J. Paediatr.

Child Health (2005) 41, 484487

Weight loss and hypernatremia in breast-fed babies:

Frequency in neonates with non-hemolytic jaundice
1
Aylin Tarcan,1 Filiz Tiker,1 Nilgun
Salk Vatandas,1 Aysegul
Haberal2 and Berkan Gurakan

Departments of 1 Pediatrics, and 2 Biochemistry, Faculty of Medicine, Baskent University, Ankara, Turkey
Objective: The aim of this study was to determine what proportion of newborns admitted with idiopathic non-hemolytic
hyperbilirubinemia exhibit severe weight loss and hypernatremia.
Methods: The prospective study involved 115 infants >48 h old who were admitted with jaundice between July 2002 and
July 2003, and had unconjugated bilirubin levels >12 mg/dL. Premature babies (gestational age <37 weeks) and those with
hemolytic jaundice and other pathologic causes of non-hemolytic jaundice were excluded. Postnatal age (days) at admission,
bodyweight at admission, weight change since birth (percentage weight loss calculated at admission) and mode of feeding
(breast-feeding, formula feeding, mixed feeding) were recorded. Severe weight loss was defined in babies who showed >10%
weight loss or had not regained enough to reach birthweight by postnatal day 10. Serum Na levels and breast-milk Na levels
were also measured.
Results: Twenty-eight (33%) of the 86 newborns with idiopathic hyperbilirubinemia in the study exhibited severe weight loss.
Almost all the 86 babies were exclusively breast-fed, and 10 babies (12%) had severe weight loss combined with hypernatremia.
The group with severe weight loss and hypernatremia had higher breast-milk Na levels than the other infants.
Conclusion: The results indicate that a large proportion of babies with non-hemolytic jaundice have severe weight loss,
and that breast-fed newborns with the combination of weight loss and hypernatremia may present with non-hemolytic
jaundice.
Key words: hyperbilirubinemia; hypernatremia; newborn; weight loss.
The combination of weight loss and hypernatremia is a serious
clinical condition that develops in breast-fed babies who receive
inadequate volumes of breast milk. In these cases, insufficient
milk is ingested, and the kidneys are matured to the point where
they can retain sodium ions (Na). In addition to low milk intake
the infant suffers water loss, predominantly via the skin and
lungs.1 The consequences of inadequate intake of breast milk
range from hyperbilirubinemia, hunger and slow weight gain to
life-threatening or even fatal dehydration and starvation.2 The
true rate of weight loss and hypernatremia in breast-fed babies in
non-industrialized countries is not known, but rates of exclusive
breast-feeding are as high as 90%, so this form of dehydration
is thought to be very widespread.
Jaundice is not a common presentation of weight loss and hypernatremia in breast-fed babies; however, breast-feeding jaundice (that due to inadequate breast-feeding) is a frequent cause
of exaggerated physiologic jaundice. It is possible that milder
forms of weight loss and hypernatremia in breast-fed babies are
involved in non-hemolytic hyperbilirubinemia.
The aim of this study was to determine what proportion
of newborns admitted with idiopathic non-hemolytic hyperbilirubinemia exhibit severe weight loss and hypernatremia. Severe weight loss and hypernatremia was defined in babies who
showed >10% weight loss or had not regained enough to reach
birthweight by postnatal day 10 and whose serum Na levels were
146 mEq/L.
METHODS
The prospective study involved 115 infants >48 h old who were
admitted with jaundice between July 2002 and July 2003, and
had unconjugated bilirubin levels >12 mg/dL. All the babies
parents gave informed consent for their child to participate. In

each case, serum was collected for maternal and infant blood
group testing, direct Coombs test, reticulocyte count, and to
measure levels of total and direct bilirubin, thyroid-stimulating
hormone, glucose 6-phosphatase and transaminases. These assessments were done to rule out non-physiologic causes of jaundice. Premature babies (gestational age <37 weeks) and those
with hemolytic jaundice and other pathologic causes of nonhemolytic jaundice were excluded from the study. The birthweight and gestational age of each infant were obtained from
birth records. Postnatal age (days) at admission, bodyweight at
admission, weight change since birth (percentage weight loss
calculated at admission) and mode of feeding (breast-feeding,
formula feeding, mixed feeding) were also recorded. Serum Na
levels and breast-milk Na levels were measured.
Based on the findings, the study population was divided as
follows:
Group 1: Babies who exhibited <10% weight loss or had regained to reach birthweight (minimum) in the period up to
postnatal day 10.
Group 2: Babies who showed >10% weight loss or had not
regained enough to reach birthweight by postnatal day 10,
and whose serum Na levels were <146 mEq/L.
Group 3: Babies with the same weight status as group 2 (severe
weight loss), but whose serum Na levels were 146 mEq/L.
Partial correlation analysis was done to assess relationships
between the following pairs of parameters: percentage weight
loss and serum Na level, percentage weight loss and indirect
bilirubin level, serum Na level and unconjugated bilirubin level,
serum Na level and human milk Na level, percentage weight loss
and human milk Na level. Age was controlled for in all the correlation testing. Scatter plots were drawn to analyse percentage
weight loss in relation to postnatal age for groups 1, 2 and 3;

