Carbon dioxide monitoring (capnography)

INTRODUCTION — This topic review will discuss the basic physiology and interpretation of capnography and its use in emergency settings.

DEFINITION AND BACKGROUND — The term capnography refers to the noninvasive 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 (figure 1). Changes in the shape of the capnogram are diagnostic of disease conditions, while changes in end-tidal CO₂ (EtCO₂), the maximum CO₂ concentration at the end of each tidal breath, can be used to assess disease severity and response to treatment. Capnography is also the most reliable indicator that an endotracheal tube is placed in the trachea after intubation.

Oxygenation and ventilation are distinct physiologic functions that must be assessed in both intubated and spontaneously breathing patients. Pulse oximetry provides instantaneous feedback about oxygenation (see "Pulse oximetry"). Capnography provides instantaneous information about ventilation (how effectively CO₂ is being eliminated by the pulmonary system), perfusion (how effectively CO₂ is being transported through the vascular system), and metabolism (how effectively CO₂ is being produced by cellular metabolism).

Capnography became a routine part of anesthesia practice in Europe in the 1970s and in the United States in the 1980s. It is now part of the standard of care for all patients receiving general anesthesia and is part of routine monitoring in the pre-hospital and acute care settings. (see "Basic patient monitoring during anesthesia", section on 'Capnography')

PRINCIPLES OF OPERATION — Carbon dioxide (CO₂) monitors measure gas concentration, or partial pressure, using one of two configurations: mainstream or sidestream. Mainstream devices measure respiratory gas (in this case CO₂) directly from the airway, with the sensor located on the airway adapter at the hub of the endotracheal tube (ETT). Sidestream devices measure respiratory gas via nasal or nasal-oral cannula by aspirating a small sample from the exhaled breath through the cannula tubing to a sensor located inside the monitor (picture 1).

Mainstream systems are configured for intubated patients. Sidestream systems are configured for both intubated and nonintubated patients.

Sidestream systems are configured to use high flow rates (around 150 cc/min) or low flow rates (around 50 cc/min). Flow rates vary according to the amount of CO₂ needed in the breath sample to obtain an accurate reading. Low-flow systems have a lower occlusion rate (from moisture or patient secretions) and are accurate in patients with low tidal volumes (eg, neonates, infants, and adult patients with hypoventilation and low tidal volume breathing). Low-flow systems are also resistant to dilution from supplemental oxygen. High-flow systems sampling at ≥100 cc/min have been shown to be inaccurate in neonates, infants, young children, and in hypoventilating adult patients [1-3].

CO₂ monitors are either quantitative or qualitative. Quantitative devices measure the precise end-tidal CO₂ (EtCO₂) either as a number (capnometry) or a number and a waveform (capnography). Qualitative devices (eg, colorimetric detectors) report the range in which the EtCO₂ falls (eg, 0 to 10 mmHg or >35 mmHg) as opposed to a precise value (eg, 38 mmHg).

The colorimetric EtCO₂ detector, which consists of a piece of specially treated litmus paper that changes color when exposed to CO₂ (purple for EtCO₂ <3 mmHg; tan for 3 to 15 mmHg; and yellow for >15 mmHg). Its primary use is for verification of ETT placement. Exhalation of CO₂ from an ETT placed in the trachea will change the color of the litmus paper from purple to yellow. An improperly placed ETT in the esophagus will not conduct CO₂ and no change will occur in the color of the litmus paper, which will remain purple. When evaluating studies of EtCO₂, it is essential to distinguish those involving qualitative measurements (colorimetric) from those describing quantitative methods with a graphic waveform display (capnography). (See "Rapid sequence intubation in adults for emergency medicine and critical care", section on 'Placement with proof'.)

Capnography uses infrared (IR) radiation to make measurements. Molecules of CO₂ absorb IR radiation at a very specific wavelength (4.26 µm), with the amount of radiation absorbed having a nearly exponential relation to the CO₂ concentration present in the breath sample. Detecting these changes in IR radiation levels, using appropriate photo-detectors sensitive in this spectral region, allows for the calculation of the CO₂ concentration in the gas sample.

CO2 WAVEFORM — The capnogram consists of four phases (figure 1).

●Phase 1 (dead space ventilation, A-B) represents the beginning of exhalation where the dead space is cleared from the upper airway and the CO₂ concentration approaches zero.

