Syncope: patients with irregular pulses were more likely to have an arrhythmia.

Clinical bottom line (level 4)

  1. Patients who described an irregular pulse were more likely to have an arrhythmia (LR+8.9) .
  2. No other symptoms were very helpful.
Zeldis et al: Chest 1980; 78 (3): 456-462
Expires June 2004

The study

Setting: university hospital, USA

477 patients (aged mean 53 years, 51% male) referred for Holter monitoring due to 'pulse irregularities' or palpitations, chest discomfort, syncope, dyspnoea, dizziness, seizures

Independent blinded reference standard, applied in all patients from a ?consecutive appropriate spectrum.
Reference standard:
  • 24 hour Holter monitoring
Diagnostic test: detailed log of symptoms kept by patient

The evidence

pre-test probability of arrhythmia: 53%, (95% CI: 49% to 57%)

diagnostic test arrhythmia no arrhythmia LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
palpitations 43 42 0.91
(0.62 to 1.3) 		51% 1.0
(0.94 to 1.1) 		54%
syncope (or near syncope) 15 24 0.55
(0.30 to 1.0) 		38% 1.1
(1.0 to 1.1) 		54%
dizziness 11 10 0.97
(0.42 to 2.3) 		52% 1.0
(0.96 to 1.0) 		53%
chest pain 13 14 0.8
(0.40 to 1.7) 		48% 1.0
(0.97 to 1.1) 		53%
pulse irregularity 10 1 8.9
(1.1 to 69) 		91% 0.96
(0.94 to 0.99) 		52%
dyspnoea 4 4 0.89
(0.22 to 3.5) 		50% 1.0
(0.98 to 1.0) 		53%
known arrythmia 11 6 1.6
(0.61 to 4.3) 		65% 0.98
(0.95 to 1.0) 		52%
stroke 4 5 0.71
(0.19 to 2.6) 		44% 1.0
(0.98 to 1.0) 		53%
seizures 3 2 1.3
(0.22 to 7.9) 		60% 1.0
(0.98 to 1.0) 		53%
total 253 224

  • No combination of symptoms was any better at diagnosing or excluding arrhythmias.

Citation

  1. Zeldis SM, Levine BJ, Michelson EL, et al: Cardiovascular complaints: correlation with cardiac arrhythmias on 24-hour electrocardiographic monitoring. Chest 1980; 78 (3): 456-462
Contributor: Chris Ball and Clare Wotton, June 2000
Reviewer:

Clinical Question.
Patient pulse irregularities
Intervention or Exposure symptoms
Outcome diagnosis of arrhythmia