- Email: [email protected]
Neonatal pharmacology
Neonatal
Neonatal pharmacology
Some factors affecting drug absorption relating to neonates
Adam Skinner
Physicochemical factors Drug formulation • Disintegration of tablets or solid phase • Dissolution of drug in gastric or intestinal fluid • Release from sustained-release preparations Molecular weight pK/Proportion of drug in ionized/un-ionized form Lipid solubility
Abstract
Patient factors General • Surface area available for absorption
The absorption, distribution, metabolism and excretion of drugs in a neonate differ markedly from those of older children. The neonate under goes rapid developmental change from birth, which alters the pharmaco kinetic profile of many drugs. The pathology seen in neonates can also have marked effects on drug handling; low cardiac output states, raised intra-abdominal pressure, hepatic and renal disease have signifi cant implications for drug pharmacokinetics. There is also considerable interpatient variability concerning all aspects of drug disposition. These factors make accurate drug administration difficult, and leave the new born particularly susceptible to adverse drug reactions, morbidity and mortality. By applying neonatal pharmacokinetic principles to individual drugs, we identify important differences between neonates and older children. Adverse effects of commonly used drugs in neonates may be reduced by applying neonatal pharmacokinetic principles. Anaesthetic agents, sedatives, analgesics and local anaesthetics are discussed in this article in relation to their particular behaviour in neonates.
Gastrointestinal • Gastric content and gastric emptying • Gastric and duodenal pH • Size of bile-salt pool • Bacterial colonization of lower intestine • Disease states (e.g. short-gut syndrome, biliary atresia) Muscle • Increased capillary density in neonatal muscle compared with adults increases absorption from muscles • Reduced cardiac output states reduce absorption Skin • Blood supply • Peripheral vasodilation • Thickness of skin/stratum corneum • Surface area
Keywords neonate; paediatric; pharmacokinetics; pharmacology
Rectal • Rectal venous drainage site • Neonatal absorption > older children
Pharmacokinetics Absorption Physicochemical and patient factors influence the ability of a drug to translocate from its site of administration to the bloodstream and site of action (Table 1).
Table 1
pool size is reduced in the neonate compared with adults. Diseases such as extrahepatic biliary obstruction worsen biliary function. Rate of drug absorption – the irregular and unpredictable peri staltic activity of the upper gastrointestinal tract contributes to a variable rate of drug absorption in the neonate.2 If gastric emptying is slowed, the drug is delayed in reaching the small intestine from where it is absorbed. The peak serum drug concentration will also be reduced.1 Slow gastric emptying is associated with: low gestational and postnatal age; type of feed (increased cal orie density and long-chain fatty acids); and disease states (e.g. pyloric stenosis, congestive cardiac failure). Human milk and low-calorie feed quicken gastric emptying. Phosphorylated glycoprotein (PGp) is a widely distributed protein in humans. It is responsible for ‘pumping’ drugs out of cells, and influences intestinal drug absorption, hepatic and renal clearance and the amount of drug that enters the CNS. The variability in PGp expression and other drugs affecting its activity may be responsible for the variability of absorption of many drugs in neonates.1,2
Enteral absorption Extent of drug absorption – neonates are able to produce gastric acid. At birth, the gastric pH is between 6 and 8; however, this falls rapidly within a few hours. Premature neonates born at 25 weeks postmenstrual age produce H-K-ATPase, and the expression increases with gestational age.1 The volume and pH of gastric secretion after birth is variable; gastric and duodenal content influences the ability of a drug to dissolve and alters the ratio of ionized to un-ionized particles. A low pH environment will render acidic drugs with a low pK more un-ionized and more able to cross lipid membranes. Changes in the bile-salt pool can alter the solubilization and absorption of lipophilic drugs or formulations. Bile-acid
Adam Skinner, MRCP, FRCA, is Consultant Paediatric Anaesthetist at the Royal Children’s Hospital, Melbourne, Australia. He has an interest in medical education and paediatric cardiac anaesthesia.
ANAESTHESIA AND INTENSIVE CARE MEDICINE 9:3
99
© 2008 Elsevier Ltd. All rights reserved.
Neonatal
Non-enteral absorption: if the enteral route is not appropriate, drugs can be administered intravenously or via the buccal, sublingual, intramuscular, subcutaneous, rectal or transcutaneous route (Table 1). Transdermal absorption in neonates is variable because of an incompletely formed stratum corneum and immature vasomotor control. The increased surface area-to-weight ratio must be taken into account and can be responsible for a higher drug exposure than intended compared with a similar dose per kg in an adult.1 Rectal administration is associated with variable plasma concentrations; factors such as variable lower gastrointestinal motility, and depth of insertion may affect bioavailability. Absorption via the upper rectal veins undergoes first-pass metabolism, whereas the inferior and middle rectal veins bypass the hepatic first-pass effect and drain directly into the inferior vena cava.
