How Antidepressants Affect the EEG

Overview

 

Understanding how psychotropic drugs affect the EEG is essential for clinicians engaged in neurofeedback and other neuromodulatory interventions. Medications can significantly influence a client’s presenting symptoms, baseline EEG patterns, and responsiveness to neurofeedback training. These effects are neither uniform nor easily predicted. Instead, they are determined by a complex interplay between drug class, degree of sedation, dosage, plasma concentration, and individual patient characteristics such as age, neurological history, and metabolic rate.

​

While members of a drug class often share common pharmacodynamic properties, their effects on the EEG may diverge. For example, within the antidepressant class, sedating tricyclics like amitriptyline increase low-frequency delta and theta power, while activating SSRIs like citalopram increase beta and decrease alpha, depending on dosage and patient arousal levels (Knott, 2000; Nissen et al., 2020). Even within the same agent, different doses can produce qualitatively different EEG effects.

 

A single dose of a psychotropic drug can markedly alter EEG activity within 1 to 3 hours, with changes often widespread and bilaterally symmetrical (Knott, 2000). The drug's effect is mediated by its plasma concentration, which in turn is shaped by factors such as liver metabolism, body weight, and drug interactions.

​

Common EEG responses to psychotropic medications include slowing of the background rhythm, elevation of beta activity (particularly in the 18–30 Hz range), emergence of epileptiform activity, or the appearance of triphasic waves, especially in toxic states. Widespread delta and increased theta activity are typical of sedative or encephalopathic effects and may signal excessive cortical suppression (Blume, 2006). These changes can complicate EEG interpretation, particularly during assessment and treatment planning.

 

Recent advances have highlighted the clinical importance of EEG biomarkers in explaining medication response and failure. Research by Swatzyna et al. (2024) has shown that certain EEG patterns—such as spindling excessive beta, focal slowing, diffuse encephalopathy, and isolated epileptiform discharges—predict poor response to specific drug classes.

 

For instance, patients with spindling beta may experience symptom worsening when treated with SSRIs or benzodiazepines, which further amplify beta activity. Similarly, patients with subclinical epileptiform discharges are more vulnerable to adverse effects from antidepressants and stimulants, which reduce seizure threshold. These biomarkers cannot be identified through clinical interviews alone but require formal EEG analysis. Their presence not only guides more effective pharmacological decision-making but also helps avoid iatrogenic complications.

 

Thus, incorporating an understanding of drug-induced EEG effects and endogenous EEG biomarkers enhances both diagnostic precision and therapeutic success. For clinicians practicing neurofeedback, reviewing a client’s full medication list—and understanding how those drugs modulate cortical activity—is not optional. It is fundamental to ensuring safe, individualized, and scientifically grounded care.

​

​

The Clinician Detective

 

Dr. Ronald Swatzyna, Director and Chief Scientist of the Houston Neuroscience Brain Center, inspired our Clinician Detective series and the EEG-informed psychiatry perspective of this post. In his Association for Applied Psychophysiology and Biofeedback (AAPB) Distinguished Scientist address, he reminded his audience that the DSM-5 advises systematically ruling out general medical conditions before assigning a psychiatric diagnosis to ensure diagnostic validity and appropriate treatment planning.

​

​

 

 

 

 

 

 

 

 

He argued that in abrupt onset and refractory cases, EEG biomarkers should challenge neurofeedback providers and their medical colleagues to become detectives to identify their causes. This collaborative approach allows each professional to contribute to assessment while "staying in their lane."

 

Dr. Swatzyna generously mentors professionals in his investigative method, including raw EEG interpretation, to train the next generation of neurofeedback clinicians.

Dr. Swatzyna and colleagues have highlighted the clinical importance of EEG biomarkers in explaining medication response and failure. Swatzyna et al. (2024) have shown that certain EEG patterns—such as spindling, excessive beta, focal slowing, diffuse encephalopathy, and isolated epileptiform discharges—predict poor response to specific drug classes.

 

For instance, patients with spindling beta may experience symptom worsening when treated with SSRIs or benzodiazepines, which further amplify beta activity. Similarly, patients with subclinical epileptiform discharges are more vulnerable to adverse effects from antidepressants and stimulants, which reduce seizure threshold. These biomarkers cannot be identified through clinical interviews alone but require formal EEG analysis. Their presence not only guides more effective pharmacological decision-making but also helps avoid iatrogenic complications.

 

Thus, incorporating an understanding of drug-induced EEG effects and endogenous EEG biomarkers enhances both diagnostic precision and therapeutic success. For clinicians practicing neurofeedback, reviewing a client’s full medication list—and understanding how those drugs modulate cortical activity—is not optional. It is fundamental to ensuring safe, individualized, and scientifically grounded care.

​

​

Case Example

 

 

Daniel, a 16-year-old high school sophomore, was referred to psychiatric services after experiencing a sudden and uncharacteristic decline in mood, energy, and motivation. His parents described a vibrant, academically engaged, and socially active adolescent who, within the span of 2 months, had become withdrawn, irritable, and disengaged from previously enjoyed activities. His teachers noted he had stopped turning in assignments, was sleeping in class, and had begun isolating himself from friends. He reported waking up unrefreshed, struggling to concentrate, and experiencing a persistent sense of heaviness and mental fatigue. He denied suicidality but expressed feeling “numb” and “slowed down.”

 

 

Daniel had no prior psychiatric diagnoses, no family history of mood disorders, and no known psychosocial stressors or traumatic events. His primary care physician started him on fluoxetine, an SSRI commonly used for adolescent depression. After 4 weeks, Daniel reported increased fatigue and flattening of emotion but no improvement in motivation or affect. When his dose was increased, he began experiencing frequent headaches, inner restlessness, and reduced sleep quality. The medication was discontinued, and he was referred for a second opinion.

​

 

During intake with a clinician using Dr. Swatzyna’s EEG-informed assessment model, it became apparent that Daniel’s symptoms had developed shortly after he began training for the school swim team in a recently reopened aquatic facility. When asked directly, he mentioned that the indoor pool area “always smelled strange” and that several teammates had also complained of headaches and fatigue during practice.

