human development class

ORIGINAL ARTICLE

Low birth weight, but not postnatal weight gain, aggravates the course of nephrotic syndrome

Christian Plank & Iris Östreicher & Katalin Dittrich & Rüdiger Waldherr & Manfred Voigt & Kerstin Amann & Wolfgang Rascher & Jörg Dötsch

Received: 12 February 2007 /Revised: 16 July 2007 /Accepted: 16 July 2007 / Published online: 14 September 2007 # IPNA 2007

Abstract Clinical and animal studies have shown a higher risk of an aggravated course of renal disease in childhood after birth for babies small for gestational age (SGA). In addition relative “supernutrition” and fast weight gain in early infancy seem to support the development of later disease. In a retrospective analysis of 62 cases of idiopathic nephrotic syndrome treated between 1994 and 2004 at a university centre for paediatric nephrology, we related the course of disease to birth weight and to the weight gain in the first 2 years of life. Six children were born SGA (birth weight <−1.5 standard deviation score), and 56 were born as appropriate for gestational age (AGA). In all SGA children renal biopsy was performed, while only 55% of the AGA children underwent renal biopsy (P=0.07), showing no difference in renal histology. In the SGA group, four of six patients developed steroid resistance (vs 12/56 AGA, P<0.05). Of the SGA children, 83% needed antihyperten- sive treatment in the course of the disease compared to 39%

of the AGA children (P=0.07). The extent of weight gain between birth and 24 months of age did not influence the course of disease. In conclusion, we were able to find evidence for an aggravated course of idiopathic nephrotic syndrome in former SGA children. Independently of birth weight, weight gain in the first 2 years of life did not influence the course of disease.

Keywords Nephrotic syndrome . Child . Low birth weight .

Small for gestational age . Intrauterine growth restriction

Introduction

Nephrotic syndrome is defined by heavy glomerular proteinuria leading to hypoalbuminaemia, hypercholester- inaemia and oedema. Most children presenting with idiopathic nephrotic syndrome are affected by minimal change nephropathy (MCGN), and only a small proportion show focal segmental glomerulosclerosis (FSGS). Over 90% of children with idiopathic syndrome respond to steroid treatment [1], but, nevertheless, 60–80% of these children will suffer a relapse. In 60% of the relapsing children five or more relapses occur over time [1]. The main prognostic factor for renal survival is steroid responsiveness. The ten-year renal survival rate for ste- roid-resistant FSGS patients is 54%, according to the analysed cohort [2].

More than 15 years ago Barker and co-workers were the first to report an inverse correlation between birth weight and later death from cardiovascular diseases [3]. Epidemi- ological and experimental investigations have presented evidence for the role of the perinatal environment in programming body functions and diseases in later life [4]. Studies in indigenous populations with a high incidence of

Pediatr Nephrol (2007) 22:1881–1889 DOI 10.1007/s00467-007-0597-9

DO00597; No of Pages

C. Plank (*) : I. Östreicher : K. Dittrich : W. Rascher : J. Dötsch Department of Paediatrics, Kinder- and Jugendklinik, University of Erlangen-Nuremberg, Loschgestrasse 15, 91056 Erlangen, Germany e-mail: [email protected]

R. Waldherr Institute for Clinical Pathology, Heidelberg, Germany

M. Voigt Department of Paediatrics, Ernst-Moritz-Arndt-University of Greifswald, Greifswald, Germany

K. Amann Department of Pathology, University of Erlangen-Nuremberg, Erlangen, Germany

neonatal hypotrophy show, in addition, a higher risk for renal insufficiency in adult life, which is difficult to separate from the risk of metabolic diseases in these populations [5–7]. Another aspect in perinatal program- ming of later diseases is postnatal growth and weight gain. It is postulated that catch-up growth until the age of 2 years restores the infant’s size back to the genetic growth trajectory [8]. Up to 90% of children born small for gestational age (SGA) show catch-up growth [9]. Barker et al. studied the influence of postnatal growth on cardiovascular events in later life and showed that adults who had suffered coronary events were smaller at birth, thin at 2 years and had showed rapid weight gain thereafter. Other groups have shown negative consequences of catch- up growth on blood pressure, death from cardiovascular diseases and occurrence of type 2 diabetes [10]. In a cohort study Ong et al. showed an association between children’s catch-up growth in the first 2 years of life and their fatness at the age of 5 years. Of the children studied, 30.7% had a weight gain of 0.67 standard deviation score (SDS). Interestingly, these children had lower birth weights, lengths and ponderal indices [11]. Data from the US Collaborative Perinatal Project (1959–1974) looked at blood pressure elevation at the age of 7 years. This study did not show an elevated risk for SGA children but demonstrated a higher risk in children who crossed weight percentiles during early childhood [12]. A smaller Korean study showed a connection between weight gain in the first 3 years of life and increased systolic blood pressure at the age of 3 years [13]. In conclusion, infant weight gain could be a risk factor for the course of later diseases indepen- dently from SGA or intrauterine growth restriction (IUGR).

To date, three studies on children have looked for SGA as a risk factor for an aggravated course of idiopathic nephrotic syndrome. Two of these studies were performed in an Asian population [14, 15] and one was done in Slovenian children [16]. They all showed an unfavourable course of nephrotic syndrome in SGA children. However, there are no studies that examine the influence of early postnatal growth on the course of kidney diseases. Two of the studies were limited to children with clinical diagnoses of idiopathic nephrotic syndrome or minimal change disease [14, 16]. Therefore, we investigated our complete cohort of children with idiopathic nephrotic syndrome due to MCGN or FSGS to test the hypothesis that low birth weight and early weight gain are risk factors for an aggravated course of idiopathic nephrotic syndrome.

