After completing this chapter, the reader should be able to

  • Define the various pediatric age group terminology

  • Discuss general pediatric considerations as they relate to blood sampling

  • Describe how pediatric reference ranges are determined

  • Discuss the age-related physiologic differences that account for variations by age in the normal reference ranges for serum sodium, potassium, bicarbonate, calcium, phosphorus, and magnesium

  • List common pediatric causes of abnormalities in the electrolytes and minerals listed above

  • Explain why age-related differences in serum creatinine and kidney function tests occur

  • Discuss the age-related differences that occur in serum albumin, liver enzyme tests, and bilirubin

  • Describe what is meant by the physiologic anemia of infancy and explain how it occurs

The interpretation of laboratory data in the pediatric patient population can be complex. Compared with adults, the pediatric population is much more dynamic. Alterations in body composition, organ function, and physiologic activity accompany the normal processes of maturation and growth that occur from birth through adolescence. These alterations can result in different normal reference ranges in pediatric patients for various laboratory tests. Pediatric patients have not only different normal laboratory values compared with adults but also normal laboratory values may differ in various pediatric age groups. It is important for the clinician to understand the reasons for these different, commonly accepted reference ranges and to use age-appropriate reference ranges when providing pharmaceutical care to pediatric patients.

The measurement of substances in neonates, infants, and young children is further complicated by a patient’s smaller physical size and difficulty in obtaining blood and urine samples. The smaller blood volume in these patients requires blood samples to be smaller; thus, special microanalytical techniques must be used. Additionally, in the neonate, substances that normally occur in higher amounts in the blood—such as bilirubin, lipids, and hemoglobin—may interfere with certain assays. This chapter briefly reviews pertinent general pediatric principles and focuses on the different age-related factors that must be considered when interpreting commonly used laboratory data in pediatric patients.


Knowledge of pediatric age group terminology is important to better understand age-related physiologic differences and other factors that may influence the interpretation of pediatric laboratory data. These terms are defined in Table 23-1 and are used throughout this chapter.1,2

TABLE 23-1.

Definition of Age Group Terminology

Gestational age (GA)

The time from conception until birth; more specifically, GA is defined as the number of weeks from the first day of the mother’s LMP until the birth of the baby; GA at birth is assessed by the date of the LMP and by physical and neuromuscular examination (eg, New Ballard Score)

Postnatal age (PNA)

Chronological age since birth

Postmenstrual age (PMA)a

Postmenstrual age is calculated as gestational age plus postnatal age (PMA = GA + PNA)


A full-term newborn 0–28 days PNA; some experts may also apply this terminology to a premature neonate who is >28 days PNA but whose PMA is ≤ 42–46 wk

Premature neonate

Neonate born at <37 wk GA

Full-term neonate

Neonate born at 37 wk 0 days to 41 wk 6 days (average ~40 wk) GA


1 mo (>28 days) to 12 mo of age


1–12 yr


13–18 yr


>18 yr

LMP = last menstrual period.

aThe term postconceptional age (PCA; age since conception) is no longer recommended for use in clinical pediatrics.2 However, this term may be found in pediatric literature. Traditionally, PCA was defined as GA + PNA. Because the exact time of conception is not generally known (except in cases of assisted reproductive technology) and GA is calculated as above (according to the mother’s LMP), PMA is considered a more accurate term to use. When PCA is used in the pediatric literature, it should be defined within the article where it is used.

Source: Adapted with permission from Taketomo CK, Hodding JH, Kraus DM. Pediatric and Neonatal Dosage Handbook. 26th ed. Hudson, OH: Lexi-Comp Inc; 2019.

The interpretation of any patient’s laboratory data must be viewed in light of the patient’s clinical status. This includes the patient’s symptoms, physical signs of disease, and physiologic parameters such as respiratory rate, heart rate, and blood pressure. For example, an elevated PaCO2 from an arterial blood gas may be clinically more significant in a patient who is extremely tachypneic (perhaps indicating impending respiratory failure) compared with a patient whose respiratory rate is mildly elevated. Thus, it is important to know the relative differences in physiologic norms that occur in the various pediatric age groups.

Normal respiratory rates are higher in neonates and young infants compared with children, adolescents, and adults. The average respiratory rate of a newborn is 60 breaths/min at 1 hour after birth but 30 to 40 breaths/min at >6 hours after birth. Mean respiratory rates of infants and young children <2 years of age (25 to 30 breaths/min) continue to be higher than in children 3 to 9 years of age (20 to 25 breaths/min) and adolescents (16 to 20 breaths/min).1

Normal heart rates follow a similar pattern, with higher heart rates in neonates and young infants, which then slowly decrease with increasing age through adolescence. For example, the mean heart rate of a newborn in the first week of life is 125 beats/min, with a normal high of 160 beats/min. The mean heart rate of a 1-month-old infant is 145 beats/min, while that of a 1-year-old child is 120 beats/min and that of a 12-year-old child is 85 beats/min.3

In pediatric patients, normal blood pressure values vary according to age, sex, and percentile height of the patient.4,5 Blood pressures are lower in neonates and increase throughout infancy and childhood. For example, typical blood pressures for a full-term newborn would be in the range of 65 to 95 mm Hg systolic and 30 to 60 mm Hg diastolic. The normal blood pressure (blood pressure <90th percentile) for a 1-year-old girl of average height (50th percentile height) would be <100/54 mm Hg, whereas that of a 15-year-old girl of average height would be <123/79 mm Hg. Blood pressures are slightly different for girls compared with boys and are higher in taller children. Appropriate references should be consulted to obtain normal blood pressure values when providing clinical care to pediatric patients.3-5

In addition to age-related physiologic differences in respiratory rates, heart rates, and blood pressures, age-related changes in body composition (eg, fluid compartments), cardiac output, organ perfusion, and organ function also exist. These age-related changes may result in different normal laboratory values for pediatric patients compared with adults. For example, age-related changes in fluid compartments affect normal laboratory values for serum electrolytes, as discussed in the Serum Electrolytes and Minerals section. Being aware of the normal laboratory values for age is important for proper monitoring of efficacy and toxicity of pediatric drug therapy.

Pediatric Blood Sampling

The smaller physical size of pediatric patients makes it more difficult to obtain blood samples. In general, venipuncture techniques used in adults can be used in older children and adolescents. However, vacuum containers used for blood sampling may collapse the small veins of younger children and are not recommended in these patients.6 Capillary puncture (also called microcapillary puncture or skin puncture) is used in patients with small or inaccessible veins. Thus, it is the blood sampling method of choice for premature neonates, neonates, and young infants. Because this method also helps preserve total blood volume, it may also be beneficial to use in infants and small children who require multiple blood tests.7

The physical sites that are used for capillary puncture include the heel, finger, and great toe.6,7 The preferred site in neonates and younger infants is the medial or lateral portion of the plantar surface of the heel. The central area of the foot is avoided because of the risk of damage to the calcaneus bone, tendons, nerves, and cartilage. Automatic lancet devices are recommended when performing heel stick capillary blood sampling as they have been associated with less pain, fewer complications, and higher precision (lower resampling rate) when compared with manual lancets. The automatic devices are available in different needle lengths and incision depths for use in premature neonates, term neonates, and younger infants and have automatic retractable needles for user safety.8 Fingersticks may be used in older infants (generally >6 months of age and weighing >10 kg) and children, whereas capillary puncture of the great toe (which is rarely used) should be reserved for nonambulatory patients >1 year of age.8,9 Once children begin to walk, heel stick and great toe capillary puncture are used less frequently because bruising from sampling may result in pain upon walking and callus formation on feet may preclude vascular access and the ability to successfully obtain a sample.

Because capillary and venous blood are similar in composition, the capillary puncture method may be used to obtain samples for most chemistry and hematology tests.7 However, differences may occur between venous and capillary blood for certain substances, such as glucose, calcium, potassium, and total protein. For example, glucose concentrations may be 10% higher when the sample is collected by capillary puncture compared with venipuncture.6 In addition, improper capillary puncture sample collection may result in hemolysis or introduction of interstitial fluid into the specimen. This may result in higher concentrations for potassium, magnesium, lactate dehydrogenase, and other substances. Therefore, using the proper procedure to collect blood by the capillary puncture method is essential. It is also important that the site of capillary puncture be warmed prior to sample collection, especially for blood gas determinations.6 Complications of capillary puncture include infection, hematoma, and bruising.

The size of the blood sample is an important issue to the pediatric clinician. Compared with adults, pediatric patients have a much smaller total blood volume (Table 23-2). For example, a full-term newborn of average weight (3.4 kg) has an approximate total blood volume of 78 to 86 mL/kg, or about 265 to 292 mL total.10 However, a 70-kg adult has an estimated total blood volume of 68 to 88 mL/kg or 4,760 to 6,160 mL total. If a standard 10-mL blood sample were to be drawn from a pediatric patient, it would represent a much higher percent of total blood volume compared with an adult. Therefore, the smaller total blood volume in pediatric patients requires blood sample sizes to be smaller. This issue is further complicated in newborns because their relatively high hematocrit (approximately 60% or higher) decreases the yield of serum or plasma from the amount of blood collected. Microanalytical techniques have reduced the required size of blood samples. However, critically ill pediatric patients may require multiple or frequent blood sample determinations. Thus, it is essential to plan pediatric laboratory tests, especially in the neonate and premature neonate, to avoid precipitating iatrogenic anemia from excessive blood drawing.

TABLE 23-2.

Total Blood Volume by Age Group





Premature neonate




Full-term neonate




1–12 mo

7.6 (6 mo)



1–3 yr

12.4 (2 yr)



4–6 yr

18.2 (5 yr)



7–18 yr

45.5 (13 yr)







aApproximate total blood volume information compiled from Nathan DG, Orkin SH, eds. Nathan and Oski’s Hematology of Infancy and Childhood. 5th ed. Philadelphia, PA: WB Saunders; 1998.

Source: Adapted with permission from Taketomo CK, Hodding JH, Kraus DM. Pediatric and Neonatal Dosage Handbook. 22nd ed. Hudson, OH: Lexi-Comp Inc; 2015.

Substances that normally occur in higher amounts in the blood of neonates, such as bilirubin, lipids, and hemoglobin, may interfere with certain assays. Hyperbilirubinemia may occur in premature and term neonates. High bilirubin concentrations may produce falsely low creatinine or cholesterol values when measured by certain analytical instruments.6 Neonates, especially those who are born prematurely, may have lipemia when receiving intravenous (IV) fat emulsions. Lipemia may interfere with spectrophotometric determinations of any substance or with flame photometer determinations of potassium and sodium. Newborns have higher hemoglobin values, and hemoglobin may interfere with certain assays. For example, hemolysis and the presence of hemoglobin may interfere with bilirubin measurements. Therefore, it is important to ensure that the assay methodology selected for measurement of substances in neonatal serum or plasma is not subject to interference from bilirubin, lipids, or hemoglobin.

Pediatric Reference Ranges

Various methods can be used to determine reference ranges, and each method has its own advantages and disadvantages. In adults, reference ranges are usually determined by obtaining samples directly from known healthy individuals (direct method). The frequency distribution of the obtained values are assessed and the extreme outliers (eg, 0 to 2.5th percentile and 97.5 to 100th percentile) are excluded. This leaves the values of the 2.5 to 97.5th percentiles to define the reference range.11 However, it also labels the 0 to 2.5th percentile and 97.5 to 100th percentile values from the healthy individuals as being outside of the reference range. If the frequency distribution of the obtained values fall in a bell-shaped or Gaussian distribution, then the mean (or average) value plus or minus two standard deviations (SDs) can then be used to define the reference range. The mean value plus or minus two SDs includes 95% of the sample. This method labels 5% of the healthy individuals as having values that fall outside of the reference range.

In the pediatric population, however, one cannot easily obtain blood samples directly from known healthy individuals. Large sample sizes of healthy pediatric individuals that include an appropriate age distribution from birth to 18 years of age would be required. Furthermore, it may be difficult to obtain blood samples from healthy pediatric patients when these individuals cannot legally give informed consent and there is no direct benefit to these individuals of obtaining the blood sample. Therefore, many pediatric reference ranges have traditionally been determined via an indirect method: by using results of tests from hospitalized sick pediatric patients and applying special statistical methods.11 The statistical methods are designed to remove outliers and distinguish the normal values from the values found in the sick patients. Obviously, problems in determining the true reference range may arise, especially when overlap between values from the diseased and nondiseased population occurs.

More recently, the Canadian Laboratory Initiative on Pediatric Reference Intervals (CALIPER) has attempted to address the limitations of using indirect methods to establish pediatric reference ranges.12 Over the course of more than a decade, the CALIPER project has developed age- and sex-specific normal pediatric reference ranges for >100 laboratory tests using blood samples obtained directly from >9,700 known healthy children and adolescents from various communities and ethnic backgrounds in Canada. The CALIPER extensive database also includes neonates and infants; however, blood samples for these individuals (<1 year of age) were obtained largely from outpatient clinics. Current applicability of reference ranges for this age group to healthy individuals may be limited, but further studies by CALIPER are underway to address this limitation.

As in adults, many factors can influence the pediatric reference range, including the specific assay methodology used, type of specimen analyzed, specific population studied, nutritional status of the individual, time of day the sample is obtained, timing of meals, medications taken, and specific patient demographics (age, sex, height, weight, body surface area [BSA], and ethnicity). These factors, if not properly identified, may also influence the determination of reference ranges. In addition, because many pediatric reference ranges are typically established in hospitalized patients, concomitant diseases may also influence the determination of the specific reference range being studied. These factors, plus the greater heterogeneity (variance) observed in the pediatric population, make determination of pediatric reference ranges more complex.

Pediatric studies that define reference ranges may not always give detailed information about factors that may have influenced the determination of the specific pediatric reference range. Furthermore, due to the variation in influencing factors, most published pediatric reference ranges are not in exact agreement with each other.1,3,11-17 Some studies report reference ranges by age for each year, others by various age groups, and others only by graphic display. This makes it difficult to ascertain standard values for pediatric reference ranges and to apply published pediatric reference ranges to one’s own patient population.

The reference ranges listed in this chapter reflect a compilation from various sources and are meant to be general guidelines. Clinicians should consult with their institution’s laboratory to determine specific age-appropriate pediatric reference ranges to be used in their patient population.

Pediatric Clinical Presentation

In general, the clinical symptoms of laboratory abnormalities in pediatric patients are similar to symptoms observed in adults. However, certain manifestations of symptoms may be different in pediatric patients. For example, central nervous system (CNS) irritability due to electrolyte imbalances (such as hypernatremia) may manifest as a high-pitched cry in infants. Hypocalcemia is more likely to manifest as seizures in neonates and young infants (compared with adults) due to the immaturity of the CNS. Neonates may also have nonspecific or vague symptoms for many disorders. For example, neonates with sepsis, meningitis, or hypocalcemia may have poor feedings, lethargy, and vomiting. In addition, young pediatric patients are unable to communicate symptoms they may be experiencing. Thus, although symptoms of laboratory abnormalities in pediatric patients are important, often a correct diagnosis relies on the physical exam and appropriate laboratory tests.


The homeostatic mechanisms that regulate fluid, electrolyte, and mineral balance in adults also apply to the pediatric patient. However, several important age-related differences exist. Compared with adults, neonates and young infants have alterations in body composition and fluid compartments; increased insensible water loss; immature (decreased) renal function; and variations in the neuroendocrine control of fluid, electrolyte, and mineral balance.18 In addition, fluid, electrolyte, and mineral intake are not controlled by the individual (ie, the neonate or young infant) but are controlled by the individual’s caregiver. These age-related physiologic differences can result in alterations in the pediatric reference range for several electrolytes and minerals and can influence the interpretation of pediatric laboratory data.

A large percent of the human body is comprised of water. Total body water (TBW) can be divided into two major compartments: intracellular water (ICW) and extracellular water (ECW). The ECW compartment consists of the interstitial water and the intravascular water (or plasma volume) (Figure 23-1). Both TBW and ECW, when expressed as a percentage of body weight, are increased in the fetus and the newborn (especially the premature neonate) and decrease during childhood with increasing age (Figure 23-2).19,20 The TBW of a fetus is 94% during the first month of gestation and decreases to 75% in a full-term newborn. The TBW of a preterm newborn may be 80%. The TBW decreases to approximately 60% by 6 to 12 months of age and to 55% in an adult. ECW is about 44% in a full-term newborn, 30% in a 3- to 6-month-old infant, 25% in a 1-year-old child, and 19% in an adult.19,20 The decrease in TBW that is seen after birth is largely due to a contraction (or mobilization) of the ECW compartment. This mobilization is, most likely, the result of an increase in renal function that is seen after birth. The ICW compartment is lower at birth, increases slowly after birth, and is greater than ECW by about 3 months of age. It is important to note that the intake of water and electrolytes can influence these postnatal changes in TBW and the distribution between ECW and ICW.18

FIGURE 23-1.
FIGURE 23-1.

Distribution of body water in a term newborn infant expressed as a percentage of body weight. (Source: Reproduced with permission from Bell EF, Oh W. Fluid and electrolyte management. In: MacDonald MG, Seshia MMK, Mullett MD, eds. Avery’s Neonatology: Pathophysiology and Management of the Newborn. 6th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2005:363.)

FIGURE 23-2.
FIGURE 23-2.

Changes in body water from birth to 15 years. (Source: Data from Friis-Hansen B. Changes in body water compartments during growth. Acta Paediatr Suppl. 1957;46(suppl 110):1–68.)

The electrolyte composition of ECW versus ICW is different (Figure 23-3). Sodium is the major cation found in intravascular water (plasma volume) of the ECW. Potassium, calcium, and magnesium make up a much smaller amount of the intravascular cations. Chloride is the primary intravascular anion and bicarbonate, protein, and other anions comprise the balance. The electrolyte composition of the interstitial component of ECW is similar to the intravascular composition, but protein content is lower. Potassium and magnesium are the major cations found in ICW. Phosphate (organic and inorganic) is the primary intracellular anion, and bicarbonate makes up a smaller amount.18

FIGURE 23-3.
FIGURE 23-3.

