Improving antimicrobial prescribing through knowledge and skills.

she developed a severe sepsis with acute renal failure and died a few days after the amputation. Two pairs of synovial fluid and serum samples (trough levels) were collected before the second dose of 400 mg of vori-conazole (i.e. Day 1 of treatment), and before the second dose of 300 mg (i.e. Day 2 of treatment). Samples were immediately frozen at 2 208C until analysis. Cortical and medullar bone samples were collected during the amputation (i.e. Day 6 of treatment). The patient medication did not involve drugs that could generate pharmacokinetic interactions with voriconazole. The patient had normal liver function and had a calculated creatinine clearance of 60 mL/min. The serum albumin concentration was low at 11.6 g/L (normal range: 38 – 46 g/L). The measurement of voriconazole concentrations was performed using high-performance liquid chromatography coupled with a diode array detector method, as previously described 5 and applied in our pharmacology department. 6 For bone analysis , as there is no published analytical method for voriconazole assay in bone, previously published procedures for extraction in bone tissues were used to develop the assay method. 7,8 The cortical bone samples were cut into small pieces. After addition of the analytical internal standard, 2 mL of 2 M acetic acid was added to 200 mg of cortical or medullar bone sample. The mixture was vigorously shaken for 10 min, boiled for 10 min and subsequently lyophilized. The stability of voriconazole through these extraction steps was verified. After addition of 500 mL of a saturated NH 4 Cl/deionized water mixture (30/70, v/v) adjusted to pH 9.5 with 25% NH 4 OH, voriconazole was assayed following the same procedure as for liquid samples. Calibration samples were obtained using voriconazole-free cortical or medullar bone samples and by means of appropriate addition of voriconazole solutions in order to obtain the following spiked bone sample concentrations: 0, 0.25, 1, 5, 10, 20 and 40 mg/g. The calibration curves were linear from 0.25 to 40 mg/g and the inter-assay precision coefficients of variation were lower than 15%. Voriconazole concentrations in serum and synovial fluid were 2.41 and 0.76 mg/L on Day 1 and 4.09 and 1.07 mg/L on Day 2, respectively. Observed bone concentrations were 20.3 mg/g of tissue in the medullar bone and 1.9 mg/g of tissue in the cortical bone. The serum concentrations were higher than those usually observed for this dosing. 9,10 One explanation could …