Methods

We retrospectively investigated every treated case of idiopathic nephrotic syndrome between 1994 and 2004 in

a university centre for paediatric nephrology. In a first step, all patients treated because of nephrotic syndrome, MCGN or FSGS were identified by the diagnostic codes 580.-, 581.-, 582.-, 583.- (ICD 9) or N00., N01., N04., N05., N06., and N08.8 (ICD 10) in a patients’ database. Then, we selected all patients with the clinical diagnosis of idiopathic nephrotic syndrome or the histological diagnosis MCGN or FSGS. Patients were 1–18 years old when nephrotic syndrome was diagnosed. Patients with chromosomal aberrations, congenital syndromes and congenital infections were excluded. The selected families were contacted. Informed consent of the parents and as far as possible assent of the patients were obtained. Copies were drawn from the patients’ prevention record with auxiology data at birth, days 3–10, weeks 4–6, months 3–4, months 6–7, months 10–12 and months 21–24, according to the German national child prevention program. Parents answered a questionnaire about parental auxiology, pregnancy and risk factors during pregnancy. Patient records were analysed for data on diagnosis, clinical course and therapy.

The study protocol was approved by the ethics commit- tee of the medical faculty of the University of Erlangen- Nuremberg. The study was conducted according to the Declaration of Helsinki and German national laws.

The definition of small for gestational age is very arbitrary, ranging from the 2.5th percentile to the 25th percentile [17]. Therefore, to distinguish between patients that were small for gestational age and those that were appropriate for gestational age, we used a birth-weight standard deviation score (SDS) ≤−1.5 (equivalent to the 7th percentile). SDS was calculated according to birth weight, gestational age and gender, using the percentiles of M. Voigt for German newborns between 1995–1997 [18]. The same references were used for calculation of the SDSs for head circumference and birth length. The ponderal index was calculated according to the formula: body weight (g)/ [body length (cm)]3.

In fact, not only patients with a birth weight below the 10th percentile or the <−1.5 SDS but also those within the entire lower birth-weight range are associated with an increased risk of later morbidity [19]. Therefore, we divided our cohort into quartiles according to different birth weight SDS ranges: <−1.0, −1.0 to 0, 0 to 1, >1.0. By this we wanted to detect differences between patients with low birth weight and patients with normal to high birth weight. SDSs for body weight at the given postnatal time points were calculated with Auxiology (version 1.0 b17, 2003) (Pfizer GmbH, Karlsruhe, Germany) according to the percentile references by Cole et al. [20]. With the Genotropin Auxiology Calculator (German version 2.0, 18.06.1999) (Pfizer) we calculated SDSs for body height, head circum- ference and body mass index (BMI) using the percentiles of Prader et al. [21] and Cole et al. [20], respectively.

1882 Pediatr Nephrol (2007) 22:1881–1889

Independent of birth weight, postnatal alimentation and weight gain in early childhood might influence the later course of kidney diseases. As a proxy for postnatal weight gain we analysed the difference between body weight SDSs at birth and at 24 months of age [11]. Percentile-crossing weight gain or loss can be shown as gain or loss of SDSs [12]. For example, a patient with a birth weight at the 2.3th percentile and weight at the 15.9th percentile at the age of 2 years would present with a gain of SDS of 1.0, indicating catch-up growth. In contrast, a SDS difference <0 would represent catch-down growth. The patients were divided in four groups according to the difference between SDS at 24 months and at birth (below −1.0, −1.0 to 0, 0 to 1, >1.0). SDSs were calculated on the basis of dry weight.

As characteristics of the clinical course we used age at manifestation, necessity of renal biopsy, histological diag- nosis, rate of primary and secondary steroid resistance, rate of primary and secondary steroid dependence, use of cyclophosphamide or cyclosporin A, rate of arterial hyper- tension in the course of disease, number of antihypertensive or antiproteinuric drugs, and number of relapses per year of follow-up. At last follow-up, number of patients with end- stage renal failure, creatinine clearance according to Schwartz [22], number of antihypertensive or antiproteinuric drugs and immunosuppressive drugs were determined.

Ambulatory blood pressure measurements are given as SDSs. SDSs were calculated on the basis of the data by de Man [23]. Formulae were provided by Elke Wühl, Department of Pediatrics, University of Heidelberg, Germany.

For the scoring of antihypertensive therapy, the following drugs were considered: captopril, enalapril, atenolol, pro- pranolol, nifedipin, amlodipin, prazosin and dihydralazin. The use of each class of antihypertensive or antiproteinuric drug during the course of the disease was given one score point, with a maximum score of 4.

Further definitions

Nephrotic syndrome was defined as proteinuria [urinary protein excretion >40 mg/m2 per hour (=1 g/m2 per 24 hours)], oedema, and hypoalbuminaemia (serum albumin level <25 g/l).

Steroid resistance was defined as ongoing proteinuria after 4 weeks of prednisone (PRD) therapy according to the Arbeitsgemeinschaft für Pädiatrische Nephrologie (APN) therapy scheme suggesting treatment with 60 mg/m2 body surface area (BSA) per day PRD divided into three single doses, with the largest dose in the morning (maximum dose 80 mg/day) [24]. Steroid dependence is defined as ongoing therapy with prednisone to sustain remission.

Relapse was defined as protein excretion >40 mg/m2 per hour (1 g/m2 per day) or as protein excretion >100 mg/dl in three successive analyses of morning spot urine. Therapy

for relapses followed the APN standard, with 60 mg/m2

BSA PRD divided into three single doses and the largest dose in the morning (maximum dose 80 mg/day) until morning spot urine showed protein excretion <30 mg/dl. This is followed by 40 mg/m2 BSA per 48 hour single doses in the morning (maximum 60 mg/day) for another 4 weeks [25]. Indications for renal biopsy were primary or secondary steroid resistance, the planned use of cyclosporin A, age below 1 year and above 10 years at the time of initial presentation and diagnostic hints for glomerulonephritis other than MCGN. One patient under- went two renal biopsies; the result of the second biopsy was included to the analysis.

An indication for oral cyclophosphamide therapy was frequent relapsing nephrotic syndrome, which means more then two relapses in 6 months or four relapses in 12 months [26]. Cyclophosphamide was given at a dose of 2–3 mg/kg body weight for 12 weeks in combination with oral administration of PRD in tapering doses on alternate days. In a total of four patients, cyclophosphamide was given as pulses every 4 weeks over 6 months in the case of steroid resistance. Cyclosporin A was given in the case of steroid resistance or steroid dependence after cyclophosphamide treatment without sustained success.

