11

Pediatr Nephrol (2016) 31:595 Pediatr Nephrol (2016) 31:595

604 10.1007/s00467-015-3285-1
DOI 604

ORIGINAL ARTICLE

Body weight-based prednisolone versus body surface area-based

prednisolone regimen for induction of remission in children
with nephrotic syndrome: a randomized, open-label, equivalence
clinical trial
1 1 2
Vaishnavi Raman & Sriram Krishnamurthy & K. T. Harichandrakumar

Received: 23 September 2015 / Revised: 23 November 2015 / Accepted: 30 November 2015 / Published online: 12 January 2016
# IPNA 2016
Abstract
Background Body surface area (BSA)-based prednisolone
dosing for childhood nephrotic syndrome (NS) leads to
higher cumulative prednisolone doses than body weight
(BW)-based dosing. The clinical effects of this higher
dosage have not been evaluated in prospective studies.
Methods This parallel-group open-label randomized
clinical
trial enrolled 100 children with idiopathic NS, to receive
BW- based (n = 50) or BSA-based (n = 50) prednisolone
dosing by block randomization in a 1:1 ratio. The time
taken for remis- sion, relapse rate per 6 months, and
adverse effects of steroids were analyzed in both groups.
Results There was no significant difference in the time
taken
for remission in the BW group versus the BSA group
(median (IQR) 7 (4.59) versus 5.5 (48) days; p = 0.082);
similar re- sults were observed on subgroup analysis in
new-onset and infrequently-relapsing NS (IFRNS). The
cumulative prednis- olone dosage during the enrolment
episode was higher in the BSA group. The incidence of
hypertension was higher (p = 0.048) in the BSA group
on per-protocol analysis. The relapse rates in the two
groups per 6 months on follow-up were comparable.
Electronic supplementary material The online version of this
article (doi:10.1007/s00467-015-3285-1) contains supplementary
material, which is available to authorized users
Conclusions Clinical outcomes with BW-based dosing are
equivalent to BSA dosing-related outcomes, although
cumu- lative prednisolone doses are lower in the former.
The practice of BW-based calculations for prescribing
prednisolone in NS is a reasonable approach.
Keywords Nephrotic syndrome . Children . Prednisolone
. Body weight . Body surface area
Introduction
Idiopathic nephrotic syndrome (NS), a common chronic
renal disorder encountered in children, responds well to
corticoste- roid therapy with remission of proteinuria. The
dosing of pred- nisolone in NS has been a matter of
controversy. A dose of
2
60 mg/m body surface area (BSA) per day of
prednisolone
was recommended arbitrarily by the International Study of
Kidney Disease in Children (ISKDC) for the therapy of NS
[1] and has been followed in several studies [28].
However, several other studies (including Indian guidelines
on the man- agement of steroid-sensitive NS) recommend 2
mg/kg body
weight (BW) per day of prednisolone as the initial starting
2
Department of Biostatistics, Jawaharlal Institute of Postgraduate
Medical Education and Research, (JIPMER), Pondicherry 605006,
India

