A cohort study of comorbidity in patients with granulomatosis with polyangiitis

A cohort study of comorbidity in patients with granulomatosis with polyangiitis Abstract Objectives To evaluate the associations between granulomatosis with polyangiitis (GPA) and a wide range of comorbidities in patients with GPA compared with population-based non-vasculitis patients. Methods Using the UK-based Clinical Practice Research Datalink we identified 570 incident patients with GPA in 1990–2014, and for each case, we selected up to 10 random non-vasculitis patients matched for age, sex, practice and years of history before the date of first GPA diagnosis. We compared the distribution of 13 pre-existing comorbidities and estimated the risk of each incident comorbidity after the cohort entry between GPA and non-vasculitis patients. Results Patients with GPA were more likely to have a history of hypothyroidism at their initial diagnosis compared with non-vasculitis patients. Patients with GPA had increased risk of incident venous thromboembolism any time during follow-up compared with non-vasculitis patients, particularly during the first 3 years of follow-up: hazard ratio (HR) of 5.24 (95% CI: 2.83, 9.71). Risks were also increased for hypertension (HR = 2.45, 95% CI: 1.84, 3.26), type 2 diabetes (HR = 2.13, 95% CI: 1.36, 3.32), dyslipidaemia (HR = 1.98, 95% CI: 1.29, 3.04) and depression (HR = 1.77, 95% CI: 1.10, 2.86) among GPA patients during the first 3 years of follow-up, but not after 3 years post-diagnosis. Conclusion Patients with GPA had a higher prevalence of hypothyroidism before the initial diagnosis, a strong risk of developing venous thromboembolism during follow-up and increased risks of hypertension, dyslipidaemia, type 2 diabetes and depression in the first few years after diagnosis, compared with non-vasculitis population. granulomatosis with polyangiitis, vasculitis, comorbidity, Clinical Practice Research Datalink, cohort Rheumatology key messages Granulomatosis with polyangiitis patients had more prior hypothyroidism than non-vasculitis patients. Granulomatosis with polyangiitis patients had an increased risk of developing venous thromboembolism compared with non-vasculitis patients. Risks were increased for hypertension, dyslipidaemia, type 2 diabetes and depression within 3 years after granulomatosis with polyangiitis diagnosis. Introduction Granulomatosis with polyangiitis (GPA, formerly known as Wegener’s granulomatosis) is the most common type of ANCA-associated vasculitis (AAV). Characteristic pathological features of GPA include arteritis of small-sized to medium-sized blood vessels, and granulomatous inflammation of the upper airways and lungs [1]. While GPA used to be frequently fatal, the introduction of CYC and glucocorticoid treatment, and more recently rituximab, greatly improved the survival rates and prognosis of GPA [2–4]. Despite this, data from the European Vasculitis Study Group trials showed a significant burden of disease manifestations involving renal and otolaryngolocical complications, and potentially treatment-related damage (including hypertension, malignancy and diabetes) in patients with AAV, with 34.4% of patients having ⩾5 areas of damage at a mean of 7 years post-diagnosis [5]. Increased carotid intima–media thickness, which indicates atherosclerosis, has also been demonstrated in GPA patients [6, 7], potentially as a consequence of glucocorticoid therapy, and GPA patients were reported to experience an increased number of cardiovascular events due to ischaemic heart disease (IHD) compared with the background population [8]. GPA patients were also reported to have increased risks of venous thromboembolism (VTE) [9, 10] and some specific cancers compared with population controls [11, 12]. Nevertheless, the epidemiology of comorbidity (including pre-existing and post-diagnosis) in GPA remains poorly understood. Therefore, we conducted this study to evaluate the associations between GPA and a wide range of comorbidities in a cohort of GPA patients compared with a matched non-vasculitis cohort, and to examine whether associations differ according to sex and age. Methods Data source Data were derived from the UK-based Clinical Practice Research Datalink (CPRD), a large primary care database that has been previously described in detail [13, 14]. Briefly, the CPRD is an ongoing longitudinal database that has collected data from over 500 general practices in the UK since 1987. It covers ∼6% of the UK population and has a UK representative age and sex distribution. The information recorded includes patient demographic characteristics, clinical diagnoses, drug prescriptions, consultant referrals, lab results and hospitalizations. Validation studies have shown the high quality of recorded diagnoses and drug exposures [13, 14]. This study was approved by the Independent Scientific Advisory Committee for Medicines and Healthcare products Regulatory Agency database research (protocol no: 15_051). Study design and base population This cohort study was conducted using CPRD data from 1 January 1990 through 31 December 2014. We identified all patients with a first-time diagnosis of GPA who had at least 1 year of recorded medical history before the first GPA diagnosis code (providing a window to ensure an incident diagnosis and to capture pre-existing comorbidities). For each GPA patient, up to 10 random patients without any forms of vasculitis were matched on age (±2 years), sex, general practice attended and years of history before cohort entry (date of first recorded GPA diagnosis or corresponding date for the matched non-vasculitis patients). GPA definition We identified all patients with a Read code for Wegener’s granulomatosis (the former eponym for GPA) during the study period, and categorized each incident case into one of three levels of likelihood (probable, possible or unlikely) according to the pre-determined algorithm for GPA cases (Table 1). This algorithm was developed by two of the authors (L.L. and T.N.) based on the definition of GPA described in a prior CPRD publication where 90% of patients with a code for GPA were confirmed to have a true diagnosis [15]. Note that the clinical judgement formulated from this process was based on an overall clinical impression from available information, such as recorded symptoms, treatments and referrals, rather than on clinical criteria, elements of which are typically not available in electronic medical records. We excluded unlikely cases from this study. Because some patients died or transferred out of their practice shortly after the initial diagnosis, they may not have had a chance to meet our GPA definition. To examine the impact of incomplete follow-up, we included patients whose data collection ended within 6 months of the cohort entry date and who did not meet the GPA definition (4.9% of GPA patients). See supplementary Fig. S1, available at Rheumatology Online, for GPA patient selection. Table 1 Classification algorithm for granulomatosis with polyangiitis Category  Criteria  Probablea (met at least one of these criteria)  Had a GPA diagnosis from a specialist (consultant) or a discharge letter with a GPA diagnosis Presented with specific symptoms suggestive of granulomatous disease affecting the upper and lower respiratory tract [15] within 6 months before or on index date (i.e. date of first diagnosis of GPA), plus either ≥1 prescriptions for immunosuppressive therapy within 6 months before or on the index date or any time after index date, or ≥2 diagnoses of GPA on different dates Had either ≥2 prescriptions for immunosuppressive therapy within 6 months before or on the index date or any time after index date, or ≥3 diagnoses of GPA on different dates in the absence of specific symptoms suggestive of granulomatous disease affecting the upper and lower respiratory tract [15] within 6 months before or on index date  Possiblea  When no probable case criteria stated above were met, a case was considered possible if they met both (i) and (ii): Had ≥2 prescriptions for glucocorticoid therapy, where one was within 6 months of the diagnosis date, and the two prescriptions were within 6 months of each other Presented with specific symptoms suggestive of granulomatous disease affecting the upper and lower respiratory tract [15] within 6 months before or on index date, or had anti-neutrophil cytoplasmic antibodies tests, chest X-ray, chest CT scan, sinus CT scan, relevant biopsy within 6 months before or on index date or 1 month after index date  Unlikely  All other cases do not meet probable or possible criteria  Category  Criteria  Probablea (met at least one of these criteria)  Had a GPA diagnosis from a specialist (consultant) or a discharge letter with a GPA diagnosis Presented with specific symptoms suggestive of granulomatous disease affecting the upper and lower respiratory tract [15] within 6 months before or on index date (i.e. date of first diagnosis of GPA), plus either ≥1 prescriptions for immunosuppressive therapy within 6 months before or on the index date or any time after index date, or ≥2 diagnoses of GPA on different dates Had either ≥2 prescriptions for immunosuppressive therapy within 6 months before or on the index date or any time after index date, or ≥3 diagnoses of GPA on different dates in the absence of specific symptoms suggestive of granulomatous disease affecting the upper and lower respiratory tract [15] within 6 months before or on index date  Possiblea  When no probable case criteria stated above were met, a case was considered possible if they met both (i) and (ii): Had ≥2 prescriptions for glucocorticoid therapy, where one was within 6 months of the diagnosis date, and the two prescriptions were within 6 months of each other Presented with specific symptoms suggestive of granulomatous disease affecting the upper and lower respiratory tract [15] within 6 months before or on index date, or had anti-neutrophil cytoplasmic antibodies tests, chest X-ray, chest CT scan, sinus CT scan, relevant biopsy within 6 months before or on index date or 1 month after index date  Unlikely  All other cases do not meet probable or possible criteria  a All cases with other causes (such as microscopic polyangiitis, Goodpasture’s syndrome, etc.) were excluded. GPA: granulomatosis with polyangiitis. Study outcomes We assessed 13 study outcomes: IHD (including myocardial infarction, angina pectoris and unspecified IHD), stroke/transient ischaemic attacks (TIA), peripheral vascular disease (PVD), VTE, heart failure, hypertension, dyslipidaemia, type 2 diabetes, cancer (except non-melanoma skin cancer), depression, psoriasis, chronic obstructive pulmonary disease and hypothyroidism. These study outcomes were selected because either they have been reported to be associated with GPA or its treatment, or they were commonly recorded in the study population after we screened GPA patient electronic records. Codes are available upon request. Covariates We collected information on smoking (current, former, never or unknown), BMI and alcoholism (yes or no), and exposure to aspirin, anti-hypertensive medicine, glucose lowering agents, lipid-lowering medicine, NSAIDs and antidepressants, defined by the presence of two or more prescriptions within 1 year before cohort entry. Statistical analysis We identified the first date that each study outcome was recorded and compared it with the cohort entry date to distinguish pre-existing and incident comorbidities. For pre-existing conditions, we compared frequencies any time before or on the cohort entry date between all GPA and non-vasculitis patients using the chi-square test or Fisher’s exact test. To account for delay in GPA diagnosis, we also calculated frequencies of pre-existing comorbidities at >1 year before the cohort entry date among those who had ⩾2 years of medical history. We compared the cumulative incidence of each incident study outcome at 1, 3, 5 and 10 years of follow-up between GPA patients and non-vasculitis patients using Gray’s test, adjusted for the competing risk of death [16, 17]. Patients with a pre-existing study outcome were excluded from the corresponding analyses of incident disease. Each eligible patient was followed from cohort entry until the date of the first of the following: the first diagnosis of a study outcome, the patient transferred out of the practice, the end of data collection or death. We used Cox regression to estimate for each study outcome the hazard ratio (HR) comparing GPA patients with non-vasculitis patients. Due to small sample size we reported herein only the ‘crude’ HRs and 95% CIs, which nevertheless, by design, take into account age, sex, practice and calendar time. For study outcomes that violated the proportional hazard assumption