●Phase 2 (ascending phase, B-C) represents the rapid rise in carbon dioxide (CO₂) concentration in the breath stream as the CO₂ from the alveoli reaches the upper airway.

●Phase 3 (alveolar plateau, C-D) represents the CO₂ concentration reaching a uniform level in the entire breath stream from alveolus to nose. Point D, occurring at the end of the alveolar plateau, represents the maximum CO₂ concentration at the end of the tidal breath and is appropriately named the end-tidal CO₂ (EtCO₂). This is the number that appears on the monitor display.

●Phase 4 (D-E) represents the inspiratory cycle where the CO₂ falls back to zero.

A normal capnogram (ie, a valid breath), for patients of all ages, is characterized by a set of specific elements: the CO₂ concentration starts at zero and returns to zero, a maximum CO₂ concentration is reached with each breath (ie, EtCO₂), the amplitude is a function of the EtCO₂ concentration, the width is a function of the expiratory time, and there is a characteristic shape with normal lung function.

●Patients with normal lung function have characteristic trapezoidal capnograms and narrow gradients between their alveolar CO₂ (ie, EtCO₂) and arterial CO₂ concentration (PaCO₂) of 0 to 5 mmHg (figure 1). Gas in the physiologic dead space accounts for this normal gradient.

●Patients with obstructive lung disease have impaired expiratory flow and uneven emptying of alveoli due to ventilation-perfusion mismatch, and demonstrate a more rounded ascending phase and an upward slope in the alveolar plateau (figure 2). In patients with abnormal lung function and ventilation-perfusion mismatch, the EtCO₂-PaCO₂ gradient widens depending on the severity of the lung disease [4,5]. 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.

CLINICAL APPLICATIONS FOR INTUBATED PATIENTS — Capnography can be used in intubated patients for:

●Verification of endotracheal tube (ETT) placement

●Continuous monitoring of tube location during transport

●Gauging effectiveness of resuscitation and prognosis during cardiac arrest

●Indicator of ROSC during chest compressions

●Titrating end-tidal carbon dioxide (EtCO₂) levels in patients with suspected increases in intracranial pressure

●Determining prognosis in trauma

●Determining adequacy of ventilation

A table summarizing capnographic findings in different clinical scenarios is provided (figure 3A-B).

Verification of ETT placement — We advocate capnography in all patients who are intubated since it is highly accurate in determining ETT location and excluding unrecognized misplaced intubation (UMI). Correct ETT placement must be verified immediately following every tracheal intubation. The role of EtCO₂ in confirming ETT placement is discussed below; other methods and our recommendations for verifying ETT placement are discussed in detail separately. (See "Direct laryngoscopy and endotracheal intubation in adults", section on 'Confirming proper tracheal tube placement'.)

UMI is defined as the placement of an ETT in a location other than the trachea that is unrecognized by the clinician. UMI has disastrous consequences and has been extensively documented in the emergency medical services (EMS) literature, with reported rates ranging from 7 to 25 percent [6-18]. UMI is unacceptable and widespread use of CO₂ monitoring along with other verification techniques has helped to prevent it.

Following intubation, the sustained presence of a waveform with all four phases (figure 1) indicates the ETT is through the vocal cords [19]. A normal waveform can occur when the tube has been placed in the right mainstem bronchus. In addition, a waveform resembling tracheal placement may be present for a few breaths with an ETT placed in the hypopharynx just above the vocal cords, but over time these waveforms are likely to become erratic due to movement of the ETT, suggesting that the ETT is not properly placed. A four-phase capnogram may be present upon initial intubation in cardiac arrest patients, indicating that the tube is correctly positioned in the trachea [20,21].

To be considered sustained exhalation of CO₂, the capnogram should meet all of the following [19]:

●Amplitude rises during exhalation and falls during inspiration

●Consistent or increasing amplitude over at least seven breaths

●Peak amplitude more than 7.5 mmHg (1 kPa) above baseline

●Reading is clinically appropriate

A flatline waveform generally indicates esophageal placement but can occur, albeit uncommonly, in several other situations, including:

●ETT obstruction

●Complete airway obstruction distal to the ETT (eg, foreign body)

●Technical malfunction of the monitor or tubing

●Prolonged cardiac arrest with diffuse cellular death (in which no CO₂ is produced because of an absence of cellular respiration and CO₂ exchange in the pulmonary bed is severely compromised). In a cardiac arrest patient who initially had a four-phase capnogram, tube placement should be verified if a flatline tracing subsequently develops.

