LAUSR

Sir, We read with great interest the article titled BNephrotoxicity in children with frequently relapsing nephrotic syndrome receiving long-term cyclosporine treatment^ by Hamasaki et al. [1]. In a retrospective study on 36 patients with continuous cyclosporine A (CsA) treatment for >3 years (median, 4.5 years), the prevalence of CsA nephropathy (CsAN) was 36.1% (13/36 patients). The authors also found that the risk of CsAN was positively correlatedwith the duration of continuous CsA treatment, with an odds ratio (OR) of 6.6 (1.18–36.98) at >5 years of treatment (vs. 0–2 years). In 2006, we reported that alternative treatment (i.e., replacement of CsA with mycophenolate mofetil) should be considered in patients who continue to exhibit steroid-dependent nephrotic syndrome (SDNS) despite the administration of CsA for >3 years [2]. Our recommendation was based on long-term CsA treatment for >3 years being an independent risk factor for the development of CsAN. However, there have been no recent reports regarding the impact of interrupted CsA treatment on the progression of CsAN. Based on our experience in a single center, we would like to discuss the significant impact of CsA treatment interruptions on irreversible CsAN prevention. Between 2005 and 2014, repeat biopsies were performed to provide histological evidence of CsAN in 28 patients with SDNS (19 boys) who had undergone long-term CsA treatment (total treatment period, >4 years) at Saitama Children’s Medical Center. The mean age of the patients at the time of nephrotic syndrome diagnosis was 5.7 ± 4.0 years. All patients developed SDNS at a mean age of 6.4 ± 4.0 years and were prescribed CsA for a mean total period of 6.3 ± 2.4 years. Although strategic treatment interruptions (CsA discontinuation for >12 months) were performed in 12 of the 28 patients, the prevalence of CsAN was 64% (18/28 patients), with the vast majority of our patients havingmild arteriolar hyalinosis without significant interstitial fibrosis (16/18 patients). There were no differences in clinical characteristics, such as gender, age at CsA initiation, total period of CsA treatment, the use of angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers, or a history of steroid-resistant nephrotic syndrome between patients with and those without CsAN. The prevalence of CsAN in patients with treatment interruption for >12 months was significantly lower than in those without interruption (5/12 vs. 13/16 patients, P < 0.05). In addition, the prevalence of CsAN in patients with continuous CsA treatment for >3 years was significantly higher than in those without continuous treatment (16/21 vs. 2/7 patients, P < 0.05). In a multivariate analysis using a logistic regression model, only the association between the risk of CsAN and prolonged continuous CsA treatment was statistically significant [OR of 4.91; 95% confidence interval (CI), 1.04–23.2; P < 0.05]. Hamahira et al. previously reported that repeat biopsies at an average of 12 months after discontinuation of CsA revealed the disappearance of arteriolopathy in six of eight patients (75%) who had received continuous CsA treatment (mean, 37months) [3]. In contrast, interstitial fibrosis did not improve after discontinuation of CsA in their study. Despite the fact that the total period of CsA treatment was relatively long in our study (mean, 6.3 years), significant interstitial fibrosis (>5% of the cortical area) did not occur in any patient with treatment interruption for >12 months. Thus, the improvement of arteriolopathy after the discontinuation of CsAmay enable our patients to receive long* Shuichiro Fujinaga [email protected]