Corticosteroids pulse therapy and oral corticosteroids therapy for IgA nephropathy patients with advanced chronic kidney disease: results of a multicenter, large-scale, long-term observational cohort study

We previously reported a multicenter retrospective cohort study of biopsy-proven IgAN patients at four kidney centers in Japan: the University of Tsukuba, Toranomon Hospital, Seirei Sakura Citizen Hospital, and Toranomon Hospital Kajigaya [15]. The 1923 patients who underwent a kidney biopsy at one of these centers during the period from March 1, 1981 to December 31, 2013 were recruited. Each patient’s observation began, and all baseline parameters were recorded, at the time of the biopsy. The study’s clinical endpoint was end-stage kidney disease (ESKD), which was defined as the initiation of renal replacement therapy. Renal replacement therapy, in turn, was defined as dialysis or pre-emptive kidney transplantation, though no patients underwent transplantation during our observation. Dialysis was initiated by a physician’s decision in each patient, in accordance with the criteria generally used for dialysis initiation in Japan, i.e., serum creatinine of ≥8 mg/dl, uremic symptoms, or lowered ADL due to chronic renal failure. This practice did not vary over the approximately 30-year study period. All patients were observed until ESKD, death, or censoring. The mean follow-up period of the whole cohort was 8.3 years.

We categorized the patients into three groups based on the method of corticosteroids administration. The steroid pulse (SP) group included patients who received intravenous methylprednisolone (mPSL) pulse therapy for at least 3 days in a row during their follow-up period. Typical steroid pulse therapy in this study was administered over a 3-week period, with a single course of methylprednisolone (500 mg/day) for 3 days followed by oral corticosteroids, according to the protocol reported by Hotta et al. [7]. However, each patient’s protocol could be altered by the treating physician according to the patient’s status, since at the time of our study there was currently no firm evidence or established regimen to guide the administration of corticosteroids in patients with IgA nephropathy. Thus, a single dosage of methylprednisolone varied from 500 mg to 1000 mg, the number of doses varied from one to three courses in a single treatment series, and the tapering schedules for the corticosteroids also varied.

The oral steroid (OS) group included the patients who received corticosteroids without an intravenous pulse. The no steroid (NS) group did not receive any corticosteroids. In light of our previous finding [15], we did not investigate the influence of tonsillectomy.

From the 1923 patients in the total cohort, we extracted the cases of the 764 patients with chronic kidney disease (CKD), stage G3–G4 (15 ≤ eGFR < 60 mL/min/1.73m²) as subjects for the present study. The eGFR was calculated by the formula for Japanese created by Matsuo et al. [16].

We stratified the cases by the amount of urine protein (UP), which was calculated as the urine protein to urine creatinine ratio (g/gCr), and investigated the renal survival. The data are summarized as proportions (percent), median (1st interquartile – 3rd interquartile) or means (± standard deviation). Categorical variables were analyzed by Fisher’s exact test, and continuous variables by Student’s t-test. The survival analysis was performed by the Kaplan-Meier method and with the log-rank test. Hazard ratios of ESKD were calculated according to a Cox-proportional hazard model. All analyses were conducted using the R software program, ver. 3.4.2 (The R Foundation for Statistical Computing Platform: x86_64-w64-mingw32/× 64 (64-bit)).

Table 1 summarizes the baseline characteristics of the 764 subjects (319 females, 445 males). The mean age was 50.3 ± 13.6 years. The median follow-up period was 70 months. The median UP at the time of the biopsy was 0.95 g/gCr. The numbers of patients with CKD stages G3a (45 ≤ eGFR < 60 mL/min/1.73m²), G3b (30 ≤ eGFR < 45 mL/min/1.73m²), and G4 (15 ≤ eGFR < 30 mL/min/1.73m²) were 397 (52.0%), 236 (30.9%), and 131 (17.1%), respectively.

Higher ages, shorter follow-up periods, and greater UP values were observed in the further-progressed G stages. The proportions of patients administered corticosteroids by each method were as follows. For the total subject group (G3a, G3b and G4 patients), 16% (n = 121), 18% (n = 139) and 66% (n = 504) of patients received SP, OS and NS treatment, respectively. Among G3a patients, the corresponding percentages were 14% (n = 56), 16% (n = 64) and 70% (n = 277). In G3b patients, they were 16% (n = 38), 19% (n = 45) and 65% (n = 153). And in G4 patients, 21% (n = 27), 23% (n = 30) and 56% (n = 74) of patients received SP, OS and NS therapy. Tonsillectomy was performed in 80 (10%) patients. Most of them (n = 68) were treated by a combination with SP. Only 8 patients underwent tonsillectomy without steroid administration.

Approximately 27% of patients had received RAAS inhibitors. Only 1.6% of subjects received immunosuppressive drugs other than corticosteroids.

During the follow-up period, 37.1% of the patients with UP ≥1.0 g/day and 11.2% of the patients with UP < 1.0 g/day reached ESKD. Figure 1a and b show the Kaplan-Meier curves of renal survival in the patients with G3a, G3b and G4 CKD for each level of UP. Among the patients with UP ≥1 g/gCr, the SP group showed significantly better renal outcomes (p < 0.001) compared to the CS and NS groups. Figure 1c shows the Kaplan-Meier curves of renal survival in the patients with G3a, G3b and G4 CKD with UP ≥1 g/gCr and taking renin-angiotensin system inhibitors. Even in this subgroup, SP showed significantly better renal outcome. On the other hand, in patients with UP < 1 g/gCr, there were no significant differences in renal survival among the three treatment groups.