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 [22-27]. However, qualitative colorimetric EtCO₂ is slightly less accurate compared with continuous quantitative waveform capnography; we prefer the latter when feasible [19]. If only qualitative colorimetric EtCO₂ is used, color change must be visualized on at least five breaths since the esophagus may yield small but detectable amounts of CO₂ during the first few positive-pressure ventilations.

In contrast, clinical assessment using physical signs is unreliable for determining ETT location. Fogging or condensation of the tube occurs in 83 percent of esophageal intubations [28], and chest wall movement can be produced by tracheal or esophageal tubes [29]. Anesthesiologists, under ideal operating room conditions, using breath sounds as the sole means of verification, incorrectly identify ETT location in 16 percent of cases [30].

Although the accuracy of EtCO₂ for verifying ETT placement in patients with spontaneous circulation approaches 100 percent, sensitivities for tracheal placement vary greatly in cardiac arrest patients, ranging from 62 to 100 percent, depending on the modality used and the duration of the arrest [4,5,10,24,25,28-31]. In patients with low perfusion states and cadavers, studies using quantitative modalities (ie, capnography) for determining tracheal location of the ETT have shown 100 percent sensitivity [20,32-34]. Studies using qualitative (ie, colorimetric) EtCO₂ methods have shown variable sensitivity, because the exhaled CO₂ concentration can fall below the detection threshold [10,24,25,30]. The small number of esophageal intubations in studies assessing the accuracy of capnography for determining ETT placement in cardiac arrest patients limits the strength of these studies. Capnography's true specificity for detecting esophageal intubation in cardiac arrest patients is not certain.

When a capnogram is present in an intubated patient in cardiac arrest, the ETT can be assumed to be within the trachea [34-36]. An absent (flatline) waveform may indicate esophageal or placement above the trachea. Early in cardiac arrest, and when the patient is receiving effective CPR, complete failure of CO₂ exchange is unlikely, and the clinician should assume esophageal intubation exists if a waveform is absent. In patients with prolonged cardiac arrest and an absent waveform, the clinician should also assume an esophageal intubation exists. Physical examination criteria, such as repeat laryngoscopy and auscultation of the lungs and epigastrium, may be used, but the clinician must be aware of their limitations in discriminating between esophageal and tracheal intubation. (See "Direct laryngoscopy and endotracheal intubation in adults", section on 'Confirming proper tracheal tube placement'.)

In addition to verification and monitoring of ETT placement, capnography provides similar information about the placement of alternate advanced airway devices (laryngeal tube airway, laryngeal mask airway, esophageal-tracheal combitube).

The accuracy of EtCO₂ for confirming ETT location can help clinicians faced with difficult airway management decisions. As an example, in a case involving one of the authors, a patient resuscitated following a submersion injury became difficult to ventilate and hypotensive following RSI, but the presence of a waveform reassured clinicians about ETT placement, allowing them to consider other reasons for decompensation. Decompression of the patient’s distended stomach with an orogastric tube resolved the difficulties with ventilation and blood pressure.

Monitoring ETT location during transport — UMI has catastrophic consequences and can occur when an ETT is dislodged during patient transport. Continuous monitoring of ETT location during transport prevents UMI. EtCO₂ confirmation of initial ETT placement and continuous capnographic monitoring of ETT location is an accepted standard of care by the American Society of Anesthesiologists [37] and is recommended as the most reliable method by other national organizations [38-40].

A prospective, observational study of 153 prehospital intubations found a 23 percent UMI rate among patients without continuous EtCO₂ monitoring, while patients with continuously monitored EtCO₂ sustained the desired 0 percent UMI rate [18]. The attached figures depict one airway confirmation algorithm using colorimetric CO₂ detectors and another using capnography. They can be used for both prehospital and in-hospital intubated patients.

Effectiveness of CPR — In the 1980s, studies using animal models showed that EtCO₂ levels reflect cardiac output during cardiopulmonary resuscitation (CPR) and can be used as a noninvasive measure of cardiac output. A landmark study in 1988 demonstrated this principle in humans [41]. During cardiac arrest, when alveolar ventilation and metabolism are essentially constant, EtCO₂ reflects pulmonary blood flow. Therefore, EtCO₂ can be used as a gauge of the effectiveness of cardiac compressions. As effective CPR leads to a higher cardiac output, EtCO₂ will rise, reflecting the increase in perfusion.

