Abstract
Objective. To quantify the occurrence of features of vasculitis that initially present after diagnosis in 6 types of primary vasculitis.
Methods. Standardized collection of data on 95 disease manifestations in 6 vasculitides, including granulomatosis with polyangiitis (GPA), microscopic polyangiitis (MPA), eosinophilic granulomatosis with polyangiitis (Churg-Strauss; EGPA), polyarteritis nodosa (PAN), giant cell arteritis (GCA), and Takayasu arteritis (TAK), was obtained within a set of multicenter longitudinal, observational cohorts. For each form of vasculitis, the frequency of disease-specific manifestations at diagnosis was compared to the cumulative frequency of each manifestation. The percentage of patients who initially developed severe manifestations after diagnosis, defined as organ- or life-threatening in the small and medium vessel vasculitides (GPA, MPA, EGPA, PAN) and as ischemic/vascular in the large vessel vasculitides (GCA, TAK), was reported.
Results. Out of 838 patients with vasculitis, 490 (59%) experienced ≥ 1 new disease manifestation after diagnosis. On average, patients with vasculitis experienced 1.3 new manifestations after diagnosis (GPA = 1.9, MPA = 1.2, EGPA = 1.5, PAN = 1.2, GCA = 0.7, and TAK = 1.0). New severe manifestations occurred after diagnosis in 224 (27%) out of 838 patients (GPA = 26%, MPA = 19%, EGPA = 21%, PAN = 23%, GCA = 24%, and TAK = 44%). Timing of onset of new manifestations was not significantly associated with disease duration.
Conclusion. A majority of patients with vasculitis develop new disease features after diagnosis, including a substantial number of new, severe manifestations. Ongoing assessment of patients with established vasculitis should remain broad in scope.
- VASCULITIS
- GRANULOMATOSIS WITH POLYANGIITIS
- TAKAYASU ARTERITIS
- GIANT CELL ARTERITIS
- EOSINOPHILIC GRANULOMATOSIS WITH POLYANGIITIS
- MICROSCOPIC POLYANGIITIS
- POLYARTERITIS NODOSA
When caring for a patient with a chronic, relapsing disease, physicians often rely on the medical history to guide ongoing clinical assessment with the assumption that features of relapse will likely resemble patterns of prior disease. The systemic vasculitides are diseases in which manifestations and patterns of disease onset are highly variable, relapse is common and unpredictable, and new features of disease can occur late into the course of illness.
The spectrum of disease involvement at diagnosis and later in the disease course has been reported for some, but not all, of the major idiopathic vasculitides. Studies focused on the antineutrophil cytoplasmic antibody (ANCA) associated vasculitides typically report the frequency of organ system involvement rather than providing information about specific manifestations1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17. Understanding the frequency and type of new clinical manifestations of vasculitis that can occur following diagnosis will inform clinical practice and assist in the development of classification and diagnostic criteria18. Our objectives were to quantify the occurrence of new features of vasculitis (organ system involvement and specific manifestations) after diagnosis in 6 types of vasculitis, including granulomatosis with polyangiitis (GPA), microscopic polyangiitis (MPA), eosinophilic granulomatosis with polyangiitis (Churg-Strauss, EGPA), polyarteritis nodosa (PAN), giant cell arteritis (GCA), and Takayasu arteritis (TAK). By systematically profiling multiple forms of vasculitis, the frequency of manifestations can be compared across types of vasculitis and discussed with reference to existing literature.
MATERIALS AND METHODS
Patients enrolled in the Vasculitis Clinical Research Consortium (VCRC) longitudinal study cohorts from 2006 to 2012, representing 6 types of vasculitis, were selected for analysis. The VCRC is an international, multicenter research infrastructure supported by US National Institutes of Health (NIC), dedicated to conducting clinical research on different forms of vasculitis. All patients with GPA fulfilled a modified version of the 1990 American College of Rheumatology (ACR) classification criteria19. All patients with MPA satisfied the Chapel Hill criteria for the disease20. Since the existing 1990 ACR classification criteria for PAN21 do not clearly differentiate PAN from MPA18, an adapted version of these criteria was used to classify PAN. Patients were eligible to enroll into the PAN longitudinal cohort if they fulfilled one major criterion for disease (arteriographic abnormality, biopsy evidence of arterial wall inflammatory infiltrate, or neuropathy) and one minor criterion (weight loss > 4 kg, skin findings, testicular pain, myalgias, diastolic blood pressure > 90 mm Hg, abnormal kidney function, or ischemic abdominal pain), or if they fulfilled 2 major criteria. Patients with hepatitis B-related PAN or with cutaneous-only PAN were excluded from analysis. All patients with EGPA, GCA, or TAK fulfilled the respective 1990 ACR classification criteria for these diseases22,23,24.
