CASE PRESENTATION

A 49-year-old Portuguese man presented five years ago with a two-day history of shortness of breath. Over the previous two to three months, he had noticed increasing swelling of his ankles, especially after prolonged standing and, for four to five days prior to admission, moderate central abdominal bloating. When directly questioned, he also reported that he had noticed increased frothiness on urination for the previous three months. He denied paroxysmal nocturnal dyspnea and orthopnea. There was no history of fever, skin rash, or arthralgias. His medical history included mild hypertension for five years, controlled by lisinopril, 10 mg/day. He had a 25 pack-year smoking history. He had no known allergies and no family history of renal disease or diabetes. He had been a construction worker until six months prior to admission, but he had had no exposure to chemicals or hydrocarbons. The patient was divorced and had no children.

Physical examination revealed a slightly overweight man whose blood pressure was 160/85 mm Hg and whose heart rate was 76 beats/min and regular. His jugular venous pressure was difficult to detect. His chest was clear. Abdominal examination revealed shifting dullness; he had severe edema of both legs and the sacrum. Urinalysis showed 4+ protein and 3+ blood; microscopic analysis disclosed 8 to 10 red blood cells and 3 to 5 granular casts per high-power field.

Laboratory examination showed normal hematology, electrolytes, and liver function tests. The serum creatinine was 1.4 mg/dL; corrected creatinine clearance, 70 mL/min; proteinuria, 13.7 g/day; and serum albumin, 1.9 g/dL. All laboratory screenings for secondary causes were negative or normal including complement profile, antinuclear antibody, antineutrophilic cytoplasmic antibodies, hepatitis B and C serology, HIV serology, and serum immunoglobulins. A Doppler ultrasound examination showed normal renal size with patent veins and no other abnormality. A chest radiograph was normal and examination of the stools for occult blood was negative 3. Serum cholesterol was 425 mg/dL; LDL, 232 mg/dL; and serum triglycerides, 180 mg/dL. Renal biopsy revealed 27 glomeruli; of these, 2 were globally sclerosed and the rest showed diffuse moderate thickening of the glomerular basement membrane (GBM) with minimal mesangial matrix increase and focal mild patchy interstitial edema and tubular atrophy, but no cellular proliferation. Immunofluorescence microscopy revealed 3+ staining for both IgG and C3 in a diffuse granular pattern along the GBM. Results were negative for all other immunoglobulins. Electron microscopy revealed diffuse epithelial cell foot process effacement and numerous, uniformly sized, electron-dense deposits along the GBM in a subepithelial location with projections of the GBM between the deposits. There were no mesangial deposits.

The patient initially was treated by his local nephrologist with corticosteroids (prednisone, 2 mg/kg) on alternate days for eight weeks. The proteinuria remained> 10 g/day and the patient remained edematous; he developed cushingoid facies and gastrointestinal distress. The prednisone was tapered and the patient was managed conservatively. After six months of continued angiotensin-converting-enzyme (ACE) inhibitor therapy and dietary protein restriction (0.8 g/kg/day), the patient remained severely nephrotic and symptomatic. He was referred for a second opinion, and three options were explained to him with their inherent risks and benefits: continued conservative treatment, a six-month course of alternating a cytotoxic agent with corticosteroids, or cyclosporine. The patient was concerned about additional steroid effects and about infertility, and he opted for the course of cyclosporine. Cyclosporine was begun at 3 mg/kg/day, with the trough level by monoclonal assay targeted between 150 and 225 ng/mL. The proteinuria gradually declined and his renal function remained stable. After eight months, his proteinuria was <0.2 g/day and his serum creatinine was 1.2 mg/dL; the corrected creatinine clearance was 90 mL/min. The cyclosporine was discontinued after 10 months. His cholesterol profile normalized but his hypertension persisted. The ACE inhibitor continued to provide good blood pressure control, and at his most recent follow-up visit two months ago, the serum creatinine was 1.2 mg/dL; creatinine clearance, 88 mL/min; and urine protein excretion, 0.12 g/day.

DISCUSSION

DR. DANIEL CATTRAN (Associate Director of Nephrology, University Health Network, Professor of Medicine, University of Toronto, Toronto, Ontario, Canada): Membranous glomerulonephropathy (MGN) remains the most common cause of adult-onset nephrotic syndrome in the world¹. Its high incidence, despite its overall good prognosis, still makes MGN the second or third most common primary glomerulonephritis in world registries to progress to end-stage renal disease^2,3. Although in the industrialized countries the most common variant is idiopathic MGN, infectious agents such as malaria in Africa^4,5 and hepatitis B in parts of Asia are more frequent causes⁶. Etiology also varies with age. In Europe and North America, a variety of malignant tumors have been associated with MGN, with increased frequency beyond the fifth decade⁷. This patient's ethnicity, site of residence, age, and the negative screening tests for secondary causes strongly support an idiopathic etiology, and I will focus on that variant of MGN in this Forum.

Pathophysiology

The Heymann model of experimental membranous nephropathy in rats, which closely mimics the human disease, suggests that the highly specialized and terminally differentiated glomerular epithelial cell or podocyte is the target of injury⁸. Research in the experimental model has focused on the identification of the responsible antigen(s) and the subsequent immune response, the role of complement, and the delineation of the injury process following the activation of complement⁹. The antigenic targets of the antibody response in the experimental model have been localized to the membrane of the glomerular epithelial cell and are specifically a 515 kD glycoprotein called megalin, a polyspecific receptor related to the low-density lipoprotein receptor family, and an associated 44 kD protein, known as the receptor-associated protein or RAP⁸. A primary role for the T-cell is clear in this model. A CD4+ T-cell-dependent, humoral response results in glomerular immunoglobulin deposition and complement activation. The CD4+ help for antibody response is a function of the Th2 cell, which produces interleukin (IL) -4, -5, -6, -10, and -13. Early in the course of the disease, both glomerular and interstitial T-cells as well as macrophage infiltrates are found. A role for cell-mediated injury is supported by the observations that depletion of the cytotoxic CD8+ T-cell reduces the injury and that monoclonal anti-CD4 and anti-CD8 treatment modifies the disease¹⁰. Complement activation also is important in producing the glomerular injury. The finding of the C5b–9 membrane attack complex (MAC) within the immune deposits and the recognition that depletion of complement by cobra venom serum prevents the subsequent proteinuria confirm the role of complement in the pathogenesis. The podocyte normally retrieves the MAC from the immune deposits by endocytosis, transports them across the cell, and discharges them by exocytosis into the urinary space¹¹. This process might account for the appearance of these fragments in the urine of patients with membranous nephropathy¹². How this sublytic injury from the complex produces proteinuria is unclear. The MAC can activate the formation of radical oxygen species in the rat model, and treatment with anti-oxidants significantly reduces the proteinuria without preventing formation of the deposit⁸. The MAC also can induce production of specific proteinases, such as gelatinase and metalloproteinase-9, and these proteinases might lead to collagen degradation in the glomerular basement membrane (GBM) and subsequently to increased protein permeability. The podocyte response to injury is not a proliferative one, but rather hypertrophy and increased matrix production. Specific isoforms of transforming growth factor beta and their receptors are up-regulated in the model, and this up-regulation might mediate accumulation of the excess matrix in experimental and human disease¹³.

