The cytotoxicity and genotoxicity of nitromethane and its halogenated analogues in mammals raise concerns about potential toxicity to humans. This study shows that halonitromethanes are not stable in human blood and undergo dehalogenation to form nitromethane. We quantified nitromethane in human blood using solid-phase microextraction (SPME) headspace sampling coupled with gas chromatography (GC) and high resolution mass spectrometry (HRMS). The limit of detection was 0.01 microg/L with a linear calibration curve spanning 3 orders of magnitude. This method employs isotope dilution to precisely quantify trace amounts of nitromethane (coefficient of variation <6%). At three spiked concentrations of nitromethane, method accuracy ranged from 88 to 99%. We applied this method to blood samples collected from 632 people with no known occupational exposure to nitromethane or halonitromethanes. Nitromethane was detected in all blood samples tested (range: 0.28-3.79 microg/L, median: 0.66 microg/L). Time-course experiments with trichloronitromethane- and tribromonitromethane-spiked blood showed that nitromethane was the major product formed (1 nmole tribromonitromethane formed 0.59 nmole of nitromethane, whereas 1 nmole trichloronitromethane formed 0.77 nmole nitromethane). Nitromethane may form endogenously from peroxynitrite: nitromethane concentrations increased proportionately in blood samples spiked with peroxynitrite. Blood nitromethane can be a biomarker of exposure to both nitromethane and halonitromethanes. This sensitive, accurate, and precise analytical method can be used to determine baseline blood nitromethane level in the general population. It can also be used to study the health impact from exposure to nitromethane and halonitromethanes in occupational environments and to assess trichloronitromethane (chloropicrin) exposure in chemical terrorism investigations.