The capnography waveform helps determine severity less by the end-tidal CO₂ number alone than by shape and trend: a normal rectangular plateau with a 35–45 mmHg ETCO₂ suggests an intact ventilation-perfusion match, while a sloped upstroke, shortened plateau, notch, or falling amplitude signals worsening obstruction, dead-space ventilation, or falling cardiac output. Watch how fast the morphology degrades. Acute bronchospasm produces that characteristic "shark fin" within minutes, and a sudden ETCO₂ drop to near zero can mean the airway or the circulation has failed. Every judgment needs the clinical context — the waveform is a monitoring signal, not a diagnosis.

Reading the Four Phases of a Normal Capnogram Before Judging Severity

A normal time-based capnogram has four phases, and severity grading starts with which phase has been distorted. The tracing begins near zero at the start of exhalation (phase I, anatomical dead space), rises steeply through phase II, flattens into the alveolar plateau of phase III, and drops sharply on inspiration through phase IV. The angle between II and III is about 90 degrees in a healthy adult, and the plateau is close to horizontal. Reference values for a normal ETCO₂ sit at 35–45 mmHg at sea level, per standard respiratory physiology teaching.

Phase III is where most diagnostic information hides. Its slope should be flat; a rising slope means the alveoli are emptying unevenly, so the last gas out is not representative of the whole lung. The height of the plateau is the ETCO₂, and the difference between arterial CO₂ and ETCO₂ — the P(a-ET)CO₂ gradient — normally runs 2–5 mmHg in healthy adults. That gradient is the number to compute before you call any value "normal."

Phase I widening tells you dead space has grown. A longer phase I means more of each breath is wasted in conducting airways or unperfused lung, which is why pulmonary embolism can widen phase I while ETCO₂ falls. Phase II slope changes point to obstructive physiology. Phase IV and the inspiratory downstroke matter less for severity but explain artifacts such as cardiogenic oscillations.

The plateau itself is the reference point for every later comparison. Record the baseline shape alongside the value at the start of monitoring — a patient arriving with a sloped plateau has different meaning than one whose plateau was square an hour earlier. In intubated patients, capnography is a continuous standard of care, and the American Society of Anesthesiologists lists exhaled CO₂ monitoring as a standard for basic anesthetic monitoring. That baseline is what makes trend interpretation possible.

Matching Waveform Shape to the Underlying Physiology

Waveform shape is a physiological fingerprint, and each recognizable pattern maps to a mechanism you can act on. The table below pairs the common abnormal shapes with their mechanisms and the severity implication, using ETCO₂ ranges and gradient logic rather than single readings.

Waveform pattern Primary mechanism Typical ETCO₂ behavior Severity implication
Shark-fin (sloped phase III) Uneven alveolar emptying, bronchospasm Often 45–55 mmHg acutely Moderate-to-severe obstruction; watch trend
Shortened plateau, wide phase I Increased dead space (PE, low perfusion) Falling, sometimes under 30 mmHg High severity even when ETCO₂ looks "safe"
Curare-cleft (bifid plateau) Muscle relaxant recovery, spontaneous effort Variable, slight rise on cleft Usually benign; verify with train-of-four
Cardiogenic oscillations Cardiac pulsation during exhalation Stable mean, oscillating trace Low severity; confirms perfusion present
Flat line or sudden near-zero Airway loss, circuit disconnection, cardiac arrest Close to 0 mmHg Immediate emergency
Gradual downward ramp Hypoperfusion, rising dead space, hyperventilation Falling over minutes Rising severity; correlate with blood pressure

The shark-fin shape is the pattern most clinicians recognize in acute asthma or COPD exacerbation. It reflects time-dependent emptying of alveoli with different time constants, and its height correlates loosely with obstruction severity, though no single waveform grade replaces spirometry. The shortened plateau with a widening phase I is the pattern that most often gets missed, because a falling ETCO₂ is easy to dismiss as hyperventilation when it may be dead-space ventilation from a pulmonary embolism or low cardiac output.

Curare-cleft describes a bifid plateau caused by partial recovery from neuromuscular blockade; it matters for ventilator management, not respiratory disease severity. Cardiogenic oscillations are small ripples on the downstroke from cardiac pulsation, and their presence is reassuring because they imply pulmonary blood flow. A flat trace where you expect a waveform should always trigger a circuit and airway check before any physiologic conclusion.