non-depolarizing neuromuscular blockers will have a higher extracellular volume of distribution, which may suggest that a higher dose per kg is required. However, the lower muscle mass and immature neuromuscular receptor function suggest that blockade is achieved at lower plasma concentrations. Protein binding: neonates have reduced albumin and total rotein concentrations. The quality of drug-protein binding also p appears to be reduced. Lower protein binding results in greater free-drug concentration and hence greater drug effect. This effect is clinically more pronounced with highly protein-bound drugs.3 Important examples of acidic drugs binding albumin include diazepam, thiopental and phenytoin. Basic drugs tend to bind α1-acid glycoprotein, which will also have reduced binding in the neonate (e.g. lidocaine and alfentanil). Neonatal jaundice is common in the premature neonate; bilirubin competes with some drugs (e.g. phenytoin) for protein binding, which can result in either further increased freedrug concentration or increased free bilirubin. The latter would increase the risk of kernicterus.4
Distribution Many factors are responsible for altered drug distribution as the neonate develops. Fluid distribution: the greatest change in body water compartments occurs in the first year of life (Figure 1). In neonates the volume of distribution for water-soluble drugs is increased. To achieve the required plasma and tissue concentrations of watersoluble drugs a higher dose per kg needs to be administered in the neonate compared with the adult.1,2
Blood–brain barrier: morphine is much less lipid soluble than other opioid analgesics, and is slower to cross the mature adult blood–brain barrier. Neonates have an immature barrier, which results in faster morphine uptake into the CNS. This is partly responsible for the considerable neonatal sensitivity to the CNS effects of morphine. Disease states such as sepsis, hypoxia and acidosis further reduce the integrity of the blood– brain barrier.
Body tissue composition: neonates have a smaller proportion of weight in the form of fat and muscle compared with adults (Figure 2).3 In adults thiopentone or fentanyl have an early peak plasma concentration followed by a rapid redistribution to muscle (and fat). In the neonate with less muscle mass, this redistribution capacity will be reduced, causing prolonged higher plasma concentrations. Opposing pharmacokinetic factors frequently need to be balanced when evaluating drug dosage in neonates. For example,
Metabolism The ability of a neonate to metabolize drugs is mainly dependent on enzyme maturation and hepatic blood flow. These factors have variable expression and development. Pathological conditions such as low cardiac output states, raised intra-abdominal pressure or liver disease can also alter the disposition of drugs.
Age-related body water compartments Preterm neonate
Term neonate
Infant 1 year
Adult man
TBW 85%
TBW 80%
TBW 60%
TBW 60%
ECF 60%
ECF 45%
ECF 25%
ECF 20%
ICF 25%
ICF 35%
ICF 35%
ICF 40%
ECF, extracellular fluid; ICF, intracellular fluid; TBW, total body water
Figure 1
ANAESTHESIA AND INTENSIVE CARE MEDICINE 9:3
100
© 2008 Elsevier Ltd. All rights reserved.
Neonatal
Enzyme systems: drug biotransformation to more polar forms occurs via phase I (oxidation, reduction and hydrolysis) and phase II conjugation reactions (e.g. glucuronidation). Many phase I reactions involve isoenzymes found in the CYP1, CYP2 and CYP3 gene families of the cytochrome P450 system.3 Premature neonates are able to metabolize drugs; however, different enzymes have variable development. Genetic polymorphism of phase I and phase II enzymes add to this variability (e.g. CYP2D6 and codeine)3,4 (Table 2).
Proportion of body composition (%)
Proportion of fat and muscle from premature neonate to adult 50
Muscle Fat
40
30
Renal excretion At birth, glomerular and tubular functional immaturity can prolong the elimination half-life of many drugs. Renal clearance rate of a drug is dependent on: glomerular filtration rate (GFR), fraction of drug unbound to protein (fu), and the balance of tubular reabsorption clearance (CLRA) and tubular secretion clearance (CLS).5
20
10
0
Preterm
Term
Adult
CL R = fu × GFR + CL S − CL RA
Figure 2
GFR is gestational age related and is reduced with increasing prematurity. At 41 weeks postconceptual age the GFR is 1.5 ml/ kg/min (20–40 ml/min/1.73m2), which increases to half of adult levels by 3 months. Adult levels of GFR 2.0 ml/kg/min (120 ml/ min/1.73m2) are achieved by 2 years of age. Reasons for this increase are: • disappearance of the placental shunt • increased renal blood flow • altered distribution of renal blood flow • increased permeability of the glomerular membrane. With low renal blood flow at birth, the fine balance between vasoconstrictor and vasodilatory renal forces are vital to maintain filtration pressure. This function in neonates is particularly vulnerable when non-steroidal anti-inflammatory dugs (NSAIDs) are administered.
Hepatic blood flow: with neonatal maturation, a greater proportion of the cardiac output is delivered to the liver. This is particularly important for the metabolism of those drugs with a high hepatic intrinsic clearance and extraction ratio.1 This can be explained by the equation: Q × f × CL int CL H= Q + (f × CL int )
where Q is the hepatic blood flow, f is the fraction of free drug, Clint is the intrinsic clearance and CLH is the hepatic clearance. If intrinsic clearance of a drug is high (e.g. propranolol), changes in liver blood flow would have much more impact on overall hepatic clearance compared with a change in the hepatocellular metabolism. Conversely, if a drug has a small intrinsic clearance (or extraction ratio), the hepatic clearance is more dependent on either the fraction of free drug (e.g. clindamicin) or intrinsic clearance (e.g. chloramphenicol). We can therefore see here how protein binding will affect the clearance of the ‘binding-sensitive drugs’ such as clindamicin, but not the hepatic clearance of chloramphenicol.1
Individual drug groups Analgesics The belief that neonates are unable to experience pain has been rejected, and the importance of neonatal analgesia has been increasingly acknowledged during the past 20 years (page 98).6
Examples of cytochrome P450 enzymes involved in phase I metabolism and how they develop with postnatal age Enzyme
Drug substrate
Enzyme development
CYPA12
Paracetamol
CYP2C9
Phenytoin
CYP2D6
β-blockers, ondansetron, codeine, tricyclic antidepressants, neuroleptics, paroxetine, fluoxetine, haloperidol Midazolam
Absent in fetus but reaches adult levels at about 4 months. The low level of cytochrome P450 activity in the liver may be responsible for the apparent reduced formation of hepatotoxic metabolites in neonates Phenytoin half-life is prolonged at 75 hours in preterm neonates; this decreases to 20 hours at term and 8 hours at 2 weeks Usually present at 1 week but only 20% of adult activity at 1 month. Variable because of genetic polymorphism: up to 47% of 3–12-yearolds cannot convert codeine to morphine Activity low at birth but increases 5-fold over 3 months of life
CYP3A4 Table 2
ANAESTHESIA AND INTENSIVE CARE MEDICINE 9:3
101
© 2008 Elsevier Ltd. All rights reserved.