The measurement of EtCO₂ varies directly with the cardiac output produced by chest compression and has been described in both EMS [42] and ICU patients [41]. Both of these prospective, observational studies found an EtCO₂ level <3 mmHg immediately after cardiac arrest, with a higher level generated during cardiac compressions and a mean peak >7.5 mmHg just before return of spontaneous circulation (ROSC) [41,42]. This peak in EtCO₂ level is the earliest sign of ROSC and may occur before return of a palpable pulse or blood pressure. Another observational study of data collected during human CPR found a positive correlation between the depth of chest compressions and EtCO₂ (10 mm increase in compression depth increased EtCO₂ by 1.4 mmHg), as well as ventilation rate and EtCO₂ (increase of 10 breaths/minute decreased ETCO₂ by 3 mmHg), and confirmed that higher EtCO₂ values during CPR correlate with increased ROSC and survival [43].

Return of spontaneous circulation — During cardiac arrest, EtCO₂ is the earliest indicator of the return of spontaneous circulation (ROSC) [41,42]. When the heart restarts, the dramatic increase in cardiac output, and resulting increase in perfusion, leads to a rapid increase in EtCO₂ as the CO₂ that has accumulated during cardiac arrest is effectively transported to the lungs and exhaled. This process manifests as a sudden rise in EtCO₂. (See 'Effectiveness of CPR' above.)

American Heart Association (AHA) guidelines for cardiac resuscitation emphasize the importance of continuing chest compressions without interruption until a perfusing rhythm is reestablished. Experimental evidence indicates that interruptions in chest compressions are followed by sustained periods of reduced blood flow, which only gradually return to pre-interruption levels. Capnogram monitoring virtually eliminates the need to stop chest compressions to check for pulses. Reestablishment of a perfusing rhythm 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 [40,44].

According to an observational study of 145 patients with out of hospital cardiac arrest (OHCA), several seconds may be required after electrical conversion to a potentially perfusing rhythm before effective mechanical contractions and a subsequent rise in EtCO₂ occur [45]. Accordingly, stopping compressions in response to a rhythm change in order to check for a pulse may result in degradation of the rhythm, as coronary perfusion had not yet been restored. EtCO2 is a more reliable harbinger of reperfusion than a pulse check. (See "Adult basic life support (BLS) for health care providers".)

Specific EtCO₂ levels have been shown to correlate with ROSC. In a systematic review of 17 observational studies including over 6100 adults with cardiac arrest, which included metaanalyses of data from five studies, EtCO₂ ≥10 to 20 mmHg during CPR was strongly associated with ROSC while persistent EtCO₂ below 10 mmHg after 20 minutes of CPR had a 0.5 percent likelihood of ROSC [46]. Subsequent retrospective studies support this general finding [47,48]. A retrospective, observational study of 526 OHCA patients with a nonshockable rhythm found that the first ETCO₂ measured after placement of an airway device predicted ROSC and survival after hospital admission [47]. Specifically, patients with an ETCO₂ >45 mmHg had an increased probability of sustained ROSC and 30-day survival. A retrospective study of 324 OHCA patients found that higher EtCO₂ levels were associated with higher rates of successful defibrillation [48].

Confounding factors in resuscitation — Drugs used in resuscitation may affect EtCO₂ values. EtCO₂ often decreases rapidly moments after administration of epinephrine. Sodium bicarbonate may produce a transient increase in EtCO₂, but the rise in EtCO₂ levels after ROSC is greater and longer lasting than after a sodium bicarbonate bolus [49,50].

Prognosis in cardiac arrest — In several prospective, observational studies, EtCO₂ levels of ≤10 mmHg measured 20 minutes after the initiation of advanced cardiac life support accurately predicted death in adult patients with cardiac arrest [51-54]. The prognostic value of measuring EtCO₂ has been demonstrated in animal [55] and human studies [51-54,56].