we reported the HRs stratified by length of follow-up (<3 and ⩾3 years). Finally, we repeated all main analyses in subpopulations stratified by sex and age (<60 and ⩾60 years). To examine the robustness of the results we re-analysed the primary results restricted to probable GPA patients and their corresponding comparison patients. All analyses were conducted using SAS statistical software, version 9.3 (SAS Institute, Cary, NC, USA), and the statistical package R with an add-on cmprsk package, version 3.1.1 (www.r-project.org). Results There were 570 GPA patients (including 512 probable cases, 30 possible cases and 28 cases with <6 months’ follow-up who did not meet the GPA definition) and 5389 non-vasculitis comparison patients in this study. Table 2 shows the baseline characteristics of patients in both groups. GPA patients were less likely to be current smokers, but more likely to be former smokers, than non-vasculitis patients. GPA patients were more likely to receive NSAIDs, but less likely to receive lipid-lowering medicine within one year before cohort entry compared with non-vasculitis patients. Mean follow-up was 5.9 years among GPA patients and 7.0 years among non-vasculitis patients. Among GPA patients, 166 (29%) patients died and 147 (26%) transferred out of the practice before the end of data collection, while 788 (15%) died and 1927 (36%) transferred out among non-vasculitis patients. Table 2 Baseline characteristics of patients with granulomatosis with polyangiitis and comparison patients Characteristics  GPA patients  Comparison patients  P-value  (n = 570)  (n = 5389)  Age, mean (s.d.), years  58.4 (15.6)  58.4 (15.5)  NAa      <60 years  264 (46.3)  2480 (46.0)  NAa      ≥60 years  306 (53.7)  2909 (54.0)  Sex      Female  254 (44.6)  2426 (45.0)  NAa      Male  316 (55.4)  2963 (55.0)  Smoking      Never  255 (44.7)  2379 (44.1)  0.0084      Current  81 (14.2)  944 (17.5)      Former  168 (29.5)  1302 (24.2)      Unknown  66 (11.6)  764 (14.2)  BMI      <18.5 kg/m2  14 (2.5)  74 (1.4)  0.21      18.5–24.9 kg/m2  134 (23.5)  1312 (24.3)      25.0–29.9 kg/m2  143 (25.1)  1456 (27.0)      ≥30 kg/m2  106 (18.6)  910 (16.9)      Unknown  173 (30.4)  1637 (30.4)  Alcoholism  16 (2.8)  179 (3.3)  0.51  Co-medications      Aspirin  57 (10.0)  644 (12.0)  0.17      Anti-hypertensive medicine  192 (33.7)  1752 (32.5)  0.57      Anti-diabetes medicine  30 (5.3)  315 (5.8)  0.57      Lipid-lowering medicine  69 (12.1)  869 (16.1)  0.012      NSAIDs  125 (21.9)  566 (10.5)  <0.0001      Anti-depressants  64 (11.2)  527 (9.8)  0.27  Year of cohort entry date      1990–94  49 (8.6)  480 (8.9)  NAa      1995–99  95 (16.7)  911 (16.9)      2000–04  127 (22.3)  1198 (22.2)      2005–09  166 (29.1)  1548 (28.7)      2010–14  133 (23.3)  1252 (23.2)  Length of recorded medical history, mean (s.d.), years      Before cohort entry  10.0 (6.1)  10.0 (6.1)  NAa      Follow-up after cohort entry  5.9 (4.9)  7.0 (5.4)  <0.0001  Characteristics  GPA patients  Comparison patients  P-value  (n = 570)  (n = 5389)  Age, mean (s.d.), years  58.4 (15.6)  58.4 (15.5)  NAa      <60 years  264 (46.3)  2480 (46.0)  NAa      ≥60 years  306 (53.7)  2909 (54.0)  Sex      Female  254 (44.6)  2426 (45.0)  NAa      Male  316 (55.4)  2963 (55.0)  Smoking      Never  255 (44.7)  2379 (44.1)  0.0084      Current  81 (14.2)  944 (17.5)      Former  168 (29.5)  1302 (24.2)      Unknown  66 (11.6)  764 (14.2)  BMI      <18.5 kg/m2  14 (2.5)  74 (1.4)  0.21      18.5–24.9 kg/m2  134 (23.5)  1312 (24.3)      25.0–29.9 kg/m2  143 (25.1)  1456 (27.0)      ≥30 kg/m2  106 (18.6)  910 (16.9)      Unknown  173 (30.4)  1637 (30.4)  Alcoholism  16 (2.8)  179 (3.3)  0.51  Co-medications      Aspirin  57 (10.0)  644 (12.0)  0.17      Anti-hypertensive medicine  192 (33.7)  1752 (32.5)  0.57      Anti-diabetes medicine  30 (5.3)  315 (5.8)  0.57      Lipid-lowering medicine  69 (12.1)  869 (16.1)  0.012      NSAIDs  125 (21.9)  566 (10.5)  <0.0001      Anti-depressants  64 (11.2)  527 (9.8)  0.27  Year of cohort entry date      1990–94  49 (8.6)  480 (8.9)  NAa      1995–99  95 (16.7)  911 (16.9)      2000–04  127 (22.3)  1198 (22.2)      2005–09  166 (29.1)  1548 (28.7)      2010–14  133 (23.3)  1252 (23.2)  Length of recorded medical history, mean (s.d.), years      Before cohort entry  10.0 (6.1)  10.0 (6.1)  NAa      Follow-up after cohort entry  5.9 (4.9)  7.0 (5.4)  <0.0001  Except where otherwise indicated, data are shown as n (%). a Not applicable because of matching factors. GPA: granulomatosis with polyangiitis. GPA patients were more likely to have a history of VTE, heart failure and hypothyroidism compared with non-vasculitis patients at the cohort entry. However, these differences were no longer present except for hypothyroidism at > 1 year before the cohort entry (Table 3). When we stratified by either sex or age, there were no material changes in the results (data not shown). When we restricted the analysis to probable GPA patients and their non-vasculitis comparison patients, the main findings did not change (see supplementary Table S1, available at Rheumatology Online). Table 3 Distributions of pre-existing comorbidities in patients with granulomatosis with polyangiitis and comparison patients Comorbidities  At cohort entry  At 1 year before cohort entrya  GPA patients  Comparison patients  P-value  GPA patients  Comparison patients  P-value  (n = 570)  (n = 5389)  (n = 527)  (n = 4983)  IHD  43 (7.5)  389 (7.2)  0.78  37 (7.0)  345 (6.9)  0.93  Stroke/TIA  20 (3.5)  168 (3.1)  0.61  12 (2.3)  143 (2.9)  0.43  PVD  9 (1.6)  84 (1.6)  0.97  8 (1.5)  77 (1.5)  0.96  VTE  15 (2.6)  76 (1.4)  0.024  6 (1.1)  69 (1.4)  0.64  Heart failure  16 (2.8)  89 (1.7)  0.046  11 (2.1)  70 (1.4)  0.22  Hypertension  140 (24.6)  1159 (21.5)  0.093  111 (21.1)  1032 (20.7)  0.85  Dyslipidaemia  45 (7.9)  412 (7.6)  0.83  41 (7.8)  363 (7.3)  0.68  Type 2 diabetes  42 (7.4)  382 (7.1)  0.80  30 (5.7)  324 (6.5)  0.47  Cancer  19 (3.3)  191 (3.5)  0.80  16 (3.0)  159 (3.2)  0.85  Depression  77 (13.5)  706 (13.1)  0.78  65 (12.3)  629 (12.6)  0.85  Psoriasis  20 (3.5)  160 (3.0)  0.47  17 (3.2)  143 (2.9)  0.64  COPD  13 (2.3)  158 (2.9)  0.38  11 (2.1)  131 (2.6)  0.46  Hypothyroidism  38 (6.7)  186 (3.5)  0.0001  33 (6.3)  159 (3.2)  0.0003  Comorbidities  At cohort entry  At 1 year before cohort entrya  GPA patients  Comparison patients  P-value  GPA patients  Comparison patients  P-value  (n = 570)  (n = 5389)  (n = 527)  (n = 4983)  IHD  43 (7.5)  389 (7.2)  0.78  37 (7.0)  345 (6.9)  0.93  Stroke/TIA  20 (3.5)  168 (3.1)  0.61  12 (2.3)  143 (2.9)  0.43  PVD  9 (1.6)  84 (1.6)  0.97  8 (1.5)  77 (1.5)  0.96  VTE  15 (2.6)  76 (1.4)  0.024  6 (1.1)  69 (1.4)  0.64  Heart failure  16 (2.8)  89 (1.7)  0.046  11 (2.1)  70 (1.4)  0.22  Hypertension  140 (24.6)  1159 (21.5)  0.093  111 (21.1)  1032 (20.7)  0.85  Dyslipidaemia  45 (7.9)  412 (7.6)  0.83  41 (7.8)  363 (7.3)  0.68  Type 2 diabetes  42 (7.4)  382 (7.1)  0.80  30 (5.7)  324 (6.5)  0.47  Cancer  19 (3.3)  191 (3.5)  0.80  16 (3.0)  159 (3.2)  0.85  Depression  77 (13.5)  706 (13.1)  0.78  65 (12.3)  629 (12.6)  0.85  Psoriasis  20 (3.5)  160 (3.0)  0.47  17 (3.2)  143 (2.9)  0.64  COPD  13 (2.3)  158 (2.9)  0.38  11 (2.1)  131 (2.6)  0.46  Hypothyroidism  38 (6.7)  186 (3.5)  0.0001  33 (6.3)  159 (3.2)  0.0003  Except where otherwise indicated, data are shown as n (%). aRestricted to patients with at least 2 years of medical history before cohort entry. COPD: chronic obstructive pulmonary disease; GPA: granulomatosis with polyangiitis; IHD: ischaemic heart disease; PVD: peripheral vascular disease; TIA: transient ischaemic attack; VTE: venous thromboembolism. The cumulative incidences for VTE, hypertension and type 2 diabetes were significantly higher in GPA patients than those in non-vasculitis patients. There were no significant differences in the cumulative incidences for other outcomes under study between GPA patients and non-vasculitis patients (Table 4). Consistent with the cumulative incidence findings, GPA patients had no increased risks for IHD, stroke/TIA, PVD and cancer compared with non-vasculitis patients (Table 5). GPA patients had a strong risk for VTE compared with non-vasculitis patients, particularly during the first 3 years of follow-up: HR = 5.24 (95% CI: 2.83, 9.71). GPA patients also had increased risks for hypertension (HR = 2.45, 95% CI: 1.84, 3.26), type 2 diabetes (HR = 2.13, 95% CI: 1.36, 3.32), dyslipidaemia (HR = 1.98, 95% CI: 1.29, 3.04) and depression (HR = 1.77, 95% CI: 1.10, 2.86) during the first 3 years of follow-up, but not after 3 years post-diagnosis, compared with non-vasculitis patients (Table 6). Table 4 Estimated cumulative incidences of study outcomes at specified times after cohort entry Incident comorbidity  Cumulative incidencea, per 100 (95% CI)  1 year  3 years  5 years  10 years  Gray’s test, P-value  Ischaemic heart disease      GPA  0.40 (0.08, 1.36)  1.73 (0.82, 3.26)  3.13 (1.74, 5.16)  5.36 (3.28, 8.17)  0.29      Comparison  0.86 (0.63, 1.15)  2.67 (2.22, 3.18)  4.07 (3.49, 4.71)  7.86 (6.90, 8.89)  Stroke/transient ischaemic attack      GPA  0.96 (0.37, 2.13)  2.27 (1.20, 3.90)  3.31 (1.93, 5.28)  5.15 (3.18, 7.80)  0.73      Comparison  0.50 (0.33, 0.73)  1.92 (1.55, 2.35)  2.69 (2.23, 3.21)  5.71 (4.90, 6.61)  Peripheral vascular disease      GPA  NAb  0.21 (0.02, 1.14)  0.99 (0.33, 2.40)  0.99 (0.33, 2.40)  0.63      Comparison  0.26 (0.14, 0.43)  0.42 (0.27, 0.64)  0.84 (0.60, 1.16)  2.10 (1.62, 2.67)  Venous thromboembolism      GPA  2.27 (1.24, 3.82)  3.16 (1.88, 4.96)  5.18 (3.40, 7.50)  6.11 (4.10, 8.67)  <0.0001      Comparison  0.24 (0.13, 0.40)  0.66 (0.46, 0.93)  1.13 (0.84, 1.48)  2.51 (1.97, 3.13)  Heart failure      GPA  0.95 (0.36, 2.10)  1.78 (0.88, 3.25)  3.11 (1.77, 5.05)  4.21 (2.50, 6.60)  0.33      Comparison  0.48 (0.32, 0.70)  1.34 (1.04, 1.71)  2.02 (1.63, 2.48)  4.06 (3.39, 4.82)  Hypertension      GPA  8.55 (6.09, 11.5)  15.4 (12.0, 19.2)  19.9 (15.9, 24.2)  27.8 (22.5, 33.3)  <0.0001      Comparison  2.34 (1.91, 2.84)  6.77 (5.99, 7.61)  10.9 (9.85, 12.0)  20.3 (18.7, 21.9)  Dyslipidaemia      GPA  2.00 (1.03, 3.53)  5.64 (3.78, 8.02)  6.74 (4.65, 9.35)  9.05 (6.39, 12.3)  0.54      Comparison  0.98 (0.73, 1.30)  2.90 (2.44, 3.43)  4.68 (4.05, 5.37)  8.43 (7.46, 9.48)  Type 2 diabetes      GPA  2.71 (1.55, 4.38)  4.93 (3.24, 7.12)  6.55 (4.51, 9.10)  10.6 (7.54, 14.1)  0.020      Comparison  1.04 (0.78, 1.35)  2.37 (1.96, 2.85)  3.97 (3.40, 4.60)  7.42 (6.51, 8.41)  Cancer      GPA  1.13 (0.47, 2.35)  3.10 (1.81, 4.95)  3.62 (2.18, 5.61)  7.17 (4.67, 10.4)  0.075      Comparison  0.99 (0.74, 1.30)  2.90 (2.44, 3.41)  4.91 (4.28, 5.60)  10.4 (9.29, 11.5)  Depression      GPA  2.12 (1.09, 3.74)  4.51 (2.84, 6.73)  6.67 (4.54, 9.36)  8.39 (5.79, 11.6)  0.51      Comparison  0.87 (0.63, 1.18)  2.91 (2.42, 3.45)  4.85 (4.19, 5.58)  8.90 (7.86, 10.0)  Psoriasis      GPA  0.38 (0.08, 1.28)  0.38 (0.08, 1.28)  0.64 (0.18, 1.77)  1.27 (0.47, 2.85)  0.23      Comparison  0.26 (0.15, 0.44)  0.87 (0.63, 1.18)  1.24 (0.94, 1.61)  2.16 (1.70, 2.71)  Chronic obstructive pulmonary disease      GPA  0.95 (0.36, 2.11)  1.38 (0.62, 2.72)  2.62 (1.43, 4.43)  4.87 (2.91, 7.55)  0.98      Comparison  0.32 (0.19, 0.51)  1.35 (1.05, 1.72)  2.27 (1.84, 2.75)  5.22 (4.44, 6.09)  Hypothyroidism      GPA  0.39 (0.08, 1.34)  1.58 (0.70, 3.10)  3.18 (1.78, 5.24)  4.56 (2.62, 7.30)  0.49      Comparison  0.24 (0.13, 0.41)  1.01 (0.75, 1.33)  1.82 (1.44, 2.27)  3.27 (2.68, 3.95)  Incident comorbidity  Cumulative incidencea, per 100 (95% CI)  1 year  3 years  5 years  10 years  Gray’s test, P-value  Ischaemic heart disease      GPA  0.40 (0.08, 1.36)  1.73 (0.82, 3.26)  3.13 (1.74, 5.16)  5.36 (3.28, 8.17)  0.29      Comparison  0.86 (0.63, 1.15)  2.67 (2.22, 3.18)  4.07 (3.49, 4.71)  7.86 (6.90, 8.89)  Stroke/transient ischaemic attack      GPA  0.96 (0.37, 2.13)  2.27 (1.20, 3.90)  3.31 (1.93, 5.28)  5.15 (3.18, 7.80)  0.73      Comparison  0.50 (0.33, 0.73)  1.92 (1.55, 2.35)  2.69 (2.23, 3.21)  5.71 (4.90, 6.61)  Peripheral vascular disease      GPA  NAb  0.21 (0.02, 1.14)  0.99 (0.33, 2.40)  0.99 (0.33, 2.40)  0.63      Comparison  0.26 (0.14, 0.43)  0.42 (0.27, 0.64)  0.84 (0.60, 1.16)  2.10 (1.62, 2.67)  Venous thromboembolism      GPA  2.27 (1.24, 3.82)  3.16 (1.88, 4.96)  5.18 (3.40, 7.50)  6.11 (4.10, 8.67)  <0.0001      Comparison  0.24 (0.13, 0.40)  0.66 (0.46, 0.93)  1.13 (0.84, 1.48)  2.51 (1.97, 3.13)  Heart failure      GPA  0.95 (0.36, 2.10)  1.78 (0.88, 3.25)  3.11 (1.77, 5.05)  4.21 (2.50, 6.60)  0.33      Comparison  0.48 (0.32, 0.70)  1.34 (1.04, 1.71)  2.02 (1.63, 2.48)  4.06 (3.39, 4.82)  Hypertension      GPA  8.55 (6.09, 11.5)  15.4 (12.0, 19.2)  19.9 (15.9, 24.2)  27.8 (22.5, 33.3)  <0.0001      Comparison  2.34 (1.91, 2.84)  6.77 (5.99, 7.61)  10.9 (9.85, 12.0)  20.3 (18.7, 21.9)  Dyslipidaemia      GPA  2.00 (1.03, 3.53)  5.64 (3.78, 8.02)  6.74 (4.65, 9.35)  9.05 (6.39, 12.3)  0.54      Comparison  0.98 (0.73, 1.30)  2.90 (2.44, 3.43)  4.68 (4.05, 5.37)  8.43 (7.46, 9.48)  Type 2 diabetes      GPA  2.71 (1.55, 4.38)  4.93 (3.24, 7.12)  6.55 (4.51, 9.10)  10.6 (7.54, 14.1)  0.020      Comparison  1.04 (0.78, 1.35)  2.37 (1.96, 2.85)  3.97 (3.40, 4.60)  7.42 (6.51, 8.41)  Cancer      GPA  1.13 (0.47, 2.35)  3.10 (1.81, 4.95)  3.62 (2.18, 5.61)  7.17 (4.67, 10.4)  0.075      Comparison  0.99 (0.74, 1.30)  2.90 (2.44, 3.41)  4.91 (4.28, 5.60)  10.4 (9.29, 11.5)  Depression      GPA  2.12 (1.09, 3.74)  4.51 (2.84, 6.73)  6.67 (4.54, 9.36)  8.39 (5.79, 11.6)  0.51      Comparison  0.87 (0.63, 1.18)  2.91 (2.42, 3.45)  4.85 (4.19, 5.58)  8.90 (7.86, 10.0)  Psoriasis      GPA  0.38 (0.08, 1.28)  0.38 (0.08, 1.28)  0.64 (0.18, 1.77)  1.27 (0.47, 2.85)  0.23      Comparison  0.26 (0.15, 0.44)  0.87 (0.63, 1.18)  1.24 (0.94, 1.61)  2.16 (1.70, 2.71)  Chronic obstructive pulmonary disease      GPA  0.95 (0.36, 2.11)  1.38 (0.62, 2.72)  2.62 (1.43, 4.43)  4.87 (2.91, 7.55)  0.98      Comparison  0.32 (0.19, 0.51)  1.35 (1.05, 1.72)  2.27 (1.84, 2.75)  5.22 (4.44, 6.09)  Hypothyroidism      GPA  0.39 (0.08, 1.34)  1.58 (0.70, 3.10)  3.18 (1.78, 5.24)  4.56 (2.62, 7.30)  0.49      Comparison  0.24 (0.13, 0.41)  1.01 (0.75, 1.33)  1.82 (1.44, 2.27)  3.27 (2.68, 3.95)  a Cumulative incidences were calculated after taking into account competing risk due to death. b Not applicable because of zero case. Table 5 Hazard ratios for granulomatosis with polyangiitis in relation to risks of study outcomes Incident comorbidity  GPA patients  Comparison patients  Hazard ratio (95% CI)  Number of events  Number of patients  Number of events  Number of patients  IHD  24  527  284  4996  0.91 (0.60, 1.38)  Stroke/TIA  22  549  228  5212  1.08 (0.70, 1.67)  PVD  7  561  81  5305  0.96 (0.45, 2.09)  VTE  30  554  108  5308  NAa  Heart failure  21  554  166  5300  1.46 (0.93, 2.30)  Hypertension  91  425  633  4223  NAa  Dyslipidaemia  37  524  323  4974  NAa  Type 2 diabetes  43  527  284  5003  NAa  Cancer  32  549  419  5188  0.86 (0.60, 1.23)  Depression  36  492  307  4670  NAa  Psoriasis  5  550  84  5229  0.66 (0.27, 1.63)  COPD  20  557  188  5231  1.18 (0.74, 1.87)  Hypothyroidism  16  532  134  5203  1.41 (0.84, 2.37)  Incident comorbidity  GPA patients  Comparison patients  Hazard ratio (95% CI)  Number of events  Number of patients  Number of events  Number of patients  IHD  24  527  284  4996  0.91 (0.60, 1.38)  Stroke/TIA  22  549  228  5212  1.08 (0.70, 1.67)  PVD  7  561  81  5305  0.96 (0.45, 2.09)  VTE  30  554  108  5308  NAa  Heart failure  21  554  166  5300  1.46 (0.93, 2.30)  Hypertension  91  425  633  4223  NAa  Dyslipidaemia  37  524  323  4974  NAa  Type 2 diabetes  43  527  284  5003  NAa  Cancer  32  549  419  5188  0.86 (0.60, 1.23)  Depression  36  492  307  4670  NAa  Psoriasis  5  550  84  5229  0.66 (0.27, 1.63)  COPD  20  557  188  5231  1.18 (0.74, 1.87)  Hypothyroidism  16  532  134  5203  1.41 (0.84, 2.37)  a Did not meet proportional hazard assumption during the whole follow-up period. COPD: chronic obstructive pulmonary disease; GPA: granulomatosis with polyangiitis; IHD: ischaemic heart disease; PVD: peripheral vascular disease; TIA: transient ischaemic attack; VTE: venous thromboembolism. Table 6 Hazard ratiosa for granulomatosis with polyangiitis in relation to risks of study outcomesb stratified by follow-up length Incident comorbidity  Follow-up length  <3 years  ≥3 years  Venous thromboembolism  5.24 (2.83, 9.71)  2.56 (1.44, 4.54)  Hypertension  2.45 (1.84, 3.26)  1.29 (0.89, 1.86)  Dyslipidaemia  1.98 (1.29, 3.04)  0.68 (0.37, 1.24)  Type 2 diabetes  2.13 (1.36, 3.32)  1.30 (0.81, 2.09)  Depression  1.77 (1.10, 2.86)  1.01 (0.60, 1.68)  Incident comorbidity  Follow-up length  <3 years  ≥3 years  Venous thromboembolism  5.24 (2.83, 9.71)  2.56 (1.44, 4.54)  Hypertension  2.45 (1.84, 3.26)  1.29 (0.89, 1.86)  Dyslipidaemia  1.98 (1.29, 3.04)  0.68 (0.37, 1.24)  Type 2 diabetes  2.13 (1.36, 3.32)  1.30 (0.81, 2.09)  Depression  1.77 (1.10, 2.86)  1.01 (0.60, 1.68)  The data are hazard ratio (95% CI). a Adjusted for age, sex and year of cohort entry. b These study outcomes did not meet proportional hazard assumption during the whole follow-up period. When we stratified by sex, female patients with GPA had an increased risk of heart failure compared with their non-vasculitis patients. When we stratified by age, patients diagnosed with GPA at age <60 years had increased risks of heart failure and hypothyroidism compared with non-vasculitis patients. Patients with GPA <60 years also had stronger risk of VTE than patients with GPA diagnosed after age 60 years. Patients with GPA ⩾60 years did not have significantly increased risks of hypertension, dyslipidaemia, type 2 diabetes or depression compared with non-vasculitis patients even during the first 3 years of follow-up (supplementary Tables S2–S4, available at Rheumatology Online). When we restricted the analysis to probable GPA patients and their non-vasculitis comparison patients, the main findings did not change (supplementary Tables S5–S9, available at Rheumatology Online). Discussion In this population-based cohort study, GPA patients had increased risks of incident VTE after GPA diagnosis compared with the matched non-vasculitis patients, particularly within the first 3 years post-diagnosis, consistent with the initial description of this phenomenon [10]. Risks were also increased for hypertension, type 2 diabetes, dyslipidaemia and depression among GPA patients during the first 3 years of follow-up. Pre-existing hypothyroidism was more common in GPA patients at their initial diagnosis, and even at > 1 year before the initial diagnosis. GPA has been reported to be associated with an increased risk of VTE [9, 10, 18–20]. Merkel et al. analysed data from the Wegener’s Granulomatosis Etanercept Trial and reported that the incidence rate of VTE among GPA patients was 7.0/100 person-years, markedly higher than the incidence rates of VTE in the general population and populations of patients with either lupus or RA, which ranged from 0.3 to 1.0/100 person-years [10]. Faurschou et al. [9] analysed data from a Danish tertiary care centre and found that within the first 2 years following the GPA diagnosis, incidence rate ratios for pulmonary embolism and deep venous thrombosis were 25.7 and 20.2 comparing GPA patients with age- and sex-matched population controls. Consistent with these findings, GPA patients in our study had a strong risk of VTE, especially during the first 3 years of follow-up, which suggests that the risk was strongest proximate to periods of active disease. We also found that pre-existing VTE (at cohort entry) was more common in GPA patients than in non-vasculitis patients, but only in the 1 year before cohort entry. Because GPA may not be diagnosed immediately upon onset of symptoms, VTEs that occurred in the year before cohort entry likely occurred after the disease onset during a period of high disease activity, but before the diagnosis was made. The pathogenetic background for the high VTE risk in GPA is unknown, but in another two studies of patients with AAV the increased risk of VTE could not be explained by traditional clinical risk factors [18, 19]. In a retrospective analysis involving 1130 patients with systemic necrotizing vasculitides, Allenbach et al. identified old age, male sex, previous VTE and stroke with motor deficit as risk factors for VTE [20]. In our study, age and sex were adjusted in the analyses, patients with prior VTE were excluded in the incident analysis, and there was no significant difference in the frequency of prior stroke between GPA patients and non-vasculitis patients. Therefore, the increased risk of VTE could not be explained by these risk factors. A previous study reported a close correlation between disease activity in GPA patients and activated coagulation. The authors also reported that an elevation of the markers of activated haemostasis was reversed when remission was obtained with specific treatment [21]. Despite this, patients with AAV in remission were still reported to be more procoagulant than healthy controls [22]. The observation that there is a stronger risk of VTE during the first 3 years post-diagnosis indirectly suggests the development of VTE is more likely to occur in patients with active or recently active vasculitis and disease- and/or treatment-related factors may play important roles in the increased risk of VTE in GPA. Further studies are needed to evaluate the extent to which the risk of VTE is associated with the disease itself or with treatment interventions. In our study, GPA patients compared with non-vasculitis patients had an increased risk of hypertension during the first 3 years of follow-up, but not after 3 years post-diagnosis. Hypertension was one of the most common medical conditions in GPA patients [23], which may be related to the typical glucocorticoid therapy used to treat GPA, renal involvement or a combination of the two [24]. In addition, GPA patients also had increased risks of dyslipidaemia, type 2 diabetes and depression compared with non-vasculitis patients during the first 3 years post-diagnosis, but not after 3 years, which is also potentially related to glucocorticoid therapy. De Leeuw et al. [25] observed that diastolic blood pressure and prevalence of dyslipidaemia significantly decreased during long-term follow-up in 23 GPA patients. Hajj-Ali et al. [26] reported that depression occurred more often in GPA patients compared with the general population. Recently Englund et al. [27] reported that AAV was associated with increased rates of several comorbidities including hypertension, diabetes and thyroid diseases compared with the general population, although they did not distinguish the prevalent and incident comorbidities under study. Faurschou et al. [28] also reported that GPA patients had an increased risk of new-onset diabetes in the first year of observation period, but not after. Note that the observation period started with the date of the second GPA-related hospital contact where the time between first-ever hospitalization for GPA and date of study inclusion did not exceed 18 months [28]. Since GPA patients were more likely to have medical visits early in the course of their disease than non-vasculitis patients, we cannot rule out the possibility that GPA patients had more opportunities to be examined and diagnosed with these comorbidities in the early stage of active disease. When we conducted analyses stratified by sex and age, we found that female patients with GPA or patients with GPA aged <60 years had an increased risk of heart failure compared with their non-vasculitis comparison patients, which could be explained by the lower background risk in female or younger patients because old age and male sex are independent risk factors for heart failure [29]. We also observed that younger patients with GPA (those <60 years) had an increased risk of new hypothyroidism, and pre-existing hypothyroidism was significantly more common in GPA patients compared with non-vasculitis patients even >1 year before cohort entry. Previously thyroid disease was reported be associated with ANCA small-vessel vasculitis [30]. Whether hypothyroidism is a risk factor for developing GPA needs further study. In our study GPA patients did not have increased risks of IHD, stroke/TIA or PVD compared with non-vasculitis patients. These findings did not change when we repeated the analyses stratified by age, sex and length of follow-up (<3 and ⩾3 years) or restricted to patients with probable GPA and their non-vasculitis comparison patients. In contrast, Faurschou et al. [8] using Danish hospital data, reported that GPA patients experienced an increased number of both early (<5 years after GPA diagnosis) and late (⩾10 years after GPA diagnosis) cardiovascular events (including myocardial infarction, angina pectoris and other IHD) compared with the background population. However, the same study group did not find an increased risk of stroke in GPA patients compared with matched population controls [9]. Similarly, Avina-Zubieta et al. [31] using population-based data from Canada reported that GPA patients had a significantly increased risk of myocardial infarction, not ischaemic stroke, compared with the age-, sex- and entry time-matched population controls and the risk was highest during the first year after GPA diagnosis. Because proteinase-3 ANCA has been associated with a reduced cardiovascular risk compared with myeloperoxidase ANCA or negative ANCA in patients