Neonatal nephrocalcinosis: long term follow up
E Porter, A McKie, T J Beattie, J H McColl, N Aladangady, A Watt, M P White
............................................................... ............................................................... .
See end of article for
Dr Porter, Southern
General Hospital, Govan
Road, Govan, Glasgow
G51 4TF, Scotland, UK;
Accepted 6 May 2006
Published Online First
16 May 2006
Arch Dis Child Fetal Neonatal Ed 2006;91:F333–F336. doi: 10.1136/adc.2006.094755
Aims: To assess the spontaneous resolution of neonatal nephrocalcinosis and its long term effects on renal
Methods: Fourteen very low birthweight preterm babies with nephrocalcinosis were followed up at 5–7
years of age; 14 controls were matched for sex, gestation, and birth weight. Height, weight, blood
pressure, and renal symptomatology were recorded, and a renal ultrasound scan was performed. Early
morning urine osmolality and creatinine ratios of albumin, phosphate, calcium, oxalate and b
microglobulin were determined. Urea and electrolytes in the study group were determined, and
glomerular filtration rate (GFR) and TmP/GFR (tubular reabsorption of phosphate per GFR) were
calculated. Statistical analysis was performed on a group basis using the Mann-Whitney confidence
Results: Mean age was 6.9 years (range 5.81–7.68). An early morning urine osmolality .700 mOsm/kg
was achieved in all cases. In two cases and four controls, the calcium/creatinine ratio was .0.7 mmol/
mmol. In all cases, the GFR was normal (median 132.6 ml/min/1.73 m2(range 104.1–173.1)). Median
TmP/GFR was 1.22 mmol/l (0.73–1.61), with two having levels below the normal range. These did not
have persisting nephrocalcinosis. Nephrocalcinosis was found in three of the 12 cases scanned and one
control. There were no significant differences in urine biochemistry.
Conclusions: Resolution of nephrocalcinosis occurred in 75% of cases. No evidence was found to suggest
that nephrocalcinosis is associated with renal dysfunction in the long term. There was evidence of
hypercalciuria in the cases and controls, suggesting that prematurity may be a risk factor.
and risk factors include low gestational age, low birth weight,
frusemide use, aminoglycoside use, male sex, length of
assisted ventilation, and length of hospital stay.2–6
Although nephrocalcinosis is thought to predispose to
urinary tract infection, renal colic, and haematuria, the long
term consequences in preterm babies are unclear, and several
studies have suggested a resolution of nephrocalcinosis in
20–60%.6–12Some reports suggest that nephrocalcinosis may
predispose to glomerular and tubular dysfunction,7 8and
others propose that prematurity alone leads to renal
dysfunction, independent of nephrocalcinosis.10
In a previous study, we determined the incidence and
aetiological factors associated with nephrocalcinosis in
,32 weeks.2The aim of the present study was to follow up
preterm babies found to have nephrocalcinosis and compare
them with matched controls from the original study to
determine whether nephrocalcinosis is associated with long
term renal functional sequelae.
