A sudden and sustained increase in end-tidal carbon dioxide during cardiopulmonary resuscitation most often indicates return of spontaneous circulation, known as ROSC. When chest compressions restart forward blood flow, that blood delivers accumulated CO2 to the lungs, and the resulting exhaled surge is large enough to register on the capnograph. The rise is not random noise; it is a physiologic signal that pulmonary perfusion has resumed.
What end-tidal CO2 measures during arrest
End-tidal CO2 is the peak partial pressure of carbon dioxide in exhaled gas at the end of each breath. During cardiac arrest, exhaled CO2 depends almost entirely on one variable: how much blood the lungs receive. No blood flow means no CO2 delivery to the alveoli, so the reading collapses.
The mechanism runs in two steps. First, chest compressions generate a fraction of normal cardiac output, and that flow carries venous CO2 into the pulmonary capillaries. Second, the CO2 diffuses across the alveolar membrane and is exhaled. Because the membrane is rarely the limiting factor in arrest, the capnograph effectively acts as a crude cardiac-output meter. Higher numbers mean better forward flow.
This makes the value dynamic rather than diagnostic in isolation. It reflects compression quality, patient metabolism, and the underlying cause of arrest at the same time. A single number tells you little. A trend, especially a sudden sustained change, tells you a great deal.
Why a sudden rise signals ROSC
When the heart regains an organized rhythm and begins ejecting blood, pulmonary perfusion jumps within seconds. Venous blood that has been stagnating and accumulating CO2 floods the lungs all at once. The capnograph records this as a sharp step up in end-tidal CO2, often by a wide margin and sustained across several breaths.
Two physical details explain why the increase is abrupt. The lungs act as a reservoir, so arriving CO2 is exhaled on the next few breaths rather than diluted slowly. And the CO2 stores in venous blood are large after minutes of low flow, so the first restored circulation delivers a concentrated bolus. The waveform steps up and stays up.
Distinguishing this from artifact matters. A transient spike that returns to baseline within one or two breaths is more consistent with a compression change, a tube adjustment, or a small leak. A sustained plateau at a new, higher level is the ROSC pattern, particularly when the rhythm also improves.
Quantified context and thresholds
Resuscitation research has established approximate ranges that guide interpretation, though guidelines caution against treating any single value as absolute. A persistent end-tidal CO2 below roughly 10 mm Hg during CPR is associated with a very low likelihood of survival and is a recognized signal of poor compression quality or poor prognosis. Values that rise above approximately 40 mm Hg without other changes warrant a ROSC check.
The change itself carries information. An abrupt sustained increase, commonly described as a step up of roughly 10 mm Hg or more from the arrest baseline, is the practical trigger to pause compressions briefly and check for a pulse and organized rhythm. The absolute number matters less than the jump and its persistence.
| Capnography pattern | Typical ETCO2 behavior | Likely interpretation | Immediate action |
|---|---|---|---|
| Persistent very low value | Stays under about 10 mm Hg | Poor compression quality or poor perfusion | Optimize depth, rate, and recoil; verify airway |
| Sudden sustained rise | Steps up 10 mm Hg or more and holds | Return of spontaneous circulation | Brief rhythm and pulse check |
| Transient spike only | Rises for 1 to 2 breaths, then falls back | Artifact from compression or airway change | Recheck technique, continue CPR |
| Gradual slow drift up | Inches upward over minutes | Improving compression or warming | Reassess quality, keep monitoring |
| Abrupt drop to near zero | Falls toward zero | Tube dislodgement or equipment failure | Verify airway and circuit immediately |
| High baseline with shock | Elevated and climbing | Hyperthermia, sepsis, or CO2 retention | Address underlying cause alongside CPR |
The time course of a ROSC signal
The sequence unfolds over seconds to minutes, and knowing the expected timing helps separate signal from noise.
| Elapsed time from compression pause | Expected observation | What it means |
|---|---|---|
| 0 to 5 seconds | Capnograph still shows prior value | Sensor lag; do not interpret yet |
| 5 to 15 seconds | First sharp increase appears | Restored ejection reaching lungs |
| 15 to 30 seconds | Value plateaus at new higher level | Sustained perfusion, ROSC likely |
| 30 to 60 seconds | Plateau holds or drifts slightly | Confirm with pulse and rhythm check |
| 1 to 3 minutes | Stabilizes or trends with ventilation | Post-ROSC phase begins |
| Beyond 3 minutes | May fall again if re-arrest occurs | Reassess immediately |
The practical lesson is patience with the first few seconds. Capnography sensors have a short response time, and a rushed reading can be misleading. Waiting through a full plateau confirms the finding.
