ICU Management & Practice, Volume 22 - Issue 2, 2022

Understanding Carbon Dioxide in Resuscitation

Peri-arrest CO2 measurement may serve to assess resuscitation efficacy and guide clinical decisions. This paper reviews current knowledge and future research directions for CO2 measurement and clinical application during CPR.

Carbon Dioxide During Normal Heart-Lung Function

CO2 is produced by the mitochondria as a major end product of tissue aerobic respiration. Approximately 70% of CO2 produced by the mitochondria undergoes a chemical reaction with water catalysed by carbonic anhydrase to form H2CO3, dissolved in the plasma as its component ions, HCO3- and H+. Another 23% of CO2 produced by the mitochondria binds to haemoglobin to form carbaminohaemoglobin. The remaining 7% is dissolved directly in the plasma. CO2 is highly soluble in the blood and therefore has a high diffusion coefficient. The dissolved CO2 is transported to the lung, thus maintaining the mixed venous partial pressure of CO2 (PmvCO2) – as measured in the pulmonary artery – at around 45 mmHg.

In the healthy individual with normal cardiac output and lung physiology, pulmonary ventilation matches pulmonary perfusion. This allows maintenance of the partial pressure of CO2 in the alveoli (PACO2) at approximately 40 mmHg, thus maintaining a CO2 diffusion gradient between the pulmonary capillaries and alveoli of 5 mmHg. The high diffusion coefficient of CO2 and efficient alveolar perfusion and diffusion across the alveolar membrane in the patient with normal heart and lung function result in an arterial partial pressure of CO2 (PaCO2) which approximates the PACO2 – about 40 mmHg.

As direct measurement of PACO2 is complex, PACO2 is usually evaluated indirectly via the end-tidal CO2 (PetCO2). In patients with normal heart and lung function, the PetCO2 is usually less than 5 mmHg below the PACO2. Thus, in the healthy patient, the PetCO2 is also less than 5 mmHg below the PaCO2 (Hall 2016).

Carbon Dioxide During Decreased Cardiac Output and the No-Flow State

Changes in cardiac output result in substantial changes in arterial and venous CO2 levels, as well as in the alveolar CO2.

A decrease in cardiac output decreases blood flow to both peripheral tissues and the lungs. In peripheral tissues, this results in less effective removal of CO2, leading to its accumulation in tissues and venous blood. In the lungs, decreased cardiac output decreases pulmonary perfusion pressures, resulting in a ventilation/perfusion (V/Q) mismatch. Thus, CO2 transport to the lungs is reduced, decreasing CO2 alveolar concentration (West 1974).

Cardiac arrest results in a no-flow state, with no organ perfusion despite ongoing cellular metabolic activity. These circumstances lead first to cellular hypoxia and subsequently to a transition to anaerobic cellular respiration, resulting in release of cellular by-products of this process to the extracellular space. These include carbon dioxide, lactate and hydrogen ions, thus resulting in a combined respiratory and metabolic acidosis (Ahn et al. 2011; Prause et al. 2001; Takasu et al. 2007). In the no-flow state of cardiac arrest, no changes in CO2 are initially noted, as the cellular by-products from unperfused tissue remain in-situ (Tucker et al. 1994).

Carbon Dioxide During the Low Flow State – Initiation of CPR

During CPR, chest compressions, along with positive pressure ventilation, restore organ perfusion and oxygenation to some extent. In ideal conditions, CPR can achieve as much as 25% of normal cardiac output, converting the no-flow state of cardiac arrest to a low-flow state (Bellamy et al. 1984; Johnson and Weil 1991; Link et al. 2015).

The low-flow state that occurs during CPR results in an increase in PmvCO2 – reflecting poor systemic perfusion. It also leads to a decoupling of the PetCO2 and the PaCO2, with a decrease in the former and an increase in the latter, both due to poor alveolar perfusion (De Backer et al. 2015; Nowak et al. 1987; Steedman and Robertson 1992). These can be expressed in the ventilation – perfusion ratio equation (Idris et al. 1994; West 1974).

