PLOS ONE RESEARCH ARTICLE "The added value of 18f-FDG PET/CT in the assessment of onset and steroid resistant polimyalgia rheumatica" Patricia Moya-Alvarado ID1*, Alejandro Fernandez Leon2, Maria Emilia Corica3, Valle Camacho Marti2, Diego Alfonso López-Mora2, Ivan Castellvı́ ID1, Hèctor Corominas ID1 1 Department of Rheumatology, Hospital Santa Creu i Sant Pau, Barcelona, Spain, 2 Department of Nuclear Medicine, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain, 3 Department of Rheumatology, Hospital José N. Lencinas, Mendoza, Argentina a1111111111 a1111111111 * pmoyaa@santpau.cat a1111111111 a1111111111 a1111111111 Abstract PMR is a common inflammatory rheumatic disease. Although its clinical characteristics are fully recognized, no specific test for its diagnosis has been established to date. Several stud- OPEN ACCESS ies have described a wide variety of diseases that present with polymyalgic symptoms. A 18 Citation: Moya-Alvarado P, Fernandez Leon A, FDG-PET/CT scan could help to deal with these differential diagnoses. The goal of our Corica ME, Camacho Marti V, López-Mora DA, study is to describe the findings of the 18FDG-PET/CT scan in a cohort of PMR patients and Castellvı́ I, et al. (2021) "The added value of 18f- to detail how the 18FDG-PET/CT scan improves accuracy when diagnosing other underlying FDG PET/CT in the assessment of onset and conditions. This cross-sectional study enrolled patients with a diagnosis of PMR who under- steroid resistant polimyalgia rheumatica" PLoS ONE 16(9): e0255131. https://doi.org/10.1371/ went to a 18FDG-PET/CT scan to rule out other diagnosis. The 18FDG-PET/CT scan was journal.pone.0255131 performed either following clinical criteria at the onset of clinical symptoms or when the Editor: Pierpaolo Alongi, Fondazione Istituto G. patient became PMR steroid resistant. Patients’ demographic, clinical and analytical data at Giglio di Cefalu, ITALY the moment of the 18FDG-PET/CT scan were recorded. The final diagnosis was confirmed Received: April 19, 2021 according to clinical judgement. A total of 103 patients with PMR were included. In 49.51% of patients, the 18FDG-PET/CT scan was ordered to study resistance to steroid therapy. Accepted: July 12, 2021 The final diagnoses of patients were PMR in 70.9% patients, large vessel vasculitis in Published: September 24, 2021 15.5%, neoplasms 4.8% and another diagnosis in the rest. The 18FDG-PET/CT scan is a Peer Review History: PLOS recognizes the very useful technique for the study of Polymyalgia Rheumatica, not only to help in the diag- benefits of transparency in the peer review nostic process, but also due to its role in the identification of a variety of PMR-like patrons. process; therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. The editorial history of this article is available here: https://doi.org/10.1371/journal.pone.0255131 Copyright: © 2021 Moya-Alvarado et al. This is an Introduction open access article distributed under the terms of Polymyalgia rheumatica (PMR) is the most common inflammatory rheumatic disease in the Creative Commons Attribution License, which permits unrestricted use, distribution, and patients over 50 years of age. Although the cause of PMR remains unknown, evidence consis- reproduction in any medium, provided the original tently suggests a multifactorial etiology leading to an immunomediated process [1,2]. author and source are credited. Although PMR’s clinical characteristics are fully recognized, no specific test for its diagnosis Data Availability Statement: All relevant data are has been confirmed to date. Several studies have described a wide variety of diseases that pres- within the manuscript and its Supporting ent with polymyalgic symptoms [2–6] and it is important to rule out all these processes, since Information files. their therapy and prognosis differ widely from the classic therapy used for PMR. Moreover, an PLOS ONE | https://doi.org/10.1371/journal.pone.0255131 September 24, 2021 1 / 13 PLOS ONE PMR and PET-CT scan Funding: The authors received no specific funding association between PMR and giant cell arteritis (GCA) has been described. Approximately for this work. 50% of patients with GCA present with the clinical symptoms of PMR [7]. The percentage of Competing interests: The authors have declared patients with PMR who present GCA ranges from five to thirty percent depending on the that no competing interests exist. series [3,8]. In one population-based study, a temporal artery biopsy yielded positive histologic findings of GCA in 9% of patients presenting with typical PMR without any clinical manifesta- tion of GCA [9]. Imaging tests could help in these differential diagnoses. These tests have gained significant prominence since the American College of Rheumatology (ACR) and the European League Against Rheumatism (EULAR) launched new classification criteria that take ultrasonography findings into account [10]. Other imaging techniques that could help in the differential diag- nosis of PMR are magnetic resonance imaging (MRI) and 18-fluorodeoxyglucose positron emission tomography (18FDG-PET/CT). 