Cause of cardiac arrest — EtCO₂ may also be helpful in determining the etiology of cardiac arrest. Two animal studies reported higher EtCO2 levels at the onset of cardiac arrest caused by primary asphyxia than arrest caused by ventricular fibrillation [57,58]. Researchers found results consistent with these animal experiments in a prospective, observational study of prehospital victims of cardiac arrest. Patients in the asphyxia group (initial rhythm of asystole or pulseless electrical activity associated with conditions such as airway foreign body, aspiration, asthma, or drowning) had higher EtCO₂ levels compared with patients in the ventricular tachycardia/fibrillation group (initial rhythm of ventricular tachycardia/fibrillation associated with acute myocardial infarction) [59]. These between group differences disappear within minutes if return of spontaneous circulation has not occurred.

Increased ICP and trauma prognosis — EtCO₂ monitoring can help clinicians avoid inadvertent hyperventilation of patients with traumatic brain injury and suspected increased intracranial pressure (ICP). It may also help determine the prognosis of trauma victims.

Arterial CO₂ tension affects blood flow to the brain. High CO₂ levels result in cerebral vasodilation, while low CO₂ levels result in cerebral vasoconstriction. Sustained hypoventilation (defined as PaCO₂ levels ≥50 mmHg) results in increased cerebral blood flow and increased ICP, which can harm head-injured patients. Sustained hyperventilation (defined as PaCO₂ ≤30 mmHg) is also detrimental and is associated with worse neurologic outcome in severely brain-injured patients. Consequently, ventilation rates to achieve eucapnia are recommended by the brain trauma foundation [60].

Use of continuous EtCO₂ monitoring in intubated trauma patients is supported by several studies:

●A prospective observational study found a lower incidence of inadvertent hyperventilation among intubated patients with severe head injury who underwent continuous EtCO2 monitoring compared with those without EtCO₂ monitoring (5.6 versus 13.4 percent; odds ratio [OR] 2.64; 95% CI 1.12-6.20) [61].

●In a controlled study of intubated blunt trauma victims, patients assigned to capnography-guided ventilation during transport were significantly more likely to arrive at the emergency department appropriately ventilated, based on PaCO₂ levels obtained by arterial blood gas [62].

Studies have found that a low EtCO₂ in trauma patients is associated with mortality and hemorrhagic shock:

●In a prospective observational study of blunt trauma patients requiring prehospital intubation, EtCO₂ values obtained 20 minutes after intubation distinguished the great majority of survivors from non-survivors [63]. Median EtCO₂ among survivors was 30.8 mmHg and among non-survivors 26.3 mmHg (95 percent CI of difference between medians 3 to 6.75 mmHg).

●In a retrospective observational study of 135 trauma patients transported by emergency medical services to a level one trauma center, the mean prehospital EtCO₂ level distinguished survivors (34 mmHg, 95% CI 32-35) from non-survivors (18 mmHg, 95% CI 9-28) and had better discrimination regarding prognosis than vital signs or pulse oximetry [64].

●In a retrospective review of 373 trauma patients receiving a massive transfusion protocol and undergoing damage control surgery, operative ETCO₂ ≤23mmHg was a predictor of mortality [65].

●In two retrospective studies of non-intubated adult trauma patients, initial side-stream ETCO₂ ≤28.5 to 29.5 mmHg was associated with higher in-hospital mortality, blood transfusion requirement, and complication rates [66,67].

CLINICAL APPLICATIONS FOR SPONTANEOUSLY BREATHING PATIENTS — In a spontaneously breathing, nonintubated patient, capnography can be used for the following:

●Performing rapid assessment of critically ill or seizing patients

●Determining response to treatment in acute respiratory distress

●Determining adequacy of ventilation in obtunded or unconscious patients, or in patients undergoing procedural sedation

●Detecting metabolic acidosis in diabetic patients and in children with gastroenteritis

●Providing prognostic indicators in patients with sepsis or septic shock

A table summarizing capnographic findings in different clinical scenarios is provided (figure 3A-B).

Assessing airway, breathing, and circulation — The airway, breathing, and circulation (ABCs) of patients can be rapidly assessed using the capnography waveform and end-tidal carbon dioxide (EtCO₂) values [68]. The presence of a normal waveform denotes a patent airway and spontaneous breathing [69]. Normal EtCO₂ levels (35 to 45 mmHg) signify adequate ventilation and perfusion [41,70]. Capnography can be used to assess unresponsive patients ranging from those who are actively seizing to victims of chemical terrorism (table 1) [71]. Unlike pulse oximetry, capnography does not misinterpret motion artifact and provides reliable readings in low perfusion states. Studies have found that an EtCO₂ ≤29.5 mmHg was independently predictive of the need for massive transfusion [66,72].