​

 

Resting EEG was conducted, revealing several abnormalities. The most prominent was diffuse slowing, with widespread increase in theta activity (4–7 Hz) across both hemispheres, particularly in frontal and temporal leads. His posterior alpha rhythm was poorly defined, with low amplitude and inconsistent peak frequency. Additionally, his EEG displayed intermittent frontal delta bursts, a pattern suggestive of low-grade encephalopathic activity. These findings indicated global cortical underactivation, a common electrophysiological correlate of toxic or metabolic encephalopathy, rather than idiopathic depression.

​

 

The EEG findings raised immediate concern for an environmental contributor to Daniel’s symptoms. A multidisciplinary evaluation was initiated. Environmental testing of the pool facility uncovered elevated levels of chloramines—chemical byproducts formed when chlorine reacts with organic matter in the water. Poor ventilation had led to accumulation of these irritants, particularly nitrogen trichloride, known to cause respiratory, cognitive, and mood-related symptoms in swimmers and pool staff. Blood testing also revealed mild elevations in inflammatory markers, including CRP and interleukin-6, supporting the suspicion of systemic neuroimmune activation.

​

 

Daniel was advised to discontinue swim practice and avoid further exposure. He began a structured treatment plan focused on neurological recovery: a nutrient-dense anti-inflammatory diet, high-dose omega-3 fatty acids, aerobic exercise limited to outdoor settings, and neurofeedback aimed at stabilizing his thalamocortical rhythm and restoring posterior alpha activity.

​

 

Psychotropic medication was not reintroduced. Within 5 weeks of environmental removal, Daniel’s affect began to lift. He resumed social activities, regained interest in school, and reported feeling more like himself. A follow-up EEG performed 6 weeks later showed substantial improvement: normalization of posterior alpha rhythm, resolution of frontal delta activity, and a shift toward dominant alpha-theta balance. His cognitive energy and motivation returned fully by the third month.

​

 

Daniel’s case exemplifies the limitations of symptom-based psychiatric diagnosis when functional brain abnormalities are overlooked. From a standard diagnostic perspective, he met criteria for major depressive disorder. However, his EEG pointed to a diffuse neurophysiological suppression, not a primary mood disorder. Without this insight, he might have remained on ineffective antidepressants, while the true etiology—environmentally mediated cortical dysfunction—was left unaddressed.

​

 

For Dr. Swatzyna and others practicing EEG-informed psychiatry, Daniel’s case illustrates the necessity of viewing psychiatric symptoms through the lens of brain function. The presence of diffuse slowing, low-amplitude alpha, and intermittent delta bursts signaled an external, reversible disturbance rather than a chronic affective illness. By identifying and removing the environmental source, his brain—and his life—recovered. The EEG didn’t just monitor progress; it redefined the clinical problem and revealed the path to resolution.

​

​​​

Clinical Presentation 

Antidepressants are prescribed for major depressive disorder and a range of affective and anxiety disorders, but each pharmacologic class exhibits distinct side effect profiles. Clinical response and tolerability vary widely based on receptor affinities, half-life, patient age, and comorbid conditions. The primary classes include monoamine oxidase inhibitors (MAOIs), tricyclic antidepressants (TCAs), selective serotonin reuptake inhibitors (SSRIs), dual-action antidepressants, and NMDA antagonists. The ✽ symbol identifies adverse effects associated with serious clinical risk.

​

Monoamine Oxidase Inhibitors (MAOIs)

Irreversible MAOIs such as selegiline (Emsam, Eldepryl) and isocarboxazid (Marplan) inhibit monoamine metabolism, leading to elevated levels of serotonin, dopamine, and norepinephrine. These drugs are generally reserved for treatment-resistant depression due to their dietary restrictions and drug interaction risks. Side effects include dizziness, confusion, headache, dyskinesia, hallucinations, and sedation. High-risk complications include ✽ hypertensive crisis following tyramine ingestion, ✽ mania, ✽ seizures, and ✽ suicidality (Stahl, 2017; Gillman, 2011). Transdermal delivery systems like the selegiline patch reduce but do not eliminate dietary concerns.

​

Tricyclic Antidepressants (TCAs)

TCAs such as imipramine (Tofranil) and amitriptyline (Elavil) are highly anticholinergic and antihistaminergic, which explains their sedative and autonomic side effects. Common symptoms include anxiety, blurred vision, dizziness, fatigue, and orthostatic hypotension. Sedation is frequent due to strong H1 receptor blockade. Serious side effects include ✽ psychotic symptoms (rare), ✽ seizures, and ✽ suicidality, particularly during early treatment or dose escalation in vulnerable individuals (Stahl, 2017). TCA overdose is particularly dangerous, leading to cardiac arrhythmias and CNS depression, and is a frequent cause of fatal antidepressant toxicity (Baldessarini, 2013).

​

Selective Serotonin Reuptake Inhibitors (SSRIs)

SSRIs such as fluoxetine (Prozac), sertraline (Zoloft), and paroxetine (Paxil) are generally better tolerated than TCAs but are not without adverse effects. Commonly reported symptoms include agitation, anxiety, headache, insomnia, sedation, and tremor. Sexual dysfunction occurs in up to 60% of users, often underreported (Clayton et al., 2002). Rare but serious reactions include ✽ seizures, ✽ mania, and ✽ suicidality, particularly in younger populations or those with bipolar spectrum traits (Fergusson et al., 2005). Serotonin syndrome may occur in combination with other serotonergic agents, marked by hyperreflexia, clonus, and autonomic instability.

​
Dual-Action and Selective Norepinephrine Reuptake Inhibitors

Dual-action antidepressants—including duloxetine (Cymbalta) and venlafaxine (Effexor)—inhibit both serotonin and norepinephrine reuptake and are widely prescribed for major depressive disorder, generalized anxiety disorder, and various chronic pain syndromes. These agents frequently cause insomnia, sedation, fatigue, hypertension, and sexual dysfunction (Stahl, 2017). Among them, venlafaxine is particularly associated with a dose-dependent increase in blood pressure, including sustained elevations in diastolic values. In certain patients, these medications have also been linked to ✽ hypomania, ✽ mania, and ✽ suicidality, especially when used in individuals with undiagnosed bipolar spectrum vulnerability or a family history of mood instability.