The cellular response to complement attack, rather than a proliferative reaction, occurs through alteration in the expression of specific cell-cycle proteins and results in podocyte hypertrophy, excess matrix formation, and glomerulosclerosis, according to studies by Shankland et al¹⁴. Other authors have suggested that the cell's response to injury is due to activation of specific signal pathways. Cybulsky et al recently have shown that cytoplasmic phospholipase A₂ was increased by C5b-9 in glomeruli of rats with Heymann nephritis¹⁵. This increase in turn released more arachidonic acid precursors, which are important in the synthesis of eicosanoids, such as prostaglandins, thromboxanes, and leukotrienes, all substances that have been implicated in GBM injury. Interruption of the phospholipase activation could lead to a new approach to the treatment of the human disease. The molecular pathogenesis of human membranous nephropathy, however, remains largely unknown. Despite the similarities in the pathology and the elegant dissection of the molecular basis of the injury in the experimental model, parallel findings in the human disease remain elusive. Neither the antigenic equivalent to megalin nor the beneficial effects of the various experimental therapeutic agents have been duplicated in humans.

The difficulty in extrapolating information from the experimental model to human disease suggests that certain variables in the human host are important. The exploration of genetic factors that influence susceptibility and/or progression are preliminary, but studies have shown a higher-than-expected association with the HLA-DR2 allele in the Japanese population¹⁶ and the literature contains descriptions of a few cases in identical twins¹⁷. A recent study using molecular techniques in a cohort of patients with idiopathic MGN from England and Greece extended an earlier HLA serologic association with DR3¹⁸. A particular variation in the gene polymorphism of the DR locus, DRB1*0301, was associated with the disease. No relationship was apparent amongst any of the tested genotypes with severity or progression of the process, but the numbers of patients were small and the observation period limited. The relationship of genetic factors to human idiopathic MGN is still preliminary, but given the variability in its natural history, it is highly likely that susceptibility factors and factors associated with progression will be found.

Natural history

The natural history of idiopathic MGN has been documented in several studies and must be integrated into any management plan. Although a wide variation in outcome has been reported, a review of 11 reports of the natural history of the disease demonstrated a 10-year renal survival within the relatively tight band of 70% to 90%¹⁹. A more current pooled analysis of 32 studies estimated survival between 65% and 75% at 10 years and 60% at 15 years Figure 1²⁰. Some of even this small scatter is probably the result of variations in diagnostic criteria, choices of end point, baseline characteristics, and statistical techniques²¹. This relatively good prognosis is likely an underestimate of today's outcome, given the introduction within the last decade of more potent antihypertensive medications and lowered ideal levels for both systolic and diastolic blood pressure.

Figure 1.

Probability of renal survival from a pooled analysis of all 32 studies (reproduced with permission from the National Kidney Foundation)²⁰.

Full figure and legend (12K)

Gender seems to influence progression. Support for this can be gleaned by looking at gender ratio at presentation versus at end stage. In France the incidence of idiopathic MGN is between 1.2 and 1.7 per 100,000 population, with an age peak between 60 and 79 years. In all age groups, the gender ratio was almost equal²². In Japan and elsewhere, a similar gender distribution exists at the start of the process^23,24. However, studies have consistently found that the ratio at end-stage renal disease has shifted to 2 to 3:1 male:female^2,3. These data indicate that gender strongly influences the severity of the disease. Moreover, the only two factors associated with spontaneous remission are persistent low-grade (subnephrotic) proteinuria and female gender²⁵. The relapse rate from complete remission is also high, varying between 30% and 50%, but prognosis for these patients remains excellent, with urinary protein excretion in the great majority remaining subnephrotic and only 5% progressing to chronic renal insufficiency^25,26. Overall, approximately 30% of patients undergo spontaneous remission (vide supra); about 30% progress to renal failure, 30% have variable degrees of proteinuria but stable function for many years, and about 10% die of nonrenal causes^27,28. Other factors at presentation that have been associated with a poor prognosis include male gender, older age, high levels of proteinuria, and abnormal renal function; histologic factors include tubular interstitial changes and degree of glomerulosclerosis^{29,30,31,32,33,34}.

The problem in associating these factors with prognosis, however, is their qualitative nature and poor specificity. An alternate approach based on dynamic changes in renal function does produce a semiquantitative risk of progression^35,36. It uses the clinical parameters of proteinuria and creatinine clearance estimates over fixed periods of time. In its simplest form, this approach demonstrated that the overall accuracy of predicting progression to chronic renal failure when proteinuria values over a 6-month period were persistently greater than 4 g/day was 71%, 6 g/day was 79%, and 8 g/day was 84% Table 1. If the patient's renal function was impaired at the beginning of these time frames and/or significantly deteriorated over the 6 months of observation, accuracy was even higher and sensitivity was substantially greater, as demonstrated by the percentage value under the R value in Table 1. The advantages of the algorithm are its reliance on very few factors, all of which are standard laboratory measurements of renal function, and its dynamic nature, which allows recalculating the risk of progression over the course of the patient's disease. Age, gender, degree of sclerosis on biopsy, and presence of hypertension are important factors in univariate analyses but are not independent of the factors in the model and hence do not add to the predictive value of the algorithm.

Table 1 - Test characteristics (%) of predicting chronic renal insufficiency in Canadian patients.

Full table

The effects of idiopathic MGN on nonrenal- and renal-related morbidity and mortality are more difficult to determine from the literature. Reviews from Finland and the United States show that a similar high proportion of deaths, between 30% and 60%, are nonrenal-related^25,28. Whether their cause is due to comorbid conditions, the effects of the disease, or the treatment is difficult to discern. But details from the reports reveal that most of the deaths are premature (mean age 51) and due either to cardiovascular disease or to cancer³⁷.

Treatment

I think it will be useful if we consider therapeutics in four categories. The first, specific immunosuppressive therapy, is aimed at modulating the immune component of the disease. The second, nonspecific therapy, focuses on reducing proteinuria and secondarily slowing disease progression. The third comprises treatment of the secondary effects of the disease, and the fourth is treatment aimed at reducing the complications of the immune-modulating drugs.

Specific immunosuppressive treatment

The observations on the natural history of idiopathic MGN and our ability to predict outcome should perhaps be the background upon which current immunosuppressive protocols are assessed. One approach would be to first establish risk of progression using the results of our algorithm. We could examine the six-month period around the entry point of the major studies, segregate the patients into risk categories, and then discuss the trial results in the context of prognostic grouping. Although somewhat imprecise given the lack of details provided in the reports, this approach would allow us to assess specific therapies from the point of view of risk as well as benefit, the overriding concern both of patients and physicians when considering treatment options.

Let us define the categories for risk of progression. Low-risk patients have normal serum creatinine and creatinine clearance values and a peak proteinuria <4 g/day over 6 months of observation. Medium-risk patients have normal or nearly normal serum creatinine and creatinine clearance values and proteinuria 4 g/day but <8 g/day over 6 months of observation. Finally, high-risk patients have either abnormal or deteriorating serum creatinine and creatinine clearance values and/or persistent proteinuria 8 g/day over 6 months of observation.

The long-term prognosis for low-risk patients is excellent. Our validation study examined more than 300 patients from three distinct geographic regions who were followed for more than 5 years. We found that only 5% of patients with these criteria went on to develop chronic renal insufficiency³⁶. Blood pressure control and strategies for reduction of urinary protein excretion should be implemented through the use of agents such as the angiotensin-converting-enzyme inhibitors. Immunosuppressive drugs are not recommended and, in the majority of cases, treatment of the secondary effects of the disease is not necessary. Because a small percentage of patients do progress, monitoring of their renal function, proteinuria, and blood pressure should be continued and assessment of their risk of progression periodically recalculated.