Ion distribution in the blood plasma, which represents extracellular fluid, and in the intracellular fluid compartment. (Source: Reproduced with permission from Bell EF, Oh W. Fluid and electrolyte management. In: MacDonald MG, Seshia MMK, Mullett MD, eds. Avery’s Neonatology: Pathophysiology and Management of the Newborn. 6th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2005:364.)

These compositional differences in ECW and ICW, along with the age-related differences in the amounts of these water compartments, can result in maturational differences in the amount of electrolytes per kilogram of body weight. For example, because premature neonates have a larger ECW compartment and ECW contains a higher amount of sodium and chloride, premature neonates contain a higher amount of sodium and chloride per kilogram of body weight compared with term neonates.18 These principles are important to keep in mind when managing neonatal fluid and electrolyte therapy. One must also remember that the management of fluid and electrolyte therapy in the mother during labor can result in alterations in the newborn’s fluid and electrolyte status. For example, if the mother is given too much fluid (ie, too much free water) during labor, the newborn may be born with hyponatremia.21

Insensible water loss is the water that is lost via evaporation from the skin and through the respiratory tract.18 Knowledge of the factors that influence insensible water loss in pediatric patients is important to estimate appropriate water intake and assess electrolyte imbalances that may occur. Compared with adults, neonates and young infants have an increase in the amount of insensible water loss, primarily due to their increased surface area to body weight ratio and higher respiratory rate. Smaller newborns and those born at a younger gestational age (GA) have an even higher insensible water loss, which is related to their immature (thinner) skin, greater skin blood flow, and larger TBW. Many other factors increase insensible water loss, such as environmental and body temperature, radiant warmers, phototherapy, motor activity, crying, and skin breakdown or injury. Congenital skin defects, such as gastroschisis, omphalocele, or neural tube defects will also increase insensible water loss. The use of high inspired or ambient humidity, plastic heat shields or blankets, occlusive dressings, and topical waterproof agents decreases insensible water loss.

The primary functions of the kidney (glomerular filtration, tubular secretion, and tubular reabsorption) are all decreased in the newborn, especially in the premature newborn, compared with adults. These functions increase with GA at birth and with postnatal age (PNA). The decreased glomerular and tubular functions in the neonatal kidney result in differences in how the neonate handles various electrolyte loads and differences in the normal reference ranges for several electrolytes, as described later.


Normal ranges are as follows:22

premature neonates (at 48 hours of life): 128 to 148 mEq/L (128 to 148 mmol/L)

neonates: 133 to 146 mEq/L (133 to 146 mmol/L)

infants: 139 to 146 mEq/L (139 to 146 mmol/L)

children: 138 to 145 mEq/L (138 to 145 mmol/L)

adults: 136 to 142 mEq/L (136 to 142 mmol/L)

Sodium is primarily excreted via the kidneys, but it also is excreted via stool and sweat.23 Usually, unless diarrhea occurs, sodium loss in the stool is minimal. In children with cystic fibrosis, aldosterone deficiency, or pseudohypoaldosteronism, the sodium concentration in sweat is increased and higher sweat losses may contribute to or cause sodium depletion.

In neonates and young infants, the renal handling of sodium is altered compared with adults.24,25 Differences in tubular reabsorption, aldosterone concentrations, and patterns of renal blood flow help to maintain a positive sodium balance, which is required for growth. In the neonate, sodium reabsorption is decreased in the proximal tubule but increased in the distal tubule. Aldosterone increases sodium reabsorption in the distal tubules, and plasma concentrations of renin, angiotensin II, and aldosterone are all increased in neonates. This increase in aldosterone may be a compensatory mechanism to help increase sodium reabsorption in the distal tubule. The pattern of renal blood flow is also different in the neonate. In adults, a larger amount of renal blood flow goes to the cortical area of the kidneys. However, in the neonate, most renal blood flow goes to the medullary area, which is more involved with sodium conservation than excretion. These factors help the neonatal kidney to retain sodium but also result in the neonate having a decreased ability to excrete a sodium load. Therefore, if an excessive amount of sodium is administered to a neonate, it will result in sodium retention with subsequent water retention and edema.

Although most infants are in a positive sodium balance, very low birth weight infants (birth weight <1.5 kg) are usually in a negative sodium balance24 due to their immature kidneys and the larger amounts of sodium that are lost in the urine. These infants are at a higher risk of sodium imbalance and may require higher amounts of sodium, especially during the first weeks of life. Compared with adults, pediatric patients may be more susceptible to imbalances of sodium and water. This may be due to their higher amount of TBW and the common pediatric occurrence of causative factors, such as diarrhea and dehydration.


In infants and children, hyponatremia is defined as a serum sodium <135 mEq/L, although slightly lower values would be considered acceptable for premature neonates and newborns.22,23 As in adults, hyponatremia occurs in pediatric patients when the ratio of water to sodium is increased. This may occur with low, normal, or high amounts of sodium in the body; likewise, the amount of water in the body may be low (hypovolemic), normal (euvolemic), or high (hypervolemic). The causes of hyponatremia in pediatric patients are the same as in adults. However, certain causes may be more commonly seen in children.

In hypovolemic hyponatremia, both sodium and water have been lost from the body, but a higher proportion of sodium has been lost. The most common cause of hypovolemic hyponatremia in children is diarrhea due to gastroenteritis.23 Emesis also can cause hyponatremia if hypotonic fluids are administered, but most children with emesis have either a normal serum sodium or hypernatremia. In addition to gastrointestinal (GI) losses, hypovolemic hyponatremia may also occur from losses of sodium through the skin (eg, excessive sweating or burns), third space losses, and renal losses.

Renal sodium loss can occur in the pediatric population from several causes, including thiazide or loop diuretics, osmotic diuresis, cerebral salt wasting, and hereditary or acquired kidney diseases. Cerebral salt wasting is thought to be caused by hypersecretion of atrial natriuretic peptide, which causes renal salt wasting. This condition is usually seen in patients with CNS disorders such as head trauma, brain tumors, hydrocephalus, cerebral vascular accidents, neurosurgery, and brain death.26 Hereditary kidney diseases that can cause hypovolemic hyponatremia include juvenile nephronophthisis, autosomal recessive polycystic kidney disease, proximal (type II) renal tubular acidosis, 21-hydroxylase deficiency, and pseudohypoaldosteronism type I. Patients with congenital adrenal hyperplasia due to 21-hydroxylase deficiency have an absence of aldosterone. Aldosterone is needed for sodium retention and potassium and acid excretion in the kidneys. The lack of aldosterone in these patients produces hyponatremia, hyperkalemia, and metabolic acidosis. Patients with pseudohypoaldosteronism have elevated aldosterone serum concentrations, but the kidneys do not respond properly to aldosterone. A lack of response to aldosterone by the renal tubules may also occur in children with urinary tract obstruction and acute urinary tract infection and result in hyponatremia.23

In euvolemic hyponatremia, patients have no real evidence of volume depletion or volume overload.23 Usually, these patients have a slight decrease in total body sodium with an excess of TBW. Although some patients may have an increase in body weight (indicating volume overload), they often appear clinically normal or have subtle signs of fluid overload. Causes of euvolemic hyponatremia include the syndrome of inappropriate antidiuretic hormone (SIADH), glucocorticoid deficiency, hypothyroidism, and water intoxication. Although SIADH is not common in children, it may occur in patients with CNS disorders or lung disease and tumors. Certain medications can cause an increase in antidiuretic hormone (ADH) secretion, as reviewed in Chapter 11.

Dilutional hyponatremia may commonly occur in hospitalized children who receive relatively large amounts of free water (eg, hypotonic IV solutions). This may even occur when medications are diluted in 5% dextrose in water, for example, and administered as 50- or 100-mL IV rider bags or piggyback riders. Neonates and young infants are more prone to this water overload (due to their lower glomerular filtration rate [GFR] and limited ability to excrete water), and should receive medications diluted in smaller volumes of IV fluid (which are usually administered via IV syringe pump). Other causes of hyponatremia due to water intoxication in pediatric patients include administration of diluted infant formula, tap water enemas, infant swimming lessons, forced water intake (child abuse), and psychogenic polydipsia23 (Minicase 1).

A Case of Hyponatremia and Seizures

Peter P., a 3-day-old male, is currently in the neonatal intensive care unit being treated with antibiotics for suspected sepsis. He was born at 39 weeks’ gestation to a mother with prolonged rupture of membranes (>72 hours). On the day of his birth, he was admitted to the neonatal intensive care unit with an elevated temperature, tachycardia (heart rate [HR] 166), and a low WBC count (3.2 × 103 cells/mm3). Blood and urine cultures were obtained, and antibiotics were started to treat his possible sepsis. Culture results are still pending. Medications include ampicillin 175 mg IV in 25 mL D5W administered as IV piggyback (IV rider bag) q 8 h (150 mg/kg/day) and gentamicin 14 mg IV in 50 mL D5W administered as IV piggyback q 24 h (4 mg/kg/day). This morning he began having rhythmic clonic twitching of his lower extremities, fluttering of his eyelids, and repetitive chewing movements, which are consistent with seizure activity.

His vital signs include blood pressure (BP) 77/50 mm Hg, HR 150 beats/min, respiratory rate (RR) 34 breaths/min, and temperature 99.1°F. Length is 51 cm (50th percentile for age), and weight is 3.5 kg (50th percentile for age). Laboratory data include sodium 119 mEq/L, potassium 4.5 mEq/L, chloride 100 mEq/L, total CO2 21 mEq/L, BUN 9 mg/dL, SCr 0.4 mg/dL, and glucose 88 mg/dL.

QUESTION: What is the most likely cause of this patient’s seizure activity and electrolyte imbalance? What other laboratory tests should be obtained to further assess his seizure disorder?

DISCUSSION: Electrolyte imbalance is a common cause of neonatal seizures. As in adults, hyponatremia may cause seizure activity in neonates and occurs when the ratio of water to sodium is increased. The total body content of sodium in patients with hyponatremia may be low, normal, or high, and the volume status may be hypovolemic, euvolemic, or hypervolemic. There are many causes of hyponatremia, but the most likely cause in this patient is the extra D5W that he received with his antibiotics. Dilutional hyponatremia may occur in neonates and young infants when medications are administered in excess fluids, such as IV piggybacks or IV riders of 5% dextrose in water. These patients are more prone to water overload due to their lower GFR and their limited ability to excrete water. Medications for these patients should be diluted in smaller amounts of IV fluid and typically administered via an IV syringe pump (or slow IV push when appropriate) so that excess fluid is not administered. The neonate in this case received 125 mL/day (~36 mL/kg/day) of exogenous IV fluid from the antibiotic administration alone (ampicillin [25 mL × 3 doses/day = 75 mL] + gentamicin [50 mL × 1 dose/day = 50 mL] = 125 mL/day). An excess fluid amount of 36 mL/kg/day of D5W would be considered very large for this neonatal patient, who would typically have a normal daily fluid requirement of 100 mL/kg/day. Standard parenteral preparation and concentrations should be followed in neonatal and pediatric patients to avoid volume overload. In a neonate with a central line, ampicillin is typically diluted to a final concentration of 100 mg/mL and administered via slow IV push. For this case, the total volume per dose of ampicillin should be 1.75 mL (175 mg) IV push q 8 h. IV gentamicin is typically diluted with normal saline to a concentration of 2 mg/mL, or a total volume of 7 mL (14 mg), and would be administered via IV syringe pump q 24 h for this patient. This would have provided a total IV fluid volume from the antibiotics of only 12.25 mL/day (3.5 mL/kg/day), a reduction of approximately 113 mL/day.

To better define this patient’s sodium and volume status, his total fluid intake and output, type of IV fluids administered, changes in body weight, and other laboratory data need to be assessed. In addition, other causes of hyponatremia, such as meningitis and SIADH, should be ruled out. A low WBC count, as observed in Peter P., often occurs in neonates with a serious bacterial infection (see section on White Blood Cell Count). Although the most likely cause of his seizure activity is his low serum sodium, his serum calcium, phosphorous, and magnesium levels also should be assessed because other electrolyte abnormalities can cause seizure activity.

In hypervolemic hyponatremia, both sodium and water are increased in the body, but there is a greater increase in water than sodium. Hypervolemic hyponatremia is typically observed in patients with congestive heart failure, cirrhosis, nephrotic syndrome, and chronic renal failure. It may also occur in patients with hypoalbuminemia or capillary leak syndrome due to sepsis.23 These conditions decrease the patient’s effective blood volume, either due to poor cardiac function or third spacing of fluid. The compensatory mechanisms in the body sense this decrease in blood volume; ADH and aldosterone are secreted and cause retention of water and sodium in the kidneys. A decrease in serum sodium occurs because the intake of water in these patients is greater than their sodium intake and ADH decreases water excretion.

Hyponatremia may also be caused by hyperosmolality (due to hyperglycemia or iatrogenic substances such as mannitol, sucrose, or glycine). For example, in pediatric patients with hyperglycemia during diabetic ketoacidosis, the high serum glucose concentration results in a high serum osmolality, which causes a shift of water from the intracellular space into the extracellular (intravascular) space. This shift of water has a dilutional effect on the serum sodium concentration, resulting in hyponatremia.23 The decrease in serum sodium is proportional to the increased serum glucose concentration; for every 100 mg/dL increase in serum glucose above normal (ie, 100 mg/dL), the measured serum sodium declines by 1.6 mEq/L. The following equation can be used in pediatric patients with hyponatremia due to hyperglycemia to correct the measured (low) serum sodium concentration.
Corrected serum sodium=Measured serum sodium+1.6×serum glucose100mg/dL/100

The calculated corrected serum sodium concentration better reflects the patient’s true ratio of total body sodium to TBW; however, it should be remembered that the calculated value is only an estimate.

Pseudohyponatremia (falsely measured low serum sodium) may occur when serum contains high concentrations of lipids (hyperlipidemia) or proteins (hyperproteinemia), for example, in patients with hypertriglyceridemia or multiple myeloma.23 This laboratory artifact may also occur in neonatal and pediatric patients who are receiving IV lipids for nutrition or IV immunoglobulin therapy.


In general for pediatric patients, hypernatremia is defined as a serum sodium concentration >145 mEq/L. As in adults, hypernatremia occurs in pediatric patients when the ratio of sodium to water is increased. This may occur with low, normal, or high amounts of sodium in the body. Hypernatremia may occur with excessive sodium intake, excess water loss, or a combination of water and sodium loss when the water loss exceeds the sodium loss.23

Excessive sodium intake or sodium intoxication may occur due to improperly mixed infant formulas, excess sodium bicarbonate administration, IV hypertonic saline solutions, intentional salt poisoning (eg, child abuse), and ingestion of sodium chloride or seawater.23 Neonates, especially premature newborns and young infants, can develop hypernatremia from excessive sodium due to the decreased ability of immature kidneys to excrete a sodium load. This becomes a problem, especially in the premature neonate when IV sodium bicarbonate is used to correct a metabolic acidosis. Excess water loss resulting in hypernatremia may occur in pediatric patients due to diabetes insipidus, increased insensible water losses, or inadequate intake. Diabetes insipidus can be of central or nephrogenic origin and either type can be acquired or congenital. Also, certain drugs may cause diabetes insipidus (see Chapter 11).

Neonates may be predisposed to hypernatremia from increased insensible water losses, especially during the first few days of life. A normal physiologic contraction of the ECW occurs after birth, resulting in a net loss of water and sodium. In term infants, this may result in a weight loss of 5% to 10% during the first week of life. In premature newborns, the weight loss may be 10% to 20%. This water loss, plus the relatively large and variable insensible water loss in neonates, can complicate the assessment of fluid and sodium balance. More premature newborns may be at higher risk for hypernatremia, as they have a more pronounced contraction of ECW and higher insensible water loss.27 The use of radiant warmers and phototherapy (used to treat hyperbilirubinemia) further increases insensible water loss.

Inadequate water intake also can cause hypernatremia in pediatric patients. This may be due to the caregiver not administering enough fluids (eg, child neglect or abuse or ineffective breastfeeding). Ineffective breastfeeding may result in severe hypernatremic dehydration. Rarely, inadequate intake may be due to adipsia (absence of thirst).23

Hypernatremia due to water losses greater than sodium losses occurs in patients with water and sodium losses through the GI tract (eg, diarrhea, emesis, nasogastric suctioning, and osmotic cathartics), skin (eg, burns and excessive sweating), and kidneys (eg, diabetes mellitus, chronic kidney disease, osmotic diuretics, and acute tubular necrosis [polyuric phase]). Hypernatremia is most likely to occur in infants or children with diarrhea who also have inadequate fluid intake due to anorexia, emesis, or lack of access to water.

It should be noted that due to the immaturity of the blood vessels in their CNS, premature neonates are especially vulnerable to the adverse effects of hypernatremia (eg, intracranial hemorrhage). These patients are also at greater risk of adverse CNS effects (eg, cerebral edema) if an elevated serum sodium is corrected too rapidly. Thus, maintaining a proper sodium balance in these patients is extremely important.


Normal ranges are as follows14,22:

premature neonates (at 48 hours of life): 3 to 6 mEq/L (3 to 6 mmol/L)

neonates: 3.7 to 5.9 mEq/L (3.7 to 5.9 mmol/L)

infants: 4.1 to 5.3 mEq/L (4.1 to 5.3 mmol/L)

children: 3.4 to 4.7 mEq/L (3.4 to 4.7 mmol/L)

adults: 3.8 to 5 mEq/L (3.8 to 5 mmol/L)

Potassium is the major intracellular cation, and <1% of total body potassium is found in the plasma.23 However, small changes in serum potassium can have large effects on cardiac, neuromuscular, and neural function. Thus, appropriate homeostasis of extracellular potassium is extremely important. Insulin, aldosterone, acid–base balance, catecholamines, and renal function all play important roles in the regulation of serum potassium. Serum potassium can be lowered quickly when potassium shifts intracellularly or more slowly via elimination by the kidneys.

