Acta Pzdiatr Scand 66: 103-109, 1977
ALDOSTERONE AND SODIUM HOMEOSTASIS IN PRETERM INFANTS J . W. HONOUR, H. B. VALMAN and C. H. L. SHACKLETON From the Divisions of Clinical Chemistry and Perinatal Medicine, Clinical Research Centre, Harrow, Middlesex, U . K .
ABSTRACT. Honour, J. W., Valman, H. B. and Shackleton, C. H. L. (Divisions of Clinical Chemistry and Perinatal Medicine, Clinical Research Centre, Harrow, U.K.). Aldosterone and sodium homeostasis in preterm infants. Acta Paediatr Scand, 66: 103, 1977.-A specific mass spectrometric method was used for tetrahydroaldosterone determination in urine of preterm infants (26-34 weeks gestational age) up to 9 weeks of age. Hyponatraemia during the first 2 weeks of life was associated with an excretion of tetrahydroaldosterone (5-50 pg/24 h) comparable with full-term infants. Excretion of tetrahydroaldosterone was significantly elevated in all infants studied during the third week of life (80-350 &24 h) and this was associated with establishment of positive sodium balance. The excretion of tetrahydroaldosterone remained high for 2 or 3 weeks. The results are discussed in relation to the development of renal tubules and control mechanisms for sodium homeostasis. KEY WORDS: Preterm infants, sodium homeostatis, tetrahydroaldosterone excretion
Sulyok reported that plasma sodium levels in healthy preterm infants fall during the first 2 weeks of life, reaching a minimum level (mean 133 mmol/l) on about the 17th day (21). A negative salt balance was observed during this period. Thereafter a positive sodium balance was restored and plasma sodium levels returned to normal. It has been generally accepted that these features of sodium homeostasis are caused by the immaturity of the renal tubules (21, 22). The role of aldosterone in the development of sodium homeostasis in preterm infants is not well understood. Other investigators have assumed that the low aldosterone secretion rate during the first week of life (23) precluded a significant role for aldosterone in sodium retention during this period. Continuous studies for longer periods have not been reported. A study of urinary tetrahydroaldosterone excretion during the first weeks of life in preterm infants is being undertaken in this laboratory. A preliminary communication reported
elevated tetrahydroaldosterone excretions in twin 14 day-old infants delivered at 29 weeks gestation (8). The values reported (60-225 pg/24 h) were greatly in excess of those found for normal full-term infants in this study and by other investigators (13). In the present report the aldosterone status has been studied in relation to the sodium balance of preterm infants of gestational ages 26-34 weeks. PATIENTS AND METHODS Nine newborn infants of gestational age 26-34 weeks have been studied (Table 1). All deliveries (except Case 3) were spontaneous and premature. Infant 3 was born by elective Caesarian section carried out due to concern arising from rapidly rising maternal blood pressure. Age was calculated from the mother’s last menstrual period and confirmed by clinical examination (5). In general, infants received breast milk collected from mothers in the maternity wards. Commercial preparations where used were similar in electrolyte content to human milk. Details of fluid and milk intake were recorded. Twenty four hour specimens of urine were generally collected in plastic paediatric urine bags, although in a few cases disposable napkins Acta
PEdiatr Scand 66
. I . W . Honour et al.
