The issue raised by Markou et al is extremely important: Does colistin penetrate significantly into lung tissue? The clinical implications are evident, but at the moment, there is not a definitive answer to this question. Li et al1 cite several studies in humans showing that IV colistin methanesulfonate (CMS) (also called colistimethate) is effective in the treatment of pneumonia caused by multidrug-resistant, gram-negative bacteria1 that indicate that colistin does indeed penetrate into lung tissue. In contrast, studies in animals and humans evaluating the concentrations of colistin or CMS in lung tissue have produced conflicting results. Albeit with methodologic weaknesses in the quantification of CMS and colistin, the pioneer studies performed by Ziv et al2 and Kunin and Bugg3 showed that after IV administration colistin penetrates and accumulates in lung tissue, with the concentration of the bound form being considerably higher than that of the free form. By contrast, Lu et al4 were unable to detect colistin in lung tissue after IV administration of CMS in their experimental model of pneumonia. The reasons for these conflicting results2-4 may lie in the different methods used for protein precipitation and CMS and colistin extraction from body tissues. Indeed, it has recently been demonstrated that the type and volume of precipitating agents used are critical for CMS and colistin recovery.5 In our study,6 we showed that colistin was undetectable in BAL 2 h after the start of an IV CMS infusion (Fig 1A), but it was present at a relevant concentration in the BAL of a patient who received CMS by aerosol (Fig 1B), suggesting significant distal lung deposition. Our findings can be interpreted as indicating either low tissue penetration or colistin tissue binding. In fact, as the result of its chemical properties, colistin (which is a polycation, while CMS is a polyanion) can bind to lung tissue, thus hindering its recovery during BAL.