* Sriram Krishnamurthy
drsriramk@yahoo.co
m

1
Department of Pediatrics, Jawaharlal Institute of Postgraduate
Medical Education and Research (JIPMER), Pondicherry
605006, India
 22
Pediatr Nephrol (2016) 31:595
dosage [912]. Some authors recommend either BW-
604
based prednisolone or BSA-based prednisolone regimens,
assuming them to be equivalent [13]. A recent study by
Feber et al. [14] has questioned such an assumption,
demonstrating that the dosages are not equivalent as per
cumulative dosing (especial- ly when BW is <30 kg
and/or the total prednisolone dose is
<60 mg/day), and recommendingPediatr Nephrol (2016) 31:595
further studies to assess
604
the clinical significance and outcome of children in
response to both dosages in terms of time taken for
remission, number of relapses, and complications due to
steroid therapy. A retrospective study by Saadeh et al.
[15] has described the clinical effects of BW-based
Brelative under-dosing^
of prednisolone. While such under-dosing did not influ-
ence the initial response; the number of relapses in the
initial 6 months appeared to be higher in this Brelatively
under-dosed^ group, thereby arguing for a BSA-based
prednisolone regimen for the initial and relapse manage-
ment of NS [15, 16]. They also recommended future
prospective studies directly comparing outcomes of
childhood NS with BW-based or BSA-based dosing to
confirm the findings of their study.
Therefore, concerns have been expressed that the
Brelative under-dosing^ due to BW-based prednisolone
dosing could lead to suboptimal therapeutic effects
compared with BSA- based prednisolone dosing, which
could in turn affect the clin- ical course of NS [14, 15]. In
view of this, we conducted the present randomized clinical
trial to analyze the clinical signif- icance of BW-based
prednisolone versus BSA-based prednis- olone regimens for
the induction of remission in children with NS. We
hypothesized that BSA-based dosing might be equiv- alent
to BW-based dosing for the induction of remission in
childhood NS.
Materials and methods
This randomized parallel group open-label equivalence
clini- cal trial was conducted at the pediatric nephrology
outpatient clinic of a tertiary hospital in south India from
March 2014 through July 2015 after obtaining approval
from the Institute Ethics Committee (Approval No.
JIP/IEC/2014/1/249). The trial recruited 100 children,
allocating the enrolled subjects into the two treatment
groups (BW- and BSA-based prednis- olone dosing
respectively) in a ratio of 1:1. The trial was registered in
the Clinical Trial Registry of India (CTRI regis- tration
No.CTRI/2014/04/004541). Informed consent was
obtained from the parents before subjects were included
in the study.
Objectives
The primary objective was to compare the time taken
for remission between children receiving a BW-based
prednisolone regimen versus those receiving a BSA-
based prednisolone regimen. The secondary objectives
were to compare the number of relapses, the adverse
effects of steroids (hypertension [systolic or diastolic
blood pressure >95th centile for age, sex, and height]
[17], Cushingoid features, serious infections, or ophthal-
mic complications), and the cumulative dose of steroids
between BW-based prednisolone and BSA-based pred-
nisolone regimens in children with NS and in different
clinical subtypes of NS after a follow-up period of
6 months.
Inclusion and exclusion criteria
Children aged 118 years with idiopathic NS were included
in the study. Children with steroid-resistant NS (SRNS),
second- ary NS (e.g. lupus nephritis, IgA nephropathy,
Henoch Schnlein purpura nephritis), children with active
infections, those already receiving daily/alternate-day
steroids or alterna- tive immunosuppressants, and children
2
with BW >30 kg and/ or BSA >1 m were excluded from
the study. Children who had already enrolled once in the
study were not enrolled dur- ing a further relapse.
Children with BW >30 kg and/or BSA >1 m2 were
exclud-
ed since the maximum dose of prednisolone per day had to
be restricted to 60 mg/day as per the guidelines formulated
by the Indian Pediatric Nephrology Group, Indian
Academy of Pe- diatrics [12]. Infants with NS were not
included in the study as they have a high chance of being
positive for genetic muta- tions, an increased likelihood of
steroid non-responsiveness, and would require renal biopsy
[12, 18]. As it is not uncom- mon to encounter
malnourished adolescents at our center, we proposed to
recruit all children aged 118 years with NS, if BW <30 kg
2
and/or BSA <1 m were noted.
Randomization, allocation concealment, and
blinding
Block randomization using 20 blocks of two block sizes
(4 and 6) was generated using random allocation software
version 2.0 (Informer Technologies, Inc.) to allocate the
enrolled subjects into one of two groups (BW-based or
BSA-based prednisolone regimen) in an allocation ratio
of 1:1 by a person not directly involved with data collec-
tion, analysis or interpretation. This randomization list
was concealed from the investigators carrying out the
study. The children were not stratified at trial entry into
different clinical subgroups of NS or by age. Allocation
was concealed placing individual assignments (folded
twice) in serially numbered, sealed opaque envelopes by
a person not involved in the trial. Patients were assessed
for eligibility and enrolled sequentially by the first and
second authors. The envelopes were opened by the first
author only after obtaining informed written consent to
enrolment, and the subjects were thereupon assigned to
the allocated interventions by the first author. The care-
givers (first and second authors) could not be blinded as
they had to check the dose of prednisolone at each visit.
However, the statistician was blinded to the assigned
interventions until initial analysis and preparation of the
first draft.
Prednisolone dosing in the two study groups
Nephrotic syndrome was diagnosed, classified, and
managed as per guidelines formulated by the Indian
Pediatric
59
Pediatr Nephrol (2016) 31:595
75
604
Nephrology Group, Indian Academy of Pediatrics [12].
After confirming the diagnosis of new-onset NS or relapse
[12], patients sequentially presenting to the pediatric
nephrology outpatient clinic were randomized into the two
study groups. The weight at enrolment (irrespective of the
presence or ab- sence of edema) and height at enrolment
were chosen for initial prednisolone dosing in both study
groups. The weight of the child was recorded with an
electronic weighing machine with a precision of 10 g. The
height was recorded with the help of a manual stadiometer
with a precision of 0.1 cm. The BSA was subsequently
calculated using Mostellers formula [19], because of its
relative simplicity and proven equivalence to other BSA
formulae [14]. Medication adherence was enforced by
pill count and a diary maintained by the parents for the
same.
1. Group ABW-based prednisolone regimen group: in
this treatment arm, new-onset NS was treated with
predniso- lone 2 mg/kg/day daily (maximum 60 mg) in
two divided doses for 6 weeks, followed by 1.5 mg/kg
on alternate days (maximum 40 mg) as a single dose in
the morning for 6 weeks. Relapses were treated with
prednisolone
2 mg/kg/day daily (maximum 60 mg) in two divided
doses until remission, followed by 1.5 mg/kg on
alternate days (maximum 40 mg) as a single dose in
the morning for 4 weeks [12].
2. Group BBSA-based prednisolone regimen group: in
this treatment arm, new-onset NS was treated with
2
pred- nisolone 60 mg/m /day daily (maximum 60 mg)
in two divided doses for 6 weeks, followed by 40
2
mg/m on alternate days (maximum 40 mg) as a single
dose in the morning for 6 weeks. Relapses were treated
2
with prednis- olone 60 mg/m /day daily (maximum 60
mg) in two divided doses until remission, followed by
2
40 mg/m on alternate days (maximum 40 mg) as a
single dose in the morning for 4 weeks [12].
If the enrolled subject did not take the dose of pred-
nisolone in the morning, they were advised to take the
tablet by the evening to ensure the correct cumulative
dose per day. If the dose was missed throughout the
day, it was planned to be considered a protocol viola-
tion. Similarly, taking incorrect doses of prednisolone
was also considered a protocol violation. Such patients
were excluded from per protocol analysis.
The dose of prednisolone was recalculated based on
the weight and/or height recorded at each follow-up visit.
The dose of prednisolone calculated according to either
BW or BSA was rounded off to the nearest multiple of 2.5
[20]. Pred- nisolone was administered in the form of
tablets (minimal strength 5 mg), which could be broken
into half, if required. No changes were made to the trial
design after the trial commenced.
Pediatr Nephrol (2016) 31:59559
604 75
Laboratory investigations
The following investigations were performed in both
dosing groups at enrolment: urine albumin, serum albu-
min, serum creatinine, and serum cholesterol. Age, sex,