Statistical analyses of medians of the appropriate for gestational age (AGA) and SGA groups were performed by unpaired Mann–Whitney t-test. For the comparison of more than two groups the Kruskal–Wallis test was used. In the case of an overall P<0.05, a post-test according to Dunns was performed to identify statistical differences between individual groups. Values are given as median, minimum and maximum. Attributive variables in SGA and AGA children were tested by Fisher‘s exact test. For more than two groups, a chi-square test was applied. A P value less than 0.05 was considered significant. GraphPad Prism, version 3.00 for Windows, GraphPad Software, San Diego, California, USA, http://www.graphpad.com, was used for statistical analysis.

Results

Between 1994 and 2004 a total number of 105 cases were identified and contacted. Eighty-two patients responded to our contacting them. Sixty-two patients (59.0%) were suitable to be included in the analysis and presented with the diagnosis of idiopathic nephrotic syndrome, biopsy proven MCGN or FSGS. Four patients declined to participate. Twenty-three families did not respond at all. In ten cases birth data or clinical follow-up details were completely missing. Four patients presented with genetic forms of nephrotic syndrome. By renal biopsy one patient was diagnosed as having Schoenlein-Henoch purpura

Pediatr Nephrol (2007) 22:1881–1889 1883

nephritis, and another patient had nephronophtisis. All these patients were excluded from further analysis. Six children were identified as SGA, 56 as AGA (Table 1). In SGA und AGA children median gestational age was similar. Both groups differed significantly in birth weight, birth length and head circumference. Birth weight SDS, birth length SDS and head circumference differed signifi- cantly after correction for gestational age and gender.

Auxiology after birth in SGA children

In the SGA group birth weight SDS was between −3.91 and −1.51. To describe catch up growth in the SGA group we analysed weight development and growth during the first 2 years of life. In one patient auxiology data from birth until the age of 2 years was completely missing.

There was no body weight catch-up in three patients; the other two patients increased their body weight at least by 0.5 SDS. No patient reached a body weight SDS over 1.2 at 24 months of age.

One patient did not attain a normal body length according to SDS till the age of 24 months. Two patients showed an increase in length SDS>1.5. Another two patients showed a normal height SDS (>−2.0) at 24 months, but catch-up details regarding height SDS between birth and 24 months were missing. Because of the small number of SGA patients, further analysis on the influence of catch- up growth after SGA was omitted.

Data of mothers and risks factors during pregnancy

In both groups age at delivery was comparable (Table 2). In the SGA group there was one twin pregnancy, but one twin died in utero. In the SGA group, four mothers were first gravida, one was second gravida and one was third gravida. In the AGA group 32 mothers were first gravida, 17 second gravida, and seven third gravida.

In the SGA group there were no mothers with alcohol abuse, diabetes mellitus before pregnancy, arterial hyper- tension before pregnancy, or gestational diabetes. Two of six mothers had smoked during pregnancy, but there was a similar proportion of mothers that had smoked in the AGA group. Two mothers developed hypertension during preg- nancy in the SGA group (Table 2). In the AGA group one mother suffered from pre-eclampsia.

Course of nephrotic syndrome in SGA and AGA children

In former SGA patients, there was a trend for a higher median age at onset of the disease than in the AGA patients (P=0.08) (Table 3). Follow-up time was 6.2 years (2.2–17.25 years) in SGA children and 5.4 years (0.17–15.75 years) in AGA children [not significant (ns)]. In four SGA and 31 AGA children laboratory results at the initial manifestation of nephrotic syndrome were available. There was no difference concerning triglycerides, serum-cholesterol and serum albumin (data not shown).

All SGA patients underwent renal biopsy, while this procedure was only necessary for 31 of the 56 AGA patients (55%, P=0.07). In the renal biopsy FSGS was found in one SGA patient and in eight of the 31 AGA patients (ns). The majority of the children with SGA showed primary or secondary steroid resistance, in contrast to the AGA children (P<0.05). The rate of steroid dependence did not differ between the groups. In both groups more than half of the patients received cyclophos- phamide therapy. One patient in the SGA group and three patients in the AGA group were additionally treated with

Table 2 Data on mothers and pregnancy risk factors. Data are given as median and range

Parameter SGA (n=6)

AGA (n=56)

Maternal age at delivery (years) 28 (22–32) 27 (18–40) Smoking during pregnancy 216 (33.3%) 6/56 (10.7%) Alcohol during pregnancy 0/6 1/56 (1.8%) Diabetes mellitus before pregnancy 0/6 0/57 Arterial hypertension before pregnancy

0/6 1/56 (1.8%)

Gestational diabetes 0/6 1/56 (1.8%) Gestational hypertension 2/6 (33.3%) 7/56 (12.5%)

Table 1 Patients’ characteristics at birth. Data are given as median and range

Characteristic SGA (n=6) AGA (n=56)

Gestational age (weeks) 39.5 (37–41) 40 (29–42) Gender (male/female) 3/3 32/24 Birth weight (g) 2,735 (1,280–

2,950) 3,400 (1,320– 4,380)a

Birth weight (SDS) −1.84 (−3.91– −1.51)

−0.28 (−1.45– +1.72)b

Birth length (cm) 49.5 (36.0–50.0) 51.5 (39.0–58.0)b

Birth length (SDS) −1.21 (−5.67– −0.79)

−0.08 (−1.88– 2.37)a

Ponderal index 2.38 (2.09–2.74) 2.41 (2.1–2.93) Head circumference at birth (cm)

33.2 (32.8–35.0) 35.0 (29.0–38.0)b

Head circumference at birth (SDS)

−1.29 (−1.72– −0.42)

−0.04 (−1.93– 1.86)b

Differences between both groups are tested with unpaired Mann– Whitney t-test a P<0.001 b P<0.01