Clinicians with expertise in vasculitis, representing 8 North American VCRC centers, performed quarterly comprehensive clinical assessments of patients enrolled in the VCRC longitudinal cohorts. Features of disease directly attributable to vasculitis were recorded at baseline and at successive study visits using standardized data collection forms. The incidence of each potential disease feature was assessed “at diagnosis” and “ever” (baseline visit) and “since last visit” (successive study visits). Demographic information was collected on each subject including age, sex, race, disease duration, length of study enrollment, and study site location. For the purpose of calculating disease duration, disease onset was defined as the date of initial diagnosis rather than the date of symptom onset or necessarily of treatment initiation.
The frequency of organ system involvement at diagnosis and at cumulative followup was described for each type of vasculitis and compared to similar published data from other vasculitis cohorts. The frequency of disease-specific manifestations at diagnosis was compared to the cumulative frequency of each manifestation, and the percentage of patients who developed one or more new disease manifestations after diagnosis was reported. A new disease manifestation was defined as a feature of disease first present after diagnosis. Severe manifestations were defined as those that were organ- or life-threatening in the small and medium vessel vasculitides (GPA, MPA, EGPA, PAN) and as ischemic/vascular in the large vessel vasculitides (GCA, TAK). For the vasculitides with granulomatous features (GPA, EGPA), a panel of vasculitis experts within the VCRC classified disease manifestations by consensus as either granulomatous, vasculitic, or “other,” and the percentage of patients who developed each type of manifestation after diagnosis was described. Manifestations were not defined at the subject level; rather, each potential disease manifestation was classified a priori as either severe or non-severe and as either granulomatous, vasculitic, or other.
A list of the 95 manifestations assessed in this study divided into 15 organ-based categories, along with the classification of each type of manifestation (severe or non-severe; granulomatous, vasculitic, or other) is provided in Appendix 1.
Data regarding thrombotic manifestations were recorded in the small and medium vessel vasculitides. Since it is unclear whether thrombosis is a disease manifestation of vasculitis or an associated condition25, occurrences of thrombotic events at diagnosis and later in the disease course were analyzed and reported separately within this study.
To determine if the development of new manifestations was related to disease duration, separate analyses were performed in a subset of patients enrolled into the VCRC within 1 year of diagnosis. Among patients who experienced at least one episode of recurrent disease activity, as determined by the treating physician, logistic regression using generalized estimated equations to account for repeated measurements was used to determine if the odds of developing a new manifestation versus purely recurrent features of disease was related to disease duration. Additionally, the median time from diagnosis to initial flare in disease activity was compared between those patients who experienced a new manifestation versus those who experienced recurrence only of prior manifestations using Wilcoxon rank-sum test. A flare in disease activity was determined based on the clinical judgment of the treating physician, and all treating physicians were experts in the clinical care of patients with vasculitis.
RESULTS
Data from 838 patients with vasculitis were available for analysis. Patient demographics are provided in Table 1. The number of patients and median disease duration for the vasculitides were: GPA, n = 341, 6.0 years; MPA, n = 26, 3.5 years; EGPA, n = 117, 3.9 years; PAN, n = 55, 3.7 years; GCA, n = 178, 2.1 years; TAK, n = 121, 5.6 years. The median time enrolled within the VCRC cohorts was 1.8 years (range 0-4.9 yrs).
Study characteristics and the 3 most frequently observed new manifestations after diagnosis.
Four-hundred ninety (59%) of 838 patients experienced at least 1 new disease manifestation after diagnosis (Figure 1). Depending on the type of vasculitis, at least 1 new disease manifestation occurred in 43% to 69% of patients. On average, patients with vasculitis experienced 1.3 new manifestations after diagnosis (GPA = 1.9, MPA = 1.2, EGPA = 1.5, PAN = 1.2, GCA = 0.7, TAK = 1.0). A subset of patients with each type of vasculitis (7%–28%) experienced ≥ 3 new manifestations after diagnosis. The 3 most frequent new manifestations for each type of vasculitis are listed in Table 1.