As one would expect, the individuals classified as medium-risk patients did not fare as well as did the low-risk patients. The entry criteria for patients in the two randomized trials that studied the effects of corticosteroid therapy alone in idiopathic MGN would place these patients in the medium risk category^38,39. Prednisone alone was ineffective in inducing and sustaining a reduction in proteinuria in these trials. The follow-up periods were less than four years, and the dose and duration varied, but the evidence does not support the use of corticosteroids as a single agent in the management of MGN patients in this risk category. Ponticelli and coworkers in Italy have found benefit when corticosteroids are combined with a cytotoxic agent. They observed a significant increase both in remissions in proteinuria and renal survival^40,41. Therapy given for six months consisted of 1 g of intravenous methylprednisolone on the first 3 days of months 1, 3, and 5, followed by 27 days of oral methylprednisolone (0.5 mg/kg/day) alternating in months 2, 4, and 6 with chlorambucil (0.2 mg/kg/day). After 10 years of follow-up, 8% of treated patients versus 40% of untreated patients had reached end-stage renal disease (P = 0.004). Of the treated group, 43% of the time versus 78% of the placebo group's time was spent in a nephrotic state (P = 0.0001). In a second trial, these authors used the same 6-month routine but substituted cyclophosphamide for chlorambucil and found a benefit similar to that in their first trial⁴². However, in this study, approximately 30% of patients in both groups relapsed within 30 months of stopping treatment. Both regimens appeared to be quite safe; only 10% of patients discontinued treatment because of adverse effects.

We recently completed a six-month randomized trial comparing cyclosporine with placebo in 51 patients with idiopathic MGN (abstract; Cattran et al for the North American Nephrotic Syndrome Study Group, J Am Soc Nephrol 9:85A, 1999). These patients were in the medium-risk category for progression, as they had 4 to 8 g/day proteinuria over a six-month observation period despite control of hypertension and a restricted-protein diet of 0.8 g/kg/day. The mean cyclosporine dose was 3.7 2.0 mg/kg. Seventy-five percent of the treated group versus 22% of the control group entered a remission (P = 0.003). This response was maintained for as long as one year after test medication in 48% of the patients receiving cyclosporine versus only 13% of the patients given a placebo (P = 0.007). Renal insufficiency occurred in two patients in each group, but the overall rate of renal functional deterioration, defined by the slope of creatinine clearance, was flat and the same in both patient populations over the active treatment and follow-up periods. Cyclosporine was well tolerated, and no one had to discontinue treatment because of adverse effects.

The percentage of high-risk patients with idiopathic MGN is small, and very few trials have selected these patients for study. In the one randomized trial of corticosteroids alone in patients with high-grade proteinuria (mean, 10.6 g/day but normal renal function), prednisolone conferred no benefit on either the progression rate or on proteinuria⁴³. In a subgroup analysis of patients with initial renal insufficiency in one of the other trials using corticosteroids alone, no difference was seen in the rate of deterioration over four years of follow-up³⁹. In a small uncontrolled study of 15 patients with idiopathic MGN and a progressive decline in renal function, treatment comprised pulse methylprednisolone for five days plus an oral prednisone taper⁴⁴. Renal function initially stabilized in 9 of the patients treated, but at last follow-up, 2 had died and 5 had gone on to end-stage renal disease. These results suggest that the benefits were transient. Three groups have used chorambucil in modified versions of the Italian regimen in high-risk patients^45,46,47,48. In the total of 34 patients in these studies, approximately 50% had a sustained improvement in function, but the rate of adverse effects was high, even with an appropriate dose adjustment of the chlorambucil. Whether the therapy was given too late or is less effective in the most severe cases of idiopathic MGN cannot be answered without a controlled trial in this select population. Pulse cyclophosphamide plus prednisone compared to prednisone alone produced no benefit in a recent randomized trial that documented progressive renal insufficiency prior to treatment (mean entry creatinine level, 2.5 mg/dL)⁴⁹. Long-term oral cyclophosphamide with or without prednisone has been used in two small, nonrandomized, case-controlled studies in high-risk patients and although both studies showed a benefit, study design limitations plus the risks of prolonged cytotoxic therapy such as infertility, infection, and malignancy are significant and have limited this approach^50,51.

In our randomized controlled trial of patients with declining function (mean creatinine, 2.2 mg/dL at entry), one year of cyclosporine (mean dose, 3.8 mg/kg) significantly reduced proteinuria (P = 0.02) and slowed the rate of disease progression (P < 0.02); both of these results were sustained in more than 50% of patients for as long as two years⁵². However, the number of patients was small, and the drug is expensive and nephrotoxic, so careful patient selection and ongoing monitoring for adverse events must be part of this approach. An earlier uncontrolled study in 15 patients with high-grade proteinuria resistant to corticosteroid therapy suggested a similar benefit with cyclosporine, but the relapse rate was high, approximately 30%⁵³. The dose and long duration of cyclosporine therapy is supported by a recent report from the German collaborative group. In a review of 41 high-risk idiopathic MGN patients treated with cyclosporine, 14 (34%) achieved a complete remission. The mean cyclosporine dose was 3.3 mg/kg, and the time to complete remission averaged 225 days (quartiles 120–459). The patients were considered to be at high risk because 70% had been resistant to other immunosuppressive agents, and their proteinuria averaged 11 g/d prior to treatment⁵⁴.

Nonimmunologic treatment

Dietary protein restriction has never been associated with complete remission of the nephrotic syndrome in idiopathic MGN. However, prospective protein-restriction studies and two meta-analyses have shown a reduction in both proteinuria and progression rate as a function of initial urinary protein level^55,56,57,58. Most patients with idiopathic MGN have heavy proteinuria, so it seems rational for us to modestly restrict their protein intake (0.8 g/kg/day), provided we continue to monitor the patient's protein and energy intake and nutritional status⁵⁹.

Treatment of hypertension also improves proteinuria and slows progression of renal failure. In several studies of diabetic and nondiabetic patients, the ACE inhibitors have shown a greater benefit than that accounted for by their antihypertensive effects alone^57,60. This additional benefit also has been demonstrated in patients with idiopathic MGN, but these studies have been small and with limited follow-up^61,62. The mechanism for ACE inhibitors' renoprotective effect is not fully understood but in part might be related to their lowering of proteinuria and the subsequent reduction in interstitial scarring from the decrease in tubular trafficking of protein moieties^63,64.

Treatment of the secondary effects of the disease

Few data have been published on the morbidity and mortality associated with the hyperlipidemia and thromboembolism that occur in patients with persistent high-grade proteinuria, impaired renal function, or both. However, the burden of cardiovascular disease seen in end-stage renal disease programs is substantial; it is by far the most common cause of death in both the dialysis and transplant populations^65,66,67. A variety of abnormalities in the lipid profile have been reported with the nephrotic syndrome, but the most frequent pattern has been an increase in total cholesterol with normal or low levels of high density lipoprotein (HDL) and increased low density lipoprotein (LDL). It is likely that the dyslipidemia plays a central role in the increased cardiovascular risk and perhaps the thromboembolic disease in these patients^68,69. The lipid metabolism regulators of the hydroxy methylglutaryl co-enzyme A reductase class (HMG-CoA) have been effective in the management of hyperlipidemia in patients with persistent nephrotic-range proteinuria⁷⁰. Although no randomized trials have been done or are likely to be carried out in this specific group, many nephrologists have applied this information and the safety and efficacy studies from other populations to justify use of these drugs in patients with idiopathic MGN⁷¹.

