Alcohol and Illicit Drug Use as Precipitants of Atrial Fibrillation in Young Adults: A Case Series and Literature Review

Methods 

City Hospital is a city center teaching hospital in Birmingham, United Kingdom, with a catchment population of 350,000. We obtained a list of all patients aged 45 years and younger admitted to City Hospital, with a discharge diagnosis of atrial fibrillation or atrial flutter, International Classification of Diseases, 10th Revision27 Code 148, over a 6-year period from June 2000 to June 2006, using the computerized Hospital Activity Analysis register. This coding system has been proved free of errors for definitive diagnoses.28 For inclusion in this case series, patients had to be ≤45 years of age on admission, have at least one electrocardiogram (ECG) for that admission demonstrating atrial fibrillation or atrial flutter, and have no cardiac or other co-morbidity that might account for the onset of atrial fibrillation (ie, “lone” atrial fibrillation patients).

The discharge letter for each patient was interrogated using the Clinical Data Archive to ascertain the precipitant for that episode of atrial fibrillation or atrial flutter. Where illicit drug use or alcohol was identified as a potential precipitant, the hospital notes of each patient were inspected further. Demographic and clinical details, including clinical investigations and the management of that episode of atrial fibrillation, were collected from the patient's hospital records. Individual patient management decisions were left to the discretion of the admitting physician.

Results 

From June 2000 to June 2006, 88 patients aged ≤45 years were admitted to City Hospital with a diagnosis of atrial fibrillation or atrial flutter, confirmed by ECG. Twenty-two (25%) patients were identified where alcohol (n = 19) or illicit drug use (n = 3) was a clear precipitant for that episode of atrial fibrillation requiring hospital admission.

A further 5 patients were identified as having an alcohol-induced episode of atrial fibrillation, but they also had other factors that may have precipitated their atrial fibrillation (ischemic heart disease [n = 2], hypertension [n = 1], thyrotoxicosis [n = 1], operated atrial septal defect [n = 1]). The probable causes for the atrial fibrillation among the remaining 61 patients were structural heart disease (n = 14), congenital heart disease (n = 12), thyrotoxicosis (n = 6), other cardiac arrhythmia (n = 5), hypertension (n = 4), postoperative atrial fibrillation (n = 3), ischemic heart disease (n = 2), exercise (n = 2), caffeine excess (n = 2), nicotine (n = 1), and fever (n = 1). Nine (10.2%) patients had no clearly identifiable precipitant to the onset of atrial fibrillation.

Table 3 (online only) presents the demographic and clinical characteristics of the 22 patients with alcohol- or illicit-drug-induced atrial fibrillation. The mean (SD) age was 33.6 (8.4) years (range 20-45 years) and the vast majority were male (90.9%). Four (18.2%) patients had a previous medical history of significance that included 2 with epilepsy, 1 with obstructive sleep apnea, and 1 with depression; all of whom were taking medications for these conditions before admission. The majority of the patients (n = 18) were current cigarette smokers. Almost all presented with palpitations, 4 with shortness of breath or dizziness, 1 with leg and ankle swelling, and 1 with a concomitant stroke. Twenty (90.9%) of the patients admitted to consumption of alcohol, 3 to smoking cannabis, and 1 to snorting cocaine, before the onset of symptoms.

Table 3. Characteristics of Patients Admitted to Hospital with Atrial Fibrillation Precipitated by Alcohol or Illicit Drugs