18FDG-PET/CT is a diagnostic imaging technique that measures metabolic activity by locating and quantifying glucose consumption. It has been used to diagnose and monitor neoplastic processes, though other specialties are now using it to study clinical symptoms such as fever of unknown origin and other processes of an inflam- matory nature. The recognized use of a 18FDG-PET/CT scan to study pathologies character- ized by high glucidic metabolism suggests that it may be a promising technique in the differential diagnostic study of patients with PMR. 18FDG-PET/CT scan findings in PMR may include increased 18FDG uptake in shoulder, hip and spinous processes [1,10]. A small num- ber of series of published cases also describe subclinical vasculitis and other pathological enti- ties of malignant origin in the 18FDG-PET/CT scans of patients with PMR. Nevertheless, the conclusive diagnostic relevance of this technique is as yet unknown when performed to com- plete the study of patients with clinical PMR. Considering all these factors, the goal of our study was to examine the findings of 18 FDG-PET/CT scans in a cohort of PMR patients and to determine whether this imaging technique offers the clinician the possibility of diagnosing underlying conditions. Material and methods Patient groups and clinical assessment A cross-sectional retrospective study was performed in a cohort of PMR patients at the Depart- ment of Rheumatology at a tertiary university hospital. Between April 2011 and April 2018, we enrolled patients who had undergone an 18FDG-PET/CT scan to rule out other diagnoses. PMR was diagnosed by a rheumatologist in accordance with the ACR/EULAR 2012 classifica- tion criteria [11] in all patients included in the study. The study population was classified into two groups: 1) patients with onset PMR (PMR_os), and 2) patients with PMR who were refrac- tory to glucocorticoid therapy (PMR_sr). The first group was defined as patients who were vis- ited in the rheumatology department during the initial diagnostic process or who underwent a 18 FDG-PET/CT scan within six weeks after the diagnosis of PMR. The second group was defined as patients with PMR who did not respond to conventional treatment with steroids or patients who flared when the dose was below 7.5 mg. The criteria for performing the 18 FDG-PET/CT scan were: 1) PMR_os: all patients who presented an onset PMR and were attended by one of the three main investigators (JM, EC, PM); and, 2) PMR_sr: all patients with PMR who were refractory to glucocorticoid therapy independently of which physician attended them. We excluded patients who had a history of neoplasia before the 18FDG-PET/ CT scan was performed. Upon admission to the study, we collected patients’ demographic, clinical and laboratory data. 18FDG-PET/CT scan variables collected were: the joint involved (shoulders, hips, periph- eral joints) and aorta, PET/CT standardized uptake values (SUV), and the tissue to PLOS ONE | https://doi.org/10.1371/journal.pone.0255131 September 24, 2021 2 / 13 PLOS ONE PMR and PET-CT scan background ratio (TBR) of each location according to the average activity of the vena cava. 18 FDG-PET/CT images were analyzed by a qualified nuclear medicine specialist. The final confirmed diagnosis was accepted according to the physician’s opinion consider- ing the following combination of data: a) clinical parameters, b) the results of the supplemental examinations (blood test and 18FDG-PET/CT scan), and c) the outcome of the disease. The study was approved by the local institutional ethics committee—Comité de Ética de investigación con medicamentos de la Fundació de Gestió Sanitaria del Hospital de la Santa Creu i Sant Pau de Barcelona—(IIBSP-VAS-2013-122) and all patients signed an informed consent form before enrollment. 18 FDG-PET/CT assessment All patients had fasted for at least six hours before 18FDG administration. After the intravenous injection of 18FDG (3.7 MBq/kg), they rested for 60 minutes. Images were acquired using a GEMINI PET/CT scanner (PHILIPS Health Systems Amsterdam Holland), integrated with a 64-slice multidetector CT. PET images were obtained for 180 seconds per position. To attenuate correction and to identify anatomical location, we performed a low-dose, non-contrast-enhanced CT scan (tube voltage: 120kV; effective tube current: 30–100 mA), which included the whole body from the top of the skull to the feet. Image analysis It was difficult to discriminate between synovitis and perisynovitis involvement at 18FDG uptake sites in the shoulder region using 18FDG/CT. Therefore, we did not classify shoulder lesions as synovitis or perisynovitis, and all sites thought to correspond to such lesions were regarded as the ‘‘shoulder”. We performed 18FDG PET/CT visual analysis by two experimented Nuclear medicine phy- sicians in order to determine positivity in vasculitis. 