reterm babies are at risk of nephrocalcinosis, first
described by Hufnagle et al in 1982.1The incidence has
been found to be 16–64% in very low birthweight babies,
In the original study, 16 out of 101 preterm babies had
nephrocalcinosis (16%). On follow up of these 16 babies, one
had died of a cause unrelated to renal dysfunction and one
was not contactable, leaving a cohort of 14. Fourteen controls
from the original study who did not have nephrocalcinosis
were matched for birth weight (¡200 g), gestational age
(¡1 week), and sex. Where more than one control in the
pool met the criteria to match a case, one of the available
controls was selected randomly. If the selected control
refused to take part, another of the appropriate controls
was selected at random from all those remaining. Both
the nephrocalcinosis cohort and controls were re-evaluated at
5–7 years of age, using a local research ethics committee
Each child and parent was interviewed, and written
informed consent obtained. The interview included specific
questions on whether there had been a history of urinary tract
infection or other urinary tract symptoms. The hospital case
records of cases and controls were examined, and contact was
made with the relevant general practitioner. Height, weight,
and blood pressure were measured and compared with normal
values.13 14An early morning urine sample was collected
before the child had anything to drink and sent for
determination of osmolality, calcium, albumin, protein,
phosphate, oxalate, b2 microglobulin, and creatinine. If early
morning urine osmolality was found to be ,700 mOsm/l,
intranasal desmopressin was administered and a follow up
was performed four hours later. The patient group also had
blood taken for determination of plasma urea and electrolytes,
calcium, and phosphate. From this, glomerular filtration rate
(GFR) in ml/min/1.73 m2was calculated using the formula
GFR = (49 6height (cm))/plasma creatinine.15
The renal tubular reabsorption of phosphate per GFR (TmP/
GFR) was calculated using the formula
TmP/GFR = PPO4– (UpPcr/Ucr),
where PPO4 = plasma phosphate (mmol/l), Up = urinary
phosphate (mmol/l), Pcr = serum creatinine (mmol/l), and
Ucr = urinary creatinine (mmol/l).16
Control patients did not have blood taken for ethical
A renal ultrasound was performed on cases and controls to
determine renal length, morphology, and presence of
nephrocalcinosis using an ATL 15000 Phillips Medical
Systems scanner. Renal length was compared with normal
values.17All scans were carried out or checked by a single
consultant radiologist, who was unaware of the neonatal
renal ultrasound result. Four children defaulted from
ultrasound leaving 24 (12 cases and 12 controls).
Statistical analysis was performed using a two tailed
Mann-Whitney U test, with significance level 5%, and
associated 95% confidence interval.
Median age at follow up for cases was 6.69 years and for
controls 7.21 years. There was no significant difference in
gestational age or birth weight between the groups (table 1).
One patient and one control had height and weight ,0.4th
centile, with a further two patients and one control having
heights and weights in the 2–9th centiles.
Only one patient had had confirmed urinary tract infection
in the past, at age 1 month. Urinary tract imaging in the form
of technetium-99m dimercaptosuccinic acid and micturating
cystourethrogram were normal. One control had a history of
frank haematuria and was found to have hypercalciuria and
has subsequently been invited for follow up. In another case,
mild hydronephrosis was noted on the neonatal scan, which
had settled by 4 years of age. Three patients and one control
had systolic hypertension, and one patient had systolic and
diastolic hypertension. Blood pressure measurement was not
possible in one control because of lack of cooperation.
There were no significant differences in any of the urinary
biochemical variables between the two groups (table 2).
One child had received frusemide for treatment of a heart
condition. He had stopped medication three weeks before the
study. He had no biochemical abnormalities of serum or
All patients had normal GFR with a median of 132.6 ml/
min/1.73 m2(104.1–173.1), and the median TmP/GFR was
1.22 mmol/l (0.73–1.61). Two patients had a TmP/GFR below
the normal for age (0.73, 0.95). These were the only two
patients who also had raised b2 microglobulin/creatinine
ratios (0.118, 0.139 mg/mmol). They did not have persisting
Six children (four controls and two cases) had evidence of
hypercalciuria (Ca/Cr ratio .0.7 mmol/mmol). This persisted
on repeat spot urine testing in four and was confirmed in
three by 24 hour urine collection (normal ,0.1 mmol/kg/
day). One child has repeatedly failed to attend follow up
appointments for 24 hour collection (table 3).
Of the three children (two cases and one control) with
hypercalciuria confirmed by 24 hour urine analysis, all had
raised phosphate/creatinine ratios on spot testing. The one
case also had raised b2 microglobulin/creatinine ratio and a
low TmP/GFR. There were no other biochemical abnormal-
On questioning, there was no family history of renal
disease, including renal stones, in any of the children.
Two cases defaulted from renal ultrasound scan appoint-
ments, and as a result 12 cases and 12 controls were scanned.