Seizure — Capnography is the only monitoring modality that is accurate and reliable in actively seizing patients because the capnogram is determined entirely by respiratory activity and is not confounded by muscle activity or movement artifact. Capnographic data (respiratory rate, EtCO₂, and capnogram) can be used to distinguish among:

●Seizing patients with apnea (flatline waveform, no EtCO₂ reading, and no chest wall movement)

●Seizing patients with ineffective ventilation (small waveforms, low EtCO₂ values), and

●Seizing patients with effective ventilation (normal CO₂ waveform, normal EtCO₂ values). (See "Convulsive status epilepticus in adults: Classification, clinical features, and diagnosis" and "Convulsive status epilepticus in adults: Management".)

Acute respiratory distress — Capnography provides dynamic monitoring of ventilatory status in patients with acute respiratory distress from any cause, including: bronchiolitis, asthma, cystic fibrosis, heart failure, and chronic obstructive pulmonary disease (COPD). Several studies have found that quantitative waveform analysis can discriminate between heart failure and COPD [73-75].

By measuring EtCO₂ and respiratory rate with each breath, capnography provides instantaneous feedback on the clinical status of the patient. Respiratory rate is measured directly from the airway (nose and mouth) with oral or nasal cannula, providing a more reliable reading than impedance respiratory monitoring. In upper airway obstruction and laryngospasm, impedance monitoring detects chest wall movement, interprets this as a valid breath, and displays a respiratory rate, even though the patient is not ventilating. In contrast, capnography detects no ventilation and shows a flatline waveform.

Clinicians can rapidly assess EtCO₂ trends. A patient with a respiratory rate of 30 will generate 150 EtCO₂ readings in five minutes. This provides sufficient information to determine whether the patient's ventilation is worsening despite treatment (increasing EtCO₂), stabilizing (stable EtCO₂), or improving (decreasing EtCO₂). As an example, increasing EtCO₂ would reflect the worsening ventilation of an acutely tachypneic patient with obstructive lung disease who is developing respiratory muscle fatigue or a more severe lower airway obstruction.

Procedural sedation — We recommend continuous capnographic monitoring in all patients undergoing procedural sedation and analgesia (PSA). Capnography can rapidly detect the common adverse airway and respiratory events associated with PSA, including respiratory depression, apnea, upper airway obstruction, laryngospasm, and bronchospasm. Patients monitored with capnography experience significantly fewer episodes of hypoxia than those monitored in standard fashion, especially since the administration of supplemental oxygen renders pulse oximetry less effective as an early warning device for respiratory depression during PSA. (See "Procedural sedation in adults in the emergency department: General considerations, preparation, monitoring, and mitigating complications", section on 'Monitoring' and "Procedural sedation in children: Approach", section on 'Monitoring'.)

Respiratory depression caused by oversedation will manifest an abnormally high or low EtCO₂ before pulse oximetry detects a falling oxyhemoglobin saturation, especially in patients receiving supplemental oxygen. In one trial, the median time between the onset of respiratory depression, as determined by EtCO₂, and hypoxia was 60 seconds (range 5 to 240 seconds) [76].

Common capnographic findings during PSA include the following and are summarized in the figures (figure 3A and figure 3B):

●Flatline waveform – This can occur from monitor calibration (will indicate "calibrating" on the monitor), cannula occlusion (will indicate "occlusion" on the monitor), or apnea (central or obstructive). The combination of absent chest wall movement and a flatline waveform differentiates central apnea from obstructive apnea (upper airway obstruction or laryngospasm), which manifests chest wall movement. Response to airway alignment maneuvers (eg, chin lift, jaw thrust) can distinguish upper airway obstruction from laryngospasm.

●Oscillating waveform progressively decreasing in height – This occurs with oversedation and is a sign of impending apnea.

●Small waveforms interspersed with flatline waveforms - This irregular waveform pattern occurs with hypopneic hypoventilation and indicates periodic breathing.