18FDG uptake was assessed in large arter- ies, proximal joints (shoulders, hips and sternoclavicular joints) and in extraarticular synovial structures (interspinous, ischiogluteal and praepubic bursae). We used the semi-quantitative values of SUV in each vascular and articular structure. Vascular SUVmax measurements were taken drawing a VOI (volume region of Interest) at the level of the most visually active seg- ment of the aorta and at the same level as the cava venous pool. Articular and synovial VOIs were also evaluated in order to measure the SUV value, taking into account the maximum activity uptake area. Vascular activity was normalized using TBR values (target–to–blood pool ratio) to divide the vascular wall SUV by the venous blood pool SUV to correct for blood uptake [12]. Statistical analysis All data are expressed as mean ± standard deviation (SD). Categorical variables are presented as absolute frequencies and percentages. Comparisons between independent means were ana- lyzed using the Student’s t-test or the Mann-Whitney test using IBM-SPSS, version 25. For the categorical variables, the chi-square test or Fisher’s exact test were used as appropriate. Corre- lations between quantitative variables were analyzed using Pearson’s correlation coefficient. The non-parametric test (Kruskal–Wallis) was used for quantitative variables without a nor- mal distribution. The Mann_Whitney U test was used in the post-hoc study. The level of statis- tical significance was established at 5% (alpha value = 0.05). PLOS ONE | https://doi.org/10.1371/journal.pone.0255131 September 24, 2021 3 / 13 PLOS ONE PMR and PET-CT scan Results A total of 103 patients with PMR (30 men and 73 women) were included in the study. The aver- age age of the patients was 72.48 ± 9.01 years. In 52 (50.48%) patients, the 18FDG-PET/CT scan was performed at the onset of the disease based on clinical opinion. In 51 (49.51%) patients, the 18 FDG-PET/CT scan was requested to study PMR refractory to glucocorticoids. The average dose of steroids (prednisone or equivalent) at the time of scanning was 11.78 ± 1.36 mg orally daily. Table 1 shows patients’ clinical characteristics at the time of the 18FDG-PET/CT scan. The final diagnosis, taking into account the results of the 18FDG-PET/CT scan, analytical parameters and clinical outcome, was PMR in 73 (70.9%) patients, large vessel vasculitis (LVV) in 16 (15.5%), neoplasms in five (4.8%) and other diagnosis (PMR with elderly-onset rheumatoid arthritis [EORA] (Fig 1), Sjögren’s syndrome, small vessel vasculitis and degenera- tive process) in the rest of the sample. Table 2 shows the TBR of the various locations according to the final diagnosis. We found significant differences between the TBR of the aorta in LVV patients and the other patients (p< 0.001). In the posthoc study, statistical significance was maintained between the final diag- nosis of LVV and the TBR in the aorta (p = 0.002). When analyzing clinical and analytical variables, we observed significant differences between the group with PMR as a final diagnosis and the patients with other diagnoses for the variables of weight loss (28.8% vs 56.7%; p = 0.013) and amaurosis (1.4% versus 13.3%; p = 0.024). 18 FDG-PET/CT in PMR & large vessel vasculitis Forty (54.8%) of the patients diagnosed with PMR showed increased glucidic metabolism in the shoulder, 27 (37%) in the hips, 23 (31.5%) in other joints, and 34 (46.6%) in one or more Table 1. Clinical characteristics of patients at the time of 18FDG-PET/CT. Patient’s clinical features Women, n (%) 73 (70.9%) Age (years), mean ± SD 72.4 ± 9.0 Disease duration (months), mean ± SD 24 ± 41.9 Asthenia, n (%) 66 (64.1%) Weight loss, n (%) 38 (36.9%) Fever, n (%) 16 (15.5%) Morning stiffness, n (%) 49 (47.6%) Cervicalgia, n (%) 71 (68.9%) Shoulder pain, n (%) 100 (97.1%) Hip pain, n (%) 89 (86.4%) Arthralgia, n (%) 26 (25.2%) Headache, n (%) 23 (22.3%) Amaurosis, n (%) 5 (4.9%) Jaw claudication, n (%) 9 (8.7%) Temporal artery tenderness, n (%) 2 (1.9%) Reactive C protein (mg/L), mean ± SD 38.0 ± 69.5 Erythrocyte sedimentation rate (mm/s), mean ± SD 55.9 ± 31.0 Hemoglobin (g/L), mean ± SD 118.7 ± 24.9 Prednisone dose (MPD), mean ± SD 11.7 ± 1.3 Duration of PDN before PET/CT (weeks), Median 13 • PMR_os (weeks), Median 2.7 • PMR_sr (weeks), Median 61.1 https://doi.org/10.1371/journal.pone.0255131.t001 PLOS ONE | https://doi.org/10.1371/journal.pone.0255131 September 24, 2021 4 / 13 PLOS ONE PMR and PET-CT scan Fig 1. 18FDG-PET/CT images. A 78-year-old woman with diagnostic of elderly onset rheumatoid arthritis [EORA]. Fused 18F-FDG PET/CT and MIP images showed increased metabolism in aortic wall and supraortic vessels which corresponded to vasculitis. Increased 18F-FDG uptake in peripheral joints was also described with shoulder involvement. Inflammatory activity in the wrists was also observed, as a characteristic arthritis in this pathology. https://doi.org/10.1371/journal.pone.0255131.g001 bursa (Fig 2). Of the 23 patients with joint uptake in the 18FDG-PET/CT scan, the area most frequently affected was the sternoclavicular (23.28%), followed by the knees (10.95%) and the wrist (8.21%). Of the 16 patients diagnosed with LVV, in addition to glucidic uptake increase in the aortic wall, 18FDG PET/CT scans showed a concomitant uptake in the shoulders, hips and bursae. We did not measure vascular diameters in the patients with PMR included in the study. Never- theless, in the visual evaluation, we did not observe dilated vessels. Table 3 shows the sites of 18 FDG accumulations in patients with PMR and LVV. Malignancies Of the five patients who were finally diagnosed with neoplastic processes (four haematological and one transverse colon) and who first presented with PMR, the 18FDG-PET/CT scan showed Table 2. TBR of the different locations according to the final diagnosis. Final diagnosis TBR _shoulder TBR _column TBR _joint TBR-aorta TBR_bursae PMR 2.03±0.9 1.78±0.9 2.74±1.09 1.52±0.217 2.42±0.91 LVV 1.80±0.55 1.95±1.05 2.51±0.77 1.98±0.64 2.77±1.07 NEOPLASIA 1.60±0.69 1.40±0.67 1.58±. 