Pharmacologically related but more selective, atomoxetine (Strattera) is a selective norepinephrine reuptake inhibitor (SNRI) prescribed primarily for attention-deficit/hyperactivity disorder, though it has also been evaluated in depressive disorders. Common side effects include abdominal discomfort, anxiety, agitation, aggression, dizziness, dysmenorrhea, and fatigue, with children and adolescents being particularly vulnerable (Michelson et al., 2001). Of particular concern are cardiovascular effects: ✽ elevated heart rate, ✽ sustained hypertension, and ✽ orthostatic hypotension have been consistently reported. Additionally, atomoxetine has been associated with psychiatric risks, including ✽ hypomania, ✽ mania, ✽ suicidality, and—though rare—✽ priapism (Stahl, 2017).

Withdrawal from dual-action antidepressants—especially venlafaxine—is associated with a well-documented discontinuation syndrome. Symptoms include dizziness, nausea, flu-like malaise, and so-called "brain zaps." These clinical signs correspond to EEG abnormalities such as intermittent delta and irregular theta bursts, consistent with transient disruption of thalamocortical regulatory circuits (Thompson & Thompson, 2016).

​

NMDA Antagonists

Glutamate antagonists are a class of agents that exert their primary effects through modulation of the N-methyl-D-aspartate (NMDA) receptor, a subtype of glutamate receptor critically involved in excitatory neurotransmission and synaptic plasticity. Drugs in this category include ketamine, esketamine, memantine, and dextromethorphan, as well as experimental agents such as AV-101. Although originally developed for anesthesia or neurodegenerative conditions, NMDA antagonists are now increasingly used in psychiatry, particularly for treatment-resistant depression, suicidal ideation, and obsessive-compulsive spectrum disorders. Their clinical effects are distinct from those of monoaminergic agents and reflect a complex interplay between rapid neuroplastic modulation and acute neuropsychiatric side effects (Zarate et al., 2006; Krystal et al., 2013). We will examine the side effects of ketamine, memantine, and dextromethorphan.

​

Ketamine, the most studied glutamate antagonist in psychiatry, is known for its rapid antidepressant effect, with symptom reduction often observed within hours. Patients receiving ketamine typically report transient dissociative symptoms, including depersonalization, derealization, and perceptual alterations, particularly during or shortly after infusion (Berman et al., 2000; Luckenbaugh et al., 2014). These effects are dose-dependent and most pronounced with intravenous administration. Cardiovascular stimulation, characterized by acute increases in blood pressure and heart rate, is also common (Short et al., 2018). Though generally self-limited, serious adverse effects such as ✽ emergence delirium, ✽ dissociative psychosis, and
✽ acute suicidality have been reported, especially in patients with comorbid bipolar spectrum illness, borderline personality traits, or latent psychotic vulnerability (Schatzberg, 2014; Niciu et al., 2014).

​

Chronic or high-dose ketamine use, including repeated subanesthetic exposure, has been associated with ✽ cognitive deficits, ✽ urinary tract dysfunction, and ✽ substance misuse, particularly in unsupervised or recreational settings (Morgan & Curran, 2012). Long-term users may present with affective blunting, executive dysfunction, and dissociative states, symptoms that can be mistaken for negative symptoms of schizophrenia or organic encephalopathy (Coyle et al., 2012).

​

Esketamine, the S-enantiomer of ketamine approved for treatment-resistant depression, shares ketamine’s dissociative profile but may have slightly reduced psychotomimetic effects when administered intranasally (Daly et al., 2019). Nevertheless, acute side effects still include sedation, derealization, hypertension, and, in some cases, ✽ hallucinations, ✽ disorientation, or ✽ transient psychosis, especially in patients with neuropsychiatric comorbidities (Popova et al., 2019).

​

Memantine, a low-affinity uncompetitive NMDA antagonist used primarily in Alzheimer’s disease, has been investigated for adjunctive use in major depressive disorder and obsessive-compulsive disorder. At therapeutic doses, memantine typically does not cause overt dissociation. However, in elderly patients or those with cognitive vulnerability, it may lead to dizziness, confusion, or worsening of agitation and psychosis (Reisberg et al., 2003). ✽ Delirium, ✽ hallucinations, and ✽ affective flattening have been documented, especially when combined with other centrally active agents or in cases of impaired renal clearance (Winblad et al., 2007).

 

Dextromethorphan, another NMDA antagonist with serotonergic and sigma-1 receptor activity, is used in combination with bupropion (as in the FDA-approved formulation Auvelity) for major depressive disorder. At standard doses, dextromethorphan is well-tolerated. However, in slow CYP2D6 metabolizers or at supratherapeutic levels, it can produce dissociative effects, tachycardia, and hallucinations (Chen et al., 2018). In recreational use, dextromethorphan has been implicated in ✽ serotonin syndrome, ✽ psychosis, ✽ seizures, and ✽ fatal overdose, particularly when combined with serotonergic or sympathomimetic drugs (Wang et al., 2008).

 

The risk of psychiatric destabilization with glutamate antagonists is heightened in patients with temporal lobe epilepsy, bipolar disorder, substance use disorders, or developmental neuropsychiatric conditions. In such populations, even low doses can unmask latent pathology, producing ✽ mania, ✽ paranoia, or ✽ catatonic features (Swanson et al., 2017). Careful screening is essential prior to initiating treatment, particularly for individuals with known cortical hyperexcitability or a family history of psychosis.

​

Given their rapid onset and neuroplastic mechanisms, glutamate antagonists often produce dramatic subjective improvement—yet this can obscure their potential to destabilize fragile cortical systems, especially when administered outside of structured, monitored environments. Clinicians should be alert to ✽ cardiovascular complications, ✽ acute dissociative states, and
✽ protracted neurocognitive effects. Their use requires close monitoring, ideally with baseline assessments of psychiatric stability, medical status, and, where possible, EEG screening to detect signs of cortical irritability or latent epileptiform activity.

​

In summary, glutamate antagonists provide a novel and powerful tool in the management of refractory psychiatric illness, particularly rapid-acting antidepressant effects in cases of suicidality or catatonic depression. However, their therapeutic promise must be balanced against a complex clinical profile that includes dissociative, cardiovascular, and neurocognitive risks, especially in vulnerable populations. Judicious selection and ongoing clinical vigilance remain essential.