with AAV [32], differences in proportions of ANCA specificities may in part explain these differences. Additionally, a cardioprotective effect of aggressive immunosuppression with CYC in cohorts of SLE patients has been reported [33, 34]. If GPA is similar in this regard to SLE, the risk of cardiovascular disease in GPA could be dependent on the relative balance of proatherogenic effects of sustained inflammation versus therapy-related control of inflammation. Differential management of GPA with regard to rapidity of glucocorticoid tapering and management of traditional risk factors of cardiovascular disease in GPA may also explain these different findings. We did not find an increased risk of incident cancer among GPA patients, nor a difference in prior cancer between GPA patients and non-vasculitis patients. GPA patients have been reported to experience a greater than expected number of specific malignancies including non-melanoma skin cancer, bladder cancer and myeloid leukaemia among patients exposed to cumulative CYC doses >36 g, while the cancer risk among CYC-naïve patients was not significantly increased [11, 12]. Moreover, the same study group did not find evidence of an increased prevalence of pre-existing cancer in GPA patients compared with age- and sex-matched controls [35]. Our study did not assess non-melanoma skin cancer, nor did we evaluate the relationship between GPA and specific cancer types taking into account CYC treatment. We did not have sufficient data available in CPRD to evaluate differences based on era of therapeutic management to compare the CYC-predominant management era vs the more recent time period when rituximab was the favoured first-line therapy. Our study has several strengths worth emphasizing. First, use of the CPRD, one of the world’s largest data resources that collects longitudinal medical history and drug information from general practices, allowed us to efficiently distinguish between pre-existing and incident morbidity and to explore many possible contributing factors involved with the burden of the disease. Second, our study assessed 13 study outcomes, allowing us to better understand comorbidity profiles in GPA patients. In this study, we not only confirmed that GPA was associated with strong risk of VTE, but also demonstrated that GPA was associated with increased risks of hypertension, dyslipidaemia, type 2 diabetes and depression in the first 3 years after diagnosis. Third, as GPA is a rare disease, reliable epidemiological studies on subpopulations of sex and age are limited. We were able to explore comorbidities in different subpopulations with GPA. There are some limitations that need to be considered in interpreting the results of this study. First, there may be misclassification of GPA cases and study outcomes, though the diagnosis codes for GPA and most of the study outcomes have been previously validated in the CPRD and their positive predictive values were reported to range from 70.3 to 98.6% [14, 15]. Second, there was a risk of detection bias in this study whereby GPA patients may have had more medical examinations and thus been more likely to be diagnosed with comorbidities, particularly in the first few years following diagnosis during a period of active disease. However, we found that GPA patients did not have increased risks of cardiovascular events or cancer; therefore, these results are not likely to be explained solely by detection bias. Third, since we only controlled for age, sex, calendar time and general practice in this study, there is a possibility of confounding due to uncontrolled or unknown confounders. Fourth, if GPA patients died shortly after the initial diagnosis, they may not meet the GPA definition. Therefore, it is likely to underestimate the risk of potentially early-occurring lethal comorbidities. However, in the original analyses we included any cases with <6 months’ follow-up regardless of whether they met the GPA definition or not and the main findings did not change after we excluded such cases, suggesting such bias is minimal. Finally, we were unable to investigate the associations between GPA and incident comorbidities of interest stratified by GPA treatment because immunosuppressive therapy is often prescribed by specialists and thus is not completely captured in the GP record, a limitation of the CPRD. In conclusion, GPA patients were confirmed to have strong risk of VTE post-diagnosis, and increased risks of hypertension, dyslipidaemia, type 2 diabetes and depression in the first few years after diagnosis, compared with a matched non-vasculitis population. Funding: This study was supported in part by the Vasculitis Foundation. Disclosure statement: The authors have declared no conflicts of interest. Supplementary data Supplementary data are available at Rheumatology Online. References 1 Seo P, Stone JH. The antineutrophil cytoplasmic antibody-associated vasculitides. Am J Med  2004; 117: 39– 50. Google Scholar CrossRef Search ADS PubMed  2 Fauci AS, Haynes BF, Katz P, Wolff SM. Wegener's granulomatosis: prospective clinical and therapeutic experience with 85 patients for 21 years. Ann Intern Med  1983; 98: 76– 85. Google Scholar CrossRef Search ADS PubMed  3 Hoffman GS, Kerr GS, Leavitt RY et al.   Wegener granulomatosis: an analysis of 158 patients. Ann Intern Med  1992; 116: 488– 98. Google Scholar CrossRef Search ADS PubMed  4 Stone JH, Merkel PA, Spiera R et al.   Rituximab versus cyclophosphamide for ANCA-associated vasculitis. N Engl J Med  2010; 363: 221– 32. Google Scholar CrossRef Search ADS PubMed  5 Robson J, Doll H, Suppiah R et al.   Damage in the anca-associated vasculitides: long-term data from the European vasculitis study group (EUVAS) therapeutic trials. Ann Rheum Dis  2015; 74: 177– 84. Google Scholar CrossRef Search ADS PubMed  6 de Leeuw K, Sanders JS, Stegeman C et al.   Accelerated atherosclerosis in patients with Wegener's granulomatosis. Ann Rheum Dis  2005; 64: 753– 9. Google Scholar CrossRef Search ADS PubMed  7 Nienhuis HL, de Leeuw K, Smit AJ et al.   Enhanced endothelium-dependent microvascular responses in patients with Wegener's granulomatosis. J Rheumatol  2007; 34: 1875– 81. Google Scholar PubMed  8 Faurschou M, Mellemkjaer L, Sorensen IJ et al.   Increased morbidity from ischemic heart disease in patients with Wegener's granulomatosis. Arthritis Rheum  2009; 60: 1187– 92. Google Scholar CrossRef Search ADS PubMed  9 Faurschou M, Obel N, Baslund B. High risk of pulmonary embolism and deep venous thrombosis but not of stroke in granulomatosis with polyangiitis (Wegener's). Arthritis Care Res  2014; 66: 1910– 4. Google Scholar CrossRef Search ADS   10 Merkel PA, Lo GH, Holbrook JT et al.   Brief communication: high incidence of venous thrombotic events among patients with Wegener granulomatosis: the Wegener's Clinical Occurrence of Thrombosis (WeCLOT) Study. Ann Intern Med  2005; 142: 620– 6. Google Scholar CrossRef Search ADS PubMed  11 Faurschou M, Sorensen IJ, Mellemkjaer L et al.   Malignancies in Wegener's granulomatosis: incidence and relation to cyclophosphamide therapy in a cohort of 293 patients. J Rheumatol  2008; 35: 100– 5. Google Scholar PubMed  12 Faurschou M, Mellemkjaer L, Voss A et al.   Prolonged risk of specific malignancies following cyclophosphamide therapy among patients with granulomatosis with polyangiitis. Rheumatology  2015; 54: 1345– 50. Google Scholar CrossRef Search ADS PubMed  13 Jick H, Jick SS, Derby LE. Validation of information recorded on general practitioner based computerised data resource in the United Kingdom. BMJ  1991; 302: 766– 8. Google Scholar CrossRef Search ADS PubMed  14 Khan NF, Harrison SE, Rose PW. Validity of diagnostic coding within the General Practice Research Database: a systematic review. Br J Gen Pract  2010; 60: e128– 36. Google Scholar CrossRef Search ADS PubMed  15 Watts RA, Al-Taiar A, Scott DG, Macgregor AJ. Prevalence and incidence of Wegener's granulomatosis in the UK general practice research database. Arthritis Rheum  2009; 61: 1412– 6. Google Scholar CrossRef Search ADS PubMed  16 Gray RJ. A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat  1988; 16: 1141– 54. Google Scholar CrossRef Search ADS   17 Scrucca L, Santucci A, Aversa F. Competing risk analysis using R: an easy guide for clinicians. Bone Marrow Transplant  2007; 40: 381– 7. Google Scholar CrossRef Search ADS PubMed  18 Weidner S, Hafezi-Rachti S, Rupprecht HD. Thromboembolic events as a complication of antineutrophil cytoplasmic antibody-associated vasculitis. Arthritis Rheum  2006; 55: 146– 9. Google Scholar CrossRef Search ADS PubMed  19 Stassen PM, Derks RP, Kallenberg CG, Stegeman CA. Venous thromboembolism in ANCA-associated vasculitis—incidence and risk factors. Rheumatology  2008; 47: 530– 4. Google Scholar CrossRef Search ADS PubMed  20 Allenbach Y, Seror R, Pagnoux C et al.   High frequency of venous thromboembolic events in Churg-Strauss syndrome, Wegener's granulomatosis and microscopic polyangiitis but not polyarteritis nodosa: a systematic retrospective study on 1130 patients. Ann Rheum Dis  2009; 68: 564– 7. Google Scholar CrossRef Search ADS PubMed  21 Hergesell O, Andrassy K, Nawroth P. Elevated levels of markers of endothelial cell damage and markers of activated coagulation in patients with systemic necrotizing vasculitis. Thromb Haemost  1996; 75: 892– 8. Google Scholar PubMed  22 Hilhorst M, Winckers K, Wilde B et al.   Patients with antineutrophil cytoplasmic antibodies associated vasculitis in remission are hypercoagulable. J Rheumatol  2013; 40: 2042– 6. Google Scholar CrossRef Search ADS PubMed  23 Tarzi RM, Pusey CD. Current and future prospects in the management of granulomatosis with polyangiitis (Wegener's granulomatosis). Ther Clin Risk Manag  2014; 10: 279– 93. Google Scholar PubMed  24 Robson J, Doll H, Suppiah R et al.   Glucocorticoid treatment and damage in the anti-neutrophil cytoplasm antibody-associated vasculitides: long-term data from the European Vasculitis Study Group trials. Rheumatology  2015; 54: 471– 81. Google Scholar CrossRef Search ADS PubMed  25 de Leeuw K, Bijzet J, van der Graaf AM et al.   Patients with Wegener's granulomatosis: a long-term follow-up study. Clin Exp Rheumatol  2010; 28(1 Suppl 57): 18– 23. 26 Hajj-Ali RA, Wilke WS, Calabrese LH et al.   Pilot study to assess the frequency of fibromyalgia, depression, and sleep disorders in patients with granulomatosis with polyangiitis (Wegener's). Arthritis Care Res  2011; 63: 827– 33. Google Scholar CrossRef Search ADS   27 Englund M, Merkel PA, Tomasson G, Segelmark M, Mohammad AJ. Comorbidities in patients with antineutrophil cytoplasmic antibody-associated vasculitis versus the general population. J Rheumatol  2016; 43: 1553– 8. Google Scholar CrossRef Search ADS PubMed  28 Faurschou M, Ahlstrom MG, Lindhardsen J, Obel N, Baslund B. Risk of diabetes mellitus among patients diagnosed with giant cell arteritis or granulomatosis with polyangiitis: comparison with the general population. J Rheumatol  2017; 44: 78– 83. Google Scholar CrossRef Search ADS PubMed  29 He J, Ogden LG, Bazzano LA et al.   Risk factors for congestive heart failure in US men and women: NHANES I epidemiologic follow-up study. Arch Intern Med  2001; 161: 996– 1002. Google Scholar CrossRef Search ADS PubMed  30 Lionaki S, Hogan SL, Falk RJ et al.   Association between thyroid disease and its treatment with ANCA small-vessel vasculitis: a case-control study. Nephrol Dial Transplant  2007; 22: 3508– 15. Google Scholar CrossRef Search ADS PubMed  31 Avina-Zubieta JA, Mai A, Amiri N et al.   Risk of myocardial infarction and stroke in patients with granulomatosis with polyangiitis (Wegener's): a population-based study. Arthritis Rheumatol  2016; 68: 2752– 9. Google Scholar CrossRef Search ADS PubMed  32 Suppiah R, Judge A, Batra R et al.   A model to predict cardiovascular events in patients with newly diagnosed Wegener's granulomatosis and microscopic polyangiitis. Arthritis Care Res  2011; 63: 588– 96. Google Scholar CrossRef Search ADS   33 Roman MJ, Shanker BA, Davis A et al.   Prevalence and correlates of accelerated atherosclerosis in systemic lupus erythematosus. N Engl J Med  2003; 349: 2399– 406. Google Scholar CrossRef Search ADS PubMed  34 Telles RW, Lanna CC, Ferreira GA et al.   Carotid atherosclerotic alterations in systemic lupus erythematosus patients treated at a Brazilian university setting. Lupus  2008; 17: 105– 13. Google Scholar CrossRef Search ADS PubMed  35 Faurschou M, Mellemkjaer L, Sorensen IJ et al.   Cancer preceding Wegener's granulomatosis: a case-control study. Rheumatology  2009; 48: 421– 4. 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

A cohort study of comorbidity in patients with granulomatosis with polyangiitis

Loading next page...