	Baseline Demographics and Clinical Information	Precipitant and Clinical Presentation	Investigations and Discharge Medication
			(i) Presenting electrocardiogram
			(ii) Blood tests
		(i) Precipitant	(iii) Chest radiograph
		(ii) Symptoms	(iv) Echocardiogram
		(iii) Heart rate (beats per minute)	(v) 24-hour electrocardiogram
		(iv) Blood pressure (mm Hg)	(vi) Discharge medication
	Male, 20 years, no previous medical history; no medication	Marjuana	AF
		Palpitations	Normal
		170	Normal
		120/70	Normal
			SR
			None
	Male, 33 years, no previous medical history; no medication	Alcohol	AF
		Palpitations	Normal
		130	Normal
		130/70	Normal
			DNA
			Flecainide & aspirin
	Male, 42 years, OSA, Bumetanide	Cocaine & alcohol	AFl 2:1
		Leg & neck swelling	Normal
		150	Normal
		134/70	Mild LVH
			DNA
			Atenolol & aspirin
	Male, 21 years, no previous medical history; no medication	Marjuana	AF
		Palpitations	Normal
		110	Normal
		151/78	Normal
			SR
			Bisoprolol & aspirin
	Female, 34 years, no previous medical history; no medication	Alcohol	AF
		Palpitations	Normal
		130	Normal
		101/61	Normal
			SR
			Flecainide
	Male, 42 years, epilepsy, Sodium valproate, B12 and folate	Alcohol	AF
		Palpitations & dizziness	Normal
		140	Normal
		131/80	Mildly dilated LA
			SR
			Verapamil & warfarin
	Male, 43 years, no previous medical history, no medication	Alcohol	AF
		Palpitations & left-sided motor weakness	Normal
		180	Cardiomegaly
		152/98	Dilated LV, moderate LVSD, EF = 35%-40%
			AF
			Warfarin
	Male, 42 years, no previous medical history, no medication	Alcohol	AF
		Palpitations	Normal
		140	Normal
		172/97	DNA
			DNA
			None
	Male, 36 years, epilepsy, Sodium valproate	Marjuana	Atrial flutter with variable block
		Palpitations	Normal
		140	Normal
		114/55	Normal
			DNA
			Bisoprolol & aspirin
	Male, 36 years, no previous medical history, no medication	Alcohol	AF
		Palpitations	Normal
		148	Normal
		155/85	Normal
			SR
			Flecainide
	Male, 42 years, no previous medical history, no medication	Alcohol	AF
		Palpitations	Normal
		100	Normal
		118/72	Normal
			DNA
			Bisoprolol
	Male, 45 years, no previous medical history, no medication	Alcohol	AF
		Palpitations	Normal
		120	Normal
		127/75	DNA
			DNA
			Bisoprolol
	Male, 41 years, no previous medical history, no medication	Alcohol	AF
		Palpitations & SOB	Normal
		140	Cardiomegaly
		110/70	Dilated LV, mild MR
			SR
			Atenolol & warfarin
	Male, 41 years, no previous medical history, no medication	Alcohol	AF
		Palpitations	Normal
		160	Normal
		140/70	Mild MR
			SR
			Verapamil
	Male, 23 years, no previous medical history, no medication	Alcohol	AF
		Palpitations	Normal
		120	Normal
		110/70	Normal
			SR
			Bisoprolol
	Male, 21 years, no previous medical history, no medication	Alcohol	AF
		Palpitations, SOB	Normal
		120	Normal
		129/70	Normal
			SR
			None
	Male, 24 years, no previous medical history, no medication	Alcohol	AF
		Palpitations, SOB	Normal
		70	Not done
		97/60	Normal
			Not done
			Warfarin 4/52 post-DC cardioversion
	Male, 36 years, no previous medical history, no medication	Alcohol	AF
		Palpitations, chest pain	Normal
		170	Normal
		130/80	Mild LV impairment, LVEF 50-60%
			Not done
			None
	Male, 33 years, depression, Fluoxetine	Alcohol	AF
		Palpitations	Abnormal
		166	Normal
		116/66	DNA
			DNA
			None
	Male, 23 years, no previous medical history, no medication	Alcohol	AF
		Palpitations	Normal
		150	Not done
		135/59	Not done
			Not done
			Flecainide & aspirin
	Male, 29 years, no previous medical history, no medication	Alcohol	AF
		Palpitations	Normal
		170	Normal
		130/60	Normal
			SR
			None
	Female, 33 years, no previous medical history, no medication	Alcohol	AF
		Palpitations	Normal
		140	Normal
		174/75	Normal
			SR
			Bisoprolol & aspirin

AF = atrial fibrillation; AFL = atrial flutter; ASA = aspirin; DNA = did not attend; ECG = electrocardiogram; EF = ejection fraction; LA = left atrium; LV = left ventricle; LVH = left ventricular hypertrophy; LVSD = left ventricular systolic dysfunction; MR = mitral regurgitation; OSA = obstructive sleep apnea; SR = sinus rhythm.

The mean heart rate and blood pressure on admission were 139 beats per minute (range 70-180 beats per minute) and 131/72 mm Hg (range 97-174 mm Hg and 55-98 mm Hg for systolic and diastolic blood pressure, respectively) (Table 3, online). Clinical examination was unremarkable in nearly all the patients. Twelve-lead ECG confirmed atrial fibrillation in 20 (90.9%) patients and atrial flutter in the other 2. All had normal baseline full blood count, thyroid function tests, and urea and electrolyte results. Chest radiograph was normal in 20 patients and demonstrated cardiomegaly in 2. Transthoracic echocardiography was performed in 18 (81.8%) patients and was normal in 12 (54.5%). Two patients had a dilated left ventricle, 2 had mild mitral regurgitation, 1 had a dilated left atrium, and 1 had mild left ventricular hypertrophy. One patient required electrical cardioversion, while the others were cardioverted back to sinus rhythm either pharmacologically or spontaneously (n = 3) (Table 3, available online). All the anti-arrhythmic drugs used were well tolerated in these patients, with variable success rates: flecainide (n = 6, 100%), verapamil (n = 3, 100%), beta-blockers (n = 6, 86%) and amiodarone (n = 3, 33%). At discharge, 14 (63.6%) patients were treated with anti-arrhythmic drugs. Anticoagulant or antiplatelet therapy was commenced in 4 (18.2%) and 6 (27.3%) patients, respectively. The patient who presented with stroke was treated appropriately with digoxin.

