Sessa A, Meroni M, Battini G (eds): Renal Involvement in Systemic Vasculitis. Contrib Nephrol. Basel, Karger, 1991, vol 94, pp 38-46
Renal Vasculitis: Microscopic Polyarteritis and Wegener's Granuloma J. Stewart Cameron' Guy's Campus UMDS, London, UK
I would like to thank all my clinical colleagues who participated in the care of these patients, and Dr. B. Hartley who evaluated their renal biopsies. Drs. M. Lockwood, A. Rees and their colleagues performed the ANCA studies.
Downloaded by: Université de Paris 193.51.85.197 - 1/14/2020 11:21:20 AM
In this short paper I would like to take the opportunity to review our experience with microscopic polyarteritis affecting the kidney during the period 1981-1990, which follows on from the analysis of 55 patients with renal vasculitis seen during 1965-1981, which we published in 1984 [1]. During this recent decade we have seen a further 53 patients with renal vasculitis, all affecting small vessels. Thus our total experience of this condition now involves more than 110 patients. Some details of the first 27 of these 53 recent patients have been published already [2]. In this recent group, unlike our analysis of the first 55 patients in 1982, we have attempted to allocate the patients with renal vasculitis into those with Wegener's granuloma, and those with microscopic polyarteritis, on a clinical basis. In 17 patients (table 1) some or all of the usual clinical criteria for a diagnosis of probable Wegener's granuloma (otitis, nasal discharge or necrosis, sinusitis, haemoptysis, pulmonary shadowing, subglottic granuloma) were present, although a histological diagnosis was available in only 3 — together with a further patient with a granulomatous glomerulonephritis (GI). The remaining 35 patients have been called `microscopic polyarteritis', including 6 patients who presented with histological lesions of necrotizing glomerulitis without evident immune aggregates and (in some) a positive antineutrophil cytoplasmic antibodies (ANCA) test. One patient (not further considered here) had the Churg-Strauss syndrome with asthma, eosinophilia and pulmonary infiltrates.
Renal Vasculitis: Microscopic Polyarteritis and Wegener's Granuloma
39
Table 1. Presentation of renal vasculitis Presentation
n
Idiopathic crescentic nephritis initially Micropolyarteritis Wegener's1 Churg-Strauss
6 296 17 1
Total
53
1 ENT
involvement, ± haemoptysis and lung infiltrates; histological diagnosis only in 3.
35
Whether it is possible, or useful, to divide patients with microscopic vasculitis affecting the kidney into microscopic polyarteritis and Wegener's granuloma, and on what grounds, has been argued elsewhere [3]. Strictly defined, Wegener's must be a histological diagnosis, but convention now allows patients with clinical manifestations in ear, nose, throat or lung and vasculitis to be called 'Wegener's' even though no granulomas have been demonstrated [3]. There ís, as yet, no clear evidence that the two conditions are different as to pathogenesis [see Coelho das Neves, this volume], and since treatment schedules are the same at the moment, as yet the allocation of patients to either category is not critical.
In the present group of 52 patients, about two thirds did not satisfy even the clinical criteria for probable Wegener's (table 1), whilst one third did so. It will become evident that these two groups of patients did not differ in their presentation or outcome to any significant extent, when severity of renal or extrarenal disease is allowed for. There was the usual slight male preponderance (M:F ratio 1.13 for Wegener's, 1.33 for micropolyarteritis), the age at presentation being identical (56 ± 14 and 58 ± 12, respectively), the majority of patients being middle-aged or elderly. Symptoms and signs were typical of vasculitis, including weight loss, myalgia, `flu-like' symptoms, arthralgia and purpura; the special features which led to a clinical diagnosis of Wegener's have been noted above. In 1 case, an empyema of the gall bladder was a precipitating feature, 1 patient had taken diclofenac, whilst 2 had had rheumatoid arthritis
Downloaded by: Université de Paris 193.51.85.197 - 1/14/2020 11:21:20 AM
Presentation and Renal Histology
Cameron
40
Table 2. Renal vasculitis 1981-1990: renal histology
Focal segmental necrotizing GI ± crescents Normal glomeruli Mesangial proliferation only Tubulointerstitial nephritis Granulomas Vasculitis in extraglomerular vessels
Wegener's
Micropolyarteritis
Total
15
25
40
1 0 1 1 1
6 1 5 0 5
7 1 6 1 6 (12%)
in the past and 1 had Waldenström's macroglobulinaemia. In 6 patients there were no clinical features suggestive of vasculitis (`idiopathic crescentic GI'), although in 1 such untreated patient, florid vasculitis developed later. ANCA were sought in 33/52 patients, in 29/52 during the acute phase before treatment (courtesy of Drs. A.J. Rees, M. Lockwood and D. Jayne). All 8 Wegener's patients were positive, but 6 of 21 with micropolyarteritis were negative, including 1 patient with `idiopathic crescentic GI'. Two patients showed antiglomerular basement membrane antibody, 1 in addition to being positive for p-ANCA. The dominant histological picture was one of focal necrotizing glomerulitis without glomerular immune aggregates, accompanied in 39/42 patients showing this appearance by crescent formation (table 2), which affected 40100% of glomeruli in 37 patients; again the two subgroups did not differ with respect to histology, except for the presence of interstitial granulomata (1 patient) or granulomatous GI (1 patient) in the Wegener group. An interesting subgroup, however, comprised 6 patients, 5 allocated to micropolyarteritis and 1 to the Wegener group, 4/5 ANCA-positive, who presented clinically as vasculitis but who showed in their renal biopsies normal glomeruli with severe interstitial infiltration; the clinical details and outcome of these patients are summarized in table 3. The relationship between presenting plasma creatinine and the severity of the renal lesion as judged by the percentage of glomeruli affected by large crescents is shown in figure 1. In general, there was a correlation between initial creatinine and the extent of crescent formation.
