Indian Pediatrics 2001; 38: 1174-1177  

Hypernatremic Dehydration in a Neonate

Swarna Rekha Bhat, P. Lewis, C. Dinakar

From the Department of Nephrology and Urology*, Sanjay Gandhi Post Graduate Institute of Medical Sciences, Lucknow 226 014, India.
Correspondence to: Dr. Sanjeev Gulati, Assistant Professor, Department of Nephrology, Sanjay Gandhi Post Graduate Institute of Medical Sciences, Lucknow 226 014, India.
E-mail: [email protected]

Manuscript received: April 26, 1999;Initial review completed: June 15, 1999; Revision accepted: April 28, 2001.

Hypernatremic dehydration can occur in neonates but is not a common problem. In this article we present a 15 day neonate with hypernatremic dehydration, probably due to inadequate breastfeeding.

A male child was born to a 38-year-old primigravida by LSCS at term gestation. The birth weight was 4.0 kg and APGAR scores were normal. The baby presented on day 15 of life with history of poor feeding and lethargy of one day duration. There was no history of diarrhea, urine output was normal and the baby was on exclusive breast feeds. The mother felt that the baby was feeding well prior to the illness. Baby had developed jaundice on day 4 for which he had received phototherapy for 3 days.

Examination revealed a sick baby who was irritable and lethargic and had signs of moderate dehydration and sclerema. His current weight was 2520 g indicating a weight loss of 37% from the time of birth. RR was 30/min. HR was 110/min and pulses were well felt. Temperature was 36.2ºC. Systemic examination was otherwise normal. A provi-sional diagnosis of sepsis with dehydration was made and the child was admitted to the NICU. Investigation revealed hypernatremia and azotemia (Table I). The initial serum sodium was 185 mEq/L and potassium was 4.8 mEq/L. The blood urea and creatinine at admission were 216 mg/dl and 2.5 mg/dl, respectively. Serum calcium was normal. Blood sugar at the time of admission was 160 mg/dl. ABG at the time of admission showed metabolic acidosis: pH - 7.29, PCO2 - 28.5, PO2 - 81, HCO3– - 13.5 and BE - 11.2. Initial blood counts were suggestive of sepsis: Hb - 14.5 g/dl, TLC - 12,300/cu mm, P66 L34 and platelet count 45,000 cu mm. However blood, CSF and urine cultures were negative. CSF examination was normal, urine routine showed pH - 6.0, specific gravity of 1005 albumin and sugar were nil and WBC were 2-4/HPF.

The baby was managed with IV fluids to correct dehydration and hypernatremia. The baby required 270 ml/kg/day of fluids on day 1 and 240 ml/kg/day on day 2 to correct hydration status. Initially half normal saline was used and this was followed by 1/5th saline with sodium bicarbonate. The serum sodium levels and renal parameters returned to normal by day 4 of admission (Table I). In view of the sclerema and abnormal blood counts cefaperazone and amikacin were used till culture reports were obtained. The baby was on complete oral feeds by day 5 of admission. Investigations were done to deter-mine the etiology of hypernatremia. Low urine specific gravity and a normal urine output in the presence of dehydration made us investigate for diabetes insipidus. The base-line investigations showed a low urine osmolality (81 mosm/kg) in the presence of normal serum osmolality (294 mosm/kg). The sample after fluid restriction showed normal serum (290 mosm/kg) and urine osmolality (320 mosm/kg). Administration of vaso-pressin did not further increase the urine osmolality (397 mosm/kg). By this test central and nephrogenic diabetes insipidus was ruled out. However, with vasopressin administra-tion the baby went into shock and brady-cardia. This was treated successfully with fluid challenge and dopamine. In view of the high blood urea and creatinine a renal scan was done to rule out renal abnormality and this was found to be normal. The baby was discharged on day 12 of admission on breast feeds and cow’s milk. The discharge weight was 2940 g. The final diagnosis was hypernatremic dehydration with acute renal failure. The cause of hypernatremia was thought to be due to inadequate breast feeding. An attempt was made to determine the sodium concentration of the breast milk but this was not possible. On follow up there was no polyuria or dehydration. Electrolytes and renal parameters remained normal. Growth and development of the baby were within normal limits. The last follow up was at 6 months of age.

Hypernatremic dehydration is not common in the neonatal period. The causes of hypernatremia can be due to excessive sodium intake or water deficit in excess of sodium deficit(1). One of the causes of excessive sodium intake in the neonatal period is excessive use of sodium bi-carbonate. Improper preparation of formula feeds and ORS can cause hypernatremia. This baby was on exclusive breast feeds and did not receive bicarbonate prior to admission. Increased insensible water loss especially in premature neonates kept under the radiant warmer can cause hypernatermic dehydra-tion. Central diabetes inspidus (DI) is rare in the neonatal period but can occur secondary to asphyxial injury. There was no history of asphyxia in this child. Nephrogenic diabetes insipidus is rare and can be primary or secondary to renal disease. Nephrogenic and central diabetes insipidus were ruled out in this child by the dehydration and vasopressin test.