Three of the 12 cases scanned (25%) and one control had
evidence of nephrocalcinosis. This was bilateral except in one
of the cases, and in this child nephrocalcinosis had been
bilateral in the neonatal period. Of these children, only the
control with nephrocalcinosis at follow up (6.2 years) had
evidence of hypercalciuria. This child had normal renal
ultrasound scans as a neonate and at 3 months of age and
had no history of urinary tract infection, but had a recent
episode of haematuria and remains under investigation.
Renal length was found to be reduced in four of 48
kidneys. One case and one control had one kidney ,5th
centile for height, and one case had bilateral small kidneys.
These children had no other evidence of renal dysfunction.
We found that nephrocalcinosis resolved in 75% of patients
scanned at a median age of 6.75 years. Downing et al7found
resolution in six of 10 (60%) preterm babies followed up at 1–
2 years of age, and Ezzedeen et al8found resolution in 44%
(four of nine), and improvement in 56% (five of nine) very
low birthweight preterm babies at 21.3 months of age. In
their follow up study, Jones et al10found that five out of 11
children (54%) had resolution at 4–5 years. Our finding of a
higher rate of resolution is possibly due to the longer
duration of follow up in our study. This conclusion is
supported by the findings of Saarela et al11and Hoppe et al.12
Hoppe et al in a retrospective study identified 16 preterm
babies with nephrocalcinosis, which had resolved on ultra-
sound in 31% by 12 months and in 75% between 3 and 6
years. Saarela et al studied a group of babies with neonatal
nephrocalcinosis classified on renal ultrasound findings into
peripheral, scattered, and extensive. Each child had annual
scans until resolution or to the age of 5–6 years. All of those
in the study
Basic characteristics of the patients and controls
Gestational age (weeks)
Birth weight (g)
Age at follow up (years)
Values are median (range).
VariableControls Patients95% CI p Value
20.36 to 0.18
2122 to 94
21.00 to 1.40
20.6 to 0.5
25.18 to 0.01
20.020 to 0.020
Values are median (range). Differences assessed by the Mann-Whitney U test.
CI, Confidence interval.
Evidence of hypercalciuria
Ca/Cr ratio 1
Ca/Cr Ratio 2
24 hour Ca collection
F334 Porter, McKie, Beattie, et al
with peripheral nephrocalcinosis had complete resolution by
1 year of age, the seven with scattered changes by 3–4 years.
Of those with extensive changes, one baby died, and two of
the remaining three had persisting nephrocalcinosis at 5–6
years, indicating a higher rate of resolution with time, with
more extensive changes taking longer to disappear. We did
not undertake such an ultrasound classification of nephro-
calcinosis at the time of the original study, but we feel it is
likely that those with persisting nephrocalcinosis will show
resolution later in life, although longer term follow up will be
Our study indicated there was no evidence of impaired
renal function or urine concentrating capacity as indicated by
normal GFR and early morning osmolalities. Only two out of
14 children had reduced TmP/GFR, which, in association
with raised b2 microglobulin/creatinine ratio, is probably a
subtle indication of proximal tubular dysfunction. One of
these children also had hypercalciuria, although in both
nephrocalcinosis had resolved. As controls did not have these
tests performed, it is not possible to conclude whether
neonatal nephrocalcinosis was causal. Other urinary bio-
chemical variables were not different between the two
groups. These results differ from those of Downing et al,7
who followed 27 children to 14.1 months of age and found
that those who had nephrocalcinosis (n = 10) had a higher
urinary calcium/creatinine ratio and a lower tubular re-
absorption of phosphate and tubular acidification capacity
than those who did not have nephrocalcinosis (n = 17).
Ezzedeen et al8suggested that there may be a reduction in
GFR in children who had neonatal nephrocalcinosis when
they were evaluated at a mean of 21.3 months. However, this
was a small study of only nine patients, with no controls.
Compared with our study, these studies were performed on
children at a significantly younger age when renal function is
not yet fully developed. Jones et al10followed up their cohort
of 11 babies at a later age of 4–5 years, and the results
showed no significant glomerular or tubular dysfunction
associated with neonatal nephrocalcinosis in keeping with
our own findings. This also concurs with Hoppe et al,12who
found that GFR and urinary osmolality in children with
neonatal nephrocalcinosis were normal for age related values
at follow up.
