Pediatr Nephrol (2001) 16:1045–1048
C L I N I C A L N E P H R O L O G Y / O R I G I N A L A RT I C L E
Sanjeev Gulati · Saubhik Sural · Raj Kumar Sharma
Amit Gupta · Ramesh Kumar Gupta
Spectrum of adolescent-onset nephrotic syndrome in Indian children
Received: 7 September 2000 / Revised: 21 June 2001 / Accepted: 26 June 2001
Abstract There are few data regarding adolescent-onset
nephrotic syndrome (NS) and no guidelines for biopsy
criteria and treatment protocol. This study was conduct-
ed to analyze the clinical spectrum of adolescent-onset
NS and evaluate possible biopsy criteria in these chil-
dren. A prospective analysis was carried out on all pa-
tients with idiopathic NS (fulfilling the ISKDC criteria)
with onset between 1 and 18 years of age. They were
evaluated clinically, followed by biochemical investiga-
tions and kidney biopsy. These characteristics of patients
with onset between 1 and 12 years (group A) were com-
pared with the same parameters in patients with onset
between 12 and 18 years of age (group B) referred to our
hospital over the same period. Among all clinical param-
eters, microhematuria was significantly more prevalent
in adolescents (P<0.001). Kidney biopsy was performed
in 88% of adolescent patients. Focal segmental glomeru-
losclerosis (FSGS) was the most-common histopatholo-
gy in group B (46.3%) compared with minimal change
disease (MCD) in group A (42.9%). Group B had a sig-
nificantly higher frequency of membranoproliferative
glomerulonephritis (MPGN) (P<0.005) and a significant-
ly lower frequency of MCD (P<0.001). The biochemical
parameters at the onset were similar. On comparing mi-
crohematuria, hypertension, and renal insufficiency at
presentation, we observed that two or more of these fea-
tures were present in all patients with MPGN and only in
19.6% of adolescents with MCD, mesangioproliferative
glomerulonephritis, and FSGS. The frequency of steroid
resistance was significantly higher in group B (P<0.001).
In conclusion, adolescent-onset NS differs from the
S. Gulati (✉) · S. Sural · R.K. Sharma · A. Gupta
Department of Nephrology,
Sanjay Gandhi Post Graduate Institute of Medical Sciences,
Raebareli Road, Lucknow 226014, India
e-mail: sgulati@sgpgi.ac.in
Tel.: +91-522-440700, Fax: +91-522-440017
R.K. Gupta
Department of Pathology,
Sanjay Gandhi Post Graduate Institute of Medical Sciences,
Lucknow, India
© IPNA 2001
childhood variety in having a significantly higher fre-
quency of hematuria, steroid resistance, and evidence of
non-MCD, especially MPGN, on histopathology. Kidney
biopsy can be restricted to those adolescents who have at
least two abnormal clinical/biochemical features or are
steroid non-responders.
Keywords Adolescent · Nephrotic syndrome ·
Focal segmental glomerulosclerosis
Introduction
There are few data regarding the clinical spectrum of ad-
olescent-onset nephrotic syndrome (NS), with variable
results in the few published reports [1, 2, 3]. There are
no clear-cut recommendations as to the biopsy criteria
and treatment protocols.
We have analyzed patients with idiopathic NS with
onset between 1 and 18 years of age, to evaluate clinical
features, biochemical parameters, steroid responsiveness,
and histopathological spectrum of adolescent-onset NS
and to compare them with childhood-onset NS.
