Low-dose rituximab is poorly effective in patients with primary membranous nephropathy.

BACKGROUND: The optimal dosing and the efficacy of rituximab for primary membranous nephropathy (PMN) has not been established. This multicentric prospective study evaluates the efficacy and safety of low-dose rituximab (RTX) therapy in patients with PMN in clinical practice. METHODS: Thirty-four consecutive patients with PMN and nephrotic syndrome were included and received RTX (375 mg/m2) once (18 patients) or twice (16 patients). RTX was the first-line therapy for 19 (56%) and the second line for 15 (44%) patients. All patients were followed for 12 months after RTX and 24 for at least 18 months (mean 23.9 ± 18.6 months). RESULTS: At 12 months, 5 patients (14.7%) achieved complete response, 10 (29.4%) partial and 19 (55.8%) no response. Response occurred ∼6 months after RTX. At 24 months, the clinical situation was unchanged: two non-responders achieved partial response and two responders relapsed. Responders had significantly higher baseline GFR and lower anti-PLA2R antibodies compared with non-responders. Outcome was similar between one or two doses of RTX (non-responders 55.5 versus 56%, respectively) and between patients who had received previous therapy versus those receiving RTX as first-line therapy (non-responders 40 versus 68%, respectively). In the 15 patients already treated, the response to RTX was comparable to that of previous therapies. CONCLUSION: Low-dose RTX obtains remission in <50% of PMN patients. Probably, higher doses and longer treatments are needed to induce and maintain a response. The balance between the costs and benefits should guide the selection of the patient and the optimal dosage.

The long-term outcome of renal transplantation of IgA nephropathy and the impact of recurrence on graft survival.

BACKGROUND: Few data are available on allograft survival at 15 years, the impact and the predictors of recurrence of the original disease in renal transplanted patients with IgA nephropathy (IgAN). METHODS: In this retrospective study, we compared the long-term outcome of renal transplant in 190 patients with IgAN with that of 380 non-diabetic controls and evaluated the impact of recurrence of IgAN on the graft outcome. RESULTS: At 15 years, the patient survival was 88.3% in IgAN patients and 82.6% in controls (P = 0.12), while the death-censored graft survival was 62.6 and 72.4%, respectively (P = 0.038). IgAN had a higher cumulative incidence of graft failures in comparison with controls even considering death as a competing risk (P = 0.025). At multivariate analysis, IgAN [relative risk (RR) = 1.468, P = 0.026], delayed graft function recovery (RR = 2.394, P = 0.000) and acute rejection (RR = 2.51, P = 0.000) were predictive of graft loss. IgAN recurred in 42 grafts (22.1%), of them, 12 were lost for recurrence and in another 6 recurrence was considered a concomitant cause of graft loss. The 15-year death censored graft survival was 68.3% in non-recurrent and 51.2% in recurrent patients (P = 0.069). Pure graft survival of non-recurrent IgAN patients was similar to that of controls (P = 0.406). At Cox analysis, the recurrence of IgAN significantly reduced from 1981 to 2010 (P = 0.0065, RR = 0.936). CONCLUSIONS: IgAN emerged as an independent predictor of worse graft outcome in the long-term. Recurrence of IgAN seems to progressively reduce in transplants performed from 1981 to 2010.

[Possible vaccine-induced immune thrombotic thrombocytopenia in a patient with diabetes and chronic kidney disease or random association?]

We report the case of a 75-year-old man who developed acute myocardial infarction 12 hours after the first dose of ChAdOx1 nCov-19 vaccine. The event was associated with a transient decrease of platelet count and the detection of anti-PF4 antibodies approximately 45 days after the event. Vaccine-induced thrombotic thrombocytopenia (VITT) is characterized by the onset of venous or arterial thrombosis in temporal relationship to the administration of anti-Sars-Cov-2 viral vector vaccines (ChAdOx1 nCov-19 and Ad26.COV2.S), thrombocytopenia and the production of anti-PF4 antibodies. It occurs mainly at a young age, even if the median age is 54 years; it is often associated with thrombosis in atypical sites, such as the cerebral sinus. Our reported case does not present all the diagnostic criteria of VITT. However, the close temporal relationship between ChAdOx1 nCov-19 vaccine administration, thrombosis, and concomitant anti-PF4 antibodies positivity makes the case suggestive of a possible slight form of VITT.

CC chemokine receptor 5 polymorphism in chronic periaortitis.

