As one of the most abundant toxic contaminants in the atmosphere, carbon monoxide (CO) plays a significant role in toxicology and public health. Every year, around half of the accidental non-fire-related poisoning deaths are attributed to CO in the USA, UK and many other countries. However, due to the non-specificity of the symptoms and often encountered inconsistency of these with the results obtained from measurements of the biomarker for CO poisonings, carboxyhemoglobin (COHb), there is a high rate of misdiagnoses. The mechanism of toxicity of CO includes not only the reduced transport of oxygen caused by COHb but also the impairment of cellular respiration and activation of oxidative metabolism by binding to other proteins. Therefore, in this study we propose the measurement of the total amount of CO in blood (TBCO) by airtight gas syringe-gas chromatography-mass spectrometry (AGS-GC-MS) as an alternative to COHb for the determination of CO exposures. The method is validated for a clinical range with TBCO concentrations of 1.63-104 nmol/mL of headspace (HS) (0.65-41.6 μmol/mL blood). The limit of quantification was found between 2 and 5 nmol/mL HS (0.8 and 2 μmol/mL blood). The method is applied to a cohort of 13 patients, who were exposed to CO under controlled conditions, and the results are compared to those obtained by CO-oximetry. Furthermore, samples were compared before and after a "flushing" step to remove excess CO. Results showed a significant decrease in TBCO when samples were flushed (10-60%), whereas no constant trend was observed for COHb. Therefore, measurement of TBCO by AGS-GC-MS suggests the presence of more dissolved CO than previously known. This constitutes a first step into the acknowledgment of a possibly significant amount of CO present not in the form of COHb, but as free CO, which might help explain the incongruences with symptoms and decrease misdiagnoses.