There have been reports of hypernatremic dehydration occuring due to inadequate breast feeding(2-6). This can occur because of a combination of poor caloric and fluid intake. Some authors have reported a high sodium content in breast milk(2,4,8) and postulate that hypernatremia can occur because of this. Livingstone et al. have reported 21 neonates with hypernatremic dehydration–all attributable to lactation failure(2). In their study the breast milk sodium content had varied from 11 mmol/liter to 80 mmol/liter and was found to be elevated in 8 of 13 milk samples tested. We could not determine the sodium content in this mother’s milk. There have been other reports of hypernatremic dehydration occuring in breastfed neonates and in one report it was fatal(4). In all the previously reported cases the mother have been elderly promigravida who were keen on breastfeeding. In the current report, the mother was a 38-year-old primigravida keen on feeding the baby. In the previous reports the neonates have been dehydrated with diminished urine output. In our patient, urine output was good despite the presence of dehydration. The good urine output in the presence of hypernatremic dehydration made us investigate for diabetes insipidus. Adequacy of breastfeeding has been based on urine output of atleast 6 times per day. This should not be the only criteria–hydration status and weight should also be evaluated in breastfed neonates.

Breastfeeding should be the mode of feeding in newborn babies though there have been reports of hypernatremic dehydration due to inadequate breastfeeding. This article and other reports should not deter anyone from continuing to promote beastfeeding. However, it is important to ensure adequate hydration and adequacy of urine output in breastfed babies and if weight loss is significant, immediate corrective measures should be adopted.

The principles of management of hyper-natremic dehydration are as follows(1). The underlying cause should be identified and treated. If the baby is in shock it should be corrected with ringer lactate or normal saline challenge. The aim of therapy is to reduce serum sodium by 10-15 mEq/liter/day. The rate of correction is important and hypotonic solution should be avoided as it can cause cerebral edema. Initial fluids should have sodium concentration of 75 mEq/liter. Fluid deficit is calculated based on severity of dehydration or per cent weight loss - 2 to 3% more can be added to the assessed dehydra-tion. Deficit and maintenance requirement for 48 hours should be calculated and this volume should be administered over 48 hours. Ongoing fluid losses should be taken into account while calculating the fluid. The sodium need for 48 hours is 80-100 meq/liter of estimated fluid deficit. Maintenance sodium requirement need not be added to this. Hyperglycemia and hypocalcemia can occur and should be treated. Convulsion in a neonate with hypernatremic dehydration could be due to rapid drop in sodium levels and should be treated with 3% saline.

In conclusion it is important to be aware that hypernatremic dehydration can occur in the neonatal period due to inadequate breastfeeding. It is therefore important to ensure adequate hydration in breastfed neonates. The other causes of hypernatremia should be ruled out before attributing hypernatremia to inadequate breastfeeding.

Contributors: LP and DC were actively involved in the management of the neonate and helped in literature search. SRB was responsible for the management of the patient and writing of the article. She will act as the guarantor for the manuscript.

Funding: None.
Competing interests: None stated.

 

1. Conley SB. Hypernatremia. Pediatr Clin North Am 1990; 37: 365-372.

2. Living Stone VH, Willis CE, Abdel-Warethg LO, Thiossen P, Lockitch G. Neonatal hyper-natremic dehydration associated with breast feeding malnutrition: A retrospective survey. Can Med Assoc J 2000; 162: 647-652.

3. Cooper WO, Athertin HD, Kahans M, Kotagal UR. Increased incidence of severe breast-feeding malnutrition and hypernatremia in a metropolitan area. Pediatrics 1995; 96: 957-960.

4. Kaplan JA, Siegler RW, Schmunk GA. Fatal hypernatremic dehydration in exclusively breastfed newborn infants due to maternal lactation failure. Am J Forensic Med Pathol 1998; 19: 19-22.

5. Roddey OF, Martin ES, Swetenburg RL Critical weight loss and malnutrition in breast-fed infants. Am J Dis Child 1981; 135: 597-599.

6. Rowland TW, Zori RT, Lkafleur WR, Reiter EO. Malnutrition and hypernatremic dehydra-tion in breastfed infants JAMA 1982; 247: 1016-1017.

7. Anand SK, Sandburg C, Robinson RG, Liberman E. Neonatal hypernatremia asso-ciated with elevated sodium concentration of breast milk. J Pediatr 1980; 96: 66-68.

8. Chilton LA. Prevention and management of hypernatremic dehydration in breastfed infants. Western J Med 1995; 163: 74-76.