blood pressure, height, weight, body mass index, and
eye examination for baseline status were noted. Data
regarding the clinical and biochemical profile were
recorded on a structured proforma.
Def initions for outcome
assessment
1. Time taken for remission: parents were taught to carry
out a dipstick test on a freshly voided morning urine
sample and maintain a diary until urine protein was nil
for 3 consecutive days (which constituted remission).
The time taken for remission was taken as the time
from the start of steroid treatment to the day before the
first day of docu- mentation of nil proteinuria (if 3
consecutive days, nil proteinuria was documented).
2. Cumulative prednisolone dosage for the enrolment epi-
sode was expressed as total mg/kg of prednisolone
calcu- lated as per the group they were assigned to.
3. Cumulative prednisolone during the observation period
(cumulative prednisolone received by the end of the 6-
month follow-up) was calculated as the cumulative
dose of prednisolone for the enrolment episode plus the
cumu- lative dose of prednisolone for each relapse plus
the cu- mulative maintenance dose of prednisolone (in
case of a frequently-relapsing course) during the study
period.
4. Number of relapses occurring during the 6-month
follow- up period were recorded and grouped as nil,
<2 or 2 relapses for analysis.
5. Adverse effects of steroids were assessed by
Cushingoid
features, hypertension, ophthalmic changes, and/or
seri- ous infections.
a) Cushingoid features, as assessed by clinical
examina- tion, were defined by the presence of
moon facies, truncal obesity or striae. (This was
limited by the sub- jectivity of the assessment).
b) Hypertension was defined as systolic or diastolic
blood pressure > 95th centile for age, sex, and
height [17]. Blood pressure recordings were taken
at each visit to the pediatric nephrology clinic using
a manual mercury manometer. Only random blood
pressure re- cordings were performed owing to
logistical con- straints. If three hospital readings
were >95th centile for age, sex, and height,
antihypertensive medications were initiated.
c) Serious infections included spontaneous bacterial
peritonitis, pneumonitis, cellulitis, intracranial
infec- tions, or exanthematous illnesses (e.g.
varicella).
 d) Steroid-induced eye changes that were assessed
in- cluded cataract, glaucoma, and hypertensive
retinop- athy. Ophthalmic examination was carried
out rou- tinely at baseline, at the end of steroid
therapy, and at the end of the follow-up period by
the ophthalmol- ogist using slit lamp examination,
applanation tonom- etry, fundoscopy, and visual
field examination by perimetry, as applicable.
e) Changes in weight as assessed by the change in
body mass index z score (Delta BMI z score)
(www.who. int/growthref/tools/en) at the end of
steroid therapy for the enrolment episode.
Follow-up
Parents were taught to perform and grade urine albumin with
a dipstick (once a day, in a freshly voided early morning
urine specimen) and to maintain a diary at home. In the
event of 3+ or 4+ for 3 consecutive days (which
constituted a relapse), subjects were asked to contact the
hospital. Children were followed up weekly until remission
and at 4-weekly intervals thereafter for a period of 6
months. During these visits, urine albumin was tested by
dipstick. Subsequent relapses were treated with the same
dosage regimen. Number of relapses, cumulative
prednisolone received, and adverse effects of ste- roids in
the two groups were noted and compared. No changes were
made to trial outcomes after the trial had commenced.
Sample size
Since there are no previous studies describing the
equivalence limits in time taken for remission between
children treated with BW- versus BSA-based prednisolone
regimens, we as- sumed that an equivalence limit of 2 days
in the time taken for remission in the two groups would be
clinically significant. A sample size of 42 subjects in each
group would have a 90 % statistical power to detect a
significant difference in mean number of days required for
remission (between the two groups) of 2 days, with a
standard deviation of 2.8 days [11], and a two-tailed test,
with p value of 0.05. Assuming an attrition of 20 % (on
account of non-compliance, loss to follow-up, steroid
resistance or mortality), a total of 100 chil- dren were
recruited to the study.
Statistical analysis
Data were analyzed to see whether they followed normal
dis- tribution by using KolmogorovSmirnov test of
normality, and expressed as mean SD or median and
inter-quartile range, as appropriate. Students t test was
used to compare continuous variables and proportions
were compared using the Chi-squared test or Fishers exact
test. Median and inter-
quartile ranges were compared using the non-parametric
MannWhitney U test. A BlandAltman plot was used to
compare the agreement between BSA- and BW-based
doses. KaplanMeier survival analysis followed by the
TaroneWare test was used for comparing the time taken
for remission and duration of remission in both groups.
Correlation between different factors (age, age at onset of
NS, and sex) and time taken for remission was calculated
using Spearmans correla- tion. A p value <0.05 was
considered significant. Effect sizes (relative risk and
absolute risk differences) were calculated using the
Practical Meta-Analysis Effect Size Calculator, George
Mason University (accessible at ht t p:/ / w w w.
campbellcollaboration.org/resources/effect_size_input.php).
Hazard ratios were calculated using the Cox proportional
hazards model. Data were evaluated using SPSS version
20.
0 (SPSS, Chicago, IL, USA). Intention to treat and per
protocol analysis were performed to study the outcome
variables. Interim analysis was planned under monitoring
of the Institute Ethics Committee, 9 months after the start
of the study. Subgroup analysis was exploratory and not pre-
defined; outcome variables such as time taken for remission,
number of relapses over a 6-month follow-up, cumulative
dose of pred- nisolone, and adverse effects of steroids for
children with new-onset NS and infrequently-relapsing NS
(IFRNS) were analyzed separately.
Results
Children were recruited and followed up from March 2014
through July 2015. One hundred and 25 children with NS
were assessed for eligibility. Twenty-five children were
ex- cluded (reasons shown in the CONSORT diagram, Fig.
1). One hundred children were randomized to receive the
inter- ventions: either BW-based prednisolone dosing
(group A) or BSA-based prednisolone dosing (group B),
with 50 in each group. Two children did not report after
randomization in group A, while one did not report after
randomization in group B; the parents were unavailable for
further contact. Therefore,
97 children received the intervention (48 in group A and 49
in group B). The two groups were comparable in terms of
their baseline clinical and laboratory characteristics (Table
1). Of the 100 children enrolled in the study, 83 subjects
were either new-onset NS or IFRNS cases. Two children
took incorrect doses of prednisolone and are mentioned in
the CONSORT diagram (Fig. 1). None of the children
failed to take the tab- lets. Hypertension was present at the
onset of the episode in
16 % of the enrolled children (14 % in group A and 18 %
in group B). None of these children had persistent
hypertension. They were normotensive within 1 week of
antihypertensive treatment (amlodipine), which could be
tapered and stopped within 1 month.
Fig. 1 CONSORT diagram depicting the inflow of patients into the study. SRNS steroid resistant nephrotic syndrome, BSA body surface area,
SBP
spontaneous bacterial peritonitis
An interim analysis was performed in November 2014,
and the data and analysis were monitored by two
independent ob- servers nominated by the Institute Ethics
Committee, who rec- ommended continuing the study until
the required sample size was attained. The randomization list
was generated for 100 pa- tients, and the trial was stopped
after recruitment of the hundredth subject as per the
directions of the Institutes ethics committee.
Table 2 depicts the analysis comparing the outcomes of
the two groups. The primary outcome variable, the time
taken for remission, was not found to be significantly
different in the
two groups (median time taken for remission 7 vs 5.5
days; p = 0.082). Cumulative dose of prednisolone for the
enrolment episode was significantly different in the two
groups (35 vs
40 mg/kg; p = 0.042) with a mean difference of 5 mg/kg.
The
cumulative dose of prednisolone at the end of the 6-month
follow-up period (cumulative prednisolone during the
observa- tional period) was not significantly different in the
two groups.
Intention to treat analysis was carried out by assuming
the worst outcome for adverse effects and relapse rate
per
6 months. Adverse effect profiles of the two groups were
not
60
Pediatr Nephrol (2016) 31:595 Pediatr Nephrol (2016) 31:59560
06
604 604 06
Table 1 Baseline characteristics in the study subjects

