for ongoing research in the PMR field. Utilization of these modalities in future studies will be vital to ensure population homogeneity beyond that which can be achieved with conventional classification criteria. There are several limitations to our study. First, small numbers may have resulted in an overrepresentation of participants with peripheral joint involvement by chance alone. Furthermore, the 2012 EULAR/ACR classification criteria have been utilized to determine a diagnosis of PMR in this instance. While these criteria have been established as highly sensitive, their ability to discriminate PMR from other inflammatory conditions with shoulder pain has been questioned [21, 22]. Misclassification therefore remains an inherent difficulty when researching PMR, with ∼8% of patients according to the 2012 criteria having a change in diagnosis after long-term follow-up [9, 22]. Three patients returned low positive RF results that were deemed clinically insignificant. All three of these were notably ACPA negative. Only one of these individuals [RF 14 IU/l (normal <14 IU/ml)] reported wrist and/or hand pain, however, synovitis could not be appreciated on physical examination. Importantly, the exclusion of these three patients from the statistical analyses does not significantly alter the study results. In particular, knee and wrist and/or hand involvement is still appreciated in the remaining 13/18 participants (72.2%) and 11/18 participants (61.1%), respectively. Finally, neither healthy controls nor controls with other rheumatic conditions (e.g. late-onset RA, OA) were enrolled as comparator groups in this study. In the former case, this was due to ethical concerns regarding the radiation exposure associated with whole-body PET/CT. It is well established, however, that 18F-FDG uptake in a distribution like that observed in our patients does not routinely occur in the absence of underlying pathology. Furthermore, a recent retrospective study comparing PET/CT findings in patients with PMR and cancer controls without rheumatologic disease demonstrated highly statistically significant differences in 18F-FDG uptake at multiple sites including the shoulders, interspinous bursae and adjacent to the ischial tuberosities (P < 0.00001 for FDG uptake score) [11]. Whether the involvement of posteromedial knee structures that we have documented is highly specific to PMR remains unknown and will, in the future, need to be compared with patterns of peripheral joint 18F-FDG uptake in other rheumatic conditions. Conclusion Intense 18F-FDG uptake adjacent to the ischial tuberosities and of posteromedial knee structures is common among untreated patients with PMR on whole-body PET/CT. Using fusion technology, we have provided the first documentation of a hamstring peritendonitis as the anatomical correlate of abnormalities detected at both of these sites. In doing so, an alternate explanation for the aetiology of the gluteal and posterior thigh pain that typically characterizes PMR is proposed. A high frequency of wrist and/or hand involvement has also been observed for the first time in this study by performing dedicated views on whole-body PET/CT. Abnormal 18F-FDG uptake in a volar distribution at the hand was found to correlate with flexor tenosynovitis on MRI. These findings add to our growing understanding of the imaging phenotype of PMR and further highlight its distinctive pathology, including a predilection for extra-articular structures distal to the shoulder and pelvic girdles. Further studies are now needed to determine the relationship between peripheral abnormalities detected on whole-body PET/CT and the longitudinal outcomes in this subset of PMR patients. Acknowledgements C.E.O. has been supported by research grants received from Arthritis Australia and Austin Medical Research Foundation. Funding: This work was supported by an unrestricted educational grant from AbbVie. Disclosure statement: The authors have declared no conflicts of interest. References 1 Takahashi H, Yamashita H, Kubota K et al.   Differences in fluorodeoxyglucose positron emission tomography/computed tomography findings between elderly onset rheumatoid arthritis and polymyalgia rheumatica. Mod Rheumatol  2015; 25: 546– 51. Google Scholar CrossRef Search ADS PubMed  2 Yamashita H, Kubota K, Takahashi Y et al.   Similarities and differences in fluorodeoxyglucose positron emission tomography/computed tomography findings in spondyloarthropathy, polymyalgia rheumatica and rheumatoid arthritis. Joint Bone Spine  2013; 80: 171– 7. Google Scholar CrossRef Search ADS PubMed  3 Wakura D, Kotani T, Takeuchi T et al.   