Studies have shown variation in the risk of thromboembolic disease in idiopathic MGN. Some studies have reported an incidence of renal vein thrombosis and thromboembolic phenomena as high as 50% in patients with severe membranous nephropathy; other investigators have quoted an incidence as low as 20%⁷². Part of this wide range is probably related to the rigor of the screening process and to the detection methods employed. The most recent report, published in 1993, found deep venous thrombosis in 11%, pulmonary emboli in another 11%, and renal vein thrombosis in 35% of their patients with idiopathic MGN⁷³. The death rate from thromboembolism in their total group was 14% but was 42% in high-risk patients over a follow-up of only 4.5 years; these data emphasize the magnitude of this issue⁷⁴. In vitro and in vivo studies have provided evidence of multiple factors that promote a hypercoagulable state in patients with the nephrotic syndrome⁷⁵. These range from platelet abnormalities to hyperfibrinogenemia. The highest risk of thromboembolism occurs in patients whose albumin levels are below 2.5 g/L, but this apparently easy screening test for identifying which patients to treat with anticoagulation is countered by the concern that patients with hypoalbuminemia have a greater propensity to bleed. It would be ideal if we could predict those who will experience thromboembolic problems, but this ability has remained elusive even with the advent of screening for common genetic risk factors for hypercoagulability⁷⁶. Data on complications of oral anticoagulant therapy in non-nephrotic populations have recently been better defined. A 14% annual risk of bleeding exists in patients with cardiovascular disease who take anticoagulants, but only 2% of these events are serious or life-threatening, and the fatality rate is 0.5%⁷⁷. In a decision analysis, Bellomo and Atkins weighed the risks of prophylactic anticoagulation versus the current approach of waiting for a symptomatic thromboembolic episode. They calculated an overall reduction in morbid events of 32% with prophylactic anticoagulation⁷⁴, but they admit that their analysis was flawed by the lack of information in the literature about the point estimates necessary for calculating event rates and the inability to balance the morbidity associated with a hemorrhagic versus a thrombotic complication. The majority of nephrologists today use oral anticoagulant therapy as primary prevention only in high-risk patients or not at all, and they reserve its general use until after occurrence of an event. The correctness of this approach remains controversial; further study is required.

Prophylaxis

Bone loss is related to both the dosage and duration of corticosteroid therapy and is greatest in the first three to six months of treatment. Well-designed trials in patients exposed to long-term corticosteroid therapy have shown that etidronate therapy prevents both lumbar and trochanteric bone loss⁷⁸. A 1997 study reported a 40% reduction in new vertebral fractures in women, especially in the postmenopausal group, and very little in the way of adverse effects of the medication. Etidronate administration should be considered in the patient with idiopathic MGN when prolonged courses of corticosteroid therapy are utilized, provided there is ongoing monitoring in patients at risk for renal stones and/or renal insufficiency. Similarly, trimethoprim-sulfamethoxazole (TMP-SMX) significantly reduces the incidence of Pneumocystis carinii pneumonia in patients undergoing prolonged immunosuppressive therapy both for solid organ transplants and certain autoimmune disorders⁷⁹. Use of TMP-SMX is recommended when patients with idiopathic MGN are exposed to such therapy.

Conclusions

Management of the patient with MGN has changed over the past two decades. An algorithm that summarizes many of the treatment approaches is outlined in Figure 2. The following general rules are also suggested:

Figure 2.

Guideline for the treatment of idiopathic membranous nephropathy. Patients can change from one category to another during the course of follow-up. Abbreviation is: ACEI, angiotensin-converting-enzyme inhibiting drug. *Supported by evidence from a controlled trial. **Introduction of risk reduction strategies for both secondary effects of disease and adverse effects of immunotherapy.

Full figure and legend (41K)

(1) Establish whether MGN is primary (idiopathic) or secondary. If the latter, identify the inciting agent and take appropriate actions for its removal or treatment.

(2) In the majority of patients who have normal renal function at presentation, renal function should be monitored over a six-month period while the patient is receiving conservative therapy. A risk of progression score should be established. If during this observation period high-grade proteinuria persists and/or renal function deteriorates, treatment for the secondary effects of the disease such as a lipid-lowering agent and possibly an anticoagulant should be introduced.

(3) In patients with medium risk of progression, the best evidence-based medicine supports the monthly cycling routine of cyclophosphamide or chlorambucil alternating with prednisone for six months.

(4) In patients with high risk of progression, cyclosporine for 6 to 12 months, or the cyclophosphamide or chlorambucil/prednisone regimen with appropriate drug dose adjustments should be implemented.

QUESTIONS AND ANSWERS

DR. JOHN T. HARRINGTON (Dean, Tufts University School of Medicine, Boston, Massachusetts): In the study chaired by Coggins³⁸, the criticism has been that there was "something wrong" with the control group. I was a participant in the direction of that study. Since we could not identify anything different in our control group as compared with other membranous control groups at time zero, might you have an explanation?

DR. CATTRAN: I do not have the answer, but it might be worthwhile to look specifically at the levels of proteinuria and their persistence in light of what we now know is one of the best indicators of prognosis. It may be that the control population had a higher and more persistent level of proteinuria compared with the treatment group and that this increase eventually produced the poor outcome. This difference could have been missed in the analysis, as only qualitative proteinuria was reported, not actual proteinuria levels over time. The magnitude of the decline in renal function in the control patients would still be unusual, however, that is, a 10% annual decline in GFR and 25% of patients doubling their initial serum creatinine level in only 15 months. This finding is not in keeping with other reports of the natural history of the disease. Remember that even the best randomized controlled study has a chance of imbalance at the time of randomization.

DR. HARRINGTON: An earlier observational study in Boston involved only nephrotic patients, including those with membranous GN. There was a clear difference in the response rate of those who were treated within six months of diagnosis of their disease compared with those treated after six months⁸⁰. No one in the latter group had a response. Did you look at that aspect of treatment?

DR. CATTRAN: This question raises several issues. I'll restrict my comments to membranous GN. The first issue is, how accurate are we at estimating when the disease started? Clinically, I think not very good. I would guess this extends over weeks to months as the patient proceeds through a series of physicians before seeing a nephrologist. Then the patient has to have a renal biopsy and histologic confirmation. Even this process assumes that following disease onset, symptomatic edema quickly occurs. All these are known to be quite variable. If not by clinical features alone, can we tell the duration of the disease from the biopsy? In most studies the relationship between the staging of the disease by electron microscopy, the time of onset of the process, and the start of treatment is lacking. Since we cannot precisely estimate onset by symptoms or pathology, it is unlikely that we can determine from either our study or the literature whether the timing of the treatment is crucial. Further complicating the relationship between onset, stage, and treatment response is the natural history of membranous GN. As many as 30% of patients will have a spontaneous remission, often in the first year or two, and few have biopsies post remission to indicate what has happened at the glomerular level. This spontaneous remission rate will confound the interpretation of any therapeutic study that starts early, especially if there is no control population. Based on this, I think that the increased accuracy gained in assessing prognosis by using an observation period might outweigh the theoretical advantages of early treatment.

DR. GERRY ARBUS (Pediatric Nephrologist, The Hospital for Sick Children, Toronto): We don't see a lot of idiopathic membranous GN in children. Is your patient cured because he's now clear of protein, or does he have a chance of relapsing? If he does progress to dialysis or transplantation, what are his chances of having a recurrence of the membranous disease?

DR. CATTRAN: About 30% of membranous GN cases will relapse subsequent to a complete remission. The great majority who do, however, will have stable long-term renal function and will relapse only to subnephrotic-range proteinuria. In our recent report of 82 patients in complete remission, only 9 developed mild renal insufficiency 5 to 25 years after disease onset, and none progressed to ESRD. Dr. Ponticelli reported the only other large group followed after a complete remission and found a similar relapse rate and stable long-term function^25,26. So relapses do occur in a significant percentage of patients, but the relapses tend to be moderate in regard to proteinuria and rarely progress to renal insufficiency.

