Nephrology Dialysis Transplantation. Course and prognosis of anti-basement membrane antibody (anti-bm-ab)-mediated disease: report of 35 cases

Transcription

1 Nephrol Dial Transplant (1994) 9: Original Article Nephrology Dialysis Transplantation Course and prognosis of anti-basement membrane antibody (anti-bm-ab)-mediated disease: report of 35 cases F. Merkel, O. Pullig, M. Marx, K. O. Netzer, and M. Weber Medizinische KJinik IV, University of Erlangen-Nurnberg, Germany Abstract. Anti-basement membrane antibody (anti- BM Ab) mediated disease is reported to be a rare disorder frequently leading to severe deterioration of renal function. It was our purpose to work out parameters necessary to predict the outcome reliably and to examine, who will benefit most from therapy. Data from 35 patients were evaluated retrospectively. Diagnosis was based on the detection of linear IgG staining (n = 28) along the glomerular basement membrane (GBM) in renal biopsies and/or on the demonstration of anti-bm Ab both by ELISA and immunoblotting (n = 35). Patients were followed up for at least 6 months. Several variables were analysed as to whether they are appropriate prognostic factors. Twenty patients (57%) presented with Goodpasture's syndrome, 13 (37%) had anti-gbm glomerulonephritis alone, whereas two patients suffered solely from pulmonary haemosiderosis. Frequent initial symptoms were haemoptysis (w=18), haematuria (/z = 26), proteinuria (w = 26) and elevated serum creatinine (n = 27). Among all, 10 patients improved, having stable renal function. Twenty-one patients developed endstage renal failure and four died. Parameters indicating a poor prognosis were a serum (s)-creatinine greater than 600 umol/1 and crescent formation in more than 50% of the glomeruli on renal biopsy. By combining these two parameters the outcome could be reliably predicted. The initial antibody titre and cigarette smoking were without predictive value. In conclusion, the earlier therapy starts, the better will be the result. Patients presenting early with a serum creatinine < 200 umol/1 and without severe glomerular alterations gained the most benefit from therapy, indicating that outcome may be improved by early diagnosis. Introduction Human anti-basement membrane antibody (anti- BM-Ab) mediated disease is an autoimmune disorder characterized by autoantibodies directed against the non-collagenous NCI domain of the a3 (IV) chain of type IV collagen [1,2]. It becomes frequently manifest as Goodpasture's syndrome with progressive glomerulonephritis and pulmonary haemorrhage [3], but it can also be confined to either the kidney or the lung [4-6]. The incidence is estimated to be 0.5 cases per million per year in Great Britain [7]. It accounts for about 20% of all cases of rapidly progressive glomerulonephritis [8]. Older studies reported that only a minority of patients recovered or maintained renal function, whereas most of the patients died or developed end-stage renal failure [9,10]. Introduced in the mid 1970s, standard therapy is a combination of immunosuppression and plasma exchange [11]. Since than recent investigations documented a better survival [6,12,13]. Nevertheless it is still not definitely clear, which parameter is most important for prognosis. Thus it is difficult to decide whether for all patients therapy will be of benefit. In our study we analysed retrospectively 35 cases of anti-bm-ab mediated disease from 1982 to 1992 with respect to the initial presentation, course of the disease, and prognosis. It was our aim to determine those parameters necessary to predict the outcome reliably. Subjects and methods Key words: Goodpasture's syndrome; anti-gbm disease; clinical outcome Correspondence and offprint requests to: Prof. Dr M. Weber, Medizinische KJinik IV, University of Erlangen-Nurnberg, Krankenhausstr. 12, W Erlangen, Germany European Dialysis and Transplant Association-European Renal Association Data from 35 patients who suffered from anti-bm antibody mediated disease were collected between 1982 and 1992 and analysed retrospectively. Case histories and follow-up were obtained from 15 hospitals. The diagnosis of anti-bm Ab mediated disease was based on renal histology («= 28) and/or on the demonstration of anti-basement membrane antibodies (H = 35) using ELISA and immunoblotting techniques [14,15]. Particular attention was focused on symptoms on admission, initial serum creatinine, anti-bm antibody titre, smoking habits and renal histology.

