End-tidal CO2 is the concentration of carbon dioxide in exhaled air at the end of expiration. It is typically expressed as a partial pressure in mm Hg (PETCO2 ). Because CO2 is a trace gas in atmospheric air, CO2 detected by capnography in exhaled air is produced in the body and delivered to the lungs by circulating blood. Under normal conditions PETCO2 is in the range of 35 to 40 mm Hg. During untreated cardiac arrest CO2 continues to be produced in the body, but there is no CO2 delivery to the lungs. Under these conditions PETCO2 will approach zero with continued ventilation.
With initiation of CPR, cardiac output is the major determinant of CO2 delivery to the lungs. If ventilation is relatively constant, PETCO2 correlates well with cardiac output during CPR. The correlation between PETCO2 and cardiac output during CPR can be transiently altered by giving IV sodium bicarbonate.208 This is explained by the fact that the bicarbonate is converted to water and CO2 , causing a transient increase in delivery of CO2 to the lungs. Therefore, a transient rise in PETCO2 after sodium bicarbonate therapy is expected and should not be misinterpreted as an improvement in quality of CPR or a sign of ROSC.
Animal and human studies have also shown that PETCO2 correlates with CPP and cerebral perfusion pressure during CPR.209,210 The correlation of PETCO2 with CPP during CPR can be altered by vasopressor therapy, especially at high doses (ie, .1 mg of epinephrine).211–214 Vasopressors cause increased afterload, which will increase blood pressure and myocardial blood flow during CPR but will also decrease cardiac output. Therefore, a small decrease in PETCO2 after vasopressor therapy may occur but should not be misinterpreted as a decrease in CPR quality.
Persistently low PETCO2 values (,10 mm Hg) during CPR in intubated patients suggest that ROSC is unlikely.171,173,174,190,191,215,216 Similar data using quantitative monitoring of PETCO2 are not available for patients with a supraglottic airway or those receiving bag-mask ventilation during CPR. One study using colorimetic end-tidal CO2 detection in nonintubated patients during CPR found that low end-tidal CO2 was not a reliable predictor of failure to achieve ROSC.217 An air leak during bag-mask ventilation or ventilation with a supraglottic airway could result in lower measured PETCO2 values. Although a PETCO2 value of ,10 mm Hg in intubated patients indicates that cardiac output is inadequate to achieve ROSC, a specific target PETCO2 value that optimizes the chance of ROSC has not been established. Monitoring PETCO2 trends during CPR has the potential to guide individual optimization of compression depth and rate and to detect fatigue in the provider performing compressions.201,218,219 In addition, an abrupt sustained increase in PETCO2 during CPR is an indicator of ROSC.91,177,196,198–201 Therefore, it is reasonable to consider using quantitative waveform capnography in intubated patients to monitor CPR quality, optimize chest compressions, and detect ROSC during chest compressions or when rhythm check reveals an organized rhythm (Class IIb, LOE C). If PETCO2 is ,10 mm Hg, it is reasonable to consider trying to improve CPR quality by optimizing chest compression parameters (Class IIb, LOE C). If PETCO2 abruptly increases to a normal value (35 to 40 mm Hg), it is reasonable to consider that this is an indicator of ROSC (Class IIa, LOE B).