Patients and methods
A prospective analysis was carried out of all patients with idio-
pathic NS, with onset between 1 and 18 years of age, referred to
our institute over the period 1990–1998. NS syndrome was diag-
nosed according to the criteria of the International Study of Kid-
ney Disease in Children (ISKDC), i.e., presence of proteinuria
(>40 mg/m2 per hour), hypoalbuminemia (<25 g/l), edema, and
usually hypercholesterolemia (>250 mg/dl) [4]. Patients with sec-
ondary NS, i.e., clinical features of systemic lupus erythematosus,
Henoch-Schonlein purpura, and/or laboratory (hepatitis B virus
surface antigen positivity) evidence of systemic disease were ex-
cluded. Patients were evaluated clinically, followed by biochemi-
cal investigations. Clinical parameters recorded were: age at onset,
gender, presence and degree of edema, urine output, blood pres-
sure, occurrence of micro- and/or macrohematuria, and other com-
plications of NS. Laboratory investigations included hematocrit,
serum blood urea nitrogen (BUN), creatinine, protein, albumin,
cholesterol, and triglycerides. All patients in group B (onset 12–18
1046
years) were recommended to undergo kidney biopsy prior to start-
ing treatment. Children in group A (onset 1–12 years) were treated
with steroids, and biopsies were carried out only in those who
were steroid resistant, planned for immunosuppressive therapy, or
had clinical and/or biochemical abnormalities [4, 5, 6]. Patients
with minimal change disease (MCD), focal segmental glomerulo-
sclerosis (FSGS), and mesangioproliferative glomerulonephritis
(MesPGN) were treated with prednisolone. At first presentation,
these patients were treated with prednisolone 60 mg/m2 per day
for 4–6 weeks, followed by 40 mg/m2 on alternate days for 4–6
weeks [4, 7]. Relapses were treated with prednisolone
60 mg/m2 per day until remission for 3 days followed by
40 mg/m2 on alternate days for 4 weeks. Based on the response to
steroids, these patients were categorized into infrequent relapsers
(IFR), frequent relapsers (FR), steroid-dependent (SD), and ste-
roid non-responders (SNR) using standard definitions [4]. Oliguria
was defined as urine output less than or equal to 400 ml/m2 per
day. Cyclophosphamide was administered in the FR, the SD group
with steroid toxicity, and in the SNR group, when consent was
available. This was given orally at a dose of 2.5 mg/kg per day for
3 months or intravenous monthly pulses for 6 months at a dose of
500 mg/m2 per dose according to the protocol described previous-
ly [8]. All clinical, biochemical, and histopathological characteris-
tics of patients with adolescent-onset NS were compared with the
same parameters in the patients with NS with onset between 1 and
12 years, who were referred to our hospital over a similar period
of time.
Statistical analysis was performed using chi-squared test and
Student’s t-test. Values are expressed as mean±standard deviation.
A P value <0.05 was considered significant.
Results
The mean age of onset of the adolescents (group B) was
14 years and there were 63 boys (69.2%) and 28 girls
(30.8%). The duration of follow-up ranged from 2 to
6 years (mean 5.6 years). On comparing childhood-onset
NS (age of onset 1–12 years, group A, n=271) with ado-
lescent-onset NS (group B, n=91), the frequency of olig-
uria (57.2% in group A vs. 45.8% in group B) and hyper-
tension (17.3% in group A and 28.6% in group B) were
similar in the two groups. There was, however, a signifi-
cantly higher prevalence of microhematuria in adoles-
cents (group A 32.8% vs. group B 54.9%, P<0.0001).
Kidney biopsy was performed in 80 of 91 patients (88%)
in group B. In 11 adolescents where consent was not
available, empirical steroid therapy was started. In group
A, 161 of 271 (59.4%) children were biopsied according
to the criteria described earlier. On comparing the histo-
pathological spectrum in the two groups, we observed
that FSGS was the most-common histopathology finding
in group B (46.3%) compared with MCD in group A
(42.9%) (Fig. 1). Adolescent-onset NS had a significant-
Table 1 Clinical and biochem-
ical profile in different sub- Clinical MCD n=13
types at presentation (MCD features (n,%)
minimal change disease, FSGS
focal segmental glomeruloscle- Microscopic hematuria 3 (23)
rosis, MesPGN mesangioprolif- Hypertension 2 (15.4)
erative glomerulonephritis, Renal insufficiency 1 (7.7)
MN membranous nephropathy, >2 of the above 1 (7.7)
MPGN membranoproliferative
glomerulonephritis)
Fig. 1 Comparison of histopathological spectrum in childhood-
onset nephrotic syndrome (group A) and adolescent-onset neph-
rotic (group B) and. Group B had a significantly higher frequency
of membranoproliferative glomerulonephritis (MPGN) (P<0.005)
and a significantly lower frequency of minimal change disease
(MCD) (P<0.001) than group A. Focal segmental glomeruloscle-
rosis (FSGS) was the commonest histopathology in group B
ly higher frequency of MPGN (P<0.005) and significant-
ly lower frequency of MCD (P<0.001) than childhood-
onset NS. The biochemical parameters at the onset were
similar in the two groups: BUN (group A 13.07±1.6 vs.
group B 15.06±1.9 mg/dl), serum creatinine (group A
0.92±0.2 vs. group B 1.02±0.3 mg/dl), total protein
(group A 4.4±1.1 vs. group B 4.3±1.4 g/dl), serum albu-
min (group A 2.2±0.6 vs. group B 2.1±0.7 g/dl), and
serum cholesterol (group A 327.1±96.2 vs. group B
319.2±1.5 mg/dl) (P>0.05).