Given that CPR effectiveness is directly related to patient survival (Ashoor et al. 2017; Sell et al. 2010) and that the CO2 changes noted during CPR can be partly reversed with effective CPR (Hartmann et al. 2015; Kim et al. 2016; Niemann et al. 1985; Ornato et al. 1985; Spindelboeck et al. 2016; von Planta et al. 1991; Yakaitis et al. 1975), the measurement of partial pressures of CO2 at various points in the systemic and pulmonary circulation, as well as measurement of alveolar CO2 as reflected by PetCO2, may be useful for optimising CPR performance, as well as for prognosticating CPR outcomes. Thus, a detailed understanding of CO2 changes in various compartments is important in assessing and optimising the quality of an ongoing resuscitation.

End-tidal CO2

When cardiac output is 5 litres per minute, as in the healthy, sedentary adult, the PetCO2 ranges between 36-40 mmHg. For every one litre decrease in cardiac output, this value decreases by 4-6 mmHg, assuming constant ventilatory conditions (Askrog 1966; Leigh et al. 1961; Maslow et al. 2001; Shibutani et al. 1994). Under constant and optimal ventilation conditions, the PetCO2 may therefore serve as an effective surrogate measure of pulmonary blood flow (Isserles and Breen 1991; Jin et al. 2000; Ornato et al. 1990). Consistent with this, animal and human studies have shown a direct correlation between quantitative waveform capnography pressure during resuscitation, cardiac output and coronary perfusion (Link et al. 2015).

Furthermore, a meta-analysis examining the relationship between PetCO2 values and resuscitation outcomes demonstrated that patients eventually achieving return of spontaneous circulation (ROSC) had a mean PetCO2 of 25.8 ± 9.8 mmHg versus a mean PetCO2 of 13.1 ± 8.2 mmHg in those not achieving ROSC (p=0.001). In contrast, administration of sodium bicarbonate, changes in minute ventilation and varying resuscitation protocols were not associated with PetCO2 changes (Hartmann et al. 2015).

Resultantly, American Heart Association and European Resuscitation Council guidelines recommend use of quantitative waveform capnography in all cardiopulmonary resuscitations in order to optimise chest compressions and identify ROSC (Kodali and Urman 2014; Link et al. 2015; Soar et al. 2021).

Arterial CO2

Arterial CO2 has been studied during CPR, immediately after ROSC and at a later stage, after ICU admission.

Multiple studies in both humans and animals have shown that the level of arterial and venous acidosis during resuscitation is determined mainly by partial pressures of CO2 (Angelos et al. 1992; Grundler et al. 1986; Martin et al. 1985; Nowak et al. 1987; Ornato et al. 1985; Sanders et al. 1988; Weil et al. 1986), with a more profound acidosis in venous versus arterial blood (Gabrielli et al. 2005; Martin et al. 1985; Nowak et al. 1987; Ralston et al. 1985).

Profound acidosis during CPR has been associated with resuscitation failure (Niemann et al. 1985; von Planta et al. 1991; Yakaitis et al. 1975), though some earlier studies have noted poor correlations between arterial blood gas values and tissue metabolism (Nowak et al. 1987; Steedman and Robertson 1992). Nevertheless, a study of 136 out-of-hospital cardiac arrest (OHCA) patients transported to the hospital during cardiac arrest showed less profound acidosis in blood samples obtained during CPR from patients who eventually achieved ROSC versus those who did not (pH=6.96 versus 6.85; P=0.009). PaCO2 and lactate levels were also lower in the former versus the latter group (74.0 versus 89.5 mmHg, P=0.009; 11.6 versus 13.6 mmol/L, P=0.044, respectively). Thus, PaCO2 during resuscitation may be a marker of ischaemia severity (Kim et al. 2016). Similarly, less acidosis and lower PaCO2 at emergency room arrival were noted in patients achieving ROSC in the pre-hospital setting versus those admitted to the emergency room in ongoing resuscitation (Ornato et al. 1985).

