Changes in clinicopathological features of primary nephrotic syndrome in children over a 10 years period: Research Paper

Background Primary nephrotic syndrome (PNS) is a common chronic disease in childhood and results in heavy illness burden. The aim of this study is to investigate the change in the clinicopathological features of children with NS worldwide. Methods Children with PNS were divided into two groups according to the date of admission. Group A was admitted from January 1, 2006 to December 31, 2010 and group B from January 1, 2011 to December 31, 2015. Clinical manifestations and pathological features of the children in both groups were retrospectively analyzed. This study includes 1072 patients. There were 783 cases of steroid-sensitive nephrotic syndrome (SSNS), 241 cases of steroid-resistant nephrotic syndrome (SRNS), and 48 cases of unknown steroid effects. The incidence of complications in group B was signicantly higher than that in group A, especially for infections. A total of 136 (12.96%) patients underwent renal biopsy. The most common lesions were minimal change disease (44.12%). The rate of mesangial proliferative glomerulonephritis (MsPGN) signicantly decreased and that of membranous nephropathy (MN) increased over the years. A total of 292 (27.24%) patients were administered immunosuppressants. The administration of Tripterygium was decreased, whereas the using of Tacrolimus increased over the years. continuous normal standard (SD); the median and interquartile range (IQR) a normal Comparisons of continuous variables between the two were performed with Student’s t-test or Mann-Whitney U test. 2 tests were used to analyse percentages.


Introduction
Primary nephrotic syndrome (PNS) is one of the most common kidney diseases in children with an incidence of 2.0 to 7/100,000 and a prevalence of 16/100,000 [1][2]. The incidence of PNS in East Asian children is approximately three to four times higher than that in Caucasians [3] [4]. The major clinical features of PNS are proteinuria, hypoalbuminemia, hyperlipidemia, and varying degrees of edema.
Glucocorticoids are the rst-line drug for children with PNS[2]. However, 60-80% of children with steroidsensitive nephrotic syndrome (SSNS) usually relapse, approximately 60% would have ve-fold or more incidences of relapses [1] and approximately 10% of children with PNS relapsed into adulthood [5].
Additionally, between 7.4% and 19.6% of children have steroid-resistant nephrotic syndrome (SRNS) in Turkey [6]. Steroid-sparing immunosuppressive medications, such as calcineurin inhibitors, cytotoxic drugs, Mycophenolate Mofetil, and Rituximab are used in SRNS and steroid-dependent/frequently relapsing nephrotic syndrome (SDNS/FRNS). Minimal change disease (MCD) is the most common pathological type of PNS in children. In recent years, focal segmental glomerulosclerosis (FSGS) has gradually increased in children with PNS [1]. Most children with refractory FSGS eventually develop endstage renal disease (ESRD). PNS is able to complicate with infection, thromboembolic disease and acute kidney injury. The mortality of PNS has been reduced from 40-0.7% because of the use of antibiotics, but infections were still the leading cause of death in children with PNS [5] [7].
Besides those complications in patient, PNS would cause a huge burden for family, and society in general. A cross-sectional analysis of a hospitalization database for children with the burden of this disease in the United States showed that an estimated 48,700 hospital days were required in 2006 and 2009[8] and the total cost was 25.9 billion dollars. This consumes large amounts of social resources and property, so we need to clarify the characteristics of PNS in order to manage it better.
There are currently no published data focusing on the clinicopathological changes of PNS. To identify the changes in clinicopathological features of children with PNS worldwide, we performed a retrospective study in a single center to examine the changes in clinicopathological features of childhood NS in recent years. This study aimed to summarize the clinicopathological features of children with PNS.

Study Design And Population
This was a retrospective study based on an electronic medical record system. We included children with PNS who were admitted to the Department of Pediatric Nephrology Rheumatology, Shandong Provincial Hospital A liated to Shandong University, from January 1, 2006 to December 31, 2015, with an onset age from 1 to 14 years. The search key words were PNS, age from 1 to 14 years. The same patient ID only counted one patient because of repeat hospitalization. Exclusion criteria were listed as follows: congenital NS and hereditary NS, or secondary NS, such as hepatitis B virus-associated nephritis and lupus nephritis.

Medical Record Collection
We collected general data (sex, age), steroid response (SRNS, SSNS), complications, renal pathological type and immunosuppressive regimen. Children who were diagnosed with PNS were divided into two groups: group A, with admission from January 1, 2006 to December 31, 2010, and group B, with admission from January 1, 2011 to December 31, 2015. The patients in group A who were admitted to hospital for recurrence during 2011-2015 wouldn't be included in group B.

