Professional Documents
Culture Documents
Diagnostic Approach To Palpitations
Diagnostic Approach To Palpitations
Diagnostic Approach To Palpitations
ALLAN V. ABBOTT, M.D., Keck School of Medicine of the University of Southern California, Los Angeles, California
A
Patient information: n increased or abnormal aware- at a university medical center who com-
A handout on heart pal- ness of the heartbeat, palpita- plained of palpitations and were followed for
pitations, written by the
author of this article, is
tions are a common symptom in one year, an etiology was determined in 84
provided on page 755. patients presenting to family phy- percent of the patients.
sicians. Palpitations can be symptomatic of Of these patients, 43 percent had palpita-
life-threatening cardiac arrhythmias.1 How- tions caused by cardiac causes (40 percent
ever, most palpitations are benign. In one had an arrhythmia, 3 percent had other
retrospective study2 in a family practice set- cardiac causes), 31 percent had palpitations
ting, there was no difference in the rates of caused by anxiety or panic disorder, 6 percent
morbidity or mortality among patients with had palpitations caused by street drugs or
palpitations compared with matched control prescription and over-the-counter medica-
subjects. tions, and 4 percent had palpitations caused
Although there are many possible cardiac by other noncardiac causes. No specific
etiologies, palpitations can be associated cause of the palpitations could be identified
with noncardiac causes such in 16 percent of the patients. Psychiatric and
Palpitations are potentially as fever, anemia, or drug use, emotional illnesses such as anxiety, panic,
more serious when they and can occur in anxious but and somatization disorders may be underly-
are associated with diz-
otherwise normal persons. The ing problems in many patients.1
ziness, near-syncope, or
differential diagnoses of pal- Although arrhythmias frequently cause
pitations are summarized in palpitations, most patients with arrhyth-
syncope because they sug-
Table 1. mias do not actually notice their arrhyth-
gest tachyarrhythmia.
Consensus or evidence-based mia and are unlikely to report having
guidelines for diagnosing and palpitations.3
managing palpitations have not been devel- This article describes the more common
oped. However, recent studies of palpitation presentations of palpitations and a rational
etiology provide improved evidence that can approach to patient evaluation, and provides
guide a family physician through diagnosis. evidence for making decisions about ambu-
In a prospective cohort study1 of 190 patients latory monitoring.
Downloaded from the American Family Physician Web site at www.aafp.org/afp. Copyright 2005 American Academy of Family Physicians. For the private, noncommercial
use of one individual user of the Web site. All other rights reserved. Contact copyrights@aafp.org for copyright questions and/or permission requests.
STRENGTH OF RECOMMENDATIONS
TABLE 1
Differential Diagnosis of Palpitations
NOTE: The categories of palpitations are arranged from most common to least common; within the categories, condi-
tions are listed in alphabetical order.
744 American Family Physician www.aafp.org/afp Volume 71, Number 4 February 15, 2005
Palpitations
Panic Disorder Questionnaire
Figure 1.
February 15, 2005 Volume 71, Number 4 www.aafp.org/afp American Family Physician 745
TABLE 2
Key Clinical Findings with Palpitations
and Suggested Diagnoses
NOTE: The information in this table is based on clinical experience and not on the EXTRACARDIAC CAUSES
results of clinical trials.
The physician should examine the patient
for extracardiac causes. The patient may
have obvious associated illness with fever,
ruled out before the diagnosis of anxiety or dehydration, hypoglycemia, anemia, or evi-
panic disorder can be accepted as the cause dence of thyrotoxicosis. Use of drugs such as
of the palpitations.1,11,12 cocaine, and alcohol, caffeine, and tobacco
Some physicians may prematurely blame can precipitate palpitations. The use of ephe-
palpitations on anxiety. In one study13 of dra and ephedrine also has been associated
patients with supraventricular tachycardia, with palpitations.15 Many prescription med-
two thirds of the patients were diagnosed ications, including digitalis, phenothiazine,
with panic, stress, or anxiety disorder, and theophylline, and beta agonists, can cause
one half of the patients had an unrecog- palpitations.