Greater arterio-alveolar CO2 difference (AaDCO2) has also been associated with resuscitation failure. A multicentre study examining AaDCO2 during or immediately post-CPR in OHCA showed an association between increased AaDC and failure to achieve sustained ROSC. No patients with an AaDCO2 greater than 33.5 mmHg during CPR achieved sustained ROSC (Spindelboeck et al. 2016). Similar findings were noted in those with an elevated AaDCO2 an hour after ROSC (Moon et al. 2007).

In contrast, in post-OHCA patients hospitalised in an ICU, lower levels of PaCO2 were associated with poorer prognosis (Helmerhorst et al. 2015; Schneider et al. 2013), and, in such patients, a relatively increased CO2 was associated with improved cerebral function (Roberts et al. 2013; Vaahersalo et al. 2014; Wang et al. 2015), possibly due to the injurious effects of hypocapnia on the brain (Aufderheide and Lurie 2004; Buunk et al. 1996; Coles et al. 2007; Schneider et al. 2013). However, it should be noted that hypercapnia, as well, has been associated with increased mortality in some studies (Roberts et al. 2013).

In conclusion, a higher PaCO2 during and immediately post-resuscitation appears to be associated with a poorer prognosis; whereas, in the later post-resuscitation period, a lower PaCO2 is associated with a poorer prognosis.

Pulmonary artery/mixed venous CO2

Though pulmonary artery sampling can provide information regarding tissue oxygen consumption in the tissues and cardiac output, use of pulmonary artery catheters has not been associated with improved patient outcomes and is therefore not recommended in most critically ill patients (Binanay et al. 2005; Harvey et al. 2005; Marik 2013; Rajaram et al. 2013; Richard et al. 2003) and is generally not possible nor recommended during CPR. Thus, PmvCO2 is generally not available during resuscitation.

In non-cardiac arrest hypoperfusion states (e.g. septic shock), Pv-aCO2 can assist in evaluation of cardiac output, tissue perfusion and anaerobic metabolic activity (Mallat et al. 2016), and increased values have been associated with increased mortality (Ospina-Tascon et al. 2013; Ospina-Tascon et al. 2015; van Beest et al. 2013).

Small retrospective studies showed substantial differences in PaCO2 and PmvCO2 during CPR due to OHCA (Nowak et al. 1987; Steedman and Robertson 1992). Indeed, PmvCO2 may differ substantially from PaCO2, as the latter reflects pulmonary gas exchange and former tissue perfusion, with differences being especially prominent in states of low cardiac output and especially during CPR (Bloom et al. 2014; Byrne et al. 2014; Kelly 2010; Spindelboeck et al. 2016). As PmvCO2 may more accurately reflect tissue perfusion it may be a better clinical decision tool than PaCO2 during resuscitation (Adrogue et al. 1989; Nowak et al. 1987; Steedman and Robertson 1992; Weil et al. 1986).

Indeed, multiple studies have noted an association between PmvCO2 levels and resuscitation outcomes (Niemann et al. 1985; von Planta et al. 1991; Yakaitis et al. 1975). Differences between PaCO2 and PmvCO2 (Pv-aCO2) may also be associated with prognosis, and thus may also serve as a useful tool during and after resuscitation (Ospina-Tascon et al. 2015).

Peripheral venous CO2

Peripheral blood sampling is a simple procedure and is the common practice in most departments of emergency medicine (Kelly 2016). As such, PvCO2 may be more useful clinically as an index of resuscitation efficacy than PmvCO2 . Several studies have also noted substantial correlation between PmvCO2 and PvCO2 (Abdelmoneim et al. 1999; Byrne et al. 2014). However, PvCO2 may be confounded by the location of sampling as there may be relative ischaemia of the tissue being drained by the sampled vein. Therefore PvCO2 may not accurately reflect global tissue perfusion (Abdelmoneim et al. 1999; Bloom et al. 2014; Byrne et al. 2014; Kelly 2010; Toftegaard et al. 2008). Unfortunately few data are available on the association between PvCO2 and other measures of CO2 during resuscitation, nor between PvCO2 and resuscitation outcomes.