Our patient stopped cyclosporine treatment 4 years ago. Although he is less likely to relapse as time goes on, we have seen it occur as long as 12 years after a complete remission. The recurrence of membranous nephropathy after transplantation is in the range of 5% to 10%, and I'm not aware of any factors that predict recurrence. Patients who do develop end-stage renal disease obviously have the worse variant of the disease, and it would not be unexpected that susceptibility to recurrence resides within the genetic make-up of the host. Currently this is pure speculation.

DR. RAMAD PRASAD (Nephrology Fellow, Toronto General Division of the University Health Network, Toronto): Is there a role for mycophenolate mofetil (MMF) in the treatment of idiopathic membranous nephropathy?

DR. CATTRAN: In studies looking at the effect of MMF on the Heymann model of membranous GN, the drug interfered with both the cell- and humoral-mediated components, although it had no effect if started after the disease was established. The results in human membranous GN are mostly sporadic. Dr. Appel's group recently reported a 50% reduction in proteinuria in 6 of 16 patients (40%) after 6 months of MMF treatment⁸¹. Almost all these patients had been resistant to other drug therapies. The jury is still out, but I think a controlled trial is worth considering. It may be that treatment of membranous nephropathy will become more like cancer chemotherapy, that is, after selection of the high-risk group, pulse treatment will be followed by a combination of agents but in lower doses than our current single-drug approach and perhaps for a longer duration. An indication of the need for a new approach is suggested by the literature that reports a relapse rate of 30% to 40% within 2 years of discontinuing the medication not only with cyclosporine treatment, but also with chlorambucil-prednisone and with cyclophosphamide-prednisone⁴².

DR. SANJEEV GULATI (Nephrology Fellow, The Hospital for Sick Children): Is the natural history of idiopathic membranous nephropathy in children any different from that in adults? How would you recommend managing the child with this disorder?

DR. CATTRAN: Idiopathic membranous GN is rare in children. As far as I'm aware, there are no published natural history studies in children and none on specific treatment in this age group. If you do see such a child, be careful to rule out secondary causes. Certainly lupus nephritis, including the membranous variant, can occur in this age group and can precede the development of systemic symptoms and/or laboratory findings of this disease.

DR. MARC GOLDSTEIN (Staff Nephrologist, St. Michael's Hospital, Toronto): I'd like to return to the question of early treatment. I agree with you that if you're going to treat with a potentially dangerous therapy, it is important to identify patients who are at risk of progression. In the group we treated with long-term oral cyclophosphamide, we treated late and still had a high response rate. These patients already had renal functional impairment, and their disease had been present for several years. I'm not convinced that early treatment is justifiable because of the risks of therapy.

DR. CATTRAN: I agree this is an important question and should perhaps be addressed in a formal study.

DR. JUDITH MILLER (Staff Nephrologist, Toronto General Division of the University Health Network): Could you expand on the impact of gender on the progression rate in membranous nephropathy and the mechanisms that might be involved?

DR. CATTRAN: A better prognosis in females has been suggested in most of the primary glomerular diseases for many years, but we still haven't figured out the precise mechanism. Your own work indicates that there are gender differences both in normal renal hemodynamics and in the kidney's response to blockade of the renin-angiotensin system⁸². Other experimental work supports an interaction between the renin-angiotensin system and intraglomerular pressure that is at least partially estrogen dependent⁸³. Also, evidence from animal data suggests that transforming growth factor beta significantly influences glomerular matrix production, and matrix production in turn might be modified by the presence of estrogen⁸⁴. Thus, gender and progression might be linked through both the hemodynamic and the cell response systems. How this occurs in women and whether it is limited to their fertile period remains to be determined. A further gender difference is suggested by a recent summary of familial cases of membranous GN. The 25 cases reported included only one female⁸⁵. In another recent case report supporting the importance of gender, two siblings developed membranous GN, but the male went on to renal failure and the female had a complete remission¹⁷. Once we determine the mechanism of the gender effect, perhaps we will be able to intervene by modifying the hormonal factor(s). A whole new avenue of therapeutics could potentially be opened.

DR. MURLI KRISHNAN (Nephrology Fellow, Toronto General Division of the University Health Network): Do any histologic findings predict a poor prognosis, and can we use them to guide therapy?

DR. CATTRAN: Yes. They include the degree of interstitial fibrosis, glomerular sclerosis, and the severity of the vascular changes. However, in most multivariate analyses, these findings are not independent risk factors. In most studies on prediction, the serum creatinine and the severity of the proteinuria over time dominate the algorithm. So although histology is associated with prognosis, it does not add anything to the risk profile providing you have the laboratory data. In the final analysis, it doesn't matter whether you're male or female or how the biopsy looks, it's the degree of proteinuria, how long it's been at that level, and the serum creatinine value that count.

DR. SANJAY MAITRA (Nephrology Fellow, Toronto General Division of the University Health Network): The high-risk patients have a significant incidence of thromboembolism. Do you recommend regular monitoring for these episodes in these patients?

DR. CATTRAN: We probably should be more vigilant, especially in patients who have persistent high-grade proteinuria and/or severe hypoalbuminemia. A recent retrospective review from Australia found a prevalence of thromboembolic events approaching 50% and a death rate that was much higher in this group⁷⁴. Therefore we should carefully monitor for such episodes and, if detected, treat them with long-term anticoagulants. Whether all patients with membranous GN should be treated prophylactically is more controversial, but most practicing nephrologists don't initiate anticoagulant therapy until after the first event.

DR. HARRINGTON: In our observational study many years ago, almost no one whose 24-hour urine protein excretion was less than 2 g progressed to renal failure⁸⁰. Have you looked at your <4 g group to see what happened to those with 0 to 2 g/day?

DR. CATTRAN: No, we have not looked specifically at this level of proteinuria, but my guess would be exactly what you suggest. It would certainly fit with the current data which indicate that the level of proteinuria is an important predictive factor. There are also accumulating data that above a certain grade, the proteinuria per se is nephrotoxic to the tubulointerstitial compartment. In our algorithm, both the proteinuria level and its persistence are used to semiquantitate the likelihood of progression. The algorithm can be used by the physician to better assess the risks of treatment versus the benefits. You can often delay or even avoid immunosuppressive drugs in many patients who are at low risk of progression by its application.

DR. HARRINGTON: Can you tell us a bit more about megalin? What is it and what does it do? Should I worry about it?

DR. CATTRAN: Megalin is a normal component of the podocyte identified in the Heymann model of experimental membranous GN. It is related to the low-density lipoprotein family and functions as a multiligand receptor for the uptake of a variety of macromolecules. A second protein, known as RAP (receptor-associated protein), is closely related to megalin and is an additional target antigen in the model. Kerjaschki and others have worked out the relationship between megalin and the receptor-protein molecule and found that the formation of the complex between these antigens and their antibodies initiates the disease process but does not proceed to structural changes and proteinuria until complement binding occurs⁸. Currently, the human equivalent of this sequence is limited to finding the C5b-9 complex or MAC in the urine of patients with membranous nephropathy. The experimental model mimics human disease in regard to the pathology, but the identification of the equivalent of megalin or RAP has not been determined. In summary, you don't have to worry about it.

DR. YORK PEI: (Staff Nephrologist, Toronto General Division of the University Health Network): Idiopathic membranous glomerulonephritis is probably a heterogenous collection of different diseases similar to what has been found in type 1 and type 2 diabetes. One of the key approaches to studying the human problem is to examine the familial form of the disease, as seen in the recent progress in other related glomerular disorders like focal glomerulosclerosis. Do you know of any such study in idiopathic membranous GN? Do such families exist?

DR. CATTRAN: The families exist, and some have been reported, as I mentioned previously⁸⁵. As far as I'm aware, however, there's no collaborative network currently in place specifically looking at the familial form of the disease.