1.46±0.22 1.33±. OTHER 2.38±0.94 1.90±0.79 3.12±2.07 1.92±0.42 3.085±0.6 p P = 0.370 P = 0.472 P = 0.516 P = <0.001 P = 0.104 TBR_bursae: Includes trochanteric and subacromial bursitis. https://doi.org/10.1371/journal.pone.0255131.t002 PLOS ONE | https://doi.org/10.1371/journal.pone.0255131 September 24, 2021 5 / 13 PLOS ONE PMR and PET-CT scan Fig 2. 18FDG-PET/CT images. 1A: Uptake in shoulders and cervical interspinous bursae.1B: Uptake in supra- and infra-diaphragmatic aorta. https://doi.org/10.1371/journal.pone.0255131.g002 accumulations suggestive of neoplastic process in three patients (mediastinal lymphadenopa- thy and increased diffuse glucidic metabolism in bone marrow in two patients with lympho- proliferative syndrome (Fig 3), along with a hypermetabolic lesion in the transverse colon with the diagnosis of a high-grade villous adenoma in the third patient). The 18FDG-PET/CT scan thus helped diagnose three of the 103 patients who had neoplastic processes. Table 3. Distribution of glycemic metabolism intake in patients diagnosed with polymyalgia rheumatica (PMR) and large vessel vasculitis (LVV). Glycemic metabolism intake in PMR and/or Vasculitis N (%) PMR 73 (70.9) Shoulder involvement 40 (54.8) Hip involvement 27 (37.0) Peripheral joints 23 (31.5) • Sternoclavicular 17 (23.2) • Knee 6 (8.2) • Wrist 8 (10.9) Bursae uptake 34 (46.6) • Bursae cervical spine 22 (30.1) • Bursae dorsal spine 3 (4.1) • Bursae lumbar spine 25 (34.2) • Ischiatic bursae 26 (35.6) LVV 16 (15.5) Shoulder involvement 9 (56.3) Hip involvement 6 (37.5) Bursae uptake 11 (68.8) • Bursae cervical spine 9 (56.3) • Bursae lumbar spine 11 (68.8) • Ischiatic bursae 6 (37.5) Table shows distribution of glycemic metabolism intake in all patients in the sample diagnosed with PMR (n = 73) or LVV (n = 16). https://doi.org/10.1371/journal.pone.0255131.t003 PLOS ONE | https://doi.org/10.1371/journal.pone.0255131 September 24, 2021 6 / 13 PLOS ONE PMR and PET-CT scan Fig 3. 18FDG-PET/CT images. A 89-year-old woman 18F-FDG PET/CT with diagnostic of polimyalgia rheumatica. PET/CT images showed multiple mediastinal nodules with increase 18F-FDG uptake and bone marrow glucose hypermetabolism. The patient was finally diagnosed of in a lymphoproliferative process. No treatment was administered due to age and other concomitant pathologies. https://doi.org/10.1371/journal.pone.0255131.g003 18 FDG-PET/CT in onset PMR versus PMR refractory to glucocorticoids When we performed the statistical analysis and compared the clinical variables of patients with PMR_os and patients with PMR_sr, we observed that the former presented more asthenia, weight loss and fever than the latter group. We did not find differences in waist pain, cervical- gia or analytical parameters. Table 4 shows the statistical differences observed in patients with PMR_os and patients with PMR_sr. We classified the final diagnoses into four groups: PMR, LVV, neoplasms and other diagno- ses (including PMR with EORA, Sjögren syndrome, small vessel vasculitis and a degenerative process). When we compared these final diagnoses considering whether they were PMR_os or PMR_sr, we found significant differences in the final diagnosis (p = 0.045) (Table 5). Discussion The present study was designed to rule out a wide spectrum of differential diagnoses in a group of patients studied by 18FDG-PET/CT. We included 103 patients with an initial diagno- sis of PMR. The confirmed diagnosis in our cohort was PMR in 70.9%, LVV in 15.5%, neo- plasms in 4.8%, and other diagnoses in the remaining 8.8%. Several studies corroborate the already known relationship between PMR and LVV studied by PET/CT scan. But unlike our study, the main objective of most of these studies was not to analyse the wide range of differential diagnoses presented by the PMR but to point out certain specifities of LVV and PMR. To date, few articles have addressed the issue of the broad range of differential diagnoses presented by PMR by PET/CT scan. PLOS ONE | https://doi.org/10.1371/journal.pone.0255131 September 24, 2021 7 / 13 PLOS ONE PMR and PET-CT scan Table 4. Clinical features of patients: a) with onset PMR vs steroid resistant PMR b) with the final diagnosis of PMR, LVV or malignancy. PMR global LVV Malignancies PMR_os PMR_sr p Women, n (%) 23 (31.5%) 4 (25%) 2 (40.0%) 33 (63.5%) 40 (78.4%) 0.129 Asthenia, n (%) 49 (67.1%) 11 (68.8%) 3 (60.0%) 40(76.9%) 26 (51.0%) 0.006� Weight loss, n (%) 21 (28.7%) 10 (62.5%) 4 (80.0%) 25 (48.1%) 13 (25.5%) 0.017� Fever, n (%) 11 (15.1%) 2 (12.5%) 0 (0%) 12 (23.1%) 4 (7.8%) 0.029� Morning stiffness, n (%) 38 (52.1%) 6 (37.5%) 2 (40.0%) 27 (51.9%) 22 (43.1%) 0.37 Cervicalgia, n (%) 52 (71.2%) 12 (75%) 2 (40.0%) 34 (65.4%) 37 (72.5%) 0.432 Shoulder pain, n (%) 71 (97.3%) 15 (93.8%) 5 (100%) 51 (98.1%) 49 (96.1%) 0.543 Hip pain, n (%) 66 (90.4%) 11(68.8%) 5 (100%) 48 (92.3%) 41 (80.4%) 0.074 Headache, n (%) 13 (17.8%) 7 (43.8%) 1 (20.0.