Drug Use and Nephrotoxicity in the Intensive Care Unit
Drug Use and Nephrotoxicity in the Intensive Care Unit
Outside innate drug toxicity and the disease states discussed above, one of the major factors leading to drug-induced systemic toxicity is disturbed medication pharmacokinetics in critically ill patients (Table 3). All phases of drug pharmacokinetics are disturbed in ICU patients, including absorption, distribution, metabolism, and clearance. These changes are often the result of organ dysfunction, particularly in the kidneys and liver, the acute-phase response of the underlying critical illness, multiple drug interactions, and therapeutic interventions, including intravenous fluids, diagnostic procedures, and various medications. Individual differences in the pharmacogenetics of drug transport systems and metabolizing enzymes may be exacerbated in this setting, further increasing the risk of toxicity.
Although drug bioavailability is less of an issue with intravenously administered medications, oral, subcutaneous, and intramuscular drugs often have altered bioavailability in ICU patients. Reduced absorption of enteral drugs occurs because of a number of factors, including increased gastric pH (alters the drug-ionized state), bowel dysmotility, formation of insoluble drug complexes in the intestines, nutrient–drug interactions, and bowel edema. Acute decompensated heart failure and cirrhosis will reduce oral bioavailability due to bowel edema and impaired intestinal perfusion. A less-appreciated cause of reduced absorption is intestinal atrophy with associated decreased surface area and cellular enzyme activity, which can occur as a result of as little as 3 days of reduced enteral feeding. On the basis of these data, it is prudent to dose medications intravenously and to initiate enteral feeds early to avoid gut atrophy. Less commonly, reduced intestinal/hepatic metabolism of enteral drugs actually increases their bioavailability.
Drug distribution is significantly altered in ICU patients for various reasons. This is due primarily to changes in the factors that determine distribution: blood delivery, tissue permeability, microenvironment pH, and drug characteristics, such as lipid solubility, pKa, and protein binding. The initial concentration of a drug after intravenous administration is determined by dose and volume of distribution (Vd), a hypothetical space representing total body drug concentration divided by plasma blood concentration. In critical illness, Vd may be increased or decreased depending on the level of total body water, level of renal function, and changes in protein concentration and binding affinity. Increased drug Vd often develops from early goal-directed therapy for sepsis, edematous states, such as cirrhosis and acute hepatic failure, nephrotic syndrome, right and left heart failure, and many illnesses associated with shock requiring aggressive volume repletion to maximize end-organ perfusion. Efficacy can suffer if drug dosing is not adjusted to these changes. Reduced Vd may be seen with older age and volume depletion from vomiting, diarrhea, blood loss, and diuretics. AKI and CKD may also reduce the Vd of certain drugs by altering tissue binding. Drug therapy that does not account for this will lead to excessive dosing and end-organ toxicity.
Drug metabolism is disturbed by many of the critical illness states. The two major metabolic pathways for drugs are categorized as phase I and II. Although the liver is the major organ involved in these pathways, the intestines, kidneys, and other organs also participate to a lesser degree. Phase I entails oxidation, reduction, and hydrolysis of certain drugs. Hepatic metabolism is influenced by the level of blood flow, hepatocyte enzyme activity, and protein binding. Vigorous hepatic blood flow allows drugs with high extraction ratios to be rapidly cleared; these drugs accumulate when perfusion is reduced by hypotension/shock, vasoconstrictor drugs, ventilation with positive end expiratory pressure, and impaired cardiac output. For drugs with low extraction ratios, the enzyme function is most important and is performed primarily by the cytochrome P-450 system. An essential component of enzyme function is tissue transporter activity, which participates in uptake and removal of drugs from the liver and other metabolizing organs. Drug transporters such as p-glycoprotein and organic anion transporter may be affected by critical illness states, such as inflammation, sepsis, CKD/AKI, acute or chronic liver disease, hypotension, burns, and trauma. In general, these processes have variable effects on transporter activity, while predominantly reducing cytochrome P-450 enzyme activity. Thus, the metabolism of drugs in ICU patients is clearly altered and likely contributes to efficacy and toxicity issues. AKI may result when potentially nephrotoxic drugs and/or their active metabolites accumulate.