746 American Family Physician www.aafp.org/afp Volume 71, Number 4 February 15, 2005
Figure 2. Sinus tachycardia with electrical alternans.
Figure 5. Sinus bradycardia. A slurred upstroke of the QRS, or delta wave, suggests Wolff-Parkinson-White syndrome.
This finding is associated with paroxysmal atrial tachycardia and other supraventricular arrhythmias.
ciated symptoms may be helpful for the of the patient after he or she exercises may
physician in diagnosis. For example, a patient reveal an arrhythmia or murmur that is
who describes single skipped beats is likely exacerbated by the resulting increased heart
to be having benign premature ventricular rate and cardiac output.
contractions. The physician should consider
ECG EVALUATION
the differential diagnoses of palpitations
(Table 1) while questioning the patient. Cer- A 12-lead ECG evaluation is appropriate in
tain clinical findings and possible associated all patients who complain of palpitations.
conditions are listed in Table 2. In the event that the patient is experienc-
Because physicians usually do not get ing palpitations at the time of the ECG, the
the chance to examine the patient dur- physician may be able to confirm the diag-
ing an episode of palpitations, the physical nosis of arrhythmia. Many ECG findings
examination primarily serves to determine warrant further cardiac investigation. These
if there are cardiac or other abnormalities findings include evidence of previous myo-
present that might predispose the patient cardial infarction, left or right ventricular
to palpitations. Careful examination of the hypertrophy, atrial enlargement, atrial ven-
heart may reveal murmurs, extra sounds, or tricular block, short PR interval and delta
cardiac enlargement. Mitral valve prolapse, waves (Wolff-Parkinson-White syndrome),
which is commonly associated with palpita- or prolonged QT interval. Occasionally, the
tions, is suggested by a midsystolic click.16 finding of an isolated premature ventricular
The physician should look for evidence of contraction or premature atrial contrac-
hyperthyroidism (e.g., nervousness, heat tion warrants further monitoring or exercise
intolerance), drug use, or other serious ill- testing. Some common arrhythmias associ-
nesses. Finally, in the occasional patient who ated with palpitations are shown in Figures
has palpitations with exercise, examination 2 through 5.
February 15, 2005 Volume 71, Number 4 www.aafp.org/afp American Family Physician 747
ECG exercise testing is
appropriate in patients
who have palpitations
with physical exertion and
patients with suspected
Further Diagnostic Testing nary artery disease or myocardial ischemia.
coronary artery disease or In patients at low risk for coro- Findings from the physical examination or
myocardial ischemia. nary heart disease who have no ECG may suggest the need for echocardiog-
palpitation-associated symp- raphy to evaluate structural abnormalities
toms such as dizziness, and who and ventricular function.
have negative physical examination and ECG High-risk patients, who require ECG
findings, palpitations may need no further monitoring, include those with organic
evaluation unless the episodes persist or the heart disease or any heart abnormality that
patient remains anxious for an explanation. could predispose the patient to arrhythmias.
Blood tests may be appropriate in the fol- Patients with a family history of arrhyth-
lowing conditions: complete blood cell count mia, syncope, or sudden death also may be
for suspected anemia or infection, electro- at higher risk. The results of one study17 of
lytes for arrhythmia from suspected elec- 24-hour ECG monitoring showed that ven-
trolyte imbalance, and thyroid-stimulating tricular tachycardia was associated with
hormone for suspected hyperthyroidism or previous myocardial infarction, idiopathic
hypothyroidism. dilated cardiomyopathy, significant valvular
ECG exercise testing is appropriate in lesions, and hypertrophic cardiomyopathies.