​

​

EEG Effects

​

Monoamine Oxidase Inhibitors (MAOIs)

Monoamine oxidase inhibitors (MAOIs) are among the oldest classes of antidepressants and are pharmacologically distinct in their mechanism and electrophysiological effects. Unlike tricyclics or selective serotonin reuptake inhibitors, MAOIs work by irreversibly inhibiting monoamine oxidase enzymes—primarily MAO-A and MAO-B—thereby preventing the breakdown of key neurotransmitters such as serotonin, norepinephrine, and dopamine. Their effects on the human EEG are complex and influenced by several factors, including the specific agent used, dosage, selectivity of inhibition, and individual neurophysiological characteristics such as baseline arousal or cortical excitability.

​

The EEG profile of MAOIs diverges significantly from that of sedative antidepressants like the tricyclics. Early pharmacological studies comparing MAOIs to TCAs noted that iproniazid, a nonselective, irreversible MAOI, produces less theta enhancement than amitriptyline but greater fast beta activity, typically in the 20–30 Hz range (Saletu et al., 1976; Thompson & Thompson, 2015). This increased fast beta activity—especially over frontocentral regions—is a signature of enhanced cortical arousal and is electrophysiologically similar to the pattern seen with CNS stimulants. Iproniazid does not promote the diffuse cortical slowing or alpha attenuation characteristic of sedating TCAs, making its EEG profile more compatible with alertness-promoting pharmacology.

​

Other nonselective, irreversible MAOIs such as isocarboxazid (Marplan) and phenelzine (Nardil) similarly produce increases in beta activity, particularly in the 20–30 Hz frequency band, while simultaneously suppressing lower frequency bands including delta and theta (Thompson & Thompson, 2015). This EEG pattern reflects a generalized increase in cortical activation, which in some patients may manifest as improved vigilance and reduction in psychomotor retardation, though it may also lead to insomnia, restlessness, or agitation, especially at higher doses or in patients with baseline beta excess. These effects are dose-dependent and appear more prominently after several days to weeks of treatment, reflecting the time course of enzyme inhibition and synaptic monoamine accumulation.

​

Importantly, not all MAOIs are equivalent in EEG impact. Selegiline, a selective MAO-B inhibitor at low doses, exhibits a milder EEG profile when compared to nonselective agents. At low transdermal doses, selegiline has minimal impact on background rhythms, but as dosage increases or enzyme selectivity is lost, it may begin to produce beta enhancement and alpha suppression similar to that observed with phenelzine or isocarboxazid (Gillman, 2011). Individual variability in metabolism, monoaminergic tone, and cortical excitability can further modulate these effects.

​

MAOIs can occasionally provoke paroxysmal discharges in predisposed individuals, particularly those with subclinical cortical irritability. This is a rare but recognized phenomenon and may be linked to dopaminergic overdrive or altered seizure threshold, particularly when used in combination with other activating agents (Blume, 2006). While MAOIs are not conventionally considered proconvulsant, caution is warranted in patients with pre-existing epileptiform patterns on EEG or a history of seizure disorder.

​

In summary, the EEG effects of MAOIs are generally activating, characterized by increased fast beta activity and reduction in slower waveforms, setting them apart from sedative antidepressants. These effects are more pronounced with nonselective, irreversible inhibitors such as iproniazid and isocarboxazid, and are influenced by dosage, agent specificity, and baseline cortical physiology. Given their resemblance to stimulant-induced EEG changes, MAOIs may paradoxically worsen symptoms in patients with beta spindling, anxiety-spectrum disorders, or subclinical epileptiform activity. As such, baseline EEG evaluation can provide valuable information to guide the safe and effective use of MAOIs in modern clinical practice.

​

​

Tricyclic Antidepressants (TCAs)

Tricyclic antidepressants (TCAs) are among the oldest and most widely studied antidepressant classes, known for their potent antihistaminergic, anticholinergic, and noradrenergic properties. Their electrophysiological effects on the EEG are dose-dependent and vary based on their relative sedative potency, with clear distinctions between sedating agents like amitriptyline and imipramine, and nonsedating variants like desipramine.

​

Sedating TCAs such as amitriptyline (Elavil) and imipramine (Tofranil) produce a robust pattern of cortical slowing. These drugs consistently increase delta (0.5–4 Hz), theta (4–8 Hz), and fast beta (20–30 Hz) activity, while concurrently reducing alpha (8–12 Hz) power and total EEG power (Knott, 2000; Malver et al., 2014; Saletu, 2010; Thompson & Thompson, 2016). The elevation in delta and theta is more pronounced at higher doses, often appearing diffusely across the scalp, particularly in frontal and central regions (Bauer & Bauer, 2005; Van Cott & Brenner, 2003). In addition, these agents can slow the peak alpha frequency, a hallmark of sedative and anticholinergic influence on cortical rhythms (Brienza et al., 2019). This constellation of changes reflects the generalized CNS depressant effect of sedating TCAs and is associated with sedation, fatigue, and cognitive dulling in clinical populations.

​

In contrast, nonsedating TCAs, such as desipramine (Norpramin), have a more activating EEG profile. These agents tend to increase alpha and fast beta activity without the marked enhancement of low-frequency (delta/theta) rhythms seen in their sedating counterparts (Knott, 2000). This may correspond to their relatively lower affinity for histamine and muscarinic receptors and greater noradrenergic specificity. Patients treated with desipramine often report less sedation, consistent with this electrophysiological pattern.

​

TCAs, particularly in high doses or in predisposed individuals, can also produce asynchronous slow-wave activity, most often in the form of diffuse theta. This may be associated with anticholinergic-induced encephalopathic states, particularly in elderly patients or those with preexisting cognitive vulnerability (Blume, 2006). Moreover, both TCAs and SSRIs have been reported to provoke epileptiform discharges, including spikes and polyspikes, especially at supratherapeutic levels or in patients with latent cortical excitability. These findings warrant caution in populations with a history of seizures or in those exhibiting isolated epileptiform discharges (IEDs) on baseline EEG.

​

Another well-documented TCA-related EEG feature is the promotion of sleep spindles—bursts of 12–14 Hz sigma activity—reflecting their sedative influence on thalamocortical networks during sleep (Thompson & Thompson, 2015). This effect is congruent with the sleep-augmenting properties of many sedating TCAs, which are often prescribed off-label for insomnia associated with depression.

