| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Rheumatology 2001; 40: 821-825
© 2001 British Society for Rheumatology
Original Papers |
Renal type AA amyloidosis associated with rheumatoid arthritis: a cohort study showing improved survival on treatment with pulse cyclophosphamide
Departments of Immunology and Rheumatology and
1 Pathology, Hôpital E. Herriot, Lyon, France
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and methodsResultsDiscussionReferences |
|---|
Objective. To determine the incidence of renal AA amyloidosis and its association with rheumatoid arthritis (RA) in a cohort of all renal biopsies at one referral hospital and to measure the effect of a monthly pulse of cyclophosphamide on renal function and survival in these RA patients.
Method. All renal biopsies with proven AA amyloidosis from a single pathology unit linked to a major nephrology referral unit in a university hospital were selected retrospectively and RA patients were identified. We studied 6931 renal biopsies. The effect of treatment with and without pulse cyclophosphamide on renal function and survival was studied in these patients.
Results. From March 1977 to February 1999, the incidence of AA amyloidosis was 2.4 cases/yr. The incidence and prevalence of the association of AA amyloidosis with RA were 0.68 cases/yr and 0.22% (15/6931) respectively. RA patients treated with cyclophosphamide (n=6) had a lower rate of renal function loss (P=0.013) and a higher median survival (P=0.026) than untreated patients (n=9). During the follow-up period, two out of six treated patients (33%) and all nine untreated patients (100%) died.
Conclusions. AA amyloidosis is a rare complication of RA and complicates the evaluation of treatment. This retrospective study suggests that treatment with cyclophosphamide is able to reduce the incidence of end-stage renal failure and to increase survival. Prospective studies are needed to clarify this issue.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
AA amyloidosis is one of most severe complications of rheumatoid arthritis (RA). The disease leads to rapid end-stage renal failure and renal dialysis is often unable to prevent early death [1]. AA amyloidosis is rarely seen with improved treatment of chronic infectious diseases, and rheumatic disease is now a leading cause of AA amyloidosis [2].
No treatment has been demonstrated to be able to reduce amyloid deposits, and the rarity of their association with RA complicates studies of the treatment of the latter disease. However, open studies have suggested that treatment with various immunosuppressive drugs may be protective [3–5]. Such treatment may slow progression by controlling the rheumatic inflammatory process.
Because placebo-controlled studies are difficult to perform in these patients, we selected all cases of renal biopsy-proven AA amyloidosis associated with RA from a single pathology unit and compared renal function and mortality in patients treated or not treated with pulses of cyclophosphamide (CYC).
|
Patients and methods |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
Patients
Between March 1977 and February 1999, we analysed 6931 renal biopsies in a single pathology unit linked to a major nephrology referral unit in a university hospital. In February 1999, the diagnoses were reviewed by examination of the pathology unit database. The list of AA amyloidosis and AL amyloidosis cases was isolated retrospectively and all patients with RA according to the American College of Rheumatology criteria [6] were selected. Sex, age, duration of RA from the first symptoms, age at diagnosis of amyloidosis by renal biopsy, serum creatinine and 24-h proteinuria, determined by routine methods, were recorded retrospectively from the medical charts.
Collection of data from RA patients
The retrospective collection of data from RA patients started when the AA amyloidosis diagnosis was obtained from the renal biopsy. Variations in serum creatinine and 24-h proteinuria were calculated from the same baseline until the beginning of renal dialysis or last contact. We calculated survival from the diagnosis of AA amyloidosis until death or, for patients still alive, the most recent contact. Information on the date of death was obtained from the hospital chart, the patient's general practitioner or the registry office of the local civil authorities.
Treatments
Treatments of RA patients were also recorded retrospectively from the medical charts. Patients with RA associated with AA amyloidosis were treated or not with intravenous (i.v.) pulse CYC (Endoxan; Asta Medica, Merignac, France) at a dose of 1 g/month for 6 months and 1 g every 3 months thereafter. Patients were also given mesna (Uromitexan; Asta Medica, Merignac) at a dose of 200 mg i.v. 1 h before the CYC pulse and 4 and 8 h after the pulse.