Edge cases that mimic ROSC
Several conditions produce a sudden sustained rise without the heart restarting. Sodium bicarbonate administration releases CO2 and can cause a sharp upward step that lasts as the drug distributes. The same is true of CO2 that builds up when ventilation is deliberately reduced or when a previously misplaced airway is corrected.
Hyperthermia and sepsis raise metabolic CO2 production, though usually as a gradual trend rather than a step. Reperfusion after thrombolysis or a revascularization procedure can also raise readings as circulation improves gradually. Each mimics part of the ROSC pattern, which is why no clinician should act on capnography alone.
The counterintuitive professional caveat: a rising end-tidal CO2 is not always good news in the moment. If it climbs steeply after bicarbonate or a ventilation change, the rise can reflect retained CO2 rather than renewed flow. That retained CO2 can worsen acidosis and does not confirm ROSC by itself. Interpret the number in the context of the rhythm, the pulse check, and what was just done to the patient.
Common misconceptions
The most frequent error is treating any upward bump as ROSC. A small transient spike after a compression pause is often just the lung emptying of accumulated gas. Sustained change, not single-breath change, is the test.
Another misconception is that a normal or high end-tidal CO2 rules out ongoing arrest. If ventilation is being overdelivered or the patient is hypermetabolic, readings can look reassuring while perfusion remains inadequate. The value must be read against the arrest baseline, not against normal physiology.
A third error is ignoring a falling value. A sustained drop during CPR usually signals declining compression quality, a displaced tube, or a failing circuit. Some clinicians interpret a drop toward zero as a poor prognostic sign. The correct response is to fix the reversible cause first, then reassess.
A final misconception treats capnography as a substitute for pulse checks. It is an adjunct. Guidelines position it as a way to reduce unnecessary interruptions and to confirm clinical findings, not to replace them.
Practical actions and red flags
Use end-tidal CO2 as a continuous trend, not a snapshot. Record the value at the start of the arrest and track changes over time. Set a mental threshold for the sustained step up and respond promptly.
Red flag: a step up of about 10 mm Hg or more that holds for several breaths: pause compressions briefly and check rhythm and pulse. Red flag: a sustained value above about 40 mm Hg without other explanation: verify the airway and check for ROSC. Red flag: a fall toward zero: immediately verify tube position and the capnography circuit. Red flag: a persistent value under about 10 mm Hg: reassess compression depth, rate, and full recoil, and consider reversible causes. Red flag: a sharp rise right after bicarbonate or a ventilation change: wait and confirm with a pulse check before concluding ROSC.
After any confirmed ROSC, shift attention to post-arrest care. Oxygenation and ventilation targets change, and the capnograph becomes a tool to keep CO2 in a normal range rather than a marker of flow. Monitor for re-arrest, since a renewed fall in end-tidal CO2 can signal the heart stopping again.
Reconciling the thresholds explicitly: the under-10 mm Hg figure is a quality and prognosis benchmark, the 40 mm Hg figure is an upper reference, and the sustained 10 mm Hg step is the practical action trigger. These are not competing rules. They answer different questions: how well are compressions working, is perfusion adequate, and has circulation returned.
Limitations and what the evidence cannot settle
Capnography reflects physiology indirectly, and several variables confound it. Body temperature, metabolic rate, the cause of arrest, and the accuracy of the sensor all shift the reading. A patient with severe lung disease may produce a different baseline than a healthy adult, so absolute numbers travel poorly between patients.
The thresholds themselves come from observational data rather than tightly controlled trials, and guidelines describe them as supportive rather than definitive. No single value has been validated to confirm ROSC across all populations and settings. Pediatric, obstetric, and trauma patients may behave differently, and evidence in those groups is thinner.
Related conditions can also drive a sustained rise. A patient who was never truly in arrest, or who has a severe metabolic or ventilatory disturbance, can produce a step up that looks like ROSC. Likewise, a rise caused by a drug or a ventilation change is not perfusion. This is the limit of what the signal alone can conclude: it narrows the possibilities but does not close them. Clinical confirmation remains essential.
This article is general educational information about physiology and monitoring, not a diagnosis or a treatment plan. It does not promise any outcome. For any decision about a patient in cardiac arrest, follow local protocols and consult a qualified clinician or emergency service immediately.