The kidney is the primary organ that regulates potassium balance and elimination. Potassium undergoes glomerular filtration and almost all filtered potassium is then reabsorbed in the proximal tubule. Urinary excretion of potassium, therefore, is dependent on distal potassium secretion by the collecting tubules. Neonates and young infants, however, have a decreased ability to secrete potassium via the collecting tubules. Thus, immature kidneys tend to retain potassium. This results in a positive potassium balance, which is required for growth (potassium is incorporated intracellularly into new tissues).24,25 Potassium retention by the immature kidneys also results in higher serum potassium concentrations compared with the adult.25


Hypokalemia is defined as a serum potassium concentration <3.5 mEq/L. As in adults, a low serum potassium may occur in pediatric patients due to an intracellular shift of potassium, decreased intake, or increased output (from renal or extrarenal losses). An intracellular shift of potassium may be seen with alkalosis, β-adrenergic stimulation, or insulin treatment. Endogenous β-adrenergic agonists (such as epinephrine released during stress) and exogenously administered β-agonists (such as albuterol) stimulate the cellular uptake of potassium. Other causes of an intracellular shift of potassium seen in pediatric patients include overdoses of theophylline, barium intoxication, and glue sniffing (toluene intoxication). A falsely low potassium concentration can be reported in a patient with an elevated white blood cell (WBC) count (eg, a patient with leukemia) if the plasma sample is inappropriately stored at room temperature. This allows the WBCs to uptake potassium from the plasma, resulting in a falsely low measurement.23

Most cases of hypokalemia in children are related to extrarenal losses of potassium due to gastroenteritis and diarrhea.23 Hypokalemia caused by diarrhea is usually associated with a metabolic acidosis, because bicarbonate is also lost in the stool. Adolescent patients with eating disorders may be hypokalemic due to inadequate intake of potassium, such as patients with anorexia nervosa. Adolescents with bulimia or laxative abuse may also have significant extrarenal losses of potassium.

Many causes of hypokalemia resulting from renal potassium loss exist. Medications commonly used in the pediatric population that are associated with hypokalemia due to renal potassium loss include loop and thiazide diuretics, corticosteroids, amphotericin B, and cisplatin (see Chapter 11). Cushing syndrome, hyperaldosteronism, and ingestion of natural licorice (containing glycyrrhizic acid) may also cause hypokalemia via this mechanism.

In the pediatric population, other causes of increased renal potassium loss, such as hereditary diseases, must be considered. Remember that many hereditary diseases are first diagnosed during infancy and childhood. Renal tubular acidosis (both distal and proximal types) may present with hypokalemia and metabolic acidosis. Patients with cystic fibrosis have greater losses of chloride in sweat, which may lead to metabolic alkalosis, low urine chloride, and hypokalemia. Certain forms of congenital adrenal hyperplasia may also lead to increased renal potassium excretion and hypokalemia. Other inherited renal diseases that are due to defects in renal tubular transporters, such as Bartter syndrome, may result in metabolic alkalosis, hypokalemia, and high urine chloride. Thus, unlike the adult population, hereditary diseases need to be considered when certain electrolyte abnormalities are not explained by common causes.


In infants, children, and adults, hyperkalemia is defined as a serum potassium >5 mEq/L. Because a normal serum potassium is slightly higher in neonates and preterm infants, hyperkalemia is defined as a serum potassium >6 mEq/L in these patients. Hyperkalemia is one of the most alarming electrolyte imbalances because it has the potential to cause lethal cardiac arrhythmias.

As in adults, hyperkalemia in pediatric patients may be due to increased intake, an extracellular shift of potassium, or decreased renal excretion. Factitious hyperkalemia is common in pediatric patients because of the difficulty in obtaining blood samples. Hemolysis often occurs during blood sampling and potassium is released from red blood cells (RBCs) in sufficient amounts to cause falsely elevated test results. This may especially happen with improperly performed heel sticks (see section on Pediatric Blood Sampling). Potassium may also be released locally from muscles after prolonged tourniquet application or from fist clenching, which may also result in false elevations of measured potassium. A falsely elevated serum potassium also can be observed in patients with leukemia or extremely elevated WBC counts (usually >200,000/mm3) caused by the release of potassium from WBCs. Prompt analysis with measurement of a plasma sample usually avoids this problem.23

Hyperkalemia may occur as a result of extracellular shifts of potassium. During metabolic acidosis, hydrogen ions move into the cells (down a concentration gradient); in exchange, potassium ions move out of the cells into the extracellular (intravascular) space. This shift leads to a significant increase in serum potassium.

In older patients with fully developed (normal) renal function, hyperkalemia rarely results from increased intake alone. However, it may occur in patients receiving large amounts of oral or IV potassium or rapid or frequent blood transfusions (due to the potassium content of blood).23 In patients with immature renal function or renal failure, increased intake of potassium can also lead to hyperkalemia due to decreased potassium excretion.

Decreased renal excretion of potassium is the most common cause of hyperkalemia. Decreased potassium excretion occurs in patients with immature renal function, renal failure, primary adrenal disease, hyporeninemic hypoaldosteronism, renal tubular disease, and with certain medications.23 Hyperkalemia is the most common life-threatening electrolyte imbalance seen in neonates. Because of the decreased ability of immature kidneys to excrete potassium, neonates, particularly premature neonates, may be predisposed to hyperkalemia. These patients also cannot tolerate receiving extra potassium. Hyperkalemia can be seen in premature infants during the first 3 days of life, even when exogenous potassium is not given and when renal dysfunction is absent.27 A rapid elevation in serum potassium is seen within the first day of life in more immature newborns. This hyperkalemia, which can be life-threatening, may be due to a shift of potassium from the intracellular space to the extracellular (intravascular) space, immaturity of the distal renal tubules, and a relative hypoaldosteronism.28

Acute or chronic renal failure in pediatric patients decreases potassium excretion and may result in hyperkalemia. Several inherited disorders may also cause decreased potassium excretion and hyperkalemia in pediatric patients, including certain types of congenital adrenal hyperplasia (eg, 21-hydroxylase deficiency), aldosterone synthase deficiency, sickle cell disease, and pseudohypoaldosteronism (types I and II).23 Medications used in pediatric patients that may also cause hyperkalemia include angiotensin-converting enzyme inhibitors, β2-adrenergic antagonists, potassium-sparing diuretics, nonsteroidal anti-inflammatory agents, heparin, trimethoprim, and cyclosporine.

Serum Bicarbonate (Total Carbon Dioxide)

Normal ranges are as follows22,25:

preterm infants: 16 to 20 mEq/L (16 to 20 mmol/L)

full-term infants: 19 to 21 mEq/L (19 to 21 mmol/L)

Infants–children 2 years: 18 to 28 mEq/L (18 to 28 mmol/L)

children >2 years and adults: 21 to 28 mEq/L (21 to 28 mmol/L)

The total carbon dioxide concentration actually represents serum bicarbonate, the basic form of the carbonic acid–bicarbonate buffer system (ie, a low serum bicarbonate may indicate acidosis). In addition to the buffer systems, the kidneys also play an important role in acid–base balance. The proximal tubule reabsorbs 85% to 90% of filtered bicarbonate. The distal tubule is responsible for the net secretion of hydrogen ions and urinary acidification.29 Compared with adults, neonates have a decreased capacity to reabsorb bicarbonate in the proximal tubule and, therefore, a decreased renal threshold for bicarbonate (the renal threshold is the serum concentration at which bicarbonate appears in the urine). The mean renal threshold for bicarbonate in adults is 24 to 26 mEq/L but only 18 mEq/L in the premature infant and 21 mEq/L in the term neonate. The renal threshold for bicarbonate increases during the first year of life and reaches adult values by about 1 year of age. Neonates also have decreased function of the distal tubules to secrete hydrogen ions and to acidify urine. The ability to acidify urine increases to adult values by about 1 to 2 months of age.25,29 The neonate’s decreased renal capacity to reabsorb bicarbonate and excrete hydrogen ions results in lower normal values for serum bicarbonate and blood pH. In addition, the neonate is less able to handle an acid load or to compensate for acid–base abnormalities.

It should be noted that for multiple reasons, the full-term newborn is in a state of metabolic acidosis immediately after birth (arterial pH 7.11 to 7.36).3 The blood pH increases to more normal values within 24 hours, mostly due to increased excretion of carbon dioxide via the lungs.


Total serum calcium—normal ranges are as follows15:

neonates 3 to 24 hours: 9 to 10.6 mg/dL (2.25 to 2.65 mmol/L)

neonates 24 to 48 hours: 7 to 12 mg/dL (1.75 to 3 mmol/L)

neonates 4 to 7 days: 9 to 10.9 mg/dL (2.25 to 2.73 mmol/L)

children: 8.8 to 10.8 mg/dL (2.2 to 2.7 mmol/L)

adolescents: 8.4 to 10.2 mg/dL (2.1 to 2.55 mmol/L)

adults: 8.7 to 10.2 mg/dL (2.2 to 2.6 mmol/L)

Ionized calcium—normal ranges are as follows:

neonates 3 to 24 hours: 4.3 to 5.1 mg/dL (1.08 to 1.28 mmol/L)

neonates 24 to 48 hours: 4 to 4.7 mg/dL (1 to 1.18 mmol/L)

infants, children, and adolescents: 4.5 to 4.92 mg/dL (1.13 to 1.23 mmol/L)

adults: 4.5 to 5.3 mg/dL (1.13 to 1.33 mmol/L)

Calcium plays an integral role in many physiologic functions, including muscle contraction, neuromuscular transmission, blood coagulation, bone metabolism, and regulation of endocrine functions. Most calcium in the body (99%) is found in the bone, primarily as hydroxyapatite. Because of the growth that occurs during infancy and childhood, bone mass increases faster than body weight.30 This increase in bone mass requires a significant increase in total body calcium. The increased calcium requirement is reflected in the higher recommended daily allowances (per kilogram of body weight) for pediatric patients compared with adults.

Calcium regulation in the body has two main goals.30 First, serum calcium must be tightly regulated to permit the normal physiologic functions in which calcium plays a role. Second, calcium intake must be adequate to permit appropriate bone mineralization and skeletal growth. It is important to remember that bone mineralization may be sacrificed (ie, calcium may be released from the bone) to allow maintenance of a normal serum calcium concentration.

As in adults, serum calcium in pediatric patients is regulated by a complex hormonal system that involves vitamin D, serum phosphate, parathyroid hormone (PTH), and calcitonin. Briefly, calcium is absorbed in the GI tract, primarily via the duodenum and jejunum.30 Although some passive calcium absorption occurs when dietary intake is high, most GI absorption of calcium occurs via active transport that is stimulated by 1,25-dihydroxyvitamin D. This occurs especially when dietary intake is low. Calcium excretion is controlled by the kidneys and influenced by multiple hormonal mediators (eg, PTH, 1,25-dihydroxyvitamin D, and calcitonin). In mature kidneys, approximately 99% of filtered calcium is reabsorbed by the tubules, with most (>50%) absorbed by the proximal tubules. Calcium also is absorbed in the loop of Henle, distal tubule, and collecting ducts.

During the first week of life, urinary calcium excretion is inversely related to GA (ie, more premature infants have a greater urinary calcium excretion).25 Compared with adults, urinary calcium excretion is higher in neonates and preterm infants. The urinary calcium-to-creatinine ratio is about 0.2 in adults but may be >2 in premature neonates and up to 1.2 in full-term neonates during the first week of life. This high rate of calcium excretion may be related to the immaturity of the renal tubules and may contribute (along with other factors) to neonatal hypocalcemia. In addition, certain medications that are commonly administered to neonates and premature infants, such as furosemide, dexamethasone, and methylxanthines, further increase urinary calcium excretion. These medications may also increase the risk for hypocalcemia as well as nephrocalcinosis and nephrolithiasis.25

Measurement of Calcium

Total serum calcium measures all three forms of extracellular calcium: complex bound, protein bound, and ionized. However, ionized calcium is the physiologically active form. Usually a parallel relationship exists between the ionized and total serum calcium concentrations. However, in patients with alterations in acid–base balance or serum proteins, the ionized serum calcium and total serum calcium are affected, respectively, and measurements of total serum calcium may no longer reflect the ionized serum concentration. Neonates have lower serum concentrations of protein (including albumin) and may be acidotic. This results in a lower total serum calcium concentration for a given ionized plasma concentration.27 Although equations exist to adjust total serum calcium measurements for low concentrations of serum albumin, these equations have limitations and may not be precise. Therefore, ionized calcium should be measured in neonates (if microtechniques are available) and other pediatric patients with hypoalbuminemia or acid–base disorders.


As in adults, hypocalcemia may occur in pediatric patients for various reasons, including inadequate calcium intake, hypoparathyroidism, vitamin D deficiency, renal failure, redistribution of plasma calcium (eg, hyperphosphatemia and citrated blood transfusions), and hypomagnesemia. Hypocalcemia may also occur due to lack of organ response to PTH (eg, pseudohypoparathyroidism) and in the neonate because of other specific causes. The exact mechanism of how hypomagnesemia causes hypocalcemia is not clearly delineated. Magnesium can affect calcium balance, and significant hypomagnesemia can result in hypocalcemia due to intracellular cationic shifts. It is also thought that hypomagnesemia impairs the release of PTH and induces resistance to PTH effects. Because hypomagnesemia can result in hypocalcemia, a serum magnesium concentration is generally obtained in patients with hypocalcemia.

In the pediatric population, hypocalcemia most commonly occurs in neonates. Early neonatal hypocalcemia occurs during the first 72 hours of life and may be due to several factors. During fetal development, a transplacental active transport process maintains a higher calcium concentration in the fetus compared with the mother. After birth, this transplacental process suddenly stops. Serum calcium concentrations then decrease, even in healthy full-term newborns, reaching a nadir at 24 hours.30 The high serum calcium concentrations in utero may also suppress the fetus’ parathyroid gland. Thus, early neonatal hypocalcemia may also be caused by a relative hypoparathyroidism in the newborn. In addition, newborns may have a decreased response to PTH. Early neonatal hypocalcemia is more likely to occur in premature and low birth weight newborns. It also occurs more commonly in infants of diabetic mothers, infants with intrauterine growth retardation, and newborns who have undergone prolonged, difficult deliveries. Inadequate calcium intake in critically ill newborns also contributes to hypocalcemia.

Late neonatal hypocalcemia, which usually presents during the first 5 to 10 days of life, is caused by high phosphate intake. It is much less common than early neonatal hypocalcemia, especially because the phosphorus content of infant formulas was decreased. It may, however, still occur if neonates are inappropriately given whole cow’s milk. Cow’s milk has a high phosphate load, which can cause hyperphosphatemia and secondary hypocalcemia in the neonate. To receive appropriate amounts of calcium and phosphorous to meet specific nutritional needs, neonates who are not breastfed should be given the proper infant formula according to their level of maturity (eg, special premature infant formula or infant formula for term infants). In general, cow’s milk should not be introduced until 9 to 12 months of age.

Hypocalcemia may also occur in neonates born to mothers with hypercalcemia. Maternal hypercalcemia is usually due to hyperparathyroidism. In utero suppression of the fetal parathyroid gland can lead to hypoparathyroidism and hypocalcemia in the neonate. Hypocalcemia due to inadequate dietary calcium intake rarely occurs in the United States but can occur if infant formula or breast milk is replaced with liquids that contain lower amounts of calcium. Hypocalcemia may be iatrogenically induced if inadequate amounts of calcium are administered in hyperalimentation solutions. Adequate amounts of calcium and phosphorus may be difficult to deliver to preterm neonates because of their high daily requirements and limitations of calcium and phosphorus solubility in hyperalimentation solutions. Certain pediatric malabsorption disorders, such as celiac disease, may also cause inadequate absorption of calcium and vitamin D.

Hypoparathyroidism can be caused by many genetically inherited disorders, such as DiGeorge syndrome, X-linked hypoparathyroidism, and PTH gene mutations.30 These and other syndromes must be considered when pediatric patients present with hypoparathyroidism.

In pediatric patients with vitamin D deficiency, hypocalcemia occurs primarily due to decreased intestinal absorption of calcium. The lower amounts of calcium in the blood stimulate the release of PTH from the parathyroid gland. PTH then prevents significant hypocalcemia via several different mechanisms. It causes bone to release calcium, increases urinary calcium reabsorption, and increases the activity of 1α-hydroxylase in the kidneys (the enzyme that converts 25-hydroxyvitamin D into 1,25-dihydroxyvitamin D, the active form of vitamin D). Hypocalcemia only develops after these compensatory mechanisms fail. In fact, most children with vitamin D deficiency present with rickets before they develop hypocalcemia.30 In addition to elevated PTH concentrations, children with vitamin D deficiency have an elevated serum alkaline phosphatase concentration (due to increased osteoclast activity) and a low serum phosphorus (secondary to decreased intestinal absorption and decreased reabsorption in the kidneys), all as a result of the effects of PTH.

Vitamin D deficiency may be caused by several factors, including inadequate intake, lack of exposure to sunlight, malabsorption, or increased metabolism of vitamin D (eg, from medications such as phenobarbital and phenytoin). Generally, patients may have more than one of these factors. For example, institutionalized children who are not exposed to sunlight and receive chronic anticonvulsant therapy may be at a greater risk for developing vitamin D deficiency and rickets. Vitamin D deficiency may also occur with liver disease (failure to form 25-hydroxyvitamin D in the liver) and with renal failure (failure to form the active moiety, 1,25-dihydroxyvitamin D, due to a loss of activity of 1α-hydroxylase in the kidneys).

Genetic disorders, such as vitamin D-dependent rickets, may also cause hypocalcemia. The absence of the enzyme, 1-α-hydroxylase, in the kidneys occurs in children with vitamin D-dependent rickets type 1. Therefore, these children cannot convert 25-hydroxyvitamin D to its active form. Children with vitamin D-dependent rickets type 2 have a defective vitamin D receptor, which prevents the normal response to 1,25-dihydroxyvitamin D30 (Minicase 2).