Neonatal
Opioid analgesics Morphine is frequently used during surgery for moderateto-severe perioperative pain relief and in neonatal intensive care for the stress of ventilation or painful conditions (e.g. necrotizing enterocolitis). The neonatal increased sensitivity to morphine is due to a combination of an immature blood–brain barrier and reduced hepatic and renal clearance, which leads to higher effector-site concentrations.4 Clearance improves rapidly with postconceptional age; the preterm neonatal clearance (approximately 3 ml/kg/min) can increase threefold by term and tenfold by 2–4 months. Adult values are reached by 6–12 months. Disease states can influence clearance; infants who have had cardiac surgery show a markedly reduced clearance compared with children who have undergone noncardiac surgery.3 Those with renal impairment have the potential to accumulate the metabolites morphine-6-glucuronide (potent μ opioid receptor activity) and morphine-3-glucuronide (antanalgesic effect).4 As with all opioid infusions, neonates should have continuous monitoring and frequent assessment with a high nurse:patient ratio. Oral morphine absorption is variable, with a high first-pass metabolism. Fentanyl – neonatal liver blood flow, bypassing flow through a patent ductus venosus, and immature enzyme systems are responsible for reduced clearance to 70–80% of mature values. Pathology, such as raised intra-abdominal pressure (IAP), that reduces hepatic flow, in turn reduces clearance. Neonates with raised IAP may require less fentanyl than those with normal IAP.7 The high lipid solubility renders the maturation of the blood–brain barrier less relevant in this situation compared with morphine (see above).
erivatives, which have an important role in nephrogenesis in d the fetus and in maintaining both glomerular and tubular function in the neonate.5 Vasodilator prostaglandins counteract the high renal vascular resistance in the first days of life by acting on prostaglandin-E receptor subtypes in the afferent glomerular vessels. In the presence of angiotensin-II-induced efferent vasoconstriction, the glomerular filtration pressure is generated to optimize GFR in the presence of poor glomerular perfusion if the first few days of life.5 It has been demonstrated that NSAID exposure can lead to a reduction in GFR, with reduced clearance of antibiotics such as gentamicin. Fatal renal failure has been reported following antenatal indomethacin exposure. NSAIDs should not be given to the mother during the renal developmental phase in utero.5 Bleeding times often remain within normal limits after NSAIDs. Intestinal perforation has been reported after treatment with indomethacin. In the future‚ stereoselective isomers of NSAIDs may have a role in neonates due to their improved safety profile (e.g. ibuprofen; r-ibuprofen is much more rapidly cleared than s-ibuprofen).9 Inhalational anaesthetics The minimum alveolar concentration (MAC) of halothane, isoflurane and desflurane is low in premature neonates, peaks at 1–6 months and then falls gradually as age increases (Figure 3).10 Interestingly, sevoflurane differs, in that the MAC in neonates (3.3%) does not change significantly through to 6 months (3.2%).11 The reason for this different pattern is unclear. The haemodynamic response to an age-adjusted MAC of volatile agents is no different from older children. Intravenous anaesthetics In the newborn, CNS sensitivity to sedatives is related to increased permeability of the developing blood–brain barrier,
Non-opioid analgesics Paracetamol – rectal absorption of paracetamol is slow and variable. Factors such as lower gastrointestinal motility, drug formulation (triglyceride or capsule) or even the height of the rectal venous plexus into which it is absorbed affects the variation of drug absorption. There are few data on rectal absorption for premature neonates. For oral administration, the rapid absorption in children aged more than 6 months is not seen in neonates because of erratic and slow gastric emptying.8 The introduction of intravenous paracetamol preparations may improve the predictability of bioavailability and increase speed of onset; however, it is currently not licensed for use in neonates. The newer intravenous paracetamol preparation appears to be safer than pro-paracetamol with fewer side effects.9 The clearance of intravenous paracetamol is reduced in neonates; however, it increases with postconceptional age. NSAIDs – indomethacin is used as a tocolytic and readily crosses the placenta. Postnatally, it is used in the closure of a patent ductus arteriosus.5,9 Although there are no linked pharmacodynamic or pharmacokinetic studies investigating analgesia in neonates, it has been used for postoperative pain (e.g. bladder extrophy repair).8 There is variable expression of the CYP2C family, and genetic polymorphism has been described. Clearance of NSAIDs increases with postconceptional and postnatal age.9 NSAIDs inhibit the two isoforms of cyclooxygenase (COX-1 and COX-2), which in turn have a key role in neonatal renal physiology. They are responsible for the synthesis of prostanoid
ANAESTHESIA AND INTENSIVE CARE MEDICINE 9:3
The minimum alveolar concentration (MAC) of isoflurane with increasing age 2.0
MAC (% isoflurane)
1–6 months 1.8
1.6
Neonates 32–37 weeks’ gestation
1.4
< 32 weeks’ gestation 1.2
1.0 0.5
1.0
5
10
50
100
Postconceptual age (years) This pattern of the MAC throughout childhood is similar for desflurane and halothane Reproduced with permission from LeDez KM, Lerman J. The minimum alveolar concentration of isoflurane in preterm neonates. Anesthesiology 1987; 67: 301–7
Figure 3
102
© 2008 Elsevier Ltd. All rights reserved.
