Capnography Waveform Interpretation: Beyond the Numbers

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-08-11

Key Takeaways

• The capnogram waveform provides critical diagnostic information beyond the numeric end-tidal CO2 (EtCO2) value, reflecting ventilation, perfusion, airway patency, and equipment function. A normal baseline (Phase I) should be zero; persistent elevation indicates rebreathing due to exhausted CO2 absorbent or faulty valves.
• An upward-sloping alveolar plateau (Phase III) signifies uneven alveolar emptying, often indicative of obstructive airway disease, bronchospasm, or mucus accumulation, which can precede audible wheezes in horses and cattle. A flattened or absent plateau may occur with rapid, shallow breathing, leading to an underestimation of arterial CO2.
• Obstructive patterns, characterized by a rounded waveform with a prolonged expiratory phase, suggest endotracheal tube kinking, bronchospasm, or foreign material; a sudden loss of waveform amplitude indicates complete obstruction or extubation. Rebreathing is identified by an elevated baseline, requiring immediate correction of the breathing circuit or fresh gas flow.
• Cardiac oscillations, small rhythmic waves on the alveolar plateau synchronous with heart rate, confirm airway patency and intact circulation, particularly at low respiratory rates. Their absence, while not always abnormal, may signal falling cardiac output if previously present.
• Artifacts such as airway leaks (sudden dips in the plateau) or slow analyzer response times (blunted waveforms) can mimic pathology and require systematic troubleshooting by disconnecting the sampling line and verifying patency before adjusting ventilator settings or administering medication.
• Species-specific interpretations are crucial: brachycephalic dogs may exhibit upward-sloping plateaus due to baseline airway anatomy, while cats' higher respiratory rates can lead to less defined plateaus and a wider EtCO2 to PaCO2 gradient.

Capnography provides a continuous, noninvasive measure of carbon dioxide in exhaled breath, but the numeric end-tidal CO2 (EtCO2) value represents only a fraction of the diagnostic information available. The waveform itself, its shape, slope, and temporal relationships, offers real-time insight into ventilation, perfusion, airway patency, and equipment function. This article serves the practicing veterinarian who already uses capnography and seeks to extract more clinical value from the trace displayed on the monitor. It answers the question: when the number looks acceptable, what can the shape still tell you, and when the number is abnormal, how does the waveform direct your next intervention?

The content assumes familiarity with basic anesthetic monitoring and focuses on waveform interpretation across species, with attention to differences between dogs, cats, and large animal patients where relevant. Equipment selection and capnograph mechanics are excluded. The discussion emphasizes pattern recognition paired with physiologic reasoning, so that the clinician can distinguish a patient problem from a sampling problem before reaching for a drug or adjusting the ventilator.

At a Glance

Physiologic Basis of the Capnogram

The capnogram traces CO2 concentration against time or volume across the respiratory cycle. Four phases describe the normal trace. Phase I is the expiratory baseline, representing gas from the anatomic dead space that contains negligible CO2. Phase II is the rapid upstroke as alveolar gas mixes with dead space gas. Phase III is the alveolar plateau, where CO2 concentration reflects gas exiting the alveoli. Phase IV is the inspiratory downstroke, when fresh gas dilutes the sample and the trace returns to baseline.

The shape of each phase carries physiologic information because CO2 elimination depends on three linked processes: alveolar ventilation, pulmonary perfusion, and the matching of the two. A normal plateau implies relatively synchronous emptying of alveoli with similar ventilation-perfusion ratios. When disease disrupts this synchrony, the plateau changes character. The numeric EtCO2 value is simply the peak of phase III, and it can remain within an acceptable range while the waveform reveals significant pathology.

The Alveolar Plateau as a Ventilation-Perfusion Window

The slope of phase III deserves particular attention. In a healthy patient, the plateau rises gently because later-emptying alveoli have slightly lower ventilation-perfusion ratios and therefore higher CO2 concentrations. A steeply rising plateau indicates progressive emptying of alveoli with increasingly high CO2 content, a pattern typical of obstructive airway disease, bronchoconstriction, or mucus accumulation. In horses and cattle, this finding may precede audible wheezes. In small animals, a steep plateau with a low EtCO2 value suggests that the sampled gas is not representative of mean alveolar CO2, and the arterial to end-tidal gradient will be widened.