2 Anti-Basement Membrane Antibody Disease ELISA and Immunoblotting Table 1. Clinical findings on admission (n = 35) 373 Microtitre plates were coated with GBM-NC1 isolated from collagenase-digested glomerular basement membranes [14-17]. Anti-BM Ab-titres were estimated by serial dilution of sera with PBS. In addition, positive sera were reanalysed by Western blotting according to standard techniques. Biopsies Percutaneous renal biopsies were evaluated by conventional light- and immunofluorescence-microscopy using standard techniques. In order to compare the severity of glomerular lesions, results were categorized in four different groups as suggested previously [18,19]: (1) segmental mesangial matrix expansion with mild to moderate glomerular hypercellularity; (2) focal and segmental proliferative glomerulonephritis with increased number of multinucleated giant cells, and glomerular crescents in less than 50% of glomeruli; (3) diffuse glomerulonephritis with endocapillary proliferation, diffuse glomerular crescents in more than 50% of the glomeruli, and tubulointerstitial disease; (4) other forms. Pulmonary biopsies were undertaken in two cases and analysed by immunofluorescence-microscopy. Statistical test Data were analysed by logistic regression to determine the variables most important for prognosis [20]. The variables were subsequently ranked depending on the level of significance. Those variables necessary to determine the outcome were calculated. Results Clinical features Twenty of 35 patients (57%) presented with pulmonorenal Goodpasture's syndrome, 13 patients (37%) had anti-gbm nephritis alone, and two patients presented solely with pulmonary haemosiderosis. In these two patients, renal biopsy showed no sign of glomerulonephritis on light-microscopy, although linear IgG staining was documented along the GBM. Altogether, 22 men and 13 women (gender ratio 1.7:1) developed anti-bm-ab-mediated disease. Most of the patients (n = 16) were between 20 and 30 years of age. A smaller second peak of incidence was found within the patients older than 60 years. Common initial symptoms were weakness, fatigue, and reduced physical strength {n = 14, Table 1). Some patients presented with flu-like symptoms («=10). Haemoptysis was found in 18 Anaemia Fatigue and/or weakness Chills and/or fever Haemoptysis Tachypnoea and cyanosis Elevated s-creatinine end-stage renal failure without end-stage renal failure Haematuria Microscopic Macroscopic Proteinuria Pulmonary infiltrates on chest X-ray No % patients. Lung haemorrhage was mild to moderate in most cases except in two patients who had to be ventilated because of respiratory insufficiency. Laboratory tests showed that 27 patients were anaemic (mean haemoglobin concentration: 8.6mg/dl). Haematuria (n = 26) and proteinuria (n = 26, protein excretion g per day) indicated renal involvement. On admission, elevated serum creatinine was noted in 27 patients (mean creatinine: umol/l, range umol/1). Anti-BM-Ab were demonstrated on admission in all patients (mean titre: 1:80, range 1:20-1:320). The results were confirmed by immunoblotting, which showed detection of type IV collagen NCI monomers and dimers in all cases. Biopsies Renal biopsies were successfully performed in 28 cases. Insufficient material was obtained in five, and no biopsy was performed in two patients. Biopsies were categorized according to the extent of glomerular lesions and crescent formation (Table 2). Three biopsies showed mild to moderate glomerular hypercellularity and were subsequently grouped into category 1, four showed crescents in less than 50% of the glomeruli and were classified into category 2, and 20 biopsies showed diffuse glomerulonephritis with endocapillary proliferation and crescent formation in more than 50% of the glomeruli (category 3). One patient presented with mesangioproliferative glomerulonephritis (category 4). Immunofluoresence studies were performed in only 20 biopsies. Linear staining for IgG could be demonstrated in all of them, staining for various complement components in eight, and staining for IgM Table 2. Renal histopathology in patients with anti-bm Ab-mediated disease (n = 28) No. Dialysis Stable 1. Segmental mesangial matrix expansion and hypercellularity 3 2. Focal and segmental proliferative GN with crescent formation < 50% of glomeruli 4 3. Diffuse GN with endocapillary proliferation, glomerular crescent formation in 20 > 50%, and tubulointerstitial disease 4. Other forms 1 19