DR. GOLDSTEIN: Is it correct that relapses are much more common after a complete remission from therapy than after a spontaneous remission, and do we understand what a relapse means at the glomerular level?

DR. CATTRAN: The answer to the first question depends on which study you review. In the randomized trials using corticosteroids, there is a trend toward a higher relapse rate in the treated group in all three studies^38,39,43, but the numbers are small and follow-up is quite short. In the long-term chlorambucil-prednisone trial from Italy, there is also only a trend but in the opposite direction, that is, the relapse rate was 11 of 24 (46%) in the treated group and 4 of 7 (57%) in the control group⁴⁰. The relapse rate is an important issue, but at least in the natural history studies, most relapses are associated with a lower grade of proteinuria and rarely associated with deteriorating function. The sequence of events at the glomerular level is not clear. Histologic studies focusing on relapse patients are rare. In most cases reported, new stage 1 deposits on electron microscopy have been found. Presumably a fresh wave of whatever prompted the original immune complexes has recurred. In some cases, however, the lesions look like they've persisted despite the complete clinical and laboratory remission. Törnoth, Honkanen, and Pettersson have suggested that the large deposits that extend deeply into the GBM but whose surface is still exposed to the subepithelial space represent this type of clinical pattern⁸⁶. This would allow for ongoing activation of complement and the subsequent chemokine release essential for the damage indicated by the return of proteinuria. It may be that both scenarios exist in the human condition. It is highly unlikely that a single factor is going to explain all the membranous GN variants, but the cases that do remit and relapse, as well as the familial ones, should be good targets for studying potential genetic influences.

DR. GULATI: Should older patients with membranous GN be considered for treatment earlier because they are at a higher risk of progression?

DR. CATTRAN: I don't think so. We have demonstrated that the rate of deterioration in the elderly is the same as in the younger patient⁸⁷. Baseline serum creatinine also did not differ, but correcting creatinine clearances for age and muscle mass significantly lowered the initial value for GFR in the older population. This means that their renal reserve is less and might suggest that starting treatment earlier would preserve more renal function, but it also means that their risk of complications from therapy are likely to be higher. Perhaps we should use the same care in selecting patients at risk of progression as younger patients, and remember that if therapy is to be used, we must make the appropriate dose adjustment given the patient's age and impaired clearance.