Characteristics Group A (BW-based; n = 50 Group B (BSA-based; n = 50 p value

Age at enrolment (years) 5.4 2.9 5.2 2.5 0.796

Age at onset of NS (years) 3.9 2.7 3.4 2.2 0.268
Males 27 (54 %) 27 (54 %) 1.000
Weight (kg) 17.5 5.7 16.9 4.9 0.604
Height (cm) 104.7 17.3 103.9 15.8 0.829
2
Body mass index (kg/m ) 15.7 1.3 15.4 1.3 0.331
Body surface area (m2) 0.7 0.2 0.7 0.1 0.685
Hypertension (systolic or diastolic BP > 95th centile) 7 (14) 9 (18) 0.616
Systolic blood pressure (mmHg) 93 10 94 9 0.779
50th90th centile 46 (92) 44 (88) 0.761
90th95th centile 2 (4) 1 (2)
>95th centile 2(4) 5 (10)
Diastolic blood pressure (mmHg) 59 8 60 9 0.787
50th90th centile 38 (76) 39 (78) 0.522
90th95th centile 6 (12) 3 (6)
>95th centile 6 (12) 8 (16)
Clinical phenotype
New onset NS 19 (38) 12 (24) 0.488
IFRNS 24 (48) 28 (56)
FRNS 7 (14) 10 (20)
Number of relapses before enrolment 2 (013) 2 (012) 0.654
Serum creatinine (mg/dL) 0.53 0.11 0.55 0.12 0.415
Serum albumin (g/dL) 1.89 0.54 2.01 0.57 0.258
Serum cholesterol (mg/dL) 421 114 413 103 0.727

All values are expressed as mean SD or n (%)