Table 1. Characteristics ofstudy group Case
Sex M M M F F F F M M
Birth weight (g)
Gestational Minimum age weight (weeks) (g)
1060 1 130 1 520 1 200 960 920 1 200 1500 1030
27 29 34 31.5 31.5 27 29 34 26
were used. In a separate study (to be published) we have shown similar steroid excretion patterns in urine collected in a bag for 24 h, and an extract prepared by centrifugation of disposable napkins used during the next 24 h period. Capillary blood samples were taken by heel puncture. Informed consent was obtained from the parents for execution of detailed sodium balance studies. Male infants were selected because of the relative ease of obtaining accurate 24 h urine collections. The normal range for plasma sodium concentration in full-term infants at this hospital is 135-145 mmol/l and for the purpose of this study hyponatraemia was defined as a persistent concentration below this range. A weight loss after birth was observed in all infants with minimum weights recorded on the 4th or 5th day. Birth weights were regained between days 7 and 18 (Table I). The infants were discharged from hospital when they had reached a weight of 2.5-3 kg and subsequent progress has been uneventful. The prenatal and intrapartum histories of all mothers were closely scrutinised. All were on unrestricted salt diets and received no diuretics during the later stages of pregnancy. Analytical methods Sodium and potassium concentrations were determined simultaneously by twin channel flame photometry. Lithium was added to the samples for use as internal standard. Urinary tetrahydroaldosterone was measured by a gas chromatographic-mass spectrometric (GC-MS) method (9, 16, 17). This steroid is excreted in urine as a 3glucosiduronate and had to be released from conjugation by enzymic hydrolysis (digestive juice of the snail, Helix pomatia). An internal standard of 3pallo-tetrahydroaldosterone was added prior to steroid extraction on Amberlite XAD-2 resin columns. The extracts were fractionated on small Sephadex LH-20 columns, and methyloxime-trimethylsilyl ethers were prepared of the fraction containing tetrahydroaldosterone. The derivatives were analysed by GC-MS using a Varian MAT 731 instrument in a selective ion monitoring mode. The fragment ions of mass to charge ratio 638 (molecular ion M') and 607 (M+-31) were alternately monitored at 1 sec intervals. Acto Pcediatr Scond 66
Days for birth weight to be regained
1 470 1 100
15 15 7 -
1 060 1 400 960
The peak height of the m / e 607 ion of urinary tetrahydroaldosterone was measured relative to the peak height of the equivalent ion derived from the internal standard separated by gas chromatography. The ion at mle 638 was only monitored to confirm the specificity of the determination. The results were calculated according to the formula:
X= Urinary tetrahydroaldosterone excretion (pg/24 h)
Hx=Peak height urinary tetrahydroaldosterone H s =Peak height 3pallo-tetrahydroaldosterone 0Sodium balance
mmol/kglday U r i n a r y sodium O excretion -I mmollkglday -2
zw 1 Tetrahydroaldosterone W a y
Fig. I . Postnatal values of sodium intake and excretion, plasma sodium and potassium concentrations, and urinary tetrahydroaldosterone in a preterm infant delivered at 27 weeks gestation (Case I ) . The sodium balance is positive above the datum line of the histogram and negative below this line.
Aldosterone and sodium homeostasis in preterm infants -
O S o d i u m balance
Sodium intake mmollkqiday Urinary sodium excretion -1 m mo likqiday
Normal ranoe for fiill-term infants
403 7 Tetrahydroaldosterone W a y
Fig. 2. Postnatal values of sodium intake and excretion, plasma electrolyte concentrations and urinary tetrahydroaldosterone in a preterm infant delivered at 29 weeks gestation (Case 2). The urinary excretion of aldosterone is the sum of free aldosterone and aldosterone 18glucosiduronate.
S =Weight of internal standard (pg) R =Response factor=HsilHxl
Hx'; H,'=Calibrating standards. Peak heights of identical weights of reference tetrahydroaldosterone and 3pd o -tetrahydroaldosterone respectively. V =24 h urine volume (ml) U=Volume of portion analysed (ml) Urinary aldosterone (free and 18-glucosiduronate) was measured by radioimmunoassay following pH 1 hydrolysis and chromatography (12).
RESULTS Fig. 1 illustrates the results of the detailed investigation of a male infant (Case 1 ) delivered at 27 weeks gestation. On the 8th day of life urinary sodium excretion exceeded dietary intake by 1 mmol/kg/day. A positive sodium balance was achieved by the 25th day of life. Little significance can be attributed to the small positive .balance observed on day 14 since in this study sodium loss other than in
urine (e.g. faeces or sweat) were not determined. Potassium intakes were around 3.2 mmol/kg/day and during the first 5 weeks of life potassium excretion did not exceed 1.6 mmol/kg/day with positive balances of around 2.0 mmol/kg/day. On the 40th day, potassium excretion rose to 2.6 mmol/kg/day and the ratio of urinary concentration of sodium to potassium fell to 0.46 from 1.30 on day 32. This infant was hyponatraemic and hyperkalaemic during the first 4 weeks of life; thereafter the plasma sodium and potassium concentrations approached normal levels for full-term infants. The infant regained birth weight by day 17. Tetrahydroaldosterone excretion during the first 15 days of life remained within the upper limits of normal for a full-term infant. A significant rise was recorded between days 14 and 18 and levels remained high for 2-3 weeks. Figs. 2 and 3 show similar data for infants delivered at 29 and 34 weeks gestation (Case 2 and 3 ) . In both these patients positive sodium balance was observed by day 14. Unfortunately sodium balance studies and urinary tetrahydroaldosterone excretions were
;w 0 S o d i u m balance
Sodium intake mmollkqiday
U r i n a r y sodium excretion -1 rnmolikqiday -2
Normal range for f u l l term infants
Plasma sodium 140 mmolilitre 130 120 Plasma potassium mmolllitre
Urinary tetrahydroaldosterone wldv
Fig. 3. Postnatal values of sodium intake and excretion, plasma electrolyte concentrations, and urinary tetrahydroaldosterone in a preterm infant delivered at 34 weeks gestation (Case 3 ) . Acta Pzdiatr Scand 66
J . W . Honour el al. ocase 4 ,case 5 AcaJeS 6 7 & 8
Plasma sodium mmalilitre
150 140 130 120
Plasma potassium mmalllitre
Fig. 4. Urinary tetrahydroaldosterone excretion and plasma electrolyte concentrations in relation to postnatal age of 5 preterm infants.