1884 Pediatr Nephrol (2007) 22:1881–1889

cyclophosphamide pulse therapy because of steroid resis- tance. Five of six patients in the SGA groups received cyclosporin A therapy; in the AGA group 23 of 56 patients (ns) received cyclosporin A. Two patients in the SGA group and four of 56 AGA patients were treated with tacrolimus (ns). The mean number of relapses per patient and follow- up year was 0.69 in the SGA and 0.65 in the AGS group (ns). Follow-up period did not differ between SGA and AGA patients. To control different follow-up times we analysed the number of relapses per patient year. There was no difference between the groups. In the SGA group five of six children (83%) needed antihypertensive or antiprotei- nuric treatment, while only 41% (22/58) of AGA children

received this kind of medication (P=0.07). Comparing the median number of antihypertensive drugs used, we found that SGA children had been treated with more antihyper- tensive or antiproteinuric drugs than had AGA children (SGA 1 (0–3) vs 0 (0–3), P<0.05). Two children in the SGA group and only one child in the AGA group had needed more than two different antihypertensive or anti- proteinuric drugs in the course of the disease (P<0.05).

At the last follow-up, SGA children showed a median SDS for systolic blood pressure of 0.99 (−0.86–3.28) vs −0.23 (−2.13–2.89) in the AGA group (ns). There was a trend towards a higher median SDS for diastolic blood pressure in the SGA group (SGA 1.24 (−0.44–4.37) vs AGA 0.05 (−1.42–2.30), P=0.06). All six SGA patients and 14/56 AGA patients still received antihypertensive or antiproteinuric treatment (P<0.01). The median number of antihypertensive drugs per patient was 1 (0–3) in SGA children and 0 (0–3) in AGA children (P<0.01).

At the last follow-up one of six SGA children presented with GFR according to Schwartz below 70 ml/min per 1.73 m2 BSA. In the AGA group no patient showed chronic renal insufficiency (ns).

Course of nephrotic syndrome in relationship to birth weight SDS

In order to search for effects over the whole spectrum of birth weight, we calculated birth weight SDS corrected for gestational age and gender (Table 4). The patients were divided into four groups with birth weight SDSs<−1.0, −1.0 to 0, 0 to 1.0, and >1.0.

Between the four groups we found no difference in age at manifestation, need for renal biopsy, histological diag-

Table 4 Clinical course of idiopathic nephrotic syndrome in relation to birth weight SDS classification. Data are given as median and range

Parameter Birth weight <−1.0 SDS (n=16)

Birth weight −1.0 to 0 SDS (n=26)

Birth weight 0 to 1.0 SDS (n=15)

Birth weight >1.0 SDS (n=5)

Birth weight SDS −1.42 (−3.91–1.04) −0.41 (−0.97–0.01) 0.34 (0.06–1.0) 1.13 (1.03–1.72)b

Age at manifestation (years) 3.8 (1.5–15.25) 3.4 (1.6–7.2) 4.1 (2.1–15.3) 4.7 (1.2–13.9) Follow-up period (years) 5.2 (1.0–17.3) 7.16 (0.2–14.3) 4.4 (0.9–15.2) 4.3 (2.5–7.8) Cyclophosphamide therapy required 11/16 (68.7%) 14/26 (53.8%) 8/15 (53.3%) 2/5 (40.0%) Cyclosporin A therapy required 9/16 (56.2%) 9/26 (34.6%) 8/15 (53.3%) 2/5 (40.0%) Steroid resistance 5/16 (31.2%) 5/26 (23.8%) 4/15 (26.6%) 2/5 (40.0%) Steroid dependence 8/16 (50.0%) 9/26 (34.6%) 3/15 (20.0%) 2/5 (40.0%) Relapses per patient per year 0.69 (0–2.4) 0.72 (0–3.0) 0 (0–2.2) 0.42 (0–1.2) Patients without relapses 3/16 (18.7%) 5/26 (19.20%) 10/15 (66.6%) 3/5 (60.0%)a

Renal biopsy performed 11/16 (68.7%) 14/26 (53.8%) 10/15 (66.6%) 2/5 (40.0%) Minimal change glomerulopathy (MCGN) 10/11 12/14 5/10 1/2 Focal segmental glomerulosclerosis (FSGS) 1/11 2/14 5/10 1/2

Differences between both groups were tested with one-way analysis of variance (ANOVA) and chi-square test a P<0.01 b P<0.0001

Table 3 Clinical course of idiopathic nephrotic syndrome in SGA and AGA children. Data are given as median and range

Parameter SGA (n=6) AGA (n=56)

Age at manifestation (years) 6.4 (1.9–15.3) 3.7 (1.2–15.5) Follow-up period (years) 6.2 (2.2–17.2) 5.4 (0.2–15.7) Cyclophosphamide therapy required

5/6 (83.3%) 30/56 (53.6%)

Cyclosporin A therapy required 5/6 (83.3%) 23/56 (41.1%) Steroid dependence 2/6 (33.3%) 20/56 (35.5%) Steroid resistance 4/6 (66.6%) 12/56 (21.4%)a

Relapses per patient/year 0.69 (0–2.41) 0.65 (0–3.0) Renal biopsy performed 6/6 (100%) 31/56 (55%) Minimal change glomerulopathy (MCGN)

5/6 (83.3%) 23/31 (74.1%)

Focal segmental glomerulosclerosis (FSGS)

1/6 (16.6%) 8/31 (25.8%)

Differences between both groups were tested with unpaired Mann– Whitney t-test and Fisher’s exact test a P<0.05

Pediatr Nephrol (2007) 22:1881–1889 1885

nosis, or use of immunosuppressive substances; neither was need for antihypertensive treatment, blood pressure at last follow-up, or use of antihypertensive drugs different between the four groups. Median number of relapses per patient year was not different, but interestingly, patients in the birth SDS group 0–1.0 had a bigger chance of facing a relapse-free remission (P<0.01).