FRNS frequently relapsing nephrotic syndrome, IFRNS infrequently relapsing nephrotic syndrome, NS nephrotic syndrome, BSA body surface area,
BP
blood pressure
significantly different on intention-to-treat analysis (Table
2). However, on per protocol analysis, there was a higher
propor- tion of hypertension occurrence in the BSA group
(p = 0.048). Of the 4 children who developed hypertension
after steroids in the BSA group, 1 child developed
hypertensive encephalopa- thy and required labetalol
infusion, although subsequently, hypertension was under
control and anti-hypertensive drugs could be tapered and
stopped within 3 months. The other 3 children also had
only transient hypertension. Other adverse effects, such as
Cushingoid features, serious infections, steroid-induced
ophthalmic complications, and weight chang- es were not
significantly different in the two groups. The number of
relapses during the 6-month follow-up were com- parable
in the two groups (p = 0.974). The survival analysis
comparing the time taken for remission (Fig. 2) was not
sig- nificantly different in the two groups.
Supplementary Table 3 depicts the subset analysis
compar- ing the outcomes between the two groups among
new-onset NS. There was no significant difference in time
taken for remission between both the groups (median
[IQR]: 7.5 [69] vs 5 [48]; p = 0.154). Duration of
remission, number of
relapses during follow-up, and adverse effect profile of the
two groups were comparable in this subgroup. However,
there was a significant difference in the mean cumulative
dose of prednisolone for the enrolment episode and
cumulative pred- nisolone received during the observation
period. Similar sub- set analysis in children with IFRNS
(Supplementary Table 4) showed that there was no
significant difference in time taken for remission (median
(IQR): 6.5 (4.59) vs 6 (4.59); p = 0.743), duration of
remission, number of relapses during follow-up, adverse
effect profile, and cumulative dose of pred- nisolone during
the observational period between the two groups. There
was a significant difference in the mean cumu- lative dose
of prednisolone for the enrolment episode between group A
and group B among the IFRNS patients (p = 0.014).
Correlation between age at enrolment, age of onset
and sex versus time taken for remission was analyzed.
There was a weak negative correlation between the age
of children at enrolment and the time taken for remis-
sion (Spearmans rho 0.308; (p = 0.003). There was no
significant correlation among sex, age of onset, and
time taken for remission.
 Table 2 Comparison of outcome variables in the two study groups

Characteristics Group A (body Group B (body Effect size Absolute risk p value
weight-based) surface area- (95%CI) difference (95 %
based) CI)
Intention to treat analysis
a e
Time taken for remission (days) (n = 50, 50) 7 (4.59) 5.5 (48) 0.7 (0.46, 1.05) 0.05 (0.45, 0.34) 0.02*
Cumulative prednisolone for enrolment episode 35 (29115) 40 (3459) 0.05 (0.48, 0.37) 0.042**
(mg/kg; n = 50, 50)
Cumulative prednisolone during observation 81(30115) 96 (36130) 0.14 (0.57, 0.28) 0.462
(mg/kg; n = 50, 50)
period
e
Duration of remission from enrolment during the 24 (1929) 20 (1326) 1.05 (0.61,1.82) 0.15 (0.28, 0.57) 0.804*
a
observation period (weeks) (n = 50,
50) Relapse in 6 months (n = 50, 50)
None 22 (44) 21 (42) 0.07 (0.1, 0.03) 0.974
<2 12 (24) 12 (24)
2 16 (32) 23 (36)
f
Adverse effects of steroid therapy (n = 50, 50) 22 (44) 23 (46) 0.96 (0.62,1.48) 0.04 (0.48, 0.39) 0.841
Per protocol analysis
Time taken for remission (days; n = 44, 49) 6.5 (4.58.5) 5.5 (48) 0.09 (0.49, 0.3) 0.261
(mg/kg; n = 44, 49) 99 (56115) 111 (64133) 0.08 (0.35, 0.51) 0.613
Cumulative prednisolone for enrolment episode 38 (31115) 40 (3459) 0.17 (0.24, 0.58) 0.190
Cumulative prednisolone during observation
period 24 (1024) 20 (1224) 0.15 (0.28, 0.57) 0.977
(mg/kg/6 m; n = 41, 43)
Duration of remission from enrolment
during observation period (weeks; n = 41,
43)
Relapses in 6 months (n = 41, 43)
 None 22 (52) 21 (49) 0.10 (0.14, 0.05) 0.791
<2 11 (27) 12 (28)
2 8 (19) 10 (23)
b f
Adverse effects of steroid therapy (n = 41, 43) 15 17 0.93 (0.54, 1.59) 0.07 (0.55, 0.42) 0.781
f
Cushingoid features (n = 41, 43) 12 10 1.25 (0.61, 2.59) 0.17 (0.36, 0.71) 0.576
f
Serious infections (n = 41, 43) 3c
4c 0.78 (0.18, 3.3) 0.14 (1.0, 0.71) 0.717
f
Eye changes (n = 41, 43) 1d
2d 0.52 (0.05, 5.56) 0.36 (1.71, 0.97) 0.584
Hypertension (n = 41, 43) 0 4 0.048**
Systolic blood pressure (n = 41, 43)
f
50th90th centile 40 38 1.95 (0.95, 1.16) 0.91 (0.29, 2.12) 0.135
90th95th centile 1 1 1.05 (0.07, 16.21)f 0.03 (-1.52, 1.57)
>95th centile 0 4
Diastolic blood pressure (n = 41, 43)
50th90th centile 38 35 1.14 (0.96, 1.34)f 0.59 (0.19, 1.36) 0.121
f
90th95th centile 3 4 0.79 (0.19, 3.3) 0.14 (1.0, 0.71)
> 95th centile 0 4 - -
Delta body mass index z score (n = 44, 49) 0.25 0.86 0.001 0.84 0.29 (0.70, 0.11) 0.166