3 374 F. Merkel et al. in three cases. No staining for IgA was detected. Pulmonary biopsies were performed in two patients. Both showed linear staining for IgG along the alveolar basement membrane. Treatment, clinical course, and prognosis Three patients were not treated because of end-stage renal failure already on admission. Twenty-five patients received plasma exchange (2-10 daily or alternate-day until anti-bm-ab disappeared) combined with steroids ( mg methylprednisolone for 3 days, followed by 1 mg per kg body weight per day) and cyclophosphamide (l-2mg per kg body weight per day or in a pulse dose of g). Seven patients were treated by immunosuppressive drugs alone. In the immunosuppression/plasma exchange group, a favourable course was observed in eight patients who had stable renal function over a follow-up period of at least 6 months. Seven of these eight patients presented with a serum creatinine of less than 600 umol/1, six had a creatinine less than 200 umol/1. Six of seven biopsies were categorized into groups 1 and 2. In contrast 14 patients developed end-stage renal failure. Among them, initial creatinine was > 600 umol/1 in nine patients. Ten of 14 biopsies were grouped into category 3. Furthermore, three patients of this group died. They all presented with an initial creatinine > 600 umol/1 and with kidney biopsies classified into category 3. In all patients with Goodpasture's syndrome haemoptysis was resolved by immunosuppressive therapy. In total, four of our patients (11%) died (3 patients in the therapy group, 1 patient without therapy), because of sepsis («= 2) or cardiac arrest («= 2). Twenty-one of 35 patients developed end-stage renal failure, remained dialysis dependent or were subsequently transplanted. Stable renal function for more than 6 months was observed in 10 patients (29%). Consecutive anti-bm Ab titres were available in 10 patients. Treatment resulted in a decline of serum antibody. After a mean of 2 months (range from 3 weeks to 4 months) anti-bm antibodies became undetectable. Predictive value of clinical, histological, and serological data Among the 35 patients with anti-bm-ab-mediated disease two parameters were found to be of important prognostic value (Table 3). The initial creatinine was determined to be the first important parameter for prognosis. Nine of 14 patients with an initial creatinine below 600 umol/1 improved, whereas most of the patients with a serum creatinine greater than 600 umol/1 required dialysis (20 versus 1). Among those who improved in the course of the disease, eight of 10 were admitted with an initial creatinine of less than 200 umol/1 (Figure 1). Statistical analysis revealed that by taking just this parameter into account, outcome can be predicted in about 86% of all cases (calculated goodness of fit = ). Renal histology was found to be the second important parameter. Taking both together, all patients can be categorized correctly (goodness of fit = 1), thus being highly predictive for prognosis (illustrated in Figure 2). Endstage renal failure on admission was seen in a high proportion (64%) among those patients who remained dialysis dependent. This supports the prognostic impor- Creatinine Initial S-Creatinine S-Creatinine after 6 month Fig. 1. Initial s-creatinine and s-creatinine level 6 month after diagnosis. Table 3. Predictive value of clinical, histological, and serological data Improvement (»=10) End-stage renal failure (" = 25) Level of significance Goodness of fit Age (years) Smokers (%) Mean ELISA-titre Pulmonary involvement (%) Sex (m/f) Initial s creatinine (umol/1) Biopsy category 1-2 («) : / : / ' 1