Course of nephrotic syndrome in relation to weight gain in the first 24 months of life

In order to look for effects of early weight gain we analysed weight development in the first 24 months of life by the difference between body weight SDS at month 24 and birth weight SDS (Table 5). We defined four groups of weight gain according to the SDS difference: <−1.0, −1.0 to 0, 0 to 1.0, and >1.0. Birth weight SDS in all four groups was not different. Different weight gain over the first 2 years led to a significant increase in body weight SDS and BMI SDS over the four groups, with the lowest in the group having an SDS difference of <−1.0 SDS and the highest in the group having an SDS difference >1.0 SDS.

Between the four groups (Table 6) we could not find differences in age at manifestation, follow-up period, need for renal biopsy, histological diagnosis, or use of immuno- suppressive substances.

Median number of relapses per patient year, need for antihypertensive and antiproteinuric treatment, and blood pressure at last follow-up did not differ between the four groups. Interestingly, there was a high proportion of children with steroid resistance in the group with the lowest weight gain.

Discussion

In our study, we demonstrated additional evidence for an aggravated course of idiopathic nephrotic syndrome in

children born small for gestational age (SGA). The propor- tion of those with steroid resistance was high, and the course of the disease was complicated by arterial hypertension, in comparison with those in the AGA children.

Most retrospective studies on intrauterine growth restric- tion (IUGR) are based on birth auxiology as an approxi- mation for poor intrauterine environment. Of course, birth weight and other birth data give only limited insight and cannot substitute prospective studies that include Doppler ultrasound diagnosis of IUGR prenatally. To exclude maternal factors as far as possible, we tried to collect this information by a questionnaire. Nevertheless, self reporting is of limited value, and we cannot provide data on prenatal maternal nutrition. Furthermore, our knowledge of postna- tal complications was limited in this cohort; therefore, we cannot exclude an influence on later renal diseases. Analysis of renal diseases in children with IUGR cannot easily be done in a prospective manner. Depending on the population, IUGR is exhibited in around 10% to 40% of all births [4]. Nephrotic syndrome is a rare disease, with a prevalence of 16 in 100,000 children [1]. For the combina- tion of nephrotic syndrome and IUGR, the chance may be as low as 1 per 100,000 children. Therefore, retrospective analyses of cohorts of diseased children are mandatory.

The definition of SGA is variable. Most authors use the 10th or below birth weight percentile for identification. In our study we used birth charts of present national origin. For a birth weight SDS of −1.5 (equivalent to the 7th percentile), we could identify six SGA children out of 62 patients (9.7%). This resembles the proportion of SGA children in the healthy population [4]. A proportion of 10% SGA children in cohorts of children with idiopathic nephrotic syndrome was shown in all published papers on IUGR and nephrotic syndrome [14–16, 27].

Zidar et al. [27] were the first to publish findings of an association between an aggravated course of idiopathic nephrotic syndrome and IUGR. In a retrospective analysis of 40 Slovenian children (aged 1–16 years) with MCGN

Differences between both groups are tested with one-way analysis of variance (ANOVA) and chi-square test a P<0.0001

Table 5 Definition of groups according to weight gain in the first 24 months. Weight gain is given as difference between body weight SDS at the age of 24 months and birth weight SDS. In addition, birth

weight SDS, weight SDS at 24 months and BMI SDS at 24 months are given. Data are given as median and range

Parameter Weight gain month 0–24

Weight gain month 0–24

Weight gain month 0–24

Weight gain month 0–24

<−1.0 SDS (n=7) −1.0–0 SDS (n=14) 0–1.0 SDS (n=19) >1.0 SDS (n=13)

Birth weight SDS 0.30 (−1.53 to 1.14) −0.23 (−1.26 to 0.85) −0.42 (−3.91 to 1.31) −0.75 (−1.88 to 0.78) Body weight SDS at 24 months −1.90 (−3.30 to −0.26) −0.66 (−1.48 to 0.67) 0.03 (−3.25 to 1.51) 0.66 (−0.05 to 2.05)a

Difference in body weight SDS month 24−month 0

−1.54 (−3.36 to −1.03) −0.18 (−0.91 to 0.0) 0.41 (0.04 to 0.95) 1.37 (1.07 to 3.06)a

BMI SDS at 24 months −1.58 (−2.95 to −1.22) −0.58 (−1.69 to −0.77) −0.02 (−1.84 to 1.35) 0.34 (−1.23 to 1.37)a

1886 Pediatr Nephrol (2007) 22:1881–1889

clinically suspected or proven by biopsy, the authors identified five children who had been SGA. In the SGA group, a higher number of relapses, a higher rate of steroid dependency, use of cytotoxic drugs, and a higher rate of renal biopsy could be observed. Sheu and Chen [14] studied 50 Taiwanese children (1–13 years) with nephrotic syndrome, half of them with biopsy proven MCGN. Eight children were identified as having been SGA (birth weight below 10th percentile). A higher rate of renal biopsy, higher serum lipids at first manifestation, more steroid depen- dence, higher number of relapses and a high proportion of hypertension in SGA children were shown. In a third paper Na and co-workers [15] analysed the records of 56 Korean children with nephrotic syndrome. In their study steroid resistance was seen significantly more often in the SGA group. Zidar et al. and Na et al. investigated only MCGN patients, Sheu and Chen identified at least one patient with FSGS in their SGA cohort by renal biopsy. We were able to confirm a high rate of renal biopsies in SGA children. In contrast to those studies, we tried to investigate all our patients with the clinical diagnosis of idiopathic nephrotic syndrome and we included patients with FSGS as well. Twenty-eight patients had a biopsy proven MCGN, and nine patients out of 62 (14.5%) had FSGS. The FSGS rate was higher than that reported in the International Study of Kidney Disease in Children (ISKDC), with an FSGS rate of 7.9% [28]. This may explain the high rate of primary and secondary steroid resistance (25%) in our cohort. Never- theless, the rate of steroid resistance was higher in the SGA cohort. This is in line with our finding that the age at manifestation was higher in the SGA group but did not reach statistical significance. A median age of 2.5 years is typical for steroid-responsive patients; patients with prima- ry steroid resistance present at a median age of 6 years [29].

