Low serum albumin levels in patients with paroxysmal atrial fibrillation: what does it mean?

BACKGROUND

Atrial fibrillation is the most common cardiac rhythm disturbance in clinical practice; its risk factors are complex and far from being clarified up to now. Previous studies demonstrated that hypoalbuminaemia significantly correlates with cardiac diseases.

OBJECTIVE

We hypothesized that the serum albumin levels were decreased in patients with paroxysmal atrial fibrillation (PAF) and that hypoalbuminaemia was a risk factor for PAF.

METHODS

We selected retrospectively 32 consecutive PAF patients as experimental group and 32 strictly age- and sex-matched paroxysmal supraventricular tachycardia patients confirmed by electrophysiology as a normal control group. Serum albumin was assayed by the bromocresol green (BCG) method. The experimental group was compared with the control group using the paired sample t test for independent samples when dealing with approximately normally distributed variables and the Wilcoxon rank-sum test when appropriate. One sample t test was used to compare albumin levels of the control group with the normal population mean value established in this laboratory so as to eliminate the selection bias. The correlations between albumin levels and cardiac function classifications (NYHA) were assessed using Spearman's rho. Categorical variables were compared using chi-squared, Fisher's exact test or RxC contingency table. Logistic regression was used for analysis of risk factors of serum albumin levels and PAF.

RESULTS

The baseline clinical characteristics between the 2 groups were comparable with no significant difference. In a univiariate analysis, albumin was significantly lower in patients with PAF (41.67 +/- 3.52 g/L) than in control patients (43.88 +/- 3.53 g/L, P = 0.009). The albumin levels among patients in the presence and the absence of baseline factors showed insignificant difference (P > 0.05). In a multivariable analysis of covariance that considered baseline factors, only PAF was an independent predictor of decreased albumin levels (P = 0.0454, OR = 0.375). Conversely, only hypoalbuminaemia was an independent predictor of PAF (P = 0.0129, OR = 0.773).

CONCLUSIONS

PAF is associated with hypoalbuminaemia and hypoalbuminaemia is an independent risk factor for PAF patients. Although the cause and effect of hypoalbuminaemia and PAF remains to be clarified, the present study provides new information with respect to the aetiology and therapy of PAF patients.