Downloaded by: Université de Paris 193.51.85.197 - 1/14/2020 11:21:20 AM
Some of the 52 biopsies showed more than one feature.
41
Renal Vasculitis: Microscopic Polyarteritis and Wegener's Granuloma
Table 3. Renal vasculitis: features in patients with tubulointerstitial nephritis and normal glomeruli Sex Age Presentation
ANCA Pcreat (type)
Initial treatment
Outcome
CAPD after 4 years
M
62
rash, iritis, weight loss, +ve (?) 217-490 pred only myalgia, old TB, single kidney
F
68
rash, eosin (diclofenac) ND
597
dx, pred, mered
M
69
rash, arthralgias, GI bleeding
550
dx, pred, cyclo, mered
died on dx 2 weeks postgastrectomy
+ve (c) 229
pred only
Πcreaί 176
+ve (c) 457
pred, cyclo, mered
died 4 weeks on dx
+ve (?) 360
pred, aza
P creat 170
F 68 rash M
70 rash, weight loss
F
58
rash, haemoptysis, pulmonary infection
—ve
Pcreaι
120
dx = Dialysis; pred = prednisolone; cyclo = cyclophosphamide; aza = azathioprine; mered = 3 g í.v. methylprednisolone; ND = not done; c = cytoplasmic pattern of ANCA.
Treatment of small vessel vasculitis affecting the kidney is arbitrary, and only one controlled study (that of Pusey et al. [5], which evaluated plasma exchange) has been performed. The initial treatment received by our 52 patients is detailed in table 4. Nineteen patients (37%), 16/35 with microscopic polyarteritis and 3/ 17 with Wegener's, required dialysis. All patients but 3 received oral prednisolone, and 30 intravenous methylprednisolone 1 g Li. on 3 consecutive days, in conformity with the more aggressive immunosuppressive treatment we adopted in 1982 when it became obvious that some patients were dying of vasculitis in the acute phase [ 1 ]. Thirty-one patients were treated in addition with oral cyclophosphamide, usually in a dose of 1-2 mg/kg/24 h for the first 12 weeks of the illness, after which they were transferred to azathioprine, 75-150 mg daily. Four patients received azathioprine but never took cyclophosphamide. Eighteen of the most severely affected patients, 11 of whom had micropolyarteritis and 7 with Wegener's, also received 5-10 (mean 7) daily 3-4 litre plasma exchanges.
Downloaded by: Université de Paris 193.51.85.197 - 1/14/2020 11:21:20 AM
Treatment
42
Cameron ∆
∆
> 1000
Ο
1000
∆
Ο ΟΟ
∆
800
∆
∆
Ο
• •
c m ώ
ø
600
ø
Φ 400 ο_" 200
∆
∆
TINS TIN TIN TIN TIN
Ο
∆
4
∆
Α
TIN ∆ ο
∆
∆
•
•
o
8
• ο
∆
•
•
Ο
∆
•
Ο
ο
80 20 40 60 Percent of surviving glomeruli with occluding crescents
100
Fig. 1. Relationship between (a) percentage of unsclerosed glomeruli affected by large crescents and plasma creatinine at the time of renal biopsy, and (b) the relationship between percentage of unsclerosed glomeruli affected by large crescents and ultimate outcome. It can be seen that whilst here is in general a relationship between plasma creatinine and extent of crescents, the same does not hold true for long-term prognosis, either for renal function or for patient survival. Data are shown separately for Wegener's granuloma (closed symbols) and microscopic polyarteritis (open symbols). TIN = Fromínent or isolated tubulointerstitial nephritis (see text for discussion); p, • = dead or on dialysis/tx; O, • = independent renal function.
n
Mean Pcτea1
Dialysis
Fex
Mpred í.v.