OBJECTIVE: Chronic periaortitis (CP) is a rare disease characterized by a fibro-inflammatory tissue surrounding the abdominal aorta, and includes non-aneurysmal [idiopathic retroperitoneal fibrosis (IRF)] and aneurysmal forms [inflammatory abdominal aortic aneurysm (IAAA)]. We investigated whether CC chemokine receptor 5 (CCR5)Δ32 polymorphism confers susceptibility to CP. METHODS: One hundred CP patients and 180 healthy controls were genotyped for CCR5Δ32 polymorphism by molecular methods. The patients were subgrouped according to the type of CP (IRF or IAAA) and the presence of established atherosclerotic disease (ischaemic heart disease, cerebrovascular disease and peripheral arterial disease). RESULTS: The distribution of the CCR5Δ32 genotype differed between CP patients and controls (P = 0.01). The CCR5Δ32 allele was more frequent in CP patients than in controls [P = 0.02, odds ratio (OR) 2.8 (95% CI 1.2, 6.4)]. The distribution of the CCR5Δ32 genotype did not differ significantly between IRF patients and controls, whereas the CCR5Δ32 allele was more frequent in IAAA patients than in controls [P = 0.0001, OR 10.0 (95% CI 3.7, 27.3)]. Furthermore, the CCR5Δ32 allele occurred more frequently in IAAA than in IRF patients [P = 0.001, OR 6.4 (95% CI 2.1, 19.1)]. The CCR5Δ32 allele frequency was higher in IAAA patients without established atherosclerotic disease compared with controls [66.7 vs 5.6%, P = 0.00001, OR 34.0 (95% CI 7.4, 156.3)], but not in IAAA patients with atherosclerotic disease and IRF patients with or without atherosclerotic disease. CONCLUSIONS: The CCR5Δ32 polymorphism might be associated with an increased risk of developing the aneurysmal form of CP, IAAA, particularly in patients without established atherosclerotic disease. Chemokines may have a role in the pathophysiology of CP.

Long-term outcome of renal transplantation in patients with idiopathic membranous glomerulonephritis (MN).

BACKGROUND: Little information is available about the long-term outcome of renal transplanted patients with idiopathic membranous nephropathy (MN). METHODS: The outcomes of 35 first renal transplants performed between 1975 and 2008 in patients with MN were compared with those of 70 controls transplanted in the same period and matched for sex, age and source of donors. RESULTS: The mean post-transplant follow-up was 117 ± 86 months for MN patients and 123 ± 83 months for controls. At 15 years, patient survival was 96% in patients with MN and 88% in the controls (P = ns), while graft survival rates were respectively 40% and 69% (P = 0.06). MN recurred in 12 patients (34%), namely in 4/8 (50%) patients who received the kidney from related living donors and in 8/27 (29.6%) who received the kidney from a deceased donor. Recurrence led to graft failure in six patients, all deceased donor kidney recipients, within 54 ± 33 months. The other six grafts are functioning 134 ± 73 months after transplantation. Patients with recurrence were more frequently females (42% vs 4.3%, P = 0.02). The recurrence occurred earlier (4.8 ± 3.0 vs 45.6 ± 46.9 months, P = 0.05), and there was a trend to develop a higher proteinuria (7.1 ± 5.5 vs 3.67 ± 2.6 g/24 h, P = 0.1) in grafts eventually lost because of recurrence. CONCLUSIONS: The long-term patient survival was similar in renal transplant recipients with MN and in controls. The graft survival was lower in MN patients than in controls, although the difference was at borderline significance. Recurrence occurred in one-third of the patients and caused graft loss in half of them.

Carotid plaques in patients with long-term lupus nephritis.