Hypocalcemia also occurs when patients receive citrated blood transfusions or exchange transfusions (citrate is used to anticoagulate blood). Citrate forms a complex with calcium and decreases the ionized calcium concentration. Pediatric patients at highest risk include those receiving multiple blood transfusions or exchange transfusions, such as neonates treated for hyperbilirubinemia and older children treated for sickle cell crisis. These patients may develop symptoms of hypocalcemia, such as tetany or seizures, due to the lower ionized calcium levels. It should be noted that the total serum calcium concentration in these patients can be normal or even elevated because the calcium-citrate complex is included in the measurement.30


Hypercalcemia is an uncommon pediatric electrolyte disorder. As in adults, it may be caused by excess PTH, excess vitamin D, excess calcium intake, excess renal reabsorption of calcium, increased calcium released from the bone, and miscellaneous factors, such as hypophosphatemia or adrenal insufficiency.30 Causes of hypercalcemia that are of particular interest in pediatric patients include neonatal hyperparathyroidism, hypervitaminosis D, excessive calcium intake, malignancy associated hypercalcemia, and immobilization. Also, several genetic syndromes and disorders may cause hypercalcemia.

Neonatal hyperparathyroidism, an autosomal recessive disorder, can be severe and life-threatening.30 Typically, these patients have defective calcium-sensing receptors in the parathyroid gland. Normally, high serum calcium concentrations would be sensed by the parathyroid gland and PTH levels would then decrease. In these patients, however, the parathyroid gland cannot sense the high serum calcium concentrations, and PTH continues to be released, which further increases serum calcium concentrations. Transient secondary neonatal hyperparathyroidism occurs in neonates born to mothers with hypocalcemia. Maternal hypocalcemia leads to hypocalcemia in the fetus with secondary hyperparathyroidism. These neonates may be born with skeletal demineralization and bone fractures. Hypercalcemia in these patients usually takes days to weeks to resolve.

Excessive intake of vitamin D or calcium may also cause hypercalcemia. Typically, it may occur in children who are being treated with vitamin D and calcium with excessive doses. Excess calcium in hyperalimentation solutions commonly results in hypercalcemia.

Compared with adults, hypercalcemia from immobilization occurs more frequently in children, especially adolescents,30 because of a higher rate of bone remodeling in these patients. Immobilization of children and adolescents may be required due to specific injuries, such as leg fractures, spinal cord paralysis, burns, or other severe medical conditions. In children with leg fractures that require traction, hypercalcemia usually occurs within 1 to 3 weeks. Immobilization may also result in isolated hypercalciuria, which may result in nephrocalcinosis, kidney stones, or renal insufficiency.


Normal ranges are as follows23:

neonates 0 to 5 days: 4.8 to 8.2 mg/dL (1.55 to 2.65 mmol/L)

children 1 to 3 years: 3.8 to 6.5 mg/dL (1.23 to 2.1 mmol/L)

children 4 to 11 years: 3.7 to 5.6 mg/dL (1.2 to 1.8 mmol/L)

adolescents 12 to 15 years: 2.9 to 5.4 mg/dL (0.94 to 1.74 mmol/L)

adolescents 16 to 19 years: 2.7 to 4.7 mg/dL (0.87 to 1.52 mmol/L)

adults: 2.3 to 4.7 mg/dL (0.74 to 1.52 mmol/L)

Phosphorus is the primary intracellular anion and plays an integral role in cellular energy and intracellular metabolism. It is also a component of phospholipid membranes and other cell structures. Most phosphorus in the body (85%) is found in bone, whereas <1% of phosphorus is found in the plasma. Like calcium, phosphorus is essential for bone mineralization and skeletal growth. During infancy and childhood, a positive phosphorus balance is required for proper growth to allow adequate amounts of phosphorus to be incorporated into bone and new cells. The higher phosphorus requirement that is needed to facilitate growth may help explain the higher serum concentrations seen in the pediatric population compared with adults.

The kidney is the primary organ that regulates phosphorus balance. Approximately 90% of plasma phosphate is filtered by the glomerulus, with most being actively reabsorbed at the proximal tubule. Some reabsorption also occurs more distally, but phosphate is not significantly secreted along the nephron.23 Unlike other active transport systems, phosphate reabsorption, both proximal and distal, is greater in the neonatal kidney compared with adults.25,29 Thus, the neonatal kidney tends to retain phosphate, perhaps as a physiologic adaptation to the high demands for phosphate that are required for growth. Neonatal renal phosphate reabsorption may be regulated by growth hormone.29


As in adults, hypophosphatemia may occur in pediatric patients for several reasons, including increased renal excretion, decreased phosphate or vitamin D intake, or intracellular shifting. Causes of excessive renal phosphorus excretion in pediatric patients include hyperparathyroidism, metabolic acidosis, diuretics, glucocorticoids, glycosuria, IV fluids and volume expansion, kidney transplantation, and inherited disorders such as hypophosphatemic rickets.

Rickets in a Child

Raymond D., a 10-year-old male, is admitted to the emergency department from a local pediatric long-term care facility with pain, tenderness, and decreased movement to his right leg. He sustained a fall at the long-term care facility when he was being moved from his bed to his wheelchair. Born at term, he suffered a traumatic birth with severe perinatal asphyxia. He subsequently developed seizures that were controlled by the combined anticonvulsant therapy of phenobarbital and phenytoin. As a result of his asphyxia at birth, he developed spastic cerebral palsy and severe neurodevelopmental delay. He was transferred to the long-term care facility at 4 months of age and has remained on phenobarbital and phenytoin since that time. Two years ago, he was diagnosed with gastroesophageal reflux disease, which has been controlled with antacids. Medications include phenobarbital elixir 60 mg (15 mL) PO BID; phenytoin suspension 75 mg (3 mL) PO BID; and aluminum hydroxide suspension 10 mL PO QID.

His vital signs include BP 102/70 mm Hg; HR 92 beats/min; RR 24 breaths/min; and temperature 98.6°F. His height is 125 cm (<3rd percentile for age), and weight is 24 kg (<5th percentile for age). Physical exam of his chest is significant for a pigeon breast deformity and slightly palpable enlargement of costochondral junctions. He has redness in his right leg, 10 cm below the knee, and pain on movement. The preliminary radiographic findings reveal a fracture of his right tibia with osteomalacia and bone changes consistent with rickets.

Significant laboratory data are as follows:

  • calcium: 8.2 mg/dL (normal for children: 8.8 to 10.8 mg/dL)

  • ionized calcium: 4 mg/dL (normal for infants to adults: 4.5 to 4.92 mg/dL)

  • phosphorus: 2.5 mg/dL (normal for 4 to 11 years: 3.7 to 5.6 mg/dL)

  • magnesium: 1.6 mg/dL (normal for 2 to 14 years: 1.5 to 2.3 mg/dL)

  • albumin 2.9 g/dL (normal for children 7 to 19 years: 3.7 to 5.6 g/dL)

  • ALT: 55 units/L (normal for 1 to 19 years: 5 to 45 units/L)

  • AST: 65 units/L (normal for children 10 to 15 years: 10 to 40 units/L)

  • alkaline phosphatase: 863 units/L (normal for children 2 to 10 years: 100 to 320 units/L)

QUESTION: What evidence exists that this patient has rickets? How did his medications affect his serum phosphorus, calcium, and liver enzymes, and how would you modify his drug therapy?

DISCUSSION: Rickets is diagnosed by both radiologic and laboratory findings. The preliminary radiographic findings and the physical findings of the pigeon breast deformity (ie, the sternum and adjacent cartilage appear to be projected forward) and the palpable enlargement of costochondral junctions (rachitic rosary sign) are compatible with the diagnosis of rickets. Serum calcium may be low or normal in patients with rickets, depending on the etiology. The primary causes of rickets in the United States are vitamin D deficiency (with secondary hyperparathyroidism), primary phosphate deficiency, and end-organ resistance to 1,25-dihydroxyvitamin D. In patients with vitamin D deficiency, serum calcium concentrations can be normal or low, phosphorus concentrations are usually low, and serum alkaline phosphatase is elevated. In patients with primary phosphate deficiency, serum calcium is normal, serum phosphorus is low, and serum alkaline phosphatase is elevated. In patients with end-organ resistance to 1,25-dihydroxyvitamin D, serum calcium is low, serum phosphorus may be low or normal, and serum alkaline phosphatase is elevated.

In this patient, ionized calcium and serum phosphorus are both low and serum alkaline phosphatase is high, all of which are consistent with a diagnosis of rickets. Serum magnesium is normal for age; ALT and AST are slightly elevated. The serum magnesium was obtained because hypomagnesemia may also cause hypocalcemia. He also has hypoalbuminemia. A total serum calcium measures all three forms of extracellular calcium: complex bound, protein bound, and ionized. In patients with low albumin, the concentration of ionized calcium will be increased for a given total serum calcium concentration. Equations can be used to “correct” total serum calcium measurements for low concentrations of serum albumin, but these equations have limitations and may not be precise. Thus, in patients with low albumin (like this patient), an ionized serum calcium should be obtained.

His medications affected his laboratory tests. He is receiving an aluminum-containing antacid, which binds phosphorus in the GI tract. This resulted in decreased absorption of phosphorus and contributed to his low serum phosphorus. Enzyme-inducing anticonvulsants, such as phenobarbital and phenytoin, increase the metabolism of vitamin D and may result in a deficiency of vitamin D with resultant anticonvulsant-induced osteomalacia and rickets. Both the aluminum-containing antacid and the anticonvulsants contributed to him developing rickets, and thus, to the elevated serum alkaline phosphatase. In addition, due to his other medical conditions, he is nonambulatory and resides at a long-term care facility. Thus, he may have a lack of exposure to sunlight and, therefore, a lack of vitamin D. This lack of vitamin D also would contribute to the development of rickets.

For treatment of his rickets, he should be started on oral supplements of calcium, phosphorous, and vitamin D. However, modifications in his preadmission medications should be made. The aluminum-containing antacid (aluminum hydroxide suspension) should be discontinued and replaced with a calcium-containing antacid (eg, calcium carbonate). The amount of calcium in this new antacid should then be subtracted from any calcium supplement that would be started in the hospital, so that the total daily dose of calcium stays the same. Alternatively, the total dose of calcium supplement can be given as calcium carbonate. Discontinuing the aluminum-containing antacid will result in a greater amount of phosphorus absorbed enterally, which will then require a decrease in the oral supplement of phosphate (depending on serum phosphorus concentrations). Once he is stable, his neurologist should be consulted to see if other anticonvulsants that have less of an enzyme-inducing effect could be used to treat his seizures.

Inadequate dietary phosphate intake is an unusual cause of hypophosphatemia in adults. However, infants are more predisposed to nutritional hypophosphatemia due to their higher phosphorus requirements.23 The phosphorus requirements of premature infants are even higher due to their rapid skeletal growth. If premature infants are fed regular infant formula (instead of premature infant formula that contains additional calcium and phosphorus), phosphorus deficiency and rickets may occur. Phosphorous deficiency and rickets also can occur in pediatric patients who receive aluminum hydroxide–containing antacids, which bind dietary and secreted phosphorous and prevent its absorption from the GI tract. Inadequate vitamin D intake and genetic causes of vitamin D deficiency (eg, vitamin D-dependent rickets type 1) also can result in hypophosphatemia in pediatric patients.

Hypophosphatemia caused by intracellular shifting of phosphorus occurs with processes that stimulate intracellular phosphorus use. For example, high serum levels of glucose stimulate insulin. Insulin then enables glucose and phosphorus to move into the cell, where phosphorus is used during glycolysis. Intracellular shifting of phosphorus also occurs during anabolism, such as in patients receiving hyperalimentation and during refeeding in patients with protein-calorie malnutrition (eg, severe anorexia nervosa). The high anabolic (growth) rate in infants (especially premature infants) and children make them more susceptible to hypophosphatemia when adequate amounts of phosphate are not supplied in the hyperalimentation solution. Hypophosphatemia as a result of refeeding malnourished children usually occurs within 5 days of refeeding. It may be prevented by a more gradual increase in nutrition and phosphate supplementation.23


Hyperphosphatemia in pediatric patients may be caused by decreased excretion of phosphorus, increased intake of phosphate or vitamin D, or a shift of intracellular phosphate to extracellular fluid. The most common cause of hyperphosphatemia in the pediatric population is decreased excretion of phosphorus due to renal failure. Excessive phosphorus intake in pediatric patients (especially in those with renal dysfunction or in neonates whose renal function is normally decreased due to immaturity) is a common cause of hyperphosphatemia.23 Hyperphosphatemia may also occur if neonates are inappropriately given whole cow’s milk. As previously mentioned, cow’s milk contains a high phosphate load, which can cause hyperphosphatemia and secondary hypocalcemia in the neonate. Administration of sodium phosphorus laxatives or enemas to infants and children may also result in excessive phosphate intake. In addition, the pediatric dosing of phosphate supplements may be confusing to some because of the multiple salts available and multiple units of measure. This may result in unintentional overdoses with resultant hyperphosphatemia.


Normal ranges are as follows15,16:

neonates 0 to 6 days: 1.2 to 2.6 mg/dL (0.49 to 1.07 mmol/L)

neonates 7 days to children 2 years: 1.6 to 2.6 mg/dL (0.66 to 1.07 mmol/L)

children 2 years to adolescents 18 years: 1.5 to 2.3 mg/dL (0.62 to 0.95 mmol/L)

adults: 1.3 to 2.1 mEq/L (0.65 to 1.05 mmol/L)

Magnesium plays an important role in neuromuscular function and is a required cofactor for many enzymatic systems in the body. Approximately 50% to 60% of magnesium is located in bone, with one-third being slowly exchangeable with extracellular fluid. About 45% of magnesium is found in the intracellular fluid, with only 1% in extracellular fluid. The kidney is the primary organ responsible for magnesium excretion. Approximately 95% to 97% of filtered magnesium is reabsorbed: 15% in the proximal tubule, 70% in the thick ascending limb of Henle, and 5% to 10% in the distal tubule.23 In the neonate, reabsorption of magnesium may be increased in the proximal tubule. Thus, the immature neonatal kidney tends to retain magnesium compared with adults.25 This results in slightly higher normal values for serum magnesium in neonates and infants compared with older children and adults. In fact, serum magnesium concentrations in the newborn have been shown to be inversely related to GA at birth and postmenstrual age (PMA). In other words, more immature neonates have slightly higher serum magnesium concentrations.31,32


Hypomagnesemia occurs in pediatric patients because of excessive renal or GI losses, decreased GI absorption, decreased intake, and specific neonatal causes.23 Hypomagnesemia may occur in neonates due to several maternal causes. Maternal diuretic use, laxative overuse or abuse, diabetes mellitus, or decreased intake due to vomiting during pregnancy may cause maternal hypomagnesemia and lead to hypomagnesemia in the newborn.33 Hypomagnesemia also commonly occurs in neonates with intrauterine growth retardation (due to deficient placental transfer of magnesium) and in neonates who receive exchange transfusions with citrated blood.

Excessive renal losses of magnesium may be caused by various reasons. Of particular pediatric concern is the use of medications (eg, diuretics, amphotericin, proton pump inhibitors, aminoglycosides, and cisplatin) that may cause magnesium wasting. Hypomagnesemia may also occur due to rare hereditary renal magnesium-losing syndromes, such as Bartter syndrome and autosomal recessive renal magnesium–wasting syndrome. Excessive GI losses of magnesium may occur in pediatric patients with diarrhea or large losses of gastric contents (eg, emesis or nasogastric suction). Decreased GI absorption of magnesium may occur in patients with short gut syndrome. These patients have had a portion of their small bowel removed, which results in poor intestinal absorption. Other important pediatric GI diseases that may result in hypomagnesemia include cystic fibrosis, inflammatory bowel disease, and celiac disease.23

Poor magnesium intake may also result in hypomagnesemia. Although this rarely occurs in children fed orally, it may occur in hospitalized children receiving inadequate amounts of magnesium in IV fluids or parenteral nutrition solutions. Hypomagnesemia also can occur during the refeeding of children with protein-calorie malnutrition (eg, severe anorexia nervosa). These patients have low magnesium reserves but a high requirement of magnesium because of cellular growth.23


As in adults, the most common cause of hypermagnesemia in pediatric patients is renal dysfunction. However, in neonates, the most common cause is the IV infusion of magnesium sulfate in the mother for the prevention and treatment of eclampsia or for fetal neuroprotection.23,33 The high levels of magnesium in the mother are delivered transplacentally to the fetus. Neonates and young infants are also more prone to hypermagnesemia because of their immature renal function. Thus, these patients cannot easily tolerate a magnesium load. Other common pediatric causes of hypermagnesemia include excessive intake due to magnesium-containing antacids, laxatives, or enemas.