Cyclo/ aza
Fred orally
Micropolyarteritis Wegener's
35 17
567 ± 347 402 ± 254
16 3
11 7
21 9
21/19 10/16
33 17
Total
52
19
18
30
30/34
49
Twenty-one patients received both drugs, all but 2 cyclophosphamide first for 8-12 weeks, followed by azathioprine. Pex = Plasma exchange; Mpred = methylprednisolone; Cyclo = cyclophosphamide; aza = azathioprine; Pred = prednisolone.
Downloaded by: Université de Paris 193.51.85.197 - 1/14/2020 11:21:20 AM
Table 4. Renal vasculitis 1981-1990: initial treatment
Renal Vasculitis: Microscopic Polyarteritis and Wegener's Granuloma
43
Table 5. Renal vasculitis: causes of death 1981-1990 In acute illness
Later deaths
age cause of death
age cause of death
Wegener's (n = 17)
75 dialysis withdrawn 61 MI
3 weeks (on dx) 2 weeks
50 ? cause (on CAPD) 71 gram —ve sepsis
Micropolyarteritis (n=35)
64 infarcted bowel 61 cardiac arrest 66 pneumonia 66 pneumonia/WBC/MI 70 bowel perf crumbled 69 GI bleeding/gastrect 37 resp failure (HIV +ve)
5 weeks (on dx) 2 weeks (on dx) 3 weeks (on dx) 3 weeks (on dx) 5 weeks (on dx) 2 weeks (on dx) 3 months (on dx)
59 resp arrest (on CAPD) 5 5 marrow failure (on RDT) 70 neutropenia 4 months 57 inanition (on RDT) 76 ?cause
dx = Acute dialysis; RDT = regular dialysis; CAPD = continuous ambulatory peritoneal dialysis; perf = perforation; gastrect = gastrectomy; resp = respiratory; WBC = neutropenia; MI = myocardial infarct.
Nine patients (2 Wegener's and 7 micropolyarteritis) died during the acute illness, all but 1 whilst still receiving dialysis; the reasons for death are detailed in table 5. Α further 5 patients who survived the acute illness through dialysis failed to recover renal function and continued on dialysis, 2 of these being transplanted successfully. Eight patients (6 with micropolyarteritis and 2 with Wegener's) required dialysis initially, but survived and recovered renal function. Seven patients died later (2 Wegener's and 5 microscopic polyarteritis), 4 after starting renal replacement therapy. The causes of death are again detailed in table 5. Five patients clearly died at least in part because of infection, with or without obvious marrow depression; 3 patients had complications of the vasculitis itself, and 2 suffered myocardial infarcts. The relationship between renal status at most recent follow-up and the percentage of crescents in the initial biopsy is shown in figure 1. There was only a poor relation between the extent of crescent formation and the likelihood of being dead, or requiring regular dialysis, after a mean of 5 years. Two patients with interstitial nephritis and no crescents died (table 3, fig. 1), as did 2 patients with minor degrees of crescent formation (fig. 1).
Downloaded by: Université de Paris 193.51.85.197 - 1/14/2020 11:21:20 AM
Outcome
Cameron
44
Table 6. Renal vasculitis 1981-1990: maintenance treatment in survivors (1-9 years later, mean 5 years) Wegener's
Micropolyarteritis
None Prednisolone alone Prednisolone + azathioprine
0 1 11
3 3 14
Total
12
201
1
Total 3 4 25 321
Details of current treatment not available in 2 patients.
Table 7. Renal vasculitis: outcome 1981-1990 (mean follow-up 4.5 years) Wegener's
Micropolyarteritis
Total 11 19 9 35 9 3 16 51 2
Alive, normal renal function Alive, reduced renal function Required long-term dialysis (deaths) Total alive Died during acute illness Died later off dialysis Total deaths
4 8 1 (1) 12 2 1 4
7 11 81 (3) 22 7 2 12
Total
162
35
1 Two 2
transplanted, both with success. Follow-up not available in 1 patient.