Neonatal
changes in the body composition and decreased elimination of drugs. Enzyme immaturity contributes to a reduced hepatic clearance of thiopental in the neonate, with an elimination halflife of 17.9 hours compared with 6.1 hours in the child and 12 hours in the adult.3 Propofol 1% is not licensed for use as an induction agent in neonates. Animal studies have suggested that ketamine isoflurane, nitrous oxide, and or midazolam exposure to the developing brain causes neuronal apoptosis; however, the question of implication for the human neonatal brain development remains unanswered.
frequently used in children to reduce pain of venepuncture. In the neonate it has been used for procedures such as heel-pricks and spinal injections. The risk of systemic absorption causing methaemaglobinaemia can be reduced by applying the agent to a limited surface area of skin. There are not enough data to support multiple daily applications. ◆
References 1 Blumer JL, Reed MD. Principles of neonatal pharmacology. In: Yaffe SJ, Aranda JV, eds. Neonatal and pediatric pharmacology: therapeutic principles in practice, 3rd edn. Philadelphia: Lippincott, Williams & Wilkins, 2005. 2 van den Anker J. Problems with medicines in the newborn infant. In: Jacqz-Aigrain E, Choonara I, eds. Paediatric clinical pharmacology. New York: Taylor and Francis Group, 2006. 3 Cote CJ, Lugo RA, Ward RM. Pharmacokinetics and pharmacology of drugs in children. In: Cote CJ, Todres ID, Goudsouzian NG, Ryan JF, eds, 3rd edn. Philadelphia: Saunders, 2001. 4 Palmer GM, Anderson BJ. Opioid analgesic drugs. In: Jacqz-Aigrain E, Choonara I, eds. Paediatric clinical pharmacology. New York: Taylor and Francis Group, 2006. 5 van den Anker J. Renal function and excretion of drugs in the newborn. In: Yaffe SJ, Aranda JV, eds. Neonatal and pediatric pharmacology: therapeutic principles in practice, 3rd edn. Philadelphia: Lippincott, Williams & Wilkins, 2005. 6 Simons SHP, Anderson BJ, Tibboel D. Analgesic agents. In: Yaffe SJ, Aranda JV, eds. Neonatal and paediatric pharmacology: therapeutic principles in practice, 3rd edn. Philadelphia: Lippincott, Williams & Wilkins, 2005. 7 Koehntop DE, Rodman JH, Brundage DM, et al. Pharmacology of fentanyl in neonates. Anesth Analg 1986; 65: 227–32. 8 Anderson BJ. Paracetamol. In: Jacqz-Aigrain E, Choonara I, eds. Paediatric clinical pharmacology. New York: Taylor and Francis Group, 2006. 9 Anderson BJ, Palmer GM. Recent developments in the pharmacological management of pain in children. Curr Opin Anaesthesiol 2006; 19: 285–92. 10 Taylor RH, Lerman J. Minimum alveolar concentration of desflurane and hemodynamic responses in neonates, infants and children. Anesthesiology 1991; 75: 975–9. 11 Lerman J, Sikich N, Kleinman S, Yentis S. The pharmacology of sevoflurane in infants and children. Anesthesiology 1994; 80: 814–24. 12 Chalkiadis GA, Anderson BJ. Age and size are the major covariates for the prediction of levobupivacaine clearance in children. Paediatr Anaesth 2006; 16: 275–82. 13 Larsson BA, Lonnqvist PA, Olsson GL. Plasma concentrations of bupivacaine in neonates after continuous epidural infusion. Anesth Analg 1997; 84: 501–5.
Benzodiazepines Diazepam: reduced hepatic blood flow and immature hepatic excretory mechanisms can prolong the elimination half-life of diazepam to up to 100 hours in the neonate (18 hours in young adults). The active metabolite N-desmethyldiazepam is similar in potency and half-life to diazepam.3 Many would consider it unsuitable for use in neonates. The intravenous solutions that contain the preservative benzyl alcohol should be avoided in neonates because of the risk of metabolic acidosis and kernicterus. Midazolam: unlike diazepam, the active metabolite of mida zolam has minimal activity; it is therefore more suited for use in neonates. The clearance is prolonged (6–12 hours) compared with older children (1.4–4.0 hours).3 This can be reduced by factors that reduce hepatic blood flow or hypovolaemic states. Local anaesthetics Regional techniques are used to reduce the use of systemic an algesics and may facilitate earlier extubation. The toxic plasma concentration of bupivacaine is unknown in children. However, many infusion regimens aim for a plasma concentration of less than 1 μg/ml. Pharmacokinetic data of epidural bupivacaine suggest that neonates will develop higher plasma concentrations of bupi vacaine compared with older children. The immature cytochrome P450 enzyme systems are the main reason for reduced clearance of the drug, although reduced hepatic blood flow may also contribute. Pharmacokinetic studies of levobupivacaine show that neonatal clearance is approximately 25% of adult values, which increases to 80% at 6 months of age. Most of the clearance can be predicted by postnatal age and weight.12 The reduced protein binding may be less important than drug clearance in determining toxic levels. Doses per kg should be reduced in neonates compared with older children. With a reduced 1.8 mg/kg epidural bolus of bupivacaine followed by a 0.2 mg/kg/hour infusion, some neonates still have rising plasma concentrations after 48 hours, which raises concern about prolonged infusions in this age group.13 Newer local anaesthetic agents such as levobupivacaine and ropivacaine may have a safer profile. Topical local anaesthetics such as EMLA (2.5% prilocaine, 2.5% lidocaine) are
ANAESTHESIA AND INTENSIVE CARE MEDICINE 9:3
103
© 2008 Elsevier Ltd. All rights reserved.