4 Anti-Basement Membrane Antibody Disease H o QOOOOO " Biopsy-category O stable renal function Dialysis Fig. 2. Serum-creatinine at presentation and category on renal biopsy. tance of the initial serum creatinine value. We could not find any correlation between antibody titre at time of admission and fate of renal function. Cigarette smoking had no influence on the clinical course of the disease or the development of haemoptysis. For this statement data were available from 21 patients. Among those who improved in the course of the disease, five were smokers and three non-smokers, whereas among those who became or remained dialysis dependend five were smokers and eight were not. Discussion Anti-basement membrane antibody mediated disease is a human autoimmune disorder with poor prognosis. Early series showed that more than 96% of the patients died [9], either due to haemoptysis (54%) or to renal failure (46%). The introduction of immunosuppressive therapy with cyclophosphamide, prednisolone and additional plasma exchange reduced the overall mortality substantially (21%/17%) [6,21]. However, a major proportion of patients (66%/78%) became dialysis dependent. In our study mortality was reduced to 11%. No patient died due to lung haemorrhage. Nevertheless, 21 of 35 patients (60%) developed endstage renal failure. Predominant findings at the beginning of the disease were haematuria, proteinuria, haemoptysis, and elevated serum creatinine. These baseline characteristics are similar to previously reported results [6,18]. Within the former studies there are several singular parameters, i.e. the initial serum creatinine [6], the glomerular alterations found on kidney biopsy [21], or end-stage renal failure on admis- sion [22] noted to be in association with poor prognosis. Still the significance of all the parameters with respect to the final outcome is less clear. Our analysis showed that the outcome could be reliably predicted when the extend of endocapillary proliferation and the degree of glomerular crescents were considered together with the initial serum creatinine. Patients with initial creatinine greater than 600 umol/1 and renal histology of category 3 never improved. More drastically, among those who improved most had an initial creatinine of less than 200 umol/1. This might be of clinical importance. It seems that the earlier immunosuppressive therapy starts, the better will be the result. This may simply reflect the duration of anti-bm Ab mediated inflammation prior to diagnosis. Moreover, potentially harmful treatment might be avoided with the exception of those patients with severe haemoptysis. In our analysis gender and age distribution were found to be without prognostic importance. Among our patients we could not confirm a previous report that cigarette smoking is associated with a deterioration of the disease [23]. Furthermore, we were not able to show any correlation between the initial antibody titre and the outcome of the disease. In the last decade considerable progress has been made to elucidate the pathogenesis of anti-bm-abmediated disease. It could be shown that the immune injury results from antibodies against the a3(iv) chain of collagen type 4 [1,2,16,24]. Hopefully, the development of recombinant oe3(iv) proteins may allow a more successful elimination of the pathogenetic relevant anti-bm-ab which possibly may improve the course of the disease. In conclusion, immunosuppression and plasma exchange has reduced mortality of anti-bm-abmediated disease compared to earlier studies. However, in patients with an initial creatinine concentrations greater than 600 umol/1, and glomerular crescents in more than 50% of the glomeruli, renal function never improved in spite of immunosuppressive therapy. Thus, early diagnosis using reliable ELISA and immunoblotting techniques seems to be the most promising approach to improve the outcome. Acknowledgements. We wish to thank M. Kleinschmidt, Institut fur Arbeit und Technik, Gelsenkirchen, Germany, for