Serum Creatinine

Jaffe Method—normal ranges are as follows34:

neonates: 0.3 to 1 mg/dL (27 to 88 µmol/L)

infants: 0.2 to 0.4 mg/dL (18 to 35 µmol/L)

children: 0.3 to 0.7 mg/dL (27 to 62 µmol/L)

adolescents: 0.5 to 1 mg/dL (44 to 88 µmol/L)

adult males: 0.6 to 1.2 mg/dL (53 to 106 µmol/L)

adult females: 0.5 to 1.1 mg/dL (44 to 97 µmol/L)

Isotope Dilution Mass Spectrometry (IDMS)–Traceable Enzymatic Method—normal ranges are as follows15:

neonates to children 4 years: 0.03 to 0.5 mg/dL (2.65 to 44.2 µmol/L)

children 4 to 7 years: 0.03 to 0.59 mg/dL (2.65 to 52.2 µmol/L)

children 7 to 10 years: 0.22 to 0.59 mg/dL (19.4 to 52.2 µmol/L)

children and adolescents 10 to 14 years: 0.31 to 0.88 mg/dL (27.4 to 77.8 µmol/L)

adolescents >14 years: 0.5 to 1.06 mg/dL (44.2 to 93.7 µmol/L)

Serum creatinine (SCr) is a useful indicator of renal function and can be used to estimate GFR. Creatinine is generated from the metabolism of creatine and creatine phosphate, a high-energy biochemical important in muscle activity. Creatinine is produced in muscles, released into the extracellular fluid, and excreted by the kidneys. Excretion of creatinine is primarily via glomerular filtration, but a smaller amount undergoes tubular secretion. The amount of creatinine that is secreted by the tubules increases in patients as GFR decreases. Thus, creatinine clearance (CrCl) overestimates the actual GFR in patients with renal insufficiency.35,36

In pediatric patients, three major factors influence the SCr concentration: muscle mass per unit of body size, GFR, and (in newborns) the exogenous (maternal) creatinine load.36 At birth, the newborn’s SCr reflects the maternal SCr because SCr crosses the placenta. If a pregnant woman has an elevated SCr, then the concentration of creatinine in the fetus also will be elevated. In fact, the plasma creatinine concentration of umbilical cord blood is almost equal to the creatinine concentration in the mother.27 In full-term newborns, SCr may increase slightly shortly after birth because of the contraction of the ECW compartment.36 SCr then decreases over the first few days of life and usually reaches 0.4 mg/dL (Jaffe method) by about 10 days of age.36 The apparent half-life of this postnatal decrease in SCr is about 2.1 days in normal full-term infants and is due to the ongoing maturation of the kidneys and progressive increase in GFR. SCr is higher at birth in premature newborns compared with full-term newborns, and the postnatal decrease in SCr may occur more slowly. This is due to the preterm newborn’s more immature kidneys and lower GFR.27,37

Compared with adults, pediatric patients have a lower muscle mass per unit of body size. Because the production of creatinine depends on muscle mass, this results in significantly lower normal values for SCr for neonates, infants, and children. The percentage of muscle mass differs with various pediatric age groups and increases with age from birth through young adulthood.36 This increase in muscle mass accounts for the increase in the normal values for SCr with increasing age (see normal values for SCr above).

Creatinine excretion depends on GFR and, as in adults, SCr becomes elevated in pediatric patients with renal dysfunction. For example, an infant with an SCr as measured by the Jaffe method of 0.8 mg/dL (twice the normal value for age) has approximately a 50% decrease in GFR. Using age-appropriate normal values to interpret SCr is essential. In the previous example, an SCr of 0.8 mg/dL (which would be considered normal in an adult) denotes significant renal dysfunction in younger patients. Correct interpretation of SCr values is extremely important because the doses of fluids, many electrolytes, and medications that are renally eliminated need to be adjusted. Misinterpretation of SCr (eg, not recognizing renal dysfunction) can result in serious and potentially fatal fluid and electrolyte imbalances and overdosing of medications.

Reliable and accurate measurement of SCr is clinically important to properly assess renal function. Historically, SCr was measured by the alkaline picrate-based method, also known as the Jaffe method. However, substances that interfere with the measurement of creatinine by this method (ie, noncreatinine chromogens such as uric acid, glucose, fructose, and acetone) can cause an overestimation of SCr and, thus, an underestimation of kidney function. In addition, certain medications, endogenous substances, and medical conditions (eg, bilirubin, lipemia, and hemolysis) may interfere with the determination of SCr by this method.6,14 This interference may be a problem in the neonatal population because neonates often have hyperbilirubinemia or lipemia, and blood sampling methods in neonates often results in hemolysis.

Inaccuracies of the Jaffe method and other methodologies have led the National Kidney Disease Education Program to recommend a recalibration and standardization of SCr measurements.38 This has resulted in implementation of improved methods of SCr determinations, such as an enzymatic assay with an isotope dilution mass spectrometry (IDMS)–traceable international standard. It is important to know what methodology a laboratory is using because measurement by newer assays results in lower SCr determinations. Thus, the normal value of SCr for a specific patient depends on the assay method being used and the age of the patient (see normal values for SCr above). Further information about laboratory measurement and reporting of SCr can be found in Chapter 10.

Age-Related Physiologic Development of Renal Function

Compared with adults, a newborn’s kidneys are anatomically and functionally immature. The primary functions of the kidney (glomerular filtration, tubular secretion, and tubular reabsorption) are all decreased in the full-term newborn. These renal functions are even further decreased in the premature infant. After birth, glomerular and tubular renal function increase (ie, mature) with PNA. During the first 2 years of life, kidney function matures to adult levels in the following order: (1) glomerular filtration, (2) tubular secretion, and (3) tubular reabsorption. The interpretation of pediatric kidney functions tests can be better understood if one knows how each function of the kidneys matures during the first 2 years.

Glomerular filtration

In the fetus, nephrogenesis (ie, the formation of new nephrons) begins at 7 to 8 weeks of gestation and continues until 34 to 36 weeks of gestation.35,39,40 Although the number of adult nephrons (~1 million) is reached at this time, the nephrons are smaller and not as functionally mature as the nephrons found in an adult kidney.24 After 36 weeks of gestation, no new nephrons are formed. However, renal mass continues to increase due to the increase in renal tubular growth. GFR is very low in the young fetus but gradually increases during gestation (Figure 23-4). Before 36 weeks of gestation, the increase in GFR is primarily due to nephrogenesis and the increase in the number of new glomeruli.39 From 36 weeks of gestation until birth, a much smaller increase in GFR occurs as renal mass and kidney function increase.

FIGURE 23-4.
FIGURE 23-4.
Maturation of GFR in relation to conceptional age. (Source: Reproduced with permission from Guignard JP. The neonatal stressed kidney. In: Gruskin AB, Norman ME, eds. Pediatric Nephrology. Philadelphia, PA: Martinus Nijhoff Publishers; 1981:507.)

At birth, GFR increases dramatically compared with what it was in utero (Figure 23-4). This dramatic increase in GFR, which occurs at birth and continues during the early postnatal period, is the result of several important hemodynamic and physiologic changes. Cardiac output and systemic blood pressure increase at birth and a significant decrease in renal vascular resistance occurs. These changes result in an increase in renal blood flow and effective glomerular filtration pressure. In addition, alterations in the pattern of renal blood flow distribution occur and the permeability of the glomerular membrane and surface area available for filtration increase.24,39,40 All of these changes help to increase GFR.

Despite the increase in GFR that occurs during this time, GFR is still very much decreased in comparison with adults. As determined by creatinine or inulin clearance, the GFR in a full-term newborn is only 10 to 15 mL/min/m2 (2 to 4 mL/min). GFR then doubles by 1 to 2 weeks of age to 20 to 30 mL/min/m2 (8 to 20 mL/min).39 Adult values of GFR are approached by about 6 to 12 months of age (70 to 90 mL/min/m2). Compared with full-term newborns, GFR in premature newborns is much lower at birth (5 to 10 mL/min/m2 or 0.7 to 2 mL/min) and increases at a less dramatic rate during the first 1 to 2 weeks after birth (10 to 12 mL/min/m2 or 2 to 4 mL/min).39 After the first postnatal week, the rate of increase in GFR is comparable in preterm and full-term infants, but the actual GFR value is still lower in preterm infants than full-term infants.

Renal tubular function

Tubular secretion and reabsorption are both decreased in the full-term newborn because of the small size and mass of the renal tubules, decreased peritubular blood flow, and immature biochemical processes that supply energy for active transport. In addition, full-term newborns have a limited ability to concentrate urine and have lower urinary pH values.39 In the preterm newborn, renal tubular functions are further decreased. Limitations of the newborn’s tubular function with respect to the renal handling of serum electrolytes are listed above in the discussion section of each serum electrolyte.

Tubular secretion transports certain electrolytes and medications from the peritubular capillaries into the lumen of the renal tubule. At birth, tubular secretion is only 20% to 30% of adult values and slowly matures by about 8 months of age. Tubular reabsorption, which also is decreased at birth, may not fully mature until 1 to 2 years of age. Thus, during infancy, a glomerulotubular imbalance occurs, with GFR maturing at a faster rate than renal tubular function.

The decreased renal function in newborns and the maturational changes in GFR and tubular function that occur throughout early infancy have important implications for the interpretation of laboratory data. For example, one must remember that even with a normal SCr for age, neonates and infants still have decreased renal function compared with adults. This decreased renal function must be taken into account, especially in the very young, when dosing electrolytes or medications that are eliminated by the kidneys. In addition, as in adults, certain medications, diseases, and medical conditions (such as hypoxic events that may occur in newborns) may cause further decreases in renal function.

Standardization of Creatinine Clearance

Creatinine clearance (CrCl) can be expressed using several different units of measure, including mL/min, mL/min/m2, or mL/min/1.73 m2. To better compare the CrCl of patients of different body sizes, CrCl is most commonly standardized to the BSA of an average-sized adult (1.73 m2). Thus, CrCl is most commonly measured as mL/min/1.73 m2. Using these units is especially helpful in pediatric patients, in whom a large range of body sizes occurs. For example, the average BSA ranges from 0.22 m2 in a full-term newborn to 1.32 m2 in a 12-year-old female and 1.77 m2 in a 17-year-old male.1 Expressing CrCl in mL/min or even mL/min/m2 over this wide of a range of BSAs would give an extremely wide range of values.

Estimating Body Surface Area in Pediatric Patients

Body surface area (BSA) can be estimated using several different methods. In pediatric patients, BSA is most commonly estimated using standard nomograms or equations and the patient’s measured height and weight. Two equations are commonly used in pediatric practice, an older equation (the DuBois formula)41:
and a more simplified equation (the Mosteller formula)42:
BSAm2=the square root ofHtcm×Wtkg/3,600

Estimation of the patient’s BSA is required to calculate CrCl from a urinary collection.

Determination of Creatinine Clearance from a Urinary Creatinine Collection

The same equation that is used in adults can be used in pediatric patients to calculate CrCl from a timed urine collection. The following equation is used:

where CrCl is in units of mL/min/1.73 m2; U is the urinary creatinine concentration in mg/dL; V (mL/min) is the total urine volume collected in milliliters divided by the duration of the collection in minutes; P is the SCr concentration in mg/dL; and BSA is the patient’s BSA in m2.

Ideally, urine should be collected over a 24-hour period. However, a full 24-hour collection period is difficult in pediatric patients, especially in those who do not have full control over their bladder and do not have a urinary catheter in place. Thus, shorter collection periods (eg, 8 or 12 hours) are sometimes used. Urinary specimen bags can be placed to collect urine in neonates and infants, but incomplete collection because of urine leakage often occurs. The incomplete collection of urine results in an inaccurate calculation of CrCl.

With any urine collection for creatinine determination, it is important to have the patient empty the bladder and discard this specimen before beginning urine collection. All urine during the time period should be collected, including the urine that would be voided at the end of the collection period. An SCr is usually obtained once during the urinary collection period (ideally at the midpoint) if the patient has stable renal function. If the patient’s renal function is changing, then two SCr samples (one at the beginning of the urine collection and one at the end of the urine collection) may be obtained. The average SCr can then be used in the above equation.3

Because of the inherent problems of collecting a 24-hour urine sample from pediatric patients and receiving inaccurate calculations, CrCl (or GFR) is often estimated using prediction equations that consider the patient’s age, height, sex, and SCr (see below). In fact, using a 24-hour timed urine specimen to calculate CrCl has been shown to be no more reliable (and often less reliable) than using equations based on SCr.43 Therefore, the National Kidney Foundation recommends that GFR should be estimated in children and adolescents using prediction equations, such as the one by Schwartz et al.36 A timed urine collection (eg, 24-hour sample) may be useful for (1) estimations of GFR in patients with decreased muscle mass (eg, muscle wasting, malnutrition, or amputation) or patients receiving special diets (eg, vegetarian diets or creatine supplements); (2) assessments of nutritional status or diets; and (3) evaluations for the need to start dialysis.43

Estimating Creatinine Clearance and Glomerular Filtration Rate from Serum Creatinine

In adults, several methods are used to estimate CrCl and GFR from an SCr. For example, the Cockcroft-Gault equation is used to estimate CrCl and the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation can be used to estimate GFR. These equations cannot be used for pediatric patients because they have a different ratio of muscle mass to SCr. In addition, the amount of pediatric muscle mass per body weight changes over time. Adult equations are based on adult muscle mass and adult urinary creatinine excretion rates. Thus, the use of adult equations in pediatric patients results in erroneous calculations.

Several predictive equations have been developed to estimate CrCl in pediatric patients. One simple equation was developed for use in children 1 to 18 years of age with stable SCr44:

This equation was found to be clinically useful in predicting CrCl in children. However, it may be less accurate in children with a height <107 cm.45

Another equation, developed by Schwartz et al, is more commonly used to estimate GFR in pediatric patients:

where k is a constant of proportionality.36 In patients with stable renal function, k is directly related to the muscle component of body weight, which correlates well with the daily rates of urinary creatinine excretion. Because the percentage of muscle mass per body weight varies for different age groups, a different value for k must be used for different age groups. In addition to age, the value of k is affected by body composition; thus, the values for k listed in Table 23-3 should be used in pediatric patients with average build.

TABLE 23-3.
Mean Values of k by Age Group



Low birth weight infants ≤1 yra


Full-term infants ≤1 yr


Children 1–12 yr


Females 13–21 yr


Males 13–21 yr


aLow birth weight infant is defined as an infant with a birthweight ≤2,500 g.

Source: Modified from Schwartz GJ, Brion LP, Spitzer A. The use of plasma creatinine concentration for estimating glomerular filtration rate in infants, children, and adolescents. Pediatr Clin North Am. 1987;34(3):571–590.

It is important to remember that these two equations were developed using the Jaffe method to assay SCr. Values for SCr as measured by newer methods (eg, the IDMS-traceable enzymatic method) are lower than those measured by the Jaffe method (especially at low concentrations of SCr). This results in an overestimation of GFR if these equations and constants are used when SCr is measured by the newer assays. In fact, use of the Schwartz equation and constants may overestimate GFR by approximately 20% to 30%.46,47 Therefore, revised age-group constants for the Schwartz equation should be determined to better estimate GFR when SCr is measured by newer assays using IDMS-based reference standards (Minicase 3).

In 2009, an “updated” Schwartz bedside equation, based on SCr measured by an IDMS-traceable enzymatic method, was developed in children 1 to 16 years of age.48

The pediatric patients in whom this equation was developed had mild-to-moderate chronic kidney disease (95% of measured GFR values were between 21 and 76 mL/min/1.73 m2) and were short for their age and sex. Further studies of this equation in children with higher GFRs and a more normal body habitus are needed before it can be widely applied to the pediatric population.48 However, in general, for children with chronic kidney disease, this equation provides a relatively accurate clinical estimate of GFR (when SCr is measured by an IDMS-traceable assay) and its use is recommended by the Kidney Disease Improving Global Outcomes (KDIGO) clinical practice guidelines for the evaluation and management of chronic kidney disease.49,50 Use of this equation (Equation 7) is clinically recommended whenever SCr is measured by an IDMS-traceable assay because it results in a more accurate estimation of GFR, as compared with using the original Schwartz equation (Equation 6).

Part I: Piperacillin–Tazobactam Dose Determination Based on Estimated Glomerular Filtration Rate in a Child

Mary M., a 5-year-old female, presents to the emergency department with a 4-day history of irritability, diffuse abdominal pain, and inability to tolerate oral intake. Mary’s parents report that in the past few hours, she has appeared more lethargic, had one episode of rectal bleeding, and has been febrile with a Tmax of 102.2°F. In the emergency department, the patient has signs of sepsis, including hypotension (BP 78/49 mm Hg) and tachycardia (HR 140 beats/min). Fluid resuscitation is initiated with a 20 mL/kg bolus of 0.9% sodium chloride, and a dose of ceftriaxone and metronidazole is administered. An abdominal radiograph is concerning for volvulus, a condition in which a part of the small intestine twists around itself, causing an obstruction that may impact blood flow to the intestinal tissue. The patient is taken immediately to the operating room for exploratory laparotomy with pediatric surgery. Her weight is 16 kg (25th percentile for age) and height is 107 cm (50th percentile for age); BSA is 0.69 m2.

The patient arrives to the pediatric intensive care unit postoperatively after reduction of the volvulus. The surgery was complicated by need for an 8-cm bowel resection of necrotic intestinal tissue. The patient is intubated and sedated and remains hypotensive despite fluid resuscitation. Laboratory data include sodium 143 mEq/L, potassium 4.8 mEq/L, chloride 110 mEq/L, total CO2 21 mEq/L, BUN 31 mg/dL, SCr 2.3 mg/dL (Jaffe method), and glucose 109 mg/dL. Intraabdominal tissue cultures were obtained during surgery; however, results are pending. The pediatric surgeon is recommending initiation of postoperative piperacillin–tazobactam for coverage of enteric gram-negative pathogens and anaerobes.

QUESTION: Knowing the following information, what dose of piperacillin–tazobactam would you recommend for this patient?

The normal dose of piperacillin–tazobactam used at this hospital for a complicated intra-abdominal infection is 300 mg piperacillin/kg/day divided q 6 h. The recommended dosing adjustment for patients with renal dysfunction is as follows1:

  • GFR >50 mL/min/1.73 m2: no adjustment required

  • GFR 30 to 50 mL/min/1.73 m2: 35 to 50 mg piperacillin/kg/dose q 6 h

  • GFR <30 mL/min/1.73 m2: 35 to 50 mg piperacillin/kg/dose q 8 h

DISCUSSION: Piperacillin–tazobactam is a combination product of two antimicrobial agents. Dosing recommendations are based on the piperacillin component of this combination product. It is primarily excreted renally with 68% of piperacillin and 80% of tazobactam excreted as unchanged drug in the urine. Mary’s BUN and SCr are elevated, indicating that she has some degree of renal impairment likely secondary to hypovolemia and septic shock. Thus, the dose of piperacillin–tazobactam must be adjusted for her renal dysfunction. To recommend an appropriate dose, her GFR needs to be calculated.

