- Split View
-
Views
-
Cite
Cite
M.A. Little, L. Nazar, K. Farrington, Polymyalgia rheumatica preceding small-vessel vasculitis: changed spots or misdiagnosis?, QJM: An International Journal of Medicine, Volume 97, Issue 5, May 2004, Pages 289–292, https://doi.org/10.1093/qjmed/hch055
- Share Icon Share
Abstract
Background: Systemic small-vessel vasculitis (SVV) is increasing in incidence and age of diagnosis. Presenting features may mimic those of polymyalgia rheumatica (PMR), a common disease of the elderly.
Aim: To test the hypotheses that SVV is frequently misdiagnosed as PMR in elderly patients, that this results in a delay in diagnosis and appropriate treatment, and that the natural history and clinical features are different.
Design: Retrospective case-control study.
Methods: Cases of glomerulonephritis due to SVV at a single centre over a 12-year period were analysed, comparing those treated previously for PMR (PMR+) to the remainder of the cohort (PMR−).
Results: Of 86 patients with complete follow-up, 13% had been treated previously for PMR. PMR+ patients had a longer duration of symptoms prior to SVV diagnosis (396 vs. 107 days, p = 0.001) and were less likely to be dialysis-dependent at diagnosis (36% vs. 68%, p < 0.05). Despite the delay in diagnosis, there was a trend towards lower serum creatinine (392 vs. 591 μM), lower relapse rate (0.04 versus 0.15 episodes/patient-year) and lower incidence of death/end-stage renal failure (27% vs. 53%) in the PMR+ group.
Discussion: SVV is frequently misdiagnosed as PMR, especially in those patients with indolent disease, although this did not appear to adversely affect outcome. We recommend that all patients suspected of suffering from PMR undergo careful urinalysis to look for haematuria or proteinuria, and that a low threshold for ANCA testing is maintained.
Introduction
Systemic small-vessel vasculitis (SVV) causes renal failure, is increasing in incidence1 and is associated with a survival probability comparable to cancer (5-year survival 65–80%).2–6 It is usually associated with anti-neutrophil cytoplasm antibodies (ANCA).7 The median age of diagnosis in the first large series of SVV (reported in 1983) was 40 years,8 whereas the median age in studies reported over the past 10 years was 56.2,3,6,9–12 The population of patients diagnosed with SVV is aging, and many other conditions now enter the differential diagnosis.
Prior to the introduction of early immunosuppressive therapy, SVV was usually fatal within 6 months, and it is generally accepted that early diagnosis is critical to prevent irreversible critical organ injury.13 However, the presenting features may be protean and non-specific, often consisting only of general malaise, myalgia or arthralgia. This clinical syndrome resembles a much more common entity that occurs in elderly patients: PMR, the diagnosis of which frequently rests upon a prompt response to empiric steroid treatment.
We observed that a proportion of our patients presenting with SVV had received prior treatment with steroids for a diagnosis of PMR. We postulated that (a) the disease process in this subgroup of patients is different, with a distinctive natural history, and (b) misdiagnosis of SVV as PMR causes a delay in diagnosis that adversely affects patient and renal survival. We tested these hypotheses by analysing the records of all patients with a diagnosis of SVV at a regional nephrology centre.
Methods
We analysed the case records of all patients diagnosed with SVV-associated glomerulonephritis (based on Chapel Hill consensus conference criteria, n = 98) at the Lister Hospital over a 12-year period (1990–2002). In all patients, the diagnosis was confirmed histologically. Of these, 12 had incomplete data or were lost to follow-up; the remaining 86 patients form the basis of this report. We established which patients had been treated previously for PMR. In those that had, we analysed family doctor or hospital records antedating the diagnosis of SVV to confirm the PMR diagnosis and to obtain clinical details pertaining to this. We specifically quantified the duration from the time of developing symptoms and the time of diagnosis.
We reviewed the renal histology reports in the 72 patients (84%) who underwent renal biopsy, to compare the fraction of glomeruli with crescents, degree of crescent chronicity, interstitial fibrosis and tubular atrophy. ANCA testing was performed by indirect immunofluorescence of sera on ethanol fixed neutrophils. Positive results were confirmed with ELISA.
It has been our policy to treat SVV with various combinations of oral prednisolone, intravenous methylprednisolone, oral cyclophosphamide (usually for 3 months), oral azathioprine (as replacement for cyclophosphamide) and, since 1999, mycophenolate mofetil. Plasma exchange was employed in patients with severe renal impairment and fresh lesions on renal biopsy. Intravenous immunoglobulin was used occasionally in those with resistant disease. Cotrimoxazole was used as Pneumocystis prophylaxis in those receiving cyclophosphamide and/or high-dose methylprednisolone. The intensity, modes of treatment and cumulative cyclophosphamide dosage (10.4 ± 3.4 vs. 11.7 ± 1.9 g, p = 0.79) were the same for PMR+ and PMR− groups.