​

In summary, the EEG effects of TCAs are highly influenced by their sedative potency. Sedating TCAs promote widespread slowing, spindle enhancement, and alpha suppression, while nonsedating TCAs favor beta and alpha augmentation without pronounced low-frequency elevation. Excessive dosing may lead to diffuse encephalopathic patterns or epileptiform abnormalities, highlighting the need for dose titration and consideration of patient-specific neurophysiological vulnerabilities. Baseline EEG evaluation may assist in identifying individuals at risk for adverse neurocognitive effects, supporting more targeted and informed prescribing.

​

​

Selective Serotonin Reuptake Inhibitors (SSRIs)

Selective serotonin reuptake inhibitors (SSRIs) are widely prescribed for depression, anxiety, obsessive-compulsive disorder, and related mood disorders. Despite their shared primary mechanism—blocking serotonin reuptake at the presynaptic cleft—SSRIs display considerable variability in electrophysiological effects, especially across different compounds, dosages, and individual neurophysiological profiles. Human EEG studies reveal distinct spectral changes that reflect each drug’s unique pharmacodynamics and its interaction with the patient’s baseline arousal and cortical excitability.

​

Several commonly prescribed SSRIs—including fluoxetine (Prozac), paroxetine (Paxil), and sertraline (Zoloft)—exhibit modest increases in beta activity in the 18–25 Hz range, particularly over frontocentral scalp regions, while simultaneously producing reductions in anterior alpha power (Thompson & Thompson, 2016). This pattern is generally interpreted as a mild increase in cortical arousal, which may be beneficial in patients with psychomotor slowing or underarousal. However, in individuals predisposed to anxiety or insomnia, this shift may exacerbate restlessness, agitation, or sleep disruption, particularly during the initial titration phase.

​

Citalopram (Celexa), considered one of the least sedating SSRIs, has been shown to produce a more pronounced shift toward cortical activation. EEG studies report decreases in total power, delta, theta, and alpha bands, along with increases in beta and gamma activity, especially in posterior and midline regions (Bauer & Bauer, 2005; Nissen et al., 2020; Saletu, 2010; Van Cott & Brenner, 2003). These findings support its role in energizing affectively blunted or hypoaroused individuals, but the increase in high-frequency activity may contribute to side effects such as insomnia, jaw tension, or increased sympathetic tone, particularly in patients already exhibiting spindling beta patterns.

​

Fluvoxamine (Luvox), often used in obsessive-compulsive disorder, shows a more complex and dose-sensitive EEG profile. In one study, administration of fluvoxamine in combination with magnesium led to an increase in delta, theta, and beta power, along with a decrease in alpha power and a slowed peak alpha frequency—suggesting a mixed sedative-activating profile (Skalski et al., 2021). The extent of these effects may depend on adjunctive supplementation, drug plasma level, and patient-specific EEG traits such as baseline alpha frequency or thalamocortical rhythmicity.

​

Vortioxetine (Trintellix), a newer multimodal SSRI that also modulates 5-HT3, 5-HT1A, and 5-HT1B receptors, produces a distinct EEG profile. Studies show that vortioxetine decreases theta (4–8 Hz) activity and increases beta (12–32 Hz) and gamma (32–45 Hz) power, particularly over frontal and parietal regions (Nissen et al., 2020). These effects may correlate with the drug’s pro-cognitive profile, including enhanced working memory and attention in both depressed and non-depressed populations. However, increases in gamma power must be interpreted cautiously, as scalp EEG at this frequency is vulnerable to contamination from muscle artifacts.

​

Importantly, dose and plasma concentration play critical roles in modulating SSRI EEG effects. At higher doses, SSRIs—especially in susceptible individuals—can provoke paroxysmal activity, including bisynchronous spikes or polyspikes, particularly in those with subclinical epileptiform discharges or reduced seizure threshold (Blume, 2006). This is of particular concern in pediatric, geriatric, or neurologically compromised populations. Moreover, in cases of serotonin syndrome, a potentially life-threatening state of serotonergic overactivation, EEG may show triphasic waves, a classic marker of toxic encephalopathy (Blume, 2006). These findings reinforce the need for careful monitoring in patients exhibiting sudden changes in mental status or autonomic function while on SSRIs.

​

In conclusion, the electrophysiological effects of SSRIs on the EEG are not uniform across the class. While many SSRIs modestly increase frontocentral beta and reduce alpha, individual drugs such as citalopram, fluvoxamine, and vortioxetine produce distinct spectral changes that can have therapeutic or adverse consequences depending on the clinical context and EEG phenotype. EEG biomarkers—such as spindling beta, diffuse slowing, or latent epileptiform discharges—may help clinicians predict which patients are more likely to benefit from serotonergic modulation versus those at increased risk for side effects or treatment failure. The integration of EEG findings into medication decision-making offers a path toward safer and more precise pharmacotherapy.

​

Dual-Action Antidepressants

Dual-action antidepressants, including venlafaxine (Effexor), desvenlafaxine (Pristiq), and duloxetine (Cymbalta), inhibit both serotonin and norepinephrine reuptake, thereby enhancing monoaminergic transmission across widespread cortical and subcortical circuits. Similarly, selective norepinephrine reuptake inhibitors (SNRIs) such as atomoxetine (Strattera) act specifically on norepinephrine transporters but are often discussed within the same neuropharmacological framework due to overlapping cortical effects. These agents exhibit characteristic EEG signatures that reflect their pro-arousal pharmacodynamics—effects that are generally beneficial in hypoaroused or sluggish cortical profiles but potentially destabilizing in patients with hyperarousal, cortical excitability, or underlying epileptiform traits.

​

Electrophysiologically, dual-action antidepressants tend to increase beta activity, particularly in the 13–21 Hz range, with a frontocentral distribution. This elevation in fast activity is consistent with enhanced noradrenergic tone and a shift toward greater cortical activation (Knott, 2000; Saletu, 2010). In individuals with baseline cortical underarousal—as seen in some subtypes of depression or attention-deficit presentations—this increase in beta may correlate with improved mood, cognitive energy, and attentional engagement. However, in patients with preexisting spindling excessive beta (SEB), further enhancement of high-frequency activity may worsen hyperarousal, leading to insomnia, irritability, or autonomic dysregulation. These effects are not merely dose-dependent but also modulated by the patient’s neurophysiological baseline and comorbid traits, particularly anxiety spectrum vulnerability (Swatzyna et al., 2024).