Values obtained during the first 10 days were low but a significant increase was observed after 15th day. The plasma sodium concentrations were low in both infants during the first 6 weeks of life but approached normal values by the 8th week. The trend of plasma potassium levels was reversed. Cases 6, 7 and 8 were only studied for one day but the results were comparable to those obtained at the same age for the infants studied more extensively. Daily sodium balance studies and urinary tetrahydroaldosterone excretions were obtained for an infant born at 26 weeks gestational age (Case 9 ) . The results obtained are illustrated in Fig. 5 and demonstrate a severe negative balance throughout the study. Sodium intake includes the sodium content of all fluids administered over the 24 h period. Fluid injected intravenously varied considerably in volume and between the 5th and 8th day of life were particularly high. Tetrahydroaldosterone excretion was extremely low during the first week of life and did not respond to the negative salt balance. An increase was observed in the 2nd week although the high excretions ob-
not obtained for Case 2 during the first 2 weeks of life. A normal daily intake of potassium for infant 2 was 2.7 mmol/kg/day. Potassium balances on days 16, 24 and 32 were 0.6, 2.1 and 1.2 mmol/kg/day. An increase in daily excretion of potassium was not observed but the ratio of sodium to potassium concentration in urine fell from 0.48 on day 16 to 0.09 at day tx O S o d i u m balance 30. Urinary aldosterone excretions were deterintake tb mined on this patient and the levels were Sodium mmolikgiday t4 higher than those found for normal infants 2 0 studied in this laboratory ( 1 to 2 pg/24 h). output -' Plasma sodium concentrations increased dur- Sodium rnmolikglday -4 ing the period of study. In one of the patients 6 (Case 2) the plasma potassium concentration 8 140 a appeared also to increase but on this occasion a Plasma radium 130 results may have been unreliable because of mrnolllitre 120 a a haemolysis. 110 Results of plasma electrolyte and tetrahydroaldosterone excretions for several other Tetrahydroinfants are illustrated in Fig; 4. Since urinary extracts were occasionally obtained from 0 1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 1 3 1 4 1 5 disposable napkins it was not possible to deAge (days) termine sodium balance because urinary sodium was not assayed. Tetrahydroaldosterone Fig. 5 . Daily changes in sodium intake and excretion in relation to plasma sodium concentrations and urinary excretion in Cases 4 and 5 show a'similar tetrahydroaldosterone excretion. The infant (Case 9 ) was pattern to that of the infants described above. delivered at 26 weeks gestation.
Actu Pazdiatr Scand 66
Aldosterone and sodium homeostasis in preterm infants
tained in the other cases were not achieved. This increased secretion of aldosterone was evidently insufficient to arrest the sodium loss and death occurred on the 15th day of life. DISCUSSION Aldosterone can be synthesised by the foetal adrenal from early in gestion. Dufau & Villee (6) achieved a synthesis of aldosterone from progesterone by adrenal homogenates from foetuses 15 weeks old and Pasqualini et al. (16) demonstrated that perfusion of a 20 week old foetus with labelled corticosterone resulted in aldosterone synthesis. It may therefore be concluded that viable preterm infants are capable of aldosterone synthesis. A significant amount of aldosterone circulating in the full-term foetus at birth is of foetal origin and relatively high levels persist in the newborn for several days (2, 3). Aldosterone catabolism is considered to be similar in fullterm neonates and adults since New and co-workers (13) found comparable proportions of urinary free aldosterone, its 18-glucosiduronate and tetrahydroaldosterone glucosiduronate. However, the problems of determining aldosterone production are exacerbated when preterm infants are being studied since metabolising enzymes in the liver and kidney may be immature, resulting in altered steroid catabolism. Premature infants aged 33-75 days appear to excrete a disproportionately high amount of “free” aldosterone compared to the other aldosterone metabolites (13). The high tetrahydroaldosterone excretions demonstrated in the present study provide in vivo evidence for the capacity of preterm infants, at a developmental age of 28 weeks gestation, to catabolise aldosterone in a similar manner to an adult. In one of our patients (Case 2 ) aldosterone (free plus 18-glucosiduronate) was determined in urine as well as tetrahydroaldosterone. The ratio of tetrahydroaldosterone to aldosterone was generally greater than 20 : 1 which compares with a ratio of about 2 : 1 for the
full-term newborns studied by New et al. (13). However, our results are not directly comparable with those of New and co-workers who studied newborns only during the first 24 h of life when urine may include metabolites of maternal aldosterone. Although the ratio of the individual aldosterone metabolites excreted by preterm infants, full-term infants and adults appears to be at variance, there is no doubt that tetrahydroaldosterone is the major metabolite of aldosterone in preterm infants and must reflect aldosterone secretion. The urinary tetrahydroaldosterone excretion during the first week of life in premature infants were generally slightly higher than those found for full-term neonates (14, 19) and increased considerably around the 15th day. Although the early tetrah ydroaldosterone excretions are higher than adults when adjusted for body surface area, the increase in aldosterone secretion suggests a transient period of mineralocorticoid deficiency after birth. The high renin activity in the first weeks of life of full-term neonates reported by Kotchen et al. (10) suggests a lower responsiveness of aldosterone synthesis to renin-angiotensin stimulation during this period and may explain why aldosterone secretion is not significantly stimulated by hyponatraemia when associated with the presence of a negative salt balance. Siege1 et al. (19) found a high level of plasma aldosterone in preterm infants during the first few days of life and state that this finding precludes aldosterone insufficiency being responsible for the sodium wasting and hyponatraemia observed. If their patients had been studied for a longer period even higher plasma aldosterone levels would almost certainly have been found and their conclusions invalidated. These authors have assumed that high aldosterone levels are synonomous with high activity. High plasma aldosterone concentrations may reflect a compensation for lower response of the renal tubules to the hormone or are required to counteract an aldosterone antagonist. The newborn synthesises many Actu Pzdiutr Srand 66
J . W . Honour et al.
steroids in large amounts which may antagonise the action of aldosterone on the renal tubules. For example 16a-hydroxyprogesterone has a natriuretic effect in adults (1 1) and while this particular steroid may not be resposible for opposing the activity of aldosterone in newborns, many similar steroids are secreted by the adrenals in large amounts. It is not known whether the increased aldosterone secretion in the third week of life reflects a response to increased stimulation or to improvement in the ability of the adrenals to synthesise larger amounts. The renal tubules are assumed to respond to aldosterone at this stage since a positive sodium balance is established with the consequent concentrations of sodium and potassium in plasma. However, the renal tubules cannot be responding optimally to aldosterone since sodium is still excreted in urine when plasma sodium concentration remains low and a direct renal exchange of sodium with potassium was not demonstrated until the 6th week of life. Insufficient sodium reabsorption may be due to earlier maturation of renal glomeruli resulting in an imbalance between tubule and glornerular mass (7). Aperia et al. (1) consider that the relatively high fractional sodium excretion in preterm infants may be attributed to an inadequate tubular surface area. These investigators concluded that aldosterone was not important in tubular sodium reabsorption on the basis of the low aldosterone secretion rates reported by Weldon and co-workers (23). In spite of sodium retention the aldosterone secretion may subsequently continue to be high in response to the plasma sodium concentration remaining below normal or to hypertrophy of the zona glomerulosa. The sodium balance figures reflect largely the amount of sodium used in growth. From the third week of life, the figures observed (1-2 mmol/kg/ day) agree with the rates of accumulation of sodium by human foetuses in relation to gestation (18). Further studies involving determinations of Actu Pzdiatr Scund 66
aldosterone metabolism and renin activity are required to resolve the outstanding problems of sodium regulation in infancy. A method for the analysis of renin activity in small amounts of blood has recently been published by Dillon (4)and determinations of renin activity using this method are to be included in future investigations. ACKNOWLEDGEMENTS We are indebted to Miss D. Davis and the nursing staff of the Special Care Baby Unit, Northwick Park Hospital, for their co-operation in these studies. The determinations of urinary aldosterone were kindly carried out by C . E. Horth of Searle Diagnostics.
REFERENCES 1. Aperia, A., Broberger, O., Thodenius, K . & Zetterstrom, R.: Developmental study of the renal response to an oral salt load in preterm infants. Acta Paediatr Scand, 63: 517, 1974. 2. Bayard, F., Ances, I. G . , Tapper, A. J . , Weldon, V. V., Kowarski, A. & Migeon, C . J.: Transplacental passage and fetal secretion of aldosterone. J Clin Invest, 49: 1389, 1970. 3. Beitins, 1. Z., Bayard, F., Levitsky, L., Ances, I. G . , Kowarski, A. & Migeon, C. J.: Plasma aldosterone concentration at delivery and during the newborn period. J Clin Invest, 51: 386, 1972. 4. Dillon, M. J.: Measurement of plasma renin activity by semi-micro radioimmunoassay of generated angiotensin. J Clin Pathol, 28: 625, 1975. 5. Dubovitz, L. M. S . , Dubovitz, V. & Goldberg, C . : Clinical assessment of gestational age in the newborn infant. JPediatr, 77: 1 , 1970. 6 . Dufau, M. & Villee, D. B.: Aldosterone biosynthesis by human fetal adrenal in vitro. Biochim Biophys Acta, 176: 637, 1969. 7. Fetterman, G. H., Shuplock, N. A., Philop, F. J. & Gregg, H. S . : The growth and maturation of human glomeruli and proximal convolutions from term to adulthood. Studies by microdissection. Pediatrics, 35:601, 1965. 8 . Honour, J. W, Valman, H . B. & Shackleton, C. H. L.: Sodium homeostasis in preterm infants. Lancet, I I : 1147, 1974. 9. Honour, J . W. & Shackleton, C . H. L.: Quantitative determination of urinary tetrahydroaldosterone by selected ion monitoring gas chromatography-mass spectrometry. 1976, in press. 10. Kotchen, T. A,, Strickland, A. L . , Rice, T. W. & Walters, D. R.: A study of the renin-angiotensin system in newborn infants. JPediatr, 80: 938, 1972. 1 1 . Jacobs, D. R.: Natriuretic activity of 16a-hydroxyprogesterone in man. Acta Endocrinol ( K b H ) , 61: 275, 1%9.
Aldosterone and sodium homeostasis in preterm infants 12. Mayes, D., Furuyama, S . , Kem, D. C., Nugen, C. A,: A radioimmunoassay for plasma aldosterone. J Clin Endocrinol, 30: 682, 1970. 13. New, M. I . , Miller, B., Peterson, R. E.: Aldosterone excretion in normal children and children with adrenal hyperplasia. J Clin Invest, 45: 412, 1966. 14. Nielson, M. D., Lund, J. 0. & Munck, 0.: Urinary excretion of tetrahydroaldosterone in normal subjects and in patients with adrenal insufficiency and Conn’s syndrome. Acta Endocrinol ( K b H ) , 71 :498, 1972. 15. Pasqualini, J. R., Wiqvist, N. & Diczfalusy, E.: Biosynthesis of aldosterone by human fetuses perfused with corticosterone at midterm. Biochim Biophys Acta, 121:430, 1966. 16. Shackleton, C. H. L. & Honour, J. W.: The use of combined gas chromatography-mass spectrometry for the quantitative measurement of urinary tetrahydroaldosterone. Z Klin Chem Klin Biochem, 12:259, 1974. 17. Shackleton, C. H. L. & Snodgrass, G . H. A. I.: Steroid excretion by an infant with an unusual salt losing syndrome. Ann Clin Biochem, 11: 91, 1974. 18. Shaw, J. C. L.: Parenteral nutrition in the management of sick low-birthweight infants. Pediatr Clin N A m , 20:333, 1973. 19. Siege], S . R., Fisher, D. A. & Oh, W.: Renal func-
tion and serum aldosterone levels in infants with respiratory distress syndrome. J Pediurr, 83: 854, 1973. Solc, J. & Knorr, D.: Untersuchungen uber die Tetrahydroaldosteron Ausscheidung im Kindesalter durch Gas Chromatographie mit Elektroneinfang detektor. Acta Endocrinol (Kbh), 70: 533, 1972. Sulyok, E.: The relationship between electrolyte and acid-base balance in the premature infant during early post-natal life. Biol Neonate, 17: 227, 1971. Thodenius, K.: Renal control of sodium homeostasis in infancy. Acta Paediatr Scand, Suppl. 243, 1974. Weldon, V. V . , Kowarski, A. & Migeon, C. J.: Aldosterone secretion rates in normal subjects from infancy to adulthood. Pediatrics, 39: 713 (1967).
Submitted April 20, 1976 Accepted July 30, 1976
(J. W. H.) Division of Clinical Chemistry Clinical Research Centre Watford Road Harrow Middlesex HA1 3UJ England