 
/lp/ou_press/a-cohort-study-of-comorbidity-in-patients-with-granulomatosis-with-31xOVZ0SmV
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/kex379
Publisher site
See Article on Publisher Site

Abstract

Abstract Objectives To evaluate the associations between granulomatosis with polyangiitis (GPA) and a wide range of comorbidities in patients with GPA compared with population-based non-vasculitis patients. Methods Using the UK-based Clinical Practice Research Datalink we identified 570 incident patients with GPA in 1990–2014, and for each case, we selected up to 10 random non-vasculitis patients matched for age, sex, practice and years of history before the date of first GPA diagnosis. We compared the distribution of 13 pre-existing comorbidities and estimated the risk of each incident comorbidity after the cohort entry between GPA and non-vasculitis patients. Results Patients with GPA were more likely to have a history of hypothyroidism at their initial diagnosis compared with non-vasculitis patients. Patients with GPA had increased risk of incident venous thromboembolism any time during follow-up compared with non-vasculitis patients, particularly during the first 3 years of follow-up: hazard ratio (HR) of 5.24 (95% CI: 2.83, 9.71). Risks were also increased for hypertension (HR = 2.45, 95% CI: 1.84, 3.26), type 2 diabetes (HR = 2.13, 95% CI: 1.36, 3.32), dyslipidaemia (HR = 1.98, 95% CI: 1.29, 3.04) and depression (HR = 1.77, 95% CI: 1.10, 2.86) among GPA patients during the first 3 years of follow-up, but not after 3 years post-diagnosis. Conclusion Patients with GPA had a higher prevalence of hypothyroidism before the initial diagnosis, a strong risk of developing venous thromboembolism during follow-up and increased risks of hypertension, dyslipidaemia, type 2 diabetes and depression in the first few years after diagnosis, compared with non-vasculitis population. granulomatosis with polyangiitis, vasculitis, comorbidity, Clinical Practice Research Datalink, cohort Rheumatology key messages Granulomatosis with polyangiitis patients had more prior hypothyroidism than non-vasculitis patients. Granulomatosis with polyangiitis patients had an increased risk of developing venous thromboembolism compared with non-vasculitis patients. Risks were increased for hypertension, dyslipidaemia, type 2 diabetes and depression within 3 years after granulomatosis with polyangiitis diagnosis. Introduction Granulomatosis with polyangiitis (GPA, formerly known as Wegener’s granulomatosis) is the most common type of ANCA-associated vasculitis (AAV). Characteristic pathological features of GPA include arteritis of small-sized to medium-sized blood vessels, and granulomatous inflammation of the upper airways and lungs [1]. While GPA used to be frequently fatal, the introduction of CYC and glucocorticoid treatment, and more recently rituximab, greatly improved the survival rates and prognosis of GPA [2–4]. Despite this, data from the European Vasculitis Study Group trials showed a significant burden of disease manifestations involving renal and otolaryngolocical complications, and potentially treatment-related damage (including hypertension, malignancy and diabetes) in patients with AAV, with 34.4% of patients having ⩾5 areas of damage at a mean of 7 years post-diagnosis [5]. Increased carotid intima–media thickness, which indicates atherosclerosis, has also been demonstrated in GPA patients [6, 7], potentially as a consequence of glucocorticoid therapy, and GPA patients were reported to experience an increased number of cardiovascular events due to ischaemic heart disease (IHD) compared with the background population [8]. GPA patients were also reported to have increased risks of venous thromboembolism (VTE) [9, 10] and some specific cancers compared with population controls [11, 12]. Nevertheless, the epidemiology of comorbidity (including pre-existing and post-diagnosis) in GPA remains poorly understood. Therefore, we conducted this study to evaluate the associations between GPA and a wide range of comorbidities in a cohort of GPA patients compared with a matched non-vasculitis cohort, and to examine whether associations differ according to sex and age. Methods Data source Data were derived from the UK-based Clinical Practice Research Datalink (CPRD), a large primary care database that has been previously described in detail [13, 14]. Briefly, the CPRD is an ongoing longitudinal database that has collected data from over 500 general practices in the UK since 1987. It covers ∼6% of the UK population and has a UK representative age and sex distribution. The information recorded includes patient demographic characteristics, clinical diagnoses, drug prescriptions, consultant referrals, lab results and hospitalizations. Validation studies have shown the high quality of recorded diagnoses and drug exposures [13, 14]. This study was approved by the Independent Scientific Advisory Committee for Medicines and Healthcare products Regulatory Agency database research (protocol no: 15_051). Study design and base population This cohort study was conducted using CPRD data from 1 January 1990 through 31 December 2014. We identified all patients with a first-time diagnosis of GPA who had at least 1 year of recorded medical history before the first GPA diagnosis code (providing a window to ensure an incident diagnosis and to capture pre-existing comorbidities). For each GPA patient, up to 10 random patients without any forms of vasculitis were matched on age (±2 years), sex, general practice attended and years of history before cohort entry (date of first recorded GPA diagnosis or corresponding date for the matched non-vasculitis patients). GPA definition We identified all patients with a Read code for Wegener’s granulomatosis (the former eponym for GPA) during the study period, and categorized each incident case into one of three levels of likelihood (probable, possible or unlikely) according to the pre-determined algorithm for GPA cases (Table 1). This algorithm was developed by two of the authors (L.L. and T.N.) based on the definition of GPA described in a prior CPRD publication where 90% of patients with a code for GPA were confirmed to have a true diagnosis [15]. Note that the clinical judgement formulated from this process was based on an overall clinical impression from available information, such as recorded symptoms, treatments and referrals, rather than on clinical criteria, elements of which are typically not available in electronic medical records. We excluded unlikely cases from this study. Because some patients died or transferred out of their practice shortly after the initial diagnosis, they may not have had a chance to meet our GPA definition. To examine the impact of incomplete follow-up, we included patients whose data collection ended within 6 months of the cohort entry date and who did not meet the GPA definition (4.9% of GPA patients). See supplementary Fig. S1, available at Rheumatology Online, for GPA patient selection. Table 1 Classification algorithm for granulomatosis with polyangiitis Category  Criteria  Probablea (met at least one of these criteria)  Had a GPA diagnosis from a specialist (consultant) or a discharge letter with a GPA diagnosis Presented with specific symptoms suggestive of granulomatous disease affecting the upper and lower respiratory tract [15] within 6 months before or on index date (i.e. date of first diagnosis of GPA), plus either ≥1 prescriptions for immunosuppressive therapy within 6 months before or on the index date or any time after index date, or ≥2 diagnoses of GPA on different dates Had either ≥2 prescriptions for immunosuppressive therapy within 6 months before or on the index date or any time after index date, or ≥3 diagnoses of GPA on different dates in the absence of specific symptoms suggestive of granulomatous disease affecting the upper and lower respiratory tract [15] within 6 months before or on index date  Possiblea  When no probable case criteria stated above were met, a case was considered possible if they met both (i) and (ii): Had ≥2 prescriptions for glucocorticoid therapy, where one was within 6 months of the diagnosis date, and the two prescriptions were within 6 months of each other Presented with specific symptoms suggestive of granulomatous disease affecting the upper and lower respiratory tract [15] within 6 months before or on index date, or had anti-neutrophil cytoplasmic antibodies tests, chest X-ray, chest CT scan, sinus CT scan, relevant biopsy within 6 months before or on index date or 1 month after index date  Unlikely  All other cases do not meet probable or possible criteria  Category  Criteria  Probablea (met at least one of these criteria)  Had a GPA diagnosis from a specialist (consultant) or a discharge letter with a GPA diagnosis Presented with specific symptoms suggestive of granulomatous disease affecting the upper and lower respiratory tract [15] within 6 months before or on index date (i.e. date of first diagnosis of GPA), plus either ≥1 prescriptions for immunosuppressive therapy within 6 months before or on the index date or any time after index date, or ≥2 diagnoses of GPA on different dates Had either ≥2 prescriptions for immunosuppressive therapy within 6 months before or on the index date or any time after index date, or ≥3 diagnoses of GPA on different dates in the absence of specific symptoms suggestive of granulomatous disease affecting the upper and lower respiratory tract [15] within 6 months before or on index date  Possiblea  When no probable case criteria stated above were met, a case was considered possible if they met both (i) and (ii): Had ≥2 prescriptions for glucocorticoid therapy, where one was within 6 months of the diagnosis date, and the two prescriptions were within 6 months of each other Presented with specific symptoms suggestive of granulomatous disease affecting the upper and lower respiratory tract [15] within 6 months before or on index date, or had anti-neutrophil cytoplasmic antibodies tests, chest X-ray, chest CT scan, sinus CT scan, relevant biopsy within 6 months before or on index date or 1 month after index date  Unlikely  All other cases do not meet probable or possible criteria  a All cases with other causes (such as microscopic polyangiitis, Goodpasture’s syndrome, etc.) were excluded. GPA: granulomatosis with polyangiitis. Study outcomes We assessed 13 study outcomes: IHD (including myocardial infarction, angina pectoris and unspecified IHD), stroke/transient ischaemic attacks (TIA), peripheral vascular disease (PVD), VTE, heart failure, hypertension, dyslipidaemia, type 2 diabetes, cancer (except non-melanoma skin cancer), depression, psoriasis, chronic obstructive pulmonary disease and hypothyroidism. These study outcomes were selected because either they have been reported to be associated with GPA or its treatment, or they were commonly recorded in the study population after we screened GPA patient electronic records. Codes are available upon request. Covariates We collected information on smoking (current, former, never or unknown), BMI and alcoholism (yes or no), and exposure to aspirin, anti-hypertensive medicine, glucose lowering agents, lipid-lowering medicine, NSAIDs and antidepressants, defined by the presence of two or more prescriptions within 1 year before cohort entry. Statistical analysis We identified the first date that each study outcome was recorded and compared it with the cohort entry date to distinguish pre-existing and incident comorbidities. For pre-existing conditions, we compared frequencies any time before or on the cohort entry date between all GPA and non-vasculitis patients using the chi-square test or Fisher’s exact test. To account for delay in GPA diagnosis, we also calculated frequencies of pre-existing comorbidities at >1 year before the cohort entry date among those who had ⩾2 years of medical history. We compared the cumulative incidence of each incident study outcome at 1, 3, 5 and 10 years of follow-up between GPA patients and non-vasculitis patients using Gray’s test, adjusted for the competing risk of death [16, 17]. Patients with a pre-existing study outcome were excluded from the corresponding analyses of incident disease. Each eligible patient was followed from cohort entry until the date of the first of the following: the first diagnosis of a study outcome, the patient transferred out of the practice, the end of data collection or death. We used Cox regression to estimate for each study outcome the hazard ratio (HR) comparing GPA patients with non-vasculitis patients. Due to small sample size we reported herein only the ‘crude’ HRs and 95% CIs, which nevertheless, by design, take into account age, sex, practice and calendar time. For