The term capnography refers to the non-invasive measurement of the partial pressure of carbon dioxide (CO2) in exhaled breath expressed as the CO2 concentration over time. The relationship of CO2 concentration to time is graphically represented by the CO2 waveform, or capnogram.
Patients with normal lung function have characteristic rectangular capnograms and narrow gradients between their alveolar CO2 (i.e., EtCO2) and arterial CO2 concentration (PaCO2) of 0 to 5 mmHg. Patients with obstructive lung disease have impaired expiratory flow, and demonstrate a more rounded ascending phase and an upward slope in the alveolar plateau. In patients with abnormal lung function and ventilation/perfusion mismatch, the EtCO2–PaCO2 gradient widens depending on the severity of the lung disease. The EtCO2 in patients with lung disease is only useful for assessing trends in ventilatory status over time; isolated EtCO2 values may or may not correlate with the PaCO2.
According to literature, in the pre-hospital setting, capnography can be used for:
- Verification of endotracheal tube (ETT) placement
- Continuous monitoring of tube location during transport
- Indicator of ROSC (return of spontaneous circulation) during chest compressions
- Gauging effectiveness of resuscitation and prognosis during cardiac arrest
Verification of endotracheal tube (ETT) placement
The accuracy of EtCO2 to confirm tracheal location of an ETT varies with the technology used and the condition of the patient. For patients who are not in cardiac arrest, studies of qualitative colorimetric EtCO2 and quantitative capnography have demonstrated 100 percent sensitivity and 100 percent specificity for tracheal placement. In contrast, clinical assessment using physical signs is unreliable for determining ETT location.
Continuous monitoring of tube location during transport
Unrecognized misplaced intubation (UMI) has catastrophic consequences, and can occur when an endotracheal tube (ETT) is dislodged during patient transport. Continuous monitoring of ETT location during transport can prevent UMI. EtCO2 confirmation of initial endotracheal tube placement and continuous capnographic monitoring of ETT location is an accepted standard of care by the American Society of Anesthesiologists and is recommended as the most reliable method by other national organizations.
Indicator of ROSC during chest compressions
American Heart Association (AHA) guidelines for cardiac resuscitation emphasize the importance of continuing chest compressions without interruption until a perfusing rhythm is re-established. Capnographic waveform monitoring virtually eliminates the need to stop chest compressions to check for pulses. Re-establishment of a perfusing rhythm organ perfusion is accompanied immediately by a dramatic increase in EtCO2. Once this rise in EtCO2 is noted, chest compressions can safely be stopped while cardiac rhythm and blood pressure are assessed.
Gauging effectiveness of resuscitation and prognosis during cardiac arrest
There is evidence that EtCO2, during cardio-pulmonary resuscitation (CPR), can be used as a predictor for survival. It has been reported that an EtCO2 value of less than 10 mmHg measured 20 min after the initiation of advanced cardiac life support (ACLS) accurately predicts death in patients with cardiac arrest.
Although EtCO2 in cardiac arrest may have a role in the prediction of death, its ability to predict favourable outcomes (ROSC/survival) is very limited. The specificity of EtCO2 in predicting ROSC at the cut-off value of 10 mmHg was reported to be between 50% and 87% for initial EtCO2 and 77% for average EtCO2.
This could be particularly true in low cardiac output states, as during cardiac arrest. In this setting, it has been demonstrated that EtCO2 values tend to be more dependent on compression-driven cardiac output and less dependent on CO2 production and ventilation. This has led to the exploration of other capnographic features to predict ROSC (e.g., area under the curve, slopes, cumulative max EtCO2) and the suggestion by researchers and companies to use the features of the capnogram (i.e., EtCO2, VCO2) as inputs to algorithms for assessing the efficacy of and controlling chest compressions.
Moreover, there is evidence coming from patients undergoing capnographic monitoring during general anaesthesia induction without muscular blockage that morphological changes of the capnographic curve could detect changes in the ventilation of the patient earlier than the simple end-tidal and pulse oximetry device.
But today’s capnography waveform displays do not capture the trending shape or the wave’s morphology. We hence start working on automatic interpretation of capnographic wave to move from “number” (i.e., EtCO2) to “shape” of the capnographic curve. The idea is to create a database source that will be able to collect the complex and comprehensive information within clinical sites and apply that information to algorithmic models for an immediate decision support analysis.
The final outcome of the project is to develop a software platform that will be able to provide an integrated digital representation and alert system that can provide constant feedback in a user-friendly format that will help the clinician manage patient care and improve outcomes.
The project, on which we are working in cooperation with SoftJam Innovation, a Microsoft Italian partner, is actually based on the Microsoft Cortana Analytics Suite® that is able to provide all the requested informatics instruments.
In more detail, the flow of data coming from the devices feeds a service of events elaboration (Event Hub) whose purpose is to pre-process data so that they can be used to construct algorithms that can learn from and make predictions on data (Machine Learning). Data are concomitantly saved in a database in the Cloud, in order to get a centered repository of all the classified data.
Simultaneous with storage and analysis, data are graphically modified in order to make possible mathematical calculations for analytical purposes (such as the average of the waveform periods over a pre-defined period of time).
We do believe that the predictive elaboration service together with the graphical consultation interface will allow a real immediate decision support analysis to be applied in all the clinical settings.