References

References

1.	Haas M. Changing etiologies of unexplained adult nephrotic syndrome: A comparison of renal biopsy findings from 1976-1979 and 1995-1997. Am J Kidney Dis 1997; 30: 621–631. | PubMed | ISI | ChemPort |
2.	US RENAL DATA SYSTEM. USRDS 1999 Annual Data Report 1999; The National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Disease Bethesda.
3.	Maisonneuve P, Agodoa L & Gellert R et al. Distribution of primary renal diseases leading to end-stage renal failure in the United States, Europe and Australia/New Zealand: Results from an international comparative study. Am J Kidney Dis 2000; 35: 157–165. | PubMed | ISI | ChemPort |
4.	Rabenantoandro R, Rakotondrajao R, Rakotondranaivo S & Rasamindrakotroka AJ. Renal glomerular lesions and Plasmodium falciparum infection. Am J Kidney Dis 1987; 10: 52–55.
5.	Goldman M & Lambert PH. Glomerulopathies associated with parasitoses. Nephrologie 1985; 6: 49–51. | PubMed | ISI | ChemPort |
6.	Kar NL, Lai FM & Chan KW et al. The clinical pathological features of hepatitis B virus associated glomerulonephritis. Q J Med 1987; 63: 323–333. | PubMed |
7.	Burstein DM, Korbet SM & Schwartz MM. Membranous glomerulonephritis and malignancy. Am J Kidney Dis 1993; 22: 5–10. | PubMed | ISI | ChemPort |
8.	Kerjaschki D & Neale TJ. Molecular mechanisms of glomerular injury in rat experimental membranous nephropathy (Heymann nephritis). Am J Kidney Dis 1996; 7: 2518–2525. | ChemPort |
9.	Shankland SJ. New insights into pathogenesis of membranous nephropathy. Kidney Int 2000; 57: 1204–1205 10.1046/j.1523-1755.2000.00950.x. | Article | PubMed | ISI | ChemPort |
10.	Penny MJ, Boyd RA & Hall BM. Mycophenolate mofetil prevents the induction of active Heymann nephritis: Association with Th2 cytokine inhibition. J Am Soc Nephrol 1998; 9: 2272–2282. | PubMed | ISI | ChemPort |
11.	Couser WG, Schultz M & Pruchno CJ. Role of C5b-9 in experimental membranous nephropathy. Nephrol Dial Transplant 1992; 1: 25–31.
12.	Kon SP, Coupes B & Short CD et al. Urinary C5b-9 excretion and clinical course in idiopathic human membranous nephropathy. Kidney Int 1995; 48: 1953–1958. | PubMed | ISI | ChemPort |
13.	Shankland SJ, Pippin J & Pichler RH et al. Differential expression of transforming growth factor- isoforms and receptors in experimental membranous nephropathy. Kidney Int 1996; 50: 116–124. | PubMed | ISI | ChemPort |
14.	Shankland SJ, Floege J & Thomas SE et al. Cyclin kinase inhibitors are increased during experimental membranous nephropathy: Potential role in limiting glomerular epithelial cell proliferation in vivo. Kidney Int 1997; 52: 404–413. | PubMed | ISI | ChemPort |
15.	Cybulsky AV, Takano T, Papillon J & McTavish AJ. Complement-induced phospholipase A2 activation in experimental membranous nephropathy. Kidney Int 2000; 57: 1052–1062 10.1046/j.1523-1755.2000.00932.x. | Article | PubMed | ISI | ChemPort |
16.	Hiki Y, Kobayashi Y, Itoh I & Kashiwagi N. Strong association of HLA-DR2 and MT1 with idiopathic membranous nephropathy in Japan. Kidney Int 1984; 25: 953–967. | PubMed | ISI | ChemPort |
17.	Guella A, Akhtar M & Ronco P. Idiopathic membranous nephropathy in identical twins. Am J Kidney Dis 1997; 29: 115–118. | PubMed | ISI | ChemPort |
18.	Vaughan RW, Tighe MR & Boki K et al. An analysis of HLA class II gene polymorphism in British and Greek idiopathic membranous nephropathy patients. Eur J Immunogenet 1995; 22: 179–186. | PubMed | ISI | ChemPort |
19.	Cattran DC, Pei Y & Greenwood C. Predicting progression in membranous glomerulonephritis. Nephrol Dial Transplant 1992; 7 Suppl 1: 48–52. | PubMed | ISI |
20.	Hogan SL, Muller KE, Jennette JC & Falk RJ. A review of therapeutic studies of idiopathic membranous glomerulopathy. Am J Kidney Dis 1995; 25: 862–875. | PubMed | ISI | ChemPort |
21.	Marx BE & Marx M. Prognosis of idiopathic membranous nephropathy: A methodologic meta-analysis. Kidney Int 1997; 51: 873–879. | PubMed | ISI | ChemPort |
22.	Simon P, Ramee MP & Autuly V et al. Epidemiology of primary glomerular diseases in a French region: Variations according to period and age. Kidney Int 1994; 46: 1192–1198. | PubMed | ISI | ChemPort |
23.	Abe S, Amagasaki SY & Kato E et al. Idiopathic membranous glomerulonephritis: Aspects of geographical differences. J Clin Pathol 1986; 39: 1193–1198. | PubMed | ISI | ChemPort |
24.	Harrison DJ, Thomson D & Macdonald MK. Membranous glomerulonephritis. J Clin Pathol 1986; 39: 167–171. | PubMed | ISI | ChemPort |
25.	Laluck BJ, Jr & Cattran DC. Prognosis after a complete remission in adult patients with idiopathic membranous nephropathy. Am J Kidney Dis 1999; 33: 1026–1032. | PubMed | ISI |
26.	Ponticelli C, Passerini P & Altieri P et al. Remissions and relapses in idiopathic membranous nephropathy. Nephrol Dial Transplant 1992; 7 Suppl 1: 85–90. | PubMed | ISI |
27.	Honkanen E, Tornroth T & Gronhagen-Riska C. Natural history, clinical course and morphological evolution of membranous nephropathy. Nephrol Dial Transplant 1992; 7 Suppl 1: 35–41. | PubMed | ISI |
28.	Donadio JV, Jr, Torres VE & Velosa JA et al. Idiopathic membranous nephropathy: The natural history of untreated patients. Kidney Int 1988; 33: 708–715. | PubMed | ISI |
29.	Tu WH, Petitti DB & Biava CG et al. Membranous nephropathy: Predictors of terminal renal failure. Nephron 1984; 36: 118–124. | PubMed | ISI | ChemPort |
30.	Davison AM, Cameron JS & Kerr DN et al. The natural history of renal function in untreated idiopathic membranous glomerulonephritis in adults. Clin Nephrol 1984; 22: 61–67. | PubMed | ISI | ChemPort |
31.	Zucchelli P, Ponticelli C, Cagnoli L & Passerini P. Long-term outcome of idiopathic membranous nephropathy with nephrotic syndrome. Nephrol Dial Transplant 1987; 2: 73–78. | PubMed | ChemPort |
32.	Murphy BF, Fairley KF & Kincaid-Smith PS. Idiopathic membranous glomerulonephritis: Long-term follow-up in 139 cases. Clin Nephrol 1988; 30: 175–181. | PubMed | ISI | ChemPort |
33.	Schieppati A, Mosconi L & Perna A et al. Prognosis of untreated patients with idiopathic membranous nephropathy. N Engl J Med 1993; 329: 85–89. | Article | PubMed | ISI | ChemPort |
34.	Magil AB. Tubulointerstitial lesions in human membranous glomerulonephritis: Relationship to proteinuria. Am J Kidney Dis 1995; 25: 375–379. | PubMed | ISI | ChemPort |
35.	Pei Y, Cattran D & Greenwood C. Predicting chronic renal insufficiency in idiopathic membranous glomerulonephritis. Kidney Int 1992; 42: 960–966. | PubMed | ISI | ChemPort |
36.	Cattran DC, Pei Y & Greenwood CM et al. Validation of a predictive model of idiopathic membranous nephropathy: Its clinical and research implications. Kidney Int 1997; 51: 901–907. | PubMed | ISI | ChemPort |
37.	Honkanen E. Survival in idiopathic membranous glomerulonephritis. Clin Nephrol 1986; 25: 122–128. | PubMed | ISI | ChemPort |
38.	A controlled study of short-term prednisone treatment in adults with membranous nephropathy. Collaborative Study of the Adult Idiopathic Nephrotic Syndrome. N Engl J Med 1979; 301: 1301–1306. | ISI |
39.	Cattran DC, Delmore T & Roscoe J et al. A randomized controlled trial of prednisone in patients with idiopathic membranous nephropathy. N Engl J Med 1989; 320: 210–215. | PubMed | ISI | ChemPort |
40.	Ponticelli C, Zucchelli P & Imbasciati E et al. Controlled trial of methylprednisolone and chlorambucil in idiopathic membranous nephropathy. N Engl J Med 1984; 310: 946–950. | PubMed | ISI | ChemPort |
41.	Ponticelli C, Zucchelli P & Passerini P et al. A 10-year follow-up of a randomized study with methylprednisolone and chlorambucil in membranous nephropathy. Kidney Int 1995; 48: 1600–1604. | PubMed | ISI | ChemPort |
42.	Ponticelli C, Altieri P & Scolari F et al. A randomized study comparing methylprednisolone plus chlorambucil versus methylprednisolone plus cyclophosphamide in idiopathic membranous nephropathy. J Am Soc Nephrol 1998; 9: 444–450. | PubMed | ISI | ChemPort |
43.	Cameron JS, Healy MJ & Adu D et al. The Medical Research Council trial of short-term high-dose alternate day prednisolone in idiopathic membranous nephropathy with nephrotic syndrome in adults. Q J Med 1990; 74: 133–156. | PubMed | ISI | ChemPort |
44.	Short CD, Solomon LR, Gokal R & Mallick NP. Methylprednisolone in patients with membranous nephropathy and declining renal function. Q J Med 1987; 65: 929–940. | PubMed | ISI | ChemPort |
45.	Mathieson PW, Turner AN & Maidment CG et al. Prednisolone and chlorambucil treatment in idiopathic membranous nephropathy with deteriorating renal function. Lancet 1988; 2: 869–872. | PubMed | ISI | ChemPort |