patients who have palpitations with physical If the etiology of palpitations is not appar-
exertion and patients with suspected coro- ent after the history, physical examination,
748 American Family Physician www.aafp.org/afp Volume 71, Number 4 February 15, 2005
Palpitations
and ECG are completed, the physician should Evidence supports the use of an initial
consider ambulatory cardiac monitoring. two-week course of continuous closed-loop
Figure 6 is an algorithm that can be used in event recording to monitor for palpitations.
the evaluation of patients with palpitations. Holter monitoring for 24 hours is an alter-
native to event monitoring in patients who
CONTINUOUS ECG MONITORS reliably experience palpitations every day, or
The Holter monitor is a simple ECG moni- who are not willing to wear an event moni-
toring device that is worn continuously to tor for two weeks, and if event monitoring is
record data for 24 or 48 hours. The patient not available locally. When palpitations are
must keep a diary of any symptoms that sustained or poorly tolerated, a referral to a
occur during the monitoring.17 Holter moni- cardiologist for an electrophysiologic evalu-
tors typically are the most expensive of the ation may be warranted.21
monitoring devices, and are maintained and
operated by hospitals or larger outpatient Management
clinics. In patients with arrhythmias, the most com-
mon finding on ambulatory monitoring is
TRANSTELEPHONIC EVENT MONITORS benign atrial or ventricular ectopic beats
Transtelephonic event monitors transmit associated with normal sinus rhythm.20-22
recordings by telephone to a central sta- Normal sinus rhythm alone is found in
tion. As with Holter monitors, patients wear about one third of patients. Many patients
continuous-loop event monitors, but unlike with palpitations have ventricular prema-
Holter monitors, these save data only for the ture contractions or brief episodes of ven-
previous and subsequent few minutes when tricular tachycardia; if the evaluation of the
the patient manually activates the monitor. heart is otherwise normal, these findings are
These monitors are smaller than a Holter not associated with increased mortality.23
monitor (i.e., the size of a beeper) and may Appropriate patient education is indicated
miss arrhythmias that are asymptomatic, in these patients. The treatment of sus-
or that occur during sleep or with syncope. tained arrhythmias involves pharmacologic
Another type of transtelephonic monitor is or invasive electrophysiologic management
not worn continuously but is carried by the and is beyond the scope of this article.
patient and held to the chest when palpita- If the patient is diagnosed with a non-
tions are perceived. This monitor records cardiac, psychiatric, or nonarrhythmia car-
ECG data for about two minutes and is likely diac etiology, the underlying condition is
to miss the onset of arrhythmia. managed according to the diagnosis. In
some patients, a thorough history, physical
Choosing an Ambulatory examination, diagnostic testing, and cardiac
Monitoring Device monitoring all fail to reveal any abnormality
The results of a review18 of studies comparing or etiology for palpitations. These patients
Holter monitors and transtelephonic event should be advised to abstain from caffeine
monitors in the diagnosis of palpitations and alcohol, as well as foods or stressful
found that the diagnostic yield was 66 to situations that appear to trigger palpitations.
83 percent when event monitors were used Fortunately, the majority of patients with
for monitoring, and 33 to 35 percent when palpitations have benign diagnoses and can
Holter monitors were used. Furthermore, be treated with reassurance.
event monitors have been found to be sig-
The author indicates that he does not have any conflicts
nificantly more cost effective than Holter of interest. Sources of funding: none reported.
monitors.19,20 The results of retrospective and
Figures 2 through 5 used with permission from Allan V.
prospective trials19,20 showed that 83 to 87 Abbott, M.D.
percent of patients had diagnostic transmis- This article is one in a series on problem-oriented diagno-
sions within the first two weeks of using a sis coordinated by the Department of Family Medicine at
transtelephonic event monitor. the University of Southern California, Los Angeles, Calif.
February 15, 2005 Volume 71, Number 4 www.aafp.org/afp American Family Physician 749
Palpitations
750 American Family Physician www.aafp.org/afp Volume 71, Number 4 February 15, 2005