●Increasing waveform height – This occurs with bradypneic hypoventilation and reflects increasing EtCO₂. An EtCO₂ >70 mmHg in a patient without COPD indicates respiratory failure.

●Decreasing waveform height – This occurs with hypopneic hypoventilation (ie, low tidal volume breathing) and reflects decreasing EtCO₂.

●Increasing waveform width – This occurs with hypoventilation and increasing expiratory time.

●Decreasing waveform width – This occurs with hyperventilation and decreasing expiratory time.

●Waveform does not return to zero – An off-baseline waveform can occur with rebreathing.

●Curved/shark-fin appearing waveform – The characteristic curved capnogram (figure 2) can be seen in a patient with obstructive lung disease (eg, asthma, COPD, cystic fibrosis) and indicates bronchospasm if it occurs in a patient without chronic lung disease.

Ventilation in obtunded or unconscious patients — Patients with altered mental status, including those with alcohol intoxication, intentional or unintentional drug overdose, and postictal patients (especially those treated with benzodiazepines), may have impaired ventilatory function. Capnography can differentiate between patients with effective ventilation and those with ineffective ventilation [77].  

Detecting metabolic acidosis — There is a positive linear correlation between serum bicarbonate (HCO3) and EtCO₂ in patients with diabetic ketoacidosis and children with gastroenteritis [78-82]. As the patient becomes acidotic, HCO3 decreases, causing an increase in minute ventilation, which results in a compensatory respiratory alkalosis. This process results in a decrease in EtCO₂.

By increasing minute ventilation, these patients are able to lower arterial CO₂ tension to help correct their underlying acidemia. The more acidotic the patient, the lower the HCO3, the higher the respiratory rate, and the lower the EtCO₂. Capnography can be used as an indicator of metabolic acidosis in these patients. In addition, EtCO₂ can be used to distinguish diabetics in ketoacidosis (metabolic acidosis, compensatory tachypnea, and low EtCO₂) from those who are not (nonacidotic, normal respiratory rate, and normal EtCO₂) [83,84]. (See "Approach to the adult with metabolic acidosis" and "Approach to the child with metabolic acidosis".)

Prognosis in sepsis — There is an inverse relationship between EtCO₂ and lactate levels in sepsis, severe sepsis, and septic shock. EtCO₂ performs similarly to lactate as a predictor for mortality in patients with suspected sepsis [85,86]. In a prospective study of 183 patients performed in the prehospital setting, patients with suspected sepsis and an EtCO₂ value ≤25 mmHg were more frequently diagnosed with sepsis (78 versus 43 percent) and severe sepsis (47 versus 7 percent), and had higher mortality (11 versus 5 percent) [87]. The use of EtCO₂ to predict mortality, sepsis, and severe sepsis in prehospital patients outperformed another well-accepted screening tool, the quick sequential organ failure assessment score (qSOFA) [88]. Patients with severely compromised respiratory reserve who cannot appropriately compensate for the lactic acidosis resulting from severe sepsis or septic shock have the worst prognosis despite ETCO₂ values that are not low. (See "Evaluation and management of suspected sepsis and septic shock in adults".)

LIMITATIONS — Capnography is most effective when assessing a pure ventilation, perfusion, or metabolism problem. Capnographic findings in mixed ventilation, perfusion, or metabolism problems are difficult to interpret. In patients with complex pathophysiology, a ventilation problem may elevate end-tidal carbon dioxide (EtCO₂), while a perfusion problem may simultaneously lower EtCO₂. As an example, capnography may not provide useful information in a patient with acute pulmonary edema (ventilation problem) from an underlying acute myocardial infarction with decreased cardiac output (perfusion problem).

In high-flow CO₂ systems, when the tidal volume of the patient drops below the flow rate of the system, the monitor will entrain room air to compensate, falsely diluting the EtCO₂ reading and slurring the ascending phase of the waveform. For this and other reasons, low flow systems are preferred. (See 'Principles of operation' above.)

Capnography is highly accurate for determining tracheal placement of an endotracheal tube (ETT) in nearly all clinical circumstances. However, the test characteristics of capnography in cardiac arrest can vary, and this is discussed above. (See 'Verification of ETT placement' above.)

