Q J Med 1999; 92: 211-218
© 1999 Association of Physicians
The very long-term prognosis and complications of lupus nephritis and its treatment
From the Renal Unit, UMDS Guy's and St Thomas' Hospitals, London, UK
Received 18 September 1998 and in revised form 5 February 1999
Professor J.S. Cameron, Elm Bank, Melmerby, Cumbria CA10 1HB.e-mail:jstewart_cameron{at}email.msn.com
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
Although the short- and medium-term (5–10 years) outcome of patients with lupus nephritis has been studied extensively, there are very few data on the second and subsequent decades. We studied outcome in 110 local patients investigated at a single centre before 1986, who all had potential follow-up of more than 10 years (actual 2–31 years, median 15.5 years). At last follow-up, 40 patients were dead and 70 alive, nine of whom were on maintenance dialysis or transplanted, actuarial survivals being 84%, 72%, 62%, 61% and 54% at 5, 10, 15, 20 and 25 years for the group as a whole. Survival was better in the cohort 1976–86 (n=60) than in that from 1963–75 (n=50) (90, 81 and 76% vs. 78, 56 and 43% at 5, 10 and 15 years, p<0.001). Sepsis (12) and myocardial infarction (8) were the principal causes of death. Of living patients with renal function, 38% had normal urine and renal function, 11 were off all treatment (19%), 62% had persistent proteinuria and 18% had reduced but generally stable renal function. Renal failure, in those patients who developed it, occurred during the first decade and none of 67 patients actually followed more than 10 years subsequently went into renal failure. Induction treatment was with prednisolone, combined with azathioprine in more severe forms of nephritis, and from the middle 1970s to 1986, 30 with methylprednisolone and in 12 cases plasma exchange. Seventeen other patients were treated using oral cyclophosphamide during the 1960s. No patient received i.v. cyclophosphamide as induction therapy, although nine patients had this form of treatment later, largely because of non-compliance. Serious complications of lupus and/or its treatment occurred in 49%: sepsis in 32, ischaemic heart disease in 20, thrombosis in one and avascular necrosis of bone in eight. In contrast, fracturing osteoporosis occurred in only three, and cataracts requiring surgery and diabetes mellitus in none. The very long-term outlook of lupus nephritis, especially its more severe forms, has improved, but that with current management strategies only a minority of patients are able to stop treatment altogether, and the incidence of serious complications is high.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Lupus is a disease with a peak incidence in adolescence and young adulthood, affecting principally young women.1,2 Until the past 30 years, the prognosis for those with lupus nephritis as part of their disease was very poor, but as prognosis has improved under empirical treatment and more patients survive long term, the later phases of the disease over decades rather than years have become increasingly important.2 This is especially so because the immunosuppressive treatment which has so dramatically altered the outlook carries with it major toxicity.
Nevertheless, data are lacking on outcome of young women with lupus nephritis who survive their first decade. Only the papers of Moroni and colleagues,3 and Donadio et al.4 give information on outcome from 10–20 years from onset. Thus we set out to analyse retrospectively the outcome of our patients studied at Guy's Hospital from 1963 to 1986, who had a potential follow-up of 10 to >30 years.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Records were available on 243 patients with lupus nephritis who had been studied and followed at Guy's Hospital renal unit between 1963 and 1996. All patients had had a renal biopsy and all had at least four manifestations of lupus as described by the American College of Rheumatology.5 Of these 243, 166 had their initial investigation in 1986 or earlier, and thus had 10 years' or more of potential follow-up. Forty-three of these 166 patients were excluded because they came from outside the UK, and the reasons for referral and selection processes might differ from local patients; in addition, follow-up was less complete in this group. Similarly, six patients whose initial investigation as carried out elsewhere within the UK and had tertiary or quaternary referral for various reasons to our unit, were excluded. This left 116 patients, in 110 of whom adequate records were available to allow description of their outcome and the complications they had suffered. All 64 local patients of the 79 included in the paper published in this journal in 19799 were included in the present study.
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Presentation and initial treatment
Details of the 110 patients at the time of presentation and investigation are shown in Tables 1–3
. Eighty-two were local Caucasians, 17 black African/Afro-Caribbean, seven Indo-Asian and four Oriental. Ninety-seven patients (88%) were female and 13 (12%) male. Median age at presentation was 30.2 years (range 7–67 years). Presentation (Table 1) was predominantly with proteinuria, the nephrotic syndrome being the commonest single renal presentation. Two-thirds of patients had had other manifestations of lupus diagnosed before their renal disease became evident.
|
|
|
Approximately half of the patients showed diminished renal function as judged by estimation of glomerular filtration rate using a single injection of 51Cr edetate. About one quarter were hypertensive, or had a requirement for hypertensive treatment.
