Evaluation of propylene glycol methyl ether as a potential challenge agent for leak detection of liquid and headspace from closed system drug transfer devices using Fourier transform infrared spectroscopy

Abstract

Choosing an appropriate surrogate of hazardous drugs for use in testing Closed System Drug-Transfer Devices (CSTDs) is a challenging endeavor with many factors that must be considered. It was suggested that the compound propylene glycol methyl ether (PGME) may meet many of the criteria we considered important in a suitable surrogate. Criteria included sufficient volatility to evaporate from aqueous liquid leaks efficiently, a Henry's constant which produced sufficient vapor phase concentrations to make headspace leaks detectable, and suitability for detection using a low-cost detection system. We evaluated the measurement of vapors from solutions containing PGME released inside a closed chamber. We present data used to quantify limits of detection, limits of quantification, bias, precision, and accuracy of Fourier Transform Infrared Spectroscopy (FTIR) measurements of vapors from 2.5 M PGME solutions. The effects of ethanol as a component of the PGME solution were also evaluated. Liquid drops of PGME solutions and headspace vapors above PGME solutions were released to simulate leaks from CSTDs. Using a calibration apparatus, an instrumental limit of detection (LOD) of 0.25 ppmv and a limit of quantitation (LOQ) of 0.8 ppmv were determined for PGME vapor. A LOD of 1.1 μL and a LOQ of 3.5 μL were determined for liquid aliquots of 2.5 M PGME solution released in a closed chamber. Accurate quantitation of liquid leaks required complete evaporation of droplets. With the upper end of the useable quantitation range limited by slow evaporation of relatively large droplets and the lower end defined by the method LOQ, the method evaluated in this research had a narrow quantitative range for liquid droplets. Displacement of 45 mL of vial headspace containing PGME vapor is the largest amount expected when using the draft NIOSH testing protocol. Release of an unfiltered 45 mL headspace aliquot within the NIOSH chamber was calculated to produce a concentration of 0.8 ppmv based on the Henry's constant, which is right at the instrumental LOQ. Therefore, the sensitivity of the method was not adequate to determine leaks of PGME vapor from a headspace release through an air filtering CSTD when using the draft NIOSH testing protocols with an FTIR analyzer.