OBJECTIVES: To evaluate the prevalence of carotid plaques in patients with long-term lupus nephritis (LN). METHODS: Intima-media thickness (IMT) and carotid plaques were evaluated with ultrasound in 75 patients after a follow-up of LN of 158+/-106 months and in 75 sex -and age-matched controls. Traditional and non-traditional atherosclerotic risks factors were also tested. RESULTS: IMT was not different between LN patients and controls, but 18% of LN patients had carotid plaques in comparison to 2.6% of controls (p=0.004). The LN patients more frequently had hypertension (p=0.0001), hypercholesterolemia (p=0.0001), were overweight (p=0.009), in menopause (p=0.01) than controls. More frequently, LN patients with carotid plaques had renal insufficiency (p=0.03), longer duration of lupus (p=0.05), anti-phospholipid antibodies (p=0.018), high C-reactive protein (p=0.03), high reactive oxygen species (p=0.001) than those without plaques. Patients with plaques were older (p=0.000001), in menopause (p=0.000001) and more frequently had cardio-vascular accidents during observation (p=0.02). The time of exposure to pathological values of systolic and diastolic blood pressure was longer (p=0.000001) and the percentage of pathological values of these variables during the follow-up was higher (p=0.000001) in patients with carotid plaques. At multivariate analysis, older age (p=0.0025), longer time of exposure to pathological values of blood pressure (p=0.015) and of cholesterol (p=0.04) were independent predictors of carotid plaques. CONCLUSIONS: Carotid plaques were more frequently found in LN patients than in controls. Although inflammatory markers and lupus related factors may contribute to the development of atherosclerosis, only traditional risk factors such as age, hypertension and hypercholesterolemia were the independent predictors.

Long-term outcome of renal transplantation in adults with focal segmental glomerulosclerosis.

SUMMARY: Little information is available about the long-term results of kidney transplantation in adults with focal segmental glomerulosclerosis (FSGS). The outcomes of 52 renal transplants performed between 1988 and 2008 in 47 adults with FSGS were compared with those of 104 matched controls (median follow-up 93.4 vs. 109.4 months respectively). At 15 years, patient survival was 100% and graft survival 56% in FSGS patients vs. 88.3% and 64% respectively in controls (P = NS). FSGS recurred in 12 out of 52 grafts (23%) and led to graft failure in seven within 10 months (median). In the other five cases, proteinuria remitted and grafts are functioning 106 months (median) after transplantation. A second recurrence developed in five out of eight re-transplanted patients (62.5%) who lost their first graft because of recurrence; only one graft was lost. Patients with recurrence were more frequently male subjects (83% vs. 40%, P = 0.02), younger at diagnosis of FSGS (16.3 +/- 6.8 vs. 24.1 +/- 11.5 years, P = 0.03) and of younger age at transplantation (28.4 +/- 7.8 vs. 35.8 +/- 12.2 years, P = 0.05). Treatment with plasmapheresis plus ACE inhibitors achieved either complete or partial remission in 80% of the cases. Long-term patient and renal allograft survivals of adults with FSGS were comparable to those of controls. Recurrence was more frequent in young patients and in patients who lost a previous graft from recurrence. Graft loss resulting from a second recurrence is lower than expected.

Rituximab monotherapy for remission induction of proliferative lupus nephritis flares: description of 3 cases.

BACKGROUND: The efficacy and safety of rituximab when not associated with other immunosuppressive therapy in induction of remission of proliferative lupus nephritis (PLN) has not until now been proven. METHODS: We report on 3 patients with PLN (class IV), 1 with a nephritic flare and 2 with a proteinuric flare (all with nephrotic syndrome (NS)) treated with 3 intravenous methylprednisolone pulses for 3 consecutive days and with rituximab at day 3 and day 18 associated with oral prednisone. At the beginning of the fourth month of therapy, mycophenolate mofetil was combined with prednisone. RESULTS: Three months after the beginning of therapy, renal function improved in the patient with the nephritic flare, and proteinuria fell to within the non-nephrotic range in all 3 patients. At the end of a follow-up of 24 months, the patient with the nephritic flare had normal renal function (serum creatinine from 1.7 to 1 mg/dL) and mild proteinuria (from 6 to 0.7 g/24 hours). The second patient was in complete remission (proteinuria from 5 to 0.127 g/24 hours) 27 months after the beginning of therapy. In the last patient, followed for 10 months, mild proteinuria persisted (from 6.6 to 0.7 g/24 hours). The therapy was well tolerated by all patients. No adverse effects occurred during the follow-up. CONCLUSION: Although our results must be confirmed by larger prospective studies, rituximab associated with methylprednisolone pulses without any other concomitant immunosuppressive drug seems to be effective and safe for induction therapy of severe flares of PLN.

Renal transplantation in adults with Henoch-Schonlein purpura: long-term outcome.