A flattened or absent plateau, where phase II rises directly into a sharp peak, occurs with rapid shallow breathing or when the sampling rate is too slow for the respiratory frequency. This pattern is common in tachypneic cats and small dogs. The EtCO2 reading under these conditions underestimates arterial CO2, and the clinician should interpret the number with caution. The AAHA anesthesia and monitoring guidelines for dogs and cats recommend integrating capnography with other monitoring parameters instead of relying on any single value in isolation.

Abnormal Waveform Patterns and Their Causes

Obstructive Patterns

Partial airway obstruction produces characteriztic changes in both phase II and phase III. The upstroke becomes less steep, the plateau slopes upward, and the overall waveform appears rounded instead of rectangular. This pattern appears with endotracheal tube kinking, bronchospasm, or foreign material in the airway. A sudden change from a normal rectangular trace to a rounded, prolonged expiratory phase should prompt immediate assessment of the breathing circuit and airway before pharmacologic intervention.

Complete obstruction or accidental extubation produces a rapid loss of waveform amplitude with a progressive decrease in EtCO2 over several breaths. The trace may become flat while the patient continues to make respiratory effort. This distinction, a flat trace with visible thoracic excursions versus a flat trace with apnea, is critical and can be made at the bedside without additional equipment.

Rebreathing Patterns

An elevated baseline that does not return to zero indicates rebreathing of CO2. In a circle system, this most commonly reflects exhausted soda lime or a faulty one-way valve. In a Mapleson system, inadequate fresh gas flow produces the same finding. The waveform shows a rising baseline with a corresponding increase in EtCO2, and the inspiratory downstroke may not reach zero. Correcting the underlying cause restores the baseline to zero within one to two breaths. Persistent elevation despite circuit inspection warrants checking the sampling line for water accumulation or partial obstruction.

Cardiac Oscillations

Small rhythmic fluctuations superimposed on the alveolar plateau, synchronous with the heart rate, represent cardiac oscillations. These occur when the heart beats against gas-filled lungs, producing small pressure changes that move CO2-rich gas past the sampling port. They are most visible at low respiratory rates and in small patients. Their presence confirms airway patency and ongoing pulmonary blood flow. Their absence is not necessarily abnormal, but in a patient with a previously visible pattern, loss of cardiac oscillations may signal falling cardiac output.

Sampling and Equipment Artifacts

The capnogram cannot distinguish patient pathology from sampling error without systematic evaluation. A common artifact is the "airway leak" pattern, where the plateau shows a sudden dip followed by recovery, caused by entrainment of room air through a leak around the endotracheal tube cuff. The EtCO2 reading will be falsely low. A slow response time in the analyzer, more common in sidestream devices with long sampling lines, blunts the waveform and reduces the measured peak.

Water or secretions in the sampling line produce erratic spikes or a dampened trace. The MSD Veterinary Manual notes that capnography interpretation requires correlation with clinical assessment, as artifacts can mimic genuine pathology. When the waveform does not match the clinical picture, disconnect the sampling line, verify patency, and confirm the sampling site before adjusting ventilator settings or administering bronchodilators.

Species-Specific Considerations

Dogs typically produce a rectangular waveform with a distinct plateau at normal respiratory rates. Cats, with their higher respiratory frequencies and smaller tidal volumes, often show a less defined plateau, and the EtCO2 value may underestimate arterial CO2 by a larger margin. Brachycephalic breeds may exhibit obstructive patterns even when the airway is considered patent, reflecting their baseline upper airway anatomy.

In horses and ruminants, the larger tidal volumes produce prominent plateaus, and cardiac oscillations are frequently visible. Recumbent large animal patients may show positional changes in the waveform related to ventilation-perfusion mismatching. The WSAVA Global Pain Council Guidelines emphasize that monitoring standards should be adapted to the species and clinical context, and this principle applies equally to capnography interpretation.

Systematic Waveform Assessment

A structured approach to capnography interpretation begins with the same question on every reading: does the waveform match the patient's ventilatory pattern? The capnogram should be evaluated as a complete shape, not as a single number. Assess the baseline, the slope of phase II, the angle between phase II and phase III, the plateau height and slope, and the end-tidal value in sequence. Each component carries distinct diagnostic weight.