Six children in the study had evidence of hypercalciuria.
Urinary Ca/Cr ratio was used, as this required only spot
fasting samples from the children and has been shown to be
useful as a screening tool in the diagnosis of hypercalciuria.18
Three of the five children who have supplied 24 hour urine
samples were found to have raised calcium excretion.
Numbers are too small to draw any conclusions, but neonatal
nephrocalcinosis does not appear to be a risk factor. This
again is in keeping with the findings of Jones et al,10who
reported abnormal calcium load tests in three cases and five
controls at 4–5 years of age, indicating that hypercalciuria
appeared to be a consequence of prematurity independent of
neonatal nephrocalcinosis. Hoppe et al12noted increased
calcium excretion in 13 of 31 preterm babies in the first
8 weeks of life, but did not comment on hypercalciuria at
follow up. In a further study of 46 preterm children and 40
term matched controls aged 7–8 years, Jones et al19found that
hypercalciuria was significantly more likely in the preterm
group and was associated with raised serum aminoglycoside
concentrations in the neonatal period. Aminoglycoside
antibiotic use is well recognised as a cause of tubular
dysfunction in the newborn period,20but whether this is
long lasting is not known. We have reviewed data from our
original study on neonatal nephrocalcinosis and ascertained
that, of the three subjects with confirmed hypercalciuria at
follow up, two had toxic gentamicin concentrations in the
neonatal period, and the other had had exposure to
gentamicin, with no toxic concentrations identified. All were
of 27 weeks gestation and birth weight less than 1000 g.
Hypercalciuria is related to renal tubular abnormalities,
abnormal handling of salt, increased prostaglandin E2
production, and abnormal synthesis of 1,25-dihydroxy
vitamin D3. It is associated with renal morbidity in child-
hood—for example, haematuria, frequency dysuria syn-
drome, and renal calculi21 22—and it has been shown to
adversely affect bone mineral density and height at 7–8 years
of age.19 23In view of this potential morbidity and the ease of
performing spot urine samples in babies, calcium/creatinine
ratios, for which normative preterm values have been
proposed,24may be a useful screening tool for distinguishing
those babies at risk of hypercalciuria.
We did not measure urinary citrate, a powerful inhibitor of
renal stone formation. This was measured in our original
study, and no significant correlation with nephrocalcinosis
Only one patient with neonatal nephrocalcinosis had
experienced urine infection or haematuria. Four patients
and one control had high blood pressure on one reading
taken on the day of the hospital visit. This clearly may be
related to anxiety and will be kept under review.
Nephrocalcinosis persisted in 25% of patients at a median of
6.7 years, but we found no evidence to suggest that it is
associated with renal dysfunction or symptoms in the long
term, but further studies on those with persistent nephro-
calcinosis are necessary. Although preterm babies may be at
higher risk of developing hypercalciuria, nephrocalcinosis
does not appear to be a risk factor, and we would not
recommend undertaking routine renal ultrasound scans on
very low birthweight preterm babies for nephrocalcinosis, as
this appears to be benign. However, a case could be made for
screening for hypercalciuria on discharge by doing a spot
urinary calcium/creatinine ratio, particularly on those who
have had exposure to toxic concentrations of aminoglyco-
E Porter, A McKie, T J Beattie, J H McColl, N Aladangady, A Watt,
M P White, Southern General Hospital, Glasgow, Scotland, UK
Competing interests: none declared
1 Hufnagle KG, Khan SN, Penn D, et al. Renal calcifications: a complication of
long term frusemide therapy. Paediatrics 1982;70:360–3.
2 Narendra A, White MP, Rolton HA, et al. Nephrocalcinosis in preterm babies.
Arch Dis Child Fetal Neonatal Ed 2001;85:F207–13.