%) 11 (21.2%) 12 (23.5%) 0.772 Amaurosis, n (%) 1 (1.4%) 4 (25%) 0 (0%) 2 (3.8%) 3 (5.9%) 0.630 Jaw claudication, n (%) 4 (5.5%) 5 (31.3%) 0 (0.0%) 4 (7.7%) 5 (9.8%) 0.704 Reactive C protein (mg/L), mean ± SD 29.7±46.1 43.3±45.3 48.9±72.17 44.8 ± 59.5 31.1 ± 78.3 0.287 Erythrosedimentation (mm/s), mean ± SD 53.2±3.1 69.3±28 59±39.8 59.3 ± 29.4 52.4 ± 32.5 0.663 Hemoglobin (g/L), mean ± SD 120.3±25.4 116.8±12.1 118±9.01 117.4 ± 23.2 120.0 ± 26.7 0.880 Prednisone dose, mean ± SD 11.5±11.64 11.0±12.2 6.6±11.5 10.7 ± 11.9 11.4 ± 10.8 0.124 PMR_os: Onset PMR patients; PMR_sr: Steroid resistant PMR patients; PMR global: PMR_os and PM_sr, LVV: Large vessel vasculitis. https://doi.org/10.1371/journal.pone.0255131.t004 Henckaerts et al. [13] prospectively included 99 consecutive patients with a possible clinical diagnosis of PMR. All patients underwent 18FDG-PET scanning before treatment with ste- roids. As in our study, the gold standard for a diagnosis of PMR was the judgment of an experi- enced clinician. A final diagnosis of isolated PMR was made in 67.6% of the patients (as in our study), while another condition was diagnosed in the remaining 32.32% of patients. Diagnoses made in non-PMR patients were for a variety of diseases that canpresent with similar clinical symptoms. Malignancies In our cohort, five patients (4.8%) presented neoplastic processes (four hematological neo- plasms and one transverse colon high-grade villous adenoma). The 18FDG-PET/CT scan helped diagnose malignant processes in three of the 103 patients included. Published data to date on this topic are scarce. In one recent paper, in a cohort of 99 patients with PMR, Henck- aerts L et al. [13] found one patient who presented a paraneoplastic syndrome secondary to a carcinoid neoplasm (visible on PET scan). In another study, Palard Novello et al. [14] studied 21 patients with “new onset PMR” in order to evaluate the use of the 18FDG-PET/CT scan in assessing tocilizumab treatment and found that one patient presented a malignant process in the 18FDG-PET/CT scan pre-treatment. In a large database study from the Swedish Hospital Discharge Register, Ji J et al. [16] included 5,918 patients with GCA and PMR and reported a clear increased risk of cancer within the first year of PMR diagnosis. Similarly, Muller et al. Table 5. Final diagnosis of patients with onset PMR (PMR_os) vs steroid resistant (PMR_sr). Final diagnosis PMR LVV NEOPLASIA OTHER PMR_os, n (%) 37 (50.7) 6 (37.5) 5 (100) 4 (44.4) PMR_sr, n (%) 36 (49.3) 10 (62.5) 0 (0) 5 (55.6) Total 73 16 5 9 p = 0.045. PMR: Polymyalgia rheumatica; LVV: Large vessel vasculitis. https://doi.org/10.1371/journal.pone.0255131.t005 PLOS ONE | https://doi.org/10.1371/journal.pone.0255131 September 24, 2021 8 / 13 PLOS ONE PMR and PET-CT scan [15] described a 69% increased risk of malignancy in patients with PMR within the first six months of diagnosis. One explanation for the high risk observed in their study may be that the high rate of health care consumption allowed for a higher rate of cancer detection. In contrast with the results of Ji J et al, several studies found no association [14–19] with an increase in the risk of cancer after PMR diagnosis. It is difficult to reach a firm conclusion or confirm this association because most studies included have a referral bias, a lack of control groups, and/or a limited follow up period. Although the association between PMR and malignant processes is well known, results from published articles and reviews do not offer consistent or strong evidence. Our data show a non-significant rate of malignant processes, but we cannot definitely conclude whether or not this rate is higher than that in the population without PMR because we did not have a control group. Large vessel vasculitis Multiple published papers describe the relationship between LVV and PMR studied by PET-CT, but the range of percentages of this association varies widely. For instance, Henck- aerts et al. [13] described associations of around 2%, which is very different from the 65% described by Lavado-Perez et al. and Prieto-Peña et al. [20,21]. Prieto-Peña studied 84 patients with classic PMR. A PET/CT scan was positive for LVV in 51 (60.7%) patients. The differences observed in the incidence of LVV described by Prieto-Peña [21] compared to our results could be explained by the differences in the populations studied. Prieto et al’s patients were all steroid resistant. When classifying the patients in our sample as steroid-resistant or onset we observed that the frequency of LVV in the PMR_sr was 62.5% compared to 37.5% in the PMR_os group, suggesting a consistency with the results of the Prieto-Peña study. Prieto-Peña also studied whether there was an association between LVV and clinical