How to Read and Interpret Capnography Waveforms

A capnography waveform is a real-time graph of the carbon dioxide leaving your lungs with each breath, and learning to read its shape tells you far more than a single number ever could. The waveform’s height reflects how much CO₂ is being exhaled, its width tracks the timing of each breath, and its contour reveals whether air is flowing freely or getting trapped. Clinicians use these visual patterns to catch airway problems, gauge circulation during cardiac arrest, and even pick up clues about metabolic emergencies. The interpretation is surprisingly intuitive once you know what each phase of the waveform represents.

The Four Phases of a Normal Waveform

A standard capnography tracing, sometimes called a capnogram, plots CO₂ concentration on the vertical axis against time on the horizontal axis. Each breath creates a characteristic shape with four recognizable segments. Phase I is the baseline at or near zero: this is the very beginning of exhalation, when the air leaving the upper airway still contains almost no CO₂ because it never reached the gas-exchanging parts of the lungs. Phase II is a sharp upstroke as CO₂-rich air from deeper in the lungs begins to arrive at the sensor. Phase III is the alveolar plateau, a gently rising or nearly flat segment representing the bulk of the exhaled gas from the alveoli. The highest point at the end of that plateau is the end-tidal CO₂ value, often abbreviated EtCO₂, and it’s the single number most people associate with capnography. Phase IV, sometimes called phase 0, is the rapid downstroke back to baseline as the next inhalation draws fresh, CO₂-free air past the sensor.

When all four phases are present and the waveform looks like a neat rectangular wave with slightly rounded corners, that’s a healthy, unobstructed breathing pattern. The shape itself, along with derived measurements like respiratory rate, dead space, and the volume of CO₂ eliminated per breath, gives clinicians a continuous window into how well ventilation is working.¹ Any distortion of that normal shape is a signal worth investigating.

Mainstream Versus Sidestream Sensors

Before interpreting a waveform, it helps to know how the measurement is being taken, because the technology itself can shape what you see on the screen. Mainstream capnography places the sensor directly in the breathing circuit, right at the airway. Because the sensor sits where the gas actually flows, it measures CO₂ with minimal delay and high accuracy. Sidestream capnography, by contrast, draws a small sample of gas through thin tubing to a sensor located away from the patient. That sampling process introduces a time lag and can slightly soften or distort the waveform, especially at faster breathing rates.²

For most bedside purposes, either method gives you a usable waveform and a reliable EtCO₂ number. But when you need precise measurements of the waveform’s slope or dead space calculations, mainstream devices are considered the reference standard. Research comparing the two has found that reliable measurement of fine details like the phase II slope and eliminated CO₂ volumes requires a mainstream device.³ Sidestream sensors remain popular in settings like procedural sedation, where attaching a lightweight nasal cannula is far more practical than clipping a bulky sensor to an endotracheal tube.

The Shark Fin and Other Obstructive Patterns

One of the most recognizable abnormal waveforms looks like the dorsal fin of a shark. Instead of a crisp, nearly vertical upstroke followed by a flat plateau, the entire waveform slopes upward gradually from start to finish, never achieving a true plateau. This “shark fin” pattern appears when air cannot leave the lungs evenly, which happens in conditions like asthma or chronic obstructive pulmonary disease. The bronchospasm forces different regions of the lungs to empty at different speeds, so CO₂-rich gas trickles out over a prolonged period rather than arriving all at once.⁴

What makes this pattern clinically powerful is that it responds to treatment in real time. As bronchodilators take effect and the airways open up, the sloping fin gradually straightens back toward a normal rectangular shape. A clinician watching the capnogram during an asthma attack can literally see whether the therapy is working breath by breath, which is faster feedback than waiting for a change in oxygen saturation or a repeat peak-flow measurement. If the shark fin persists or worsens despite treatment, that’s a sign the patient may need more aggressive intervention.

Subtle Waveform Clues in the Operating Room

Two waveform oddities that tend to catch people off guard during surgery are the curare cleft and cardiac oscillations. Neither one is alarming on its own, but misreading them could lead to unnecessary interventions.