Renal Biopsy
Indications of renal biopsy were SRNS, SDNS/FRNS. After seeking consent of the guardians of children, an ultrasound-guided percutaneous renal biopsy was performed. Kidney pathologies were based on the World Health Organization WHO 1995 classi cation scheme for glomerular diseases [9].
PNS was divided into the following two types according to steroid response: (1) SSNS was de ned as when patients' urinary protein turned negative within 4 weeks of prednisone treatment in su cient amounts (2 mg/kg/d or 60 mg/m 2 /d).
(2) SRNS was de ned as when urinary protein was still positive after prednisone treatment for more than 4 weeks.

Pns Recurrence And Frequency Of Recurrence
Relapse was identi ed when morning urine protein changed from negative to (+++) or (++++) for 3 consecutive days, 24-h urinary protein quantitation was ≥ 50 mg/kg, or urinary protein/creatinine (mg/mg) was ≥ 2.0. Frequent relapse was de ned as relapse that occurred more than two times in the course of disease within 6 months, or relapse that occurred more than three times within 1 year.
Infections included lower respiratory tract infection, urinary tract infection, and the others that required hospitalization.
The criteria for admission were the emergence of complications and recurrence of proteinuria.

Statistical analysis
Statistical analyses were performed with SPSS software (version 22.0, IBM, NY, USA). Using the Kolmogorov-Smirnov test, continuous variables with normal distribution are presented with the mean and standard deviation (SD); the median and interquartile range (IQR) is shown when a normal distribution was not observed. Comparisons of continuous variables between the two groups were performed with Student's t-test or Mann-Whitney U test. 2 tests were used to analyse percentages. Descriptive statistics are presented as percentages and means and standard deviations. P values of < 0.05 were considered statistically signi cant.

Baseline characteristics
From January 1, 2006 to December 31, 2015, 8193 inpatients were admitted to our department, of whom 2832 were diagnosed with PNS. This number accounted for 34.57% (2832/8193) of the hospitalized children in the same period. A total of 2832 hospitalizations were recorded for 1072 patients (806 boys and 266 girls), the ratio of boys to girls with PNS was 3:1. There were 372 patients in group A and 700 patients in group B. The age of onset was from 1 to 14 years (mean age 4.92 ± 2.87 years). The frequency of hospitalization varied from 1 to 17 times with a mean time of 2.75 ± 3.03. There were 59 patients in Group A who still needed to be admitted to hospital in the period of Group B. The number of hospitalization days ranged from 1 to 65 days, with a mean of 10.41 ± 8.83 days. There were 783 cases of SSNS, 241 cases of SRNS, and 48 cases of unknown steroid effects. There were no signi cant differences in baseline characteristics between the two groups ( Table 1).  The main pathological types of SSNS were MCD and MsPGN, and those of SRNS were mainly MCD, MsPGN, and FSGS. The associations between pathological type and clinical classi cation and steroid response are shown in Table 3.

