Epileptic Seizures and VOC Detection
Yes, genuine epileptic seizures do produce measurable changes in volatile organic compounds (VOCs) in both sweat and breath, with robust evidence demonstrating that trained detection dogs can identify these unique chemical signatures with high accuracy.
Evidence for Seizure-Related VOC Changes
The most compelling evidence comes from prospective canine detection studies. In a rigorously designed EMU study, trained service dogs demonstrated a 93.7% probability of correctly distinguishing between ictal (during seizure) and interictal (between seizures) sweat samples (OR: 14.89,95% CI: 9.27,23.90) 1. This represents strong evidence that epileptic seizures produce a distinct, measurable VOC signature.
Pre-Seizure Detection Window
More remarkably, dogs identified the unique seizure scent before 78.7% of all captured seizures, with an 82.2% probability (OR: 4.60,95% CI: 0.98,21.69) that a positive detection would predict an upcoming seizure 1. The warning phase averaged 68.2 minutes before clinical-electrical seizure onset, while the scent persisted for approximately 81 minutes post-ictally. This temporal pattern strongly suggests that VOC changes are not merely a consequence of seizure activity but may reflect underlying neurophysiological changes preceding the clinical event.
Specificity to Epileptic Seizures
Critical to establishing this as a true epileptic phenomenon: 18 of 19 non-epileptic seizure (NES) events did not produce the unique seizure scent 1. This specificity distinguishes genuine epileptic activity from psychogenic events, addressing a major confounding factor in earlier literature.
The Chemical Identity: Menthone as a Potential Biomarker
Research suggests the VOC menthone may be the key compound involved 2. In a hypothesis-testing study, trained seizure-detection dogs identified fear-scented sweat (collected during scary movie viewing) as their seizure scent in 4 of 5 participants, with 82% sensitivity and 100% specificity when distinguishing fear or seizure sweat from other samples 2. The near-perfect inter-rater agreement among dogs (kappa = 0.814) supports the reliability of this finding.
This connection between fear and seizure scent implies shared neuroanatomical pathways—specifically, that seizure propagation may involve primitive neural networks associated with alarm pheromone production. This represents a fascinating convergence of epilepsy neurobiology and evolutionary stress response systems.
Technical Context: VOC Detection in Medicine
While the epilepsy-specific research is relatively recent, the broader field of VOC biomarker detection is well-established. The European Respiratory Society's 2017 technical standards document confirms that VOCs arise in exhaled breath, urine, blood, saliva, and are emitted by skin 3. These compounds can have endogenous metabolic origins or arise from bacteria, with concentrations influenced by blood-gas partition coefficients, cardiac output, and alveolar ventilation 3.
Important Caveats for VOC Measurement
The ERS guidelines emphasize multiple confounding factors that affect VOC detection 3:
- Diet, exercise, and smoking (typically prohibited 2-3 hours before testing)
- Medications (including antiepileptic drugs themselves, which can produce volatile metabolites)
- Environmental VOCs (requiring ambient air correction or filtered breathing)
- Humidity and water vapor (affecting sensor responses)
- Breathing patterns and flow rates (influencing VOC concentrations)
These technical challenges explain why standardized protocols are still under development, and why the field remains largely in the discovery phase 3.
Clinical Implications and Current Limitations
The measurement of seizure-related VOCs is not yet ready for routine clinical application. While canine detection has proven remarkably accurate in controlled research settings 1, 4, several barriers remain:
- Lack of standardized collection methods for clinical use
- Need for validation across diverse epilepsy populations and seizure types
- Absence of portable, real-time detection devices (though VOC profiling technology is advancing 5)
- Unknown generalizability across different epilepsy syndromes
The 2018 National Association of Medical Examiners guidelines for epilepsy death investigation do not yet incorporate VOC testing, focusing instead on traditional autopsy, toxicology, and antiepileptic drug levels 6. This reflects the current gap between research findings and clinical implementation.
Future Directions
The identification of specific VOCs like menthone opens pathways for developing rapid, unbiased seizure detection methods that could eventually replace or supplement canine detection 4. Advanced VOC profiling using gas chromatography-mass spectrometry (GC-MS) or electronic nose (e-nose) technology may enable real-time seizure prediction devices, potentially transforming epilepsy management by addressing the unpredictability that significantly impairs quality of life 4.
Bottom line: The science confirms that epileptic seizures produce measurable VOC changes in sweat and breath, with a temporal window that may allow pre-seizure detection. However, translation to clinical practice requires further validation and technological development of practical detection systems.