Drugs are eliminated from the body by both renal and nonrenal pathways. Hepatic clearance is considerable for lipophilic drugs, whereas hydrophilic drug excretion is largely a function of renal clearance. Currently used markers of kidney function, such as serum creatinine concentration, are suboptimal, especially in the ICU setting. Resuscitation with large volumes of crystalloid will dilute serum creatinine concentration masking the presence of AKI or misinforming about the stage of CKD. Creatinine production is also reduced in sepsis, further limiting its potential as a marker of kidney injury. Thus, abrupt decreases in GFR initially may be accompanied by little or no change in the serum creatinine. As a result, parent drugs and active metabolites from liver metabolism that are excreted by the kidneys, accumulate and can cause toxicity. In addition, drug clearance is affected by the etiology of critical illness. As compared with normal subjects, burn patients more often have increased clearance because of younger age, aggressive fluid resuscitation, and hypermetabolic state. In contrast, studies on patients in medical and surgical ICUs demonstrate no change or reduced drug clearance as compared with normal subjects. However, these are only generalizations, and patients with any of the noted illnesses can have increased, unchanged, or decreased drug clearance, emphasizing the need for individualized dosing. Thus, ICU patients are at risk for underdosing with loss of efficacy and adverse toxic events from drug overdosing, especially when GFR is overestimated by an inaccurate parameter, such as serum creatinine.
With severe AKI in the ICU, some form of RRT is often required. Acute intermittent hemodialysis and various forms of continuous RRT are used to provide clearance, metabolic control, and fluid balance. Drug clearances and changes in pharmacokinetics must be considered with addition of this therapy. Not only must drug dosing take into account RRT-associated clearances but also dialysis-induced changes in Vd and drug distribution, as well as the associated improvement in hepatic and other organ drug metabolism. The latter point is important and not commonly appreciated with dialytic therapy of the uremic patient. In addition, one must recognize that extracorporeal drug clearances are different with chronic RRT (CRRT) and intermittent hemodialysis. As will be discussed later, continuous convective clearance requires dosing regimens that are distinct from intermittent diffusion-based hemodialysis.
Finally, pharmacokinetic changes that occur with critical illness are by no means static but rather represent a dynamic process. Many change as the clinical status of the patient changes. For example, eradication of infection, recovery of end-organ dysfunction, improvement in intravascular volume status, and enhanced nutritional status are but a few of the changes that occur in this unique subset of patients. Alternatively, patients may deteriorate and develop worsening anasarca, hypotension, new end-organ failure, and a progressive decline in clinical status. As such, these changes must be kept in mind when prescribing medications, with vigilant monitoring as integral to appropriate therapy.
Pharmacokinetic Changes in ICU Patients
Outside innate drug toxicity and the disease states discussed above, one of the major factors leading to drug-induced systemic toxicity is disturbed medication pharmacokinetics in critically ill patients (Table 3). All phases of drug pharmacokinetics are disturbed in ICU patients, including absorption, distribution, metabolism, and clearance. These changes are often the result of organ dysfunction, particularly in the kidneys and liver, the acute-phase response of the underlying critical illness, multiple drug interactions, and therapeutic interventions, including intravenous fluids, diagnostic procedures, and various medications. Individual differences in the pharmacogenetics of drug transport systems and metabolizing enzymes may be exacerbated in this setting, further increasing the risk of toxicity.
Bioavailability
Although drug bioavailability is less of an issue with intravenously administered medications, oral, subcutaneous, and intramuscular drugs often have altered bioavailability in ICU patients. Reduced absorption of enteral drugs occurs because of a number of factors, including increased gastric pH (alters the drug-ionized state), bowel dysmotility, formation of insoluble drug complexes in the intestines, nutrient–drug interactions, and bowel edema. Acute decompensated heart failure and cirrhosis will reduce oral bioavailability due to bowel edema and impaired intestinal perfusion. A less-appreciated cause of reduced absorption is intestinal atrophy with associated decreased surface area and cellular enzyme activity, which can occur as a result of as little as 3 days of reduced enteral feeding. On the basis of these data, it is prudent to dose medications intravenously and to initiate enteral feeds early to avoid gut atrophy. Less commonly, reduced intestinal/hepatic metabolism of enteral drugs actually increases their bioavailability.
Distribution
Drug distribution is significantly altered in ICU patients for various reasons. This is due primarily to changes in the factors that determine distribution: blood delivery, tissue permeability, microenvironment pH, and drug characteristics, such as lipid solubility, pKa, and protein binding. The initial concentration of a drug after intravenous administration is determined by dose and volume of distribution (Vd), a hypothetical space representing total body drug concentration divided by plasma blood concentration. In critical illness, Vd may be increased or decreased depending on the level of total body water, level of renal function, and changes in protein concentration and binding affinity. Increased drug Vd often develops from early goal-directed therapy for sepsis, edematous states, such as cirrhosis and acute hepatic failure, nephrotic syndrome, right and left heart failure, and many illnesses associated with shock requiring aggressive volume repletion to maximize end-organ perfusion. Efficacy can suffer if drug dosing is not adjusted to these changes. Reduced Vd may be seen with older age and volume depletion from vomiting, diarrhea, blood loss, and diuretics. AKI and CKD may also reduce the Vd of certain drugs by altering tissue binding. Drug therapy that does not account for this will lead to excessive dosing and end-organ toxicity.