We have continued to treat most survivors (25/32) with prednisolone and azothioprine for a mean of 5 years (table 6), and it is noteworthy that only 2 of these patients experienced relapses, a much lower proportion than that reported from other series in which treatment was tapered or stopped earlier. The outcome at latest follow-up during 1990 is shown in table 7: after a mean of 5 years, 35/52 (67%) patients survived, 12/17 (71%) with Wegener's and 22/35 (63%) with microscopic polyarteritis, 5 patients with microscopic polyarteritis surviving on substitution therapy, 3 on dialysis and
Downloaded by: Université de Paris 193.51.85.197 - 1/14/2020 11:21:20 AM
Maintenance Treatment
Renal Vasculitis: Microscopic Polyarteritis and Wegener's Granuloma
45
100 —
80 —
Wegen er's ø Ζ 60
Ç
ιι
3 w
η
L
`η1
~l
Φ
ώ 40 —
All
Micro-PAN
~--- —► Serra et al., 1982
ο-
20 —
ι
0
I
Discharge from hospital
2
ι
4
ι
6 Years
ι
8
10
Fig. 2. Survival of .patients calculated by the Kaplan-Meler method. Data are displayed (continuous lines) for the whole group of renal small vessel vasculitis (52 patients), and separately for those with Wegener's granuloma (18 patients) and microscopic polyarteritis (35 patients). None of these curves is statistically different at a 1:20 level (Kaplan-Meier). Also shown (interrupted line) are data from the previous cohort of 55 patients described by Serra et al. 1984 [ 1]. This line differs from the recent group of 52 patients at less than 1:20 level, patient survival being inferior in the 1968-81 cohort compared with the 1981-90 cohort.
functioning transplant. Survival curves for the whole group of patients, and for those with Wegener's and microscopic polyarteritis separately, are shown in figure 2, compared with data from our previous cohort of 5 5 patients [1].
Thus, although the prognosis for small vessel vasculitis affecting the kidney has improved from almost inevitable death 30 years ago, through a 30-35% 5-year survival 10-15 years ago [1], many problems remain. Although it seems likely that the more intense immunosuppression used in most units over the past 10-15 years has improved immediate survival, a number of these elderly and often frail patients die directly as a result of these treatment regimens [4]; in our present series, 5/16 deaths (31%) could be related directly to immunosuppression. In some patients extrarenal complications of the vasculitis, particularly in the gastrointestinal tract, still may
Downloaded by: Université de Paris 193.51.85.197 - 1/14/2020 11:21:20 AM
Summary and Conclusions
Cameron
46
lead to early death. The relative roles (if any) of methylprednisolone and/or plasma exchange in treatment of renal vasculitis are not clear yet, although a recent controlled trial in patients with crescentic nephritis, mainly the result of small vessel renal vasculitis, showed a modest benefit from plasma exchange in addition to prednisolone and cyclophosphamide [4], but only in those patients requiring dialysis. In the longer term, we do not know for how long, with what agent and with what intensity immunosuppression must be maintained. In our series, in which most patients were maintained on modest immunosuppression for many years, relapse of the vasculitis was almost absent, suggesting some merit in this strategy. We used azathioprine rather than intermittent intravenous cyclophosphamide, as others have advocated; there are no data to choose between these regimens at the moment, although even in this relatively elderly population long-term oral cyclophosphamide is better avoided because of risks to the bladder, gonadal toxicity and oncogenesis. However, there may well be a subgroup of patients in whom this long-term treatment is unnecessary and therefore dangerous. It may be that the measurement of ANCA in the serum, even if it does not play an immediate pathogenic role, will be useful in predicting relapses and monitoring treatment. However, the continuing controversy about the utility of anti-dsDNA antibodies for this purpose in systemic lupus, now 45 years after their first description, suggests that clear-cut answers may not be evident immediately. References Serra A, Cameron JS, Turner DR, Hartley B, Ogg CS, Neild GH, Williams DG, Taube D, Brown CB, Jicks HJA: Vasculitis affecting the kidney: presentation, histopathology and long-term outcome. Q J Med 1984;53:181-207. 2 Fuiano G, Cameron JS, Raftery M, Hartley BH, Williams DG, Ogg CS: Improved prognosis of renal microscopic polyarteritis in recent years. Nephrol Dialysis Transplant 1988;3:383-391. Serra A, Cameron JS: Clinical and pathologic aspects of renal vasculitis. Serif 3 Nephrol 1985;5:15-33. Neild GH: Infectious complications in the management of systemic vasculitis and 4 rapidly progressive glomerulonephritis (submitted). 5 Pusey CD, Rees AJ, Evans DJ, Peters DK, Lockwood CM: A randomised controlled trial of plasma exchange in focal necrotising glomerulonephritis without anti-GBM antibodies. Kidney Int 1990;38 (in press). Prof. J.S. Cameron, Clinical Science Labs, 17th Floor Guy's Tower, Guy's Hospital, London SE 1 9RT (UK)
Downloaded by: Université de Paris 193.51.85.197 - 1/14/2020 11:21:20 AM
1
Renal Vasculitis: Microscopic Polyarteritis and Wegener's Granuloma J. Stewart Cameron' Guy's Campus UMDS, London, UK
I would like to thank all my clinical colleagues who participated in the care of these patients, and Dr. B. Hartley who evaluated their renal biopsies. Drs. M. Lockwood, A. Rees and their colleagues performed the ANCA studies.