Neonatal pharmacology
Some factors affecting drug absorption relating to neonates
Adam Skinner
Physicochemical factors Drug formulation • Disintegration of tablets or solid phase • Dissolution of drug in gastric or intestinal fluid • Release from sustained-release preparations Molecular weight pK/Proportion of drug in ionized/un-ionized form Lipid solubility
Abstract
Patient factors General • Surface area available for absorption
The absorption, distribution, metabolism and excretion of drugs in a neonate differ markedly from those of older children. The neonate under goes rapid developmental change from birth, which alters the pharmaco kinetic profile of many drugs. The pathology seen in neonates can also have marked effects on drug handling; low cardiac output states, raised intra-abdominal pressure, hepatic and renal disease have signifi cant implications for drug pharmacokinetics. There is also considerable interpatient variability concerning all aspects of drug disposition. These factors make accurate drug administration difficult, and leave the new born particularly susceptible to adverse drug reactions, morbidity and mortality. By applying neonatal pharmacokinetic principles to individual drugs, we identify important differences between neonates and older children. Adverse effects of commonly used drugs in neonates may be reduced by applying neonatal pharmacokinetic principles. Anaesthetic agents, sedatives, analgesics and local anaesthetics are discussed in this article in relation to their particular behaviour in neonates.
Gastrointestinal • Gastric content and gastric emptying • Gastric and duodenal pH • Size of bile-salt pool • Bacterial colonization of lower intestine • Disease states (e.g. short-gut syndrome, biliary atresia) Muscle • Increased capillary density in neonatal muscle compared with adults increases absorption from muscles • Reduced cardiac output states reduce absorption Skin • Blood supply • Peripheral vasodilation • Thickness of skin/stratum corneum • Surface area
Keywords neonate; paediatric; pharmacokinetics; pharmacology
Rectal • Rectal venous drainage site • Neonatal absorption > older children
Pharmacokinetics Absorption Physicochemical and patient factors influence the ability of a drug to translocate from its site of administration to the bloodstream and site of action (Table 1).
Table 1
pool size is reduced in the neonate compared with adults. Diseases such as extrahepatic biliary obstruction worsen biliary function. Rate of drug absorption – the irregular and unpredictable peri staltic activity of the upper gastrointestinal tract contributes to a variable rate of drug absorption in the neonate.2 If gastric emptying is slowed, the drug is delayed in reaching the small intestine from where it is absorbed. The peak serum drug concentration will also be reduced.1 Slow gastric emptying is associated with: low gestational and postnatal age; type of feed (increased cal orie density and long-chain fatty acids); and disease states (e.g. pyloric stenosis, congestive cardiac failure). Human milk and low-calorie feed quicken gastric emptying. Phosphorylated glycoprotein (PGp) is a widely distributed protein in humans. It is responsible for ‘pumping’ drugs out of cells, and influences intestinal drug absorption, hepatic and renal clearance and the amount of drug that enters the CNS. The variability in PGp expression and other drugs affecting its activity may be responsible for the variability of absorption of many drugs in neonates.1,2
Enteral absorption Extent of drug absorption – neonates are able to produce gastric acid. At birth, the gastric pH is between 6 and 8; however, this falls rapidly within a few hours. Premature neonates born at 25 weeks postmenstrual age produce H-K-ATPase, and the expression increases with gestational age.1 The volume and pH of gastric secretion after birth is variable; gastric and duodenal content influences the ability of a drug to dissolve and alters the ratio of ionized to un-ionized particles. A low pH environment will render acidic drugs with a low pK more un-ionized and more able to cross lipid membranes. Changes in the bile-salt pool can alter the solubilization and absorption of lipophilic drugs or formulations. Bile-acid
Adam Skinner, MRCP, FRCA, is Consultant Paediatric Anaesthetist at the Royal Children’s Hospital, Melbourne, Australia. He has an interest in medical education and paediatric cardiac anaesthesia.
ANAESTHESIA AND INTENSIVE CARE MEDICINE 9:3
99
© 2008 Elsevier Ltd. All rights reserved.
Neonatal
Non-enteral absorption: if the enteral route is not appropriate, drugs can be administered intravenously or via the buccal, sublingual, intramuscular, subcutaneous, rectal or transcutaneous route (Table 1). Transdermal absorption in neonates is variable because of an incompletely formed stratum corneum and immature vasomotor control. The increased surface area-to-weight ratio must be taken into account and can be responsible for a higher drug exposure than intended compared with a similar dose per kg in an adult.1 Rectal administration is associated with variable plasma concentrations; factors such as variable lower gastrointestinal motility, and depth of insertion may affect bioavailability. Absorption via the upper rectal veins undergoes first-pass metabolism, whereas the inferior and middle rectal veins bypass the hepatic first-pass effect and drain directly into the inferior vena cava.
non-depolarizing neuromuscular blockers will have a higher extracellular volume of distribution, which may suggest that a higher dose per kg is required. However, the lower muscle mass and immature neuromuscular receptor function suggest that blockade is achieved at lower plasma concentrations. Protein binding: neonates have reduced albumin and total rotein concentrations. The quality of drug-protein binding also p appears to be reduced. Lower protein binding results in greater free-drug concentration and hence greater drug effect. This effect is clinically more pronounced with highly protein-bound drugs.3 Important examples of acidic drugs binding albumin include diazepam, thiopental and phenytoin. Basic drugs tend to bind α1-acid glycoprotein, which will also have reduced binding in the neonate (e.g. lidocaine and alfentanil). Neonatal jaundice is common in the premature neonate; bilirubin competes with some drugs (e.g. phenytoin) for protein binding, which can result in either further increased freedrug concentration or increased free bilirubin. The latter would increase the risk of kernicterus.4
Distribution Many factors are responsible for altered drug distribution as the neonate develops. Fluid distribution: the greatest change in body water compartments occurs in the first year of life (Figure 1). In neonates the volume of distribution for water-soluble drugs is increased. To achieve the required plasma and tissue concentrations of watersoluble drugs a higher dose per kg needs to be administered in the neonate compared with the adult.1,2
Blood–brain barrier: morphine is much less lipid soluble than other opioid analgesics, and is slower to cross the mature adult blood–brain barrier. Neonates have an immature barrier, which results in faster morphine uptake into the CNS. This is partly responsible for the considerable neonatal sensitivity to the CNS effects of morphine. Disease states such as sepsis, hypoxia and acidosis further reduce the integrity of the blood– brain barrier.