Because this laboratory used the Jaffe method to measure SCr, an estimation of GFR from her SCr and height can be obtained by using Equation 6 and the k value of 0.55 for children 2 to 12 years of age (Table 23-3). Her estimated GFR is ~26 mL/min/1.73 m2. Because she weighs 16 kg, the normal dose of piperacillin–tazobactam for this patient would be 1,200 mg piperacillin IV q 6 h (300 mg piperacillin/kg/day divided q 6 h). Because her GFR is 26 mL/min/1.73 m2, she should receive a dose of 35 to 50 mg piperacillin/kg/dose q 8 h. The higher end of this dosing range would be appropriate for this patient given her severity of illness. Therefore, a dose of 800 mg piperacillin IV q 8 h (150 mg piperacillin/kg/day divided q 8 h) should be recommended. The patient’s clinical response to antimicrobial therapy and changes in renal function should be evaluated daily to ensure appropriate response and/or toxicity. It should be noted that if SCr were measured by a newer enzymatic assay using IDMS-based reference standards, then Equation 7 should be used to estimate GFR. If Equation 6 and the constants from Table 23-3 were used, it would result in an overestimation of GFR by approximately 20% to 30%. Additionally, almost all drug dosage guidelines for patients with renal impairment (including those recommended by pharmaceutical manufacturers) were developed using SCr as measured by older assay methodologies. Thus, estimated GFR values may not correlate well with these dosage guidelines. These limitations further emphasize the need for appropriate follow-up and monitoring of laboratory parameters and clinical response to determine if dosage adjustments are required.

Part II: Piperacillin–Tazobactam Dose Determination Based on Estimated Glomerular Filtration Rate in a Child

A few days later, Mary is extubated, is hemodynamically stable (BP 98/62 mm Hg), and appears to be clinically improving. This morning, the hospital instituted the newer assay using IDMS-traceable enzymatic reference standards for SCr measurement. Mary’s SCr is now 1.1 mg/dL (IDMS-traceable method) and the medical resident suggests increasing the piperacillin–tazobactam dose to the normal dose of 300 mg piperacillin/kg/day divided q 6 h given the patient’s GFR has “normalized.”

QUESTION: Should Mary’s piperacillin–tazobactam dose be adjusted to the normal dose of 300 mg piperacillin/kg/day divided q 6 h?

DISCUSSION: To answer this question and recommend an appropriate antibiotic dose, Mary’s GFR needs to be calculated using the appropriate equation based on the laboratory assay utilized. Because the laboratory has switched to the IDMS method to measure SCr, an estimation of GFR from her SCr and height should be obtained by using Equation 7. Her estimated GFR is now ~40 mL/min/1.73 m2 (= 0.413 × 107 cm/1.1 mg/dL). Although her GFR has improved from 26 mL/min/1.73 m2, it has not completely normalized for her age. Her SCr of 1.1 mg/dL remains elevated with a normal SCr (IDMS method) for a child this age being 0.03 to 0.59 mg/dL. This shows that Mary is still experiencing some renal dysfunction or acute kidney injury. Although it is appropriate to increase her piperacillin–tazobactam dose for her improvement in renal function, a dose of 35 to 50 mg piperacillin/kg/dose IV q 6 h should be used. A calculated dose of 800 mg piperacillin IV q 6 h (200 mg piperacillin/kg/day divided q 6 h) would be recommended based on the renal dose adjustments provided above and her illness severity. It should be noted that if Equation 6 and the constants from Table 23-3 had been used, the patient’s GFR would have been calculated as 53.5 mL/min/1.73 m2 (0.55 × 107 cm/1.1 mg/dL). This would be approximately a 30% overestimation of GFR. The medical resident’s recommendation of increasing the piperacillin–tazobactam dose to 300 mg piperacillin/kg/day divided q 6 h was likely a result of using the wrong equation for estimating GFR. Utilization of this higher piperacillin-tazobactam dose, based on the GFR estimation by Equation 6 would have resulted in too high of a dosage for this patient’s renal function, ultimately placing her at a higher risk of toxicity. Daily monitoring of laboratories is required because multiple dose adjustments may need to occur as her renal function continues to improve. An alternate approach to estimating GFR would have been to use cystatin C and Equation 8. The benefit of using the cystatin C method of GFR estimation is the lack of influence of other factors, such as muscle mass, sex, body composition, and age on GFR estimation. At this time, cystatin C is not widely utilized in clinical practice. Some hospital laboratories may not have the ability to process this test, requiring transfer of the sample to an offsite laboratory (with a delay in receiving results) and making the use of a cystatin C level in the acute setting limited.

These equations may not be accurate in certain pediatric populations, including patients with unstable renal function (eg, acute kidney injury), abnormal body habitus (eg, obesity or malnutrition), decreased muscle mass (eg, cardiac patients), severe chronic renal failure, or insulin-dependent diabetes.36,51 If clinically indicated, CrCl should be determined by a timed urine collection in these patients.

Estimating Glomerular Filtration Rate from Serum Cystatin C

To improve calculated estimations of GFR, other endogenous markers of renal function have been investigated. Cystatin C is a cysteine protease inhibitor that is produced by all nucleated cells at a relatively constant rate; it is freely filtered by the kidneys. Its rate of production is so constant that serum cystatin C levels are not thought to be influenced by muscle mass, sex, body composition, and age (after 12 months). Reciprocal values of serum cystatin C have been correlated to measured GFR values in adults and children.46 However, serum cystatin C alone may not accurately estimate GFR in renal transplant patients, patients with high C-reactive protein, diabetes with ketonuria, or thyroid dysfunction. Thus, other equations have been developed which incorporate multiple patient parameters and endogenous markers of renal function. One such pediatric equation includes terms for the ratio of height to SCr, the reciprocals of serum cystatin C and BUN, a factor for sex, and a separate factor for height alone.48,49 This complex equation is not clinically friendly but is being further investigated in the National Institutes of Health–sponsored Chronic Kidney Disease in Children study. Results of this investigation may yield a more accurate equation to estimate GFR in pediatric patients.

Recently, a user-friendly, simplified single variate serum cystatin C equation for clinical use in children (1 to 18 years) with chronic kidney disease has been developed.49
GFRmL/min/1.73m2=70.69×serum cystatin Cmg/L0.931

This equation is recommended by current KDIGO guidelines to estimate GFR in pediatric patients, if a serum cystatin C is measured.50 Unfortunately at this time, serum cystatin C assays are not readily available at many institutions.


Serum Albumin

Normal ranges are as follows11:

neonates 0 to 5 days, body weight <2.5 kg: 2 to 3.6 g/dL (20 to 36 g/L)

neonates 0 to 5 days, body weight >2.5 kg: 2.6 to 3.6 g/dL (26 to 36 g/L)

children 1 to 3 years: 3.4 to 4.2 g/dL (34 to 42 g/L)

children 4 to 6 years: 3.5 to 5.2 g/dL (35 to 52 g/L)

children and adolescents 7 to 19 years: 3.7 to 5.6 g/dL (37 to 56 g/L)

adults: 4 to 5 g/dL (40 to 50 g/L)

Serum proteins, including albumin, are synthesized by the liver. Thus, measurements of serum total protein, albumin, and other specific proteins are primarily a test of the liver’s synthetic capability. Maturational differences in the liver’s ability to synthesize protein help determine the normal range for serum albumin concentrations. The liver of the fetus is able to synthesize albumin beginning at approximately 7 to 8 weeks of gestation. However, the predominant serum protein in early fetal life is α-fetoprotein. As gestation continues, the concentration of albumin increases, whereas α-fetoprotein decreases. At approximately 3 to 4 months of gestation, the fetal liver is able to produce each of the major serum protein classes. However, serum concentrations are much lower than those found at maturity.52

At birth, the newborn liver is anatomically and functionally immature. Because of immature liver function and a decreased ability to synthesize protein, full-term neonates have decreased concentrations of total plasma proteins, including albumin, γ-globulin, and lipoproteins. Concentrations in premature newborns are even lower, with serum albumin levels as low as 1.8 g/dL.15 Adult serum concentrations of serum albumin (~3.5 g/dL) are reached only after several months of age. Conditions that cause abnormalities in serum albumin in pediatric patients are the same as in adults and can be reviewed in Chapter 15.

Liver Enzymes

Alanine aminotransferase (ALT, also called SGPT)—normal ranges are as follows12,15:

neonates 0 to 7 days: 6 to 40 units/L (0.1 to 0.67 μkat/L)

neonatal males 8 to 30 days: 10 to 40 units/L (0.17 to 0.67 μkat/L)

neonatal females 8 to 30 days: 8 to 32 units/L (0.13 to 0.53 μkat/L)

infants: 12 to 45 units/L (0.2 to 0.75 μkat/L)

children and adolescents 1 to 19 years: 5 to 45 units/L (0.08 to 0.75 μkat/L)

adults: 7 to 41 units/L (0.12 to 0.68 μkat/L)

Aspartate aminotransferase (AST, also called SGOT)—normal ranges are as follows:

neonatal males 0 to 7 days: 30 to 100 units/L (0.5 to 1.67 μkat/L)

neonatal females 0 to 7 days: 24 to 95 units/L (0.4 to 1.59 μkat/L)

neonates 8 to 30 days: 22 to 71 units/L (0.37 to 1.19 μkat/L)

infants: 22 to 63 units/L (0.37 to 1.05 μkat/L)

children 1 to 3 years: 20 to 60 units/L (0.33 to 1 μkat/L)

children 3 to 9 years: 15 to 50 units/L (0.25 to 0.84 μkat/L)

children and adolescents 10 to 15 years: 10 to 40 units/L (0.17 to 0.67 μkat/L)

adolescent males 16 to 19 years: 15 to 45 units/L (0.25 to 0.75 μkat/L)

adolescent females 16 to 19 years: 5 to 30 units/L (0.08 to 0.5 μkat/L)

adults: 12 to 38 units/L (0.2 to 0.63 μkat/L)

Alkaline phosphatase—normal ranges are as follows:

neonates 0 to 14 days: 90 to 273 units/L (1.5 to 4.56 µkat/L)

neonates 15 days to infants: 134 to 518 units/L (2.24 to 8.65 µkat/L)

children 1 to <10 years: 156 to 369 units/L (2.61 to 6.16 µkat/L)

children 10 years to adolescents <13 years: 141 to 460 units/L (2.35 to 7.68 µkat/L)

adolescent males 13 to <15 years: 127 to 517 units/L (2.12 to 8.63 µkat/L)

adolescent females 13 to <15 years: 62 to 280 units/L (1.04 to 4.68 µkat/L)

adolescent males 15 to <17 years: 89 to 365 units/L (1.49 to 6.1 µkat/L)

adolescent females 15 to <17 years: 54 to 128 units/L (0.9 to 2.14 µkat/L)

adolescent males 17 to <19 years: 59 to 164 units/L (0.99 to 2.74 µkat/L)

adolescent females 17 to <19 years: 48 to 95 units/L (0.8 to 1.59 µkat/L)

adults: 33 to 96 units/L (0.55 to 1.6 μkat/L)

Lactate dehydrogenase—normal ranges are as follows:

neonates 0 to 14 days: 309 to 1,222 units/L (5.16 to 20.41 μkat/L)

neonates 15 days to infants: 163 to 452 units/L (2.72 to 7.55 μkat/L)

children 1 to <10 years: 192 to 321 units/L (3.21 to 5.36 μkat/L)

adolescent males 10 to <15 years: 170 to 283 units/L (2.84 to 4.73 μkat/L)

adolescent females 10 to <15 years: 157 to 272 units/L (2.62 to 4.54 μkat/L)

adolescents 15 to <19 years: 130 to 250 units/L (2.17 to 4.18 μkat/L)

adults: 115 to 221 units/L (1.92 to 3.69 μkat/L)

The normal reference ranges for liver enzymes are higher in pediatric patients compared with adults. This may be due to the fact that the liver makes up a larger percentage of total body weight in infants and children compared with adults. For certain enzymes, such as alkaline phosphatase, the higher normal concentrations in childhood represent higher serum concentrations of an isoenzyme from other sources, specifically bone. Approximately 80% of alkaline phosphatase originates from liver and bone. Smaller amounts come from the intestines, kidneys, and placenta. Normally, growing children have higher osteoblastic activity during the bone growth period and an influx into serum of the alkaline phosphatase isoenzyme from bone.14 Thus, the higher normal concentrations of alkaline phosphatase in childhood primarily represent a higher rate of bone growth. After puberty, the liver is the primary source of serum alkaline phosphatase. Differences in normal values for liver enzymes also exist between male and female patients at certain ages. For example, during adolescence, males have higher alkaline phosphatase compared with females.

One must keep these sex and age-related differences in mind when interpreting liver enzyme test results. For example, an isolated increase in alkaline phosphatase in a rapidly growing male adolescent—whose other liver function tests are normal—would not indicate hepatic or biliary disease but merely a rapid increase in bone growth. As in adults, increases in AST and ALT in pediatric patients are associated with hepatocellular injury whereas elevations of alkaline phosphatase are associated with cholestatic disease. Cholestasis and bone disorders (such as osteomalacia and rickets) are common causes of elevated serum alkaline phosphatase concentrations in pediatric patients.


Total bilirubin, premature neonates—normal ranges are as follows53:

  • 0 to 1 day: <8 mg/dL (<137 µmol/L)

  • 1 to 2 days: <12 mg/dL (<205 µmol/L)

  • 3 to 5 days: <16 mg/dL (<274 µmol/L)

  • >5 days: <2 mg/dL (<34 µmol/L)

Total bilirubin, full-term neonates—normal ranges are as follows:

  • 0 to 1 day: <8 mg/dL (<137 µmol/L)

  • 1 to 2 days: <11.5 mg/dL (<197 µmol/L)

  • 3 to 5 days: <12 mg/dL (<205 µmol/L)

  • >5 days: <1.2 mg/dL (<21 µmol/L)

Total bilirubin, adults—normal range is as follows:

  • adults: 0.3 to 1.3 mg/dL (5.1 to 22 µmol/L)

Conjugated bilirubin—normal ranges are as follows:

  • neonates: <0.6 mg/dL (<10 µmol/L)

  • infants and children: <0.2 mg/dL (<3.4 µmol/L)

  • adults: 0.1 to 0.4 mg/dL (1.7 to 6.8 µmol/L)

To better understand the age-related differences in serum bilirubin concentrations, a brief review of bilirubin metabolism is needed. A detailed discussion of bilirubin metabolism and abnormalities can be found in Chapter 15. Bilirubin is a breakdown product of hemoglobin. Hemoglobin, which is released from senescent or hemolyzed RBCs, is degraded by heme oxygenase into iron, carbon monoxide, and biliverdin. Biliverdin undergoes reduction by biliverdin reductase to bilirubin. Unconjugated bilirubin then enters the liver and is conjugated with glucuronic acid to form conjugated bilirubin, which is water soluble. Conjugated bilirubin is excreted in the bile and enters the intestines, where it is broken down by bacterial flora to urobilinogen. However, conjugated bilirubin also can be deconjugated by bacteria in the intestines or by β-glucuronidase in intestinal tissue and reabsorbed into the circulation.54

Compared with adults, newborns have higher concentrations of bilirubin. This results from a higher production of bilirubin in the neonate and a decreased ability to excrete it. A higher rate of production of bilirubin occurs in newborns due to the shorter life span of neonatal RBCs and the higher initial neonatal hematocrit. The average RBC life span is only 65 days in very premature neonates and 90 days in full-term neonates, compared with 120 days in adults.55 In addition, full-term neonates have a mean hematocrit of about 50%, compared with adult values of approximately 44%. The shorter RBC life span plus the higher hematocrit both increase the load of unconjugated bilirubin to the liver. Newborn infants, however, have a decreased ability to eliminate bilirubin. The activity of neonatal uridine diphosphate glucuronosyltransferase, the enzyme responsible for conjugating bilirubin in the liver, is decreased. In addition, newborns lack the intestinal bacteria needed to break down conjugated bilirubin into urobilinogen. However, the newborn’s intestine does contain glucuronidase, which can deconjugate bilirubin and allow unconjugated bilirubin to be reabsorbed back into the circulation (enterohepatic circulation). This enterohepatically reabsorbed bilirubin further increases the unconjugated bilirubin load to the liver.