Renal histological findings at presentation
All patients had a renal biopsy (Table 2
): two-thirds of these showed aggressive patterns of lupus nephritis (WHO classes III and IV) whilst one-fifth (21 patients) showed a membranous pattern of glomerulopathy, although the majority of these had some mesangial deposits. Patients with `mixed' patterns of membranous and focal or diffuse proliferation were classified as class III or IV, respectively.6
Induction treatment
The specific immunosuppression received by the patients as induction therapy is shown in Table 3. The majority of patients with more severe histological appearances received prednisolone plus azathioprine, although 17 patients in the 1960s and early 1970s had courses of oral cyclophosphamide lasting from 3 months to 2 years in duration.7 No patient received intravenous bolus injections of cyclophosphamide for induction, although nine patients received this form of treatment later in their course as part of maintenance therapy, all within the past 5–10 years. In contrast, patients with milder forms of histological lupus nephritis were in general treated with prednisolone alone, or in a few cases no specific treatment to begin with.
Follow-up: treatment, outcome and complications
One hundred and six of the 110 patients were followed until 1996, or until death, for 2–31 years (median 15.5 years). Only four patients were lost to follow-up, 2–18 years from initial investigation. Maintenance treatment consisted of oral prednisolone together with azathioprine in 70 patients and oral cyclophosphamide in nine patients, all treated between 1965 and 1970. Nine patients subsequently received 6–18 months' treatment with intravenous cyclophosphamide, using the protocol of the National Institutes of Health,8 because of relapses failing to respond to azathioprine and prednisolone or intravenous methylprednisolone with suspicion of non-compliance. In general, treatment was continued for a minimum of 5 years from onset before withdrawal of treatment was considered, and in six patients, had to be re-started because of relapse following deliberate or patient-motivated cessation of immunosuppression, in two patients as late as 22 and 25 years following presentation of renal disease. A further patient taking 10 mg of prednisolone had a severe biopsy-proven renal relapse 28 years after presentation followed by irreversible loss of renal function. Because of this prolonged maintenance treatment, relapses were infrequent and no analysis was done of outcome in relation to number of relapses.
Survival
Actuarial analyses of survival of patients form the onset of renal disease are shown in Figure 1. Data are shown also separated for the first decade and the second decade of this study. There is a marked improvement in survival in the 1976–86 cohort compared with the 1963–75 cohort (p<0.01, Kaplan-Meier). For comparison the survival of a more recent cohort, not included in this study, is shown who were investigated and treated between 1987 to 1996: survival was 83% at 10 years, similar to the 1976–86 cohort. Thus during the past decade there has been no further improvement in survival.
|
It was not the purpose of this study to analyse the predictive value of features at onset, in view of our approach of varying therapy according to clinical and histological severity of disease (Table 3
). However, in a univariant analysis no clinical parameters emerged as significant in this respect. Figure 2 shows actuarial analyses of patient survival according to a glomerular appearances on renal biopsy, and b normal or reduced GFR at onset. In neither case does the difference exceed a likelihood of 0.05 (Kaplan-Meier).
|
Figure 3
shows the timing of onset of renal failure in the study group. Despite the fact that in 67 patients actual follow-up exceeded 10 years, no further cases of renal failure were observed from 10–25 years (although since this study was completed in 1996 the patient with the very late relapse mentioned above required dialysis after 29 years follow-up).
|
Most recent status
The most recent status of the 110 patients is shown in Table 4
. Two-thirds of the patients were still alive with renal function, their median age being by this time 46 years. Altogether, 18 patients developed end-stage renal disease and received dialysis or transplantation, except one, in whom renal replacement treatment was withheld. The renal status of those with surviving renal function in relation to treatment is shown also in Table 4. Eleven patients were well and off all treatment. The majority were still receiving immunosuppression, the majority in the form of prednisolone, but some received prednisolone plus azathioprine in an attempt to permit reduction in the dose of corticosteroids.
|
Forty patients had died, and the causes of death (as far as they could be determined) are shown in Table 5
. The main causes of death were sepsis and cardiovascular disease; only three patients had developed malignancy, all lymphomas. Of 12 patients who died from causes other than vascular disease and who had post mortems, eight showed more or less severe coronary atheroma. In a number of cases the causes of death were either obscure or multiple, and only the major cause of death is listed for each patient in Table 5.