​

EEG profiles observed with venlafaxine and duloxetine reflect this pro-arousal pattern. These agents have been shown to increase beta activity while simultaneously suppressing alpha and, at times, theta power—creating an EEG signature more aligned with stimulant-like activation than with sedative antidepressants (Bauer & Bauer, 2005). Notably, venlafaxine, at higher doses, exhibits a pronounced noradrenergic effect and has been associated with sustained diastolic hypertension, a marker of central adrenergic load that may mirror cortical excitation (Stahl, 2017). Withdrawal from dual-action agents, particularly venlafaxine, can provoke an EEG rebound phenomenon, with increases in low-frequency activity (delta/theta) and loss of fast coherence, corresponding clinically to "brain zaps," dizziness, and sensorimotor disorientation (Thompson & Thompson, 2016).

​

In parallel, atomoxetine, a selective norepinephrine reuptake inhibitor primarily indicated for ADHD, demonstrates similar electrophysiological effects. EEG studies have shown that atomoxetine decreases frontal theta and increases beta power, particularly in individuals with high baseline theta—such as those with ADHD or frontal underarousal (Arns et al., 2008). This pattern likely reflects enhanced cortical vigilance and executive control, aligning with its therapeutic effect in attention disorders. However, the same excitatory influence may carry iatrogenic risk in patients with latent cortical instability. Individuals exhibiting isolated epileptiform discharges (IEDs), focal slowing, or prior head trauma may experience adverse neurobehavioral outcomes, including increased impulsivity, agitation, or subclinical seizure activation when treated with atomoxetine (Swatzyna et al., 2024).

​

While dual-action agents and SNRIs can restore engagement and attentional focus in underactivated cortical systems, their EEG-enhancing properties in the beta band require careful interpretation. In patients with pre-existing EEG biomarkers of hyperexcitability, such as SEB or IEDs, the beta-enhancing, seizure-threshold-lowering effects of these agents may lead to treatment-emergent instability, both affectively and behaviorally. These risks are heightened in youth and in individuals with neurological comorbidity, such as traumatic brain injury or developmental encephalopathy.

​

Consequently, EEG-guided prescribing is especially valuable when considering noradrenergic or dual-action agents. In the presence of spindling beta, diffuse slowing, or epileptiform activity, clinicians may consider alternative pharmacologic strategies, such as mood-stabilizing anticonvulsants (e.g., lamotrigine or valproate), which offer more neurostabilizing effects and avoid further amplifying cortical excitability. EEG profiling can therefore not only optimize medication selection but also prevent treatment-emergent complications in complex or treatment-resistant presentations.

​

NMDA Antagonists

The electroencephalographic (EEG) effects of NMDA receptor antagonists—including ketamine, memantine, and dextromethorphan—reflect the profound influence these agents have on cortical excitability, synaptic plasticity, and thalamocortical dynamics. Unlike serotonergic or noradrenergic antidepressants, NMDA antagonists modulate the glutamatergic system directly, often leading to non-linear and state-dependent changes in the EEG. These effects vary by drug, dose, route of administration, and the neurophysiological baseline of the patient, including factors such as age, cortical stability, and the presence of psychiatric or neurological comorbidities.

​

Ketamine, a noncompetitive NMDA receptor antagonist, is well known for its rapid antidepressant and dissociative effects. EEG studies in healthy individuals and depressed patients reveal that ketamine administration typically results in an increase in gamma activity (30–50 Hz), particularly in frontal and medial cortical regions, often within minutes of administration (Muthukumaraswamy et al., 2015). This gamma elevation is thought to reflect cortical disinhibition, due to ketamine's preferential blockade of NMDA receptors on inhibitory interneurons. The consequence is a release of excitatory activity, leading to transient cortical hyper-synchrony in high-frequency bands.

​

Simultaneously, ketamine also induces reductions in alpha (8–12 Hz) and increases in theta (4–8 Hz) and delta (1–4 Hz) power, depending on the dose and level of consciousness (Shaw et al., 2015). During subanesthetic infusions used in psychiatry, these changes often co-exist with dissociative symptoms and correlate with the subjective experience of depersonalization and derealization. In EEG terms, the combination of elevated gamma, suppressed alpha, and frontal theta bursting is distinctive of the ketamine state and may help to differentiate it from other classes of psychotropics. Higher doses, or use in sedated states, may shift the EEG toward slow-delta dominance, reflecting NMDA blockade across thalamocortical relay systems and the approach to unconsciousness (Sarasso et al., 2015).

​

However, ketamine’s EEG effects are not purely sedative. In treatment-resistant depression, increases in gamma power following ketamine infusion have been associated with antidepressant response, suggesting that gamma enhancement may be a biomarker of treatment efficacy (Cornwell et al., 2012). Yet, this same gamma hypersynchrony may also indicate cortical instability, particularly in patients with baseline excitability, such as those with epileptiform discharges, bipolar disorder, or traumatic brain injury. In such cases, ketamine’s EEG effects can potentiate irritability, agitation, or transient psychosis, reflecting a pathological amplification of excitatory signaling (Blume, 2006; Krystal et al., 2013).

​

Memantine, by contrast, is a low-affinity, voltage-dependent NMDA antagonist used primarily in neurodegenerative conditions. Its EEG profile is more subtle and does not produce the gamma enhancement or profound alpha suppression typical of ketamine. Instead, memantine tends to normalize or preserve alpha rhythms and may modestly increase theta activity, particularly in frontotemporal regions (Günther et al., 2009). In Alzheimer’s patients, memantine has been observed to enhance alpha coherence and reduce diffuse delta activity, possibly reflecting improved thalamocortical regulation (van Straaten et al., 2012). Unlike ketamine, memantine rarely induces dissociation or hyperexcitability on EEG. However, at high doses or in individuals with renal impairment, memantine can result in cognitive slowing, excessive theta, or even encephalopathic patterns, particularly in older adults with cortical atrophy or polypharmacy (Winblad et al., 2007). Although memantine is generally neuroprotective in design, EEG findings suggest that vulnerability to excitatory-inhibitory imbalance still exists in specific populations.