Metabolism
Drug metabolism is disturbed by many of the critical illness states. The two major metabolic pathways for drugs are categorized as phase I and II. Although the liver is the major organ involved in these pathways, the intestines, kidneys, and other organs also participate to a lesser degree. Phase I entails oxidation, reduction, and hydrolysis of certain drugs. Hepatic metabolism is influenced by the level of blood flow, hepatocyte enzyme activity, and protein binding. Vigorous hepatic blood flow allows drugs with high extraction ratios to be rapidly cleared; these drugs accumulate when perfusion is reduced by hypotension/shock, vasoconstrictor drugs, ventilation with positive end expiratory pressure, and impaired cardiac output. For drugs with low extraction ratios, the enzyme function is most important and is performed primarily by the cytochrome P-450 system. An essential component of enzyme function is tissue transporter activity, which participates in uptake and removal of drugs from the liver and other metabolizing organs. Drug transporters such as p-glycoprotein and organic anion transporter may be affected by critical illness states, such as inflammation, sepsis, CKD/AKI, acute or chronic liver disease, hypotension, burns, and trauma. In general, these processes have variable effects on transporter activity, while predominantly reducing cytochrome P-450 enzyme activity. Thus, the metabolism of drugs in ICU patients is clearly altered and likely contributes to efficacy and toxicity issues. AKI may result when potentially nephrotoxic drugs and/or their active metabolites accumulate.
Excretion
Drugs are eliminated from the body by both renal and nonrenal pathways. Hepatic clearance is considerable for lipophilic drugs, whereas hydrophilic drug excretion is largely a function of renal clearance. Currently used markers of kidney function, such as serum creatinine concentration, are suboptimal, especially in the ICU setting. Resuscitation with large volumes of crystalloid will dilute serum creatinine concentration masking the presence of AKI or misinforming about the stage of CKD. Creatinine production is also reduced in sepsis, further limiting its potential as a marker of kidney injury. Thus, abrupt decreases in GFR initially may be accompanied by little or no change in the serum creatinine. As a result, parent drugs and active metabolites from liver metabolism that are excreted by the kidneys, accumulate and can cause toxicity. In addition, drug clearance is affected by the etiology of critical illness. As compared with normal subjects, burn patients more often have increased clearance because of younger age, aggressive fluid resuscitation, and hypermetabolic state. In contrast, studies on patients in medical and surgical ICUs demonstrate no change or reduced drug clearance as compared with normal subjects. However, these are only generalizations, and patients with any of the noted illnesses can have increased, unchanged, or decreased drug clearance, emphasizing the need for individualized dosing. Thus, ICU patients are at risk for underdosing with loss of efficacy and adverse toxic events from drug overdosing, especially when GFR is overestimated by an inaccurate parameter, such as serum creatinine.
With severe AKI in the ICU, some form of RRT is often required. Acute intermittent hemodialysis and various forms of continuous RRT are used to provide clearance, metabolic control, and fluid balance. Drug clearances and changes in pharmacokinetics must be considered with addition of this therapy. Not only must drug dosing take into account RRT-associated clearances but also dialysis-induced changes in Vd and drug distribution, as well as the associated improvement in hepatic and other organ drug metabolism. The latter point is important and not commonly appreciated with dialytic therapy of the uremic patient. In addition, one must recognize that extracorporeal drug clearances are different with chronic RRT (CRRT) and intermittent hemodialysis. As will be discussed later, continuous convective clearance requires dosing regimens that are distinct from intermittent diffusion-based hemodialysis.
Finally, pharmacokinetic changes that occur with critical illness are by no means static but rather represent a dynamic process. Many change as the clinical status of the patient changes. For example, eradication of infection, recovery of end-organ dysfunction, improvement in intravascular volume status, and enhanced nutritional status are but a few of the changes that occur in this unique subset of patients. Alternatively, patients may deteriorate and develop worsening anasarca, hypotension, new end-organ failure, and a progressive decline in clinical status. As such, these changes must be kept in mind when prescribing medications, with vigilant monitoring as integral to appropriate therapy.