Effect of Calcium Carbonate on Bioavailability of Orally Administered Gemifloxacin

We investigated the effect of calcium carbonate on the oral bioavailability of gemifloxacin. Gemifloxacin was administered alone, 2 h before, simultaneously, or 2 h after calcium carbonate in 16 volunteers. Data for 320 mg of gemifloxacin alone were as follows: maximum concentration of drug in serum ( C max ),13 (cid:1) g/ml; half-life, 7.33 h; and area under the concentration-time curve from 0 h toinfinity (AUC (cid:2) ), 6.79 (cid:1) g (cid:1) h/ml. Only simultaneous coadministration of calcium carbonate reduced C max ( (cid:3) 17%) and AUC (cid:2) ( (cid:3) 21%) significantly. Gemifloxacin is a new fluoroquinolone with a broad spec-trum activity against gram-positive and gram-negative bacteria. Plasma samples were analyzed following protein precipitation with acetonitrile. The urine sam- ples were prepared for analysis by dilution and analyzed by high-pressure liquid chromatography and mass spectrometry, using positive-ion spray ionization. The lower limits of quantification for each of the assays for gemifloxacin in plasma and urine were each 0.01 (cid:3) g/ml using a 50- (cid:3) l aliquot. The intra-assay and interassay coefficients of variation for determination of plasma and urine were less than 10%. Pharmacokinetic calculations. Plasma concentration data were analyzed by standard noncompartmental methods using the software program WinNonlin- Professional Edition (version 2.1). The primary pharmacokinetic parameters to determine the effect of calcium carbonate on the bioavailability of gemifloxacin were the maximal concentration in plasma ( C max ) and the area under the plasma concentration-time curve extrapolated to infinity (AUC (cid:4) ). Primary parameters were log e -transformed and compared by analysis of variance using the factor sequence, subject nested within sequence, period, and regimen. The point estimates and adjusted 95% confidence intervals for the difference between the regimens (B-A, C-A, and D-A) were constructed using the residual variance from the model, and the appropriate adjustments were made for the three multiple comparisons using Dunnett’s procedure (4). The overall significance level was maintained at 5%. The point and interval estimates on the log scale were then back-transformed to obtain the estimates of the ratios (B:A, C:A, and D:A). Within subject coefficient of variation for AUC (cid:4) and C max were calculated based on the log e -normal distribution.