Automated CPR devices are quickly gaining acceptance, especially since the 2005 American Heart Association (AHA) guidelines recognized that effective and uninterrupted cardiac compressions are critical to successful resuscitation.(1) These devices, however, do require a somewhat different approach in the use and evaluation of capnography in the cardiac arrest patient. In order to gain a better understanding of these differences, it’s important to first have an in-depth knowledge of the physiology of automated compressions and also of capnography.

A large volume of research has been generated over the past 15 years on the physiological effects of mechanical CPR, in particular as it relates to active compression-decompression cardiopulmonary resuscitation (ACD-CPR). The research is consistent in the finding that mechanical CPR generates a greater mean systolic and diastolic blood pressure, a higher coronary perfusion pressure limiting cardiac ischemia, higher cardiac output, a higher cerebral perfusion pressure and higher end-tidal carbon dioxide (EtCO2) values than standard manual compressions.(2-4)

Figures 1 and 2 demonstrate the increased coronary perfusion pressure and cortical cerebral blood flow provided by the LUCAS automated CPR device as compared with standard or manual CPR.

Capnography is directly reflective of the differences that occur with automated CPR devices. It’s well documented that EtCO2 has a direct relationship with cardiac output, which is logarithmic.(5) As cardiac output decreases or increases, pulmonary blood flow also decreases or increases. The EtCO2then decreases or increases relative to the change in pulmonary blood flow.

Table 1 shows the relationship of EtCO2 with cardiac output found in 24 patients undergoing aortic aneurysm surgery with constantly maintained ventilatory rates.(6)

Similar relationships have been found between EtCO2 and coronary perfusion pressure. One study noted that expired partial pressure of carbon dioxide (PCO2) positively correlated with coronary perfusion pressure and that inasmuch as coronary perfusion pressure correlates with survival in prolonged CPR, capnography may be a useful measure for non-invasively assessing the adequacy of CPR.(7) The correlation of capnography with cerebral perfusion pressure is somewhat less clear; although studies do show a relationship, it’s suggested that this is more than likely related to the EtCO2 response to changes in cardiac output.(8)

Impact of Automated Compressions
The first impact of automated compressions on capnography relates to the EtCO2 values. With manual compressions, the patient in cardiac arrest rarely achieves EtCO2 values greater than 18–20 mmHg even in the very early stages of resuscitation. These values then generally deteriorate even further as the rescuer tires. On the other hand, automated ACD-CPR often presents and maintains EtCO2 values in the 25–35 mmHg range. These values are then maintained for prolonged periods of time. One study demonstrated compression times that averaged 105 minutes with a range of 45–240 minutes.(9)

The second impact of automated cardiac compressions is related to the capnography waveforms. Because automated cardiac compressions create significant chest recoil during the decompression stage, air will move freely into the lungs in the presence of an open airway. This air movement is often significant enough to result in CO2 detection by capnography, and therefore observable on capnography waveforms. This also results in an observable respiratory rate that is high, often very near to the compression rate. The waveforms are uniform and consistent in size and shape and have a short plateau. With manually delivered ventilations, the waveforms are usually longer, differently shaped and less frequent (see Figure 3).

It’s very important to ensure the rescuer delivers assisted ventilations at a rate consistent with the AHA guidelines of eight to 10 breaths per minute. This very slow respiratory rate will prove beneficial to the patient by providing an improved neurological outcome in the event of survival.(10)

Finally, automated compressions will impact the determination of a return of spontaneous circulation (ROSC). Because the EtCO2 is low (5–20 mmHg) with manual cardiac compressions, when the patient has ROSC the EtCO2 increases to a more normal range. The change is rapid and dramatic, even preceding the detection of pulses.

With automated cardiac compressions, the EtCO2 range is already significantly higher, meaning when the patient achieves ROSC, the increase in EtCO2 is not as dramatic. The EtCO2 is already close to the normal range due to the higher cardiac output delivered, so it doesn’t need to increase at the same ratio. The higher cardiac output delivered by the automated CPR device may even result in detectable peripheral pulses during CPR. Because of this, the rescuer must be carefully observant for even a minimal increase in EtCO2 and evaluate the patient for the possibility of ROSC.

Despite or perhaps even because of these differences in capnography assessment with the use of the automated CPR device, it remains important to continuously observe the trend of the EtCO2 and capnography waveform. Changes should be carefully evaluated and correlated with clinical findings to determine not only the possibility of ROSC but also any evidence of other physiological changes.

A decreased EtCO2 may result from decreased cardiac output, which can alert the rescuer to a CPR device malfunction, blood loss or fluid third-spacing, which might also allow the rescuer to take appropriate corrective actions. A decreased EtCO2 may also be the result of an increase in ventilatory rate, which should be corrected expediently to prevent a negative impact on neurological outcome. An increase in EtCO2 could be the result of ROSC, but it also could result from hypoventilation. Observing the respiratory rate would allow the rescuer to make the needed adjustments.

Cardiac arrest patients sometimes experience bronchospasms, so the observation for the presence of, or change to, a shark-fin-shaped waveform will alert the rescuer to this development. Bronchodilator therapy can then be administered via the bag-valve mask (BVM) to improve ventilations.

Also, even with the use of an automated CPR device, capnography continues to be useful in the cardiac arrest patient in the same ways it is useful in the patient receiving manual compressions. These include ensuring accurate and continued endotracheal (ET) tube placement, predicting survival in cardiac arrest and confirming the effectiveness of cardiac compressions.