his support on statistical analysis. We further greatfully acknowledge the assistance of Dr K. Andrassy, Heidelberg; Dr R. Baethke, Oldenburg; Dr W.H. Boesken, Trier; Dr R. Fiinfstuck, Jena; Dr P Hecking, Bochum; Dr K..M. Koch, Hannover; Dr H. Kohler, Mainz; Dr K.F. Kopp, Munchen; Dr M. Molzahn, Berlin; Dr Th. Philipp, Essen; Dr G. Schutterle, GieBen; Dr V. Schulz, Heidelberg; Dr W. Schulz, Bamberg; and Dr R. Valentin, Bielefeld for their communication of clinical data. This work was supported by a grant of the Deutsche Forschungsgemeinschaft, SFB 263, C 6 to MW. References 375 I. Hudson BG, Wieslander J, Wisdom BJ, Noelken ME. Goodpasture syndrome: molecular architecture and function of basement membrane antigen. Lab Invest 1989; 61:

5 Weber M. Basement membrane proteins. Kidney Int 1992; 41: Stan ton MC, Tange JD. Goodpature's syndrome. Pulmonary haemorrhage associated with glomerulonephritis. Aust NZ J Med\95%;l: Mathew TH, Holbs JB, Kalowski S, Sutherland PW, Kincaid- Smith P. Goodpasture's syndrome: normal renal diagnostic findings. Ann Intern Med 1975; 82: Zimmermannn SW, Varanasi VR, Hoff B. Goodpasture's syndrome with normal renal function. Am J Med 1979; 66: Savage COS, Pusey CD, Bowman C, Rees AJ, Lockwood CM. Antiglomerular basement membrane mediated disease in the British Isles Br Med J 1986; 292: Turner N, Rees AJ. Antiglomerular basement membrane disease. Oxford Textbook of Nephrology. Oxford University Press, Oxford; 1991: Couser WG. Rapidly progressive glomerulonephritis: classification, pathogenetic mechanisms, and therapy. Am J Kidney Dis 1988; 11: Benoit FL, Rulon DB, Theil GB, Doolan PD, Watten RH. Goodpasture's syndrome. A clinicopathologic entity. Am J Med 1964; 37: Wilson CB, Dixon FJ. Antiglomerular basement membrane antibody induced glomerulonephritis. Kidney Int 1973; 3: Lockwood CM, Rees AJ, Pearson TA, Evans DJ, Peters DK. Immunosuppression and plasma-exchange in the treatment of Goodpasture's syndrome. Lancet 1976; 1: Johnson JP, Whitman W, Briggs WA, Wilson CB. Plasmapheresis and immunosuppressive agents in antibasement membrane antibody-induced Goodpasture's syndrome. Am J Med 1978; 64: Walker RG, Scheinkestel C, Becher GJ, Owen JE, Dowling IP, Kincaid-Smith P. Clinical and morphological aspects of the management of cresentic anti-glomerular basement membrane F. Merkel et al. antibody (anti-gbm) nephritis/goodpasture's syndrome. Q J Med 1985; 54: Weber M, Meyer zum Buschenfelde KH, Kohler H. Immunological properties of the human Goodpasture antigen. Clin Exp Immunol 1988; 74: Weber M, Manns M, Meyer zum Buschenfelde KH, Kohler H. Dot-immunobinding assay with globular domain of collagen type IV for antiglomerular basement membrane antibodies. J Clin Lab Anal 1988; 2: Weber M, Pullig O. Different immunological properties of the globular NCI domain of collagen type IV isolated from various human basement membranes. Eur J Clin Invest 1992; 22: Wieslander J, Bygren P, Heinegard D. Antiglomerular basement membrane antibody: antibody specifity in different forms of glomerulonephritis. Kidney Int 1983; 23: Teague CA, Doak PB, Simpson IJ, Rainer SP, Herdson PB. Goodpasture's syndrome: an analysis of 29 cases. Kidney Int 1978; 13: Couser WG. Goodpasture's Syndrome. The Principles and Practice of Nephrology. B.C. Decker, Philadelphia 1991: Norusin MJ. SPSS. Advanced statistic user's guide. SPSS Chicago 1990: Briggs WA, Johnson JP, Teichmann S, Yeager HC, Wilson CB. Antiglomerular basement membrane antibody-mediated glomerulonephritis and Goodpasture's syndrome. Medicine 1979; 58: Macandoro J, Sobel A. Goodpasture's syndrome. Development of its prognosis from 1955 to Presse Med 1983; 12: Donaghy M, Rees AJ. Cigarette smoking and lung haemorrhage in glomerulonephritis caused by autoantibodies to glomerular basement membrane. Lancet 1983; 2: Saus J, Wieslander J, Langeveld JPM, Quinones S, Hudson BG. Identification of the Goodpasture's antigen as the a3(iv) chain of collagen IV. J Biol Chem 1988; 263: Received for publication Accepted in revised form