Newer studies observe increasing rates of steroid resistance in childhood nephrotic syndrome [30]. Steroid responsive- ness is an important prognostic parameter of idiopathic nephrotic syndrome [31]. Nevertheless, we could not find patients with renal failure in our study, which can be expected in FSGS patients. A sampling error because the data had been collected at a paediatric nephrology referral centre, and the short observation period, may be explanations for the difference in steroid resistance and renal survival.

In contrast to the published studies, we tried to analyse the influence of postnatal weight gain on the later course of disease. In our small SGA cohort we could not find a clear pattern of catch-up growth, and, therefore, we could not address the influence of catch-up growth of SGA children on the course of nephrotic syndrome in later life in this study. To find a connection between postnatal weight gain and the course of disease independent of birth weight, we tried to analyse the clinical course of nephrotic syndrome in relation to the weight gain shown as SDS difference between birth and 24 months of age. Between the four groups with an SDS difference <−1.0, 1.0–0, 0–1.0 and >1.0, there was no difference in the main aspects of clinical course of nephrotic syndrome. Even though the number of patients was too low for multiple regression analysis, the comparison of the different groups may give an approxima- tion for the missing influence of postnatal percentile-crossing weight gain. This question should be further pursued in a bigger and, as far as possible, prospective study.

Another interesting feature of our study and the cited papers on nephrotic syndrome in former SGA children is the higher rate of hypertension in the SGA cohort. The high rate of hypertension and the need for hypertensive treatment may be confounded by the use of cyclosporine A and recurrent prednisone treatment. Nevertheless, for long-term treatment

Table 6 Clinical course of idiopathic nephrotic syndrome in relation to weight gain in the first 24 months of life. Data are given as median and range

Parameter Weight gain month 0–24

Weight gain month 0–24

Weight gain month 0–24

Weight gain month 0–24

<−1.0 SDS (n=7) −1.0–0 SDS (n=14) 0–1.0 SDS (n=19) >1.0 SDS (n=13)

Age at manifestation (years) 4.5 (1.2–8.0) 3.9 (1.6–7.2) 3.9 (2.1–15.3) 2.9 (1.5–14.8) Follow-up period (years) 5.4 (4.2–17.25) 4.96 (1.7–11.7) 4.8 (0.9–14.2) 5.3 (0.2–15.2) Cyclophosphamide therapy required 5/7 (71.4%) 6/14 (42.8%) 8/19 (42.1%) 9/13 (69.2%) Cyclosporin A therapy required 4/7 (57.5%) 6/14 (42.8%) 4/19 (21.0%) 6/13 (46.1%) Steroid resistance 4/7 (57.1%) 3/14 (21.4%) 3/19 (15.8%) 4/13 (30.7%) Steroid dependence 3/7 (42.9%) 5/14 (35.7%) 4/19 (21.0%) 3/13 (23.0%) Relapses per patient per year 1.0 (0–2.4) 0.56 (0–2.2) 0.57 (0–3.0) 0.7 (0–1.8) Renal biopsy performed 4/7 (57.1%) 9/14 (64.2%) 7/19 (36.0%) 9/13 (69.2%) Minimal change glomerulopathy (MCGN)

2/4 8/9 5/7 6/9

Focal segmental glomerulosclerosis (FSGS)

2/4 1/9 2/7 3/9

Differences between both groups were tested with one-way analysis of variance (ANOVA) and chi-square test

Pediatr Nephrol (2007) 22:1881–1889 1887

with cyclosporin A, the rate of hypertension is 10% [32] or even below [33]. Development of elevated blood pressure is one of the key features of perinatal programming by IUGR [4], even though a meta-analysis of epidemiological studies saw a weaker association between birth weight and later hypertension [34] than initially suspected. Animal studies analysed potential mechanisms of perinatal programming. The main renal phenotype of IUGR is nephron reduction, which could be demonstrated in different animal models of IUGR and SGA [7, 35]. Autopsy studies confirmed the inverse correlation between birth weight and nephron number in humans [36]. Nephron reduction itself is seen as a possible risk factor for the later development of hyperten- sion [37]. Studies in IUGR animals revealed further influences on the development of hypertension, such as salt intake [38, 39] or postnatal nutrition [40]. Nephron deficit, perhaps in combination with other postnatal factors, is a risk for the progression of secondary renal diseases. Zimanyi et al. demonstrated an increased susceptibility to secondary renal injury due to advanced gylcation products in a low protein model of IUGR [41]. Our group showed increased renal damage in anti-Thy1 nephritis as a model of mesangioproliferative glomerulonephritis in a similar IUGR model [42]. Even though studies on possible mechanisms in nephrotic children are still missing, these animal studies support the hypothesis that IUGR is a risk factor for the progression of secondary renal injury such as idiopathic nephrotic syndrome. Therefore, rather than nephron number, altered glomerular inflammatory reaction might be involved in the pathogenesis of nephrotic syndrome.

Analyses of genetic forms of nephrotic syndrome have provided huge knowledge on the structure and function of the slit diaphragma [1, 43]. The pathogenesis of idiopathic nephrotic syndrome is still unclear, although studies of FSGS have suggested the existence of a putative permeability factor [44–46] probably derived from T cells [47]. The relationship between MCGN and primary allergic reaction is still under discussion [1]. More interesting are differences in the phenotype, cytokine profiles and function of lymphocytes of nephrotic patients during relapse in remission [48–51]. In immunological studies of SGA cohorts, a reduced number of T cells, a higher number of CD8-positive cells and delayed hypersensitivity reaction were detectable [52, 53]. Knowl- edge of the immunological consequences of IUGR is limited at the moment. Therefore, since idiopathic nephrotic syn- drome is, at least partly, mediated by T cells, one might speculate that an alteration in T cell response might be involved in the pathogenesis of more severe nephrotic syndrome in children with low birth weight.

In conclusion, we were able to find further evidence for the influence of low birth weight, but not for postnatal weight gain, on the clinical course of secondary renal injury in a cohort of children with idiopathic nephrotic syndrome. The

mechanisms involved are not well understood. Perinatal programming as an additional pathogenic principal in renal disease needs further investigation. Investigation of the role of early catch-up growth should be included in further studies.