The curare cleft (also called a curare crest) appears as a small dip or notch in the middle of the alveolar plateau. It happens when a patient who has been given a neuromuscular blocking agent starts to recover muscle function. The diaphragm makes a weak, spontaneous effort to breathe against the ventilator, briefly disrupting the smooth flow of exhaled gas. Recognizing this dip is useful because it’s an early indicator that the paralytic is wearing off and the patient is beginning to breathe on their own, potentially before a nerve stimulator would pick up the change.⁵

Cardiac oscillations look like small, rhythmic ripples superimposed on the plateau phase. They are caused by the physical motion of the beating heart nudging the lungs and the gas inside them. These ripples match the heart rate and tend to show up in patients with an enlarged heart, high cardiac output, or a thin chest wall. They are generally harmless and can be eliminated by adjusting ventilator settings. In one reported case, adding a small amount of positive end-expiratory pressure abolished the oscillations entirely.⁶

Confirming an Airway Is in the Right Place

One of capnography’s most consequential roles is verifying that an endotracheal tube actually sits in the trachea and not in the esophagus. A tube that accidentally ends up in the esophagus delivers no oxygen to the lungs, and every second it stays there is dangerous. Because the esophagus produces essentially no CO₂, an esophageal placement will show a flat line where a waveform should be. A sustained, repeating four-phase waveform is strong evidence that the tube is in the airway.

In a cadaver study that tested this principle under controlled conditions, every tracheal ventilation produced a characteristic waveform and every esophageal or hypopharyngeal ventilation registered zero CO₂, yielding perfect sensitivity and specificity for distinguishing tracheal from extratracheal placement.⁷ Real-world emergency intubations are messier, though. A meta-analysis based on over 2,000 intubations found that capnography had about 93% sensitivity and 97% specificity in emergency settings. The roughly 7% false-negative rate, where the tube was correctly placed but capnography failed to confirm it, can occur in cardiac arrest patients who have very low pulmonary blood flow and therefore exhale very little CO₂.⁸ That gap is why guidelines recommend using capnography alongside clinical assessment rather than relying on it as the sole check.

Reading the Waveform During Cardiac Arrest

During cardiopulmonary resuscitation, capnography becomes a real-time gauge of how much blood the chest compressions are actually pushing through the lungs. CO₂ can only reach the capnography sensor if blood carries it from the tissues to the pulmonary capillaries, so the EtCO₂ reading during CPR reflects cardiac output. Higher numbers mean the compressions are generating better circulation. Advanced cardiac life support guidelines now recommend using capnography to monitor compression quality for exactly this reason.⁹

Capnography also serves as an early warning system for the return of spontaneous circulation, or ROSC. When the heart starts beating effectively on its own again, there’s a sudden surge of CO₂-rich blood returning from the tissues. On the monitor, this shows up as an abrupt, unmistakable jump in the EtCO₂ value. One study of out-of-hospital cardiac arrests found that EtCO₂ climbed from a median of about 41 mmHg three minutes before ROSC to 57 mmHg one minute after, and adding EtCO₂ data to detection algorithms achieved over 96% sensitivity and 94% specificity for identifying when a pulse had returned.¹⁰

Persistently low readings carry a different message. During prolonged resuscitation efforts, EtCO₂ levels that stay below about 10 mmHg are consistently associated with poor outcomes.¹¹ That threshold has been discussed as a possible criterion for considering whether to continue resuscitation efforts or transition to more advanced rescue strategies like extracorporeal membrane oxygenation.

Catching Respiratory Depression During Sedation

Outside the operating room and the resuscitation bay, capnography has earned a growing role during procedural sedation, where patients receive medications that can slow or stop breathing. The traditional safety net is pulse oximetry, which measures oxygen levels in the blood. The problem is that oxygen saturation can remain normal for a surprisingly long time after someone stops breathing adequately, especially if they’re receiving supplemental oxygen. Capnography, by contrast, detects changes in ventilation almost immediately because it measures breathing itself rather than its downstream effect on oxygen levels.

A prospective study of pediatric procedural sedation illustrated this gap: among episodes where both monitors detected a respiratory event, capnography identified the problem a median of 35 seconds earlier than pulse oximetry. In about 62% of those cases, capnography flagged the issue first. In roughly a quarter of cases the two detected it at the same time, and in only about 14% did pulse oximetry catch it before capnography.¹² Thirty-five seconds may not sound like much, but during sedation it can be the difference between a simple verbal prompt to take a breath and a full airway rescue.