and analytical variables. They described that pelvic girdle pain, inflammatory low back pain and lower limb pain were predictors of a positive 18FDG-PET/CT scan result for LVV in patients with PMR. However, they did not find an association with any analytical parameters or with the presence of constitutional symptoms. Like Prieto- Peña, we found no-significant associa- tion between LVV and PMR in patients who presented analytical alterations. Conversely, Gon- zalez Gay et al. [22] reported that patients with PMR associated with GCA had significant alterations in erythrocyte sedimentation rate, platelet count and hemoglobin compared to patients with isolated PMR. Furthermore, as in Prieto-Peña et al’s study., the work of Lavado-Perez et al. [20] described 40 patients diagnosed with PMR using PET/CT scan who showed a high prevalence of LVV (65%), which is considerably higher than our prevalence. Once again, this difference could be explained by the fact that 33 of their 40 patients had suspected LVV before 18FDG-PET/CT, strengthening the notion that despite being PMR, the cohorts of patients studied and pub- lished to date with 18FDG-PET/CT scan are not homogenous. In contrast, Henckaerts et al. [13] reported fewer cases of LVV than those described by pre- vious studies. They found that only 15% of PMR patients had an increased FDG uptake in the larger thoracic vessels, compared with 6% of control patients. Of the patients with a vascular 18 FDG uptake, two were diagnosed with GCA based on a positive temporal artery biopsy. They attributed these findings to blood pool activity or no activity in the vessel wall, and to defects in the technique in the collection of images in some of the scans. Concerning TBR aorta data, in our cohort, the prevalence of TBR of the aorta in the patients included in the “other group” was similar to that in patients in the LLV group. There may be several plausible explanations for this. The first is the small size and heterogeneity of PLOS ONE | https://doi.org/10.1371/journal.pone.0255131 September 24, 2021 9 / 13 PLOS ONE PMR and PET-CT scan the "other group". Furthermore, the final diagnosis was based on 18FDG-PET/CT results and the presence of additional symptoms (clinical, analytical and evolutionary data). For these rea- sons the patients were not labelled LVV. Taken together, it is evident that the frequency of association between PMR and LVV dif- fers significantly depending on the characteristics of the cohorts studied. The time of disease evolution, the treatment administered and the presence of new symptoms of LVV are variables that must be considered when describing the association between PMR and LVV studied by 18 FDG-PET/CT. Distribution of increased glucidic metabolism in 18FDG-PET/CT scan In our study, the majority of patients diagnosed with PMR showed an increase in glucidic metabolism in the shoulders, followed by hips and peripheral joints. Almost half of the patients showed uptake in a bursa. The most frequently affected peripheral areas were the sternoclavi- cular, followed by the carpus and the knees. In a restrospective study of 50 PMR patients undergoing a 18FDG-PET/CT scan, Sondag et al. [23] found greater uptake in the shoulders than in the hips, with percentages of distribu- tion that were very similar to ours (54% and 36% for shoulders and hips, respectively). Two other authors, Rehak et al. [24] and Henckaerts et al. [13], also observed greater uptake in the shoulders than in the hips, though their percentages were considerably higher than ours. Rehak et al. evaluated 67 patients and found that 87% and 70%, respectively, of the patients showed uptake in the shoulders and hips. Henckaerts et al. evaluated 67 patients and also described a greater uptake in the shoulders than the hips than in our study (97% in shoulders and 67% in hips compared with our figures of 54.8% and 37%, respectively). This higher per- centage of hip and shoulder involvement in Rehak’s and Henckaert’s studies [13,24] compared to our findings, as well as those reported by Sondag [23], could be because no patients in their study were undergoing steroid treatment. We also analyzed the 18FDG-PET/CT scan uptake frequency in bursae. We observed a 35.6% uptake in the ischial bursa in patients with PMR, frequencies that are lower than those described by the groups in the studies by Rehak [24], Sondag [22] and Henckaerts [13], who found an uptake from 52% to 67% in patients with PMR. Likewise, we observed uptake in 30.1% and 34.2% of cervical and lumbar spinous processes, respectively. We also found a simi- lar frequency and involvement when comparing lumbar and cervical processes. Rehak and Sondag found a greater involvement in lumbar processes (57% and 38% respectively) than in cervical processes (19.5% and 10% respectively). Interestingly, we found that dorsal processes were involved in 4.1% of patients, a finding that had not been described previously. Our results showed peripheral joint uptake. This is controversial considering the possible differential diagnosis with EORA in patients with PMR. Although we describe 8.2% of involve- ment in knees and 