Statistical analysis
Using SPSS 9.0 software, baseline variables in the two groups were compared with the unpaired t-test, Kruskal-Wallis, Mann-Whitney or χ2 tests, as appropriate. Intragroup variability was expressed as the standard error of the mean (SEM) for normally distributed datasets and range for non-normal data. The probability of death or developing ESRF was assessed with the Kaplan-Meier method with censoring in the event of patient death or passing the date 30 June 2002; the two groups were compared in this respect using the log-rank test.
Results
PMR+patients
Of the 86 patients with SVV, 11 (13%) had a prior diagnosis of PMR. Five of these had been managed in primary care, and the remainder in a hospital setting. One had undergone a temporal artery biopsy, which was negative. We could find evidence of a urinalysis record at the time of PMR diagnosis in three (27%); all showed haematuria. All 11 PMR+ patients had been treated with prednisolone from the time of PMR diagnosis, to which all had a partial response. The atypical features that prompted a change in diagnosis were deterioration in renal function (n = 4), haemoptysis (n = 3), new neurological signs (n = 2), dipstick evidence of haematuria (n = 1) and skin rash (n = 1).
Clinical and histological characteristics
Table 1 summarizes the two groups. The PMR+ patients were older, although not significantly so. The time from the development of symptoms to diagnosis was 3.7 times longer in the PMR+ group than in the PMR− group (p = 0.001). Despite this delay, renal function was worse at diagnosis in the PMR− group, with a greater proportion of patients requiring urgent dialysis, and a trend towards a higher serum creatinine. There were no significant differences between the groups with respect to histological features suggestive of chronic disease (crescent fibrosis, obsolescent glomeruli, tubular trophy, interstitial fibrosis) or severity of glomerulonephritis. All PMR+ patients were ANCA-positive, with antibodies directed against either proteinase-3 or myeloperoxidase detectable in serum, and none had renal-limited vasculitis. ANCA results were otherwise similar between the two groups. The number of organs affected was also similar (2.4 ± 0.2 vs. 2.2 ± 0.1, for PMR+ and PMR− groups, respectively).
. | Prior PMR diagnosis . | No prior PMR diagnosis . | p . |
---|---|---|---|
Age (mean ± SEM) (years) | 68.8 ± 2.9 | 60.6 ± 1.9 | 0.10 |
Gender (% male) | 64% | 64% | 0.99 |
Symptom duration (mean ± SEM) | 396 ± 93 days | 107 ± 24 days | 0.001 |
Number of organs affected | |||
1 | 0% | 29% | 0.11 |
2 | 72% | 39% | |
>2 | 28% | 32% | |
Glomeruli with crescents (±SEM) | 41% (±10) | 43% (±3) | 0.87 |
ANCA status | |||
C | 45% | 36% | 0.69 |
P | 45% | 42% | |
Atypical | 9% | 7% | |
Negative | 0% | 15% | |
Vasculitis type | |||
MPA | 55% | 36% | 0.35 |
WG | 45% | 35% | |
RLV/Churg Strauss | 0% | 29% | |
Creatinine at presentation (mean ± SEM) | 392 ± 80 μM | 591 ± 37 μM | 0.07 |
Dialysis dependent at diagnosis | 36% | 68% | 0.04 |
. | Prior PMR diagnosis . | No prior PMR diagnosis . | p . |
---|---|---|---|
Age (mean ± SEM) (years) | 68.8 ± 2.9 | 60.6 ± 1.9 | 0.10 |
Gender (% male) | 64% | 64% | 0.99 |
Symptom duration (mean ± SEM) | 396 ± 93 days | 107 ± 24 days | 0.001 |
Number of organs affected | |||
1 | 0% | 29% | 0.11 |
2 | 72% | 39% | |
>2 | 28% | 32% | |
Glomeruli with crescents (±SEM) | 41% (±10) | 43% (±3) | 0.87 |
ANCA status | |||
C | 45% | 36% | 0.69 |
P | 45% | 42% | |
Atypical | 9% | 7% | |
Negative | 0% | 15% | |
Vasculitis type | |||
MPA | 55% | 36% | 0.35 |
WG | 45% | 35% | |
RLV/Churg Strauss | 0% | 29% | |
Creatinine at presentation (mean ± SEM) | 392 ± 80 μM | 591 ± 37 μM | 0.07 |
Dialysis dependent at diagnosis | 36% | 68% | 0.04 |
MPA, microscopic polyangiitis; WG, Wegener's granulomatosis; RLV, renal limited vasculitis.