All values are expressed as median (IQR), n (%), n or mean SD

p values were generated using the MannWhitney U test for median (IQR), Students t test for mean SD, and Chi-squared test for proportions
CI confidence interval, IQR interquartile range, SD standard deviation
*By TaroneWare test
**p value is significant
a
By KaplanMeier survival analysis
b
Four children had 2 adverse effects
c
One case of varicella and 2 of peritonitis in group A, and 2 cases of cellulitis and 2 cases of pneumonitis in group B
d
One case of cataract in group A; 1 of glaucoma and 1 of hypertensive retinopathy in group B
e
Hazard ratio (95 % CI)
f
Relative risk (95 % CI)
and their respective BSA dose per day. The mean theoretical
dose difference (mg/day) was 7.35 2 (p < 0.001). Bland
Altman agreement analysis was carried out between
average daily dose of prednisolone (average of BW- and
BSA-based dose) and the ratio of BW-based prednisolone
and BSA-based pred- nisolone, as depicted in Fig. 3. The
curve shows that the ratio

Fig. 2 Survival analysis curve comparing time taken for remission in

the two groups. BSA body surface area. The KaplanMeier survival
function with TaroneWare test comparing the probability of
remission within
28 days in the two groups. The median survival time taken for
remission in group A is 7 days compared with 5.5 days in group B,
which was not statistically significant (p = 0.028)

The theoretical dose difference was calculated for the

chil- dren from the difference between their calculated BW

was lower when the average dose is less, and approaches 1
when the dose becomes 60 mg.
Discussion
This randomized clinical trial examined the clinical effects
of dose discrepancies between BSA- and BW-based dosing
regimens. Our study found that there was no significant
difference in the time taken for remission between the
two groups. There was higher cumulative prednisolone
dosing in the BSA-based regimen group (median cumula-
tive dose difference 5 mg/kg); with a significant difference
in the cumulative dosage of prednisolone in children with
new-onset NS and IFRNS (mean difference 29 mg/kg and
8 mg/kg respectively). These two subgroups accounted for
83 % of the subjects recruited. Other secondary outcome
variables such as cumulative dosage of prednisolone re-
ceived by the end of the 6-month follow-up period, dura-
tion of remission, and number of relapses in the subsequent
6-month follow-up period were not found to be significant-
ly different in the two groups. Overall, there was no sig-
nificant difference in the adverse effects of steroids be-
tween the two groups, except for an increased number of
children with hypertension in the BSA-based group (on per
protocol analysis).
Saadeh et al. [15] and Feber et al. [14] documented a
Brelative under-dosing^ in a BW-based prednisolone
regimen when compared with BSA. The argument put
forward in favor of a BSA-based prednisolone dosing
regimen for initial and relapse treatment of NS [15, 16] is
that this Bunder-dosing^ leads to a sub-optimal dose that is
not sufficient to achieve the desired therapeutic effects.
Although the data of Saadeh et al. [15] did suggest that
BW-based dosing resulted in a sub-