Frequency of new disease manifestations occurring after diagnosis in 6 types of vasculitis. GPA: granulomatosis with polyangiitis; MPA: microscopic polyangiitis; EGPA: eosinophilic granulomatosis with polyangiitis (Churg-Strauss); PAN: polyarteritis nodosa; GCA: giant cell arteritis; TAK: Takayasu arteritis.
New severe manifestations occurred after diagnosis in 224 (27%) of 838 patients (GPA, 26%; MPA, 19%; EGPA, 21%; PAN, 23%; GCA, 26%; TAK, 45%). Non-severe manifestations occurred after diagnosis in 311 (37%) of 838 patients (GPA, 52%; MPA, 15%; EGPA, 49%; PAN, 37%; GCA, 22%; TAK, 13%). An equal proportion of patients with GPA or EGPA developed new granulomatous manifestations (GPA, 35%; EGPA, 32%) compared to new vasculitic manifestations (GPA, 31%; EGPA, 30%).
The frequency of organ system involvement for each type of vasculitis at diagnosis and at cumulative followup is listed and compared to literature from other vasculitis cohorts in Table 2. Limited data were available from published reports on organ system involvement in the large vessel vasculitides because most of these studies preferentially reported the frequency of individual manifestations rather than organ system involvement. New involvement of the musculoskeletal system after diagnosis was common across all types of vasculitis (range 8%–12%). Among the 484 patients with small vessel vasculitis (GPA, MPA, EGPA), new ocular involvement (range 10%–12%) and cutaneous involvement (range 9%–18%) were frequently observed after diagnosis. Among the 299 patients with large vessel vasculitis, new nervous system involvement (GCA, 9%; TAK, 14%) frequently occurred after diagnosis.
Organ system involvement within the Vasculitis Clinical Research Consortium (VCRC) and comparison to existing literature.
Thrombotic disease was observed in a substantial number of patients with small and medium vessel vasculitis. Thrombotic disease at diagnosis was noted in GPA (5%), MPA (0%), EGPA (3%), and PAN (14%). New thrombotic disease, mostly venous thromboembolic events, first occurred after diagnosis in GPA (9%), MPA (12%), EGPA (6%), and PAN (13%).
The frequency of selected, severe disease manifestations at diagnosis and at cumulative followup, and the percentage of patients in whom the manifestation initially occurred after diagnosis for each type of vasculitis, is provided in Table 3. There was a striking spectrum of severe manifestations that first occurred after diagnosis, including alveolar hemorrhage (4%), claudication (11%), glomerular disease (7%), deafness (2%), and vision loss (2%).
Frequency and timing of selected disease manifestations.
A subset of 388 patients were enrolled into the VCRC longitudinal study cohorts within 1 year of diagnosis and followed for a median of 1.9 years (range 0–6.6 yrs). At least one active disease visit after diagnosis was documented in 131 (34%) of 388 patients and 53 patients had > 1 active disease visits. New manifestations were noted in 115 (49%) of active visits. The odds of developing a new disease manifestation compared to recurrence of prior manifestations decreased by 11% for every 1 year increase in disease duration following diagnosis, but the results were not statistically significant (OR 0.89, 95% CI 0.5–1.1, p = 0.2). The median time from diagnosis to initial flare in disease activity did not significantly differ among those who experienced a new manifestation versus a recurrence of prior disease features (0.87 vs 0.93 yrs; p = 0.8).
DISCUSSION
This study is the first of its kind to systematically evaluate clinical features of vasculitis across several forms of small, medium, and large vessel vasculitis. Based on comprehensive clinical assessment within the VCRC longitudinal observational cohorts, the frequency of new manifestations of vasculitis occurring after diagnosis was described in 6 different types of vasculitis. New features of disease after diagnosis were frequently observed across all types of vasculitis under study. Fifty-nine percent of patients with vasculitis developed new features of disease after diagnosis, and 27% of patients first developed life-threatening, organ-threatening, or ischemic features of disease later into the disease course. Adding to the complexity of clinical assessment of patients with established vasculitis, the likelihood of developing a new disease manifestation versus purely a recurrence of previous disease features was not significantly associated with the timing of disease flare in relationship to disease duration.