Correspondence: Dr Aylin Tarcan, Konutkent 1, Safir sok. D35/3 Cayyolu, Ankara 06530, Turkey. Fax: +90 312 2157597; email: aylint@baskentank.edu.tr or aylintarcan@yahoo.com
Accepted for publication 18 April 2005.

Weight loss and hypernatremia in breast-fed babies

485

Table 1 Comparison of study parameters in the three groups

Birthweight (g)
Gestational age (weeks)
Age at admission (postnatal days)
Weight at admission (g)
Weight loss at admission (% loss relative to birthweight)
Feeding method (breast milk/breast milk + formula)
Indirect bilirubin (mg/dL)
Serum Na (mEq/L)
Breast milk Na (mEq/L)

Group 1 (n = 58)

Group 2 (n = 18)

Group 3 (n = 10)

3236 449
3300 (22004000)
38.6 1.1
39 (3741)
6.3 3
6 (320)
3153 448
3190 (21004090)
2.4 4.4
2.9 [(9)13]
53/5
20.1 4.7
19 (1237)
137 5.3
138 (128155)
25.3 14.7
21.5 (478)

3295 584
3400 (22504900)
38.2 1.1
38 (3741)
8.1 3.5
8 (317)
2967 498
2990 (21404390)
9.4 5.4
10.5 [(21)(2.2)]
16/2
19.5 4
19.2 (1229)
136.1 4.7
136 (126145)
23.5 11.7
21 (645)

3214 426
3200 (25004000)
38.4 0.92
38 (3740)
6.2 3.2
5 (312)
2760 451
2810 (20503590)
13.9 3.8
12.5 [(21)(8.5)]
9/1
18.4 2.8
18.7 (1221)
150.4 7.4
147 (146168)
37.7 11.4
39.8 (1960)

Values are given as mean standart deviation median (minimummaximum). Na, sodium ions.

RESULTS
Of the 115 newborns initially included in the study, 29 were
excluded. The reasons for exclusion were prematurity (n = 17),
ABO and Rh-rh incompatibility (n = 8 and n = 2, respectively)
and cephal hematoma (n = 2). Twenty-eight of the babies (33%
of the 86 infants total) had lost more than 10% of their birthweight and had not regained enough to reach birthweight by
postnatal day 10. Of these 28 babies, 18 had serum Na levels <146 mEq/L (group 2, 21% of 86) and 10 had Na levels
146 mEq/L (group 3, 12% of 86 total). Two (2.3%) of the
babies in group 3 had severe hypernatremia, defined as Na
>150 mEq/L (infants values: 160 mEq/L and 168 mEq/L, respectively). Interestingly, five (6%) of the babies were hyponatremic (serum Na <130 mmol/L).
The total number of admissions to the neonatal intensive
care unit at our centre during the study period was 398. The
incidence of combined weight loss/hypernatremia in babies with
non-hemolytic jaundice was 2.5% (10/398). During this period,
two babies (0.5% of the total) were admitted with weight loss
and hypernatremia (Na >150 mEq/L) but without jaundice.
Of the 86 neonates in the study, 78 (91%) were being breastfed exclusively and eight were receiving breast milk plus supplementary formula. None of the infants was being fed formula
exclusively. Two of the mothers of the 10 group-3 babies had
inverted nipples, and one of the infants in this group had urinary tract infection. The descriptive data for the three groups of
babies are summarized in Table 1.
As specified earlier, all the correlation analyses were corrected for age. Testing revealed a significant correlation between
serum Na level and percentage weight loss, with greater weight
loss linked to higher Na levels (r = 0.4, P = 0.001). There
was also a significant positive correlation between breast-milk
Na and serum Na levels (r = 0.25, P = 0.01), and a marginally
significant correlation between breast-milk Na and percentage weight loss (r = 0.20 P = 0.055). However, unconjugated
bilirubin level was not correlated with serum Na level or with
percentage weight loss.

Figure 1 shows a scatter plot of the group data for percentage

weight loss in relation to postnatal age, and Figure 2 shows a
scatter plot of the group data for breast-milk Na and postnatal
age. The lines of best fit in Figure 2 show that group 1 and
group 2 babies had normal breast-milk Na levels according to
their postnatal ages at admission, whereas babies in group 3
had elevated breast-milk Na levels. This was true even for the
group 3 infants who were of older postnatal age at admission.