We compared microhematuria, hypertension, and re-
nal insufficiency at presentation in children with MCD,
MesPGN, and FSGS with those of MPGN (Table 1). We
observed that two or more of these three features were
present in all patients with MPGN, and only in 11 of the
other 56 children (19.6%) with MCD, MesPGN, and
FSGS (P<0.01) (Table 1). Low C3 was detected only
in patients with MPGN. After a mean follow-up of
5.6 years, the prevalence of renal insufficiency was 7.7%
in children with MCD, 21.6% in patients with FSGS,
33.4% in patients with MesPGN, and 54.5% in patients
with MPGN (includes children who already had renal in-
sufficiency at onset).
The steroid response pattern in group B revealed that
all steroid responders had either MCD, FSGS, or
MesPGN. All the adolescents with MPGN (9/22) and
membranous nephropathy (2/2) who received a course of
FSGS n=37 MesPGN n=6 MN n=2 MPGN n=22
(n,%) (n,%) (n,%) (n,%)
17 (46) 6 (100) 1 (50) 20 (90.9)
8 (21.6) 2 (33.3) 0 14 (63.6)
4 (10.8 1 (16.7) 0 8 (36.4)
8 (21.6) 2 (33.4) 0 22 (100)

Fig. 2 Comparison of response to steroid therapy in childhood-
onset nephrotic syndrome (group A) and adolescent-onset neph-
rotic syndrome (group B), excluding patients with MPGN and
MGN. The frequency of steroid non-responsiveness (SNR) was
significantly higher (P<0.001) in group B (IFR infrequent relaps-
ers, FR frequent relapsers, SD steroid-dependent)
Fig. 3 Comparison of clinical and biochemical parameters be-
tween steroid responders and non-responders with adolescent-
onset nephrotic syndrome. Steroid non-responders had a signifi-
cantly higher (P<0.001) frequency of renal insufficiency at onset
(HTN hypertension)
steroids before biopsy were steroid non-responders. Ex-
cluding patients with MPGN and MGN, the comparison
of response to steroid therapy in group A and group B is
shown in Fig. 2. The frequency of steroid resistance was
significantly higher in group B (P<0.001). On comparing
the clinical and biochemical parameters of steroid re-
sponders (IFR, FR, and SD) and non-responders, we ob-
served that steroid non-responsive patients had a signifi-
cantly higher frequency of renal insufficiency at onset
(Fig. 3). Of 56 adolescents with MCD, FSGS, and
MesPGN, 34 (61%) were steroid sensitive. None of these
patients had two or more features of microhematuria, hy-
pertension, and renal insufficiency at presentation. All
the 11 children in this subgroup who had two or more
abnormal features were also SNR.
1047
Discussion
This study, the largest to date, analyses the spectrum of
adolescent-onset NS in Indian children – the demograph-
ic pattern, various clinical and biochemical parameters,
response to steroid therapy, and histopathological le-
sions. This is the first study of this type in a homogenous
Asian population of Indian ethnic origin. It has previous-
ly been shown that there is a variation in the prevalence
and steroid response in NS in different ethnic groups
[9, 10]. So far there have been only three other published
studies [1, 2, 3] describing exclusively adolescent pa-
tients with NS. All of these were retrospective studies
and included only those patients who underwent kidney
biopsy. None of these have correlated the clinical and
biochemical features with steroid responsiveness or have
compared adolescents with childhood-onset NS. More-
over, these studies have not addressed the guidelines to
be adopted for kidney biopsy and steroid therapy in this
group of patients.
We observed that microscopic hematuria is more fre-
quent in adolescents (group B). This is probably due to
the fact that there was a higher prevalence of non-MCD
in this group [5]. The frequency of steroid resistance was
higher in group B. Microscopic hematuria was more
common in steroid non-responders, although not statisti-
cally significant. Renal insufficiency at presentation in
our patients was associated with steroid unresponsive-
ness. There was no difference in biochemical parameters
at onset.
FSGS was the most-common etiology (46.3%) in our
adolescent patients, which is less than that in non-white
adolescents [3], but much higher than the 14%–18.5%
reported in the other two studies [1,2]. In the study of
Takada et al. [1], 54% of 50 Japanese adolescents with
NS had MCD. Their conclusion was that proliferative
and membranous lesions were more common in adoles-
cents, with a lower incidence of MCD. In the report of
the Southwest Pediatric Nephrology Study Group, of
65 patients, 31% had MCD, 18.5% each had FSGS and
MGN, and 12% had MPGN [2]. Baqi et al. [3] also re-
ported a low incidence of MCD in an adolescent popula-
tion. However in their patient population consisting of
African-Americans, there was a preponderance of FSGS
(55%). We observed a significantly higher frequency of
MPGN (27.5%) among adolescents compared with
younger children. In the study of Hogg et al. [2], the fre-
quency of MPGN was 12%, while Baqi et al. [3] ob-
served the prevalence of MPGN to be only 7%. This
variation reflects geographical difference in distribution
of glomerular diseases, as the prevalence of MPGN is re-
ported to be declining in the western world.