Context decides how much a shape means. An obese patient with a chronically sloped plateau and a 48 mmHg ETCO₂ may be stable at their baseline, whereas the same tracing in a previously healthy young adult after a bee sting suggests severe bronchospasm. The 2022 Global Initiative for Asthma guidelines emphasize objective measurement over symptom impressions, and waveform shape is one such objective clue. A single abnormal shape is a prompt to investigate, not a severity score.

Using ETCO₂ Values, the P(a-ET)CO₂ Gradient, and Dead-Space Fractions to Grade Severity

The absolute ETCO₂ value ranks severity only when you pair it with the arterial-to-end-tidal gradient. In a patient with healthy lungs, ETCO₂ of 40 mmHg implies PaCO₂ near 42–45 mmHg. In a patient with large dead space, ETCO₂ of 30 mmHg can coexist with PaCO₂ of 50 mmHg or higher. The gradient — PaCO₂ minus ETCO₂ — normally 2–5 mmHg, can exceed 10 mmHg in severe ventilation-perfusion mismatch. That arithmetic is the single most underused severity step in bedside capnography.

Dead-space fraction gives you another graded number. Bohr's equation expresses the ratio of dead space to tidal volume; a physiologically normal value is around 0.2–0.3, while ARDS literature reports markedly elevated fractions in severe disease. The exact threshold you use should come from your local protocol; the principle is that a rising dead-space fraction over serial measurements indicates worsening lung efficiency. Without an arterial blood gas to anchor it, ETCO₂ alone cannot tell you whether the patient is hyperventilating or shunting.

Apply the gradient to common scenarios. In a post-arrest patient, an ETCO₂ below 10 mmHg during CPR predicts poor resuscitation outcomes in published resuscitation literature. In a mechanically ventilated patient with a rising ETCO₂ and a normal gradient, suspect hypoventilation rather than dead space. In a spontaneously breathing asthmatic, a rising ETCO₂ despite increased work of breathing suggests fatigue, which is a late and dangerous sign.

Numbers must be corrected for the patient. Pregnancy lowers baseline ETCO₂ because of chronic hyperventilation, and one-lung ventilation during thoracic surgery deliberately broadens the gradient. Without a baseline measurement at the start of monitoring, you have nothing to trend against. The P(a-ET)CO₂ gradient recalibrated after each blood gas converts the capnogram from a shape you watch into a number you can grade.

Trend direction matters more than any snapshot, because the speed of change tells you which emergency you face. A sudden ETCO₂ fall to near zero with loss of waveform signals airway dislodgement, circuit disconnection, or cardiac arrest. A gradual decline over 10–20 minutes with a preserved waveform suggests a developing dead-space process or hypoperfusion. A slow rise with an increasingly sloped plateau suggests progressive obstruction or hypoventilation. The three patterns require different responses and different timelines. The American Heart Association's resuscitation guidance relies on waveform capnography as a quality measure during CPR, and that use depends on trend recognition rather than a single threshold.

Sudden collapse is the most actionable trend. If ETCO₂ drops from 35 to under 5 mmHg in seconds and the trace disappears, check the tube, circuit, and pulse simultaneously — the differential is short. In intubated patients, unrecognized esophageal intubation shows a flat trace, and sustained absence of a waveform is considered confirmation of a misplaced tube in emergency practice. The response is immediate repositioning or re-intubation.

Steady deterioration is harder because each individual reading may look acceptable. A decline from 38 to 32 to 27 mmHg over 30 minutes in a ventilated patient with rising airway pressures suggests worsening dead space. Correlate with blood pressure and consider pulmonary embolism, pneumothorax, or worsening ARDS. The number needs a companion measurement: a falling ETCO₂ with stable vital signs may be increasing dead space, while the same number with a falling blood pressure suggests cardiac output failure.

Improvement is also a trend. A shark-fin that squares off and a plateau that rises toward 35–40 mmHg after bronchodilator treatment is objective evidence of response. GINA's 2022 report emphasizes that treatment response should be assessed with objective measures, and waveform normalization is one of them. Document serial waveforms at defined intervals — every few minutes during acute interventions — or the pattern lives only in memory.