All patients were offered follow-up appointments, although 3 did not attend any follow-up. Almost all the patients were followed-up for at least 12 months: 16 (72.3%) did not experience any further paroxysms of atrial fibrillation. Of the 6 patients who had further paroxysms, 5 had continued to abuse alcohol and 1 used cannabis.

Discussion 

In this case series, the predominant precipitant for atrial fibrillation was alcohol (86.4%), followed by marijuana (13.6%) and cocaine (4.5%). The most common symptom at presentation was palpitations, occurring in virtually every patient. All the patients were offered appropriate investigations including routine bloods, thyroid function tests, transthoracic echocardiogram, and 24-hour Holter monitoring according to current guidelines.2, 3 All but 4 of the patients were pharmacologically cardioverted to sinus rhythm, with one electrically cardioverted and the remaining 3 spontaneously reverting. Anti-arrhythmic drugs for prophylaxis were offered to the majority (63.6%) of patients, although thromboprophylaxis with either aspirin or warfarin was relatively rare. Only 6 patients experienced further paroxysms of atrial fibrillation, and all continued to abuse either alcohol or illicit drugs. In the previous case reports demonstrating an association between marijuana use and atrial fibrillation,23, 24, 25, 26 total abstinence from marijuana prevented further paroxysms, suggesting that the cardiovascular effects of marijuana are probably reversible and that the atrial fibrillation burden may relate to continuous abuse. The patient who experienced a concomitant stroke was treated appropriately postdischarge with both anti-arrhythmic and thromboprophylatic drugs according to current guidelines.2, 3

Several reports linking alcohol4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21 and drug abuse22, 23, 24, 25, 26, 29, 30, 31, 32, 33, 34, 35 to the onset of atrial fibrillation have been published over the last 30 years, but there is currently no clear consensus of opinion or treatment guidelines for such patients. This lack of guidelines raises the following questions. Do you treat an episode of atrial fibrillation precipitated by alcohol or illicit drug use in the same manner as atrial fibrillation caused by another precipitant? What about the use of anti-arrhythmic therapy for prophylaxis? What happens with thromboprophylaxis in these patients? What are the long-term effects of these substances on rhythm problems? How should we investigate such patients? How do we make decisions about the type and length of follow-up?

Based on guidelines for atrial fibrillation management,2, 3 patients with life-threatening hemodynamic instability should be offered emergency electrical cardioversion, irrespective of the precipitant. In non-life-threatening situations, pharmacological cardioversion may be achieved with anti-arrhythmic drugs. Beta-blockers may be the ideal drug of choice, as in most of the cases initiation of the arrhythmia is due to hyperadrenergic state. The use of beta-blockers, in particular with opiates and amphetamines overdoses, can be hazardous, inducing further coronary ischemia, especially with its unopposed alpha agonist activity worsening further vasoconstriction and thereby the ischemia and blood pressure.34, 35 Recent guidelines22 suggest avoiding the use of beta-blockers in those situations, but there is no such guideline in atrial fibrillation patients. With limited data in those patients developing atrial fibrillation with illicit drugs, it is advisable to avoid beta-blockers, if the culprit (cocaine or amphetamines) is clearly known.

Calcium channel blockers and amiodarone can be used safely to achieve pharmacological success for rhythm control and are preferable in patients with coronary ischemia and left ventricular dysfunction. A Class Ic agent, like flecainide, also is well tolerated and is the preferred option in patients with no coronary ischemia, structural heart disease, or left ventricular dysfunction. In our small case series, patients who had further recurrences are the ones who continued to abuse either alcohol or illicit substances. Chronic anti-arrhythmic prophylaxis may be needed in patients with recurrent symptoms, and we suggest that counseling the patient on the need for complete abstinence from alcohol or the offending drug is mandatory.

In our case series, one patient experienced a stroke and more than one quarter suffered further paroxysms. The majority of the latter patients tend to continue abusing alcohol and illicit drugs, which likely increases the arrhythmic risk, although we need long-term follow up data to confirm this. The patients, who continue to have further paroxysms, are often offered thromboprophylaxis with aspirin in the absence of contraindications, but the benefit of such an approach is uncertain. Given the uncertainty of aspirin benefit in low-risk atrial fibrillation patients36 and the possibility of harm from the combined effects of alcohol and aspirin intake on the gastric mucosa,37 this needs to be carefully considered.

In conclusion, alcohol and illicit drugs are arrhythmogenic and are associated with atrial fibrillation. Apart from abstinence, the optimal management of such patients and the long-term effects of these substances on the heart and atrial fibrillation recurrences are still unclear.