BACKGROUND: Little information is available about the long-term outcome of renal transplantation in adults with Henoch-Schonlein purpura (HSP). METHODS: We compared the outcomes of 17 patients with HSP who received 19 renal transplants with those of 38 controls matched for time of transplantation, age, gender and source of donor. The mean post-transplant follow-up was 109 +/- 99 months for HSP patients and 110 +/- 78 months for controls. RESULTS: The actuarial 15-year patient survival was 80% in HSP patients and 82% in controls, and the death-censored graft survival was 64% in HSP patients and in controls. The risks of acute rejection, chronic graft dysfunction, arterial hypertension and infection were not different between the two groups. In eight grafts (42%) recurrence of HSP nephritis was found (0.05/patient/year). In spite of therapy, one patient died and four eventually restarted dialysis respectively 10, 32, 35 and 143 months after renal transplant. Seventy-one percent of grafts transplanted in patients with necrotizing/crescentic glomerulonephritis of the native kidney had HSP recurrence in comparison to 12% of recurrences in patients with mesangial nephritis (P = 0.05) CONCLUSIONS: Long-term patient and allograft survival of HSP patients was good. However, 42% of HSP patients, particularly those with necrotizing/crescentic glomerulonephritis of the native kidneys, developed a recurrence of HSP nephritis that eventually caused the loss of the graft function in half of them.

Skin infection due to Alternaria species in kidney allograft recipients: report of a new case and review of the literature.

A kidney allograft recipient developed a cutaneous infection 29 months after transplantation, due to the dematiaceous fungus Alternaria infectoria on his right forearm and left leg. Since the lesions were too large to be excised, the patient was treated only with systemic itraconazole and a reduction of the immunosuppressive therapy. After 4 months, the lesions were completely healed, and no relapses were observed at follow-up of 22 months. Twenty-seven other cases of cutaneous alternariosis have been described so far in renal transplant recipients. All types of immunosuppressive treatment can be associated with Alternaria infection, for which predisposing factors are jobs with frequent contact with earth, diabetes mellitus and skin trauma. In 70% of cases the infection occurred within the first year after transplantation. More frequently the lower limbs were involved and the lesions were multiple. Alternaria alternata was the commonest causative agent, followed by Alternaria tenuissima,Alternaria infectoria and Alternaria chartarum. The treatment is far from being standardized, but the best results are obtained with the surgical excision of the lesion(s) associated with systemic antifungal therapy. Since relapses are possible, strict control of the patients over time is essential.

The long-term prognosis of renal transplantation in patients with lupus nephritis.

BACKGROUND: Few data are available about the long-term outcome of renal transplantation in patients with systemic lupus erythematosus (SLE). METHODS: Between June 1982 and 2004, a total of 33 adults with lupus nephritis received 35 kidney allografts. Outcomes of these grafts and those of 70 controls matched for age, sex, and donor source who underwent transplantation during the same period were compared. RESULTS: Mean follow-up after renal transplantation was 91 +/- 59 months for patients with lupus and 90 +/- 64 months for controls. Actuarial 15-year patient (80% versus 83%) and death-censored graft survival rates (69% versus 67%) were not significantly different between patients with lupus and controls. Risks for acute and chronic rejection, arterial hypertension, and infection were not different between the 2 groups. Mean serum creatinine levels also were similar in the 2 groups at the last follow-up visit. Intravascular thrombotic events occurred in 9 patients with SLE (26%) and 6 controls (8.6%; P = 0.038). In the SLE group, 6 of 7 antiphospholipid (aPL) antibody-positive versus 3 of 17 aPL antibody-negative patients experienced thrombotic events ( P = 0.015). Recurrence of lupus nephritis was documented in 3 renal grafts (8.6%), but no graft was lost because of recurrent lupus nephritis. CONCLUSION: Long-term patient and graft survival probabilities were similar in patients with SLE and matched controls. The risk for thrombotic complications was greater in patients with SLE, particularly aPL-positive patients. Nephritis recurred in less than 10% of patients with SLE and did not influence graft survival.

Membranous nephropathy in systemic lupus erythematosus: long-term outcome and prognostic factors of 103 patients.