Future Possibilities
As additional research is completed, our understanding of the use of capnography with automated CPR devices will evolve. Some areas require additional exploration.

As mentioned, automated compressions result in sufficient air movement in and out of the lungs to produce an observable EtCO2 and capnography waveform. How this ventilation impacts the patient hasn’t been studied. Some EMS personnel have suggested this phenomenon is similar to high-frequency jet or percussive ventilation, where low tidal volumes are delivered at very rapid ventilatory rates.

This type of ventilation has been used to manage adult and pediatric patients with acute respiratory distress syndrome, and it has been shown to provide favorable gas exchange and improve oxygenation, as well as provide adequate ventilation at lower peak pressures than conventional ventilation.(11) Is it possible that the “ventilations” seen with automated CPR devices might be beneficial to the patient and might even replace BVM ventilations?

Another area that needs further evaluation is the impact of higher EtCO2 ranges achieved with automated CPR devices.(12) Although numerous studies have confirmed that higher EtCO2 values are achieved with automated CPR devices, no formal research has been done on how this affects the determination of ROSC. A retroactive study of all patients with ROSC in which both an automated CPR device and continuous waveform capnography were in use may provide a pattern or template for the rescuer to follow.

In addition, the higher EtCO2 ranges produced by automated CPR devices may change the guidelines for termination of CPR. Current research shows that an EtCO2 of less than 10 mmHg after 20 minutes of ALS therapy has a predictive value for death in the prehospital setting.(13) Because the vast majority of cardiac arrest patients with CPR provided by an automated CPR device have continuous EtCO2 values higher than 10 mmHg, does this mean that none of these patients will meet the criteria for termination of resuscitation? Does this mean the values of EtCO2 for termination of resuscitation will need to be different in the patient receiving automated compressions? Only solid, scientific research will provide the answers to those questions.

It’s extremely important that EMS personnel understand the implications of capnography use and assessment in the cardiac arrest patient undergoing automated CPR. There are many similarities between capnography use with manual and automated CPR, including the necessity to use waveform capnography to verify accurate and ongoing ET tube placement, to monitor the efficacy of cardiac compressions and to continuously monitor the appropriate assisted-ventilation rate.

However, it’s equally important to recognize the differences you’ll see in the capnography assessment, especially in regard to the increased EtCO2 values seen with automated compressions and the waveform changes precipitated by the active movement of air into and out of the lungs with each automated compression.

Capnography is a valuable tool in the clinical assessment of the cardiac arrest patient, guiding many aspects of treatment and clinical decision-making. Understanding its use in conjunction with automated CPR devices only increases that value.

Disclosure: The author has reported receiving honoraria and/or research support, either directly or indirectly, from Physio-Control, Inc.


  1. AHA: “Part 4: Adult Basic Life Support.” Circulation. 112(24 Suppl):IV 12–IV 18, 2005.
  2. Bonnemeier H, Olivecrona G, Simonis G, et al: “Automated continuous chest compression for in-hospital cardiopulmonary resuscitation of patients with pulseless electrical activity: A report of five cases.” International Journal of Cardiology. 136(2):e39-50, 2009.
  3. Lindner KH, Pfenninger EG, Lurie KG, et al: “Effects of active compression-decompression resuscitation on myocardial and cerebral blood flow in pigs.” Circulation. 88(3):1254–1263, 1993.
  4. Rubertsson S, Karlsten R: “Increased cortical cerebral blood flow with LUCAS; A new device for mechanical chest compressions compared to standard external compressions during experimental cardiopulmonary resuscitation.” Resuscitation. 65(3):357–363, 2005.
  5. Ornato JP, Garnett AR, Glasuer FL, et al: “Relationship between cardiac output and the end-tidal carbon dioxide tension.” Annals of Emergency Medicine. 19(10):1104–1106, 1990.
  6. Shibutani K, Muraoka M, Shirasaki S, et al: “Do changes in end-tidal PCO2 quantitatively reflect changes in cardiac output?” Anesthesia and Analgesia. 79(5):829–833, 1994.
  7. Sanders AB, Atlas M, Ewy GA, et al: “Expired PCO2 as an index of coronary perfusion pressure.” American Journal of Emergency Medicine. 3(2):147–149, 1985.
  8. Lewis LM, Stothert J, Standeven J, et al: “Correlation of end-tidal CO to cerebral perfusion during CPR.” Annals of Emergency Medicine. 21(9):1131–1134, 1992.
  9. Larsen AI, Hjørnevik AS, Ellingsen CL, et al: “Cardiac arrest with continuous mechanical chest compression during percutaneous coronary intervention. A report on the use of the LUCAS device.” Resuscitation. 75(3):454–459, 2007.
  10. AHA: “Part 7.2: Management of Cardiac Arrest” in 2005 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care supplement to Circulation. 112(24 Suppl):IV 51–IV 57, 2005.
  11. Salim A & Martin M: “High-frequency percussive ventilation.” Critical Care Medicine. 33(3 Suppl):S241–245, 2005.
  12. Steen S, Liao Q, Pierre L, et al: “Evaluation of LUCAS, a new device for automatic mechanical compression and active decompression resuscitation.” Resuscitation. 55(3):285–299, 2002.
  13. Wayne MA, Levine RL, Miller CC, et al: “Use of end-tidal carbon dioxide to predict outcome in prehospital cardiac arrest.” Annals of Emergency Medicine. 25(6):762–767, 1995.