Body tissue composition: neonates have a smaller proportion of weight in the form of fat and muscle compared with adults (Figure 2).3 In adults thiopentone or fentanyl have an early peak plasma concentration followed by a rapid redistribution to muscle (and fat). In the neonate with less muscle mass, this redistribution capacity will be reduced, causing prolonged higher plasma concentrations. Opposing pharmacokinetic factors frequently need to be balanced when evaluating drug dosage in neonates. For example,
Metabolism The ability of a neonate to metabolize drugs is mainly dependent on enzyme maturation and hepatic blood flow. These factors have variable expression and development. Pathological conditions such as low cardiac output states, raised intra-abdominal pressure or liver disease can also alter the disposition of drugs.
Age-related body water compartments Preterm neonate
Term neonate
Infant 1 year
Adult man
TBW 85%
TBW 80%
TBW 60%
TBW 60%
ECF 60%
ECF 45%
ECF 25%
ECF 20%
ICF 25%
ICF 35%
ICF 35%
ICF 40%
ECF, extracellular fluid; ICF, intracellular fluid; TBW, total body water
Figure 1
ANAESTHESIA AND INTENSIVE CARE MEDICINE 9:3
100
© 2008 Elsevier Ltd. All rights reserved.
Neonatal
Enzyme systems: drug biotransformation to more polar forms occurs via phase I (oxidation, reduction and hydrolysis) and phase II conjugation reactions (e.g. glucuronidation). Many phase I reactions involve isoenzymes found in the CYP1, CYP2 and CYP3 gene families of the cytochrome P450 system.3 Premature neonates are able to metabolize drugs; however, different enzymes have variable development. Genetic polymorphism of phase I and phase II enzymes add to this variability (e.g. CYP2D6 and codeine)3,4 (Table 2).
Proportion of body composition (%)
Proportion of fat and muscle from premature neonate to adult 50
Muscle Fat
40
30
Renal excretion At birth, glomerular and tubular functional immaturity can prolong the elimination half-life of many drugs. Renal clearance rate of a drug is dependent on: glomerular filtration rate (GFR), fraction of drug unbound to protein (fu), and the balance of tubular reabsorption clearance (CLRA) and tubular secretion clearance (CLS).5
20
10
0
Preterm
Term
Adult
CL R = fu × GFR + CL S − CL RA
Figure 2
GFR is gestational age related and is reduced with increasing prematurity. At 41 weeks postconceptual age the GFR is 1.5 ml/ kg/min (20–40 ml/min/1.73m2), which increases to half of adult levels by 3 months. Adult levels of GFR 2.0 ml/kg/min (120 ml/ min/1.73m2) are achieved by 2 years of age. Reasons for this increase are: • disappearance of the placental shunt • increased renal blood flow • altered distribution of renal blood flow • increased permeability of the glomerular membrane. With low renal blood flow at birth, the fine balance between vasoconstrictor and vasodilatory renal forces are vital to maintain filtration pressure. This function in neonates is particularly vulnerable when non-steroidal anti-inflammatory dugs (NSAIDs) are administered.
Hepatic blood flow: with neonatal maturation, a greater proportion of the cardiac output is delivered to the liver. This is particularly important for the metabolism of those drugs with a high hepatic intrinsic clearance and extraction ratio.1 This can be explained by the equation: Q × f × CL int CL H= Q + (f × CL int )
where Q is the hepatic blood flow, f is the fraction of free drug, Clint is the intrinsic clearance and CLH is the hepatic clearance. If intrinsic clearance of a drug is high (e.g. propranolol), changes in liver blood flow would have much more impact on overall hepatic clearance compared with a change in the hepatocellular metabolism. Conversely, if a drug has a small intrinsic clearance (or extraction ratio), the hepatic clearance is more dependent on either the fraction of free drug (e.g. clindamicin) or intrinsic clearance (e.g. chloramphenicol). We can therefore see here how protein binding will affect the clearance of the ‘binding-sensitive drugs’ such as clindamicin, but not the hepatic clearance of chloramphenicol.1
Individual drug groups Analgesics The belief that neonates are unable to experience pain has been rejected, and the importance of neonatal analgesia has been increasingly acknowledged during the past 20 years (page 98).6
Examples of cytochrome P450 enzymes involved in phase I metabolism and how they develop with postnatal age Enzyme
Drug substrate
Enzyme development
CYPA12
Paracetamol
CYP2C9
Phenytoin
CYP2D6
β-blockers, ondansetron, codeine, tricyclic antidepressants, neuroleptics, paroxetine, fluoxetine, haloperidol Midazolam
Absent in fetus but reaches adult levels at about 4 months. The low level of cytochrome P450 activity in the liver may be responsible for the apparent reduced formation of hepatotoxic metabolites in neonates Phenytoin half-life is prolonged at 75 hours in preterm neonates; this decreases to 20 hours at term and 8 hours at 2 weeks Usually present at 1 week but only 20% of adult activity at 1 month. Variable because of genetic polymorphism: up to 47% of 3–12-yearolds cannot convert codeine to morphine Activity low at birth but increases 5-fold over 3 months of life
CYP3A4 Table 2
ANAESTHESIA AND INTENSIVE CARE MEDICINE 9:3
101
© 2008 Elsevier Ltd. All rights reserved.
