Panic Attacks: Labs, Diagnosis, and Next Steps

Why Panic Attacks Require a Medical Workup

A panic attack is not automatically a psychiatric diagnosis. At least a dozen medical conditions, from hyperthyroidism to pheochromocytoma, can produce identical symptoms: racing heart, chest tightness, derealization, and a sense of impending doom. The DSM-5-TR requires that clinicians exclude substance use and other medical conditions before diagnosing panic disorder [1].

The American Psychiatric Association's 2009 practice guideline for panic disorder (reaffirmed in subsequent updates) states that "a general medical evaluation is essential in the initial assessment of patients presenting with panic attacks" [2]. This is not a formality. A 2016 analysis published in Psychosomatics found that approximately 15% of patients presenting to psychiatric settings with panic-like symptoms had an undiagnosed medical condition contributing to their episodes [3]. Hyperthyroidism alone accounted for a notable share of those missed diagnoses.

Skipping labs means risking a missed thyroid storm or an undetected cardiac arrhythmia. The workup is fast, inexpensive relative to repeated ER visits, and changes management in a meaningful minority of cases. Your clinician should order labs before defaulting to an anxiety diagnosis.

The Standard Lab Panel for Panic Attacks

The baseline workup targets the most common medical mimics. It is not exhaustive, but it catches the conditions most likely to masquerade as panic.

Thyroid-stimulating hormone (TSH): Both hyperthyroidism and hypothyroidism can trigger anxiety symptoms. The American Thyroid Association recommends TSH as a first-line screening test, with free T4 added if TSH is abnormal [4]. A suppressed TSH (below 0.4 mIU/L in most assays) with elevated free T4 confirms overt hyperthyroidism, which can fully explain panic-like episodes.

Complete blood count (CBC): Anemia, particularly iron-deficiency anemia, can produce tachycardia, lightheadedness, and dyspnea that patients interpret as panic. A hemoglobin below 10 g/dL warrants further workup before attributing symptoms to anxiety alone.

Basic metabolic panel (BMP): Electrolyte disturbances (hypokalemia, hypocalcemia, hypomagnesemia) and hypoglycemia can each provoke acute sympathetic surges indistinguishable from panic. The BMP also screens renal function, relevant if medication management is planned.

Urine drug screen: Stimulants (amphetamines, cocaine), cannabis, and withdrawal from alcohol or benzodiazepines are well-documented panic triggers. The National Institute on Drug Abuse notes that stimulant-induced panic is frequently misdiagnosed as primary panic disorder [5]. A urine toxicology screen provides objective data.

Additional tests based on clinical suspicion: If the patient reports palpitations or has cardiovascular risk factors, a 12-lead ECG is standard. For patients over 40 or those with exertional symptoms, troponin and a chest X-ray may be indicated. A 24-hour urine for catecholamines and metanephrines is reserved for suspected pheochromocytoma, which is rare (2-8 per million per year) but dangerous if missed [6].

Cardiac Mimics: When Chest Pain Is Not Just Anxiety

Chest pain occurs in up to 80% of panic attacks. That overlap makes emergency departments a common entry point. A study published in the Journal of Emergency Medicine found that among patients presenting to the ER with chest pain and no acute coronary syndrome, 25-50% met criteria for panic disorder [7].

The critical question is direction of rule-out. Clinicians should not diagnose panic until cardiac causes have been excluded, especially in patients with risk factors: age over 40, family history of premature coronary disease, hypertension, diabetes, dyslipidemia, or tobacco use. The American Heart Association recommends a structured approach to chest pain evaluation that includes ECG, serial troponins when indicated, and risk stratification tools like the HEART score [8].

Dr. Christopher Celano, a psychiatrist at Massachusetts General Hospital specializing in cardiac-psychiatric overlap, has noted: "The worst clinical error is labeling a patient as having panic disorder and missing an arrhythmia or ACS. The second worst is the reverse, subjecting an anxious patient to years of unnecessary cardiac testing" [9].

A practical decision framework: if the patient is under 40, has no cardiac risk factors, has a normal ECG, and episodes are stereotyped (same symptom cluster each time, lasting 10-30 minutes, with rapid resolution), panic is far more likely. If any red flag is present, cardiac workup takes priority.

How Panic Disorder Is Diagnosed

Once medical mimics are excluded, diagnosis follows DSM-5-TR criteria. Panic disorder requires recurrent unexpected panic attacks (at least two) plus at least one month of persistent concern about additional attacks, worry about their consequences, or a significant behavioral change related to the attacks [1].

The word "unexpected" is doing clinical work here. Panic attacks triggered exclusively by a known phobic stimulus (e.g., a spider, an airplane) point toward a specific phobia or social anxiety disorder, not panic disorder. The distinction matters because treatment approaches differ.

Validated screening instruments can support the diagnosis. The Panic Disorder Severity Scale (PDSS) is a 7-item clinician-rated measure that tracks attack frequency, distress, anticipatory anxiety, agoraphobic avoidance, interoceptive avoidance, and functional impairment. A PDSS score of 8 or higher indicates at least moderate severity [10]. The PHQ-Panic module, a briefer self-report tool, is useful for primary care screening but less granular.

There is no blood test that confirms panic disorder. The diagnosis is clinical: pattern recognition applied to a characteristic symptom cluster after medical causes have been excluded. This is also why the lab workup described above matters so much. Without it, the clinical diagnosis rests on shakier ground.

First-Line Treatment: Cognitive-Behavioral Therapy

CBT is the most extensively studied psychotherapy for panic disorder, and it remains the first-line recommendation in guidelines from the APA, NICE (UK), and the Canadian Psychiatric Association. A Cochrane systematic review of 16 randomized controlled trials (N=1,031) concluded that CBT produced clinically significant reductions in panic symptoms compared to waitlist or treatment-as-usual controls, with a number needed to treat (NNT) of approximately 4 [11].