​

Dextromethorphan, which functions as an NMDA receptor antagonist and also has serotonin and sigma-1 receptor activity, exhibits a dose-dependent EEG profile that spans normal cognition to dissociative and psychotomimetic states. At therapeutic doses—particularly in combination with bupropion (as in the FDA-approved formulation Auvelity)—dextromethorphan may produce modest reductions in theta and increases in beta activity, paralleling mild increases in arousal and vigilance. However, at higher doses or in slow CYP2D6 metabolizers, dextromethorphan can induce delta and theta slowing, alpha suppression, and high-amplitude beta or gamma bursts, particularly in frontal and parietal leads (Wang et al., 2008; Gahimer et al., 2022). These changes correlate with hallucinations, euphoria, and dissociative states commonly observed in recreational use.

​

At toxic levels, dextromethorphan may produce triphasic waves, spike-wave discharges, or generalized slowing, all of which resemble toxic-metabolic encephalopathy or serotonin syndrome, particularly when used with MAOIs or SSRIs (Blume, 2006). The EEG profile becomes erratic, often with alternating low-voltage fast activity and paroxysmal slow bursts, a pattern indicative of cortical overstimulation followed by compensatory inhibition.

In clinical use, these EEG features can offer valuable insight into treatment suitability. For example, patients with spindling excessive beta (SEB) or epileptiform discharges may experience symptom exacerbation with NMDA antagonists, even at subpsychedelic doses. In contrast, patients with diffuse alpha slowing, cognitive suppression, or flattened affective EEG patterns may benefit from the cortical desuppression induced by ketamine or dextromethorphan. EEG monitoring may also help clinicians detect early signs of emergent cortical instability, guide dose adjustments, and avoid pharmacological overdrive in sensitive populations.

​

In sum, NMDA receptor antagonists produce distinct, state-dependent EEG signatures characterized by alterations in high-frequency and low-frequency bands. Ketamine is associated with robust increases in gamma and theta activity, coupled with alpha suppression, mirroring its dissociative and antidepressant effects. Memantine has a modest normalizing influence, especially in populations with baseline cortical slowing. Dextromethorphan, depending on dose and metabolism, ranges from mild beta enhancement to full dissociative EEG disruption. While these drugs open new frontiers in psychiatric treatment, their use in clinical practice—particularly in patients with neurological comorbidities—requires nuanced interpretation of EEG biomarkers to guide safe and effective implementation.

​

​​

EEG Effects

 

​Antidepressants produce distinct electrophysiological effects on the human EEG, reflecting their pharmacological mechanisms and clinical profiles. These effects are modulated by sedation, dosage, patient age, and baseline EEG characteristics. The following synthesis organizes current findings by antidepressant class and reconciles interpretive differences in the literature.

 

Monoamine Oxidase Inhibitors (MAOIs)

Monoamine oxidase inhibitors (MAOIs), particularly irreversible nonselective agents like iproniazid and isocarboxazid, induce an EEG profile characterized by increased fast beta activity in the 20–30 Hz range and decreased power in slower frequencies. These effects are consistent with increased central catecholaminergic transmission and resemble those observed with mild CNS stimulants. Saletu et al. (1986) directly compared iproniazid and amitriptyline, showing that iproniazid produced greater enhancement in fast beta and less theta elevation, indicating an activating rather than sedative effect.

​

These EEG changes are not associated with sedation and appear relatively dose-independent within therapeutic ranges. However, due to the activating nature of MAOIs, older adults may exhibit increased sensitivity to cortical overactivation, potentially manifesting as impaired sleep or executive dysfunction, though no EEG studies have explicitly examined this by age subgroup.

​

Tricyclic Antidepressants (TCAs)

TCAs demonstrate the most pronounced sedative EEG effects among antidepressants, particularly with agents like amitriptyline and imipramine. These drugs reliably increase delta and theta power, reduce alpha, and in some cases increase fast beta (20–30 Hz), particularly in frontal and central regions (Saletu et al., 1991; Brienza et al., 2019). This combination of slowing and fast beta enhancement likely reflects both sedation and anticholinergic disinhibition.

​

Nonsedating TCAs, such as desipramine, produce a distinct EEG profile, with increases in alpha and fast beta and little change in slow-wave power (Knott, 2000). This suggests that EEG slowing is tied more to sedative burden than to the class as a whole.

​

At toxic levels, TCAs are associated with asynchronous slow waves, diffuse theta, and epileptiform discharges, including polyspikes (Blume, 2006). These findings emphasize the dose-dependent nature of EEG changes and the risk of cortical excitability in overdose situations.

​

As with SSRIs, older adults may exhibit more prominent EEG slowing in response to sedating TCAs, which may complicate the interpretation of EEGs in this population. Clinicians must distinguish between drug-induced slowing and early encephalopathic changes or degenerative pathology.

​

Selective Serotonin Reuptake Inhibitors (SSRIs)

Selective serotonin reuptake inhibitors (SSRIs) present a more heterogeneous EEG profile, dependent on drug-specific properties such as sedation and receptor binding profiles. Most SSRIs, including fluoxetine, paroxetine, and sertraline, produce a modest increase in frontocentral beta activity (18–25 Hz) and a decrease in anterior alpha power (Davidson et al., 2000). These effects likely reflect enhanced serotonergic tone and altered prefrontal cortical regulation.

​

More sedating SSRIs (e.g., paroxetine) may induce subtle increases in delta and theta activity, particularly in resting-state EEG, whereas activating SSRIs (e.g., fluoxetine and citalopram) show decreased power in delta, theta, and alpha bands and increased beta, especially in frontal and parietal regions (Bauer & Bauer, 2005; Nissen et al., 2020).

​

Vortioxetine, a multimodal SSRI that also modulates 5-HT1A, 5-HT1B, and 5-HT3 receptors, shows a unique EEG profile: increased beta and gamma power and decreased theta. These changes are interpreted as neurophysiological markers of improved working memory and executive control (Nissen et al., 2020). However, findings involving gamma activity require cautious interpretation due to potential contamination by muscle artifact and relatively low signal-to-noise ratios in scalp EEG.

​

At high doses or in serotonin toxicity, SSRIs have been associated with triphasic waves, suggestive of toxic metabolic encephalopathy (Blume, 2006). This phenomenon highlights the need for clinical vigilance in overdose situations.

​

Age is also a significant moderating variable. EEG slowing—specifically increased delta and theta with reduced alpha—may be more pronounced in older adults, potentially mimicking or exacerbating age-related cognitive slowing (Davidson et al., 2000). Thus, age-specific EEG norms are essential when interpreting SSRI effects in geriatric populations.