10.95% in wrists, the final diagnosis of PMR stands, though not that of EORA. A few authors describe peripheral joint uptake in detail in patients with PMR [25,26]. Kaneko et al. [27] studied 20 patients with PMR, describing an exceptionally high frequency of knee involvement (96.2%). However, only four patients complained of knee symptoms. The same group also described a high frequency of involvement of the wrists, with figures of up to 40%. Cimmino et al. [28] reported a higher frequency of knee involvement (84%) in a study using 18FDG-PET/CT. On the other hand, in 2018, Yuge et al. [25] studied 16 patients with a definitive diagnosis of PMR and described a significant incidence (6%) of 18FDG uptake in the wrists. Considering the small number of patients in these previous studies, it is difficult to reach a firm conclusion concerning peripheral uptake. We still need to define those findings in pure PMR and observe whether they can develop into EORA over time. PLOS ONE | https://doi.org/10.1371/journal.pone.0255131 September 24, 2021 10 / 13 PLOS ONE PMR and PET-CT scan In our cohort, FDG-PET was helpful to rule out vasculitis, malignancy and peripheral arthritis. Nevertheless, the TBRs in the typical PMR locations (spine/shoulders/hips) did not differ significantly from those in the”other diagnosis” group. TBR at typical PMR locations can be similar in groups for two reasons. First, other diagnoses, such as LVV and neoplasms, can be accompanied by PMR, and second, the group of “other diagnoses” is small, making it more difficult to find statistically significant differences. Limitations This study has several limitations. The main limitation is its retrospective character. Neverthe- less, our results clearly reflect observations in our daily clinical practice. Second, as some patients were under treatment with steroids at the time of the 18FDG-PET/CT scan, our results cannot be compared statistically to studies involving steroid-naïve patients only. A third limi- tation is the lack of a control group. However, we would not have been able to compare the diagnosis found in patients with PMR using a 18FDG-PET/CT scan with the general popula- tion for ethical and procedural reasons. Fourth, it has been suggested that a late PET scan at 180 minutes increase the accuracy of the method in diagnosing LVV, especially for the evalua- tion of the thoracic aorta; but unfortunately this scan is not available in our cohort. Fifht, the inclusion of patients who concomitantly presented GCA symptoms may entail a bias. Conclusion Based on our findings, we consider the 18FDG-PET/CT scan to be a very useful tool, not only to help diagnose PMR, but also to identify diseases such as LVV and malignant processes that are associated with this disorder. However, the best moment to perform this technique (PMR_os or PMR_sr) remains to be defined. Supporting information S1 File. (PDF) Author Contributions Conceptualization: Patricia Moya-Alvarado, Alejandro Fernandez Leon, Maria Emilia Corica. Data curation: Patricia Moya-Alvarado, Alejandro Fernandez Leon, Maria Emilia Corica, Valle Camacho Marti, Diego Alfonso López-Mora. Formal analysis: Patricia Moya-Alvarado, Diego Alfonso López-Mora, Ivan Castellvı́. Investigation: Patricia Moya-Alvarado, Alejandro Fernandez Leon. Methodology: Patricia Moya-Alvarado, Alejandro Fernandez Leon. Supervision: Ivan Castellvı́, Hèctor Corominas. Validation: Alejandro Fernandez Leon, Hèctor Corominas. Writing – original draft: Patricia Moya-Alvarado, Hèctor Corominas. References 1. González-Gay MA, Matteson EL, Castañeda S. Polymyalgia rheumatica. Lancet 2017; 390:1700– 1712. https://doi.org/10.1016/S0140-6736(17)31825-1 PMID: 28774422 PLOS ONE | https://doi.org/10.1371/journal.pone.0255131 September 24, 2021 11 / 13 PLOS ONE PMR and PET-CT scan 2. Pipitone N, Salvarani C. Update on polymyalgia rheumatica. Eur J Intern Med. 2013; 24: 583–589. https://doi.org/10.1016/j.ejim.2013.03.003 PMID: 23579169 3. Gonzalez-Gay MA, Garcia-Porrua C, Salvarani C, et al. Diagnostic approach in a patient presenting with polymyalgia. Clin Exp Rheumatol. 1999; 17: 276–8. PMID: 10410257 4. Gonzalez-Gay MA, Garcia-Porrua C, Salvarani C, et al. The spectrum of conditions mimicking poly- myalgia rheumatica in Northwestern Spain. J Rheumatol. 2000; 27: 2179–84. PMID: 10990231 5. Deal CL, Meenan RF, Goldenberg DL, et al. The clinical features of elderly-onset rheumatoid arthritis. A comparison with younger-onset disease of similar duration. Arthritis Rheum. 1985; 28: 987–94. https:// doi.org/10.1002/art.1780280905 PMID: 4038365 6. Salvarani C, Cantini F, Macchioni P, et al. Distal musculoskeletal manifestations in polymyalgia rheu- matica: a prospective followup study. Arthritis Rheum. 1998; 41: 1221–6. https://doi.org/10.1002/1529- 0131(199807)41:7<1221::AID-ART12>3.0.CO;2-W PMID: 9663479 7. Salvarani C, Gabriel SE, O’Fallon WM, et al. The incidence of giant cell arteritis in Olmsted County, Min- nesota: apparent fluctuations in a cyclic pattern. Ann Intern Med. 1995; 123: 192–4. https://doi.org/10. 