Because of these limitations in bilirubin metabolism and the resultant unconjugated (indirect) hyperbilirubinemia, a “physiologic jaundice” commonly occurs in newborns. Typically in full-term neonates, high serum bilirubin concentrations occur in the first few days of life, with a decrease over the next several weeks to values seen in adults. High bilirubin concentrations may occur later in premature newborns, up to the first week of life, and are usually higher and persist longer than in full-term newborns. Physiologic jaundice is usually transient and benign; however, sometimes hyperbilirubinemia and jaundice in a neonate can be a symptom of an underlying medical condition.54

Pathological (or nonphysiological) jaundice due to unconjugated hyperbilirubinemia may occur in newborns for many reasons, including increased production of bilirubin, decreased uptake of unconjugated bilirubin into the liver, decreased conjugation of bilirubin in the liver, and increased enterohepatic circulation of bilirubin.54 Increased production of bilirubin may occur with RBC hemolysis because of blood group incompatibilities (eg, Rh and ABO incompatibility), enzyme deficiencies of the erythrocytes (eg, glucose-6-phosphate dehydrogenase deficiency), erythrocyte structural defects (eg, hereditary spherocytosis), or certain racial or ethnic makeup (eg, Asian and Native American). Certain genetic disorders may cause neonatal hyperbilirubinemia. For example, patients with Gilbert syndrome have decreased activity of bilirubin uridine diphosphate glucuronosyltransferase, the enzyme that is responsible for conjugation of bilirubin in the liver. Infants with Crigler-Najjar syndrome may have absent (type I) or decreased (type II) activity of uridine diphosphate glucuronosyltransferase. Breastfeeding also is associated with neonatal hyperbilirubinemia and jaundice. Newborns who are exclusively breastfed, not feeding well, or not being enterally fed (ie, newborns who are not taking anything by mouth) may have increased intestinal reabsorption of bilirubin that can cause or worsen hyperbilirubinemia. Breastfeeding may also increase bilirubin concentrations by other mechanisms. Breast milk may contain substances that decrease the conjugation of bilirubin by inhibiting the enzyme uridine diphosphate glucuronosyltransferase. Some medications (eg, sulfonamides and certain cephalosporins, such as ceftriaxone) can displace unconjugated bilirubin from albumin-binding sites and may worsen unconjugated hyperbilirubinemia.1,54 Use of these medications in neonates, especially those with or at risk for hyperbilirubinemia, is not recommended. For example, sulfonamides are not indicated for use in infants <2 months of age (unless as a last drug option to treat a life-threatening infection), and ceftriaxone is contraindicated for use in hyperbilirubinemic neonates and in premature neonates <41 weeks’ postmenstrual age.1

Appropriate monitoring of serum bilirubin is important in neonates because high concentrations of unconjugated bilirubin can cause acute bilirubin encephalopathy or chronic bilirubin encephalopathy, also termed kernicterus (ie, deposits of unconjugated bilirubin in the brain). The neurotoxic effects of bilirubin are serious and potentially lethal. Clinical features of acute kernicterus include poor sucking, stupor, seizures, fever, hypotonia, hypertonia, opisthotonus, and retrocollis. Neonates who survive may develop mental retardation, cerebral palsy, delayed motor skills, movement disorders, and sensorineural hearing loss.54,56 Phototherapy (which converts unconjugated bilirubin to a more water-soluble isomer for easier excretion) and exchange transfusion (which removes bilirubin from the bloodstream) are common treatments for neonatal unconjugated hyperbilirubinemia. IV immunoglobulin may be used in patients with isoimmune hemolytic disease (RBC hemolysis due to Rh and ABO incompatibility) to decrease the need for exchange transfusions.54,56

Pathologic jaundice due to conjugated (direct) hyperbilirubinemia (also termed cholestasis) may also occur in newborns. Etiological factors may include obstruction of the biliary system (eg, biliary atresia), defects of bile acid transport or synthesis, metabolic liver diseases (eg, α1-antitrypsin deficiency), or systemic conditions such as infection (eg, (t)oxoplasmosis, (o)ther agents, (r)ubella, (c)ytomegalovirus, and (h)erpes simplex [TORCH] infections), parenteral nutrition-associated cholestasis, or acute liver injury from hypoxia or liver ischemia.54



Mean values and lower limit of normal (minus 2 standard deviations) for red blood cell count, hemoglobin and hematocrit are as follows.57

Red Blood Cell Count

birth (cord blood): 4.7 (3.9) × 1012 cells/L (4.7 (3.9) × 106 cells/µL)

1 to 3 days: 5.3 (4) × 1012 cells/L (5.3 (4) × 106 cells/µL)

1 week: 5.1 (3.9) × 1012 cells/L (5.1 (3.9) × 106 cells/µL)

2 weeks: 4.9 (3.6) × 1012 cells/L (4.9 (3.6) × 106 cells/µL)

1 month: 4.2 (3) × 1012 cells/L (4.2 (3) × 106 cells/µL)

2 months: 3.8 (2.7) × 1012 cells/L (3.8 (2.7) × 106 cells/µL)

3 to 6 months: 3.8 (3.1) × 1012 cells/L (3.8 (3.1) × 106 cells/µL)

0.5 to 2 years: 4.5 (3.7) × 1012 cells/L (4.5 (3.7) × 106 cells/µL)

2 to 6 years: 4.6 (3.9) × 1012 cells/L (4.6 (3.9) × 106 cells/µL)

6 to 12 years: 4.6 (4) × 1012 cells/L (4.6 (4) × 106 cells/µL)

12 to 18 years, female: 4.6 (4.1) × 1012 cells/L (4.6 (4.1) × 106 cells/µL)

12 to 18 years, male: 4.9 (4.5) × 1012 cells/L (4.9 (4.5) × 106 cells/µL)

18 to 49 years, female: 4.6 (4) × 1012 cells/L (4.6 (4) × 106 cells/µL)

18 to 49 years, male: 5.2 (4.5) × 1012 cells/L (5.2 (4.5) × 106 cells/µL)


birth (cord blood): 16.5 (13.5) g/dL (10.2 (8.4) mmol/L)

1 to 3 days: 18.5 (14.5) g/dL (11.5 (9) mmol/L)

1 week: 17.5 (13.5) g/dL (10.9 (8.4) mmol/L)

2 weeks: 16.5 (12.5) g/dL (10.2 (7.8) mmol/L)

1 month: 14 (10) g/dL (8.7 (6.2) mmol/L)

2 months: 11.5 (9) g/dL (7.1 (5.6) mmol/L)

3 to 6 months: 11.5 (9.5) g/dL (7.1 (5.9) mmol/L)

0.5 to 2 years: 12 (10.5) g/dL (7.4 (6.5) mmol/L)

2 to 6 years: 12.5 (11.5) g/dL (7.8 (7.1) mmol/L)

6 to 12 years: 13.5 (11.5) g/dL (8.4 (7.1) mmol/L)

12 to 18 years, female: 14 (12) g/dL (8.7 (7.4) mmol/L)

12 to 18 years, male: 14.5 (13) g/dL (9 (8.1) mmol/L)

18 to 49 years, female: 14 (12) g/dL (8.7 (7.4) mmol/L)

18 to 49 years, male: 15.5 (13.5) g/dL (9.6 (8.4) mmol/L)


birth (cord blood): 51 (42)% (0.51 (0.42))

1 to 3 days: 56 (45)% (0.56 (0.45))

1 week: 54 (42)% (0.54 (0.42))

2 weeks: 51 (39)% (0.51 (0.39))

1 month: 43 (31)% (0.43 (0.31))

2 months: 35 (28)% (0.35 (0.28))

3 to 6 months: 35 (29)% (0.35 (0.29))

0.5 to 2 years: 36 (33)% (0.36 (0.33))

2 to 6 years: 37 (34)% (0.37 (0.34))

6 to 12 years: 40 (35)% (0.40 (0.35))

12 to 18 years, female: 41 (36)% (0.41 (0.36))

12 to 18 years, male: 43 (37)% (0.43 (0.37))

18 to 49 years, female: 41 (36)% (0.41 (0.36))

18 to 49 years, male: 47 (41)% (0.47 (0.41))

Compared with adults, normal newborn infants have higher hemoglobin and hematocrit values. For example, the mean hemoglobin value in a full-term newborn on the first day of life is 18.5 g/dL, compared with 15.5 g/dL in adult males. Hemoglobin and hematocrit start to decrease within the first week of life and reach a minimum level at 8 to 12 weeks in term infants.58 This normal decrease in hemoglobin and hematocrit is called the physiologic anemia of infancy. This physiologic anemia is normochromic and microcytic and is accompanied by a low reticulocyte count. Physiologic anemia of infancy (in term infants) does not require medical treatment.

Age-related changes in hemoglobin that occur during the first few months of life are due to several reasons. In utero, a low arterial pO2 exists, which stimulates the production of erythropoietin in the fetus. This results in a high rate of erythropoiesis and accounts for the high levels of hemoglobin and hematocrit that exist at birth. At birth, pO2 and oxygen content of blood significantly increase with the newborn’s first breaths. The higher amount of oxygen that is available to the tissues downregulates erythropoietin production and decreases the rate of erythropoiesis.59 Without the stimulation of erythropoietin to produce new RBCs, hemoglobin concentrations decrease as aged RBCs are removed from the circulation. The shorter life span of neonatal RBCs (90 days versus 120 days in adults) also contributes to the decline in hemoglobin.

Hemoglobin continues to decline in full-term infants until 8 to 12 weeks of age, when values reach 9 to 11 g/dL. These levels of hemoglobin result in lower amounts of oxygen delivery to tissues. Usually at this point, oxygen requirements exceed oxygen delivery and the relative hypoxia stimulates the production of erythropoietin. Erythropoiesis then increases and the reticulocyte count and hemoglobin concentrations begin to rise. It is important to remember that the iron from the aged RBCs that were previously removed from the circulation has been stored. The amount of this stored iron is usually adequate to meet the requirements of hemoglobin synthesis.

In premature infants, physiologic anemia occurs at 3 to 6 weeks of age (sooner than in full-term infants) and the nadir of the hemoglobin concentrations is lower (eg, 7 to 9 g/dL).59 This can be explained by the even shorter life span of a premature infant’s RBCs (65 days versus 90 days in full-term newborns) and inadequate synthesis of erythropoietin in response to anemia.55,59 In addition, total body iron stores in premature infants are smaller and are depleted sooner. Thus, anemia of prematurity requires treatment with iron and may require blood transfusions or the use of recombinant human erythropoietin.58,59

Differences in RBC indices also exist for different pediatric ages. For example, compared with adults, newborns have larger erythrocytes (mean corpuscular volume of 108 fL compared with an adult value of 90 fL). Mean values and the lower limits of normal for RBC indices according to different ages are listed in Table 23-4.

TABLE 23-4.
Red Blood Cell Indices by Age: Mean Values and Lower Limits of Normal (−2 SD)


MCH pg/cell


Birth (cord blood)

108 (98)

34 (31)

33 (30)

1–3 days

108 (95)

34 (31)

33 (29)

1 wk

107 (88)

34 (28)

33 (28)

2 wk

105 (86)

34 (28)

33 (28)

1 mo

104 (85)

34 (28)

33 (29)

2 mo

96 (77)

30 (26)

33 (29)

3–6 mo

91 (74)

30 (25)

33 (30)

0.5–2 yr

78 (70)

27 (23)

33 (30)

2–6 yr

81 (75)

27 (24)

34 (31)

6–12 yr

86 (77)

29 (25)

34 (31)

12–18 yr


90 (78)

30 (25)

34 (31)


88 (78)

30 (25)

34 (31)

18–49 yr

90 (80)

30 (26)

34 (31)

MCV = mean corpuscular volume; MCH = mean corpuscular hemoglobin; MCHC = mean corpuscular hemoglobin concentration.

Source: Adapted from Osberg IM. Chemistry and hematology reference (normal) ranges. In: Current Pediatric Diagnosis and Treatment. 16th ed. New York, NY: Lange Medical Books/McGraw-Hill; 2003.

Causes of the various types of anemias in pediatric patients are similar to the causes in adults. Of particular note is the iron deficiency anemia that occurs in infants who are fed whole cow’s milk. The iron in whole cow’s milk is less bioavailable and may cause inadequate iron intake. Typically infants with iron deficiency anemia from whole cow’s milk have chronically consumed large amounts of cow’s milk (>24 oz per day) and foods that are not supplemented with iron. Some infants receiving whole cow’s milk develop severe iron deficiency due to chronic intestinal blood loss. The blood loss is thought to be due to intestinal exposure to a specific heat-labile protein found in whole cow’s milk. Breast feeding, delaying introduction of whole cow’s milk until 12 months of age, and decreasing the amount of whole cow’s milk to <24 oz per day have been recommended to decrease the loss of blood.60

In pediatric patients with anemia, genetic disorders that produce inadequate RBC production (eg, Diamond-Blackfan anemia), hemolytic anemias (eg, hereditary spherocytosis), or hemoglobin disorders (eg, sickle cell disease) also must be considered.


White Blood Cell Count

Normal ranges are as follows61:

neonates 1 day: 9 to 30 × 103 cells/mm3 (9 to 30 × 109/L)

neonates 2 weeks: 5 to 21 × 103 cells/mm3 (5 to 21 × 109/L)

infants 3 months: 6 to 18 × 103 cells/mm3 (6 to 18 × 109/L)

children 0.5 to 6 years: 6 to 15 × 103 cells/mm3 (6 to 15 × 109/L)

children 7 to 12 years: 4.5 to 13.5 × 103 cells/mm3 (4.5 to 13.5 × 109/L)

adults: 4.4 to 11 × 103 cells/mm3 (4.4 to 11 × 109/L)

Normal WBC counts are higher in neonates and infants compared with adults. Usually in adults with systemic bacterial infections, the WBC becomes elevated. However, in neonates and infants with a systemic bacterial infection, the WBC count is typically decreased; but it may sometimes be increased or within the normal range.58 Neonates have a lower storage pool of neutrophils, and an overwhelming infection (eg, neonatal sepsis) may deplete this pool and cause neutropenia. Therefore, although an increase in WBCs is a nonspecific finding in neonates (ie, it may occur in many conditions other than sepsis), neutropenia is a highly significant finding and may be the first abnormal laboratory result that indicates neonatal bacterial infection. Not recognizing that neutropenia in neonates indicates a serious infection could result in a delay in treatment and significant morbidity or even mortality for the patient.

In addition to the age-related differences in total WBC count, the age-related differences in WBC differential also need to be taken into consideration when interpreting laboratory results (Table 23-5). After the newborn period and up until 5 to 6 years of age, lymphocytes represent the most prevalent circulating WBC type. Subsequent to this, neutrophils predominate in the blood for the remainder of life.

TABLE 23-5.
White Blood Cell Differential by Age











2 wk





3 mo





0.5–6 yr





7–12 yr










Source: Adapted from Glader B. The anemias. In: Kliegman RM, Behrman RE, Jenson HB, et al, eds. Nelson Textbook of Pediatrics. 18th ed. Philadelphia, PA: Saunders Elsevier; 2007.


Platelet Count

Normal ranges are as follows15:

newborn ≤ 1 week: 84,000 to 478,000/mm3 (84 to 478 × 109/L)

neonate >1 week, infants, children, adolescents, and adults: 150,000 to 400,000/mm3 (150 to 400 × 109/L)

Compared with adults, the normal platelet count in the newborn may be lower. Adult values are reached after 1 week of age, although platelet counts may range higher in children (up to 600,000/mm3) than in adults. Platelet counts are discussed in detail in Chapter 17.


Interpreting pediatric laboratory data can be complex. Age-related differences in normal reference ranges occur for many common laboratory tests. These differences may be the result of changes in body composition and the normal anatomic and physiologic maturation that occurs throughout childhood. Changes in various body compartments, the immature function of the neonatal kidney, and the increased electrolyte and mineral requirements necessary for proper growth help to explain many age-related differences in serum electrolytes and minerals. Alterations in skeletal muscle mass and the pattern of kidney function maturation account for the various age-related differences in SCr and kidney function tests. Neonatal hepatic immaturity and subsequent maturation help to explain the age-related differences in serum albumin, liver enzymes, and bilirubin. Likewise, the immature hematopoietic system of the newborn and its maturation account for the age-related differences in various hematologic tests.

This chapter also reviews several general pediatric considerations, including differences in physiologic parameters, pediatric blood sampling considerations, and the determination of pediatric reference ranges. Interpretation of pediatric laboratory data must take into account the various age-related differences in normal values. If these differences are not taken into consideration, inappropriate diagnoses and treatment may result.


1. Would the dose of a medication that is primarily excreted by the kidney ever have to be adjusted in a patient with an SCr of 0.8 mg/dL?

ANSWER: The age of the patient must be taken into consideration when interpreting laboratory tests, especially SCr. In addition, the methodology used to assay SCr needs to be considered. For the Jaffe method, an SCr of 0.8 mg/dL indicates significant renal dysfunction in an infant whose normal SCr should be 0.2 to 0.4 mg/dL and mild or moderate renal dysfunction in a child whose normal SCr should be 0.3 to 0.7 mg/dL. However, an SCr of 0.8 mg/dL in an adolescent or adult would be considered within the normal range. Thus, medications that are primarily excreted by the kidney would need to have a dosage adjustment in infants and children with an SCr of 0.8 mg/dL as measured by the Jaffe method. For the IDMS-traceable enzymatic method, measured SCr values and normal values for age are lower than with the Jaffe method. An SCr of 0.8 mg/dL indicates significant renal dysfunction in an infant or young child (newborn to 4 years of age) whose normal SCr should be 0.03 to 0.5 mg/dL and mild-to-moderate renal dysfunction in a child 4 to 7 years of age whose normal SCr should be 0.03 to 0.59 mg/dL or in a child 7 to 10 years of age whose normal SCr should be 0.22 to 0.59 mg/dL. However, an SCr of 0.8 mg/dL in an adolescent or adult would be considered within the normal range.

2. Are there any concerns with using ceftriaxone in a neonate?

ANSWER: Ceftriaxone has been shown in vitro to displace bilirubin from its albumin binding sites. Thus, ceftriaxone should not be used in hyperbilirubinemic neonates, especially premature neonates, because displacement of bilirubin from albumin-binding sites may lead to bilirubin encephalopathy. These concerns are so significant that current labeled contraindications to the use of ceftriaxone include premature neonates <41 weeks’ postmenstrual age and hyperbilirubinemic neonates.62 In addition, if ceftriaxone is used in a neonate, it should be administered slowly, over 60 minutes, to reduce the risk of bilirubin encephalopathy. Ceftriaxone may also cause sludging in the gallbladder and pseudolithiasis. Fatal reactions have been reported in neonates due to ceftriaxone–calcium precipitates in the lungs and kidneys when ceftriaxone and calcium-containing IV solutions were coadministered. In some cases, the ceftriaxone and calcium-containing solutions were administered through the same IV infusion line and a precipitate was sometimes observed. In at least one fatal case, the ceftriaxone and calcium-containing solutions were administered in different infusion lines and at different times. Therefore, ceftriaxone must not be administered to neonates who also are receiving (or who are expected to receive) calcium-containing IV solutions, including continuous infusions of parenteral nutrition that contain calcium.62 This is also a labeled contraindication.

3. Would a hemoglobin of 9.5 g/dL in a 10-week-old infant who was born at full-term require initiation of iron therapy?

ANSWER: No. Iron therapy would not be required because this anemia would be considered a normal physiologic anemia of infancy. In full-term infants, hemoglobin values of 9 to 11 g/dL normally occur at 8 to 12 weeks of age. After this time, the reticulocyte count and hemoglobin concentration should begin to rise. If an infant’s hemoglobin concentration remained at 9.5 g/dL after 12 weeks of age, a further workup of the infant’s anemia would be required. If the anemia was found to be the result of iron deficiency, then iron therapy would be required. Dietary causes of iron deficiency, such as consuming large amounts of whole cow’s milk, also would need to be considered.