Downloaded by: Université de Paris 193.51.85.197 - 1/14/2020 11:21:20 AM
In this short paper I would like to take the opportunity to review our experience with microscopic polyarteritis affecting the kidney during the period 1981-1990, which follows on from the analysis of 55 patients with renal vasculitis seen during 1965-1981, which we published in 1984 [1]. During this recent decade we have seen a further 53 patients with renal vasculitis, all affecting small vessels. Thus our total experience of this condition now involves more than 110 patients. Some details of the first 27 of these 53 recent patients have been published already [2]. In this recent group, unlike our analysis of the first 55 patients in 1982, we have attempted to allocate the patients with renal vasculitis into those with Wegener's granuloma, and those with microscopic polyarteritis, on a clinical basis. In 17 patients (table 1) some or all of the usual clinical criteria for a diagnosis of probable Wegener's granuloma (otitis, nasal discharge or necrosis, sinusitis, haemoptysis, pulmonary shadowing, subglottic granuloma) were present, although a histological diagnosis was available in only 3 — together with a further patient with a granulomatous glomerulonephritis (GI). The remaining 35 patients have been called `microscopic polyarteritis', including 6 patients who presented with histological lesions of necrotizing glomerulitis without evident immune aggregates and (in some) a positive antineutrophil cytoplasmic antibodies (ANCA) test. One patient (not further considered here) had the Churg-Strauss syndrome with asthma, eosinophilia and pulmonary infiltrates.
Renal Vasculitis: Microscopic Polyarteritis and Wegener's Granuloma
39
Table 1. Presentation of renal vasculitis Presentation
n
Idiopathic crescentic nephritis initially Micropolyarteritis Wegener's1 Churg-Strauss
6 296 17 1
Total
53
1 ENT
involvement, ± haemoptysis and lung infiltrates; histological diagnosis only in 3.
35
Whether it is possible, or useful, to divide patients with microscopic vasculitis affecting the kidney into microscopic polyarteritis and Wegener's granuloma, and on what grounds, has been argued elsewhere [3]. Strictly defined, Wegener's must be a histological diagnosis, but convention now allows patients with clinical manifestations in ear, nose, throat or lung and vasculitis to be called 'Wegener's' even though no granulomas have been demonstrated [3]. There ís, as yet, no clear evidence that the two conditions are different as to pathogenesis [see Coelho das Neves, this volume], and since treatment schedules are the same at the moment, as yet the allocation of patients to either category is not critical.
In the present group of 52 patients, about two thirds did not satisfy even the clinical criteria for probable Wegener's (table 1), whilst one third did so. It will become evident that these two groups of patients did not differ in their presentation or outcome to any significant extent, when severity of renal or extrarenal disease is allowed for. There was the usual slight male preponderance (M:F ratio 1.13 for Wegener's, 1.33 for micropolyarteritis), the age at presentation being identical (56 ± 14 and 58 ± 12, respectively), the majority of patients being middle-aged or elderly. Symptoms and signs were typical of vasculitis, including weight loss, myalgia, `flu-like' symptoms, arthralgia and purpura; the special features which led to a clinical diagnosis of Wegener's have been noted above. In 1 case, an empyema of the gall bladder was a precipitating feature, 1 patient had taken diclofenac, whilst 2 had had rheumatoid arthritis
Downloaded by: Université de Paris 193.51.85.197 - 1/14/2020 11:21:20 AM
Presentation and Renal Histology
Cameron
40
Table 2. Renal vasculitis 1981-1990: renal histology
Focal segmental necrotizing GI ± crescents Normal glomeruli Mesangial proliferation only Tubulointerstitial nephritis Granulomas Vasculitis in extraglomerular vessels
Wegener's
Micropolyarteritis
Total
15
25
40
1 0 1 1 1
6 1 5 0 5
7 1 6 1 6 (12%)
in the past and 1 had Waldenström's macroglobulinaemia. In 6 patients there were no clinical features suggestive of vasculitis (`idiopathic crescentic GI'), although in 1 such untreated patient, florid vasculitis developed later. ANCA were sought in 33/52 patients, in 29/52 during the acute phase before treatment (courtesy of Drs. A.J. Rees, M. Lockwood and D. Jayne). All 8 Wegener's patients were positive, but 6 of 21 with micropolyarteritis were negative, including 1 patient with `idiopathic crescentic GI'. Two patients showed antiglomerular basement membrane antibody, 1 in addition to being positive for p-ANCA. The dominant histological picture was one of focal necrotizing glomerulitis without glomerular immune aggregates, accompanied in 39/42 patients showing this appearance by crescent formation (table 2), which affected 40100% of glomeruli in 37 patients; again the two subgroups did not differ with respect to histology, except for the presence of interstitial granulomata (1 patient) or granulomatous GI (1 patient) in the Wegener group. An interesting subgroup, however, comprised 6 patients, 5 allocated to micropolyarteritis and 1 to the Wegener group, 4/5 ANCA-positive, who presented clinically as vasculitis but who showed in their renal biopsies normal glomeruli with severe interstitial infiltration; the clinical details and outcome of these patients are summarized in table 3. The relationship between presenting plasma creatinine and the severity of the renal lesion as judged by the percentage of glomeruli affected by large crescents is shown in figure 1. In general, there was a correlation between initial creatinine and the extent of crescent formation.