|
Complications of disease and treatment
The complications recorded are shown in Table 6
. No patient developed diabetes mellitus, although one patient had suffered type I diabetes before developing lupus. Systematic ophthalmoscopic examination was not carried out, although a number of patients were recorded as showing small posterior cataracts. However no lens removal was needed in any patient in this cohort, although we are looking after one other patient biopsied abroad, who required bilateral operations after 25 years' corticosteroid treatment. Likewise, although a number of women had DEXA bone density estimations after up to 20 years or more follow-up from onset under continuous corticotherapy, with results ranging from high normal to major thinning, only three patients had severe osteoporosis, in two leading to actual fractures. Cardiac echocardiography was not systematically practised during this period, but only one patient required valve replacement for Libman-Sachs endocarditis, in association with a persistently high titre of IgG anti-phospholipid antibody. Altogether 54 (49%) of patients suffered 68 major complications, and 12 patients died as result of these.
|
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The data presented here concern the largest series of patients with lupus nephritis and follow-up of more than a decade hitherto reported in detail. Nevertheless, the deficiencies in this type of study are well known. It is a retrospective open case-note study, gathered over a period during which management changed, not only of the lupus but also of hypertension, infections and other associated problems. Data were not gathered in a systematic and prospective fashion on likely complications. Inevitably, the outcomes reported are to some extent a historical record, and do not represent the likely outcome of patients presenting today, as our own more recent data show the apparent improvement in survival noted in almost all series. However they do indicate qualitatively, and to some extent quantitatively, the type of problems faced in the long term by lupus patients.
Certain features are surprising and some reassuring, for example the lack of induction of steroid-related diabetes, in sharp contrast to findings over a similar period of time following transplantation. It may be that the genotype associated with predisposition to lupus in some way protects against induction of diabetes. With better use of use of immunosuppression, infections are likely to play a lesser role, at least in patients seen today, but the actual incidence of infections (for example herpes zoster) does not seem to have diminished during the period of study, despite improved survival.
Some of the common complications noted in our cohort of patients (thrombosis, infection, induction of lymphoproliferative disorders, avascular necrosis of bone) are shared by the disease of lupus itself and its immunosuppressive treatment,1,2 and are unlikely to disappear. It is a sobering thought moreover that half the patients suffered one or more major complications, and this must be set against the improvement in outcome discussed below. Most of the deaths resulted from complications in whole or in part induced by treatment, particularly sepsis, and not by the lupus per se, although it might be argued that if the treatment is being applied appropriately then death because of a complication of treatment is a secondary effect of the disease. We have analysed causes of death in lupus, including some of the present patients, in more detail in a previous paper.10
The genesis of some of the complications found in lupus such as thrombosis or vascular is multifactorial.2 Antiphospholipid antibodies are only one mechanism that operates to promote thrombosis in patients with lupus, and low plasma factor S concentrations probably are of equal importance. In addition, the many nephrotic patients will suffer from the pro-coagulant effects of hypoproteinaemia. We have reported previously that 44% of our patients with lupus nephritis had anti-phospholipid antibodies,11 and more than half were nephrotic, so a high incidence of thrombosis is not surprising. The pathogenesis of the grossly increased incidence in coronary artery disease compared with normal young or middle-aged women (eight deaths from ischaemic heart disease and eight other women with atheroma at post mortem) is not clear, but apart from possible involvement of enhanced coagulation, interactions between hypercholesterolaemia and circulating immune complexes may contribute;11 a role for corticosteroids remains controversial.
As in the published literature on short-term outcome of lupus nephritis,2 long-term outcome in our study improved during the period of study, and short-term outcome in a subsequent cohort improved even further, both for survival and chronic renal failure. The latter event is now quite rare in lupus nephritis, affecting only about 15% of our patients even in the very long term. It is likely that current cohorts of patients may experience even less renal failure. In our study, almost all cases of end-stage renal disease emerged during the first decade, although we are following a number of patients who have a reduced GFR and increased plasma creatinine concentrations long after this point, and one patient required dialysis 29 years from onset after the present study had been completed. However, many patients in moderate renal insufficiency appear to have stable renal function.