The baseline should sit at zero. Any elevation indicates rebreathing of carbon dioxide from the apparatus dead space or exhausted absorbent. Phase II, the rapid upstroke, reflects the transition from carbon dioxide-free gas in the conducting airways to alveolar gas. A prolonged or slurred phase II suggests slow gas sampling, a kinked sampling line, or low fresh gas flow diluting the sample. The angle between phase II and phase III, sometimes called the alpha angle, widens with obstructive disease. The plateau, phase III, should be nearly horizontal in a healthy patient. An upward-sloping plateau indicates progressive emptying of alveoli with different time constants, a hallmark of small airway disease or partial obstruction.

The end-tidal value is read at the peak of the plateau. Compare it against the arterial carbon dioxide partial pressure when an arterial catheter is available. The gradient between arterial and end-tidal carbon dioxide normally ranges from 2 to 5 mm Hg in dogs and cats. A widening gradient suggests increased alveolar dead space, as occurs with hypovolemia, pulmonary embolism, or positioning that reduces cardiac output. The gradient narrows when alveolar dead space decreases, such as after fluid resuscitation. The AAHA anesthesia and monitoring guidelines for dogs and cats recommend continuous capnography during general anesthesia and describe the arterial to end-tidal gradient as a clinically useful indicator of ventilation-perfusion matching.

Decision Points in Waveform Interpretation

When the waveform appears abnormal, the first decision is whether the cause is physiologic or artifactual. This distinction changes the immediate response. Artifacts require equipment troubleshooting. Physiologic abnormalities require patient assessment and possibly intervention.

The following table prioritizes differential diagnoses based on the dominant waveform feature.

The second decision point is whether the end-tidal value is changing rapidly. A sudden decrease in end-tidal carbon dioxide with an unchanged waveform shape suggests a fall in cardiac output or pulmonary perfusion. A sudden decrease accompanied by loss of the plateau suggests disconnection, leak, or esophageal intubation. A gradual increase in end-tidal carbon dioxide with a normal waveform indicates hypoventilation or increasing metabolic production. A gradual decrease with a widening arterial to end-tidal gradient suggests improving ventilation or deteriorating perfusion.

Waveform-Guided Ventilator Adjustments

In mechanically ventilated patients, the capnogram provides real-time feedback for ventilator settings. The plateau shape and slope guide adjustments to inspiratory time, respiratory rate, and tidal volume. A steep phase III slope with a high end-tidal value suggests that the expiratory time is too short for complete alveolar emptying. Increasing expiratory time or reducing respiratory rate allows more complete exhalation and lowers the end-tidal value.

The waveform also reveals auto-positive end-expiratory pressure. When the expiratory limb does not return to baseline before the next inspiration begins, gas trapping is present. This pattern appears as an elevated baseline that rises progressively over successive breaths. The response is to lengthen expiratory time, reduce respiratory rate, or decrease tidal volume. The MSD Veterinary Manual describes capnography as a useful adjunct for detecting rebreathing and assessing the adequacy of ventilation during mechanical ventilation in small animal patients.

Pressure-controlled ventilation produces a characteriztic capnogram. The plateau is often more rounded than with volume-controlled ventilation because inspiratory flow decelerates as the pressure target is reached. This is a normal finding and should not be mistaken for obstruction. The waveform should still show a clear plateau and a sharp transition at end-expiration.

Documentation and Trend Monitoring

Capnography data should be recorded as trends, not isolated readings. A single waveform captures one breath. The clinical value emerges from the pattern over time. Record the end-tidal carbon dioxide value, the respiratory rate, and a description of the waveform shape at regular intervals during anesthesia. Note the arterial carbon dioxide value whenever an arterial sample is obtained and document the gradient.

Trend monitoring detects deterioration before the end-tidal value reaches an alarming threshold. A progressive increase in the arterial to end-tidal gradient with a stable end-tidal value indicates worsening perfusion. A progressive decrease in end-tidal carbon dioxide with stable ventilation suggests falling cardiac output. Both patterns warrant immediate investigation even when the absolute values remain within reference ranges.