. | Prior PMR diagnosis . | No prior PMR diagnosis . | p . |
---|---|---|---|
Age (mean ± SEM) (years) | 68.8 ± 2.9 | 60.6 ± 1.9 | 0.10 |
Gender (% male) | 64% | 64% | 0.99 |
Symptom duration (mean ± SEM) | 396 ± 93 days | 107 ± 24 days | 0.001 |
Number of organs affected | |||
1 | 0% | 29% | 0.11 |
2 | 72% | 39% | |
>2 | 28% | 32% | |
Glomeruli with crescents (±SEM) | 41% (±10) | 43% (±3) | 0.87 |
ANCA status | |||
C | 45% | 36% | 0.69 |
P | 45% | 42% | |
Atypical | 9% | 7% | |
Negative | 0% | 15% | |
Vasculitis type | |||
MPA | 55% | 36% | 0.35 |
WG | 45% | 35% | |
RLV/Churg Strauss | 0% | 29% | |
Creatinine at presentation (mean ± SEM) | 392 ± 80 μM | 591 ± 37 μM | 0.07 |
Dialysis dependent at diagnosis | 36% | 68% | 0.04 |
. | Prior PMR diagnosis . | No prior PMR diagnosis . | p . |
---|---|---|---|
Age (mean ± SEM) (years) | 68.8 ± 2.9 | 60.6 ± 1.9 | 0.10 |
Gender (% male) | 64% | 64% | 0.99 |
Symptom duration (mean ± SEM) | 396 ± 93 days | 107 ± 24 days | 0.001 |
Number of organs affected | |||
1 | 0% | 29% | 0.11 |
2 | 72% | 39% | |
>2 | 28% | 32% | |
Glomeruli with crescents (±SEM) | 41% (±10) | 43% (±3) | 0.87 |
ANCA status | |||
C | 45% | 36% | 0.69 |
P | 45% | 42% | |
Atypical | 9% | 7% | |
Negative | 0% | 15% | |
Vasculitis type | |||
MPA | 55% | 36% | 0.35 |
WG | 45% | 35% | |
RLV/Churg Strauss | 0% | 29% | |
Creatinine at presentation (mean ± SEM) | 392 ± 80 μM | 591 ± 37 μM | 0.07 |
Dialysis dependent at diagnosis | 36% | 68% | 0.04 |
MPA, microscopic polyangiitis; WG, Wegener's granulomatosis; RLV, renal limited vasculitis.
Renal and patient outcome
Mean durations of follow-up from time of diagnosis were 2.3 years (±0.3) and 3.6 years (±0.3) for the PMR+ and PMR− groups, respectively. Outcomes are summarized in Table 2. There was a trend towards a reduction in the combined end-point of death or permanent dialysis dependence and relapse rate in the PMR group. While short-term outcome was better in the PMR+ group, there were no significant differences in the actuarial renal or patient survival between the two groups.
. | Prior PMR diagnosis . | No prior PMR diagnosis . | p . |
---|---|---|---|
Dead or ESRF at last follow-up | 27% | 53% | 0.1 |
Patient survival (1 year/3 years) (%) | 90/50 | 82/73 | 0.9 |
Renal survival (1 year/3 years) (%) | 100/55 | 88/77 | 0.8 |
Relapse rate (episodes/pt-year, ±SEM) | 0.04 ± 0.03 | 0.15 ± 0.08 | 0.4 |
. | Prior PMR diagnosis . | No prior PMR diagnosis . | p . |
---|---|---|---|
Dead or ESRF at last follow-up | 27% | 53% | 0.1 |
Patient survival (1 year/3 years) (%) | 90/50 | 82/73 | 0.9 |
Renal survival (1 year/3 years) (%) | 100/55 | 88/77 | 0.8 |
Relapse rate (episodes/pt-year, ±SEM) | 0.04 ± 0.03 | 0.15 ± 0.08 | 0.4 |
Patient and renal survival probability were estimated using the Kaplan-Meier method.
. | Prior PMR diagnosis . | No prior PMR diagnosis . | p . |
---|---|---|---|
Dead or ESRF at last follow-up | 27% | 53% | 0.1 |
Patient survival (1 year/3 years) (%) | 90/50 | 82/73 | 0.9 |
Renal survival (1 year/3 years) (%) | 100/55 | 88/77 | 0.8 |
Relapse rate (episodes/pt-year, ±SEM) | 0.04 ± 0.03 | 0.15 ± 0.08 | 0.4 |
. | Prior PMR diagnosis . | No prior PMR diagnosis . | p . |
---|---|---|---|
Dead or ESRF at last follow-up | 27% | 53% | 0.1 |
Patient survival (1 year/3 years) (%) | 90/50 | 82/73 | 0.9 |
Renal survival (1 year/3 years) (%) | 100/55 | 88/77 | 0.8 |
Relapse rate (episodes/pt-year, ±SEM) | 0.04 ± 0.03 | 0.15 ± 0.08 | 0.4 |
Patient and renal survival probability were estimated using the Kaplan-Meier method.