4. Differences in TBW, ECW, and ICW occur in neonates (compared with older pediatric patients and adults) and are described in the beginning of this chapter. Could the differences in TBW, ECW, and ICW that occur in neonates have an impact on drug distribution?

ANSWER: Yes. Both TBW and ECW, when expressed as a percentage of body weight, are increased in the newborn (especially the premature neonate). The increase in TBW and ECW help explain why water-soluble drugs have an increased volume of distribution in these patients. In fact, the volume of distribution for gentamicin, a water-soluble drug that primarily distributes to the ECW compartment, correlates well with the volume of ECW. In the neonate, ECW is about 44% of body weight and the volume of distribution for gentamicin is approximately 0.45 L/kg.1,19,20 Thus, the total amount of body water at birth, distribution between ECW and ICW, and the age-related changes that occur with time not only impact electrolyte composition and their respective normal laboratory values but also affect the distribution of medications and required doses.

5. Why would a pediatric patient presenting with diabetic ketoacidosis (DKA) have low serum sodium?

ANSWER: The three key criteria in the diagnosis of DKA are the presence of hyperglycemia (blood glucose level >200 mg/dL [>11.1 mmol/L]), acidosis (venous pH <7.3 or serum bicarbonate <15 mmol/L), and accumulation of ketoacids (ketonemia and/or ketonuria).63 In the setting of hyperglycemia, the plasma osmolarity rises and results in an osmotic shift of water from the intracellular to the extracellular (intravascular) space. This expanded extracellular volume results in measured serum sodium levels that appear low. This is referred to as dilutional hyponatremia, an excess of water relative to the amount of solute. In general, serum sodium concentrations decrease by 1.6 mmol/L for every 100 mg/dL increase in glucose concentration over normal (ie, normal value of 100 mg/dL). To account for this, sodium levels should be corrected for the degree of hyperglycemia using Equation 1. The corrected sodium value is an estimate of what the expected serum sodium would be in the absence of hyperglycemia. Because the calculated value is only an estimate, careful monitoring of serum sodium and other electrolytes is recommended. In most pediatric patients with DKA, corrected sodium levels are higher than the measured serum sodium levels. As a patient’s hyperglycemia is treated with fluid resuscitation and insulin administration, the dilutional hyponatremia reverses and a rise in the measured serum sodium and corrected sodium levels should occur. In addition to the dilutional hyponatremia, a low serum sodium in patients with DKA may be caused by an increase in renal sodium loss. The hyperglycemia-associated increase in plasma osmolarity also results in an osmotic diuresis, leading to water loss (dehydration) and electrolyte losses, including sodium, potassium, chloride, and phosphate.63,64 Failure of sodium levels to increase with correction of hyperglycemia may indicate a need to increase the sodium concentration of IV fluids.


  • 1.

    Taketomo CK, Hodding JH, Kraus DM. Pediatric and Neonatal Dosage Handbook. 26th ed. Hudson, OH: Lexi-Comp Inc; 2019.

  • 2.

    Engle WA, American Academy of Pediatrics Committee on Fetus and Newborn. Age terminology during the perinatal period. Pediatrics. 2004;114(5):13621364.PubMed

    • Search Google Scholar
    • Export Citation
  • 3.

    Kleinman K, McDaniel L, Molloy M, eds. The Harriet Lane Handbook. 22nd ed. Philadelphia, PA: Elsevier; 2021.

  • 4.

    Task Force on Blood Pressure Control in Children. Report of the second task force on blood pressure control in children—1987. Pediatrics. 1987;79(1):125.PubMed

    • Search Google Scholar
    • Export Citation
  • 5.

    National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescents. The fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents. Pediatrics. 2004;114(2)(Suppl 4th Report):555576.PubMed

    • Search Google Scholar
    • Export Citation
  • 6.

    Hicks JM. Pediatric clinical biochemistry: why is it different? In: Soldin SJ, Rifai N, Hicks JM, eds. Biochemical Basis of Pediatric Disease. 2nd ed. Washington, DC: AACC Press; 1995:117.

    • Search Google Scholar
    • Export Citation
  • 7.

    Malarkey LM, McMorrow ME. Specimen collection procedures. In: Saunders Nursing Guide to Laboratory and Diagnostic Tests. 2nd ed. Philadelphia, PA: Elsevier Saunders; 2012:1838.

    • Search Google Scholar
    • Export Citation
  • 8.

    Heelstick (Capillary Blood Sampling). In: Gomella T, Cunningham M, Eyal FG, et al, eds. Neonatology: Management, Procedures, On-Call Problems, Diseases, and Drugs. 7th ed. New York, NY: McGraw-Hill; 2013.

    • Search Google Scholar
    • Export Citation
  • 9.

    World Health Organization. WHO Guidelines on Drawing Blood: Best Practices in Phlebotomy. Geneva: World Health Organization; 2010.

  • 10.

    Nathan DG, Orkin SH, eds. Nathan and Oski’s Hematology of Infancy and Childhood. 5th ed. Philadelphia, PA: WB Saunders; 1998.

  • 11.

    Soldin SJ, Brugnara C, Wong EC, eds. Pediatric Reference Intervals. 5th ed. Washington, DC: AACC Press; 2005.

  • 12.

    Adeli K, Higgins V, Trajcevski K, White-Al Habeeb N. The Canadian laboratory initiative on pediatric reference intervals: A CALIPER white paper. Crit Rev Clin Lab Sci. 2017;54(6):358413.PubMed

    • Search Google Scholar
    • Export Citation
  • 13.

    Malarkey LM, McMorrow ME. Saunders Nursing Guide to Laboratory and Diagnostic Tests. 2nd ed. Philadelphia, PA: Elsevier Saunders; 2012.

  • 14.

    Jacobs DS, DeMott WR, Oxley DK. Laboratory Test Handbook. 5th ed. Hudson, OH: Lexi-Comp Inc; 2001.

  • 15.

    Lo SF. Reference intervals for laboratory tests and procedures. In: Kliegman RM, St Geme JW, et al, eds. Nelson Textbook of Pediatrics. 21st ed. Philadelphia, PA: Elsevier; 2020.

    • Search Google Scholar
    • Export Citation
  • 16.

    Hoppe JE, Snyder R, Accurso FJ, et al.Chemistry and hematology reference intervals. In: Hay WW, Levin MJ, Deterding RR, et al, eds. Current Diagnosis and Treatment: Pediatrics. 24th ed. New York, NY: McGraw-Hill Education; 2018.

    • Search Google Scholar
    • Export Citation
  • 17.

    Jagarinec N, Flegar-Mestrić Z, Surina B, et al.Pediatric reference intervals for 34 biochemical analytes in urban school children and adolescents. Clin Chem Lab Med. 1998;36(5):327337.PubMed

    • Search Google Scholar
    • Export Citation
  • 18.

    Bell EF, Segar JL, Oh W. Fluid and electrolyte management. In: MacDonald MG, Seshia MMK, eds. Avery’s Neonatology: Pathophysiology and Management of the Newborn. 7th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2016.

    • Search Google Scholar
    • Export Citation
  • 19.

    Friis-Hansen B. Water distribution in the foetus and newborn infant. Acta Paediatr Scand Suppl. 1983;305:711.PubMed

  • 20.

    Friis-Hansen B. Changes in body water compartments during growth. Acta Paediatr Suppl. 1957;46(suppl 110):168.PubMed

  • 21.

    Tarnow-Mordi WO, Shaw JC, Liu D, et al.Iatrogenic hyponatraemia of the newborn due to maternal fluid overload: a prospective study. Br Med J (Clin Res Ed). 1981;283(6292):639642.PubMed

    • Search Google Scholar
    • Export Citation
  • 22.

    Malarkey LM, McMorrow ME. Serum electrolytes. In: Nurses Manual of Laboratory Tests and Diagnostic Procedures. 2nd ed. Philadelphia, PA: WB Saunders; 2000:82103.

    • Search Google Scholar
    • Export Citation
  • 23.

    Greenbaum LA. Electrolyte and acid-base disorders. In: Kliegman RM, St Geme JW, et al, eds. Nelson Textbook of Pediatrics. 21st ed. Philadelphia, PA: Elsevier; 2020.

    • Search Google Scholar
    • Export Citation
  • 24.

    Blackburn ST. Renal function in the neonate. J Perinat Neonatal Nurs. 1994;8(1):3747.PubMed

  • 25.

    Nafday SM, Woda CB, Saland JM, et al.Renal disease. In: MacDonald MG, Seshia MMK, eds. Avery’s Neonatology: Pathophysiology and Management of the Newborn. 7th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2016.

    • Search Google Scholar
    • Export Citation
  • 26.

    Breault DT, Majzoub JA. Other abnormalities of arginine vasopressin metabolism and action. In: Kliegman RM, St Geme JW, et al, eds. Nelson Textbook of Pediatrics. 21st ed. Philadelphia, PA: Elsevier; 2020.

    • Search Google Scholar
    • Export Citation
  • 27.

    Lorenz JM. Assessing fluid and electrolyte status in the newborn. Clin Chem. 1997;43(1):205210.PubMed

  • 28.

    Papageorgiou A, Pelausa E, Kovacs L. The extremely low birth weight infant. In: MacDonald MG, Seshia MMK, eds. Avery’s Neonatology: Pathophysiology and Management of the Newborn. 7th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2016.

    • Search Google Scholar
    • Export Citation
  • 29.

    Jones DP, Chesney RW. Development of tubular function. Clin Perinatol. 1992;19(1):3357.PubMed

  • 30.

    Greenbaum LA. Electrolytes and acid-base disorders. In: Behrman RE, Kliegman RM, Jenson HB, eds. Nelson Textbook of Pediatrics. 17th ed. Philadelphia, PA: WB Saunders; 2004.

    • Search Google Scholar
    • Export Citation
  • 31.

    Stigson L, Kjellmer I. Serum levels of magnesium at birth related to complications of immaturity. Acta Paediatr. 1997;86(9):991994.PubMed

    • Search Google Scholar
    • Export Citation
  • 32.

    Ariceta G, Rodríguez-Soriano J, Vallo A. Magnesium homeostasis in premature and full-term neonates. Pediatr Nephrol. 1995;9(4):423427.PubMed

    • Search Google Scholar
    • Export Citation
  • 33.

    Geven WB, Monnens LA, Willems JL. Magnesium metabolism in childhood. Miner Electrolyte Metab. 1993;19(4–5):308313.PubMed

  • 34.

    Pesce MA. Reference ranges for laboratory tests and procedures. In: Kliegman RM, Behrman RE, et al, eds. Nelson Textbook of Pediatrics. 18th ed. Philadelphia, PA: Saunders Elsevier; 2007.

    • Search Google Scholar
    • Export Citation
  • 35.

    Nehus EJ. Introduction to glomerular diseases. In: Kliegman RM, St Geme JW, et al, eds. Nelson Textbook of Pediatrics. 21st ed. Philadelphia, PA: Elsevier; 2020.

    • Search Google Scholar
    • Export Citation
  • 36.

    Schwartz GJ, Brion LP, Spitzer A. The use of plasma creatinine concentration for estimating glomerular filtration rate in infants, children, and adolescents. Pediatr Clin North Am. 1987;34(3):571590.PubMed

    • Search Google Scholar
    • Export Citation
  • 37.

    Bueva A, Guignard JP. Renal function in preterm neonates. Pediatr Res. 1994;36(5):572577.PubMed

  • 38.

    Myers GL, Miller WG, Coresh J, et al.Recommendations for improving serum creatinine measurement: a report from the Laboratory Working Group of the National Kidney Disease Education Program. Clin Chem. 2006;52(1):518.PubMed

    • Search Google Scholar
    • Export Citation
  • 39.

    Alcorn J, McNamara PJ. Ontogeny of hepatic and renal systemic clearance pathways in infants: part I. Clin Pharmacokinet. 2002;41(12):959998.PubMed

    • Search Google Scholar
    • Export Citation
  • 40.

    Botwinski CA, Falco GA. Transition to postnatal renal function. J Perinat Neonatal Nurs. 2014;28(2):150154. PubMed

  • 41.

    DuBois D, Dubois EF. A formula to estimate the approximate surface area if height and weight be known. Arch Intern Med (Chic). 1916;17:863871.

    • Search Google Scholar
    • Export Citation
  • 42.

    Mosteller RD. Simplified calculation of body-surface area. N Engl J Med. 1987;317(17):1098.PubMed

  • 43.

    Hogg RJ, Furth S, Lemley KV, et al.National Kidney Foundation’s Kidney Disease Outcomes Quality Initiative clinical practice guidelines for chronic kidney disease in children and adolescents: evaluation, classification, and stratification. Pediatrics. 2003;111(6 Pt 1):14161421.PubMed

    • Search Google Scholar
    • Export Citation
  • 44.

    Traub SL, Johnson CE. Comparison of methods of estimating creatinine clearance in children. Am J Hosp Pharm. 1980;37(2):195201.PubMed

    • Search Google Scholar
    • Export Citation
  • 45.

    Hernandez de Acevedo L, Johnson CE. Estimation of creatinine clearance in children: comparison of six methods. Clin Pharm. 1982;1(2):158161.PubMed

    • Search Google Scholar
    • Export Citation
  • 46.

    Schwartz GJ, Work DF. Measurement and estimation of GFR in children and adolescents. Clin J Am Soc Nephrol. 2009;4(11):18321843.PubMed

    • Search Google Scholar
    • Export Citation
  • 47.

    Mian AN, Schwartz GJ. Measurement and estimation of glomerular filtration rate in children. Adv Chronic Kidney Dis. 2017;24(6):348356.PubMed

    • Search Google Scholar
    • Export Citation
  • 48.

    Schwartz GJ, Muñoz A, Schneider MF, et al.New equations to estimate GFR in children with CKD. J Am Soc Nephrol. 2009;20(3):629637.PubMed

    • Search Google Scholar
    • Export Citation
  • 49.

    Schwartz GJ, Schneider MF, Maier PS, et al.Improved equations estimating GFR in children with chronic kidney disease using an immunonephelometric determination of cystatin C. Kidney Int. 2012;82(4):445453.PubMed

    • Search Google Scholar
    • Export Citation
  • 50.

    CKD Work Group, KDIGO. Kidney Disease: Improving Global Outcomes (KDIGO) 2012 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int Suppl. 2013;3:1150.

    • Search Google Scholar
    • Export Citation
  • 51.

    Waz WR, Quattrin T, Feld LG. Serum creatinine, height, and weight do not predict glomerular filtration rate in children with IDDM. Diabetes Care. 1993;16(8):10671070.PubMed

    • Search Google Scholar
    • Export Citation
  • 52.

    Huppert SS, Balistreri WF. Morphogenesis of the liver and biliary system. In: Kliegman RM, St Geme JW, et al, eds. Nelson Textbook of Pediatrics. 21st ed. Philadelphia, PA: Elsevier; 2020.

    • Search Google Scholar
    • Export Citation
  • 53.

    Hughes HK, Kahl LK, eds. The Harriet Lane Handbook. 21st ed. Philadelphia, PA: Elsevier; 2018.

  • 54.

    Pan DH, Rivas Y. Jaundice: Newborn to age 2 months. Pediatr Rev. 2017;38(11):499510.PubMed

  • 55.

    Lockitch G. Beyond the umbilical cord: interpreting laboratory tests in the neonate. Clin Biochem. 1994;27(1):16.PubMed

  • 56.

    Gurria JP, Brown RL. Digestive system disorders. In: Kliegman RM, St Geme JW, et al, eds. Nelson Textbook of Pediatrics. 21st ed. Philadelphia, PA: Elsevier; 2020.

    • Search Google Scholar
    • Export Citation
  • 57.

    Osberg IM. Chemistry and hematology reference (normal) ranges. In: Current Pediatric Diagnosis and Treatment. 16th ed. New York, NY: Lange Medical Books/McGraw-Hill; 2003.

    • Search Google Scholar
    • Export Citation
  • 58.

    Dror Y, Chan A, Backer JM, et al.Hematology. In: MacDonald MG, Seshia MMK, eds. Avery’s Neonatology: Pathophysiology and Management of the Newborn. 7th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2016.

    • Search Google Scholar
    • Export Citation
  • 59.

    Thornburg CD. Physiologic anemia of infancy. In: Kliegman RM, St Geme JW, et al, eds. Nelson Textbook of Pediatrics. 21st ed. Philadelphia, PA: Elsevier; 2020.

    • Search Google Scholar
    • Export Citation
  • 60.

    Rothman JA. Iron-deficiency anemia. In: Kliegman RM, St Geme JW, et al, eds. Nelson textbook of Pediatrics. 21st ed. Philadelphia, PA: Elsevier; 2020.

    • Search Google Scholar
    • Export Citation
  • 61.

    Glader B. The anemias. In: Kliegman RM, Behrman RE, Jenson HB, et al, eds. Nelson Textbook of Pediatrics. 18th ed. Philadelphia, PA: Saunders Elsevier; 2007.

    • Search Google Scholar
    • Export Citation
  • 62.

    Rocephin [package insert]. South San Francisco, CA: Genentech USA Inc; 2018.

  • 63.

    Wolfsdorf J, Craig ME, Daneman D, et al.Diabetic ketoacidosis in children and adolescents with diabetes. Pediatr Diabetes. 2009;10(suppl 12):118133.PubMed

    • Search Google Scholar
    • Export Citation
  • 64.

    Castellanos L, Tuffaha M, Koren D et al. Management of diabetic ketoacidosis in children and adolescents with type 1 diabetes mellitus. Pediatric Drugs Published ahead of print May 25, 2020 10.1007/s40272-020-00397-0.

    • Search Google Scholar
    • Export Citation