7326/0003-4819-123-3-199508010-00006 PMID: 7598301 8. Gran JT, Myklebust G, Wilsgaard T, et al. Survival in polymyalgia rheumatica and temporal arteritis: a study of 398 cases and matched population controls. Rheumatology (Oxford) 2001; 40:1238. https:// doi.org/10.1093/rheumatology/40.11.1238 PMID: 11709607 9. Gonzalez-Gay MA. Giant cell arteritis and polymyalgia rheumatica: two different but often overlapping conditions. Semin. Arthritis Rheum. 2004; 33: 289–93. https://doi.org/10.1016/j.semarthrit.2003.09. 007 PMID: 15079759 10. Dasgupta B, Cimmino MA, Maradit-Kremers H, et al. 2012 provisional classification criteria for poly- myalgia rheumatica: a European League Against Rheumatism/American College of Rheumatology col- laborative initiative. Ann Rheum Dis. 2012; 71: 484–92. https://doi.org/10.1136/annrheumdis-2011- 200329 PMID: 22388996 11. 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. Rheuma- tology 2006; 46: 672–677. https://doi.org/10.1093/rheumatology/kel376 PMID: 17114803 12. Slart, Riemer H J Aet al. FDG-PET/CT(A) imaging in large vessel vasculitis and polymyalgia rheuma- tica: joint procedural recommendation of the EANM, SNMMI, and the PET Interest Group (PIG), and endorsed by the ASNC. Eur J Nucl Med Mol Imaging 2018; 45:1250–69. https://doi.org/10.1007/ s00259-018-3973-8 PMID: 29637252 13. Henckaerts L, Gheysens O, Vanderschueren S, et al. Use of 18F-fluorodeoxyglucose positron emission tomography in the diagnosis of polymyalgia rheumatica-A prospective study of 99 patients. Rheumatol- ogy (Oxford) 2018; 57: 1908–1916. 14. Palard-Novello X, Querellou S, Gouillou M, et al. Value of (18)F-FDG PET/CT for therapeutic assess- ment of patients with polymyalgia rheumatica receiving tocilizumab as first-line treatment. Eur J Nucl Med Mol Imaging. 2016; 43: 773–9. https://doi.org/10.1007/s00259-015-3287-z PMID: 26753600 15. Muller S, Hider SL, Belcher J, et al. Is cancer associated with polymyalgia rheumatica? A cohort study in the General Practice Research Database. Ann Rheum Dis. 2014; 73: 1769–1773. https://doi.org/10. 1136/annrheumdis-2013-203465 PMID: 23842460 16. Ji J, Liu X, Sundquist K, Sundquist J, et al. Cancer risk in patients hospitalized with polymyalgia rheuma- tica and giant cell arteritis: a follow-up study in Sweden. Rheumatology (Oxford) 2010; 49: 1158–63. https://doi.org/10.1093/rheumatology/keq040 PMID: 20299378 17. Haga HJ, Eide GE, Brun J, et al. Cancer in association with polymyalgia rheumatica and temporal arteri- tis. J Rheumatol. 1993; 20: 1335–9. PMID: 8230015 18. Myklebust G, Wilsgaard T, Jacobsen BK, et al. No increased frequency of malignant neoplasms in poly- myalgia rheumatica and temporal arteritis. A prospective longitudinal study of 398 cases and matched population controls. J Rheumatol. 2002; 29: 2143–7. PMID: 12375324 19. Pfeifer EC, Crowson CS, Major BT, et al. Polymyalgia Rheumatica and its Association with Cancer. Rheumatology (Sunnyvale). 2015; Suppl 6: 003. https://doi.org/10.4172/2161-1149.S6-003 PMID: 26688777 20. Lavado-Pérez C, Martı́nez-Rodrı́guez I, Martı́nez-Amador N, et al. (18)F-FDG PET/CT for the detection of large vessel vasculitis in patients with polymyalgia rheumatica. Rev Esp Med Nucl Imagen Mol.; 34: 275–81. https://doi.org/10.1016/j.remn.2015.05.011 PMID: 26159505 21. Prieto-Peña D, Martı́nez-Rodrı́guez I, Loricera J, et al. Predictors of positive 18F-FDG PET/CT-scan for large vessel vasculitis in patients with persistent polymyalgia rheumatica. Semin Arthritis Rheum. 2019; 48: 720–727. https://doi.org/10.1016/j.semarthrit.2018.05.007 PMID: 29903537 PLOS ONE | https://doi.org/10.1371/journal.pone.0255131 September 24, 2021 12 / 13 PLOS ONE PMR and PET-CT scan 22. González-Gay MA, Garcı́a-Porrúa C, Vázquez-Caruncho M. Polymyalgia rheumatica in biopsy proven giant cell arteritis does not constitute a different subset but differs from isolated polymyalgia rheumatica. J. Rheumatol 1998; 25: 1750–5. PMID: 9733456 23. 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. 2015; 159: 629–636. 24. Yamashita H, Kubota K, Takahashi Y, et al. Whole-body fluorodeoxyglucose positron emission tomog- raphy/computed tomography in patients with active polymyalgia rheumatica: evidence for distinctive bursitis and large-vessel vasculitis. Mod Rheumatol. 2012; 22: 705–11. https://doi.org/10.1007/ s10165-011-0581-x PMID: 22205118 25. Sondag M, Guillot X, Verhoeven F, et al. Utility of 18F-fluoro-dexoxyglucose positron emission tomogra- phy for the diagnosis of polymyalgia rheumatica: a controlled study. Rheumatology (Oxford) 2016; 55: 1452–7. 26. Yuge S, Nakatani K, Yoshino K, et al. Diagnosing polymyalgia rheumatica on 18F-FDG PET/CT: typical uptake patterns. Ann Nucl Med. 2018; 32: 573–577. https://doi.org/10.1007/s12149-018-1269-5 PMID: 29948622 27. Kaneko K, Suematsu E, Miyamura T, et al. Differences of articular and extra-articular involvement in polymyalgia rheumatica: A comparison by whole-body FDG-PET/CT. Mod Rheumatol. 2019; 7: 1–7. https://doi.org/10.1080/14397595.2019.1591065 PMID: 30843747 28. 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 (Oxford) 2013; 52: 1865–72. https://doi.org/10.1093/rheumatology/ket229 PMID: 23850896 PLOS ONE | https://doi.org/10.1371/journal.pone.0255131 September 24, 2021 13 / 13