Carbon monoxide reduces near-infrared spectroscopy determined 'total' hemoglobin: a human volunteer study

Carbon monoxide (CO) increases middle cerebral artery mean flow velocity (MCAVmean), but the effect of CO on the near-infrared spectroscopy (NIRS) determined cerebral oxygenation (ScO2) is not detailed. In our study, 11 non-smoking subjects breathed 100% O2 through a closed circuit. A CO2 scrubber with CO (1.5 mL kg-1) was added to the circuit. Two NIRS systems (NIRO-200NX and INVOS-5100) assessed ScO2 as the ratio of oxygenated to deoxygenated hemoglobin, while venous blood samples were analyzed for carboxyhemoglobin (COHb). After CO/O2 rebreathing COHb increased to 8.7% (IQR; 7.9-9.4; p = .004) vs. normoxia, but MCAVmean remained stable (55.6 cm s-1; 53.1-69.7) compared to inhalation of O2 (54.6 cm s-1; 48.4-62.9; p = .178) and normoxia (54.1 cm s-1; 44.5-66.9; p = .055). Also, INVOS-5100 determined ScO2 increased during CO/O2 (74.4 ± 7.5%) and O2 inhalation (73.1 ± 7.2%) compared to normoxia (68.9 ± 6.9%; p < .001). In contrast, NIRO-200NX determined ScO2 remained unchanged during CO/O2 and O2 inhalations but oxygenated and deoxygenated hemoglobin decreased (by 19.7 μM (median; IQR 2.8-34.8; p = .016) and 37.3 μM (30.8-46.6; p = .004), respectively) during inhalation of CO/O2 compared to inhalation of O2. Therefore, NIRO-200NX determined 'total' hemoglobin (sum of O2Hb and HHb) decreased (by 62.1 μM; 44.5-78.2; p = .001). In conclusion, exposure to CO did not increase MCAVmean, and neither NIRO-200NX nor INVOS-5100 detected a change in ScO2 when CO was added to inhalation of oxygen. Unaffected ScO2 after exposure to CO reflected a similar decrease in oxygenated and deoxygenated hemoglobin suggesting that detection of exposure to CO by NIRS should focus on 'total' hemoglobin rather than on ScO2.