OBJECTIVES: The objective of this study was to evaluate the clinical features, course, outcome, and prognostic indicators in lupus membranous nephritis (LMN) and to compare data of "pure" LMN vs "mixed" forms. METHODS: We retrospectively examined medical records and kidney biopsies of 103 patients with a diagnosis of LMN. RESULTS: Sixty-seven patients had "pure" LMN and 36 had "mixed" forms. Patients with mixed LMN had more frequent nephrotic syndrome (66.6 vs 44.7%, P = 0.05), low C3 (83.3 vs 62.6%, P = 0.05) and C4 (80.5 vs 52.2%, P = 0.005), anti-DNA positivity (86.0 vs 62.6%, P = 0.03), and a tendency toward a lower creatinine clearance (93 ± 29 vs 112 ± 50 mL/min, P = 0.07). Moreover, mixed membranous nephritis had a higher activity and chronicity index (6.5 ± 2.1 vs 1.4 ± 2.03, P = 0.005 and 2.4 ± 1.7 vs 1.4 ± 1.8, P = 0.0001, respectively). Methylprednisolone pulses and immunosuppressive therapy were more often used in patients with mixed forms (86.1 vs 60.6%, P = 0.016 and 83.3 vs 57.5%, P = 0.008, respectively). After a mean follow-up of 156.5 ± 104.5 months, there was no difference in the 2 subgroups concerning the number of patients achieving remission and patient/renal survival (94.5 vs 94.0% and 85.8 vs 86% at 10 years). At multivariate analysis, serum creatinine at presentation (P = 0.0013), chronicity index (P = 0.007), failure of achieving remission (P = 0.000001), and occurrence of nephritic flares (P = 0.00167) were independent predictors of chronic renal insufficiency. CONCLUSIONS: Despite the differences in clinical and histological presentation, a therapy tailored on the grounds of clinical and histological features may reduce the differences in the outcome of white patients with mixed and pure membranous nephritis.

Membranoproliferative glomerulonephritis type I in renal transplantation patients: a single-center study of a cohort of 68 renal transplants followed up for 11 years.

BACKGROUND: To evaluate the long-term outcome of renal transplant patients with membranoproliferative glomerulonephritis (MPGN) type I and the impact of recurrence. METHODS: The outcomes of 68 renal transplants performed between 1976 and 2009 in 63 patients with MPGN were compared with those of 136 controls matched for time of transplantation, sex, age, and source of donors. RESULTS: The mean posttransplant follow-up was 131.3±83.8 months for patients with MPGN and 139.21±88.7 months for controls. At 15 years, patient survival rates were 76.2% in patients with MPGN and 78.8% in controls (P=ns), whereas pure graft survival rates were 68% in MPGN and 67.9% in controls (P=ns). MPGN recurred in 16 patients (23.5%) 44±30.3 months after transplant (range, 3.5-105 months). Of recurrent grafts, nine were lost for recurrence within 116.5±51.36 months, three patients died with functioning kidney, the other 4 grafts are functioning 156.7±47.5 months after transplantation. Graft survival at 15 years was 73.5% in nonrecurrent and 40.4% in recurrent patients (P=0.02). Patients with recurrence were younger at diagnosis of MPGN (17.64±5.02 years vs. 22.9±9.6 years; P=0.037) and had low C3 more frequently than nonrecurrent patients (75% vs. 28.8%; P=0.01). Proteinuria was higher in recurrent patients who lost the graft in comparison with those with functioning graft (7.14±4.05 vs. 2.86±1.95; P=0.02). CONCLUSIONS: The long-term patient and graft survival were similar in patients with MPGN and in controls. Recurrence occurred in one-fourth of patients and caused graft loss in 56%. Younger age at diagnosis of MPGN and low C3 during transplantation seems to be predictive of recurrence.

Anti-C1q autoantibodies in lupus nephritis.

Anti-C1q antibodies are found in a variety of diseases, in addition to systemic lupus erythematosus (SLE), and in 3-5% of normal individuals. In particular, anti-C1q antibodies are detected at a high titer in 100% of patients with hypocomplementemic urticarial vasculitis and in 30-48% of SLE patients. Their titer correlates with active renal disease with a sensitivity of 44-100% and a specificity of 70-92%. An increase in anti-C1q antibody titer has been suggested to be able to predict renal flares in lupus nephritis so that monitoring anti-C1q might be valuable for the clinical management of SLE patients as a noninvasive biological marker. Recently our group studied 228 patients affected by lupus nephritis and found that the association of anti-C1q, C3, and C4, in a multivariate analysis, provided the best prediction of renal flares, particularly in patients with focal and diffuse proliferative lupus nephritis and in the absence of antiphospholipid antibodies.

The long-term outcome of 93 patients with proliferative lupus nephritis.