Fig. 3 BlandAltman curve

comparing the theoretical ratio of
the body weight-based dose: body
surface area dose in an individual
patient versus the average dose of
prednisolone. BSA body surface
area, SD standard deviation. The
BlandAltman curve comparing
the agreement between body
weight and BSA dose. The
average dose is plotted against the
ratio of body weight to BSA dose.
The curve demonstrates that the
ratio of body weight:BSA is
lower when the average dose of
prednisolone per day is less.
Upper and lower limit of
agreements represent +2 SD and
2 SD respectively
optimal clinical effect, the study was small, retrospective,
and not a randomized controlled trial. Hence, it was
subject to bias.
It is notable that the clinical characteristics of the
first episode of NS are different from those of relapses.
For example, water overload during the first episode of
NS is often higher. This leads to a discrepancy between
the Breal^ (edematous) weight and the Bappropriate^
(dry) weight. The situation is different in a relapse as
steroid therapy is often initiated before the occurrence
of marked edema (based on 3+ or 4+ proteinuria being
detected by parents). Thus, prednisolone dosing accord-
ing to Breal^ weight corresponds to a dose higher than
2 mg/kg of Bappropriate^ weight in the first episode of
NS. These differences may explain why in the study by
Saadeh et al. [15], BW-based Bunder-dosing^ did not
influence the initial response, but only further along
the clinical course.
The reason for the time taken to remission not being
affect- ed by the higher doses provided by BSA-based
prednisolone regimens, as observed in this study, could be
related to mech- anisms other than immunosuppressive
actions of glucocorti- coids (which are known to be dose-
dependent). Mehls and Hoyer [16] argued that the doses
required for stabilization of the podocyte cytoskeleton may
actually be lower than those needed for
immunosuppression. Glucocorticoids stabilize the podocyte
cytoskeleton by the synthesis of glycosylated nephrin.
Additionally, the sensitivity to corticosteroids varies from
patient to patient [16].
We found that the number of relapses after enrolment
into either of the two treatment groups did not vary
significantly. Recent evidence suggests that a cumulative
dosage of pred- nisolone, including longer periods of
prednisolone treatment, do not determine the risk for an
increased number of relapses on follow-up [21]. Indeed,
despite more cumulative prednis- olone in an Indian trial
[22], the percentage of children with a frequently relapsing
course was greater and the time taken to frequently
relapsing status was shorter than in a Japanese trial [23]. In
fact, younger age has been demonstrated to be a major risk
factor for a frequently relapsing course rather than the
cumulative dosage of prednisolone [24]. Our study also
dem- onstrated an inverse relationship between time taken
to remis- sion and the age of the children, implying that age
might be a predictor of the severity of the disease, although
we could not demonstrate a relationship between age and
frequently relaps- ing course.
This study has some merits. The authors compared
the clinical effects of BW- and BSA-based treatment
regimens for children with NS in a randomized clinical
trial, unlike previous studies [15]. Our study was suffi-
ciently powered to examine the time taken to remission
in the two groups, as the sample size was adequate and
the loss to follow-up was within the acceptable limits.
The study also provides valuable confirmation regarding
significantly increased prednisolone doses in children
receiving BSA-based prednisolone regimens.
The present study has some limitations too. First, it
is difficult to draw definitive conclusions regarding the
secondary outcome variables as the study was under-
powered to examine these. Second, the dose of prednis-
olone calculated according to either BW or BSA was
rounded off to the nearest multiple of 2.5. In small
children, rounding off could have a significant impact,
also on the side effects. Third, as the clinical character-
istics of the first episode of NS, such as the extent of
edema, are often different from those in children suffer-
ing a relapse, the outcome variables were analyzed sep-
arately in new-onset NS and IFRNS. However, as the
children were not stratified at trial entry into different
clinical subgroups of NS, the data on the subgroups of
new-onset NS and IFRNS, while supporting the overall
results of the study, represent a post-hoc analysis and
should be confirmed in other studies. It must also be
acknowledged that the study was conducted on South
Indian Tamilian children. Inter-ethnic differences limit
the generalizability of the study results. Also, we were
unable to assess the inter-subject variability in steroid
levels. Finally, regarding the adverse effects of steroids,
it must be acknowledged that the observation period
was short and the numbers were small for the assess-
ment of outcome variables such as hypertension. The
assessment of Cushingoid features was additionally lim-
ited by subjectivity. A longer follow-up would be re-
quired to conclusively assess the impact of an increased
dose of prednisolone on the risk of future relapses, in
addition to the adverse effects, e.g., cardiovascular ef-
fects, final height, bone mineral density, etc. Only a few
studies on the long-term outcome of children with idio-
pathic NS have been published, all showing good out-
come and low morbidity [2528].
Conclusions
Considering the lack of demonstrable beneficial thera-
peutic effects of the higher doses provided by BSA-
based treatment and the undesirable side effects of
higher-dose prednisolone, e.g., hypertension, in addition
to the practical simplicity of a body weight regimen, it
would be desirable to continue with the body weight-
based calculations for prescribing prednisolone in chil-
dren with NS until future studies with longer follow-up
periods are available to assess prednisolone-related ad-
verse effect profiles. The optimal dosage of predniso-
lone required for podocyte cytoskeleton stabilization
could be an area for future research.
Compliance with ethical standards Approval was obtained from
the Institute Ethics Committee (Approval No. JIP/IEC/2014/1/249)
and the trial was registered in the Clinical Trial Registry of India
(CTRI registra- tion No. CTRI/2014/04/004541). Informed consent
was obtained from the parents before the inclusion of subjects in the
