Written informed consent was obtained in all patients who participated in this study. An admission blood sample was provided by patients followed by serial samples at 1, 4, 12, and 24 h, then once daily until discharge or death, as allowed by clinical factors. Blood was collected into an EDTA tube which was promptly centrifuged and the plasma was removed and frozen at −23 °C until the time of analysis. Ethics approval for this observational study was obtained

from Sri Lanka (the Universities of Colombo, Peradeniya and Sri Lankan Medical Association) and the grantholder’s universities (Oxfordshire Clinical Research Ethics Committee (UK) and Australian Gefitinib datasheet National University). The total and free NU7441 (unbound) concentrations of MCPA were quantified in the samples collected above in addition to admission samples collected for a previous study (Roberts et al., 2005). The total MCPA concentration was measured in the above-mentioned plasma samples. 300 μL of plasma was then ultrafiltered using Millipore Centrifree Micropartition Device® (Millipore, Bedford, MA, USA) yielding approximately 100 μg of plasma ultrafiltrate. The concentration of MCPA in the ultrafiltrate is the free (unbound) concentration. The concentrations of MCPA were determined by Queensland Health

Scientific Services (Australia) using a method derived from that of the United States Environmental Protection Agency (EPA, 1980). 100 μg of plasma or ultrafiltrate was hydrolysed in diluted sodium hydroxide and then buffered with acetic acid. The concentration of MCPA was determined by HPLC–MS/MS using an AB/Sciex API4000Q mass spectrometer in the negative ion mode equipped with an electrospray (TurboV) interface. This was coupled to a Shimadzu Prominence HPLC system (Shimadzu Corp., Kyoto, Japan) and a 50 mm × 2 mm C6-phenyl column (Phenomenex, Torrance, CA). The limit of reporting for the LCMSMS method was 1 μg/L for MCPA and the method was linear from at least 1–300 μg/L. Method recovery was confirmed using MPCA concentrations of 2.5 mg/L to around 300 mg/L with an average recovery of 105% and a standard deviation 0.25. Therefore, the limit

of detection (LOD; 3× standard Thiamine-diphosphate kinase deviation) is 0.75 mg/kg and the limit of reporting (LOR; using 6× LOD) is 4.5 mg/kg. The resulting concentrations (mg/kg plasma) were multiplied by 1.0205 which is the specific gravity of plasma at 37 °C (Trudnowski and Rico, 1974) to allow reporting with the unit mg/L. To validate the Centrifree® ultrafiltration device, plasma from a patient with MCPA poisoning was ultrafiltered, analysed for MCPA, re-ultrafiltered and then re-analysed for MCPA. There was no change in the concentration of MCPA between the 2 ultrafiltrates so MCPA does not appear to be adsorbed to the ultrafiltration device. Further, control plasma which did not contain MCPA was ultrafiltered and the filtrate was analysed for protein. It was noted that <0.