That these improvements are largely the result of immunosuppressive treatment, first with prednisolone alone in the 1950s and 1960s, and then together with cytotoxic agents in the 1970s and since, has never been formally tested against a control group receiving no specific treatment.12 Nevertheless it is generally accepted that this is almost certainly the case, and both a single-agent trial13 and meta-analyses of controlled trials14,15 suggest that the addition of a cytotoxic agent improves outcome over prednisolone alone, although others dissent from this view.16,17 Our data do not permit us to compare different immunosuppressive regimens usefully, and no study to date has demonstrated a superior effect of one immunosuppressive regimen over another (including intravenous bolus cyclophosphamide) when added to prednisolone.2,8,12,16,17,19
We did not attempt any detailed analysis of the predictive power of clinical and histological parameters in this group of patients, given its small size and the variable treatments received during the long period of the study, but a simple univariate analysis did not show any clinical parameter at presentation to be predictive of survival, including glomerular filtration. Nor did glomerular appearance (Figure 2b) predict outcome, although previously we have shown the value of interstitial changes in determining this.18 The loss of predictive value probably results from the very success of current empirical treatment regimens, graded according to clinical and histological severity, in improving survival.
The ultimate goal of management in lupus is the complete suppression of disease and cessation of treatment. Only 11 (19%) of our patients were able to or were allowed to stop treatment altogether. The only other detailed study of patients with lupus beyond 10 years' follow-up is that of Moroni et al.3 They studied 25 Caucasian patients followed more than 10 years (mean 16.7 years), 10 of whom (42%) had discontinued all treatment. These data suggest that at least half of patients with lupus will experience disease requiring immunosuppression for two decades or more. It this connection, it is worth noting that neither our study nor that of Moroni et al. used intravenous bolus cyclophosphamide as induction treatment, and only nine of our patients received it during follow-up for resistant disease, or suspected non-compliance with oral treatment. The short-and medium-term survival of the patients in both these series using predominantly azathioprine as long-term maintenance treatment is equally as good as that reported using i.v. cyclophosphamide.2 It would be interesting to compare these data with similar figures for patients treated with intermittent i.v. cyclophosphamide for one or two years. However only nine patients in the i.v. cyclophosphamide group of the NIH trials had been followed for more than 10 years, and only one for 15 years;19 and so far no data have been reported on their very long-term outcome or final treatment status.
|
Acknowledgments |
|---|
We would like to thank the many colleagues who helped in the care of these patients in the Adult and Paediatric Nephrology units, and the Histopathology department at Guy's over the period of more than 30 years during which these data were collected. Fred Compton gave valuable assistance with the actuarial analyses.
|
Notes |
|---|
* Present address: Servizio Nefrologia e Dialisi, Ospedale `G di Cristina`, Pizza Montalto, 90134 Palermo, Italy
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. Wallace DH, Hahn BH. Dubois' lupus erythematosus, 4th edn. Baltimore, Lea & Febiger, 1993.
2. Cameron JS. Systemic lupus erythematosus. In: Nielson EG, Couser WG, eds. Immunologic renal disease. Philadelphia, Lippincott-Raven; 1997:1055–94.
3.
Moroni G, Banfi G, Ponticelli C. Clinical status of patients after 10 years of lupus nephritis.Q J Med 1992; 84:681–9.
4.
Donadio JV, Hart GM, Bergstralh EJ, Holley KE. Prognostic determinants in lupus nephritis: a long-term clinicopathologic study. Lupus 1995; 4:109–15.
5. Tan EM, Cohen AS, Fries JF, et al. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982; 25:1276–82.
6. Churg J, Bernstein J, Glassock RJ. Renal disease. Classification and atlas of glomerular diseases, 2nd edn. World Health Organization. New York, Ikagu Shoin, 1996:151–78.
7. Cameron JS, Boulton-Jones M, Robinson R, Ogg CS. Treatment of lupus nephritis with cyclophosphamide. Lancet 1970; ii:846–9.
8. Austin HA III, Klippel JH, Balow JE, et al. Therapy of lupus nephritis. Controlled trial of prednisone and cytotoxic drugs. N Engl J Med 1986; 314:614–19.[Abstract]
9.
Cameron JS, Turner DR, Ogg CS et al. Lupus with nephritis: a long term study.Q J Med 1979; 48:1–24.
10. Correia P, Cameron JS, Lian JD, Hicks J, Ogg CS, Williams DG, Chantler C, Haycock G. Why do patients with lupus nephritis die? Br Med J 1985; 290:126–31.
11. Frampton G, Hicks J, Cameron J. Significance of anti-phospholipid antibodies in patients with lupus nephritis. Kidney Int 1991; 39:1225–31.[Web of Science][Medline]
12. Cameron JS. What is the role of long-term cytotoxic agents in the treatment of lupus nephritis? J Nephrol 1993; 6:172–6.
