Can breath isoprene be measured by ozone chemiluminescence?

Isoprene, involved in the biosynthetic pathway to cholesterol, is the prevalent hydrocarbon in breath. Breath isoprene measurement is of great interest as a measure of basal cholesterol production rate. We investigated the merits and pitfalls of isoprene measurement via its chemiluminescence (CL) reaction with ozone. For many subjects, apparent concentrations measured are higher than those obtained by a gas chromatography (GC) reference method that can be traced to ozone-induced CL with simultaneously present lower olefins and sulfur compounds. A warm column preconcentration method eliminates the lower olefins and greatly improves sensitivity while a silver-form, ion-exchange resin can remove the sulfur gases. The breath sample is captured on a miniature synthetic carbon sorbent column maintained at 55 degrees C, under which conditions ethylene, propylene, and water vapor are not significantly captured while the preconcentration process greatly improves the limit of detection for isoprene to 0.6 ppbv (S/N=3). The captured isoprene is released by heating the column to 150 degrees C. Breath samples from different subjects were collected both before and after meals and analyzed in a double-blind fashion in two laboratories, with the second laboratory performing quantitation by cryofocusing GC-flame ionization detection with parallel measurement by mass spectrometry to provide identity confirmation. For all individuals studied, the CL and the GC results agreed when both warm column preconcentration and passage through Ag+-form cation-exchange resin, which removes divalent sulfur gases, were implemented prior to CL measurement. The intensity of CL from the reaction with ozone can be much higher for some sulfur gases than for isoprene. Even though present at lower concentrations than isoprene, unless removed prior to CL measurement, for some individuals sulfur gases can cause unacceptably large (up to 500%) errors, making the sulfur gas removal step critical.