Spotting the Failure Cases That Mimic Respiratory Deterioration

Some waveform changes look respiratory but are actually perfusion or equipment failures, and misreading them delays the real fix. The table below separates mimics, causes, and the discriminating check that settles each one.

Mimic Looks like Actual cause Discriminating check
Falling ETCO₂ with preserved trace Worsening dead space or obstruction Low cardiac output or shock Blood pressure, cap refill, lactate
Flat trace during CPR Loss of airway Poor chest compressions or arrest Compression quality, pulse check
Sloped plateau with normal spirometry New obstruction Cardiogenic oscillations artifact Tracheal tug, cardiac exam
Sudden ETCO₂ spike to >50 mmHg Rising CO₂ production Absorption of CO₂ during laparoscopy Surgical phase, IAP reading
Erratic trace with cough artifact Waveform instability Secretions, condensation in line Suction, water trap inspection
Gradual ETCO₂ fall during transport Ventilator failure Circuit leak or disconnection System check, pressure trace

Hemodynamic coupling is the mimic most often mistaken for a respiratory problem. A falling ETCO₂ with a preserved plateau can reflect reduced pulmonary blood flow from a massive pulmonary embolism, tamponade, or profound hypotension. A published case series in the journal Anaesthesia describes massive pulmonary embolism presenting primarily as a sudden ETCO₂ fall. The discriminating check is blood pressure and cardiac output assessment, not a respiratory diagnosis.

Equipment artifacts mimic deterioration too. A partially obstructed sampling line produces a damped trace with a low amplitude, and a water trap saturated with condensate produces a wavy or flat trace. During laparoscopic surgery, CO₂ insufflation can produce an ETCO₂ rise that reflects absorption rather than hypoventilation, so surgery phase is essential context. If the trace doesn't match the clinical picture, fix the equipment before treating the patient.

A failure case worth remembering is the false reassurance of a normal ETCO₂ in late respiratory failure. A patient with severe asthma nearing fatigue can have a falling ETCO₂ as their tidal volume shrinks and they hyperventilate relative to a shrinking alveolar volume. The waveform becomes smaller in amplitude while the plateau remains short. That pattern is described in respiratory physiology literature as a pre-arrest sign, and it needs escalation, not reassurance.

Comparing Waveform Severity Signs Across Asthma, COPD, ARDS, and Pulmonary Embolism

The following waveform severity signs differ by disease, so the same shape carries different weight depending on the diagnosis. The comparison below maps each condition to the pattern, the tracked number, and the sign that matters most.

Condition Dominant waveform sign Key tracked number Most serious trend
Asthma exacerbation Shark-fin with prolonged phase II ETCO₂ rise with falling amplitude Falling amplitude at high ETCO₂
COPD exacerbation Sloped plateau, delayed emptying Baseline ETCO₂ compared to prior Rising ETCO₂ with rising work of breathing
ARDS Sloped plateau, shortened trace Dead-space fraction, gradient Rising dead space with stable minute ventilation
Pulmonary embolism Wide phase I, low ETCO₂ P(a-ET)CO₂ gradient Sudden fall with stable airway

Asthma's waveform behavior is the most studied. The shark-fin and its correlation with obstruction severity is described in the British Journal of Anaesthesia and in physiology reviews of capnography. The most serious sign is not the degree of finning but the falling amplitude — a sign that tidal volume is collapsing as fatigue sets in. That makes amplitude tracking a stronger severity marker than plateau slope during an acute attack.

COPD patients usually have a chronically sloped baseline plate, so the severity signal is deviation from their own baseline. A patient whose ETCO₂ has been 45 mmHg for years and climbs to 60 mmHg with a flatter plateau represents different severity than a healthy patient reaching the same value. ARDS waveform signs are subtler: the plate shortens, minute ventilation rises, and dead space grows. The gradient recalibrated with serial blood gases is the practical severity tool.