13. Boumpas DT, Austin HA III, Vaughan EM et al. Controlled trial of pulse methylprednisolone versus two regimes of pulse cyclophosphamide in severe lupus nephritis. Lancet 1992; 340:741–5.[Web of Science][Medline]
14. Felson DT, Anderson J. Evidence for the superiority of immunosuppressive drugs and prednisone over prednisone alone in lupus nephritis. N Engl J Med 1984; 311:1528–33.[Abstract]
15. Bansal VK, Beto JA. Treatment of lupus nephritis: a meta-analysis of clinical trials. Am J Kidney Dis 1997; 29:193–9.[Web of Science][Medline]
16. Gruppo Italiano per lo Studio della Nefrite Lupica (GISNEL). Lupus nephritis: prognostic factors and probability of maintaining life-support renal function 10 years after the diagnosis. Am J Kidney Dis 1992; 19:473–9.[Web of Science][Medline]
17. Donadio JV, Glassock RJ. Immunosuppressive drug therapy in lupus nephritis. Am J Kidney Dis 1993; 21:239–50.[Web of Science][Medline]
18. Alexopoulos E, Cameron JS, Hartley BH. Lupus Nephritis: correlation of interstitial cells with glomerular function. Kidney Int 1990; 37:100–9.[Web of Science][Medline]
19. Steinberg AD, Steinberg SC. Long-term preservation of renal function in patients with lupus nephritis receiving treatment that includes cyclophosphamide versus those treated with prednisolone alone. Arthritis Rheum 1991; 34:945–50.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. A. Austin III, G. G. Illei, M. J. Braun, and J. E. Balow Randomized, Controlled Trial of Prednisone, Cyclophosphamide, and Cyclosporine in Lupus Membranous Nephropathy J. Am. Soc. Nephrol., April 1, 2009; 20(4): 901 - 911. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.O. Sun, W.X. Hu, H.L. Xie, H.T. Zhang, H.P. Chen, C.H. Zeng, Z.H. Liu, and L.S. Li Long-term outcome of Chinese patients with membranous lupus nephropathy Lupus, January 1, 2008; 17(1): 56 - 61. [Abstract] [PDF]
|
|
|
|
|
|
V Baca, T Catalan, M Villasis-Keever, G Ramon, A. Morales, and F Rodriguez-Leyva Effect of Low-Dose Cyclosporine A in the Treatment of Refractory Proteinuria in Childhood-Onset Lupus Nephritis Lupus, August 1, 2006; 15(8): 490 - 495. [Abstract] [PDF]
|
|
|
|
|
|
C C Mok Prognostic factors in lupus nephritis Lupus, January 1, 2005; 14(1): 39 - 44. [Abstract] [PDF]
|
|
|
|
|
|
H A Austin and G G Illei Membranous lupus nephritis Lupus, January 1, 2005; 14(1): 65 - 71. [Abstract] [PDF]
|
|
|
|
|
|
D Jayne and A Tyndall Autologous stem cell transplantation for systemic lupus erythematosus Lupus, May 1, 2004; 13(5): 359 - 365. [Abstract] [PDF]
|
|
|
|
|
|
P I Sidiropoulos and D T Boumpas Lessons learned from anti-CD40L treatment in systemic lupus erythematosus patients Lupus, May 1, 2004; 13(5): 391 - 397. [Abstract] [PDF]
|
|
|
|
 |
|
 C Fiehn, Y Hajjar, K Mueller, R Waldherr, A D Ho, and K Andrassy Improved clinical outcome of lupus nephritis during the past decade: importance of early diagnosis and treatment Ann Rheum Dis, May 1, 2003; 62(5): 435 - 439. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. G. Illei, H. A. Austin III, M. Crane, L. Collins, M. F. Gourley, C. H. Yarboro, E. M. Vaughan, T. Kuroiwa, C. L. Danning, A. D. Steinberg, et al. Combination Therapy with Pulse Cyclophosphamide plus Pulse Methylprednisolone Improves Long-Term Renal Outcome without Adding Toxicity in Patients with Lupus Nephritis Ann Intern Med, August 21, 2001; 135(4): 248 - 257. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Moroni, G. C. Greloni, and C. Ponticelli Late recurrence of lupus nephritis after long-term clinical remission Nephrol. Dial. Transplant., April 1, 2001; 16(4): 849 - 852. [Full Text] [PDF]
|
|
|
|
 |
|
 J. Font, M. Ramos-Casals, R. Cervera, M. Garcia-Carrasco, A. Torras, A. Siso, A. Darnell, and M. Ingelmo Cardiovascular risk factors and the long-term outcome of lupus nephritis QJM, January 1, 2001; 94(1): 19 - 26. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||