SUMMARY AND RECOMMENDATIONS

●Definition and background – Carbon dioxide (CO₂) monitoring is a versatile noninvasive diagnostic tool for continuous assessment of the ventilatory status of both intubated and nonintubated patients. Capnography provides instantaneous information about ventilation (how effectively CO₂ is being eliminated by the pulmonary system), perfusion (how effectively CO₂ is being transported through the vascular system), and metabolism (how effectively CO₂ is being produced by cellular metabolism). A table summarizing capnographic findings in different clinical scenarios is provided (figure 3A-B). (See 'Definition and background' above.)

●Principles of operation – Colorimetric detectors provide a qualitative measurement of end-tidal CO₂ (EtCO₂); capnometry provides a quantitative EtCO₂ measurement; capnography provides a quantitative measure and a waveform depicting CO₂ levels over time. (See 'Principles of operation' above.)

●CO₂ waveform – The capnogram consists of four phases: dead space ventilation, ascending phase, alveolar plateau, and descending inspiratory phase, which are depicted in the figure (figure 1). (See 'CO2 Waveform' above.)

●Clinical applications – Capnography is of greatest value for emergency departments and emergency medical services for confirming placement of endotracheal tubes (ETTs), continuous monitoring of tube position during transport, assessing the status of patients in cardiac arrest, and monitoring the respiratory status of patients undergoing procedural sedation. (See 'Clinical applications for intubated patients' above.)

●Verification of ETT placement – In all intubated patients, we recommend that an EtCO₂ monitor be used to help confirm proper placement of the ETT. EtCO₂ monitoring is the primary means of confirming tracheal placement of the ETT. Clinical techniques may be used adjunctively, but all have significant limitations in discriminating between esophageal and tracheal intubation. (See 'Verification of ETT placement' above.)

Following intubation, the presence of a sustained waveform with all four phases (figure 1) is a highly accurate indication that the ETT is placed in the trachea, including upon initial intubation in cardiac arrest patients. If only qualitative colorimetric EtCO₂ is used, color change must be visualized on at least five breaths since the esophagus may yield small but detectable amounts of CO₂ during the first few positive-pressure ventilations.

A flatline waveform generally indicates esophageal placement but can occur with: ETT obstruction, complete airway obstruction distal to the ETT (eg, foreign body), technical malfunction of the monitor or tubing, and prolonged cardiac arrest. When the amplitude of the tracing is very small or CO₂ is not detectable by waveform or qualitative EtCO₂ detection, clinicians should assume an esophageal intubation exists.

●Monitoring resuscitation in cardiac arrest – In a patient in cardiac arrest, we suggest that EtCO₂ be used to monitor resuscitation efforts. EtCO₂ monitoring can be performed on patients in cardiac arrest to determine the adequacy of chest compressions and the return of spontaneous circulation. During cardiac arrest, EtCO₂ reflects pulmonary blood flow and can be used as a gauge of the effectiveness of cardiac compressions. As effective CPR leads to a higher cardiac output, EtCO₂ will rise. (See 'Effectiveness of CPR' above.)

●Assessing return of spontaneous circulation – EtCO₂ is also the earliest indicator of the return of spontaneous circulation. When the heart restarts, the dramatic increase in cardiac output, and resulting increase in perfusion, leads to a rapid increase in EtCO₂. (See 'Return of spontaneous circulation' above.)

●Monitoring during procedural sedation – We recommend continuous capnographic monitoring in all patients undergoing procedural sedation and analgesia (PSA). Capnography can rapidly detect common adverse airway and respiratory events including central apnea, obstructive apnea (upper airway obstruction, laryngospasm), bronchospasm, and respiratory failure (algorithm 1). The evidence for capnographic monitoring is presented separately. (See 'Procedural sedation' above and "Procedural sedation in adults in the emergency department: General considerations, preparation, monitoring, and mitigating complications", section on 'Monitoring'.)

●Applications in spontaneously breathing patients – Capnography can be helpful in monitoring the status of patients with seizures, acute respiratory distress, unconsciousness, and metabolic acidosis. (See 'Clinical applications for spontaneously breathing patients' above.)

●Limitations – Capnography is most effective when assessing a pure ventilation, perfusion, or metabolism problem. Capnographic findings in mixed ventilation, perfusion, or metabolism problems are difficult to interpret. (See 'Limitations' above.)

ACKNOWLEDGMENT — The editorial staff at UpToDate acknowledge Salvatore Silvestri, MD, now deceased, who contributed to an earlier version of this topic review.