Variation in the pharmacologic action of many drugs correlates more closely with the serum concentration of active drug or metabolite or the free fraction in serum than with the dose.  Monitoring of serum levels is useful in a variety of situations:

• When there is wide pharmacokinetic variation among individuals.
• When zero order kinetics apply; e.g., phenytoin, where the drug is eliminated at a constant rate rather than at a rate proportional to the amount of remaining drug.
• When the therapeutic index of a drug is low; i.e., when there is a small difference between the toxic and effective concentrations.
• When signs of toxicity are difficult to recognize clinically.
• When physiologic factors are present which may affect drug pharmacokinetics; e.g., age, nutritional state, etc.
• When intercurrent illness is present that may affect drug pharmacokinetics; e.g., kidney damage or liver damage.
• When the usual dose does not give expected results.
• When noncompliance is suspected.
• When surreptitious use is suspected.
• When drug interactions are suspected.

Time of Sampling

Steady state drug concentrations are most generally useful when the drug has reached equilibrium and the amount of the drug systematically available is equal to the amount being eliminated.  Trough levels, drawn immediately before the next dose, are used for constant maintenance in the therapeutic range.  Peak levels, usually drawn one to two hours after the dose, may be useful to detect possible toxic levels.

Therapeutic drug levels are also affected by age, disease states, metabolism, and the presence of other drugs.  Therapeutic decisions should never be based solely on drug concentration in serum.