Neonatal
Opioid analgesics Morphine is frequently used during surgery for moderateto-severe perioperative pain relief and in neonatal intensive care for the stress of ventilation or painful conditions (e.g. necrotizing enterocolitis). The neonatal increased sensitivity to morphine is due to a combination of an immature blood–brain barrier and reduced hepatic and renal clearance, which leads to higher effector-site concentrations.4 Clearance improves rapidly with postconceptional age; the preterm neonatal clearance (approximately 3 ml/kg/min) can increase threefold by term and tenfold by 2–4 months. Adult values are reached by 6–12 months. Disease states can influence clearance; infants who have had cardiac surgery show a markedly reduced clearance compared with children who have undergone noncardiac surgery.3 Those with renal impairment have the potential to accumulate the metabolites morphine-6-glucuronide (potent μ opioid receptor activity) and morphine-3-glucuronide (antanalgesic effect).4 As with all opioid infusions, neonates should have continuous monitoring and frequent assessment with a high nurse:patient ratio. Oral morphine absorption is variable, with a high first-pass metabolism. Fentanyl – neonatal liver blood flow, bypassing flow through a patent ductus venosus, and immature enzyme systems are responsible for reduced clearance to 70–80% of mature values. Pathology, such as raised intra-abdominal pressure (IAP), that reduces hepatic flow, in turn reduces clearance. Neonates with raised IAP may require less fentanyl than those with normal IAP.7 The high lipid solubility renders the maturation of the blood–brain barrier less relevant in this situation compared with morphine (see above).
erivatives, which have an important role in nephrogenesis in d the fetus and in maintaining both glomerular and tubular function in the neonate.5 Vasodilator prostaglandins counteract the high renal vascular resistance in the first days of life by acting on prostaglandin-E receptor subtypes in the afferent glomerular vessels. In the presence of angiotensin-II-induced efferent vasoconstriction, the glomerular filtration pressure is generated to optimize GFR in the presence of poor glomerular perfusion if the first few days of life.5 It has been demonstrated that NSAID exposure can lead to a reduction in GFR, with reduced clearance of antibiotics such as gentamicin. Fatal renal failure has been reported following antenatal indomethacin exposure. NSAIDs should not be given to the mother during the renal developmental phase in utero.5 Bleeding times often remain within normal limits after NSAIDs. Intestinal perforation has been reported after treatment with indomethacin. In the future‚ stereoselective isomers of NSAIDs may have a role in neonates due to their improved safety profile (e.g. ibuprofen; r-ibuprofen is much more rapidly cleared than s-ibuprofen).9 Inhalational anaesthetics The minimum alveolar concentration (MAC) of halothane, isoflurane and desflurane is low in premature neonates, peaks at 1–6 months and then falls gradually as age increases (Figure 3).10 Interestingly, sevoflurane differs, in that the MAC in neonates (3.3%) does not change significantly through to 6 months (3.2%).11 The reason for this different pattern is unclear. The haemodynamic response to an age-adjusted MAC of volatile agents is no different from older children. Intravenous anaesthetics In the newborn, CNS sensitivity to sedatives is related to increased permeability of the developing blood–brain barrier,
Non-opioid analgesics Paracetamol – rectal absorption of paracetamol is slow and variable. Factors such as lower gastrointestinal motility, drug formulation (triglyceride or capsule) or even the height of the rectal venous plexus into which it is absorbed affects the variation of drug absorption. There are few data on rectal absorption for premature neonates. For oral administration, the rapid absorption in children aged more than 6 months is not seen in neonates because of erratic and slow gastric emptying.8 The introduction of intravenous paracetamol preparations may improve the predictability of bioavailability and increase speed of onset; however, it is currently not licensed for use in neonates. The newer intravenous paracetamol preparation appears to be safer than pro-paracetamol with fewer side effects.9 The clearance of intravenous paracetamol is reduced in neonates; however, it increases with postconceptional age. NSAIDs – indomethacin is used as a tocolytic and readily crosses the placenta. Postnatally, it is used in the closure of a patent ductus arteriosus.5,9 Although there are no linked pharmacodynamic or pharmacokinetic studies investigating analgesia in neonates, it has been used for postoperative pain (e.g. bladder extrophy repair).8 There is variable expression of the CYP2C family, and genetic polymorphism has been described. Clearance of NSAIDs increases with postconceptional and postnatal age.9 NSAIDs inhibit the two isoforms of cyclooxygenase (COX-1 and COX-2), which in turn have a key role in neonatal renal physiology. They are responsible for the synthesis of prostanoid
ANAESTHESIA AND INTENSIVE CARE MEDICINE 9:3
The minimum alveolar concentration (MAC) of isoflurane with increasing age 2.0
MAC (% isoflurane)
1–6 months 1.8
1.6
Neonates 32–37 weeks’ gestation
1.4
< 32 weeks’ gestation 1.2
1.0 0.5
1.0
5
10
50
100
Postconceptual age (years) This pattern of the MAC throughout childhood is similar for desflurane and halothane Reproduced with permission from LeDez KM, Lerman J. The minimum alveolar concentration of isoflurane in preterm neonates. Anesthesiology 1987; 67: 301–7
Figure 3
102
© 2008 Elsevier Ltd. All rights reserved.
