Comment on: pharmacokinetics of inhaled colistin in patients with cystic fibrosis.

Sir, Ratjen et al. recently published an important paper in JAC reporting the pharmacokinetics of colistin after inhalation of colistin methanesulphonate in patients with cystic fibrosis (CF). HPLC was used to measure the concentrations of the formed colistin. The paper raises a number of important issues that deserve comment. It is important not to use the terms colistin and colistin methanesulphonate (sodium salt, or colistimethate-Na in the paper) interchangeably. The chemistry, pharmacokinetics, pharmacodynamics and toxicity of these two entities are substantially different; most importantly, colistin is formed from colistin methanesulphonate both in vitro and in vivo. Even though some manufacturers (such as Grünenthal, Germany) label the content of injection vials of colistin methanesulphonate with ‘158 mg colistin methanesulphonate is equal to 66 mg colistin’, this labelling is misleading because it is based on antibacterial activity measured with microbiological assays and does not mean that there is 66 mg of colistin in each vial. Therefore, we strongly recommend that any pharmacokinetic calculations on colistin or colistin methanesulphonate should not be based on ‘66 mg colistin’. The HPLC assay used by Ratjen et al. was very similar to the one reported previously from another laboratory using derivatization with 9-fluorenylmethyl chloroformate on solid-phase extraction C18 cartridges; in the previously reported HPLC assay, colistin sulphate was employed to prepare calibration curves for measurement of colistin (base) in biological fluids. In the paper of Ratjen et al., it is not clear whether colistin sulphate, colistin A or colistin methanesulphonate (sodium) was used for the calibration curves of the HPLC assay. Ratjen et al. reported the amount (0.18–16.13 mg) of ‘colistin methate’ recovered in urine was equal to a mean 4.3 – 1.3% (range 0.3–24.2%) of the inhaled dose. Unfortunately, it is not clear how the concentrations of ‘colistin methate’ in urine were measured. It appears that the urinary recovery was calculated based on the recovered amount of colistin (base) and the dose of ‘66 mg colistin’. Considering the hydrolysis of colistin methanesulphonate to colistin in vivo and in vitro, such calculation of urinary recovery using the dose of ‘66 mg colistin’ is inappropriate. In addition, it is unclear whether the authors have allowed for the molecular weight difference between colistin methanesulphonate (sodium, average molecular weight 1743), colistin sulphate (average molecular weight 1403, if they used it for calibration curves for colistin) and colistin (average molecular weight 1163) when the concentrations of ‘colistin’ were calculated. Ratjen et al. used the HPLC peak for polymyxin E1 to generate calibration curves for colistin concentrations. Usually, the two major components of colistin, colistin A (polymyxin E1) and colistin B (polymyxin E2), account for more than 85% of colistin, but the ratio of these two components may vary substantially between 3:1 and 1:1. It is not clear how the authors calculated the concentrations of polymyxin E1 in the HPLC assay as the commercially available colistin sulphate is a mixture of many components. In addition, we are not aware of any literature indicating that polymyxin E2 has no antibacterial activity; hence, the claim by Ratjen et al. that ‘colistin is a pro-drug that is converted into its active component polymyxin E1’ is questionable. Indeed, polymyxin E1 is one of the constituents of colistin. It should be noted that the clearance of colistin reported by Ratjen et al. should be regarded as an apparent clearance because the amount of the inhaled dose of colistin methanesulphonate reaching the systemic circulation in the form of colistin is unknown. In summary, Ratjen et al. provided some potentially useful information on the pharmacokinetics of colistin after inhalation of colistin methanesulphonate. However, because of lack of clarity regarding the quantification of colistin in biological fluids and the pharmacokinetic analysis, caution is required when these data are used as reference for inhalation therapy of colistin methanesulphonate in CF patients.