Article Figures & Data
Figures
- FIG 1
Observed colistimethate and colistin plasma concentration data in the three studies included in the model development (6, 7). The data from the current study are shown in the two right panels. inf, infusion; MU, million IU; q8h, every 8 h; q12h, every 12 h. (Reprinted from reference 1 with permission of the publisher.)
- FIG 2
Final structural model describing disposition of CMS and colistin. CMS (CMS1) and its derivatives (CMS2) occur in two states and distribute in both a central and peripheral compartment. For the previous studies (6, 7), a portion of the CMS dose (F2) was estimated to enter in the form of derivatives, while F2 was negligible for the new study. In addition, F1 for the new study was higher than F1 plus F2 for the previous studies. Vc, central volume of distribution; Vp, peripheral volume of distribution.
- FIG 3
Prediction-corrected visual predictive check for the final model. Dots represent the observed data, and lines represent the 5th, 50th, and 95th percentiles of the observed data. The shaded areas are the 95% confidence intervals of the 5th, 50th, and 95th percentiles as calculated from 500 simulations from the final PK model. LD, loading dose; MD, maintenance dose. Data from studies 1 and 2 were collected earlier (6, 7), and study 3 is the current study.
- FIG 4
Predicted profiles of colistimethate (top) and colistin (bottom) following the first dose and on day 5 (steady state) for the three dosing regimens available to generate the population PK model. In all predictions, an infusion time of 30 min was applied, and CLCR was set to 80 ml/min. F1 and F2 are the fractions of the administered dose that enters the CMS compartments as A or B forms (Fig. 2). F3 is the corresponding fraction directly entering the colistin compartment (applied in the 4th column). The two left columns show the predictions based on the estimated fractions of F1 and F2 in the current study, and the new formulation is set to include 100% A or B forms of CMS. The old formulation is set to include the estimated relative percentage of 61% A or B forms of CMS. The two right columns are two hypothetical scenarios where a part of the dose enters the CMS2 compartment (F2 > 0; 3rd and 4th columns) or colistin compartment (F3 > 0; 4th column).
Tables
- TABLE 1
Patient characteristicsa
No. Gender Age (yr) BW (IBWb; kg) Dose (MU; q12h) Serum CR (mg/dl) CLCRc (ml/min) Serum Alb (g/dl) APACHE II score Reason for colistin use Pathogen or purpose 1 F 29 100 (45) 4.5 0.6 98 2.5 10 cUTI K. pneumoniae 2 F 54 65 (50) 4.5 0.6 85 2.8 31 VAP A. baumannii 3 M 77 70 (75) 4.5 0.4 153 2.8 20 VAP A. baumannii 4 F 46 60 (55) 4.5 0.7 87 2.4 19 VAP P. aeruginosa 5 M 77 60 (70) 2.0 2.1 29 2.5 29 VAT Citrobacter spp. 6 F 66 60 (50) 3.0 0.9 48 3.3 23 VAT Empirical use 7 M 67 95 (80) 4.5 1 81 3.0 21 cIAI Empirical use 8 M 34 75 (80) 4.5 0.5 220 3.0 16 VAP P. aeruginosa 9 F 63 70 (60) 3.0 1.4 39 2.5 25 VAP P. aeruginosa 10 F 62 75 (60) —d 0.4 138 2.9 22 BSI K. pneumoniae 11 F 43 65 (60) 4.5 0.6 114 2.8 14 VAT A. baumannii 12 M 66 120 (70) 5.0 0.7 70 3.1 22 VAT A. baumannii 13 M 36 100 (85) 4.5 0.6 204 2.0 11 Necrotizing fasciitis A. baumannii 14 M 61 95 (60) 3.0 1.6 41 3.0 20 VAP A. baumannii 15 F 81 110 (60) 3.0 1.1 38 2.7 14 VAP A. baumannii 16 M 86 80 (65) 4.5 1.1 44 2.6 22 CRBSI A. baumannii 17 M 18 50 (55) 4.5 1 85 2.9 20 BSI K. pneumoniae 18 M 80 70 (65) 4.5 0.9 60 4.7 10 VAT P. aeruginosa 19 M 21 55(65) 4.5 0.8 115 2.0 9 cIAI P. aeruginosa ↵a Abbreviations: Alb, albumin; BSI, bloodstream infection; IBW, ideal body weight; cIAI, complicated intra-abdominal infection; CR, creatinine; CRBSI, catheter-related bloodstream infection; cUTI, complicated urinary tract infection; q12h, every 12 h; F, female; M, male; VAP, ventilator-associated pneumonia; VAT, ventilator-associated tracheobronchitis.
↵b IBW was calculated based on the formula (for males) IBW = 50 kg + 2.1 kg for each inch (2.54 cm) over 5 feet (152.4 cm) and (for females) IBW = 45.5 kg + 2.1 kg for each inch over 5 feet.
↵c For the estimation of creatinine clearance, ideal body weight was used and was calculated according to the Cockcroft-Gault formula (10).
↵d In patient 10, samples were collected only after the loading dose because the administration of CMS was discontinued.
- TABLE 2
Estimated population pharmacokinetic parameters of CMS and colistina
Parameter (unit) Explanation CMS Colistin Typical value IIV (%) IOV (%) Typical value IIV (%) IOV (%) CLNR,CMS (liters/h) Nonrenal clearance of CMS1 and CMS2 or of formed colistin (CL/fm) 5.84 (11) 16 (37) 40 (13) 4.99 (25) 71b (37) 41 (14) SlCRCL Proportionality factor for renal CL of CMS1 and CMS2; CLR (liters/h) = SlCRCL × CRCL (liters/h) 0.541 (16) 16 (37) 40 (13) V1 (liters) Volume of distribution of central compartment for CMS1 and for CMS2 or of formed colistin (V/fmcol) 1.42 (13) 80.4 (11) 41 (14) Q1 (liters/h) Intercompartmental clearance for CMS1 550 (31) Q2 (liters/h) Intercompartmental clearance for CMS2 7.75 (11) V2 (liters) Volume of distribution of peripheral compartment for CMS1 and CMS2 12.5 (10) 30 (13) Fstudy 1 and 2 Relative availability of CMS A and B in earlier studies compared to the current work 0.610 (7.2) F1study 1 and 2 Fraction of the available CMS dose entering CMS1 compartment for earlier studies 0.892 (4.4) Additive residual error (μmol/liter) 0.159 (10) 0.0629 (14) 48 (16) Proportional error (%) 0.157 (6.7) 0.0884 (13) 48 (16) ↵a Shown are the estimated population pharmacokinetic parameters of CMS and colistin and their percent relative standard errors (RSE) for the final model based on simultaneous fit of the data from the current and previous studies in critically ill patients (6, 7). IIV, interindividual variability; IOV, interoccasion variability.
↵b The value obtained when the variability of CL/fm was scaled from the variability of CLCMS by the estimated value of 4.52 (41% RSE).