DISCUSSION
Twenty-eight (33%) of the 86 newborns with idiopathic hyperbilirubinemia in our series exhibited severe weight loss. Ten
(36%) of these 28 babies had serum Na >146 mEq/L, and two
of them had serum Na >150 mEq/L. Therefore, of the 86 total

15

% weight loss

and to analyse postnatal age in relation to breast-milk Na level

for each group. SPSS for Windows version 10.0 was used for all
statistical analyses and to generate all graphs.

15

Group 3
Group 2

25

Group 1
0

10

20

Postnatal age (days)

Fig. 1 Scatter plot of percentage weight loss versus postnatal age for
each group.

486

Fig. 2 Scatter plot of breast milk sodium ions (Na) versus postnatal
age showing the line of best fit for each group.

jaundiced babies, almost all of whom were exclusively breastfed, 12% had severe weight loss combined with hypernatremia.
Geiger et al.3 investigated risk factors in babies who were rehospitalized for jaundice within 14 days after birth. They found
that 24.2% of the infants in this group were dehydrated, and
that 26.8% had difficulty feeding. Two studies4,5 have compared findings in babies with bilirubin levels 12.9 mg/dL versus babies with bilirubin levels <12.9 mg/dL. Both these reports
revealed significantly greater weight loss in the hyperbilirubinemic infants, and both sets of investigators concluded that, according to their results, fasting plays an important role in neonatal
hyperbilirubinemia.
The literature also contains some research on weight loss in
breast-fed babies. In a population screening study, Manganaro
et al.6 observed that 7.7% of healthy, breast-fed, term newborns
experienced weight loss exceeding 10% of birthweight in the
early postnatal period. Only 36% of the neonates with >10%
weight loss had hypernatremia, and the maximum serum Na
concentration in this group was 160 mEq/L. The incidence of the
combination of severe weight loss (>10%) and hypernatremia
in that study was 3%. Oddie et al.7 found that 34 of 904 newborn
babies who were readmitted before 29 days of age showed >10%
weight loss. Eight (0.9%) of these 904 infants had serum Na
>150 mEq/L, although only one was admitted with jaundice.
In contrast, our group of babies readmitted for jaundice had a
higher rate (2.3%) of severe weight loss plus hypernatremia,
with hypernatremia defined as Na >150 mEq/L.
The reported rates of hospital readmission for breast-fed babies with the combination of weight loss and hypernatremia
range from 0.25 to 2.1 per 1000 live births;7,8 however, the
combination of life-threatening dehydration and hypernatremia
is rare.911 None of the 10 babies with breast-feeding failure/hypernatremia that we investigated was in life-threatening
condition, and the only presenting complaint in this group was
jaundice.
Breast-feeding jaundice is a condition of the first week of
life, and most cases are due to failure of breast-feeding in the
days after delivery. The jaundice that results from this problem
aggravates physiologic jaundice. In addition, insufficient milk
intake may delay the passage of meconium (which is rich in
bilirubin), thus leading to increased bilirubin absorption.12,13 The
combination of weight loss and hypernatremia usually presents
after the first week, and this may, in part, be due to the lower

A Tarcan et al.

requirement for fluid and calories in the initial few days of

life. However, prolonged failure of breast-feeding can lead to
hypernatremic dehydration.1,1416
One detailed study examined the relationship between jaundice and low calorie intake.17 Although the findings indicated
that not all breast-fed babies with weight loss and hypernatremia
have jaundice, they showed that babies with this combination of
problems may be readmitted to hospital with exaggerated jaundice. Our study did not include an age-matched control group
without jaundice, so it is impossible to state with certainty that
jaundice was not a coincidental finding in the babies with weight
loss and hypernatremia that we investigated. Another weakness
of our study was the difficulty of discriminating between normal and abnormal weight loss, because the infants who were
admitted for non-hemolytic hyperbilirubinemia were of different ages and it is uncertain what constitutes normal weight
loss in breast-fed infants. For example, a study by MacDonald
et al.18 revealed that healthy breast-fed infants may lose up to
12.8% (97.5th percentile) of birthweight in 10.3 days, and that
they may take as long as 21 days (97.5th percentile) to regain to
birthweight status.
In our study, we found that unconjugated bilirubin levels in
the infants with severe weight loss/hypernatremia were similar
to the levels in the infants without this combination of problems. Although, the difference was not statistically significant,
the newborns with severe weight loss/hypernatremia tended to
have lower unconjugated bilirubin levels. This suggests that hyperbilirubinemia was likely not the primary problem in these
babies; rather, it is likely that hyperbilirubinemia developed secondary to starvation.
Another important finding in our study was that the group
with severe weight loss and hypernatremia had higher breastmilk Na levels than the other infants, and that these levels did
not decline with age, as normally occurs. Research has shown
that Na concentrations in antenatal and postnatal colostrum are
elevated, that the breast-milk Na level falls dramatically by day 3,
and that it continues to decline at a slower rate for at least the
first 6 months of life.19 According to a study by Morton,20 the
normal drop in breast-milk Na is a strong predictor of successful
lactation, and prolonged Na elevation signifies impaired lactogenesis and a high risk of lactation failure. The latter may
also be a simple marker of poor interaction between baby and
breast, as was suggested in a report that noted disparity between
the Na levels in one womans right and left breasts.21 The high
breast-milk Na levels in our babies with severe weight loss and
hypernatremia confirmed lactation failure in that group. Inadequate breast-feeding education and missing follow-up visits
after discharge were the causes of lactation failure.
In conclusion, our results show that severe weight loss is
relatively common in infants who present with jaundice, and that
the combination of weight loss and hypernatremia may present
as non-hemolytic jaundice.