Acknowledgements This study was supported by a grant from the Deutsche Forschungsgemeinschaft, Bonn, Germany; SFB 423, Collaborative Research Centre of the German Research Foundation Kidney Injury: Pathogenesis and Regenerative Mechanisms, project B13, to Wolfgang Rascher and Jörg Dötsch, and project Z2 to Kerstin Amann. We thank Elke Wühl for the data for SDS calculation of spontaneous blood pressure measurement in children. We gratefully appreciate the support of Melek Düz in conducting this study.

References

1. Eddy AA, Symons JM (2003) Nephrotic syndrome in childhood. Lancet 362:629–639

2. Abrantes MM, Cardoso LS, Lima EM, Penido Silva JM, Diniz JS, Bambirra EA, Oliveira EA (2006) Predictive factors of chronic kidney disease in primary focal segmental glomerulosclerosis. Pediatr Nephrol 21:1003–1012

3. Barker DJ, Winter PD, Osmond C, Margetts B, Simmonds SJ (1989) Weight in infancy and death from ischaemic heart disease. Lancet 2:577–580

4. McMillen IC, Robinson JS (2005) Developmental origins of the metabolic syndrome: prediction, plasticity, and programming. Physiol Rev 85:571–633

5. Lackland DT, Bendall HE, Osmond C, Egan BM, Barker DJ (2000) Low birth weights contribute to high rates of early-onset chronic renal failure in the southeastern United States. Arch Intern Med 160:1472–1476

6. Lackland DT, Egan BM, Fan ZJ, Syddall HE (2001) Low birth weight contributes to the excess prevalence of end-stage renal disease in African Americans. J Clin Hypertens (Greenwich) 3:29–31

7. Hoy WE, Hughson MD, Bertram JF, Douglas-Denton R, Amann K (2005) Nephron number, hypertension, renal disease, and renal failure. J Am Soc Nephrol 16:2557–2564

8. Tanner JM (1986) Childhood epidemiology. Physical development. Br Med Bull 42:131–138

9. Karlberg J, Albertsson-Wikland K (1995) Growth in full-term small-for-gestational-age infants: from birth to final height. Pediatr Res 38:733–739

10. Hales CN, Ozanne SE (2003) For debate: Fetal and early postnatal growth restriction lead to diabetes, the metabolic syndrome and renal failure. Diabetologia 46:1013–1019

11. Ong KK, Ahmed ML, Emmett PM, Preece MA, Dunger DB (2000) Association between postnatal catch-up growth and obesity in childhood: prospective cohort study. BMJ 320:967–971

12. Hemachandra AH, Howards PP, Furth SL, Klebanoff MA (2007) Birth weight, postnatal growth, and risk for high blood pressure at 7 years of age: results from the Collaborative Perinatal Project. Pediatrics 119:e1264–e1270

13. Min JW, Kong KA, Park BH, Hong JH, Park EA, Cho SJ, Ha EH, Park H (2007) Effect of postnatal catch-up growth on blood pressure in children at 3 years of age. J Hum Hypertens DOI 10.1038/sj.jhh.1002215

14. Sheu JN, Chen JH (2001) Minimal change nephrotic syndrome in children with intrauterine growth retardation. Am J Kidney Dis 37:909–914

15. Na YW, Yang HJ, Choi JH, Yoo KH, Hong YS, Lee JW, Kim SK (2002) Effect of intrauterine growth retardation on the progression of nephrotic syndrome. Am J Nephrol 22:463–467

1888 Pediatr Nephrol (2007) 22:1881–1889

16. Zidar N, Cavic MA, Kenda RB, Koselj M, Ferluga D (1998) Effect of intrauterine growth retardation on the clinical course and prognosis of IgA glomerulonephritis in children. Nephron 79:28–32

17. Gardosi J (2006) New definition of small for gestational age based on fetal growth potential. Horm Res 65 [Suppl 3]:15–18

18. Voigt M, Friese K, Pawlowski P, Schneider R, Wenzlaff P, Wermke K (2001) Analysis of newborns in Germany between 1995 and 1997. Part 6: differences in birth weight classification among states. Geburtshilfe Frauenheilkd 61:700–706

19. Barker DJ, Osmond C, Forsen TJ, Kajantie E, Eriksson JG (2005) Trajectories of growth among children who have coronary events as adults. N Engl J Med 353:1802–1809

20. Cole TJ, Freeman JV, Preece MA (1998) British 1990 growth reference centiles for weight, height, body mass index and head circumference fitted by maximum penalized likelihood. Stat Med 17:407–429

21. Prader A, Largo RH, Molinari L, Issler C (1989) Physical growth of Swiss children from birth to 20 years of age. First Zurich longitudinal study of growth and development. Helv Paediatr Acta Suppl 52:1–125

22. Schwartz GJ, Gauthier B (1985) A simple estimate of glomerular filtration rate in adolescent boys. J Pediatr 106:522–526

23. de Man SA, Andre JL, Bachmann H, Grobbee DE, Ibsen KK, Laaser U, Lippert P, Hofman A (1991) Blood pressure in childhood: pooled findings of six European studies. J Hypertens 9:109–114

24. Ehrich JH, Brodehl J (1993) Long versus standard prednisone therapy for initial treatment of idiopathic nephrotic syndrome in children. Arbeitsgemeinschaft fur Padiatrische Nephrologie. Eur J Pediatr 152:357–361

25. Hodson EM, Craig JC, Willis NS (2005) Evidence-based management of steroid-sensitive nephrotic syndrome. Pediatr Nephrol 20:1523–1530

26. Brodehl J (1981) Alternate-day prednisone is more effective than intermittent prednisone in frequently relapsing nephrotic syndrome. Eur J Pediatr 135:229–237

27. Zidar N, Avgustin Cavic M, Kenda RB, Ferluga D (1998) Unfavorable course of minimal change nephrotic syndrome in children with intrauterine growth retardation. Kidney Int 54:1320–1323