Discussion
We have found that one in eight patients diagnosed with glomerulonephritis due to SVV at a single centre had been treated for some time prior to diagnosis for PMR. Despite the fact that these patients had a much longer duration of symptoms and a delay of an average of a year before diagnosis, they did not have a worse outcome. In fact, PMR+ patients had less severe renal failure at the time of SVV diagnosis, and there was a trend towards a lower incidence of death/ESRF. It appears that this group had a more indolent form of SVV, in which organ injury that was kept in check by prednisolone monotherapy for a period of time.
Is SVV presenting with features typical of PMR a separate pathology? Certainly there are distinct features in that there were no cases of renal limited vasculitis (not surprising, given that these patients initially presented with myalgia), they were slightly older (perhaps biasing physicians towards a diagnosis of PMR), had less severe renal failure and all were ANCA-positive. Based on the results presented here, it appears that PMR+ patients represented the benign end of the SVV spectrum. However, there was insufficient power in our cohort to detect more subtle differences in histology, pattern of organ involvement or relapse rate.
It may be argued that the PMR+ patients had two diagnoses: PMR/giant cell arteritis in addition to SVV, or that PMR ‘evolved’ into SVV. We believe the latter to be highly improbable. The pathogenesis of the two diseases is very different, there is no reported association between PMR and ANCA and, to our knowledge, there are no reported cases of PMR evolving into SVV. We cannot exclude categorically that the two diseases were present. However, for PMR (prevalence 7/1000 in those over 50 years) to occur in such a high proportion of patients with SVV (a rare condition) would be implausible. Rather, it is much more likely that a mild, indolent form of SVV was misdiagnosed as PMR. The differential diagnosis of PMR is wide,14 and such a misdiagnosis could easily occur; the hallmarks of PMR (myalgia, raised ESR, steroid responsiveness) are frequently present in early SVV. However, haematuria and a positive ANCA test are not features of PMR. While it is probably not appropriate to perform an ANCA test in all patients suspected of PMR, we believe that it is imperative that a urinalysis be performed in every patient in whom this diagnosis is considered and that there is a high index of suspicion for SVV. The finding of microscopic haematuria or proteinuria in a patient diagnosed with PMR should prompt further investigation, as should any other atypical feature in their clinical course.
References
USRDS annual report: Survival, mortality, and causes of death.
Aasarod K, Iversen BM, Hammerstrom J, Bostad L, Vatten L, Jorstad S. Wegener's granulomatosis: clinical course in 108 patients with renal involvement.
Aasarod K, Bostad L, Hammerstrom J, Jorstad S, Iversen BM. Renal histopathology and clinical course in 94 patients with Wegener's granulomatosis.
Guillevin L, Cordier JF, Lhote F, et al. A prospective, multicenter, randomized trial comparing steroids and pulse cyclophosphamide versus steroids and oral cyclophosphamide in the treatment of generalized Wegener's granulomatosis.
Adu D, Pall A, Luqmani RA, Richards NT, Howie AJ, Emery P, Michael J, Savage CO, Bacon PA. Controlled trial of pulse versus continuous prednisolone and cyclophosphamide in the treatment of systemic vasculitis.
Allen A, Pusey C, Gaskin G. Outcome of renal replacement therapy in antineutrophil cytoplasmic antibody-associated systemic vasculitis.
Falk RJ, Jennette JC. Anti-neutrophil cytoplasmic autoantibodies with specificity for myeloperoxidase in patients with systemic vasculitis and idiopathic necrotizing and crescentic glomerulonephritis.
Fauci AS, Haynes BF, Katz P, Wolff SM. Wegener's granulomatosis: prospective clinical and therapeutic experience with 85 patients for 21 years.
Haroun MK SJ, Nair R, Racusen L, Hellmann DB, Eustace JA. Correlation of percentage of normal glomeruli with renal outcome in Wegener's granulomatosis.
Matteson EL, Gold KN, Bloch DA, Hunder GG. Long-term survival of patients with Wegener's granulomatosis from the American College of Rheumatology Wegener's Granulomatosis Classification Criteria Cohort.
Aasarod K, Iversen BM, Hammerstrom J, Bostad L, Jorstad S. Clinical outcome of patients with Wegener's granulomatosis treated with plasma exchange.
Boomsma MM, Stegeman CA, van der Leij MJ, Oost W, Hermans J, Kallenberg CG, Limburg PC, Cohen Tervaert JW. Prediction of relapses in Wegener's granulomatosis by measurement of antineutrophil cytoplasmic antibody levels: a prospective study.