46.	Warwick GL, Geddes CG & Boulton-Jones JM. Prednisolone and chlorambucil therapy for idiopathic membranous nephropathy with progressive renal failure. Q J Med 1994; 87: 223–229. | PubMed | ISI | ChemPort |
47.	Wetzels JFM, Hoitsma AJ & Koene RAP. Immunosuppression for membranous nephropathy. letterLancet 1989; 1: 211. | PubMed | ISI |
48.	Stirling CM, Simpson K & Boulton-Jones JM. Immunosuppression and outcome in idiopathic membranous nephropathy. Q J Med 1998; 91: 159–164. | ISI | ChemPort |
49.	Falk RJ, Hogan SL, Muller KE & Jennette JC et al. Treatment of progressive membranous glomerulopathy: A randomized trial comparing cyclophosphamide and corticosteroids with corticosteroids alone. Ann Intern Med 1992; 116: 438–445. | PubMed | ISI | ChemPort |
50.	Jindal K, West M, Bear R & Goldstein M. Long-term benefits of therapy with cyclophosphamide and prednisone in patients with membranous glomerulonephritis and impaired renal function. Am J Kidney Dis 1992; 19: 61–67. | PubMed | ISI | ChemPort |
51.	Bruns FJ, Adler S, Fraley DS & Segel DP. Sustained remission of membranous glomerulonephritis after cyclophosphamide and prednisone. Ann Intern Med 1991; 114: 725–730. | PubMed | ISI | ChemPort |
52.	Cattran DC, Greenwood C & Ritchie S et al. A controlled trial of cyclosporine in patients with progressive membranous nephropathy. Kidney Int 1995; 47: 1130–1135. | PubMed | ISI | ChemPort |
53.	Rostoker G, Belghiti D & Maadi AB et al. Long-term cyclosporine A therapy for severe idiopathic membranous nephropathy. Nephron 1993; 63: 335–341. | PubMed | ISI | ChemPort |
54.	Fritsche L, Budde K & Farber L et al. Treatment of membranous glomerulopathy with cyclosporine A: How much patience is required? Nephrol Dial Transplant 1999; 14: 1036–1037 10.1093/ndt/14.4.1036. | PubMed | ISI | ChemPort |
55.	Amico GD, Gentile MG & Fellin G et al. Effect of dietary protein restriction on the progression of renal failure: A prospective randomized trial. Nephrol Dial Transplant 1994; 9: 1590–1594. | PubMed | ChemPort |
56.	Gansevoort RT, de Zeeuw D & de Jong PR. Additive antiproteinuric effect of ACE inhibition and low-protein diet in human renal disease. Nephrol Dial Transplant 1995; 10: 497–504. | PubMed | ISI | ChemPort |
57.	Peterson JC, Adler S & Burkart JM et al. Blood pressure control, proteinuria, and the progression of renal disease. Ann Intern Med 1995; 123: 754–762. | PubMed | ISI | ChemPort |
58.	Pedrini MT, Levey AS & Lau J et al. The effect of dietary protein restriction on the progression of diabetic and nondiabetic renal disease: A meta-analysis. Ann Intern Med 1996; 124: 627–632. | PubMed | ISI | ChemPort |
59.	Kopple JD, Levey AS & Greene T et al. Effect of dietary protein restriction on nutritional status in the Modification of Diet in Renal Disease study. Kidney Int 1997; 52: 778–791. | PubMed | ISI | ChemPort |
60.	Gansevoort RT, Sluiter WJ & Hemmelder MH et al. Antiproteinuric effect of blood-pressure-lowering agents: A meta-analysis of comparative trials. Nephrol Dial Transplant 1995; 10: 1963–1974. | PubMed | ISI | ChemPort |
61.	Thomas DM, Hillis AN & Coles GA et al. Enalapril can treat the proteinuria of membranous glomerulonephritis without detriment to systemic or renal hemodynamics. Am J Kidney Dis 1991; 18: 38–43. | PubMed | ISI | ChemPort |
62.	Rostoker G, Ben Maadi A & Remy P et al. Low-dose angiotensin-converting-enzyme inhibitor captopril to reduce proteinuria in adult idiopathic membranous nephropathy: A prospective study of long-term treatment. Nephrol Dial Transplant 1995; 10: 25–29. | PubMed | ISI | ChemPort |
63.	Ruggenenti P, Perna A & Gherardi G et al. Renoprotective properties of ACE-inhibition in non-diabetic nephropathies with non-nephrotic proteinuria. Lancet 1999; 354: 359–364 10.1016/s0140-6736(98)10363-x. | Article | PubMed | ISI | ChemPort |
64.	Abbate M, Benigni A, Bertani T & Remuzzi G. Nephrotoxicity of increased glomerular protein traffic. Nephrol Dial Transplant 1999; 14: 304–312 10.1093/ndt/14.2.304. | PubMed | ISI | ChemPort |
65.	Ojo AO, Hanson JA & Wolfe RA et al. Long-term survival in renal transplant recipients with renal grafts. Kidney Int 2000; 57: 307–313 10.1046/j.1523-1755.2000.00816.x. | Article | PubMed | ISI | ChemPort |
66.	US RENAL DATA SYSTEMS. USRDS 1995 Annual Data Report 1995; The National Institutes of Health National Institute of Diabetes and Digestive and Kidney Diseases Bethesda.
67.	ANZDATA REGISTRY REPORT 1999. Australia and New Zealand Dialysis and Transplant Registry Adelaide, South Australia.
68.	Wheeler DC & Bernard DB. Lipid abnormalities in the nephrotic syndrome: Causes, consequences, and treatment. Am J Kidney Dis 1994; 23: 331–346. | PubMed | ISI | ChemPort |
69.	Ordonez JD, Hiatt RA, Killebrew EJ & Fireman BH. The increased risk of coronary heart disease associated with nephrotic syndrome. Kidney Int 1993; 44: 638–642. | PubMed | ISI | ChemPort |
70.	Spitalewitz S, Porush JG, Cattran DC & Wright N. Treatment of hyperlipidemia in the nephrotic syndrome: The effects of pravastatin therapy. Am J Kidney Dis 1993; 22: 143–150. | PubMed | ISI | ChemPort |
71.	Scandinavian Simvastatin Survival Study. Randomized trial of cholesterol lowering in 4444 patients with coronary heart disease. Lancet 1994; 344: 1383–1389. | PubMed | ISI |
72.	Wagoner RD, Stanson AW, Holley KE & Winter CS. Renal vein thrombosis in idiopathic membranous glomerulopathy and nephrotic syndrome: Incidence and significance. Kidney Int 1983; 23: 368–374. | PubMed | ISI | ChemPort |
73.	Bellomo R, Wood C & Wagner I et al. Idiopathic membranous nephropathy in an Australian population: The incidence of thromboembolism and its impact on the natural history. letterNephron 1993; 63: 240–241. | PubMed | ISI | ChemPort |
74.	Bellomo R & Atkins RC. Membranous nephropathy and thromboembolism: Is prophylactic anticoagulation warranted? editorialNephron 1993; 63: 249–254. | PubMed | ISI | ChemPort |
75.	Rabelink TJ, Zwaginga JJ, Koomans HA & Sixma JJ. Thrombosis and hemostasis in renal disease. Kidney Int 1994; 46: 287–296. | PubMed | ISI | ChemPort |
76.	Fabri D, Belangero VM, Annichino-Bizzacchi JM & Arruda VR. Inherited risk factors for thrombophilia in children with nephrotic syndrome. Eur J Pediatr 1998; 157: 939–942 10.1007/s004310050972. | Article | PubMed | ISI | ChemPort |
77.	Baggio D, Madore F & Lalonde G. Oral anticoagulant therapy for heart disease: Results in actual cardiology practice. Can J Cardiol 2000; 16: 153–161. | PubMed | ISI | ChemPort |
78.	Adachi J, Bensen WG & Brown J et al. Intermittent etidronate therapy to prevent corticosteroid induced osteoporosis. N Engl J Med 1997; 337: 382–387. | Article | PubMed | ISI | ChemPort |
79.	Ognibene FP, Shelhamer JH & Hoffman GS et al. Pneumocystis carinii pneumonia: A major complication of immunosuppressive therapy in patients with Wegener's granulomatosis. Am J Resp Crit Care Med 1995; 151: 795–799. | PubMed | ISI | ChemPort |
80.	Miller RB, Harrington JT & Ramos CP et al. Long-term results of steroid therapy in adults with idiopathic nephrotic syndrome. Am J Med 1969; 46: 919–929. | Article | PubMed | ISI | ChemPort |
81.	Miller G, Zimmerman R, III, Radhakrishan J & Appel G. Use of mycophenolate mofetil in resistant membranous nephropathy. Am J Kidney Dis 2000; 36: 250–256. | PubMed | ISI | ChemPort |
82.	Miller JA, Anacta LA & Cattran DC. Impact of gender on the renal response to angiotensin II. Kidney Int 1999; 55: 278–285 10.1046/j.1523-1755.1999.00260.x. | Article | PubMed | ISI | ChemPort |
83.	Evans JK, Naish PF & Aber GM. Oestrogen-induced changes in renal haemodynamics in the rat: influence of plasma and intrarenal renin. Clin Sci 1986; 71: 613–619. | PubMed | ISI | ChemPort |
84.	Silbiger S, Lei J, Ziyadeh FN & Neugarten J. Estradiol reverses TGF-beta 1-stimulated type IV collagen gene transcription in murine mesangial cells. Am J Physiol 1998; 274: F1113–F1118. | PubMed | ISI | ChemPort |
85.	Maccario M, Segagni S & Efficace E et al. Idiopathic membranous nephropathy in two siblings. Nephron Dial Transplant 1995; 10: 108–110. | ChemPort |
86.	Törnroth T, Honkanen E & Pettersson E. The evolution of membranous glomerulonephritis reconsidered: New insights from a study on relapsing disease. Clin Nephrol 1987; 28: 107–117. | PubMed |
87.	Zent R, Nagai R & Cattran DC. Idiopathic membranous nephropathy in the elderly: A comparative study. Am J Kidney Dis 1997; 29: 200–206. | PubMed | ISI | ChemPort |