In conclusion, measurements of CO2 during and after resuscitation, including PetCO2, PaCO2 and PmvCO2, may serve to assess resuscitation efficacy. Such measurements may therefore be useful adjuncts in clinical decision-making during resuscitation as they have been shown to correlate overall with patient prognosis. Given that PmvCO2 is rarely available during resuscitation, research is needed regarding the usefulness of substituting PvCO2 for PmvCO2. More research is needed into effective interventions towards modifying PCO2 indices during resuscitation and their possible effects on outcomes.

Conflict of Interest


«« electronic Poor Outcome Screening (ePOS) Score

Delirium Motor Subtypes in ICU Patients »»


Abdelmoneim T et al. (1999) Acid-base status of blood from intraosseous and mixed venous sites during prolonged cardiopulmonary resuscitation and drug infusions. Crit Care Med. 27(9):1923-1928.


Adrogue HJet al. (1989) Assessing acid-base status in circulatory failure. Differences between arterial and central venous blood. N Engl J Med. 320(20):1312-1316.


Ahn S et al. (2011)Potassium values in cardiac arrest patients measured with a point-of-care blood gas analyzer. Resuscitation. 82(12):e25-26.


Angelos MG et al. (1992) Arterial pH and carbon dioxide tension as indicators of tissue perfusion during cardiac arrest in a canine model. Crit Care Med. 20(9):1302-1308.


Ashoor HM et al. (2017) Effectiveness of different compression-to-ventilation methods for cardiopulmonary resuscitation: A systematic review. Resuscitation. 118:112-125.


Askrog V (1966) Changes in (aA) CO2 difference and pulmonary artery pressure in anesthetized man. Journal of applied physiology. 21(4):1299-1305.


Atwood C et al. (2005) Incidence of EMS-treated out-of-hospital cardiac arrest in Europe. Resuscitation. 67(1):75-80.


Aufderheide TP, Lurie KG (2004). Death by hyperventilation: a common and life-threatening problem during cardiopulmonary resuscitation. Crit Care Med. 32(9 Suppl):S345-351.


Bellamy RF et al. (1984). Coronary blood flow during cardiopulmonary resuscitation in swine. Circulation. 69(1):174-180.

Binanay C et al. (2005) Evaluation study of congestive heart failure and pulmonary artery catheterization effectiveness: the ESCAPE trial. JAMA. 294(13):1625-1633.


Bloom BM et al. (2014) The role of venous blood gas in the emergency department: a systematic review and meta-analysis. Eur J Emerg Med. 21(2):81-88.


Buunk G et al. (1996) Cerebral vasoconstriction in comatose patients resuscitated from a cardiac arrest? Intensive Care Med. 22(11):1191-1196.


Byrne AL et al. (2014) Peripheral venous and arterial blood gas analysis in adults: are they comparable? A systematic review and meta-analysis. Respirology. 19(2):168-175.


Callans DJ (2004) Out-of-hospital cardiac arrest--the solution is shocking. N Engl J Med. 351(7):632-634.


Coles JP et al. (2007) Hyperventilation following head injury: effect on ischemic burden and cerebral oxidative metabolism. Crit Care Med. 35(2):568-578.


De Backer D et al. (2015) Microcirculation Alterations in Patients With Severe Sepsis. Clinical Pulmonary Medicine. 22(1):31-35.


Gabrielli A et al. (2005) Physiology of ventilation during cardiac arrest. Cardiopulmonary Resuscitation. Humana Press. 41.


Grundler W et al. (1986) Arteriovenous carbon dioxide and pH gradients during cardiac arrest. Circulation 74(5):1071-1074.


Hall JE (2016) Guyton and Hall Textbook of Medical Physiology: Elsevier.


Hartmann SMet al. (2015) Systematic Review and Meta-Analysis of End-Tidal Carbon Dioxide Values Associated With Return of Spontaneous Circulation During Cardiopulmonary Resuscitation. J Intensive Care Med. 30(7):426-435.