Statistical analysis
Results are given as mean±S.D. (range). Comparisons between groups were performed with Student's t-test or 
2 analysis as appropriate. Within-group comparisons of serum creatinine and proteinuria changes were performed with the paired t-test. Survival curves were obtained using the Kaplan–Meier method. Differences between curves were evaluated using the log-rank method. Analysis was performed with SAS software.
|
Results |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
Incidence of RA in AA amyloidosis patients
Over a period of 22 yr, amyloidosis was diagnosed in 87 out of 6931 renal biopsies (1.26%) and AA amyloidosis in 53 (0.76%) of these. For AA amyloidosis, rheumatic diseases were the leading cause (18/53, 34.0%), with RA (15/53, 28.3%) representing 0.22% of all renal biopsies, followed by familial Mediterranean fever (9/53, 17.0%), chronic infections (8/53, 15.1%), granulomatous infections (6/53, 11.3%), cancers (2/53, 3.8%), sarcoidosis (2/53, 3.8%) and inflammatory bowel disease (1/53, 1.9%) (Table 1
). The incidence of AA amyloidosis was 2.4 cases/yr in our cohort of renal biopsies. The incidence of RA in this cohort of renal AA amyloidosis patients was 0.68 cases/yr.
|
Characteristics of RA patients
The two sexes were represented similarly in the 15 RA patients (eight women, seven men). Age at onset of RA was 41±11 (20–58) yr and age at diagnosis of amyloidosis 60±13 (29–77) yr. The median duration between the first symptoms of RA and a biopsy-proven diagnosis of AA amyloidosis was 19±11 (5–38) yr.
Renal follow-up and survival of RA patients
The 15 RA patients were not assigned at random to a treatment group, but all treated patients with CYC (n=6) were followed in a single unit and all untreated patients (n=9) in another unit (Table 2). In the latter group, patients received either penicillamine D alone (one patient) or associated with colchicine (two patients) or no drug regimen (six patients). Patients treated with CYC received a mean dose of 22±19 (7–58) g over a period of 27±25 (9–77) months. None received methotrexate or azathioprine from the date of diagnosis. Creatinine concentration increased significantly only in the untreated group (P=0.013). During the follow-up period, two of six (33%) treated patients died compared with all nine untreated patients. Figure 1 shows Kaplan–Meier curves of the survival of the treated and untreated groups. By log-rank test, treated patients had a significantly greater survival rate (P=0.026). The median survival of the untreated group was 46 months vs 165 months in the treated group. Causes of death in the treated group were ischaemic heart disease (one patient) and sepsis (one patient) and in the untreated group heart failure (five patients), sepsis (two patients) and multiorgan failure following hip fracture (one patient). No autopsy was performed.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
AA amyloidosis is mostly associated with chronic infectious diseases in developing countries, while in industrialized countries non-infectious inflammatory conditions are the most common cause [7]. The two series of patients from the Mayo Clinic described in 1949 [8] and 1990 [1] clearly indicate the switch from infectious diseases to rheumatic diseases as the leading cause (Table 1
). In our study, the distribution of the disorders associated with AA amyloidosis was similar to that reported in the recent literature [1, 9]. Furthermore, RA is now the most common associated condition. In addition, the ratio of AA to AL amyloidosis shows marked variation with time and the location of the study, ranging from 1.7:1 to 1:17 [1, 8, 9]; the ratio in our study was 1.6:1. AA amyloidosis is a rare complication of RA, but there is considerable variation regarding its reported prevalence. In the literature, the prevalence varies from 4 to 26% of patients with RA [2], with a median of 14% when several series are combined [10]. The prevalence was much higher (14–61%) when AA amyloidosis in RA was detected post-mortem [11]. In a recent population-based series of subjects with RA who had died but in whom AA amyloidosis was detected before death, AA amyloidosis was detected in 5.8% of the cases [12]. While prevalence at autopsy seems to be similar in most countries, results obtained by tissue sampling and clinical assessment vary considerably. Moreover, it is difficult to assess the incidence of AA amyloidosis in a RA population. It was estimated to be 0.94 cases/yr within one referral unit [1], similar to the 0.68 cases/yr observed in our study.
Our study further evaluated the delay between the onset of RA and that of AA amyloidosis. Surprisingly, the median interval between the onset of RA and a biopsy-proven diagnosis of AA amyloidosis appears to show little variation. The interval in two previous studies was 17 yr [4] and 19 yr [12], identical to the 19 yr observed here. It is possible that the use of better treatments (such as methotrexate) early in the course of RA may further reduce the incidence and delay the onset of secondary AA amyloidosis [7].
Previous studies of AA amyloidosis populations have indicated a median survival of 24.5 months [1] and 38 months [12] after AA amyloidosis diagnosis without treatment, or treatment without demonstrated efficacy. The median survival period of 46 months observed in our study is in the same range. The two main causes of death in our study were ischaemic heart disease and infections.
Few studies of large AA amyloidosis populations have evaluated the effect of treatment. In two successive small studies, with almost the same series of 10 RA patients treated mainly with chlorambucil, no significant variation was seen in serum creatinine over 10 yr [3, 4]. Our study indicated a loss of renal function in the absence of CYC treatment; however, the increase in creatinine was not significant in the treated patients, indicating a reduction in the rate of loss of renal function.