BACKGROUND: Few data are available about the very long-term outcome of patients with proliferative lupus nephritis. METHODS: Ninety-three Italian patients with biopsy-proven proliferative lupus nephritis (15 with class III, 9 with class III+V, 64 with class IV and 5 with class IV+V) followed for a median follow-up of 15 years in a single renal unit were considered for this observational study. Patients were treated with an induction treatment consisting of high doses of corticosteroids plus immunosuppressive agents in the more severe cases. This treatment was repeated in the event of a renal flare. Then corticosteroids and immunosuppressive agents were reduced to the minimal effective dose for maintenance. RESULTS: Renal survival including death was 97% at 10 years and 82% at 20 years. At the last follow-up visit, 59 patients were in complete renal remission, 18 were in partial renal remission, four patients had chronic renal insufficiency, six had entered end-stage renal disease and six patients had died. At multivariate analysis the lack of achievement of complete renal remission and the occurrence of nephritic flares were significantly correlated both with the risk of doubling plasma creatinine and death or dialysis. Those patients who entered complete renal remission had significantly less probability of developing nephritic flares. CONCLUSION: The long-term prognosis of Caucasian patients with proliferative lupus nephritis may be better than usually thought. Favorable factors for good long-term outcome are the achievement of complete renal remission, the absence of nephritic flares and their complete reversibility after therapy.

Long-term outcome of idiopathic retroperitoneal fibrosis treated with surgical and/or medical approaches.

BACKGROUND: Retroperitoneal fibrosis is a severe disease that affects the ureters, causing renal insufficiency in three-quarters of patients. The optimal treatment is far from being established. METHODS: Seventeen patients with idiopathic retroperitoneal fibrosis and ureteral entrapment followed in our unit for at least 1 year were selected for this study. At presentation 13 patients had renal insufficiency. All patients received steroids, associated with ureterolysis in five (group 1), with azathioprine in six (group 2) and with tamoxifen in six (group 3). Four patients of group 2 and five of group 3 received ureteral stenting or nephrostomy. There were no significant differences among the three groups or the clinical and biochemical characteristics at presentation. RESULTS: All patients of groups 1 and 2 entered remission after therapy. One patient from group 3 did not respond to therapy. During a mean follow-up of 56 +/- 41 months, three patients (two from group 1, one from group 2, 18%) had a recurrence of the disease, which fully responded to retreatment in all three cases. At the last observation, all patients were alive; three patients (18%) had renal insufficiency, of them one from group 1 had to start dialysis 6 years after ureterolysis, one patient from group 2 and one from group 3 had serum creatinine of 1.5 mg/dl. Renal survival was 100% at 5 years and 80% at 10 years. CONCLUSIONS: In most patients, each of the three different therapeutic approaches restored renal function and significantly reduced the fibrotic mass in the short-term and maintained stable serum creatinine in the long-term.

Withdrawal of therapy in patients with proliferative lupus nephritis: long-term follow-up.

BACKGROUND: Whether corticosteroid and immunosuppressive therapy may be safely withdrawn in patients with proliferative lupus nephritis is still unclear. METHODS: In 32 patients with biopsy-proven proliferative lupus nephritis previously put into remission, therapy was gradually tapered off. RESULTS: When immunosuppressive therapy was stopped (median: 38 months; 25th-75th percentile: 24-81 months, after biopsy), 24 patients were in complete remission and eight had a median proteinuria of 1.05 g/24 h (25th-75th percentile: 0.91-1.1 g/24 h) with normal renal function. After stopping therapy, patients were followed for a median of 203 months (25th-75th percentile: 116-230 months). Fifteen patients (Group 1) never developed lupus activity. The other 17 patients (Group 2) developed lupus exacerbations in a median of 34 months (25th-75th percentile: 29-52 months) after stopping therapy and were re-treated. The only significant differences between the two groups were the longer median durations of treatment, 57 months (25th-75th percentile: 41.5-113.5 months) vs 30 months (25th-75th percentile: 18-41 months; P<0.009), and remission, 24 months (25th-75th percentile: 18-41) vs 12 months (25th-75th percentile: 7-20 months; P<0.02), before stopping therapy in Group 1 than in Group 2. At last follow-up, 12 patients of Group 1 were in complete remission, two had mild proteinuria and one had died. In Group 2, one patient died, 14 were in complete remission, one had mild proteinuria and in another patient serum creatinine doubled. CONCLUSIONS: Some patients with severe lupus nephritis who enter stable remission can be maintained without any specific treatment for many years. Those patients who have new flares can again go into remission with an appropriate treatment. The longer the treatment and remission before withdrawal, the lower the risk of relapse.