Pulmonary embolism is the condition where waveform signs most often precede diagnosis. A sudden low ETCO₂ with a widened phase I and a sharp rise in the P(a-ET)CO₂ gradient, in a patient with sudden dyspnea or chest pain, should prompt urgent evaluation. No single waveform sign confirms PE, but the pattern combination raises the suspicion enough to justify imaging. Each disease needs its own severity rubric, and the waveform is one input.

Integrating Waveform Data With Blood Gases, Spirometry, and Mechanical Ventilation Metrics

Waveform data become severity data only when integrated with a few anchors. The first anchor is an arterial blood gas near the start of monitoring. PaCO₂ minus ETCO₂ gives you the gradient, and the gradient at baseline lets you project severity from later capnography without repeated gases. The second anchor is spirometry in spontaneously breathing patients — the forced expiratory volume in one second (FEV₁) quantifies obstruction and calibrates what the waveform slope means. The third anchor is the ventilator's derived metrics: plateau pressures, driving pressure, and minute ventilation.

The gradient is not a fixed constant; it changes with the clinical state. After a pulmonary embolism, the gradient may jump from 5 to 15 mmHg because ETCO₂ falls while PaCO₂ stays high. After a recruitment maneuver in ARDS, the gradient can narrow as dead space decreases. That responsiveness makes the gradient a serial severity metric, and the 2022 ARDS consensus guidance from the American Thoracic Society emphasizes using physiologic data, including dead space estimates, to guide management.

Spirometry in awake patients offers a different calibration. In a COPD patient with an FEV₁ of 30% predicted, a sloped plateau is expected at baseline; severity grading shifts to how much the slope worsens during an exacerbation. The waveform's amplitude at known tidal volume can also detect leaks and under-delivery in non-invasive ventilation. Interpret capnometry together with ventilator waveforms and patient effort.

One integration failure is common in teaching: treating the ETCO₂ number as if it were PaCO₂. The gradient reintroduces that distinction, and repeated recalculation keeps the estimate honest. The final integration step is documentation — record the waveform description (slope, amplitude, phase I width) alongside ETCO₂ at defined intervals. Without documentation, the severity trend across a shift disappears.

Building a Bedside Severity Checklist From the Capnogram

A three-step bedside workflow moves from raw observation to a severity judgment without overcomplicating the routine. Each step has a specific trigger and a defined response.

1: Inspect the shape first: measure slope, amplitude, and phase I width relative to the start of monitoring. 2: Pair ETCO₂ with a recent arterial blood gas to compute the P(a-ET)CO₂ gradient, and repeat after interventions. 3: Log the trend at fixed intervals — typically every 5 minutes during acute instability, every 15–30 minutes during stabilization.

The first inspection classifies the pattern. A square plateau with an ETCO₂ of 35–45 mmHg and a gradient of 2–5 mmHg is the low-concern pattern. A shark-fin with rising ETCO₂ and preserved amplitude is moderate concern. A falling amplitude or widening phase I is high concern and warrants rapid escalation. Document which phase changed and by how much.

The gradient computation converts observation into a number. If PaCO₂ is 55 mmHg and ETCO₂ is 30 mmHg, the gradient is 25 mmHg — a severe dead-space or mismatch signal even though the ETCO₂ looks mild. If no recent blood gas exists, order one. Use a locally validated protocol for interpreting the gradient.

The trend log resolves acute versus chronic. Two readings 15 minutes apart are sufficient to classify rapid change. If the severity is worsening, escalate monitoring, oxygen therapy, or airway support, and consider ICU or higher-level review. If it's improving, continue standard care and reduce monitoring intensity. Write the plan into the chart, and keep it consistent with the institution's protocol.

Verifying Severity Differentiation Against the Published Evidence

The evidence base for waveform-based severity is a mix of physiology and clinical study, and knowing its limits prevents overconfidence. Waveform capnography became an anesthesia standard after the American Society of Anesthesiologists adopted exhaled CO₂ monitoring as part of its standards for basic anesthetic monitoring; the standard was adopted by the ASA House of Delegates in 1986 and last amended in 2020 (confirm on the ASA's site). Published clinical evidence for specific waveform shapes, however, is mostly observational and physiologic literature rather than randomized trials.