Downloaded by: Université de Paris 193.51.85.197 - 1/14/2020 11:21:20 AM
Some of the 52 biopsies showed more than one feature.
41
Renal Vasculitis: Microscopic Polyarteritis and Wegener's Granuloma
Table 3. Renal vasculitis: features in patients with tubulointerstitial nephritis and normal glomeruli Sex Age Presentation
ANCA Pcreat (type)
Initial treatment
Outcome
CAPD after 4 years
M
62
rash, iritis, weight loss, +ve (?) 217-490 pred only myalgia, old TB, single kidney
F
68
rash, eosin (diclofenac) ND
597
dx, pred, mered
M
69
rash, arthralgias, GI bleeding
550
dx, pred, cyclo, mered
died on dx 2 weeks postgastrectomy
+ve (c) 229
pred only
Πcreaί 176
+ve (c) 457
pred, cyclo, mered
died 4 weeks on dx
+ve (?) 360
pred, aza
P creat 170
F 68 rash M
70 rash, weight loss
F
58
rash, haemoptysis, pulmonary infection
—ve
Pcreaι
120
dx = Dialysis; pred = prednisolone; cyclo = cyclophosphamide; aza = azathioprine; mered = 3 g í.v. methylprednisolone; ND = not done; c = cytoplasmic pattern of ANCA.
Treatment of small vessel vasculitis affecting the kidney is arbitrary, and only one controlled study (that of Pusey et al. [5], which evaluated plasma exchange) has been performed. The initial treatment received by our 52 patients is detailed in table 4. Nineteen patients (37%), 16/35 with microscopic polyarteritis and 3/ 17 with Wegener's, required dialysis. All patients but 3 received oral prednisolone, and 30 intravenous methylprednisolone 1 g Li. on 3 consecutive days, in conformity with the more aggressive immunosuppressive treatment we adopted in 1982 when it became obvious that some patients were dying of vasculitis in the acute phase [ 1 ]. Thirty-one patients were treated in addition with oral cyclophosphamide, usually in a dose of 1-2 mg/kg/24 h for the first 12 weeks of the illness, after which they were transferred to azathioprine, 75-150 mg daily. Four patients received azathioprine but never took cyclophosphamide. Eighteen of the most severely affected patients, 11 of whom had micropolyarteritis and 7 with Wegener's, also received 5-10 (mean 7) daily 3-4 litre plasma exchanges.
Downloaded by: Université de Paris 193.51.85.197 - 1/14/2020 11:21:20 AM
Treatment
42
Cameron ∆
∆
> 1000
Ο
1000
∆
Ο ΟΟ
∆
800
∆
∆
Ο
• •
c m ώ
ø
600
ø
Φ 400 ο_" 200
∆
∆
TINS TIN TIN TIN TIN
Ο
∆
4
∆
Α
TIN ∆ ο
∆
∆
•
•
o
8
• ο
∆
•
•
Ο
∆
•
Ο
ο
80 20 40 60 Percent of surviving glomeruli with occluding crescents
100
Fig. 1. Relationship between (a) percentage of unsclerosed glomeruli affected by large crescents and plasma creatinine at the time of renal biopsy, and (b) the relationship between percentage of unsclerosed glomeruli affected by large crescents and ultimate outcome. It can be seen that whilst here is in general a relationship between plasma creatinine and extent of crescents, the same does not hold true for long-term prognosis, either for renal function or for patient survival. Data are shown separately for Wegener's granuloma (closed symbols) and microscopic polyarteritis (open symbols). TIN = Fromínent or isolated tubulointerstitial nephritis (see text for discussion); p, • = dead or on dialysis/tx; O, • = independent renal function.
n
Mean Pcτea1
Dialysis
Fex
Mpred í.v.