Harvey S et al. (2005) Assessment of the clinical effectiveness of pulmonary artery catheters in management of patients in intensive care (PAC-Man): a randomised controlled trial. Lancet. 366(9484):472-477.


Helmerhorst HJet al. (2015) Associations of arterial carbon dioxide and arterial oxygen concentrations with hospital mortality after resuscitation from cardiac arrest. Crit Care. 19 348.


Idris AH et al. (1994) Effect of ventilation on acid-base balance and oxygenation in low blood-flow states. Crit Care Med. 22(11):1827-1834.


Isserles SA, Breen PH (1991) Can changes in end-tidal PCO2 measure changes in cardiac output? AnesthAnalg. 73(6):808-814.


Jin X et al. (2000) End-tidal carbon dioxide as a noninvasive indicator of cardiac index during circulatory shock. Crit Care Med. 28(7):2415-2419.

Johnson BA, Weil MH (1991) Redefining ischemia due to circulatory failure as dual defects of oxygen deficits and of carbon dioxide excesses. Crit Care Med. 19(11):1432-1438.


Kelly AM (2010) Review article: Can venous blood gas analysis replace arterial in emergency medical care. Emerg Med Australas. 22(6):493-498.


Kelly AM (2016) Can VBG analysis replace ABG analysis in emergency care? Emerg Med J. 33(2):152-154.


Kim YJ et al. (2016) Role of blood gas analysis during cardiopulmonary resuscitation in out-of-hospital cardiac arrest patients. Medicine (Baltimore). 95(25):e3960.


Kodali BS, Urman RD (2014) Capnography during cardiopulmonary resuscitation: Current evidence and future directions. J Emerg Trauma Shock. 7(4):332-340.


Leigh MD et al. (1961) The expired carbon dioxide as a continuous guide of the pulmonary and circulatory systems during anesthesia and surgery. The Journal of Thoracic and Cardiovascular Surgery. 41(5):597-610.


Link MS et al. (2015). Part 7: Adult Advanced Cardiovascular Life Support: 2015 American Heart Association Guidelines Update for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation. 132(18 Suppl 2):S444-464.


Mallat J et al. (2016) Use of venous-to-arterial carbon dioxide tension difference to guide resuscitation therapy in septic shock. World J Crit Care Med. 5(1):47-56.


Marik PE (2013) Obituary: pulmonary artery catheter 1970 to 2013. Ann Intensive Care. 3(1):38.


Martin GB et al. (1985) Comparison of central venous and arterial pH and PCO2 during open-chest CPR in the canine model. Ann Emerg Med. 14(6):529-533.


Maslow A et al. (2001)Monitoring end-tidal carbon dioxide during weaning from cardiopulmonary bypass in patients without significant lung disease. AnesthAnalg. 92(2):306-313.


Moon SW et al. (2007) Arterial minus end-tidal CO2 as a prognostic factor of hospital survival in patients resuscitated from cardiac arrest. Resuscitation. 72(2):219-225.


Niemann JTet al. (1985) Predictive indices of successful cardiac resuscitation after prolonged arrest and experimental cardiopulmonary resuscitation. Ann Emerg Med. 14(6):521-528.


Nowak RM et al. (1987). Selective venous hypercarbia during human CPR: implications regarding blood flow. Ann Emerg Med. 16(5):527-530.


Ornato JP et al. (1990) Relationship between cardiac output and the end-tidal carbon dioxide tension. Ann Emerg Med. 19(10):1104-1106.

Ornato JP, et al. (1985) Arterial pH in out-of-hospital cardiac arrest: response time as a determinant of acidosis. Am J Emerg Med. 3(6):498-502.


Ospina-Tascon GA et al. (2013) Persistently high venous-to-arterial carbon dioxide differences during early resuscitation are associated with poor outcomes in septic shock. Crit Care. 17(6):R294.