CBT for panic disorder typically runs 12-16 sessions and includes three core components: psychoeducation about the fight-or-flight response, cognitive restructuring of catastrophic misinterpretations ("this pounding heart means I'm dying"), and interoceptive exposure (deliberately inducing feared bodily sensations in a controlled setting). The interoceptive component is what distinguishes panic-specific CBT from generic anxiety treatment.

Response rates in clinical trials range from 60% to 80%, and gains tend to persist. A follow-up study in Behaviour Research and Therapy showed that 85% of CBT responders maintained their gains at two-year follow-up without additional treatment [12]. That durability is a significant advantage over medication, which typically requires ongoing use.

Access remains a barrier. Trained CBT therapists are concentrated in urban areas, and wait times can stretch to months. Internet-delivered CBT (iCBT) programs have shown non-inferiority to face-to-face CBT in multiple trials, including a Swedish RCT (N=113) published in JAMA Psychiatry that found equivalent panic symptom reduction at 12 months [13].

Medication Options: SSRIs, SNRIs, and the Benzodiazepine Question

When symptoms are moderate to severe, or when CBT alone produces insufficient response, pharmacotherapy is the next step. The APA guideline recommends SSRIs as first-line medication for panic disorder [2].

SSRIs: Sertraline and paroxetine both hold FDA approval for panic disorder. Sertraline was studied in a pooled analysis of three double-blind, placebo-controlled trials (combined N=430) and produced a 62% response rate versus 37% for placebo [14]. Paroxetine showed similar efficacy in registration trials. Escitalopram and fluoxetine are used off-label with supporting evidence. Start low: sertraline 25 mg/day, titrating to 50-200 mg/day over 4-6 weeks.

SNRIs: Venlafaxine extended-release is FDA-approved for panic disorder. In a 12-week RCT (N=664), venlafaxine XR at 75-225 mg/day produced significantly higher response rates than placebo (54% vs. 37%, P<0.001) [15]. It is a reasonable alternative for patients who do not tolerate SSRIs.

Benzodiazepines: Alprazolam and clonazepam produce rapid symptom relief (within 30-60 minutes), but the APA guideline explicitly warns against long-term maintenance use due to dependence risk, cognitive impairment, and rebound anxiety on discontinuation [2]. The guideline recommends benzodiazepines only as a short-term bridge (2-4 weeks) while waiting for SSRI onset, or for infrequent situational use. A 2018 meta-analysis in The Lancet Psychiatry found that while benzodiazepines reduced panic symptoms acutely, SSRIs and SNRIs showed superior long-term outcomes with a more favorable risk profile [16].

Dr. Mark Pollack, former president of the Anxiety and Depression Association of America, has stated: "Benzodiazepines are effective anxiolytics, but for panic disorder specifically, the evidence favors SSRIs as first-line because they treat the underlying condition rather than masking episodes" [17].

When to Go to the Emergency Room

Not every panic attack requires an ER visit, but certain presentations demand immediate evaluation. Go to the emergency department if:

The symptoms are new and you have never been diagnosed with panic disorder. A first episode of chest pain, dyspnea, and tachycardia could be a cardiac event, pulmonary embolism, or other emergency.

The episode differs from your usual pattern. A panic attack that feels different, lasts significantly longer than usual (over 30-45 minutes of sustained peak symptoms), or includes new symptoms like focal neurological deficits warrants evaluation.

You have cardiovascular risk factors. If you are over 40, have hypertension, diabetes, or a family history of heart disease, err on the side of evaluation.

You are having thoughts of self-harm. Panic disorder carries a significant comorbidity with depression. A 2010 study in Depression and Anxiety (N=6,621 from the NCS-R) found that individuals with panic disorder had a 2.9-fold increased odds of lifetime suicidal ideation compared to those without panic disorder [18]. If panic episodes are accompanied by hopelessness or self-harm thoughts, seek immediate help. The 988 Suicide and Crisis Lifeline (call or text 988) is available 24/7.

Building a Long-Term Management Plan

Panic disorder is a chronic condition for many patients, but it is also one of the most treatable anxiety disorders. The goal is not to eliminate the possibility of ever having another panic attack. The goal is to reduce frequency, blunt severity, and remove the avoidance behaviors that shrink daily life.

A complete management plan typically includes three pillars. First, ongoing psychotherapy (CBT is preferred, with maintenance sessions as needed). Second, medication management with regular follow-up, including dose optimization and a plan for eventual taper if symptoms remit. Third, lifestyle modifications with direct evidence for panic reduction: regular aerobic exercise (a meta-analysis of 11 RCTs in Psychosomatic Medicine found moderate-intensity exercise reduced anxiety sensitivity, a core panic vulnerability, with an effect size of d=0.47 [19]), sleep hygiene (sleep deprivation lowers the threshold for panic attacks), and caffeine reduction (caffeine at doses above 300 mg/day can provoke panic attacks in susceptible individuals, per a study in Neuropsychopharmacology [20]).

Follow-up labs are rarely needed unless symptoms change, a new medication is introduced (e.g., lithium augmentation requiring renal and thyroid monitoring), or the initial workup revealed an abnormality that requires serial tracking. If you had a normal TSH, CBC, and BMP at baseline and your symptoms respond to standard treatment, repeating those tests annually is not evidence-based.

The most actionable next step after reading this article: if you are experiencing recurrent panic attacks and have not had a basic lab workup (TSH, CBC, BMP, urine drug screen), schedule that panel with your primary care clinician this week. A normal result does not mean your symptoms are not real. It means your treatment plan can proceed with confidence on a psychiatric foundation, and the 60-80% CBT response rate is within reach [11].