Cyclo/ aza
Fred orally
Micropolyarteritis Wegener's
35 17
567 ± 347 402 ± 254
16 3
11 7
21 9
21/19 10/16
33 17
Total
52
19
18
30
30/34
49
Twenty-one patients received both drugs, all but 2 cyclophosphamide first for 8-12 weeks, followed by azathioprine. Pex = Plasma exchange; Mpred = methylprednisolone; Cyclo = cyclophosphamide; aza = azathioprine; Pred = prednisolone.
Downloaded by: Université de Paris 193.51.85.197 - 1/14/2020 11:21:20 AM
Table 4. Renal vasculitis 1981-1990: initial treatment
Renal Vasculitis: Microscopic Polyarteritis and Wegener's Granuloma
43
Table 5. Renal vasculitis: causes of death 1981-1990 In acute illness
Later deaths
age cause of death
age cause of death
Wegener's (n = 17)
75 dialysis withdrawn 61 MI
3 weeks (on dx) 2 weeks
50 ? cause (on CAPD) 71 gram —ve sepsis
Micropolyarteritis (n=35)
64 infarcted bowel 61 cardiac arrest 66 pneumonia 66 pneumonia/WBC/MI 70 bowel perf crumbled 69 GI bleeding/gastrect 37 resp failure (HIV +ve)
5 weeks (on dx) 2 weeks (on dx) 3 weeks (on dx) 3 weeks (on dx) 5 weeks (on dx) 2 weeks (on dx) 3 months (on dx)
59 resp arrest (on CAPD) 5 5 marrow failure (on RDT) 70 neutropenia 4 months 57 inanition (on RDT) 76 ?cause
dx = Acute dialysis; RDT = regular dialysis; CAPD = continuous ambulatory peritoneal dialysis; perf = perforation; gastrect = gastrectomy; resp = respiratory; WBC = neutropenia; MI = myocardial infarct.
Nine patients (2 Wegener's and 7 micropolyarteritis) died during the acute illness, all but 1 whilst still receiving dialysis; the reasons for death are detailed in table 5. Α further 5 patients who survived the acute illness through dialysis failed to recover renal function and continued on dialysis, 2 of these being transplanted successfully. Eight patients (6 with micropolyarteritis and 2 with Wegener's) required dialysis initially, but survived and recovered renal function. Seven patients died later (2 Wegener's and 5 microscopic polyarteritis), 4 after starting renal replacement therapy. The causes of death are again detailed in table 5. Five patients clearly died at least in part because of infection, with or without obvious marrow depression; 3 patients had complications of the vasculitis itself, and 2 suffered myocardial infarcts. The relationship between renal status at most recent follow-up and the percentage of crescents in the initial biopsy is shown in figure 1. There was only a poor relation between the extent of crescent formation and the likelihood of being dead, or requiring regular dialysis, after a mean of 5 years. Two patients with interstitial nephritis and no crescents died (table 3, fig. 1), as did 2 patients with minor degrees of crescent formation (fig. 1).
Downloaded by: Université de Paris 193.51.85.197 - 1/14/2020 11:21:20 AM
Outcome
Cameron
44
Table 6. Renal vasculitis 1981-1990: maintenance treatment in survivors (1-9 years later, mean 5 years) Wegener's
Micropolyarteritis
None Prednisolone alone Prednisolone + azathioprine
0 1 11
3 3 14
Total
12
201
1
Total 3 4 25 321
Details of current treatment not available in 2 patients.
Table 7. Renal vasculitis: outcome 1981-1990 (mean follow-up 4.5 years) Wegener's
Micropolyarteritis
Total 11 19 9 35 9 3 16 51 2
Alive, normal renal function Alive, reduced renal function Required long-term dialysis (deaths) Total alive Died during acute illness Died later off dialysis Total deaths
4 8 1 (1) 12 2 1 4
7 11 81 (3) 22 7 2 12
Total
162
35
1 Two 2
transplanted, both with success. Follow-up not available in 1 patient.