study outcomes that violated the proportional hazard assumption we reported the HRs stratified by length of follow-up (<3 and ⩾3 years). Finally, we repeated all main analyses in subpopulations stratified by sex and age (<60 and ⩾60 years). To examine the robustness of the results we re-analysed the primary results restricted to probable GPA patients and their corresponding comparison patients. All analyses were conducted using SAS statistical software, version 9.3 (SAS Institute, Cary, NC, USA), and the statistical package R with an add-on cmprsk package, version 3.1.1 (www.r-project.org). Results There were 570 GPA patients (including 512 probable cases, 30 possible cases and 28 cases with <6 months’ follow-up who did not meet the GPA definition) and 5389 non-vasculitis comparison patients in this study. Table 2 shows the baseline characteristics of patients in both groups. GPA patients were less likely to be current smokers, but more likely to be former smokers, than non-vasculitis patients. GPA patients were more likely to receive NSAIDs, but less likely to receive lipid-lowering medicine within one year before cohort entry compared with non-vasculitis patients. Mean follow-up was 5.9 years among GPA patients and 7.0 years among non-vasculitis patients. Among GPA patients, 166 (29%) patients died and 147 (26%) transferred out of the practice before the end of data collection, while 788 (15%) died and 1927 (36%) transferred out among non-vasculitis patients. Table 2 Baseline characteristics of patients with granulomatosis with polyangiitis and comparison patients Characteristics  GPA patients  Comparison patients  P-value  (n = 570)  (n = 5389)  Age, mean (s.d.), years  58.4 (15.6)  58.4 (15.5)  NAa      <60 years  264 (46.3)  2480 (46.0)  NAa      ≥60 years  306 (53.7)  2909 (54.0)  Sex      Female  254 (44.6)  2426 (45.0)  NAa      Male  316 (55.4)  2963 (55.0)  Smoking      Never  255 (44.7)  2379 (44.1)  0.0084      Current  81 (14.2)  944 (17.5)      Former  168 (29.5)  1302 (24.2)      Unknown  66 (11.6)  764 (14.2)  BMI      <18.5 kg/m2  14 (2.5)  74 (1.4)  0.21      18.5–24.9 kg/m2  134 (23.5)  1312 (24.3)      25.0–29.9 kg/m2  143 (25.1)  1456 (27.0)      ≥30 kg/m2  106 (18.6)  910 (16.9)      Unknown  173 (30.4)  1637 (30.4)  Alcoholism  16 (2.8)  179 (3.3)  0.51  Co-medications      Aspirin  57 (10.0)  644 (12.0)  0.17      Anti-hypertensive medicine  192 (33.7)  1752 (32.5)  0.57      Anti-diabetes medicine  30 (5.3)  315 (5.8)  0.57      Lipid-lowering medicine  69 (12.1)  869 (16.1)  0.012      NSAIDs  125 (21.9)  566 (10.5)  <0.0001      Anti-depressants  64 (11.2)  527 (9.8)  0.27  Year of cohort entry date      1990–94  49 (8.6)  480 (8.9)  NAa      1995–99  95 (16.7)  911 (16.9)      2000–04  127 (22.3)  1198 (22.2)      2005–09  166 (29.1)  1548 (28.7)      2010–14  133 (23.3)  1252 (23.2)  Length of recorded medical history, mean (s.d.), years      Before cohort entry  10.0 (6.1)  10.0 (6.1)  NAa      Follow-up after cohort entry  5.9 (4.9)  7.0 (5.4)  <0.0001  Characteristics  GPA patients  Comparison patients  P-value  (n = 570)  (n = 5389)  Age, mean (s.d.), years  58.4 (15.6)  58.4 (15.5)  NAa      <60 years  264 (46.3)  2480 (46.0)  NAa      ≥60 years  306 (53.7)  2909 (54.0)  Sex      Female  254 (44.6)  2426 (45.0)  NAa      Male  316 (55.4)  2963 (55.0)  Smoking      Never  255 (44.7)  2379 (44.1)  0.0084      Current  81 (14.2)  944 (17.5)      Former  168 (29.5)  1302 (24.2)      Unknown  66 (11.6)  764 (14.2)  BMI      <18.5 kg/m2  14 (2.5)  74 (1.4)  0.21      18.5–24.9 kg/m2  134 (23.5)  1312 (24.3)      25.0–29.9 kg/m2  143 (25.1)  1456 (27.0)      ≥30 kg/m2  106 (18.6)  910 (16.9)      Unknown  173 (30.4)  1637 (30.4)  Alcoholism  16 (2.8)  179 (3.3)  0.51  Co-medications      Aspirin  57 (10.0)  644 (12.0)  0.17      Anti-hypertensive medicine  192 (33.7)  1752 (32.5)  0.57      Anti-diabetes medicine  30 (5.3)  315 (5.8)  0.57      Lipid-lowering medicine  69 (12.1)  869 (16.1)  0.012      NSAIDs  125 (21.9)  566 (10.5)  <0.0001      Anti-depressants  64 (11.2)  527 (9.8)  0.27  Year of cohort entry date      1990–94  49 (8.6)  480 (8.9)  NAa      1995–99  95 (16.7)  911 (16.9)      2000–04  127 (22.3)  1198 (22.2)      2005–09  166 (29.1)  1548 (28.7)      2010–14  133 (23.3)  1252 (23.2)  Length of recorded medical history, mean (s.d.), years      Before cohort entry  10.0 (6.1)  10.0 (6.1)  NAa      Follow-up after cohort entry  5.9 (4.9)  7.0 (5.4)  <0.0001  Except where otherwise indicated, data are shown as n (%). a Not applicable because of matching factors. GPA: granulomatosis with polyangiitis. GPA patients were more likely to have a history of VTE, heart failure and hypothyroidism compared with non-vasculitis patients at the cohort entry. However, these differences were no longer present except for hypothyroidism at > 1 year before the cohort entry (Table 3). When we stratified by either sex or age, there were no material changes in the results (data not shown). When we restricted the analysis to probable GPA patients and their non-vasculitis comparison patients, the main findings did not change (see supplementary Table S1, available at Rheumatology Online). Table 3 Distributions of pre-existing comorbidities in patients with granulomatosis with polyangiitis and comparison patients Comorbidities  At cohort entry  At 1 year before cohort entrya  GPA patients  Comparison patients  P-value  GPA patients  Comparison patients  P-value  (n = 570)  (n = 5389)  (n = 527)  (n = 4983)  IHD  43 (7.5)  389 (7.2)  0.78  37 (7.0)  345 (6.9)  0.93  Stroke/TIA  20 (3.5)  168 (3.1)  0.61  12 (2.3)  143 (2.9)  0.43  PVD  9 (1.6)  84 (1.6)  0.97  8 (1.5)  77 (1.5)  0.96  VTE  15 (2.6)  76 (1.4)  0.024  6 (1.1)  69 (1.4)  0.64  Heart failure  16 (2.8)  89 (1.7)  0.046  11 (2.1)  70 (1.4)  0.22  Hypertension  140 (24.6)  1159 (21.5)  0.093  111 (21.1)  1032 (20.7)  0.85  Dyslipidaemia  45 (7.9)  412 (7.6)  0.83  41 (7.8)  363 (7.3)  0.68  Type 2 diabetes  42 (7.4)  382 (7.1)  0.80  30 (5.7)  324 (6.5)  0.47  Cancer  19 (3.3)  191 (3.5)  0.80  16 (3.0)  159 (3.2)  0.85  Depression  77 (13.5)  706 (13.1)  0.78  65 (12.3)  629 (12.6)  0.85  Psoriasis  20 (3.5)  160 (3.0)  0.47  17 (3.2)  143 (2.9)  0.64  COPD  13 (2.3)  158 (2.9)  0.38  11 (2.1)  131 (2.6)  0.46  Hypothyroidism  38 (6.7)  186 (3.5)  0.0001  33 (6.3)  159 (3.2)  0.0003  Comorbidities  At cohort entry  At 1 year before cohort entrya  GPA patients  Comparison patients  P-value  GPA patients  Comparison patients  P-value  (n = 570)  (n = 5389)  (n = 527)  (n = 4983)  IHD  43 (7.5)  389 (7.2)  0.78  37 (7.0)  345 (6.9)  0.93  Stroke/TIA  20 (3.5)  168 (3.1)  0.61  12 (2.3)  143 (2.9)  0.43  PVD  9 (1.6)  84 (1.6)  0.97  8 (1.5)  77 (1.5)  0.96  VTE  15 (2.6)  76 (1.4)  0.024  6 (1.1)  69 (1.4)  0.64  Heart failure  16 (2.8)  89 (1.7)  0.046  11 (2.1)  70 (1.4)  0.22  Hypertension  140 (24.6)  1159 (21.5)  0.093  111 (21.1)  1032 (20.7)  0.85  Dyslipidaemia  45 (7.9)  412 (7.6)  0.83  41 (7.8)  363 (7.3)  0.68  Type 2 diabetes  42 (7.4)  382 (7.1)  0.80  30 (5.7)  324 (6.5)  0.47  Cancer  19 (3.3)  191 (3.5)  0.80  16 (3.0)  159 (3.2)  0.85  Depression  77 (13.5)  706 (13.1)  0.78  65 (12.3)  629 (12.6)  0.85  Psoriasis  20 (3.5)  160 (3.0)  0.47  17 (3.2)  143 (2.9)  0.64  COPD  13 (2.3)  158 (2.9)  0.38  11 (2.1)  131 (2.6)  0.46  Hypothyroidism  38 (6.7)  186 (3.5)  0.0001  33 (6.3)  159 (3.2)  0.0003  Except where otherwise indicated, data are shown as n (%). aRestricted to patients with at least 2 years of medical history before cohort entry. COPD: chronic obstructive pulmonary disease; GPA: granulomatosis with polyangiitis; IHD: ischaemic heart disease; PVD: peripheral vascular disease; TIA: transient ischaemic attack; VTE: venous thromboembolism. The cumulative incidences for VTE, hypertension and type 2 diabetes were significantly higher in GPA patients than those in non-vasculitis patients. There were no significant differences in the cumulative incidences for other outcomes under study between GPA patients and non-vasculitis patients (Table 4). Consistent with the cumulative incidence findings, GPA patients had no increased risks for IHD, stroke/TIA, PVD and cancer compared with non-vasculitis patients (Table 5). GPA patients had a strong risk for VTE compared with non-vasculitis patients, particularly during the first 3 years of follow-up: HR = 5.24 (95% CI: 2.83, 9.71). GPA patients also had increased risks for hypertension (HR = 2.45, 95% CI: 1.84, 3.26), type 2 diabetes (HR = 2.13, 95% CI: 1.36, 3.32), dyslipidaemia (HR = 1.98, 95% CI: 1.29, 3.04) and depression (HR = 1.77, 95% CI: 1.10, 2.86) during the first 3 years of follow-up, but not after 3 years post-diagnosis, compared with non-vasculitis patients (Table 6). Table 4 Estimated cumulative incidences of study outcomes at specified times after cohort entry Incident comorbidity  Cumulative incidencea, per 100 (95% CI)  1 year  3 years  5 years  10 years  Gray’s test, P-value  Ischaemic heart disease      GPA  0.40 (0.08, 1.36)  1.73 (0.82, 3.26)  3.13 (1.74, 5.16)  5.36 (3.28, 8.17)  0.29      Comparison  0.86 (0.63, 1.15)  2.67 (2.22, 3.18)  4.07 (3.49, 4.71)  7.86 (6.90, 8.89)  Stroke/transient ischaemic attack      GPA  0.96 (0.37, 2.13)  2.27 (1.20, 3.90)  3.31 (1.93, 5.28)  5.15 (3.18, 7.80)  0.73      Comparison  0.50 (0.33, 0.73)  1.92 (1.55, 2.35)  2.69 (2.23, 3.21)  5.71 (4.90, 6.61)  Peripheral vascular disease      GPA  NAb  0.21 (0.02, 1.14)  0.99 (0.33, 2.40)  0.99 (0.33, 2.40)  0.63      Comparison  0.26 (0.14, 0.43)  0.42 (0.27, 0.64)  0.84 (0.60, 1.16)  2.10 (1.62, 2.67)  Venous thromboembolism      GPA  2.27 (1.24, 3.82)  3.16 (1.88, 4.96)  5.18 (3.40, 7.50)  6.11 (4.10, 8.67)  <0.0001      Comparison  0.24 (0.13, 0.40)  0.66 (0.46, 0.93)  1.13 (0.84, 1.48)  2.51 (1.97, 3.13)  Heart failure      GPA  0.95 (0.36, 2.10)  1.78 (0.88, 3.25)  3.11 (1.77, 5.05)  4.21 (2.50, 6.60)  0.33      Comparison  0.48 (0.32, 0.70)  1.34 (1.04, 1.71)  2.02 (1.63, 2.48)  4.06 (3.39, 4.82)  Hypertension      GPA  8.55 (6.09, 11.5)  15.4 (12.0, 19.2)  19.9 (15.9, 24.2)  27.8 (22.5, 33.3)  <0.0001      Comparison  2.34 (1.91, 2.84)  6.77 (5.99, 7.61)  10.9 (9.85, 12.0)  20.3 (18.7, 21.9)  Dyslipidaemia      GPA  2.00 (1.03, 3.53)  5.64 (3.78, 8.02)  6.74 (4.65, 9.35)  9.05 (6.39, 12.3)  0.54      Comparison  0.98 (0.73, 1.30)  2.90 (2.44, 3.43)  4.68 (4.05, 5.37)  8.43 (7.46, 9.48)  Type 2 diabetes      GPA  2.71 (1.55, 4.38)  4.93 (3.24, 7.12)  6.55 (4.51, 9.10)  10.6 (7.54, 14.1)  0.020      Comparison  1.04 (0.78, 1.35)  2.37 (1.96, 2.85)  3.97 (3.40, 4.60)  7.42 (6.51, 8.41)  Cancer      GPA  1.13 (0.47, 2.35)  3.10 (1.81, 4.95)  3.62 (2.18, 5.61)  7.17 (4.67, 10.4)  0.075      Comparison  0.99 (0.74, 1.30)  2.90 (2.44, 3.41)  4.91 (4.28, 5.60)  10.4 (9.29, 11.5)  Depression      GPA  2.12 (1.09, 3.74)  4.51 (2.84, 6.73)  6.67 (4.54, 9.36)  8.39 (5.79, 11.6)  0.51      Comparison  0.87 (0.63, 1.18)  2.91 (2.42, 3.45)  4.85 (4.19, 5.58)  8.90 (7.86, 10.0)  Psoriasis      GPA  0.38 (0.08, 1.28)  0.38 (0.08, 1.28)  0.64 (0.18, 1.77)  1.27 (0.47, 2.85)  0.23      Comparison  0.26 (0.15, 0.44)  0.87 (0.63, 1.18)  1.24 (0.94, 1.61)  2.16 (1.70, 2.71)  Chronic obstructive pulmonary disease      GPA  0.95 (0.36, 2.11)  1.38 (0.62, 2.72)  2.62 (1.43, 4.43)  4.87 (2.91, 7.55)  0.98      Comparison  0.32 (0.19, 0.51)  1.35 (1.05, 1.72)  2.27 (1.84, 2.75)  5.22 (4.44, 6.09)  Hypothyroidism      GPA  0.39 (0.08, 1.34)  1.58 (0.70, 3.10)  3.18 (1.78, 5.24)  4.56 (2.62, 7.30)  0.49      Comparison  0.24 (0.13, 0.41)  1.01 (0.75, 1.33)  1.82 (1.44, 2.27)  3.27 (2.68, 3.95)  Incident comorbidity  Cumulative incidencea, per 100 (95% CI)  1 year  3 years  5 years  10 years  Gray’s test, P-value  Ischaemic heart disease      GPA  0.40 (0.08, 1.36)  1.73 (0.82, 3.26)  3.13 (1.74, 5.16)  5.36 (3.28, 8.17)  0.29      Comparison  0.86 (0.63, 1.15)  2.67 (2.22, 3.18)  4.07 (3.49, 4.71)  7.86 (6.90, 8.89)  Stroke/transient ischaemic attack      GPA  0.96 (0.37, 2.13)  2.27 (1.20, 3.90)  3.31 (1.93, 5.28)  5.15 (3.18, 7.80)  0.73      Comparison  0.50 (0.33, 0.73)  1.92 (1.55, 2.35)  2.69 (2.23, 3.21)  5.71 (4.90, 6.61)  Peripheral vascular disease      GPA  NAb  0.21 (0.02, 1.14)  0.99 (0.33, 2.40)  0.99 (0.33, 2.40)  0.63      Comparison  0.26 (0.14, 0.43)  0.42 (0.27, 0.64)  0.84 (0.60, 1.16)  2.10 (1.62, 2.67)  Venous thromboembolism      GPA  2.27 (1.24, 3.82)  3.16 (1.88, 4.96)  5.18 (3.40, 7.50)  6.11 (4.10, 8.67)  <0.0001      Comparison  0.24 (0.13, 0.40)  0.66 (0.46, 0.93)  1.13 (0.84, 1.48)  2.51 (1.97, 3.13)  Heart failure      GPA  0.95 (0.36, 2.10)  1.78 (0.88, 3.25)  3.11 (1.77, 5.05)  4.21 (2.50, 6.60)  0.33      Comparison  0.48 (0.32, 0.70)  1.34 (1.04, 1.71)  2.02 (1.63, 2.48)  4.06 (3.39, 4.82)  Hypertension      GPA  8.55 (6.09, 11.5)  15.4 (12.0, 