In our racially homogeneous Indian population, only
16.3% of adolescents had MCD, compared with 42.9%
in group A. In group A only 59.5% of patients under-
went kidney biopsy. The biopsy criteria were selective in
group A according to current recommendations. This
practice helps to minimize biopsies in steroid-responsive
cases without any atypical features and who are most
1048
likely to have MCD. Thus the actual prevalence of MCD
is likely to be even higher in group A. In group B, 29.9%
of patients were steroid non-responsive, and only 16.3%
displayed underlying MCD. In the other three studies all
patients were subjected to routine pre-treatment biopsies.
There are no clear recommendations regarding restrict-
ing biopsies in this subgroup, and no study has so far
evaluated this problem. MCD, MesPGN, and FSGS are
treated with a common therapeutic protocol in children.
We also used a similar protocol in these adolescent chil-
dren. In our study, they accounted for 70% of the pa-
tients who were subjected to biopsies. It is possible to
predict MPGN in adolescents based on clinical and bio-
chemical features. Thus one could subject the other ado-
lescents to standard prednisolone therapy and limit biop-
sies to the subgroup of steroid non-responders and/or
those with two or more abnormal clinical or biochemical
features. We tried to quantify the expected benefit of
guidelines for selective biopsy. Of the 56 children with
MCD, MesPGN, and FSGS, 34 were steroid sensitive
and 22 steroid non-responders. All the 11 children who
had two or more biochemical abnormalities were SNR.
Thus, using this approach, biopsy would have been
restricted to 22 of these children and avoided in 34 of
80 adolescent patients. One draw back of this approach
is that 2 patients with MGN did not have any clinical or
biochemical abnormalities. Hence they would have been
exposed to 4 weeks of steroid treatment before labeling
them SNR and undertaking biopsy. With the incidence of
MGN in this age group being only 2.5%, the overall haz-
ard of overexposing MGN patients to 4 weeks of steroids
is minimal.
In conclusion, adolescent-onset NS differs from the
childhood variety, in having a significantly higher fre-
quency of microscopic hematuria, steroid resistance, and
evidence of histopathology other than MCD, especially
MPGN. Kidney biopsy in this subgroup could be re-
stricted to children who have at least two abnormal clini-
cal/biochemical features or are steroid non-responders.
References
1. Takada T, Yangihahra T, Kuwabra H, Igarrhic T, Yoshigumi A,
Kihora I (1989) Clinicopathologic study in 50 adolescent pa-
tients with primary nephrotic syndrome. Pediatr Nephrol 3:C9
2. Hogg RJ, Silva FG, Berry PL, Weng JE (1993) Report of
Southwest Pediatric Nephrology study group. Glomerular le-
sions in adolescents with gross hematuria or the nephrotic syn-
drome. Pediatr Nephrol 7:27–31
3. Baqi N, Singh A, Balachandra S, Ahmad H, Nicastri A,
Kytinski S, Homel P, Tejani A (1998) The paucity of minimal
change disease in adolescents with primary nephrotic syn-
drome. Pediatr Nephrol 12:105–107
4. Gulati S, Kher V, Sharma RK, Gupta A (1994) Steroid re-
sponse pattern in Indian children with nephrotic syndrome.
Acta Pediatr Scand 83:530–533
5. Nash MA, Edlemann CM Jr, Bernstein J, Barnett HL (1992)
The nephrotic syndrome. In: Edlemann CM Jr (ed) Pediatric
kidney disease. Little Brown, Boston, pp 1247–1266
6. Broyer M, Meyrier A, Niaudet P, Habib R (1998) Minimal
change and focal glomerular sclerosis. In: Davison AM,
Cameron JS, Grunfeld JP, Kerr DNS, Ritz E, Winerals CG
(eds) Oxford textbook of clinical nephrology. Oxford Univer-
sity Press, Oxford, pp 493–535
7. Gulati S, Kher V, Elhence R, Kumar P, Sharma RK, Gupta A
(1994) Treatment of nephrotic syndrome. Indian Pediatr
31:165–170
8. Elhence R, Gulati S, Kher V, Gupta A, Sharma RK (1994) In-
travenous pulse cyclophosphamide – a new regimen for ste-
roid resistant minimal change nephrotic syndrome. Pediatr
Nephrol 8:1–3
9. Sorof JM, Hawkins EP, Breven ED, Boydstun II, Kale AS,
Powel DR (1998) Age and ethnicity effect the risk and out-
come of focal segmental glomerulosclerosis. Pediatr Nephrol
12:764–768
10. Sharples PM, Boulton J, White RHR (1985) Steroid respon-
sive nephrotic syndrome is more common in Asians. Arch Dis
Child 60:1014–1017