Aminoglycosides such as gentamicin and amikacin are antibiotics, which are mainly used against aerobic gram‐negative bacteria. They can also be used synergistically with beta‐lactam antibiotics against certain gram‐positive organisms (Hammett‐Stabler and Johns [80]). Traditional dosing follows standard 8‐hour dosing regimens. Drug‐level monitoring revolves around the determination of peak and trough drug concentrations. Subsequent changes to the doses, dose intervals or both are made with the aim of achieving a drug concentration within the accepted therapeutic range. The timing of serum sampling will be determined by the patient's clinical manifestations, particularly for renal function. For gentamicin, the trough level should be lower than 2 mg/L and that for amikacin should be <10 mg/L. The peak level for gentamicin should be 5–8 mg/L and that for amikacin should be 20–30 mg/L (Walker and Whittlesea [249], Winter [262]).

Administering extended‐interval dosing of aminoglycosides (high doses that are given less frequently) instead of traditional dosing has been found to be associated with fewer adverse events. The benefit of high doses results from their rapid ‘concentration‐dependent’ action; the rate and extent of bacterial cell death are increased (Owens and Shorr [175], Pagkalis et al. [178]). Extended‐interval dosing also reduces the adaptive resistance of gram‐negative bacilli to aminoglycosides and is known as the first exposure effect (Urban and Craig [246]). Even when the level of the drug falls below the minimum inhibitory concentration (MIC) of the organism, there is suppression of the regrowth of organisms that have been exposed to aminoglycosides; this is the post‐antibiotic effect. Peak‐level tests are not performed in extended‐interval dosing as such tests are designed to ensure a high peak level and a low trough level. Instead, a single drug‐level test performed 6 to 14 hours after giving the dose is used to monitor therapy.

Vancomycin is the glycopeptide antibiotic most widely used for the treatment of serious infections caused by gram‐positive pathogens, such as those suspected to be meticillin resistant (Jones [110]). In vitro models suggest that vancomycin exhibits time‐dependent effects on Staphylococcus aureus and coagulase‐negative staphylococci. This means the effectiveness of vancomycin increases the longer the antibiotic serum concentration is above the MIC of susceptible pathogens (Lowdin et al. [129]). However, a study of 108 patients with S. aureus lower respiratory tract infections demonstrated highly significant associations between clinical cure and an AUC/MIC (area under the curve/minimum inhibitory concentration) over 400 and between bacteriological cure and an AUC/MIC over 850 (Moise‐Broder et al. [149]). This suggests that there is time‐dependent killing and that there are moderate to prolonged persistent effects, and that it is the amount of drug received that determines the ideal dosing regime.

The use and monitoring of vancomycin have increased due to the emergence of meticillin‐resistant S. aureus (MRSA). However, the emergence of vancomycin‐resistant S. aureus has further complicated this drug's clinical use (Roberts et al. [209]).

How often serum blood levels of vancomycin are determined depends on the frequency of administration. If it is given as an intermittent infusion (usually twice‐daily doses), trough levels should be taken at set times, immediately before the administration of the next bolus dose (Jones [110]). The trough level should be between 10 and 20 mg/L (Rybak et al. [216]).

An alternative method of administration is the continuous infusion of vancomycin. This is as effective as intermittent administration and is associated with reduced nephrotoxicity (Cataldo et al. [30]). Continuous infusions aim to maintain serum concentrations at between 15 and 25 mg/L.