Because of the use of pooled data obtained from animal eyes, pharmacokinetic profiles of various ocular tissues are limited in their interpretation of the data. Sophisticated models that more adequately explain the data are often too complex and tend to be overly scattered and not as smooth as data obtained in a serial fashion. Therefore, the models have to be oversimplified. As a result of the oversimplification of the modeling and the use of animal eyes instead of human eyes, the value of ocular pharmacokinetic studies are limited to:
1. Validating the minimum effective concentrations in various tissues from different routes of administration
2. Determining the best route of administration for entry into the eye as well as accumulation at the site of action
3. Determining the elimination half-life, which can serve as a guide to selecting dosing regimens for further study
4. Estimating the safety of various drugs by determining their accumulation in tissues of interest, such as the lens or retina
Approaches to extending the application of pharmacokinetics have recently focused on two approaches: serial sampling and microdialysis.
Serial sampling has not been advocated as a reliable method for obtaining samples from animal eyes because of the possibility of a breakdown of the blood-aqueous barrier, resulting in an alteration of the drug clearance from the eye. Studies by Tang-Liu and coworkers (2,3) established that for a single intracameral injectkion of 5 mL solutions containing flurbiprofen or levobunolol, no significant breakdown of the blood aqueous barrier had occurred since protein concentration was < 1 mg/mL. However, serial sampling is a repetitive process not just a single injection. Miller et al. (4) studied the effect of paracentesis on the pharmacokinetics of fleroxacin following direct intravitreal or systemic drug administration and serially measuring drug levels in serum, aqueous, and vitreous humor. Samples were obtained from aqueous humor with the use of a sterile 30-gauge needle fused to a calibrated 20 mL capillary tube. A volume of 7 mL was removed at each time interval. A 28-gauge needle was used to sample vitrous humor and 20 mL samples were removed. Figure 1 gives the results, for which no statistically significant differences were observed in assigning a correct model for the data, nor were there significant differences in the half-lives for serum, aqueous, and vitreous humor—2.34,3.20, and 3.88 hours, respectively. Similar experiments with similar results were conducted for the pharmacokinetics of amikacin and chloramphenicol in rabbit aqueous humor following anterior chamber injection and serial sampling using 30-gauge needle and removing 7 mL of aqueous (5).
The microdialysis approach was widely used in the analysis of drugs in brain tissue and cerebral spinal fluid with the intention of measuring free drug concentration at the sampling site (6,7). More recently the approach has been adapted for use in the eye, particularly in the measurement of aqueous and vitreal humor concentrations of drug (8-12). Its primary advantage is to measure complete concentration-time profiles in individual animal eyes, reducing the number of animals required, and to more accurately define the pharmacokinetics of ocular drugs.
The procedure consists of surgically placing a probe within the aqueous or vitreous chambers of an anesthetized animal. The microdialysis
probe is designed to allow for perfusate to flow into and out of the unit at a fixed rate, usually 2-3.5 mL/min. The base of the probe is constructed with a semipermeable membrane (e.g., a 4 mm polycarbonate membrane with a 20,000 dalton cutoff), which allows drug to diffuse into a perfusate (11). Probe inlet and outlet lines allow for collection of dialysate and analysis of drug over time. It is essential for the percent relative recovery to be determined so that accurate concentrations can be estimated, which may likely vary depending on the membrane used in the probe, the size of the probe and the perfusion rate. For example, reported values for propranolol and two nucleoside antivirals (ganciclovir and acyclovir) were 46 and 15%, respectively (12,13), when probe and perfusion rate varied. Of concern is the insertion of the probe, which appears to result in an increase in protein concentrations in aqueous humor (12), suggesting an alteration of clearance. The use of systemic anesthesia has also been shown to alter the pharmaco-kinetics of propranolol by increasing the area under the aqueous humor time-concentration plot, possibly by decreasing aqueous humor turnover (12). Nevertheless, results using this technique have reduced the number of animals needed to estimate ocular pharmacokinetic parameter values and have produced curves that are smooth in appearance and, therefore, more reliable in itnerpreting the fitting of models to the results.
Was this article helpful?