Antibiotic tissue penetration and its relevance: impact of tissue penetration on infection response.

Fluid penetration. Given that distribution in extracellular fluid (ECF) is important, a variety of methods have been applied to the collection and study of data on ECF distribution. Antibiotic concentrations have been measured in the lymphatic drainage of organs and tissues (95, 96), fluids obtained from tissue cage reservoirs (82, 90), chemically or mechanically induced skin blisters (1, 81), surgically implanted cotton threads (28, 81), implanted fibrin clots (3), and directly from inflammatory exudates (72). This diversity of models has confused the interpretation of antibiotic extravascular fluid distribution, particularly in comparisons between members of the same class of compounds. Whenever possible, comparisons of drugs should be made under identical conditions. Antibiotics leave the vasculature and enter ECF via passive diffusion through the spaces between vascular endothelial cells. Thus, the surface area of vascular tissue in relation to the total volume of the tissue to be sampled is the most important consideration in the modeling of ECF (26). The commonly used tissue penetration models differ in the vascular tissue surface area/volume ratio at the measurement site (28, 77, 81, 93). In addition, the inflammatory response induced by the procedure (81, 104) and the presence of cellular debris in the sample (i.e., paper disks or cotton threads) (81) may influence results. Given these multiple sources of confusion, it is vital to define general principles of antibiotic extravascular fluid penetration and then interpret the outcomes of specific fluid models within this context. In the tissue cage method, small, inert spheres or tubes are surgically implanted in subcutaneous tissue (12). The surface of the device contains multiple pores, each 1 to 2 mm in diameter. The implantation procedure causes an inflammatory response that may continue for 2 to 3 weeks (31). After 4 weeks, the fluid contained in the reservoir is biochemically similar to ECF (25, 31). One great advantage of this method is that multiple samples can be obtained over a period of days or weeks. One must be careful to maintain sterility when puncturing the device to collect fluid samples, or infection may complicate the data. After several weeks, the growth of vascular or fibrous tissue may limit the proper functioning of the device. The tissue cage device itself is usually spherical, with a relatively small diffusion surface area in relation to a large fluid volume. As a result, the peak antibiotic concentration in the device generally lags behind the peak antibiotic concentration in the blood or in true ECF (30, 93). Steady-state concentrations normally reflect complete equilibration between the fluid in the device and the free drug concentration in serum (30, 92). Skin blister models are perhaps more useful in estimating antibiotic concentrations in ECF. Noninflammatory blisters