There were some surprising findings in our study. In the GCA group, 14% of patients experienced vision loss at time of diagnosis, which is consistent with the 14% to 20% incidence of vision loss in the modern era of glucocorticoid therapy26. Among patients with no evidence of vision loss at diagnosis, 5 (3%) of 178 patients with GCA developed vision loss later in the disease course. Although new vision loss after diagnosis of GCA is uncommon, it can occur despite therapeutic intervention27,28. Additionally, the most frequent new manifestation after diagnosis of GCA was polymyalgia rheumatica (PMR). Epidemiologic studies indicate that 16% to 20% of patients with PMR develop GCA, highlighting the prodromal aspect of PMR in relationship to GCA29,30,31. However, symptoms of PMR have been observed in 40% to 60% of patients with GCA and can occur at any timepoint during the disease course30. In one clinical trial, 16% of patients with GCA developed PMR later in the disease course28, and clinical experience suggests that symptoms of PMR can be unmasked later in the disease course during tapering of glucocorticoids.
The number of patients with MPA in this study was fairly small, therefore, the findings in MPA should be interpreted with caution. However, the frequency and severity of new manifestations after diagnosis of MPA was similar to other forms of vasculitis. In our study, 3 (12%) of 25 patients with MPA developed new ocular symptoms after diagnosis (all 3 developed conjunctivitis/episcleritis) and an additional 3 patients developed new ear/nose/throat symptoms (sinusitis, 2; nasal crusts, 1). It is possible that some patients diagnosed initially as MPA progressed to a GPA phenotype.
Overall, the frequency of disease involvement at diagnosis and later in the disease course was consistent with that reported in the literature, when these data were available1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,32,33,34,35,36,37,38,39,40. Notably, there is considerable variability in the literature regarding the prevalence of cardiac involvement in GPA, with reports of prevalence of cardiac disease in GPA ranging from 4% to 25%, depending on the extent of screening and inclusion of patients without overt clinical cardiac disease. Increased morbidity from ischemic heart disease has been observed in GPA41, but it is unclear whether coronary atherosclerosis is an associated condition or a direct manifestation of disease. In our study, coronary arteritis was not observed in any patients with GPA at any point in the disease course.
Thrombotic events were assessed and observed in a substantial number of patients with small and medium vessel vasculitis. Ample evidence supports an increased incidence of thromboembolic disease in ANCA-associated vasculitis, but there is minimal literature regarding thrombotic disease in PAN25. Surprisingly, the highest prevalence of thrombotic disease in our study was observed in PAN (27%) rather than ANCA-associated vasculitis, and new thrombotic disease after diagnosis was the most frequent new manifestation later in the course of PAN.
This study has several strengths including relatively large study cohorts for these rare diseases, a comprehensive and standardized system to collect data on disease manifestations, and inclusion of multiple centers with expertise in the evaluation of patients with vasculitis.
This study also has some limitations to consider. It is possible that retrospective assessment of disease features at diagnosis, as performed for some patients in the cohorts, was biased towards preferential reporting of only the more severe manifestations. However, severe disease manifestations, such as glomerulonephritis, which are less susceptible to recall bias, were also frequently first observed after diagnosis. Referral bias may have influenced the data if patients with more complicated disease courses may have been preferentially referred to tertiary care centers for ongoing management of vasculitis. However, recruitment of subjects from 8 centers may have helped reduce such bias. The influence of treatment on the development of new disease manifestations was not assessed; however, a significant number of new manifestations occurred in patients while they were under direct observation and presumably receiving standard of care therapy at tertiary care centers with expertise in vasculitis. As therapeutic advances continue to transform the systemic vasculitides from potentially fatal diseases if untreated into more chronic disorders, our findings highlight that the majority of patients with vasculitis will experience new disease manifestations after initial diagnosis. Continued improvement in the management of patients with established vasculitis is needed.
A substantial number of variable disease features occur after diagnosis across many forms of vasculitis. These data could influence development of classification and diagnostic criteria since such criteria often focus on the initial presentation of disease. Similarly, these results provide insight important to physicians caring for patients with vasculitis. While previous patterns of disease may help guide clinical assessment of patients with relapsing systemic disease syndromes, ongoing clinical assessment of a patient with established systemic vasculitis should remain comprehensive and broad in scope.
APPENDIX 1. History and review of systems related to vasculitis.
Footnotes
-
Supported by the Vasculitis Clinical Research Consortium, which has received support from the National Institute of Arthritis and Musculoskeletal and Skin Diseases (U54AR057319 and U01 AR51874 04), the National Center for Research Resources (U54 RR019497); and the Office of Rare Diseases Research. Dr. Grayson receives support from an NIH Training Program in Rheumatic Disease grant (AR 007598) and an American College of Rheumatology (ACR-REF) Rheumatology Scientist Development award.
- Accepted for publication June 6, 2013.