Ospina-Tascon GA et al. (2015) Combination of arterial lactate levels and venous-arterial CO2 to arterial-venous O2 content difference ratio as markers of resuscitation in patients with septic shock. Intensive Care Med. 41(5):796-805.


Prause G et al. (2001) Comparison of lactate or BE during out-of-hospital cardiac arrest to determine metabolic acidosis. Resuscitation. 51(3):297-300.


Rajaram SS et al. (2013) Pulmonary artery catheters for adult patients in intensive care. Cochrane Database of Systematic Reviews(2).


Ralston SH et al. (1985) Venous and arterial blood gases during and after cardiopulmonary resuscitation in dogs. Am J Emerg Med. 3(2):132-136.


Richard C et al. (2003) Early use of the pulmonary artery catheter and outcomes in patients with shock and acute respiratory distress syndrome: a randomized controlled trial. JAMA. 290(20):2713-2720.


Roberts BW et al. (2013) Association between postresuscitation partial pressure of arterial carbon dioxide and neurological outcome in patients with post-cardiac arrest syndrome. Circulation. 127(21):2107-2113.


Sanders AB et al. (1988) Acid-base balance in a canine model of cardiac arrest. Ann Emerg Med. 17(7):667-671.


Schneider AG et al. (2013) Arterial carbon dioxide tension and outcome in patients admitted to the intensive care unit after cardiac arrest. Resuscitation. 84(7):927-934.


Sell RE et al. (2010) Minimizing pre- and post-defibrillation pauses increases the likelihood of return of spontaneous circulation (ROSC). Resuscitation. 81(7):822-825.


Shibutani K et al. (1994) Do changes in end-tidal PCO2 quantitatively reflect changes in cardiac output? AnesthAnalg. 79(5):829-833.


Soar J et al. (2021) European Resuscitation Council Guidelines 2021: Adult advanced life support. Resuscitation. 161:115-151.


Spindelboeck Wet al. (2016) Arterial blood gases during and their dynamic changes after cardiopulmonary resuscitation: A prospective clinical study. Resuscitation. 106:24-29.


Steedman DJ, Robertson CE (1992) Acid base changes in arterial and central venous blood during cardiopulmonary resuscitation. Arch Emerg Med. 9(2):169-176.

Takasu A et al. (2007) Arterial base excess after CPR: the relationship to CPR duration and the characteristics related to outcome. Resuscitation. 73(3):394-399.


Toftegaard M et al. (2008) Correlation between acid-base parameters measured in arterial blood and venous blood sampled peripherally, from vena cavae superior, and from the pulmonary artery. Eur J Emerg Med. 15(2):86-91.


Tucker KJet al. (1994) Changes in arterial and mixed venous blood gases during untreated ventricular fibrillation and cardiopulmonary resuscitation. Resuscitation. 28(2):137-141.


Vaahersalo J et al. (2014) Arterial blood gas tensions after resuscitation from out-of-hospital cardiac arrest: associations with long-term neurologic outcome. Crit Care Med 42(6):1463-1470.


van Beest PA et al. (2013) Central venous-arterial pCO2 difference as a tool in resuscitation of septic patients. Intensive Care Med. 39(6):1034-1039.


von Planta I et al. (1991). Hypercarbic acidosis reduces cardiac resuscitability. Crit Care Med. 19(9):1177-1182.


Wang C-H et al. (2015) Association between early arterial blood gas tensions and neurological outcome in adult patients following in-hospital cardiac arrest. Resuscitation. 89:1-7.


Weil MH et al. (1986) Difference in acid-base state between venous and arterial blood during cardiopulmonary resuscitation. N Engl J Med. 315(3):153-156.


West JB (1974) Respiratory Physiology. The Essentials: Williams and Wilkins.

Yakaitis RW et al. (1975) Influence of pH and hypoxia on the success of defibrillation. Crit Care Med. 3(4):139-142.

CPR, cardiopulmonary resuscitation, resuscitation, carbon dioxide Understanding Carbon Dioxide in Resuscitation

No comment

Highlighted Products