Discussion
Our study showed that there were some clinicopathological changes of PNS in children over 10 years (2006-2015). Children with PNS accounted for 34.57% of all hospitalized children in the Department of Pediatric Nephrology Rheumatology over the 10-year period. In 1982, children with PNS in some provinces and cities in China accounted for 21% of all hospitalized children and this rate was 31% in 1992 [12]. This nding suggests that hospitalization for PNS had increased over these years. In this study, the group B was twice the size of group A, which may have been a result from the economic development and convenient public transportation. In our study, the ratio of boys to girls with PNS was 3:1 and the subjects were mainly pre-school children, which was consistent with domestic and international studies [12] [13]. Because of disease relapse or infused cyclophosphamide, the number of hospitalizations uctuated between 1 and 17 times, and 59 patients in Group A still needed to be admitted to hospital in the period of Group B. The days of hospitalization varied from 1 to 65 days with a mean length of 10.41 ± 8.83 days. This hospital stay was longer than the mean length of stay in Gipson DS' study (5 ± 0.1 days)[8].
In our study, there was no signi cant difference in the response to glucocorticoids between the two groups. Dincel et al [14] found that 88.3% of children with PNS were SSNS and 11.7% of patients were SRNS, which was in accordance with our ndings. Another study showed that approximately 10% of patients with PNS had SRNS, and SRNS pathology was predominantly FSGS [13]. Even though our study did not show any difference in the rate of SRNS, the proportion of SRNS was higher than that in Dincel et al's study [14], and slightly higher than that (10-20%) previously reported in China [11]. This suggested an increasing trend of SRNS in recent years. There were 48 patients with an unknown steroid response because we were unable to follow them up.
MCD, MsPGN, and FSGS are the most common pathological types of PNS in children [14][15] [1]. Our ndings further supported this conclusion. In our study, the rate of MsPGN gradually decreased in recent years, while that of MN increased. The speci c etiology of this nding is unclear. One reason for the decrease in the percentage of MsPGN is due to the increasing tendency of MPGN or MN. In addition, a multicenter study in China [16] showed that an increased incidence of idiopathic membranous nephropathy (IMN) was associated with an increase in Particulate Matter 2.5(PM2.5) in the environment. How the mechanisms of exposure to PM2.5 increase the risk of MN remains to be elucidated. Furthermore, a previous study in our center [17] indicated that the incidence of children with FSGS and MN showed an upward trend in the past 20 years. Especially in the latest 5 years, the incidence of FSGS and MN in renal pathological types has signi cantly increased. However, other results [18] demonstrated that the main pathology of PNS in children was FSGS, which may have been related to hospitals receiving di cult referral cases. On the other hand, in our study, there were two children with FSGS and recurrent disease who had repeated renal biopsies, including one with pathological ndings of MN (II-III), which progressed to MN (III) and FSGS. The other child showed a change from minimal changes to FSGS, and the pathological ndings were more severe than the former ones. For those repeatedly refractory patients, repeated renal biopsy can assess the severity of disease, predict the outcome and adjust the treatment regimen timely.
In our study, 512 (47.76%) children with PNS had complications in hospital. In particular, the incidence of infections was signi cantly higher in group B than in group A. However, there were no signi cant differences in other types of complications between the groups. Some previous studies have shown that 60% of patients [19][8] have at least one complication, and14-16% of patients have at least one serious complication in the course of PNS. Historically, infections were the leading cause of mortality in children with PNS, and even now, infection is the leading cause of morbidity in children with PNS[20][2]. Infections are due to urinary loss of immunoglobulins and complement, impaired lymphocytic function. The reason for higher incidence of infection in group B in our nding was not clear, but it may be related to more air pollution. Song et al [21] have shown that the hospitalization rate of children with pneumonia was positively correlated with levels of the air pollution marker PM 2.5, and ambient PM 2.5 contributed as much as 33.1% (95% con dence interval: 22.6%, 42.4%) to total acute lower respiratory infection (< 5 years) deaths in China. Therefore, we speculated that the increase in hospitalization is due to the rising rate of infection.
Glucocorticoids are still the rst-line regimen for children with PNS. However, for SRNS or FRNS, the option of immunosuppressants has changed over recent years. The proportion of CsA and Triptolide administrated was decreased, while the proportion of TAC used was increased in our study. The reasons for this nding were related with the recommendation of TAC in the treatment guidelines in China and some other countries. However, CsA and Triptolide have severe side effects. In addition, the administration of immunosuppressive drugs improves the remission rate of refractory PNS. A previous report [19] showed that 51% of patients with PNS received two or more immunosuppressive agents. For children with SRNS, Trautmann A et.al showed that using enhanced immunosuppressive agents had a good effect [13]. CTX was the most used immunosuppressant [22], and our study con rmed it. Several case series suggest that TAC is effective for FRNS or SDNS[2]. Moreover, TAC was superior to MMF in maintaining remission of SRNS [23]. Reports [24] [25] found that Adalimumab and Rituximab can help patients to achieve a long-term remission period, improve quality of life, and reduce the hospitalization rate. However, Adalimumab and Rituximab are not widely used because of high costs. However, using immunosuppressive drugs also increased the rate of infection in children with PNS. Therefore, when using immunosuppressive agents, clinicians need to consider the advantages and disadvantages of these agents.
Our study also has some limitations. Primarily, this was a retrospective, single-center study. Future studies need to include other medical centers to identify changes in the clinicopathological features of PNS. In addition, the biopsy reports were not reported by the same pathologist because the study lasted 10 years.
In conclusion, our study showed the rate of complications has signi cantly increased with mainly infectious complications. The rate of MsPGN has decreased in recent years, while that of MN increased. Glucocorticoids are still the rst-line treatment for children with PNS. However, due to the choice of immunosuppressive agents, the use of Tacrolimus in children increased whereas Tripterygium decreased.
Haiyan Guo was a major contributor in writing the manuscript. Shunzhen Sun analysed and interpreted the patient data and revised the nal manuscript. Qian Li and Chen Yuan performed the kidney biopsy, and interpreted the patient data. Jing Wang Yue Su and Lichun Yu analysed data and interpreted the patient's clinical data and contributed to writing the manuscript. All authors read and approved the nal manuscript.