Obviously, a possible effect on survival remains the major goal of treatment. Chlorambucil was evaluated in juvenile chronic arthritis (JCA) [13, 14]. In a retrospective study of 79 JCA patients with reactive amyloidosis, chlorambucil appeared to confer long-term survival and remission [14]. One prospective randomized trial performed in patients with RA and AA amyloidosis compared various treatments (mainly podophyllotoxin derivatives, but also chlorambucil, azathioprine and CYC) with no treatment [5]. The death rate in the untreated group (n=11) at 60 months was 63 vs 11% in the treated group (n=11, P<0.04). In our study, patients were not assigned randomly to a treatment group, as the treatment was related to the unit where the follow-up was performed. Although diagnosis was obtained by renal biopsy performed in a single unit, all treated patients were followed in a single unit and all untreated patients in other units. The results demonstrated a dramatically increased median survival period in the treated group.
No placebo-controlled study, using a single treatment and assessing survival, has been performed. In the present study, patients treated with i.v. CYC, as used for lupus nephritis, seemed to show stabilized renal function, a reduced incidence of end-stage renal failure and increased survival compared with patients not treated in this way. The small size of our population, although it was among the largest in which a single treatment was used, reflects the low incidence of the disease. It is obvious that multicentre trials are needed. However, these preliminary results may limit the need for a placebo arm. Unanswered questions remain about the type and mode of immunosuppression and its duration. Intensive immunosuppressive and myeloablastic therapy may be of interest in this context.
|
Acknowledgments |
|---|
We thank F. Munoz for performing the statistical analysis.
|
Notes |
|---|
Correspondence to: P. Miossec, Department of Immunology and Rheumatology, Hôpital E. Herriot, 5 place d'Arsonval, 69437 Lyon Cedex 03, France
|
References |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
- Gertz MA, Kyle RA. Secondary systemic amyloidosis: Response and survival in 64 patients. Medicine 1991;70:246–56.[Medline]
- Buxbaum J. The amyloidoses. In: Klippel JH, Dieppe PA, eds. Rheumatology, 2nd edn. London: Mosby1998; 8:27/21:10.
-
Berglund K, Keller C, Thysell H. Alkylating cytostatic treatment in renal amyloidosis secondary to rheumatic disease. Ann Rheum Dis1987;46:757–62.
[Abstract/Free Full Text] - Berglund K, Thysell H, Keller C. Results, principles and pitfalls in the management of renal AA-amyloidosis: a 10–21 year followup of 16 patients with rheumatic disease treated with alkylating cytostatics. J Rheumatol 1993;20:2051–7.[Web of Science][Medline]
- Ahlmen M, Ahlmen J, Svalander C, Bucht H. Cytotoxic drug treatment of reactive amyloidosis in rheumatoid arthritis with special reference to renal insufficiency. Clin Rheumatol1987;6:27–38.[Web of Science][Medline]
- Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Pries F, Cooper NS et al. The American Rheumatoid Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum1988;31:315–24.[Web of Science][Medline]
-
Hazenberg BPC, van Rijswijk MH. Where has secondary amyloid gone? Ann Rheum Dis2000;59:577–9.
[Free Full Text] -
Dahlin DC. Secondary amyloidosis. Ann Intern Med 1949;31:105–19.
[Abstract/Free Full Text] - Janssen S, van Rijswijk MH, Meijer S, Ruinen L, van der Hem GK. Systemic amyloidosis: a clinical survey of 144 cases. Neth J Med1986;29:376–85.[Web of Science][Medline]
- Missen GAK, Taylor JD. Amyloidosis in rheumatoid arthritis. J Pathol Bacteriol1956;71:179–92.[Medline]
-
Dhillon V, Woo P, Isenberg D. Amyloidosis in the rheumatic diseases. Ann Rheum Dis1989;48:696–701.
[Free Full Text] -
Myllykangas-Luosujärvi R, Aho K, Kautiainen H, Hakala M. Amyloidosis in a nationwide series of 1666 subjects with rheumatoid arthritis who died during 1989 in Finland. Rheumatology1999;38:499–503.
[Abstract/Free Full Text] - Ansell BM, Eghtedari A, Bywater EGL. Chlorambucil in the management of juvenile chronic polyarthritis complicated by amyloidosis. Ann Rheum Dis1971;30:331.
- David J, Vouyiouka O, Ansell BM, Hall A, Woo P. Amyloidosis in juvenile chronic arthritis: a morbidity and mortality study. Clin Exp Rheumatol1993;11:85–90.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. Nakamura, S. Higashi, K. Tomoda, M. Tsukano, S. Baba, and M. Shono Significance of SAA1.3 allele genotype in Japanese patients with amyloidosis secondary to rheumatoid arthritis Rheumatology, January 1, 2006; 45(1): 43 - 49. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Ortiz-Santamaria, A. Olive, M. Valls-Roc, and X. Tena Treatment of AA amyloid with chlorambucil Rheumatology, July 1, 2002; 41(7): 833 - 833. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||