We have continued to treat most survivors (25/32) with prednisolone and azothioprine for a mean of 5 years (table 6), and it is noteworthy that only 2 of these patients experienced relapses, a much lower proportion than that reported from other series in which treatment was tapered or stopped earlier. The outcome at latest follow-up during 1990 is shown in table 7: after a mean of 5 years, 35/52 (67%) patients survived, 12/17 (71%) with Wegener's and 22/35 (63%) with microscopic polyarteritis, 5 patients with microscopic polyarteritis surviving on substitution therapy, 3 on dialysis and
Downloaded by: Université de Paris 193.51.85.197 - 1/14/2020 11:21:20 AM
Maintenance Treatment
Renal Vasculitis: Microscopic Polyarteritis and Wegener's Granuloma
45
100 —
80 —
Wegen er's ø Ζ 60
Ç
ιι
3 w
η
L
`η1
~l
Φ
ώ 40 —
All
Micro-PAN
~--- —► Serra et al., 1982
ο-
20 —
ι
0
I
Discharge from hospital
2
ι
4
ι
6 Years
ι
8
10
Fig. 2. Survival of .patients calculated by the Kaplan-Meler method. Data are displayed (continuous lines) for the whole group of renal small vessel vasculitis (52 patients), and separately for those with Wegener's granuloma (18 patients) and microscopic polyarteritis (35 patients). None of these curves is statistically different at a 1:20 level (Kaplan-Meier). Also shown (interrupted line) are data from the previous cohort of 55 patients described by Serra et al. 1984 [ 1]. This line differs from the recent group of 52 patients at less than 1:20 level, patient survival being inferior in the 1968-81 cohort compared with the 1981-90 cohort.
functioning transplant. Survival curves for the whole group of patients, and for those with Wegener's and microscopic polyarteritis separately, are shown in figure 2, compared with data from our previous cohort of 5 5 patients [1].
Thus, although the prognosis for small vessel vasculitis affecting the kidney has improved from almost inevitable death 30 years ago, through a 30-35% 5-year survival 10-15 years ago [1], many problems remain. Although it seems likely that the more intense immunosuppression used in most units over the past 10-15 years has improved immediate survival, a number of these elderly and often frail patients die directly as a result of these treatment regimens [4]; in our present series, 5/16 deaths (31%) could be related directly to immunosuppression. In some patients extrarenal complications of the vasculitis, particularly in the gastrointestinal tract, still may
Downloaded by: Université de Paris 193.51.85.197 - 1/14/2020 11:21:20 AM
Summary and Conclusions
Cameron
46
lead to early death. The relative roles (if any) of methylprednisolone and/or plasma exchange in treatment of renal vasculitis are not clear yet, although a recent controlled trial in patients with crescentic nephritis, mainly the result of small vessel renal vasculitis, showed a modest benefit from plasma exchange in addition to prednisolone and cyclophosphamide [4], but only in those patients requiring dialysis. In the longer term, we do not know for how long, with what agent and with what intensity immunosuppression must be maintained. In our series, in which most patients were maintained on modest immunosuppression for many years, relapse of the vasculitis was almost absent, suggesting some merit in this strategy. We used azathioprine rather than intermittent intravenous cyclophosphamide, as others have advocated; there are no data to choose between these regimens at the moment, although even in this relatively elderly population long-term oral cyclophosphamide is better avoided because of risks to the bladder, gonadal toxicity and oncogenesis. However, there may well be a subgroup of patients in whom this long-term treatment is unnecessary and therefore dangerous. It may be that the measurement of ANCA in the serum, even if it does not play an immediate pathogenic role, will be useful in predicting relapses and monitoring treatment. However, the continuing controversy about the utility of anti-dsDNA antibodies for this purpose in systemic lupus, now 45 years after their first description, suggests that clear-cut answers may not be evident immediately. References Serra A, Cameron JS, Turner DR, Hartley B, Ogg CS, Neild GH, Williams DG, Taube D, Brown CB, Jicks HJA: Vasculitis affecting the kidney: presentation, histopathology and long-term outcome. Q J Med 1984;53:181-207. 2 Fuiano G, Cameron JS, Raftery M, Hartley BH, Williams DG, Ogg CS: Improved prognosis of renal microscopic polyarteritis in recent years. Nephrol Dialysis Transplant 1988;3:383-391. Serra A, Cameron JS: Clinical and pathologic aspects of renal vasculitis. Serif 3 Nephrol 1985;5:15-33. Neild GH: Infectious complications in the management of systemic vasculitis and 4 rapidly progressive glomerulonephritis (submitted). 5 Pusey CD, Rees AJ, Evans DJ, Peters DK, Lockwood CM: A randomised controlled trial of plasma exchange in focal necrotising glomerulonephritis without anti-GBM antibodies. Kidney Int 1990;38 (in press). Prof. J.S. Cameron, Clinical Science Labs, 17th Floor Guy's Tower, Guy's Hospital, London SE 1 9RT (UK)
Downloaded by: Université de Paris 193.51.85.197 - 1/14/2020 11:21:20 AM
1