The shark-fin pattern's association with obstruction is described in physiology reviews and in clinical studies of acute asthma, but no waveform grading system has been validated as a substitute for FEV₁ or blood gases. The P(a-ET)CO₂ gradient's behavior in PE and ARDS is described in critical care literature, including reviews in Critical Care and the American Journal of Respiratory and Critical Care Medicine. Those reviews support the gradient as a severity adjunct, not a standalone classification tool.

Evidence for prehospital waveform capnography as a resuscitation quality tool comes from the American Heart Association guidelines, which rank waveform capnography as a Class I recommendation for confirming and monitoring tracheal tube placement. That is a strong procedural claim, not a claim about diagnosing respiratory severity. The distinction matters: the waveform can confirm a tube, but grading a disease requires the physiology and clinical data described above.

Treat published values as reference points and apply them to your patient. A gradient of 2–5 mmHg is normal only when measured correctly and in a stable patient. A shark-fin is abnormal only when the baseline was square or the patient is symptomatic. Publish the limits alongside the numbers. The strongest verification path is local protocol combined with arterial blood gas correlation — that's the combination that survives audit.

FAQ

Can a normal-looking capnogram hide a severe respiratory condition?

Yes, in specific cases. A patient with severe asthma nearing fatigue can show a shrinking amplitude with a relatively preserved short plateau while their alveolar ventilation is falling, and a pulmonary embolism can lower ETCO₂ to 25–30 mmHg with only subtle phase I widening. The gradient with an arterial blood gas is what exposes the hidden severity. Call the response team if the waveform and the clinical picture disagree.

What does a sudden drop in ETCO₂ to near zero mean?

Sudden near-zero ETCO₂ with loss of the waveform usually means airway loss, circuit disconnection, or cardiac arrest — check the tube, the circuit, and a pulse at the same time. In intubated patients, a flat trace is used to help confirm esophageal intubation, and sustained absence is treated as an emergency. Do not wait for a second reading before acting. Reference values come from emergency airway protocols, not the capnography number alone.

Is the shark-fin waveform a reliable indicator of bronchospasm severity?

The shark-fin correlates with obstruction, but its degree does not reliably grade severity across patients. The more important sign in acute asthma is the falling amplitude as tidal volume shrinks, which appears before the trace goes quiet. That makes amplitude tracking stricter than slope tracking for escalation decisions in acute bronchospasm. Confirm the trend with spirometry or clinical scoring.

How does capnography during CPR guide resuscitation quality?

Waveform capnography provides an ETCO₂ value that reflects pulmonary blood flow during compressions, and a very low ETCO₂ is associated with poor outcomes in resuscitation literature. The American Heart Association's guidelines include waveform capnography as a Class I recommendation for confirming and monitoring tracheal tube placement during CPR. It also provides feedback on compression quality. Check the AHA's current guidelines for the exact thresholds applied in your setting.

Does the P(a-ET)CO₂ gradient cost more than standard monitoring?

The gradient costs nothing beyond the monitoring you already use. An arterial blood gas is the main expense, and it is standard of care in the ICU and ED settings where severity questions arise. The computation is arithmetic from the blood gas and the capnogram. Compare the price locally rather than assuming a national figure, because blood gas pricing varies by lab contract and region.

Which conditions produce a sloped plateau besides asthma?

A sloped or prolonged plateau appears in COPD with uneven emptying, in ARDS with low compliance and dead space, and in any condition where alveolar time constants differ. A sloped plateau in a patient with a prior square plateau warrants a search for new obstruction or worsening dead space. The slope alone does not identify the condition; the gradient and the clinical context do. Treat it as a prompt to investigate.

Can capnography replace an arterial blood gas in respiratory assessment?

No. The capnogram provides continuous ETCO₂ and waveform information, but PaCO₂ requires a blood gas or a validated surrogate like transcutaneous CO₂. The gradient recalibrates the capnogram's meaning, and without a blood gas you can be confidently wrong. Use capnography for continuous trend and use gases to anchor the trend. Consult local protocols for the interval between recalibrations.

This article is general educational information about physiologic monitoring and is not a diagnosis, treatment plan, or substitute for clinical training. It does not guarantee any outcome for any patient. Qualified local clinicians, guidelines, and institutional protocols govern individual care. In emergencies, contact local emergency services immediately; for self-harm risk, contact local crisis services.