REFERENCES
1 Laing IA, Wong CM. Hypernatremia in the first few days: Is the
incidence rising? Arch. Dis. Child. Fetal Neonatal Ed. 2002; 87:
F15862.
2 Neifert MR. Prevention of breastfeeding tragedies. Pediatr. Clin.
North Am. 2001; 48: 27397.
3 Geiger AM, Petitti DB, Yao JFF. Rehospitalisation for neonatal
jaundice: Risk factors and outcomes. Paediatr. Perinat. Epidemiol.
2001; 15: 3528.
4 Maisels JM, Gifford K, Antle CE, Leib GR. Jaundice in the healthy
newborn infant: A new approach to an old problem. Pediatrics
1988; 81: 50511.

Weight loss and hypernatremia in breast-fed babies

5
6
7
8
9
10
11
12
13

Bertini G, Dani C, Tronchin M, Rubaltelli FF. Is breastfeeding

really favoring early neonatal jaundice. Pediatrics 2001; 107: e41.
Manganaro R, Mami C, Marrone T, Marseglia L, Gemelli M. Incidence of dehydration and hypernatremia in exclusively breastfed
infants. J. Pediatr. 2001; 139: 6735.
Oddie S, Richmond S, Coulthard M. Hypernatremic dehydration
and breastfeeding: A population study. Arch. Dis. Child. 2001; 85:
31820.
Escobar GJ, Gonzales VM, Armstrong MA, Folck BF, Xiong B,
Newman TB. Rehospitalization for neonatal dehydration. Arch.
Pediatr. Adolesc. Med. 2002; 156: 15561.
Ghishan FK, Roloff JS. Malnutrition and hypernatremic dehydration in two breastfed infants. Clin. Pediatr. 1983; 22: 5924.
van Amerangen RH, Monetta AC, Gaeha TJ. Severe hypernatremic
dehydration and death in a breastfed infant. Pediatr. Emerg. Care
2001; 17: 17580.
Bhat SR, Lewis P, Dinakar C. Hypernatremic dehydration in a
neonate. Indian Pediatr. 2001; 38: 11747.
Gartner LM, Herschel M. Jaundice and breastfeeding. Pediatr. Clin.
North Am. 2001; 48: 389400.
Gartner LM, Lee K. Jaundice in the breastfed infant. Clin. Perinatol.
1999; 26: 43146.

487

14 Clarke TA, Markarian M, Griswold W, Mendoza S. Hypernatremic

dehydration resulting from inadequate breastfeeding. Pediatrics
1979; 63: 9312.
15 Arboit JM, Gildengers E. Breastfeeding and hypernatremia. J. Pediatr. 1980; 97: 3356.
16 Cooper WO, Atherton HD, Kahana M, Kotagal U. Increased incidence of severe breastfeeding malnutrition and hypernatremia in a
metropolitan area. Pediatrics 1995; 96: 95760.
17 Whitmer DI, Gollan JL. Mechanisms and significance of fasting and dietary hyperbilirubinemia. Semin. Liver Dis. 1983; 3:
4251.
18 Macdonald PD, Ross SRM, Grant L, Young D. Neonatal weight loss
in breast and formula fed infants. Arch. Dis. Child. Fetal Neonatal
Ed. 2003; 88: F4726.
19 Allen JC, Keller RP, Archer P, Neville MC. Studies in human
lactation: Milk composition and daily secretion rates of macronutrients in the first year of lactation. Am. J. Clin. Nutr. 1991; 54:
6980.
20 Morton JA. The clinical usefulness of breast milk sodium in the
assessment of lactogenesis. Pediatrics 1994; 93: 8026.
21 Chilton LA. Prevention and management of hypernatremic dehydration in breastfed infants. West. J. Med. 1995; 163: 746.