28. The International Study of Kidney Disease in Children (1981) The primary nephrotic syndrome in children. Identification of patients with minimal change nephrotic syndrome from initial response to prednisone. J Pediatr 98:561–564

29. Clark AG, Barratt TM (1999) Steroid responsive nephrotic syndrome. In: Barratt TM, Avner ED, Harmon WE (eds) Pediatric nephrology, 4th edn. Lippincott, Williams and Wilkins, Baltimore pp 731–747

30. Kim JS, Bellew CA, Silverstein DM, Aviles DH, Boineau FG, Vehaskari VM (2005) High incidence of initial and late steroid resistance in childhood nephrotic syndrome. Kidney Int 68:1275–1281

31. Dötsch J, Dittrich K, Plank C, Rascher W (2006) Is tacrolimus for childhood steroid-dependent nephrotic syndrome better than ciclosporin A? Nephrol Dial Transplant 21:1761–1763

32. El-Husseini A, El-Basuony F, Mahmoud I, Sheashaa H, Sabry A, Hassan R, Taha N, Hassan N, Sayed-Ahmad N, Sobh M (2005) Long-term effects of cyclosporine in children with idiopathic nephrotic syndrome: a single-centre experience. Nephrol Dial Transplant 20:2433–2438

33. Ponticelli C, Rizzoni G, Edefonti A, Altieri P, Rivolta E, Rinaldi S, Ghio L, Lusvarghi E, Gusmano R, Locatelli F, Pasquali S, Castellani A, Della Casa-Alberighi O (1993) A randomized trial of cyclosporine in steroid-resistant idiopathic nephrotic syndrome. Kidney Int 43:1377–1384

34. Huxley R, Neil A, Collins R (2002) Unravelling the fetal origins hypothesis: is there really an inverse association between birth- weight and subsequent blood pressure? Lancet 360:659–665

35. Amann K, Plank C, Dötsch J (2004) Low nephron number—a new cardiovascular risk factor in children? Pediatr Nephrol 19:1319–1323

36. Hughson M, Farris AB 3rd, Douglas-Denton R, Hoy WE, Bertram JF (2003) Glomerular number and size in autopsy kidneys: the relationship to birth weight. Kidney Int 63:2113–2122

37. Keller G, Zimmer G, Mall G, Ritz E, Amann K (2003) Nephron number in patients with primary hypertension. N Engl J Med 348:101–108

38. Manning J, Vehaskari VM (2005) Postnatal modulation of prenatally programmed hypertension by dietary Na and ACE inhibition. Am J Physiol Regul Integr Comp Physiol 288:R80–R84

39. Woods LL, Weeks DA, Rasch R (2004) Programming of adult blood pressure by maternal protein restriction: role of nephro- genesis. Kidney Int 65:1339–1348

40. Hoppe CC, Evans RG, Moritz KM, Cullen-McEwen LA, Fitzgerald SM, Dowling J, Bertram JF (2007) Combined prenatal and postnatal protein restriction influences adult kidney structure, function, and arterial pressure. Am J Physiol Regul Integr Comp Physiol 292:R462–R469

41. Zimanyi MA, Denton KM, Forbes JM, Thallas-Bonke V, Thomas MC, Poon F, Black MJ (2006) A developmental nephron deficit in rats is associated with increased susceptibility to a secondary renal injury due to advanced glycation end- products. Diabetologia 49:801–810

42. Plank C, Östreicher I, Hartner A, Marek I, Struwe FG, Amann K, Hilgers KF, Rascher W, Dötsch J (2006) Intrauterine growth retardation aggravates the course of acute mesangioproliferative glomerulonephritis in the rat. Kidney Int 70:1974–1982

43. Niaudet P (2004) Genetic forms of nephrotic syndrome. Pediatr Nephrol 19:1313–1318

44. Savin VJ, Sharma R, Sharma M, McCarthy ET, Swan SK, Ellis E, Lovell H, Warady B, Gunwar S, Chonko AM, Artero M, Vincenti F (1996) Circulating factor associated with increased glomerular permeability to albumin in recurrent focal segmental glomerulo- sclerosis. N Engl J Med 334:878–883

45. Kemper MJ, Wolf G, Muller-Wiefel DE (2001) Transmission of glomerular permeability factor from a mother to her child. N Engl J Med 344:386–387

46. Carraro M, Caridi G, Bruschi M, Artero M, Bertelli R, Zennaro C, Musante L, Candiano G, Perfumo F, Ghiggeri GM (2002) Serum glomerular permeability activity in patients with podocin muta- tions (NPHS2) and steroid-resistant nephrotic syndrome. J Am Soc Nephrol 13:1946–1952

47. Koyama A, Fujisaki M, Kobayashi M, Igarashi M, Narita M (1991) A glomerular permeability factor produced by human T cell hybridomas. Kidney Int 40:453–460

48. Yan K, Nakahara K, Awa S, Nishibori Y, Nakajima N, Kataoka S, Maeda M, Watanabe T, Matsushima S, Watanabe N (1998) The increase of memory T cell subsets in children with idiopathic nephrotic syndrome. Nephron 79:274–278

49. Topaloglu R, Saatci U, Arikan M, Canpinar H, Bakkaloglu A, Kansu E (1994) T-cell subsets, interleukin-2 receptor expression and production of interleukin-2 in minimal change nephrotic syndrome. Pediatr Nephrol 8:649–652

50. Tomizawa S, Suzuki S, Oguri M, Kuroume T (1979) Studies of T lymphocyte function and inhibitory factors in minimal change nephrotic syndrome. Nephron 24:179–182

51. Cunard R, Kelly CJ (2002) T cells and minimal change disease. J Am Soc Nephrol 13:1409–1411

52. Chandra RK, Ali SK, Kutty KM, Chandra S (1977) Thymus- dependent lymphocytes and delayed hypersensitivity in low birth weight infants. Biol Neonate 31:15–18

53. Chatrath R, Saili A, Jain M, Dutta AK (1997) Immune status of full-term small-for-gestational age neonates in India. J Trop Pediatr 43:345–348

Pediatr Nephrol (2007) 22:1881–1889 1889

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.