Document the waveform description using standard terms: normal, elevated baseline, prolonged phase II, sloping plateau, shark-fin, cardiac oscillations, or flat. Include the suspected cause and the intervention taken. This documentation supports anesthetic record review and provides a baseline for subsequent procedures. The WSAVA global pain council guidelines emphasize that continuous monitoring of physiologic variables, including ventilation, is an expected standard during procedures requiring general anesthesia in companion animals.

Species and Patient Factors That Change Interpretation

The same waveform shape can carry different implications across species. Brachycephalic dogs commonly show an upward-sloping plateau during spontaneous ventilation because of upper airway obstruction and increased work of breathing. This finding is expected in this population but still warrants attention to airway positioning and the depth of anesthesia. Cats are more prone to bronchospasm and may develop a shark-fin pattern with minimal provocation. The pattern should resolve with bronchodilator administration if airway reactivity is the cause.

Obese patients show a sloping plateau because of reduced chest wall compliance and early airway closure in dependent lung regions. The slope may improve with positioning changes or increased positive end-expiratory pressure. Pregnant patients have reduced functional residual capacity and may show a steeper phase III slope as a result of cephalad displacement of the diaphragm.

In large animal species, the sampling system and respiratory rate change the waveform appearance. Horses have lower respiratory rates and larger tidal volumes, producing a prolonged plateau that may show cardiac oscillations. Ruminants may show an irregular baseline because of eructation, which introduces methane and carbon dioxide into the airway. This artifact is transient and should not be mistaken for rebreathing.

Patient temperature alters the relationship between end-tidal carbon dioxide and metabolic production. Hypothermia reduces carbon dioxide production and lowers the end-tidal value without a change in ventilation. The waveform shape remains normal. Rewarming increases carbon dioxide production and raises the end-tidal value. These changes should be interpreted in the context of the temperature trend instead of as isolated ventilation events.

The correct response to an abnormal waveform depends on the patient's cardiovascular status. A sloping plateau in a hypotensive patient may reflect low cardiac output and increased alveolar dead space instead of airway disease. Assess blood pressure, pulse quality, and mucous membrane color before treating the waveform as a primary airway problem. The waveform is one component of the anesthetic monitoring picture and should be interpreted alongside heart rate, blood pressure, oxygen saturation, and depth of anesthesia.

Recognized Complications and Early Detection

Capnography failures fall into three categories: patient-related, equipment-related, and interpretation-related. The most consequential patient-related complication is unrecognised hypoventilation progressing to hypercapnia with respiratory acidosis. Early detection relies on trend monitoring instead of single readings. A rising baseline with an unchanged plateau indicates rebreathing, while a rising plateau with a stable baseline indicates increasing alveolar dead space or falling cardiac output. The AAHA anesthesia and monitoring guidelines for dogs and cats recommend continuous monitoring of ventilation parameters with capnography as part of the minimum monitoring standard.

A second failure mode is the false reassurance of a normal numeric end-tidal CO2 when the waveform is distorted. A short, steep plateau with a normal numeric value can reflect a rapid shallow breath where alveolar gas never fully sampled. The waveform shape, not the number, reveals the problem. Similarly, a normal capnogram during spontaneous ventilation does not exclude atelectasis or shunt, because the waveform reflects ventilation, not oxygenation. Pulse oximetry and blood gas analysis remain complementary.

Equipment failure modes include water condensation in the sampling line, partial line occlusion from secretions, and sidestream sampling rates too low for the patient's respiratory rate. In very small patients, a slow sampling rate can produce a falsely low end-tidal CO2 because the sample dilutes with room air. Detecting this requires comparing the waveform to the respiratory rate observed on the rebreathing bag or thoracic excursions.

Common Interpretation Errors and Corrections

Less experienced clinicians frequently mistake the expiratory upstroke for the inspiratory downstroke, inverting the entire interpretation. The capnogram reads left to right: a rapid rise at the start of expiration, a plateau, then a rapid fall at inspiration. Correcting this error requires anchoring on the respiratory cycle, not the screen.