What is already known on this topic
N Preterm babies are at risk of developing nephrocalci-
N Hypercalciuria occurs in some premature children
What this study adds
N Nephrocalcinosis resolves in most cases by 7–8 years
of age and does not cause significant renal impairment
N Prematurity not nephrocalcinosis appears to be a risk
factor for hypercalciuria
3 Jacinto JS, Modanlou H, Crade M, et al. Renal calcification incidence in VLBW Download full-text
infants. Pediatrics 1988;81:31–5.
4 Dayer-Zamara VM, Sauve SR, Yee W, et al. Incidence, risk factors and long
term outcome of nephrocalcinosis in infants born weighing 1250 g or less in
Southern Alberta, Canada. Paediatr Res, 1999;45(4)Part 2, 241.
5 Short A, Cooke RWI. The incidence of renal calcification in preterm infants.
Arch Dis Child 1991;66:412–17.
6 Schell-Fieth EA, Kist-van Hothe JE, Conneman N, et al. Etiology of
nephrocalcinosis in preterm neonates: association of nutritional intake and
urinary parameters. Kidney Int 2000;58:2102–10.
7 Downing GJ, Egelhoff JC, Daily DK, et al. Kidney function in VLBW infants
with furosemide-related renal calcifications at ages 1–2 years. J Paediatr
8 Ezzedeen F, Adelman R, Ahlfors CE. Renal calcification in preterm infants:
pathophysiology and long-term sequelae. J Paediatr 1988;113:532.
9 Pope JC, Trusler LA, Klein AM, et al. The natural history of nephrocalcinosis in
premature infants treated with loop diuretics. J Urol 1996;156:709–12.
10 Jones CA, King S, Shaw NJ, et al. Renal calcification in preterm infants: follow
up at 4–5 years. Arch Dis Child Fetal Neonatal Ed 1997;76:F185–9.
11 Saarela T, Vaarala A, Lanning P, et al. Incidence, ultrasonic patterns and
resolution of nephrocalcinosis in very low birthweight infants. Acta Paediatr
12 Hoppe B, Duran I, Martin A, et al. Nephrocalcinosis in preterm infants: a
single centre experience. Pediatr Nephrol 2002;17:264–8.
13 National High Blood Pressure Education Program Working Group on
Hypertension Control in Children and Adolescents. Update on the 1987 Task
Force report on high BP in children and adolescents: a working group report
from the National High Blood Pressure Education Program. Paediatrics
14 Child Growth Foundation 1996/1.
15 Schwartz GJ, Brion LP, Spitzer A. The use of plasma creatinine clearance for
estimating glomerular filtration rate in infants, children and adolescents.
Paediatr Clin North Am 1987;34:571–90.
16 Payne RB. Renal tubular reabsorption of phosphate (TmP/GFR: indications
and interpretation. Ann Clin Biochem 1998;35:201–6.
17 Han BK, Babcock DS. Sonographic measurements and appearance of normal
kidneys. Am J Radiol 1985;145:611–16.
18 Akashi S, Motizuki H. Screening for hypercalciuria. Acta Paediatr Jpn
19 Jones CA, Bowden LS, Watling R, et al. Hypercalciuria in ex-preterm children,
aged 7–8 years. Pediatr Nephrol 2001;16:665–71.
20 Giapros VI, Androikou S, Cholevas VI, et al. Renal function in
premature infants during aminoglycoside therapy. Pediatr Nephrol
21 Stapleton FB. Hematuria associated with hypercalciuria and hyperuricosuria:
a practical approach. Pediatr Nephrol 1994;8:756–61.
22 Alon US, Warady BA, Hellerstein S. Hypercalciuria in the frequency-dysuria
syndrome of childhood. J Paediatr 1990;116:103–5.
23 Garcia-Nieto V, Ferrandez C, Monge M, et al. Bone mineral density in
paediatric patients with idiopathic hypercalciuria. Pediatr Nephrol
24 Aladangady N, Coen PG, White MP, et al. Urinary excretion of calcium and
phosphate in preterm infants. PediatrNephrol 2004;19:1225–31.
25 White MP, Aladangady N, Rolton HA, et al. Urinary citrate in term and
preterm babies. Early Hum Dev 2005;81:191–5.
F336 Porter, McKie, Beattie, et al