Background: Atrial fibrillation (AF) is the most common sustained arrhythmi
a diagnosed in non-uremic patients and its prevalence increases in older su
bjects, however, information concerning AP in dialysis patients is scarce.
Therefore, we carried out a prospective cross-sectional study from Septembe
r 1996 to December 1996 in order to evaluate the prevalence and some of the
clinical characteristics associated to AF in hemodialysis (HD) patients. S
ubjects and methods: 316 HD patients (age 63 +/- 12 years, dialysis duratio
n 69 +/- 71 months) treated in three different hospital-based units were st
udied. Standard 12-lead electrocardiograms (ECGs) carried out in the interd
ialytic day during the study period were reviewed. Data concerning age, his
tory of ischemic heart disease (IHD), cerebrovascular disease (CVD), periph
eral vascular disease (PVD), presence of diabetes, smoking history and anti
hypertensive therapy were collected. Systolic and diastolic blood pressure,
fasting cholesterol and triglycerides, albumin and hemoglobin were also de
rived from the clinical records. Performance status was assessed by Karnofs
ky index (Ki). Results: 74 patients (23.4%) had persistent AF, i.e, presenc
e of AF in all (at least two) ECGs performed in the study time. Patients wi
th AF were older (age 69+/- 10 vs 62+/-12 years, p <0.001), had lower Ki (5
4 +/- 20 vs 68 +/- 17, p < 0,01), cholesterol (182+/- 46 vs 198 +/- 52 mg/d
l, p < 0.01) and albumin (3.9 +/- 0.5 vs 4.1 +/- 0.5 g/dl, p < 0.001) compa
red to those with no AF. Prevalence of IHD (44.5% vs 19%, p < 0.05) and PVD
(23% vs 11%, p < 0.05) was higher among AF patients. Logistic regression a
nalysis showed that IHD (p < 0.001) and Ki (p < 0.01) were independently as
sociated to AF. Conclusion: We conclude that AF is a frequent arrhythmia in
HD patients treated in hospital-based dialysis units, especially in those
with low performance status. It appears to be associated to the atheroscler
otic damage of coronary arterial tree. Prospective studies are necessary to
assess whether it could contribute to cardiovascular morbidity and mortali
ty in end-stage renal disease.