study.
The protocol is accessible at Clinical Trial Registry of India
(CTRI)
www.ctri.nic.in. (Registration No.
CTRI/2014/04/004541)
Contributors Vaishnavi Raman and Sriram Krishnamurthy were
in- volved in the management of the patients; Vaishnavi Raman
collected the data, reviewed the literature, and drafted the
manuscript; Sriram Krishnamurthy conceptualized the study, reviewed
the literature, and crit- ically reviewed the manuscript; K.T.
Harichandrakumar performed the statistical analysis; all the authors
contributed to writing the paper and approved the final version of the
manuscript. Sriram Krishnamurthy shall act as guarantor of the paper.
Funding Supported, in part, by institutional and departmental
funds.
Conflicts of interest The authors declare that they have no conflicts
of interest.
References
1. Barnett HL (1979) Nephrotic syndrome in children: a
randomized trial comparing two prednisone regimens in steroid-
responsive pa- tients who relapse early. Report of the
international study of kidney disease in children. J Pediatr
95:228233
2. Hoyer PF, Brodehl J (2006) Initial treatment of idiopathic
nephrotic syndrome in children: prednisone versus prednisone
plus cyclo- sporine a: a prospective, randomized trial. J Am Soc
Nephrol 17:
11511157
3. Brodehl J, Ehrich JH (1988) Short versus standard prednisone
ther- apy for initial treatment of idiopathic nephrotic syndrome
in chil- dren. Arbeitsgemeinschaft fr Pdiatrische
Nephrologie. Lancet 1:380383
4. Melvin T, Bennett W (1991) Management of nephrotic syndrome
in childhood. Drugs 42:3051
5. Ehrich JH, Brodehl J (1993) Long versus standard prednisone
ther- apy for initial treatment of idiopathic nephrotic syndrome
in chil- dren. Arbeitsgemeinschaft fr Pdiatrische
Nephrologie. Eur J Pediatr 152:357361
6. Gauthier B, Trachtman H (1999) Pharmacological treatment of
ne- phrotic syndrome. Drugs Today (Barc) 35:1326
7. Filler G (2003) Treatment of nephrotic syndrome in children
and controlled trials. Nephrol Dial Transplant 18 [Suppl 6]:
vi75vi78
8. Hiraoka M, Tsukahara H, Matsubara K, Tsurusawa M, Takeda
N, Haruki S, Hayashi S, Ohta K, Momoi T, Ohshima Y,
Suganuma N, Mayumi M, West Japan Cooperative Study
Group of Kidney Disease in Children (2003) A randomized
study of two long course prednisolone regimens for nephrotic
syndrome in children. Am J Kidney Dis 41:11551162
9. Ksiazek J, Wyszyska T (1995) Short versus long initial
prednisone treatment in steroid-sensitive nephrotic syndrome in
children. Acta Paediatr 84:889893
10. Bagga A, Hari P, Srivastava RN (1999) Prolonged versus
standard prednisolone therapy for initial episode of nephrotic
syndrome. Pediatr Nephrol 13:824827
11. Ekka BK, Bagga A, Srivastava RN (1997) Single-versus
divided- dose prednisolone therapy for relapses of nephrotic
syndrome. Pediatr Nephrol 11:597599
12. Indian Pediatric Nephrology Group, Indian Academy of
Pediatrics, Bagga A, Ali U, Banerjee S, Kanitkar M, Phadke KD,
Senguttuvan P, Sethi S, Shah M (2008) Management of steroid
sensitive nephrot- ic syndrome: revised guidelines. Indian
Pediatr 45:203214
13. Hodson EM, Willis NS, Craig JC (2007) Corticosteroid therapy
for
nephrotic syndrome in children. Cochrane Database Syst Rev 4,
CD001533
14. Feber J, Al-Matrafi J, Farhadi E, Vaillancourt R, Wolfish N
(2009) Prednisone dosing per body weight or body surface area
in children with nephrotic syndromeis it equivalent? Pediatr
Nephrol 24:
10271031
15. Saadeh SA, Baracco R, Jain A, Kapur G, Mattoo TK, Valentini
RP (2011) Weight or body surface area dosing of steroids in
nephrotic syndrome: is there an outcome difference? Pediatr
Nephrol 26:
21672171
16. Mehls O, Hoyer PF (2011) Dosing of glucocorticosteroids in
ne- phrotic syndrome. Pediatr Nephrol 26:20952098
17. National High Blood Pressure Education Program Working
Group on High Blood Pressure in Children and Adolescents
(2004) The fourth report on the diagnosis, evaluation, and
treatment of high blood pressure in children and adolescents.
Pediatrics 114:555576
18. Hinkes BG, Mucha B, Vlangos CN, Gbadegesin R, Liu J,
Hasselbacher K, Hangan D, Ozaltin F, Zenker M, Hildebrandt
F, Arbeitsgemeinschaft fr Paediatrische Nephrologie Study
Group (2007) Nephrotic syndrome in the first year of life: two
thirds of cases are caused by mutations in 4 genes (NPHS1,
NPHS2, WT1, and LAMB2). Pediatrics 119:e907e919
19. Mosteller RD (1987) Simplified calculation of body-surface area.
N Engl J Med 317:1098
20. Neuhaus TJ, Laube GF (2012) Weight dosing of steroids in ne-
phrotic syndrome. Pediatr Nephrol 27:505
21. Hahn D, Hodson EM, Willis NS, Craig JC (2015)
Corticosteroid therapy for nephrotic syndrome in children.
Cochrane Database Syst Rev 3:CD001533
22. Sinha A, Saha A, Kumar M, Sharma S, Afzal K, Mehta A,
Kalaivani M, Hari P, Bagga A (2015) Extending initial
predniso- lone treatment in a randomized control trial from 3 to 6
months did not significantly influence the course of illness in
children with steroid-sensitive nephrotic syndrome. Kidney Int
87:217224
23. Yoshikawa N, Nakanishi K, Sako M, Oba MS, Mori R, Ota E,
Ishikura K, Hataya H, Honda M, Ito S, Shima Y, Kaito H,
Nozu K, Nakamura H, Igarashi T, Ohashi Y, Iijima K,
Japanese Study Group of Kidney Disease in Children (2015) A
multicenter ran- domized trial indicates initial prednisolone
treatment for childhood nephrotic syndrome for two months is
not inferior to six-month treatment. Kidney Int 87:225232
24. Hoyer PF (2015) New lessons from randomized trials in steroid-
sensitive nephrotic syndrome: clear evidence against long
steroid therapy. Kidney Int 87:1719
25. Fakhouri F, Bocquet N, Taupin P, Presne C, Gagnadoux MF,
Landais P, Lesavre P, Chauveau D, Knebelmann B, Broyer
M, Grnfeld JP, Niaudet P (2003) Steroid-sensitive nephrotic
syndrome: from childhood to adulthood. Am J Kidney Dis
41:550557
26. Leonard MB, Feldman HI, Shults J, Zemel BS, Foster BJ,
Stallings VA (2004) Long-term, high-dose glucocorticoids and
bone mineral content in childhood glucocorticoid-sensitive
nephrotic syndrome. N Engl J Med 351:868875
27. Rth EM, Kemper MJ, Leumann EP, Laube GF, Neuhaus TJ
(2005)
Children with steroid-sensitive nephrotic syndrome come of
age:
long-term outcome. J Pediatr 147:202
207
28. Kyrieleis HA, Lwik MM, Pronk I, Cruysberg HR, Kremer JA,
Oyen WJ, van den Heuvel BL, Wetzels JF, Levtchenko EN
(2009) Long-term outcome of biopsy-proven, frequently
relapsing minimal-change nephrotic syndrome in children. Clin
J Am Soc Nephrol 4:15931600