A second common error is attributing every abnormal waveform to equipment malfunction. A persistently flat waveform with a low numeric value may represent apnoea, a disconnected sampling line, or a sampling line kinked at the endotracheal tube connector. The discriminating check is the presence of thoracic movement and the anesthesia machine's fresh gas flow. If the patient is breathing and the waveform is flat, the problem is sampling. If the patient is not breathing, the problem is ventilatory.

A third error is overinterpreting minor waveform variations. Small notches on the descending limb can represent the closing of the expiratory valve or a leak around the endotracheal tube cuff, neither of which requires intervention. The corrective action is to verify the waveform is reproducible across three consecutive breaths before acting.

Limitations of Current Evidence

The veterinary literature on capnography waveform interpretation relies heavily on extrapolation from human medicine and from physiologic first principles. Direct comparative studies across species are limited. The MSD Veterinary Manual notes that normal end-tidal CO2 values and waveform morphology vary with species, body position, and ventilatory mode, but the quantitative thresholds for abnormality are not uniformly established.

Expert opinion differs on the clinical significance of cardiac oscillations. Some authors consider them a benign artifact, while others view them as an early indicator of reduced thoracic compliance or increased airway resistance. The evidence base does not resolve this disagreement. Similarly, the role of capnography in detecting pulmonary embolism in veterinary patients is extrapolated from human data and should not be treated as a diagnostic standard.

A further limitation is the absence of validated waveform classification systems for veterinary patients. Human scoring systems for capnogram shape have not been adapted or validated across species. Clinicians should therefore treat waveform interpretation as a pattern-recognition skill supported by physiology, not as a validated diagnostic test.

Referral, Consultation, and Reporting

Referral or specialist consultation is warranted when capnography reveals a pattern that cannot be explained by the current anesthetic protocol, when the waveform deteriorates despite corrective intervention, or when the patient requires mechanical ventilation beyond the immediate post-anesthetic period. Persistent hypercapnia with a normal waveform in a spontaneously breathing patient warrants blood gas analysis to confirm the severity of respiratory acidosis before adjusting ventilator settings.

Laboratory involvement is indicated when capnography suggests a metabolic component, such as a falling end-tidal CO2 with a normal respiratory rate, which may indicate metabolic acidosis with compensatory hyperventilation. Electrolyte and blood gas panels clarify the picture.

Regulatory reporting obligations vary by jurisdiction and production system. In food animal practice, anesthetic complications are not typically reportable, but adverse events associated with specific drugs may be subject to pharmacovigilance reporting. The WOAH terrestrial animal health standards address welfare during procedures, and clinicians should consult regional requirements where anesthetic complications intersect with welfare standards. The AVMA practice resources provide guidance on professional obligations in these circumstances.

Troubleshooting Table

Frequently Asked Questions

How Should I Interpret Capnography When Only a Numerical EtCO2 Value Is Available?

A number alone cannot distinguish hypoventilation from rebreathing, airway obstruction from equipment malfunction, or a valid alveolar sample from a diluted one. Treat the numeric value as a screening tool, not a diagnosis. If the reading conflicts with the patient's clinical status, pulse oximetry, or blood gas analysis, assume the waveform is abnormal until proven otherwise. Check the sampling line for kinks, water, or disconnection, and confirm the respiratory rate matches the displayed waveform frequency. When waveform display is unavailable, intermittent arterial or venous blood gas sampling provides the reference standard for ventilation assessment. The AAHA anesthesia and monitoring guidelines for dogs and cats recommend integrating multiple monitoring modalities instead of relying on any single parameter.

What Can I Do When My Practice Lacks Capnography Equipment?

Clinical assessment of ventilation relies on observation of thoracic excursions, reservoir bag movement, and auscultation of breath sounds, supplemented by mucous membrane color and pulse oximetry. These methods detect advanced hypoventilation but miss subtle changes in alveolar gas exchange. Capnography is the most sensitive noninvasive ventilation monitor, yet its absence does not preclude safe anesthesia. Increase the frequency of patient assessment, monitor trends in heart rate and blood pressure, and use blood gas analysis when available for prolonged procedures or compromised patients. Document the monitoring limitations in the anesthetic record. The MSD Veterinary Manual provides species-specific guidance on anesthetic monitoring and the clinical signs that indicate inadequate ventilation when electronic monitoring is unavailable.