Volumetric Capnography and Dead Space

Standard time-based capnography plots CO₂ against time. Volumetric capnography plots CO₂ against the volume of exhaled gas instead. That switch unlocks a different set of measurements, most notably dead space: the portion of each breath that fills the airways but never participates in gas exchange because it doesn’t contact functioning pulmonary capillaries.¹³

Volumetric capnography can derive an impressive list of values at the bedside, including airway dead space, alveolar dead space, the dead-space-to-tidal-volume ratio, CO₂ output per breath and per minute, and alveolar ventilation.¹⁴ These numbers are especially valuable in the intensive care unit when managing patients on mechanical ventilators. If dead space increases, the ventilator may be set to deliver breaths that look adequate on paper but are actually wasting a large fraction of each tidal volume on gas that never participates in oxygen and CO₂ exchange. Tracking dead space over time helps clinicians fine-tune the ventilator and judge whether a lung recruitment maneuver or a change in body position is actually improving gas exchange.

One technical point worth knowing: there are two different ways to calculate dead space from volumetric capnography, and they don’t measure the same thing. The original approach gives you true dead space, meaning parts of the lung being ventilated but not perfused. A widely used modification conflates true dead space with other causes of ventilation-perfusion mismatch, including blood shunting past non-ventilated lung regions. Clinicians interpreting dead space values need to know which formula was used, because the numbers can tell very different physiological stories.¹³

Using EtCO₂ as a Metabolic Clue

Capnography primarily monitors ventilation, but the EtCO₂ number can also hint at metabolic problems because the body’s acid-base balance directly influences how much CO₂ you breathe out. In diabetic ketoacidosis, for example, the blood becomes acidic and the body compensates by breathing faster and deeper to blow off extra CO₂. That drives EtCO₂ down. Prehospital providers have explored whether a low EtCO₂ reading in a patient with high blood sugar could serve as an early clue to ketoacidosis before lab results are available.

Studies have found a meaningful correlation between EtCO₂ and serum bicarbonate, the standard lab marker for metabolic acidosis. One prehospital study reported a statistically significant correlation between EtCO₂ and both bicarbonate and the anion gap in hyperglycemic patients.¹⁵ An emergency department study found an even stronger correlation between EtCO₂ and bicarbonate levels.¹⁶ The correlation is not tight enough to replace a blood gas measurement, but it can help paramedics and emergency physicians triage more efficiently when lab work is still pending.

Special Challenges in Newborns and Small Children

Interpreting capnography in neonates introduces a set of technical headaches that don’t apply to adults. Newborns breathe fast and move tiny volumes of air, which means the capnography sensor needs to respond almost instantaneously. A slow sensor will blur one breath into the next, producing a waveform that looks like a series of smudged humps rather than clean rectangles. Mainstream sensors need fast response times and minimal added dead space so they don’t force the baby to rebreathe its own exhaled gas. Sidestream sensors need low suction flow rates for the same reason.¹⁷

Beyond equipment challenges, neonatal waveforms can look different from adult waveforms even when nothing is wrong. The plateau phase is often shorter and steeper, simply because the exhaled volume is so small and the breath cycle so rapid. Clinicians who are accustomed to reading adult capnograms sometimes misread a normal neonatal waveform as abnormal, or miss a genuine abnormality because the baseline already looked “off.” Getting comfortable with what normal looks like in this population takes deliberate practice.

How Capnography Became a Safety Standard

Capnography’s journey from specialized research tool to mandatory monitor is one of the clearest success stories in patient safety. In the 1980s, a group of Harvard-affiliated hospitals developed formal monitoring standards for anesthesia that included continuous CO₂ monitoring. The American Society of Anesthesiologists subsequently adopted similar standards for use across the United States, and other countries followed. The cumulative effect has been dramatic: current estimates place anesthesia-related mortality in healthy patients at roughly 1 in 400,000, perhaps ten times lower than in the early 1980s.¹⁸ Capnography was not the only change that drove those numbers down, but it was a key piece of a monitoring revolution that made anesthesia vastly safer.

Today, capnography’s reach extends well beyond the operating room. Emergency departments, intensive care units, ambulances, and procedural sedation suites all use it routinely. As portable sensors become cheaper and smaller, capnography is showing up in settings where it would have been impractical a decade ago, from pediatric dental offices to field hospitals. The waveform remains the same four phases it has always been, but the range of clinical questions it can help answer keeps expanding.