19.2)  19.9 (15.9, 24.2)  27.8 (22.5, 33.3)  <0.0001      Comparison  2.34 (1.91, 2.84)  6.77 (5.99, 7.61)  10.9 (9.85, 12.0)  20.3 (18.7, 21.9)  Dyslipidaemia      GPA  2.00 (1.03, 3.53)  5.64 (3.78, 8.02)  6.74 (4.65, 9.35)  9.05 (6.39, 12.3)  0.54      Comparison  0.98 (0.73, 1.30)  2.90 (2.44, 3.43)  4.68 (4.05, 5.37)  8.43 (7.46, 9.48)  Type 2 diabetes      GPA  2.71 (1.55, 4.38)  4.93 (3.24, 7.12)  6.55 (4.51, 9.10)  10.6 (7.54, 14.1)  0.020      Comparison  1.04 (0.78, 1.35)  2.37 (1.96, 2.85)  3.97 (3.40, 4.60)  7.42 (6.51, 8.41)  Cancer      GPA  1.13 (0.47, 2.35)  3.10 (1.81, 4.95)  3.62 (2.18, 5.61)  7.17 (4.67, 10.4)  0.075      Comparison  0.99 (0.74, 1.30)  2.90 (2.44, 3.41)  4.91 (4.28, 5.60)  10.4 (9.29, 11.5)  Depression      GPA  2.12 (1.09, 3.74)  4.51 (2.84, 6.73)  6.67 (4.54, 9.36)  8.39 (5.79, 11.6)  0.51      Comparison  0.87 (0.63, 1.18)  2.91 (2.42, 3.45)  4.85 (4.19, 5.58)  8.90 (7.86, 10.0)  Psoriasis      GPA  0.38 (0.08, 1.28)  0.38 (0.08, 1.28)  0.64 (0.18, 1.77)  1.27 (0.47, 2.85)  0.23      Comparison  0.26 (0.15, 0.44)  0.87 (0.63, 1.18)  1.24 (0.94, 1.61)  2.16 (1.70, 2.71)  Chronic obstructive pulmonary disease      GPA  0.95 (0.36, 2.11)  1.38 (0.62, 2.72)  2.62 (1.43, 4.43)  4.87 (2.91, 7.55)  0.98      Comparison  0.32 (0.19, 0.51)  1.35 (1.05, 1.72)  2.27 (1.84, 2.75)  5.22 (4.44, 6.09)  Hypothyroidism      GPA  0.39 (0.08, 1.34)  1.58 (0.70, 3.10)  3.18 (1.78, 5.24)  4.56 (2.62, 7.30)  0.49      Comparison  0.24 (0.13, 0.41)  1.01 (0.75, 1.33)  1.82 (1.44, 2.27)  3.27 (2.68, 3.95)  a Cumulative incidences were calculated after taking into account competing risk due to death. b Not applicable because of zero case. Table 5 Hazard ratios for granulomatosis with polyangiitis in relation to risks of study outcomes Incident comorbidity  GPA patients  Comparison patients  Hazard ratio (95% CI)  Number of events  Number of patients  Number of events  Number of patients  IHD  24  527  284  4996  0.91 (0.60, 1.38)  Stroke/TIA  22  549  228  5212  1.08 (0.70, 1.67)  PVD  7  561  81  5305  0.96 (0.45, 2.09)  VTE  30  554  108  5308  NAa  Heart failure  21  554  166  5300  1.46 (0.93, 2.30)  Hypertension  91  425  633  4223  NAa  Dyslipidaemia  37  524  323  4974  NAa  Type 2 diabetes  43  527  284  5003  NAa  Cancer  32  549  419  5188  0.86 (0.60, 1.23)  Depression  36  492  307  4670  NAa  Psoriasis  5  550  84  5229  0.66 (0.27, 1.63)  COPD  20  557  188  5231  1.18 (0.74, 1.87)  Hypothyroidism  16  532  134  5203  1.41 (0.84, 2.37)  Incident comorbidity  GPA patients  Comparison patients  Hazard ratio (95% CI)  Number of events  Number of patients  Number of events  Number of patients  IHD  24  527  284  4996  0.91 (0.60, 1.38)  Stroke/TIA  22  549  228  5212  1.08 (0.70, 1.67)  PVD  7  561  81  5305  0.96 (0.45, 2.09)  VTE  30  554  108  5308  NAa  Heart failure  21  554  166  5300  1.46 (0.93, 2.30)  Hypertension  91  425  633  4223  NAa  Dyslipidaemia  37  524  323  4974  NAa  Type 2 diabetes  43  527  284  5003  NAa  Cancer  32  549  419  5188  0.86 (0.60, 1.23)  Depression  36  492  307  4670  NAa  Psoriasis  5  550  84  5229  0.66 (0.27, 1.63)  COPD  20  557  188  5231  1.18 (0.74, 1.87)  Hypothyroidism  16  532  134  5203  1.41 (0.84, 2.37)  a Did not meet proportional hazard assumption during the whole follow-up period. COPD: chronic obstructive pulmonary disease; GPA: granulomatosis with polyangiitis; IHD: ischaemic heart disease; PVD: peripheral vascular disease; TIA: transient ischaemic attack; VTE: venous thromboembolism. Table 6 Hazard ratiosa for granulomatosis with polyangiitis in relation to risks of study outcomesb stratified by follow-up length Incident comorbidity  Follow-up length  <3 years  ≥3 years  Venous thromboembolism  5.24 (2.83, 9.71)  2.56 (1.44, 4.54)  Hypertension  2.45 (1.84, 3.26)  1.29 (0.89, 1.86)  Dyslipidaemia  1.98 (1.29, 3.04)  0.68 (0.37, 1.24)  Type 2 diabetes  2.13 (1.36, 3.32)  1.30 (0.81, 2.09)  Depression  1.77 (1.10, 2.86)  1.01 (0.60, 1.68)  Incident comorbidity  Follow-up length  <3 years  ≥3 years  Venous thromboembolism  5.24 (2.83, 9.71)  2.56 (1.44, 4.54)  Hypertension  2.45 (1.84, 3.26)  1.29 (0.89, 1.86)  Dyslipidaemia  1.98 (1.29, 3.04)  0.68 (0.37, 1.24)  Type 2 diabetes  2.13 (1.36, 3.32)  1.30 (0.81, 2.09)  Depression  1.77 (1.10, 2.86)  1.01 (0.60, 1.68)  The data are hazard ratio (95% CI). a Adjusted for age, sex and year of cohort entry. b These study outcomes did not meet proportional hazard assumption during the whole follow-up period. When we stratified by sex, female patients with GPA had an increased risk of heart failure compared with their non-vasculitis patients. When we stratified by age, patients diagnosed with GPA at age <60 years had increased risks of heart failure and hypothyroidism compared with non-vasculitis patients. Patients with GPA <60 years also had stronger risk of VTE than patients with GPA diagnosed after age 60 years. Patients with GPA ⩾60 years did not have significantly increased risks of hypertension, dyslipidaemia, type 2 diabetes or depression compared with non-vasculitis patients even during the first 3 years of follow-up (supplementary Tables S2–S4, available at Rheumatology Online). When we restricted the analysis to probable GPA patients and their non-vasculitis comparison patients, the main findings did not change (supplementary Tables S5–S9, available at Rheumatology Online). Discussion In this population-based cohort study, GPA patients had increased risks of incident VTE after GPA diagnosis compared with the matched non-vasculitis patients, particularly within the first 3 years post-diagnosis, consistent with the initial description of this phenomenon [10]. Risks were also increased for hypertension, type 2 diabetes, dyslipidaemia and depression among GPA patients during the first 3 years of follow-up. Pre-existing hypothyroidism was more common in GPA patients at their initial diagnosis, and even at > 1 year before the initial diagnosis. GPA has been reported to be associated with an increased risk of VTE [9, 10, 18–20]. Merkel et al. analysed data from the Wegener’s Granulomatosis Etanercept Trial and reported that the incidence rate of VTE among GPA patients was 7.0/100 person-years, markedly higher than the incidence rates of VTE in the general population and populations of patients with either lupus or RA, which ranged from 0.3 to 1.0/100 person-years [10]. Faurschou et al. [9] analysed data from a Danish tertiary care centre and found that within the first 2 years following the GPA diagnosis, incidence rate ratios for pulmonary embolism and deep venous thrombosis were 25.7 and 20.2 comparing GPA patients with age- and sex-matched population controls. Consistent with these findings, GPA patients in our study had a strong risk of VTE, especially during the first 3 years of follow-up, which suggests that the risk was strongest proximate to periods of active disease. We also found that pre-existing VTE (at cohort entry) was more common in GPA patients than in non-vasculitis patients, but only in the 1 year before cohort entry. Because GPA may not be diagnosed immediately upon onset of symptoms, VTEs that occurred in the year before cohort entry likely occurred after the disease onset during a period of high disease activity, but before the diagnosis was made. The pathogenetic background for the high VTE risk in GPA is unknown, but in another two studies of patients with AAV the increased risk of VTE could not be explained by traditional clinical risk factors [18, 19]. In a retrospective analysis involving 1130 patients with systemic necrotizing vasculitides, Allenbach et al. identified old age, male sex, previous VTE and stroke with motor deficit as risk factors for VTE [20]. In our study, age and sex were adjusted in the analyses, patients with prior VTE were excluded in the incident analysis, and there was no significant difference in the frequency of prior stroke between GPA patients and non-vasculitis patients. Therefore, the increased risk of VTE could not be explained by these risk factors. A previous study reported a close correlation between disease activity in GPA patients and activated coagulation. The authors also reported that an elevation of the markers of activated haemostasis was reversed when remission was obtained with specific treatment [21]. Despite this, patients with AAV in remission were still reported to be more procoagulant than healthy controls [22]. The observation that there is a stronger risk of VTE during the first 3 years post-diagnosis indirectly suggests the development of VTE is more likely to occur in patients with active or recently active vasculitis and disease- and/or treatment-related factors may play important roles in the increased risk of VTE in GPA. Further studies are needed to evaluate the extent to which the risk of VTE is associated with the disease itself or with treatment interventions. In our study, GPA patients compared with non-vasculitis patients had an increased risk of hypertension during the first 3 years of follow-up, but not after 3 years post-diagnosis. Hypertension was one of the most common medical conditions in GPA patients [23], which may be related to the typical glucocorticoid therapy used to treat GPA, renal involvement or a combination of the two [24]. In addition, GPA patients also had increased risks of dyslipidaemia, type 2 diabetes and depression compared with non-vasculitis patients during the first 3 years post-diagnosis, but not after 3 years, which is also potentially related to glucocorticoid therapy. De Leeuw et al. [25] observed that diastolic blood pressure and prevalence of dyslipidaemia significantly decreased during long-term follow-up in 23 GPA patients. Hajj-Ali et al. [26] reported that depression occurred more often in GPA patients compared with the general population. Recently Englund et al. [27] reported that AAV was associated with increased rates of several comorbidities including hypertension, diabetes and thyroid diseases compared with the general population, although they did not distinguish the prevalent and incident comorbidities under study. Faurschou et al. [28] also reported that GPA patients had an increased risk of new-onset diabetes in the first year of observation period, but not after. Note that the observation period started with the date of the second GPA-related hospital contact where the time between first-ever hospitalization for GPA and date of study inclusion did not exceed 18 months [28]. Since GPA patients were more likely to have medical visits early in the course of their disease than non-vasculitis patients, we cannot rule out the possibility that GPA patients had more opportunities to be examined and diagnosed with these comorbidities in the early stage of active disease. When we conducted analyses stratified by sex and age, we found that female patients with GPA or patients with GPA aged <60 years had an increased risk of heart failure compared with their non-vasculitis comparison patients, which could be explained by the lower background risk in female or younger patients because old age and male sex are independent risk factors for heart failure [29]. We also observed that younger patients with GPA (those <60 years) had an increased risk of new hypothyroidism, and pre-existing hypothyroidism was significantly more common in GPA patients compared with non-vasculitis patients even >1 year before cohort entry. Previously thyroid disease was reported be associated with ANCA small-vessel vasculitis [30]. Whether hypothyroidism is a risk factor for developing GPA needs further study. In our study GPA patients did not have increased risks of IHD, stroke/TIA or PVD compared with non-vasculitis patients. These findings did not change when we repeated the analyses stratified by age, sex and length of follow-up (<3 and ⩾3 years) or restricted to patients with probable GPA and their non-vasculitis comparison patients. In contrast, Faurschou et al. [8] using Danish hospital data, reported that GPA patients experienced an increased number of both early (<5 years after GPA diagnosis) and late (⩾10 years after GPA diagnosis) cardiovascular events (including myocardial infarction, angina pectoris and other IHD) compared with the background population. However, the same study group did not find an increased risk of stroke in GPA patients compared with matched population controls [9]. Similarly, Avina-Zubieta et al. [31] using population-based data from Canada reported that GPA patients had a significantly increased risk of myocardial infarction, not ischaemic stroke, compared with the age-, sex- and entry time-matched population controls and the risk was highest during the first year after GPA diagnosis. Because proteinase-3 ANCA has been associated with a