How Does Capnography Interpretation Differ in Birds and Exotic Mammals?

Birds possess a cross-current gas exchange system and a more efficient carbon dioxide excretion mechanism than mammals, producing lower arterial carbon dioxide tensions and different alveolar plateau characteriztics. Their high respiratory rates and small tidal volumes challenge sidestream sampling, often yielding falsely low EtCO2 values. Rabbits and other small mammals similarly have rapid, shallow breathing patterns that may not produce a stable plateau. The capnogram in these species frequently appears as a continuous upslope without a distinct phase III. Interpretation should therefore emphasize trends instead of absolute values, and capnography should be corroborated with blood gas analysis when precise ventilation assessment is required. Reference ranges for normal EtCO2 in these species differ from those for dogs and cats.

How Should I Document Capnography Findings in the Anesthetic Record?

Record the numeric EtCO2 value, the respiratory rate, and a description of the waveform shape at regular intervals, typically every five minutes during stable anesthesia and more frequently during changes. Note the inspired carbon dioxide level when rebreathing is suspected. Describe abnormal waveforms using standardized terminology such as "absent plateau," "shark-fin pattern," or "cardiac oscillations" instead of subjective terms like "weird" or "irregular." Document the intervention performed and the waveform response. Serial waveform descriptions allow detection of gradual deterioration that isolated numeric values may miss. The AVMA practice resources emphasize that complete medical records support continuity of care and provide a defensible basis for clinical decisions.

How Do I Explain an Abnormal Capnogram to a Client or Referring Veterinarian?

Use language that conveys the clinical significance without alarming the owner unnecessarily. Explain that the monitor measures carbon dioxide in the breath and that the shape of the tracing helps the team assess how well the patient is breathing during anesthesia. State what action was taken, for example repositioning the airway, adjusting the ventilator, or changing the breathing circuit. Avoid speculative language about causes that remain unconfirmed. For referring veterinarians, describe the waveform pattern, the suspected cause, the intervention, and the patient's response in the discharge summary. The WSAVA Global Pain Council guidelines note that clear communication about monitoring findings supports trust and informed decision-making throughout the perioperative period.

When Should I Abandon Waveform-Guided Adjustments and Switch to Blood Gas Analysis?

Blood gas analysis is indicated when the capnogram shows a progressive loss of the alveolar plateau despite corrective interventions, when the EtCO2 to arterial carbon dioxide gradient is suspected to be abnormally large, or when the patient's clinical status deteriorates without a corresponding waveform change. Patients with severe pulmonary disease, shock, or marked body temperature changes may have unstable gradients that make capnographic estimation of arterial carbon dioxide unreliable. In these situations, the waveform remains useful for detecting rebreathing and circuit malfunction, but ventilation should be guided by arterial blood gas values. The AAHA anesthesia and monitoring guidelines for dogs and cats recommend blood gas analysis as the definitive method for assessing ventilation when noninvasive monitoring is ambiguous or conflicting.

Related Clinical & Scientific Guides

• Anesthetic Machine Leak Testing and Pressure Checks: A Step-by-Step Protocol
• Anesthetic Depth Assessment: Reflexes, Eye Position, and Ventilation
• Anesthesia for Patients with Obesity: Challenges and Solutions

References and Further Reading

• AAHA Anesthesia and Monitoring Guidelines for Dogs and Cats. AAHA.
• WSAVA Global Pain Council Guidelines. WSAVA.
• MSD Veterinary Manual, Professional Edition. MSD Veterinary Manual.
• American Veterinary Medical Association Practice Resources. American Veterinary Medical Association.
• WOAH Terrestrial Animal Health Code. WOAH.

Related Articles

• Capnography Waveform Interpretation in Veterinary Patients
• Anesthetic Risk Assessment: Beyond the ASA Score
• Anesthetic Complications in Brachycephalic Dogs: Beyond Airway Obstruction
• Blood Pressure Monitoring During Anesthesia: Methods and Interpretation
• Preanesthetic Bloodwork Interpretation: Minimum Database and Clinical Decision Points

This article is educational professional reference material for veterinary audiences. It is not a substitute for veterinary diagnosis, individual clinical judgment, current product labeling, or applicable regulatory requirements.