Factors influencing nonoxynol-9 permeation and bioactivity in cervical mucus.

The effects of delivery gel pH and osmolarity on both the mass transport and 'biodiffusion' of the spermicide nonoxynol-9 (N9) in bovine cervical mucus were evaluated. Delivery gels were calcium chloride crosslinked alginate containing 3% N9, and were manufactured over a pH range of 3.4 to 5.9 and an osmolarity range of 300 to 900 mosmol. Mass transfer parameters (diffusion coefficients and total drug loading) were determined using a new UV spectrophotometric technique while biodiffusion (the diffusion distance into mucus at which sperm are killed) was assessed using the Double Ended Test. It was found that delivery gel pH had a significant effect on spermicidal efficacy of the alginate-N9 system; biodiffusion increased with decreasing pH. Actual N9 diffusion into mucus was found to be influenced by both the delivery gel pH and osmolarity. At high N9 concentration (near the gel/mucus interface), mass transport tended to decrease with decreasing pH at the highest osmolarity. At low concentration, mass transport tended to decrease with increasing osmolarity and decrease with increasing pH at the highest osmolarity. The difference between low and high concentration behavior can be attributed to N9 micelle formation. These findings are interpreted in the context of the design of intravaginal drug delivery vehicles for spermicides.

Spectrophotometric analysis of molecular transport in gels.

An automated spectrophotometric method has been developed for analyzing molecular transport out from and into gels. A Beckman DU7500 diode-array UV-visible spectrophotometer with gel scanner was modified to accept and longitudinally scan a quartz diffusion cell, 0.3x10x40 mm. Molecules of interest are identified and concentrations quantitated via analysis of spectrophotometric absorbance peaks relative to background absorbance of the gel. Thus, concentration profiles are obtained as functions of both position and time. Test data are fitted to a diffusion model via nonlinear least-squares regression. Precision and accuracy of the method were assessed via analysis of several test molecules and gels: (1) 30 mg/ml nonoxynol-9 (N9), contained in 1% sodium alginate gel cross-linked with 2.5 mM calcium chloride, permeating standardized, reconstituted bovine cervical mucus (BCM); (2) 2.5 mg/ml sodium fluorescein, contained in and permeating 10 mg/ml and 100 mg/ml gelatin gels; and (3) 1.0 mg/ml sodium ganciclovir, contained in and permeating 10 mg/ml sodium hyaluronate gel. Diffusion coefficients for (1) and (3) were 3.8x10-7 and 54.1x10-7 cm2/s, respectively. All measurements of diffusion coefficients, partition coefficients, and solute loads obtained in this study were highly repeatable (most C.V.'s<8%). The mean diffusion coefficient for (2) was within 3% of values predicted from theory for the 100 mg/ml gel; the mean partition coefficient for (3) was within 2% of the expected value. This new technique is simpler than many traditional ones in that it does not require labeling of test molecules nor changes in refractive index of target materials. It is particularly well-suited to situations in which the target material is a gel, because no stirring of the target is necessary.