Differentiation between polymyalgia rheumatica (PMR) and elderly-onset rheumatoid arthritis using 18F-fluorodeoxyglucose positron emission tomography/computed tomography: is enthesitis a new pathological lesion in PMR? PLoS One  [Electronic Resource] 2016; 11: e0158509. Google Scholar CrossRef Search ADS PubMed  4 Blockmans D, De Ceuninck L, Vanderschueren S et al.   Repetitive 18-fluorodeoxyglucose positron emission tomography in isolated polymyalgia rheumatica: a prospective study in 35 patients. Rheumatology  2006; 46: 672– 7. Google Scholar CrossRef Search ADS PubMed  5 Salvarani C, Barozzi L, Boiardi L et al.   Lumbar interspinous bursitis in active polymyalgia rheumatica. Clin Exp Rheumatol  2013; 31: 526– 31. Google Scholar PubMed  6 Salvarani C, Barozzi L, Cantini F et al.   Cervical interspinous bursitis in active polymyalgia rheumatica. Ann Rheum Dis  2008; 67: 758– 61. Google Scholar CrossRef Search ADS PubMed  7 Yamashita H, Kubota K, Takahashi Y et al.   Whole-body fluorodeoxyglucose positron emission tomography/computed tomography in patients with active polymyalgia rheumatica: evidence for distinctive bursitis and large-vessel vasculitis. Mod Rheumatol  2012; 22: 705– 11. Google Scholar CrossRef Search ADS PubMed  8 Cimmino MA, Camellino D, Paparo F et al.   High frequency of capsular knee involvement in polymyalgia rheumatica/giant cell arteritis patients studied by positron emission tomography. Rheumatology  2013; 52: 1865– 72. Google Scholar CrossRef Search ADS PubMed  9 Dasgupta B, Cimmino MA, Kremers HM et al.   2012 Provisional classification criteria for polymyalgia rheumatica: a European League Against Rheumatism/American College of Rheumatology collaborative initiative. Arthritis Rheum  2012; 64: 943– 54. Google Scholar CrossRef Search ADS PubMed  10 Mackie SL, Koduri G, Hill CL et al.   Accuracy of musculoskeletal imaging for the diagnosis of polymyalgia rheumatica: systematic review. RMD Open  2015; 1: e000100. Google Scholar CrossRef Search ADS PubMed  11 Sondag M, Guillot X, Verhoeven F et al.   Utility of 18F-fluoro-dexoxyglucose positron emission tomography for the diagnosis of polymyalgia rheumatica: a controlled study. Rheumatology  2016; 55: 1452– 7. Google Scholar CrossRef Search ADS PubMed  12 Palard-Novello X, Querellou S, Gouillou M et al.   Value of 18F-FDG PET/CT for therapeutic assessment of patients with polymyalgia rheumatica receiving tocilizumab as first-line treatment. Eur J Nucl Med Mol Imaging  2016; 43: 773– 9. Google Scholar CrossRef Search ADS PubMed  13 Rehak Z, Sprlakova-Pukova A, Bortlicek Z et al.   PET/CT imaging in polymyalgia rheumatica: praepubic 18F-FDG uptake correlates with pectineus and adductor longus muscles enthesitis and with tenosynovitis. Radiol Oncol  2017; 51: 8– 14. Google Scholar CrossRef Search ADS PubMed  14 Rehak Z, Vasina J, Nemec P et al.   Various forms of 18F-FDG PET and PET/CT findings in patients with polymyalgia rheumatica. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub  2015; 159: 629– 36. Google Scholar PubMed  15 Hitora T, Kawaguchi Y, Mori M et al.   Ischiogluteal bursitis: a report of three cases with MR findings. Rheumatol Int  2009; 29: 455– 8. Google Scholar CrossRef Search ADS PubMed  16 Cantini F, Niccoli L, Nannini C et al.   Inflammatory changes of hip synovial structures in polymyalgia rheumatica. Clin Exp Rheumatol  2005; 23: 462– 8. Google Scholar PubMed  17 Moore KL., Dally AF. Clinically Oriented Anatomy , 4th edn. Baltimore, MD: Lippincott Williams & Wilkins, 1999. 18 Cimmino MA, Parodi M, Zampogna G, Barbieri F, Garlaschi G. Polymyalgia rheumatica is associated with extensor tendon tenosynovitis but not with synovitis of the hands: a magnetic resonance imaging study. Rheumatology  2011; 50: 494– 9. Google Scholar CrossRef Search ADS PubMed  19 McGonagle D, Pease C, Marzo-Ortega H et al.   Comparison of extracapsular changes by magnetic resonance imaging in patients with rheumatoid arthritis and polymyalgia rheumatica. J Rheumatol  2001; 28: 1837– 41. Google Scholar PubMed  20 Mackie SL, Pease CT, Fukuba E et al.   Whole-body MRI of patients with polymyalgia rheumatica identifies a distinct subset with complete patient-reported response to glucocorticoids. Ann Rheum Dis  2015; 74: 2188– 92. Google Scholar CrossRef Search ADS PubMed  21 Macchioni P, Boiardi L, Catanoso M, Pazzola G, Salvarani C. Performance of the new 2012 EULAR/ACR classification criteria for polymyalgia rheumatica: comparison with the previous criteria in a single-centre study. Ann Rheum Dis  2014; 73: 1190– 3. Google Scholar CrossRef Search ADS PubMed  22 Ozen G, Inanc N, Unal AU et al.   Assessment of the new 2012 EULAR/ACR clinical classification criteria for polymyalgia rheumatica: a prospective multicenter study. J Rheumatol  2016; 43: 893– 900. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oup.com

Journal

RheumatologyOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial