Abstract
Despite the fact that hundreds of thousands of preterm infants receive parenteral fluids each year, study of optimal fluid and electrolyte management in this population is limited. Compared to older children and adults, preterm infants have an impaired capacity to regulate water and electrolyte balance. Appropriate fluid and electrolyte management is critical for optimal care of low birth weight or sick infants, as fluid overload and electrolyte abnormalities pose significant morbidity. This review highlights basic physiological principles which need to be applied when prescribing parenteral fluids and builds upon published literature to outline a rational approach to initial fluid and electrolyte management of the preterm infant.
Introduction
The provision of water and electrolytes is one of the most basic and longest standing tenets of care of the preterm infant. However, clinical approaches of fluid management continue to evolve. Into the 1960’s, it was common for fluids to be withheld from premature infants during the first 48 of life for concerns of worsening pulmonary pathology [1]. Recognition of the frequent development of hypoglycemia, hyperbilirubinemia and hypernatremia in fasted infants, coupled with evidence of limited impact on respiratory disease, led to the earlier administration of fluid and nutrition in preterm infants [2]. Over the subsequent 50–60 years, the practice of neonatology has included extensive study of neonatal fluid and electrolyte homeostasis. Despite this work, there have been relatively few clinical trials which have sought to identify optimal fluid and electrolyte management strategies [3]. Maintaining water and sodium balance following preterm birth poses a unique challenge due to high insensible water losses and immaturity of the kidneys, which lack fully functional regulatory systems. This review seeks to highlight investigative work and provide a physiological understanding of contemporary fluid and electrolyte management of preterm infants.
Fluid homeostasis
Water is the major component of cells and tissues, being responsible for about 75% of an infant’s weight at term birth, and a greater percentage in preterm infants [4]. Bent Friis-Hansen called water “the major nutrient,” proclaiming “water is the basis for the physical and chemical conditions of life . . . .the whole organic would as we know it, is built up around the properties of water [5].” The initial prescription of fluids, which most preterm infants require, and influences short- and long-term health outcomes, depends upon many factors, including birthweight, gestational age, housing environment and local practices [6–9].
The quantity of water necessary to provide during the first few days of life in otherwise healthy preterm infants is determined by balance of the following formula:
Because metabolically derived water, fecal water loss and water required for growth are negligible (<10 ml/kg) immediately after birth, the equation can be simplified:
To determine the appropriate water intake to prescribe, one needs to understand IWL, urine water requirements and appropriate postnatal changes in body water compartments after birth.
Insensible water losses
Insensible water loss includes evaporative losses from the skin and respiratory tract (Table 1). In preterm infants, approximately 70–85% of IWL normally occurs through skin losses, with the remaining 15–30% lost through the respiratory tract. In term infants, skin and respiratory IWL are about equal [10, 11]. Skin IWL, or transepidermal water loss (TEWL) constitutes a loss of body heat and is a function of energy expenditure. TEWL is inversely proportional to birth weight and gestational age, a function of the increased surface area to body weight ratio as birth weight decreases, as well as thinner skin and greater skin blood flow present in increasingly premature infants [12, 13]. In fact, TEWL increases exponentially with decreasing gestational age, such that TEWL of a 24-week gestation infant may be 10–15 times greater than that of a term neonate on the first day of life [12]. Agren et al reported 23–27 week gestation infants cared for in 50% relative humidity have a TEWL of approximately 60 g/m2/h during the first few days of life (150 ml/kg/d), decreasing to about half that at 7 days of age [14, 15]. Skin evaporative water loss is significantly reduced by increasing ambient relative humidity; this inverse relationship being greatest in extremely preterm infants. As gestational and postnatal ages increase, TEWL decreases, though remains elevated on a per kg basis in premature compared to term infants even beyond a month of age. Interestingly, caring for preterm infants in higher levels of humidity may delay postnatal skin barrier maturation [15]. Additional factors, including exposure to nonionizing radiant energy, as occurs from radiant warmers or some forms of phototherapy, may further increase skin IWL. In contrast, use of plastic heat shields or blankets, semipermeable membranes and topical agents which act as skin barriers reduce IWL [16].
Estimated insensible water losses of preterm infants during first 24–48 hours of life cared for in differing environments
Estimated insensible water losses of preterm infants during first 24–48 hours of life cared for in differing environments
Respiratory water loss is dependent upon respiratory rate and tidal volume, as well as air temperature and humidity. Respiratory IWL decreases significantly when breathing warmed, humidified air, as may occur with infants cared for in incubators and may approach zero for infants being mechanically ventilated with high humidity gas at body temperature [17]. Increased minute ventilation, which may occur as a result of cardiopulmonary dysfunction, metabolic acidosis, increased body temperature or increased motor activity (crying) results in higher respiratory IWL.
Urine production is necessitated by the need to excrete soluble waste products. This renal solute load is typically of dietary origin, related to nitrogen (protein) and electrolyte content. The potential renal solute load (RSL) is calculated by the equation:
Postnatal adaptation of body water compartments
Infants born prematurely have higher total body water (TBW) and extracellular water (ECW) per kilogram than do term infants [23]. After birth, TBW content falls, primarily due to contraction of the extracellular space [24]. The degree of contraction of the ECW compartment is inversely proportional to gestational age and greater in small for gestational age than appropriate for gestational age infants. Preterm infants may exhibit a 10–15% weight loss during the first week of life related to loss of ECW, whereas term infants generally exhibit a 5–7% weight loss [25, 26]. While the optimal amount of weight loss is not known, failure to allow the normal postnatal contraction in ECW and weight loss in preterm infants may increase the risk of patent ductus arteriosus, necrotizing enterocolitis, and bronchopulmonary dysplasia [3, 27–29].
Maintenance fluid therapy
With this understanding, one can estimate initial total water requirements for the preterm neonate. Consider the 28 weeks gestation, 1000 g infant who in the first 24 hours of life receives parenteral fluids with minimal electrolytes (2 mEq/kg NaCl, no KCl) and two grams/kg protein. The infant needs to excrete a solute load of approximately 16 mOsm/kg (4mOsm/kg from NaCl intake+approximately 12 mOsm/kg of urea from protein intake). To excrete this load with a urine osmolarity of 300 mOsm/L, a urine volume of approximately 50 ml/kg is needed. Estimating an IWL of 60 ml/kg/d (incubator with a relative humidity of 50%) and allowing for a 2% weight loss (20 ml), the resulting prescribed water intake would be 90 ml/kg (50 ml/kg for renal solute excretion+60 ml IWL –20 ml (2% weight loss)). If cared for on an open radiant warmer, estimated IWL would be increased 40–50% and total intake adjusted accordingly.
Ongoing fluid requirements are dependent upon the physiological maturity of the infant, nutritional intake goals and environmental factors. A short-term goal of therapy is achieving a 6–12% loss of weight from birth over the first few days of life, suggesting appropriate contraction of the extracellular space. Careful attention is directed to serial determinations of serum sodium and chloride concentrations, which provide an assessment of TBW status. Hypernatremia (serum sodium concentration > 150 mEq/L), most frequently caused by an intake of water insufficient to meet ongoing losses rather than excessive administration of sodium, should be avoided. Conversely, hyponatremia in the first few days of life frequently results from an excessive supply of water rather than deficient sodium intake.
With an understanding the physiological rationale for initial fluid therapy in the preterm infant, it is important to incorporate knowledge gained from clinical trials in prescribing fluid for preterm infants. Unfortunately, the study of such is limited. Bell et al.randomized 170 infants with birth weights between 750 and 2000 grams to a “high” or “low” maintenance fluid therapy beginning on day of life three and continuing until 30 days of age [28]. Fluid goals in each group depended upon birth weight, postnatal age and environmental factors, urine and stool water loss, IWL and water for growth. Infants in the “high” fluid group (169 ± 20 ml/kg/d over study period) had increased incidence of patent ductus arteriosus (PDA) with or without congestive heart failure and necrotizing enterocolitis compared to “low” fluid group (122 ± ml/kg/d). The authors concluded that “limitation of fluid intake to amounts carefully calculated to provide for solute excretion, stool losses, insensible water loss, and growth can reduce the risks of both patent ductus arteriosus and necrotizing enterocolitis, two of the most common and serious complications of prematurity.” Thus, too much fluid is detrimental, though the optimal amount was not identified.
Lorenz et al.randomized 88 infants with birthweights 750–1500 grams to a fluid intake protocol over the first five days of life allowing a 1 to 2% loss of birthweight per day to a maximum loss of 8 to 10% (group 1), or fluids allowing 3 to 5% loss of birthweight per day to a maximum loss of 13 to 15% (group 2) [25]. Although mean cumulative fluid intakes for the two groups over the first five days of life differed by 220 ml/kg, their mean cumulative weight losses differed by only 41 gm/kg, reflecting the infants’ ability to maintain water balance by regulating urine osmolarity and thus water excretion. No significant differences in the rates of PDA, bronchopulmonary dysplasia, intraventricular hemorrhage, necrotizing enterocolitis or mortality were identified. Kavvadia et al.assigned 168 infants, 23–33 week gestation, to receive either “standard” volumes of fluid, starting with 60–70 ml/kg/d and progressing to 150 ml/kg/d by day of life 7 or restricted to 80% of standard [30]. Mean weight loss in both groups was similar, being approximately 8% of birth weight of day of life seven. No significant differences in any short- or long-term morbidities were observed. Randomizing 100 infants < 1750 grams to a “dry” group (50 ml/kg/d increasing to 120 ml/kg/d over first week and 150 ml/kg/d until four weeks of age) or “control” group (80 up to 150 ml/kg/d over first week and 200 ml/kg/d until 4 weeks of age), Temmela et al.found the dry group had significantly lower rates of death and BPD [9]. Palta et al.demonstrated that in infants with moderate respiratory disease at birth, increased fluid intake in the first four days of life was associated with oxygen dependency at 28 d[8]. In a post-hoc analysis of a large parenteral glutamine supplementation trial, higher fluid intakes and lack of postnatal weight loss were significantly associated with increased risk of death or BPD [7]. Taken together, the above studies suggest fluid regimens that allow for contraction of the extracellular compartment with resultant weight loss of 6–12% of birth weight, and avoidance of hypernatremia (plasma sodium > 150 mEq/L) likely result in optimal outcomes for premature infants. Because of the ability to regulate urine osmolarity, and thus free water loss, there is a range of acceptable fluid intakes early in life, which in infants with normal renal function, result in this targeted loss of weight.
Sodium homeostasis
Sodium is the major cation of extracellular fluid, which includes blood plasma and interstitial fluid. Over 80 years ago, James Gamble highlighted the importance of the extracellular fluid in maintaining the “stability of physicochemical conditions within the organism” and emphasized that the “innumerable and interrelated chemical reactions that together accomplish what we call metabolism would rapidly fall out of adjustment, if these underlying conditions were not held at fairly constant values [31].” Thus, understanding the changing physiological sodium requirements of neonate that occur with advancing gestational age and postnatal age is critical for those prescribing parenteral fluids to provide optimal care.
Immediate postnatal sodium requirements
During the initial two to three days after birth, contraction of the extracellular fluid compartment is manifested by natriuresis and diuresis. Negative sodium balance is achieved during this time by obligate renal sodium losses related to renal tubular immaturity, and in breastfed infants by the limited volume of breastmilk coupled with the low sodium content of breastmilk [32, 33]. For term infants requiring parenteral fluid therapy, initial therapy should, with few exceptions, consist of dextrose containing fluids devoid of electrolytes to allow for contraction of the extracellular fluid space.
Although the initial fluid requirements of preterm infants are greater than those of term infants, there is little reason to provide sodium in the first 24–48 hours of life and the use of electrolyte free nutritional solutions is recommended. While moderate intake of sodium (<3 mEq/kg/d) does not impede postnatal contraction of the extracellular space, preterm infants 25–31 weeks gestation are at an increased risk for hypernatremia resulting from the loss of free water in excess of sodium [26]. In infants 25–30 weeks gestation, delaying sodium supplementation until a weight loss of 6% of birth weight compared to initiating sodium administration of 4 mmol/k/d on the second day of life resulted in a significantly greater percentage of infants on room air at seven days of life without changes in time to regain birth weight and weight at 36 weeks and six months of postmenstrual age [34]. In extremely low birth weight infants (<1000 grams), Costarino et al. compared sodium restriction (<1 mEq/kg/d) in the first 5 days of life with sodium supplementation (3–4 mEq/kg/d) [35]. Sodium restriction was associated with decreased incidence of hypernatremia and bronchopulmonary dysplasia. Recognizing that avoidance of sodium administration is difficult as many centers administer sodium containing solutions through umbilical arterial catheters, results from these limited studies suggest that concerted efforts to limit sodium intake in the first 48 hours of life in preterm infants are appropriate.
Sodium needs for growth
Beyond the first few days after birth, sodium requirements increase secondary to both sodium accretion necessary for growth, and ongoing renal sodium losses. Preterm infants have higher sodium requirements (mEq/kg) than term infants and older children, primarily related to impaired renal tubular sodium retention [33, 36]. However, limited attention has been given to identifying absolute sodium requirements in preterm infants and how these requirements change with maturation. Following birth, significant changes in renal sodium handling occur, the extent of which is dependent upon gestational age as well as postnatal age. Siegal and Oh found the magnitude of urinary sodium excretion decreased from approximately 200μEq/kg/h in infants at 27 weeks gestation to less than 25μEq/kg/h at term [37]. Additionally, Gabhaju et al. reported fractional excretion of sodium (FENa) exceeded 6% in infants < 28 weeks gestation on day of life three, decreasing to approximately 4% by the end of the first week of life and 2% at a month of age [33]. Less premature infants (29–36 weeks gestation at birth) had lower FENa at three days of age and showed a similar maturational decrease in FENa over the first month of life. By 28 days of life, there was no statistically significant difference in FENa among the gestation age groups, though infants born at < 28 weeks gestation appeared to have a FENa twice that of older infants.
Current recommendations for preterm infants from the American Academy of Pediatrics (AAP) and European Society for Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) include enteral or parenteral intake of sodium of 3–5 mEq/kg/d during the stable, growth phase of postnatal care [38, 39]. The recommendations, based on a factorial approach calculated from fetal accretion of body components, likely fail to account for the degree of renal immaturity present in extremely preterm infants who are surviving now far earlier in gestation [40]. Infants born at 23–31 weeks gestation receiving the currently recommended sodium intakes do not consistently achieve a state of positive sodium balance until after 32 weeks postmenstrual age [41]. Considering these calculations did not account for non-renal sodium losses or sodium accretion associated with growth, (approximately 1.4 mEq/kg/d at these gestational ages), it is unlikely sufficient sodium was provided at any time point [40].
The issue of identifying infants in need of and determining the quantity of sodium supplementation is not insignificant given impact of total body sodium status on somatic growth. Animal studies from over 90 years ago demonstrated greater weight gain and nitrogen retention when provided rations with added NaCl [42]. In young rats, a sodium deficient diet impairs weight and length growth, diminishes nitrogen retention and decreases muscle protein synthesis and RNA concentrations [43]. Caloric intake was similar between groups, suggestion the utilization of food energy for growth is impaired by inadequate NaCl intake. In infants at high risk for sodium depletion and growth failure, such as those with cystic fibrosis and ileostomies, correction of the sodium deficit is associated with improved growth [44, 45].
Studies in preterm infants suggest sodium supplementation may optimize weight gain. Vanpee et al.randomized ten infants with a gestational age 29–34 weeks to receive sodium supplementation of 4 mEq/kg/d from 4–14 days of life [46]. At 2 weeks of age, supplemented infants weighed about 6% above birthweight while non-supplemented infants (n = 10) were a little less than 2% below birthweight (<0.01); fluid intake and urine output were similar between groups. Isemann et al.randomized infants < 32 weeks gestation to receive 4 mEq/kg/d of sodium or placebo from days of life 7–35, with an average daily sodium intake of 6.3 mEq/kg in the supplemented group and 2.9 mEq/kg in the placebo group [47]. Though the sample size was small, at six weeks of postnatal age, 79% of supplemented infants maintained their birthweight percentile compared with only 13% in the placebo group. No significant differences in length or head circumference were identified. Taken together, these studies suggest that avoiding sodium deficiency in preterm infants may result in improved weight gain.
Decisions regarding sodium administration are often based on serum sodium concentrations despite the known poor relationship between plasma sodium concentration and body sodium content [48]. Given the importance of sodium to growth alternative approaches are needed to guide the clinician in providing supplemental sodium to preterm infants. Though limitations exist, urine sodium concentration provides an assessment of total body sodium homeostasis; an abnormally low urine sodium concentration reflecting active renal sodium conservation in response to decreased total body sodium. Infants with short bowel syndrome at risk for sodium depletion due to intestinal losses demonstrate improved growth with sodium supplementation and sodium repletion based upon urine sodium concentration [44]. We recently developed and put in to clinical practice an algorithm based on urine sodium concentrations to identify sodium deficiency in preterm infants 25–29 weeks gestational age and guide sodium supplementation between two and eight weeks of postnatal age [49]. Use of the algorithm resulted in significantly increased sodium intake and weight gain between two and eight weeks of postnatal age compared to a recent historical cohort, despite similar caloric, protein and fluid intakes. Though not advocating this approach, as it has currently undergoing clinical trial (NCT03889197), the findings support the argument that many extremely premature infants have sodium requirements not currently being met to promote optimal growth.
In developing the urine algorithm we used published measures of preterm infant glomerular and tubular function to calculate expected daily urinary sodium losses and urine sodium concentrations at advancing gestational and postnatal ages [33, 49]. Based upon these data and accounting for sodium accretion associated with growth, recommended sodium intake according to gestational and postnatal ages are provided in Table 2. Clinical conditions of infants, including renal impairment, states of vasopressin excess and administration of diuretics will impact the desired quantity of supplemental sodium. Ongoing study is necessary to define the optimal sodium needs in preterm infant and in assessment of total body sodium status.
Estimated sodium requirements (mEq/kg/d) for preterm infants at various gestational and postnatal ages
Estimated sodium requirements (mEq/kg/d) for preterm infants at various gestational and postnatal ages
Potassium is the most abundant cation of intracellular fluids and plays an important role in numerous intracellular functions. Because a positive potassium balance is required for growth, homeostatic regulatory mechanisms are different in infants than adults [50]. Despite the importance of potassium, few studies have examined nutritional requirements in preterm infants.
Plasma potassium concentration is higher in preterm than in term neonates, and typically falls in in the first few days of life [51]. Non-oliguric hyperkalemia is frequently seen in very low birth weight infants in the first 48 hours after birth despite minimal potassium intake. Elevated plasma potassium levels likely results from a shift from the intracellular to the extracellular space, in part due to decreased Na-K-ATPase activity, and limited renal potassium excretion associated with a low glomerular filtration rate [52, 53]. Exposure to antenatal glucocorticoids and early provision of amino acids (>1.5–2 mg/kg/d) reduce the risk of hyperkalemia [54].
A positive potassium balance, or potassium retention, is necessary for postnatal growth. Based on data from chemical analyses of human fetuses 24–40 weeks gestation, Ziegler et al.reported fetal potassium accretion rates of 1.1–1.3 mEq/30 gram weight gain [40]. These values for potassium retention are similar to those reported in growing preterm infants. A longitudinal study of potassium balance in infants 23–31 weeks gestation demonstrated that fractional excretion of potassium decreases with advancing gestation and postnatal age, although urine potassium excretion (1.5–2.0 mEq/kg/d) is relatively stable over the range of intakes provided (2–3 mEq/kg/d after the first few days of life) [41]. In this study, a state of positive potassium balance was generally achieved in infants > 30 weeks gestation, although the impact of increased potassium intake on more premature infants was not evaluated.
Recommendations for daily potassium intake are derived not only from balance studies, but the knowledge of the amount of potassium provided in human milk. Though potassium content of human milk decreases slightly over time following delivery, reported values range from 1.1–1.5 mEq/100 ml [55]. Intakes of 150–200 ml/kg of term mother’s milk would provide on average 2–3 mEq/kg of potassium. Therefore, once the postnatal contraction of the extracellular space has occurred (diuresis) and plasma potassium levels are below 4–4.5 mEq/L, intakes of 2–3 mEq/Kg/d are usually sufficient to maintain a normal plasma potassium concentration and allow for growth. Hypokalemia in the presence of recommended potassium intake may result from drug administration, particularly diuretics and amphotericin, increased renal losses from mineralocorticoid excess or renal tubular acidosis, or a redistribution to the intracellular space from alkalosis or insulin therapy. Conversely, hyperkalemia may result from sampling technique, acute kidney injury, drug administration and increased potassium load through the diet, blood transfusion or hemolysis.
Recommended guidelines
Any strategy for providing fluid and electrolytes to preterm infants must be flexible, taking into consideration the multiple variables, discussed above, that affect water and electrolyte requirements. The nutritional needs of infants must also be addressed when prescribing fluids. Table 3 provides general guidelines for fluid and electrolyte management of preterm infants over the first week of life. Optimal management requires the frequent assessment of fluid and electrolyte status, which can be obtained through measurement of weight, urine output and serum electrolytes. In adjusting water and electrolyte intakes in response to these measurements, the provider should utilize his/her understanding of the postnatal physiological changes of the skin and kidney. For example, a rising serum sodium or chloride concentration in the first few days of life likely reflects a developing free water deficit related to an underestimation of insensible water loss. Increasing free water intake 20–40 ml/kg/d may the appropriate response to this condition, depending upon the rate of change.
Guidelines for initial fluid and electrolyte management of preterm infants
Guidelines for initial fluid and electrolyte management of preterm infants
GA, gestational age, DOL, day of life, PN, parenteral nutrition, Q, every.; *<28 weeks should have a barrier placed (plastic heat shield) to minimize heat and insensible water losses.; PN (D10 5% amino acids; 30 ml/kg/d for < 26 wk gestation, provides 1.5 g/kg/d protein and 2.1 mg/kg/min carbohydrate; 60 ml/kg/d for 27–29 wk gestation, provides 3 g/kg/d protein and 4.2 mg/kg/min carbohydrate). M, milk (mothers or donor) or formula is often provided as an initial fluid.; **Initial fluid composition only. Sodium supplementation should begin according to Table 2 in this review. Potassium supplementation should be provided once serum potassium is < 4 mEq/L.; T -The frequency of electrolyte monitoring can often be reduced after 24–48 hours.
No guideline is appropriate for all situations. The recommended fluid intakes consider the decreasing absolute fluid requirements as TEWL decrease while addressing the nutritional needs of the infant. Parenteral nutrition should be initiated soon after birth to limit the development of a protein deficit. An approach that uses separate parenteral nutrition fluid and dextrose in water allows for adjustments of water and carbohydrate intake without compromising protein intake. Monitoring of serum electrolytes (sodium and/or chloride) is vital to assess water balance. Frequent assessment is needed for the extremely premature infant as insensible water losses are often over or under estimated. The frequency of electrolyte determination can often be reduced after the first couple of days of life. While phlebotomy blood loss is a concern with frequent monitoring, the use of whole blood benchtop analyzers or point-of-care laboratory devices, which utilize no more than 25–100μL per sample minimizes blood loss. In addition, phlebotomy overdraw (excess volumes of blood taken per sample) can be avoided through understanding of device requirements and education of personnel [56].
The maintenance of fluid and electrolyte homeostasis is essential for optimal outcomes in premature infants. Recognition and understanding of physiological and environmental factors that influence water and electrolyte needs are vital for those providing care to this vulnerable population. Any rational approach to fluid and electrolyte therapy must allow for individualized care with consideration of gestational and postnatal age dependent changes in water and salt requirements. Continued research and reevaluation of currently recommended fluid and sodium intakes are necessary to improve the long-term outcomes of preterm infants.
Disclosures
The author has no financial or ethical conflict of interest to declare.
Funding sources
None.
Footnotes
Acknowledgments
None.