​

Dual-Action Antidepressants

Dual-action antidepressants, also referred to as serotonin-norepinephrine reuptake inhibitors (SNRIs), exert their effects by inhibiting the reuptake of both serotonin and norepinephrine, thereby enhancing synaptic concentrations of these monoamines. Common representatives include venlafaxine, desvenlafaxine, and duloxetine. These agents are frequently prescribed for major depressive disorder, generalized anxiety disorder, fibromyalgia, and neuropathic pain syndromes.

 

Their electrophysiological impact on the EEG reflects their mixed pharmacological profile and varies with dose, degree of sedation, and the baseline arousal state of the patient.

At low to moderate doses, dual-action antidepressants often produce mild increases in beta activity—particularly in the low to mid-beta range (13–21 Hz)—predominantly in frontocentral regions, which likely corresponds to enhanced cortical vigilance. This increase is more pronounced in individuals with lower baseline beta or increased frontal theta activity, such as those with attentional or mood dysregulation (Knott, 2000; Arns et al., 2008). Duloxetine, in particular, has been shown to reduce theta and alpha power while increasing fast beta activity, consistent with an activating electrophysiological profile, especially during wake EEG recordings (Brienza et al., 2019).

​

At higher doses, however, the noradrenergic component of these medications becomes more dominant, which may lead to elevated high-frequency beta, reduced alpha, and in some cases, emergence of excessive cortical arousal, especially in patients with spindling excessive beta (SEB). These patients may report increased anxiety, insomnia, and cognitive overstimulation—effects that may be reflected in beta spindles and hyper-synchronous beta rhythms on EEG (Swatzyna et al., 2024).

​

In certain vulnerable individuals, particularly those with underlying isolated epileptiform discharges (IEDs) or focal slowing, dual-action antidepressants can exacerbate subclinical epileptogenicity or provoke adverse behavioral responses. This is likely due to the pro-convulsant effect associated with increased monoaminergic tone, particularly when the seizure threshold is already reduced. As with SSRIs, these agents may lower seizure threshold and precipitate abnormal EEG activity when administered without biomarker-informed screening (Swatzyna et al., 2024; Blume, 2006).

​

Notably, venlafaxine is associated with dose-dependent blood pressure elevation, which may contribute to sympathetic nervous system hyperactivation and further amplify EEG beta activity, particularly in the upper 20–30 Hz range. Conversely, abrupt discontinuation of dual-action agents—especially venlafaxine—can produce electrophysiological withdrawal patterns, such as increased slow-wave activity, frontal intermittent delta, and irregular theta bursts, which may correlate with dizziness, disorientation, or "brain zaps" reported clinically (Thompson & Thompson, 2016).

​

In summary, dual-action antidepressants typically enhance beta activity and suppress theta/alpha rhythms, producing an activating EEG profile that may be therapeutic in under-aroused patients but destabilizing in those with hyperarousal or epileptiform biomarkers. Their effect is clearly dose-dependent, and caution is warranted in individuals with predisposing EEG abnormalities. Incorporating baseline EEG screening can improve treatment matching, reduce adverse outcomes, and support the implementation of biologically informed psychiatric care.

​​​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​​

​

​

​

​

​

​

​

​

​

​

​

​

​

 

 

 

 

 

 

 

 

​​

An EEG-Informed Psychiatry Perspective

From Dr. Ronald Swatzyna's EEG-informed psychiatry perspective, the widespread variability in antidepressant efficacy—and the often unpredictable emergence of adverse effects—can be better understood through systematic evaluation of neurophysiological biomarkers. His research consistently demonstrates that patients with treatment-resistant symptoms frequently present with one or more EEG abnormalities—such as spindling excessive beta (SEB), focal slowing (FS), diffuse encephalopathy (EN), or isolated epileptiform discharges (IEDs)—which have significant implications for medication selection and treatment outcomes (Swatzyna et al., 2024).

​

Accepting the clinical importance of these biomarkers necessitates a shift from symptom-based prescribing to biomarker-informed decision-making. If an EEG reveals SEB, for example, prescribing SSRIs—known to increase beta activity—may worsen hyperarousal, anxiety, or agitation. Similarly, prescribing activating antidepressants like atomoxetine or venlafaxine to a patient with IEDs or cortical instability increases the risk of inducing neuropsychiatric destabilization, including mania, psychosis, or even seizure activity. In patients with diffuse slowing or encephalopathy, any psychotropic drug may be ineffective or harmful until the underlying medical or metabolic cause is identified and addressed.

​

This approach implies that prescription practices should not proceed in complex or refractory cases without first ruling out functional neurobiological contributors. When patients present with overlapping symptoms of depression, anxiety, cognitive dysfunction, or behavioral dysregulation, it becomes essential to ask: Is this truly a chemical imbalance, or a sign of impaired cortical regulation, injury, or latent excitability? In such cases, EEG offers a cost-effective, noninvasive tool for differentiating primary psychiatric disorders from functional neurological disturbances.

​

Antidepressants, particularly those that are highly activating or serotonergically potent, may be the worst choice in patients with SEB, IEDs, or focal abnormalities. Examples include SSRIs in patients with beta spindling, SNRIs in those with epileptiform discharges, or TCAs in the context of encephalopathy, where their anticholinergic burden can exacerbate cognitive dysfunction. These mismatches not only lead to treatment failure, but can also worsen the underlying condition, mislead the diagnostic process, or result in harmful polypharmacy.

EEG biomarkers, then, serve not only as predictors of medication response but as red flags alerting the clinician to circumstances where functional neuroanatomy—not neurotransmitter deficiency—is driving the symptomatology. In such cases, the treatment priority should shift from pharmacologic modulation to addressing root causes: medical, metabolic, structural, or developmental. Only after this foundation is addressed should one consider antidepressant therapy or neuromodulation techniques such as neurofeedback.

​

Ultimately, incorporating EEG biomarkers into psychiatric practice supports a paradigm of precision psychiatry—one that aligns the intervention with the patient’s neurophysiological profile. This approach does not replace clinical judgment but strengthens it, offering a more objective, individualized framework for improving outcomes and avoiding harm. It challenges us to stop treating the brain as a black box and to begin, in earnest, testing the organ we intend to heal.​​​​

​​​​​

​