Neonatal
changes in the body composition and decreased elimination of drugs. Enzyme immaturity contributes to a reduced hepatic clearance of thiopental in the neonate, with an elimination halflife of 17.9 hours compared with 6.1 hours in the child and 12 hours in the adult.3 Propofol 1% is not licensed for use as an induction agent in neonates. Animal studies have suggested that ketamine isoflurane, nitrous oxide, and or midazolam exposure to the developing brain causes neuronal apoptosis; however, the question of implication for the human neonatal brain development remains unanswered.
frequently used in children to reduce pain of venepuncture. In the neonate it has been used for procedures such as heel-pricks and spinal injections. The risk of systemic absorption causing methaemaglobinaemia can be reduced by applying the agent to a limited surface area of skin. There are not enough data to support multiple daily applications. ◆
References 1 Blumer JL, Reed MD. Principles of neonatal pharmacology. In: Yaffe SJ, Aranda JV, eds. Neonatal and pediatric pharmacology: therapeutic principles in practice, 3rd edn. Philadelphia: Lippincott, Williams & Wilkins, 2005. 2 van den Anker J. Problems with medicines in the newborn infant. In: Jacqz-Aigrain E, Choonara I, eds. Paediatric clinical pharmacology. New York: Taylor and Francis Group, 2006. 3 Cote CJ, Lugo RA, Ward RM. Pharmacokinetics and pharmacology of drugs in children. In: Cote CJ, Todres ID, Goudsouzian NG, Ryan JF, eds, 3rd edn. Philadelphia: Saunders, 2001. 4 Palmer GM, Anderson BJ. Opioid analgesic drugs. In: Jacqz-Aigrain E, Choonara I, eds. Paediatric clinical pharmacology. New York: Taylor and Francis Group, 2006. 5 van den Anker J. Renal function and excretion of drugs in the newborn. In: Yaffe SJ, Aranda JV, eds. Neonatal and pediatric pharmacology: therapeutic principles in practice, 3rd edn. Philadelphia: Lippincott, Williams & Wilkins, 2005. 6 Simons SHP, Anderson BJ, Tibboel D. Analgesic agents. In: Yaffe SJ, Aranda JV, eds. Neonatal and paediatric pharmacology: therapeutic principles in practice, 3rd edn. Philadelphia: Lippincott, Williams & Wilkins, 2005. 7 Koehntop DE, Rodman JH, Brundage DM, et al. Pharmacology of fentanyl in neonates. Anesth Analg 1986; 65: 227–32. 8 Anderson BJ. Paracetamol. In: Jacqz-Aigrain E, Choonara I, eds. Paediatric clinical pharmacology. New York: Taylor and Francis Group, 2006. 9 Anderson BJ, Palmer GM. Recent developments in the pharmacological management of pain in children. Curr Opin Anaesthesiol 2006; 19: 285–92. 10 Taylor RH, Lerman J. Minimum alveolar concentration of desflurane and hemodynamic responses in neonates, infants and children. Anesthesiology 1991; 75: 975–9. 11 Lerman J, Sikich N, Kleinman S, Yentis S. The pharmacology of sevoflurane in infants and children. Anesthesiology 1994; 80: 814–24. 12 Chalkiadis GA, Anderson BJ. Age and size are the major covariates for the prediction of levobupivacaine clearance in children. Paediatr Anaesth 2006; 16: 275–82. 13 Larsson BA, Lonnqvist PA, Olsson GL. Plasma concentrations of bupivacaine in neonates after continuous epidural infusion. Anesth Analg 1997; 84: 501–5.
Benzodiazepines Diazepam: reduced hepatic blood flow and immature hepatic excretory mechanisms can prolong the elimination half-life of diazepam to up to 100 hours in the neonate (18 hours in young adults). The active metabolite N-desmethyldiazepam is similar in potency and half-life to diazepam.3 Many would consider it unsuitable for use in neonates. The intravenous solutions that contain the preservative benzyl alcohol should be avoided in neonates because of the risk of metabolic acidosis and kernicterus. Midazolam: unlike diazepam, the active metabolite of mida zolam has minimal activity; it is therefore more suited for use in neonates. The clearance is prolonged (6–12 hours) compared with older children (1.4–4.0 hours).3 This can be reduced by factors that reduce hepatic blood flow or hypovolaemic states. Local anaesthetics Regional techniques are used to reduce the use of systemic an algesics and may facilitate earlier extubation. The toxic plasma concentration of bupivacaine is unknown in children. However, many infusion regimens aim for a plasma concentration of less than 1 μg/ml. Pharmacokinetic data of epidural bupivacaine suggest that neonates will develop higher plasma concentrations of bupi vacaine compared with older children. The immature cytochrome P450 enzyme systems are the main reason for reduced clearance of the drug, although reduced hepatic blood flow may also contribute. Pharmacokinetic studies of levobupivacaine show that neonatal clearance is approximately 25% of adult values, which increases to 80% at 6 months of age. Most of the clearance can be predicted by postnatal age and weight.12 The reduced protein binding may be less important than drug clearance in determining toxic levels. Doses per kg should be reduced in neonates compared with older children. With a reduced 1.8 mg/kg epidural bolus of bupivacaine followed by a 0.2 mg/kg/hour infusion, some neonates still have rising plasma concentrations after 48 hours, which raises concern about prolonged infusions in this age group.13 Newer local anaesthetic agents such as levobupivacaine and ropivacaine may have a safer profile. Topical local anaesthetics such as EMLA (2.5% prilocaine, 2.5% lidocaine) are
ANAESTHESIA AND INTENSIVE CARE MEDICINE 9:3
103
© 2008 Elsevier Ltd. All rights reserved.