Page 239 - Academic Press Encyclopedia of Physical Science and Technology 3rd BioTechnology
P. 239

P1: GTV/GWT  P2: GRB/GJP  QC: FYDFinal Pages
 Encyclopedia of Physical Science and Technology  EN011J-559  July 25, 2001  18:57







              Pharmacokinetics                                                                            817

              E.  Patients with Liver Disease                   renal clearance appears to be increased 1 week after the
                                                                injury, while hepatic metabolism is quite variable. Poorly
              Most data concerning drug elimination in liver disease
                                                                extracted drugs may have a significantly decreased hepatic
              have been obtained in patients with cirrhosis. Current the-
                                                                clearance. Therapeutic drug monitoring is recommended
              ories regard the decreased permeability of the hepatic si-
                                                                for all critical medications in burn patients.
              nusoid as a major feature in the changes brought about
              in the liver. Oxidative metabolism is typically impaired in
              cirrhosis with the sparing of drug glucuronidation until se-  VI.  QUANTITATIVE APPROACHES TO
              vere impairment occurs. Biliary excretion is also impaired.  PHARMACOKINETIC MODELING
              Moderate liver impairment decreases the clearance of the
              native drug or metabolites from the kidney, often in spite
                                                                The goal of pharmacokinetics is the quantitative descrip-
              of normal creatinine clearance. Because renal tubular se-
                                                                tion  of  drug  entry,  distribution,  and  elimination  in  the
              cretion of creatinine is increased in cirrhosis, creatinine
                                                                body. These processes are typically integrated into a math-
              clearance may not reflect the actual degree of renal dys-
                                                                ematical model which uses a system of equations such as
              function. In cases of liver disease without cirrhosis (car-
                                                                Eqs. (1)–(20) to calculate the systemic concentration due
              cinoma, viral hepatitis), dose changes are probably not
                                                                to a certain dose administered to the patient. If the phar-
              necessary. In general, chronic liver disease with cirrho-
                                                                macodynamic characteristics of the drug can be clearly
              sis requires reduction in dose, regardless of the route of
                                                                defined, the desired concentration at the target or in the
              elimination.
                                                                plasma can be specified, and the pharmacokinetic model
                                                                can be used to calculate the dose to attain the effective
              F.  Congestive Heart Failure                      concentration.
              Congestive heart failure includes cardiac “pump” failure
                                                                A. Compartmental Approaches
              and results in poor perfusion of the liver and kidney. Con-
              gestion also occurs in the liver and the gut. Pharmacoki-  The classic approach to pharmacokinetic modeling is to
              netics is altered in heart failure: the volume of distribution  describe the system as a group of arbitrarily sized com-
              is reduced and clearance is decreased. The net result of  partments with one compartment designated as the cen-
              changes is often not predictable, but plasma drug con-  tral compartment which receives the drug via some route
              centrations are usually higher in patients with congestive  and from which the excretion occurs. Figure 7 displays
              failure than in healthy persons. Drug levels of such com-  a three-compartment model. In this example, we will as-
              mon cardiac drugs as lidocaine rise proportionately to the  sume that the drug has been given directly into the central
              degree of cardiac failure. Plasma concentrations in cardiac  compartment (1); however, a fourth compartment could
              shock can rise to toxic levels. The half-life of antipyrine,  be added for the route of absorption or an input function
              used as a marker of hepatic metabolism, is prolonged in  of drug mass entering compartment 1 versus time. The
              cardiac failure but returns to normal during convalescence.  compartments may or may not match anatomic compart-
              It is therefore important to monitor drug levels of toxic  ments such as the plasma space or the extracellular space.
              drugs during conditions of myocardial infarction and se-  The peripheral compartments 2 and 3 may represent parts
              vere congestive failure.                          of the body which have different rates of perfusion than
                                                                the central compartment. Each compartment is considered

              G.  Patients with Burns
              Burn injuries cause a variety of changes in drug pharma-
              cokinetics. Drug absorption of orally administered drugs
              may be enhanced because of increased permeability of the
              intestine. Unfortunately this condition permits bacteria to
              pass into the blood as well and often results in septicemia
              or shock. For the first few days, the levels of proteins do
              not change, but after 4–5 days, albumin progressively falls
              and α 1 -acid glycoprotein (AAG) increases. Decreased al-
              bumin concentrations but increased AAG levels affect the
              distribution and elimination of drugs. The binding has ma-
                                                                FIGURE 7 Three-compartment model. The k i are the rate coef-
              jor  effects  on  the  clearance  and  V d .  Clearance  is  often  ficients, C i is the concentration in compartment, i , and V i is the
              enhanced directly through the burned area. In addition,  volume in compartment i . See text for detailed discussion.
   234   235   236   237   238   239   240   241   242   243   244