reduced cardiovascular risk compared with myeloperoxidase ANCA or negative ANCA in patients with AAV [32], differences in proportions of ANCA specificities may in part explain these differences. Additionally, a cardioprotective effect of aggressive immunosuppression with CYC in cohorts of SLE patients has been reported [33, 34]. If GPA is similar in this regard to SLE, the risk of cardiovascular disease in GPA could be dependent on the relative balance of proatherogenic effects of sustained inflammation versus therapy-related control of inflammation. Differential management of GPA with regard to rapidity of glucocorticoid tapering and management of traditional risk factors of cardiovascular disease in GPA may also explain these different findings. We did not find an increased risk of incident cancer among GPA patients, nor a difference in prior cancer between GPA patients and non-vasculitis patients. GPA patients have been reported to experience a greater than expected number of specific malignancies including non-melanoma skin cancer, bladder cancer and myeloid leukaemia among patients exposed to cumulative CYC doses >36 g, while the cancer risk among CYC-naïve patients was not significantly increased [11, 12]. Moreover, the same study group did not find evidence of an increased prevalence of pre-existing cancer in GPA patients compared with age- and sex-matched controls [35]. Our study did not assess non-melanoma skin cancer, nor did we evaluate the relationship between GPA and specific cancer types taking into account CYC treatment. We did not have sufficient data available in CPRD to evaluate differences based on era of therapeutic management to compare the CYC-predominant management era vs the more recent time period when rituximab was the favoured first-line therapy. Our study has several strengths worth emphasizing. First, use of the CPRD, one of the world’s largest data resources that collects longitudinal medical history and drug information from general practices, allowed us to efficiently distinguish between pre-existing and incident morbidity and to explore many possible contributing factors involved with the burden of the disease. Second, our study assessed 13 study outcomes, allowing us to better understand comorbidity profiles in GPA patients. In this study, we not only confirmed that GPA was associated with strong risk of VTE, but also demonstrated that GPA was associated with increased risks of hypertension, dyslipidaemia, type 2 diabetes and depression in the first 3 years after diagnosis. Third, as GPA is a rare disease, reliable epidemiological studies on subpopulations of sex and age are limited. We were able to explore comorbidities in different subpopulations with GPA. There are some limitations that need to be considered in interpreting the results of this study. First, there may be misclassification of GPA cases and study outcomes, though the diagnosis codes for GPA and most of the study outcomes have been previously validated in the CPRD and their positive predictive values were reported to range from 70.3 to 98.6% [14, 15]. Second, there was a risk of detection bias in this study whereby GPA patients may have had more medical examinations and thus been more likely to be diagnosed with comorbidities, particularly in the first few years following diagnosis during a period of active disease. However, we found that GPA patients did not have increased risks of cardiovascular events or cancer; therefore, these results are not likely to be explained solely by detection bias. Third, since we only controlled for age, sex, calendar time and general practice in this study, there is a possibility of confounding due to uncontrolled or unknown confounders. Fourth, if GPA patients died shortly after the initial diagnosis, they may not meet the GPA definition. Therefore, it is likely to underestimate the risk of potentially early-occurring lethal comorbidities. However, in the original analyses we included any cases with <6 months’ follow-up regardless of whether they met the GPA definition or not and the main findings did not change after we excluded such cases, suggesting such bias is minimal. Finally, we were unable to investigate the associations between GPA and incident comorbidities of interest stratified by GPA treatment because immunosuppressive therapy is often prescribed by specialists and thus is not completely captured in the GP record, a limitation of the CPRD. In conclusion, GPA patients were confirmed to have strong risk of VTE post-diagnosis, and increased risks of hypertension, dyslipidaemia, type 2 diabetes and depression in the first few years after diagnosis, compared with a matched non-vasculitis population. Funding: This study was supported in part by the Vasculitis Foundation. Disclosure statement: The authors have declared no conflicts of interest. Supplementary data Supplementary data are available at Rheumatology Online. References 1 Seo P, Stone JH. The antineutrophil cytoplasmic antibody-associated vasculitides. Am J Med  2004; 117: 39– 50. Google Scholar CrossRef Search ADS PubMed  2 Fauci AS, Haynes BF, Katz P, Wolff SM. Wegener's granulomatosis: prospective clinical and therapeutic experience with 85 patients for 21 years. Ann Intern Med  1983; 98: 76– 85. Google Scholar CrossRef Search ADS PubMed  3 Hoffman GS, Kerr GS, Leavitt RY et al.   Wegener granulomatosis: an analysis of 158 patients. Ann Intern Med  1992; 116: 488– 98. Google Scholar CrossRef Search ADS PubMed  4 Stone JH, Merkel PA, Spiera R et al.   Rituximab versus cyclophosphamide for ANCA-associated vasculitis. N Engl J Med  2010; 363: 221– 32. Google Scholar CrossRef Search ADS PubMed  5 Robson J, Doll H, Suppiah R et al.   Damage in the anca-associated vasculitides: long-term data from the European vasculitis study group (EUVAS) therapeutic trials. Ann Rheum Dis  2015; 74: 177– 84. Google Scholar CrossRef Search ADS PubMed  6 de Leeuw K, Sanders JS, Stegeman C et al.   Accelerated atherosclerosis in patients with Wegener's granulomatosis. Ann Rheum Dis  2005; 64: 753– 9. Google Scholar CrossRef Search ADS PubMed  7 Nienhuis HL, de Leeuw K, Smit AJ et al.   Enhanced endothelium-dependent microvascular responses in patients with Wegener's granulomatosis. J Rheumatol  2007; 34: 1875– 81. Google Scholar PubMed  8 Faurschou M, Mellemkjaer L, Sorensen IJ et al.   Increased morbidity from ischemic heart disease in patients with Wegener's granulomatosis. Arthritis Rheum  2009; 60: 1187– 92. Google Scholar CrossRef Search ADS PubMed  9 Faurschou M, Obel N, Baslund B. High risk of pulmonary embolism and deep venous thrombosis but not of stroke in granulomatosis with polyangiitis (Wegener's). Arthritis Care Res  2014; 66: 1910– 4. Google Scholar CrossRef Search ADS   10 Merkel PA, Lo GH, Holbrook JT et al.   Brief communication: high incidence of venous thrombotic events among patients with Wegener granulomatosis: the Wegener's Clinical Occurrence of Thrombosis (WeCLOT) Study. Ann Intern Med  2005; 142: 620– 6. Google Scholar CrossRef Search ADS PubMed  11 Faurschou M, Sorensen IJ, Mellemkjaer L et al.   Malignancies in Wegener's granulomatosis: incidence and relation to cyclophosphamide therapy in a cohort of 293 patients. J Rheumatol  2008; 35: 100– 5. Google Scholar PubMed  12 Faurschou M, Mellemkjaer L, Voss A et al.   Prolonged risk of specific malignancies following cyclophosphamide therapy among patients with granulomatosis with polyangiitis. Rheumatology  2015; 54: 1345– 50. Google Scholar CrossRef Search ADS PubMed  13 Jick H, Jick SS, Derby LE. Validation of information recorded on general practitioner based computerised data resource in the United Kingdom. BMJ  1991; 302: 766– 8. Google Scholar CrossRef Search ADS PubMed  14 Khan NF, Harrison SE, Rose PW. Validity of diagnostic coding within the General Practice Research Database: a systematic review. Br J Gen Pract  2010; 60: e128– 36. Google Scholar CrossRef Search ADS PubMed  15 Watts RA, Al-Taiar A, Scott DG, Macgregor AJ. Prevalence and incidence of Wegener's granulomatosis in the UK general practice research database. Arthritis Rheum  2009; 61: 1412– 6. Google Scholar CrossRef Search ADS PubMed  16 Gray RJ. A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat  1988; 16: 1141– 54. Google Scholar CrossRef Search ADS   17 Scrucca L, Santucci A, Aversa F. Competing risk analysis using R: an easy guide for clinicians. Bone Marrow Transplant  2007; 40: 381– 7. Google Scholar CrossRef Search ADS PubMed  18 Weidner S, Hafezi-Rachti S, Rupprecht HD. Thromboembolic events as a complication of antineutrophil cytoplasmic antibody-associated vasculitis. Arthritis Rheum  2006; 55: 146– 9. Google Scholar CrossRef Search ADS PubMed  19 Stassen PM, Derks RP, Kallenberg CG, Stegeman CA. Venous thromboembolism in ANCA-associated vasculitis—incidence and risk factors. Rheumatology  2008; 47: 530– 4. Google Scholar CrossRef Search ADS PubMed  20 Allenbach Y, Seror R, Pagnoux C et al.   High frequency of venous thromboembolic events in Churg-Strauss syndrome, Wegener's granulomatosis and microscopic polyangiitis but not polyarteritis nodosa: a systematic retrospective study on 1130 patients. Ann Rheum Dis  2009; 68: 564– 7. Google Scholar CrossRef Search ADS PubMed  21 Hergesell O, Andrassy K, Nawroth P. Elevated levels of markers of endothelial cell damage and markers of activated coagulation in patients with systemic necrotizing vasculitis. Thromb Haemost  1996; 75: 892– 8. Google Scholar PubMed  22 Hilhorst M, Winckers K, Wilde B et al.   Patients with antineutrophil cytoplasmic antibodies associated vasculitis in remission are hypercoagulable. J Rheumatol  2013; 40: 2042– 6. Google Scholar CrossRef Search ADS PubMed  23 Tarzi RM, Pusey CD. Current and future prospects in the management of granulomatosis with polyangiitis (Wegener's granulomatosis). Ther Clin Risk Manag  2014; 10: 279– 93. Google Scholar PubMed  24 Robson J, Doll H, Suppiah R et al.   Glucocorticoid treatment and damage in the anti-neutrophil cytoplasm antibody-associated vasculitides: long-term data from the European Vasculitis Study Group trials. Rheumatology  2015; 54: 471– 81. Google Scholar CrossRef Search ADS PubMed  25 de Leeuw K, Bijzet J, van der Graaf AM et al.   Patients with Wegener's granulomatosis: a long-term follow-up study. Clin Exp Rheumatol  2010; 28(1 Suppl 57): 18– 23. 26 Hajj-Ali RA, Wilke WS, Calabrese LH et al.   Pilot study to assess the frequency of fibromyalgia, depression, and sleep disorders in patients with granulomatosis with polyangiitis (Wegener's). Arthritis Care Res  2011; 63: 827– 33. Google Scholar CrossRef Search ADS   27 Englund M, Merkel PA, Tomasson G, Segelmark M, Mohammad AJ. Comorbidities in patients with antineutrophil cytoplasmic antibody-associated vasculitis versus the general population. J Rheumatol  2016; 43: 1553– 8. Google Scholar CrossRef Search ADS PubMed  28 Faurschou M, Ahlstrom MG, Lindhardsen J, Obel N, Baslund B. Risk of diabetes mellitus among patients diagnosed with giant cell arteritis or granulomatosis with polyangiitis: comparison with the general population. J Rheumatol  2017; 44: 78– 83. Google Scholar CrossRef Search ADS PubMed  29 He J, Ogden LG, Bazzano LA et al.   Risk factors for congestive heart failure in US men and women: NHANES I epidemiologic follow-up study. Arch Intern Med  2001; 161: 996– 1002. Google Scholar CrossRef Search ADS PubMed  30 Lionaki S, Hogan SL, Falk RJ et al.   Association between thyroid disease and its treatment with ANCA small-vessel vasculitis: a case-control study. Nephrol Dial Transplant  2007; 22: 3508– 15. Google Scholar CrossRef Search ADS PubMed  31 Avina-Zubieta JA, Mai A, Amiri N et al.   Risk of myocardial infarction and stroke in patients with granulomatosis with polyangiitis (Wegener's): a population-based study. Arthritis Rheumatol  2016; 68: 2752– 9. Google Scholar CrossRef Search ADS PubMed  32 Suppiah R, Judge A, Batra R et al.   A model to predict cardiovascular events in patients with newly diagnosed Wegener's granulomatosis and microscopic polyangiitis. Arthritis Care Res  2011; 63: 588– 96. Google Scholar CrossRef Search ADS   33 Roman MJ, Shanker BA, Davis A et al.   Prevalence and correlates of accelerated atherosclerosis in systemic lupus erythematosus. N Engl J Med  2003; 349: 2399– 406. Google Scholar CrossRef Search ADS PubMed  34 Telles RW, Lanna CC, Ferreira GA et al.   Carotid atherosclerotic alterations in systemic lupus erythematosus patients treated at a Brazilian university setting. Lupus  2008; 17: 105– 13. Google Scholar CrossRef Search ADS PubMed  35 Faurschou M, Mellemkjaer L, Sorensen IJ et al.   Cancer preceding Wegener's granulomatosis: a case-control study. Rheumatology  2009; 48: 421– 4. 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