Introduction: Diabetes mellitus is frequently accompanied by cardiac rhythm disorders. On the other hand, atrial fibrillation is the most frequent cardiac arrhythmia in adult population [1, 2]. According to some of the large epidemiological studies diabetes mellitus is among independent risk factors for development and persistence of atrial fibrillation . Both diabetes mellitus and atrial fibrillation independently increase the risk of thromboembolism, especially of stroke [3-5]. It is obvious that rhythm control, i.e. restoration and maintenance of sinus rhythm, may be essential for prevention of thromboembolism in these patients.
The aim of the study: The aim of this study is to analyse the impact of diabetes mellitus on rhythm control in patients with persistent atrial fibrillation.
Methods: We analysed the impact of diabetes mellitus and other clinical and echocardiographic parameters (age, gender, current arrhythmia duration, presence of previous episodes of persistent atrial fibrillation, cardiac and/or noncardiac diseases, left atrial diameter and left ventricular ejection fraction) on outcome of attempted cardioversion in patients with persistent atrial fibrillation admitted to Cardiologic Department of the Institute of Cardiovascular Diseases, Clinical Centre of Serbia, between January 1992 and December 1999. We also analysed retrospectively the impact of diabetes mellitus and other parameters listed above on the presence of previous episodes of atrial fibrillation in our patients, that at our opinion reflected the possibilities of sinus rhythm maintenance in these patients. All continuous parameters were expressed as mean value and standard deviation. Statistical significance of differences between variables was examined using Chi-square test. For identification of independent predictors of examined outcomes we used multiple logistic regression model with 95% of confidence interval. Statistical analysis was performed using Statistical Package for Social Sciences (SPSS) programme.
Results: Of 378 patients with currently persistent atrial fibrillation, aged mean 53.98 +/- 11.69 years, there were 266 (70.4%) men. Diabetes mellitus was previously diagnosed in 27 (7.1%) patients, cardiac diseases in 223 (59.0%), noncardiac diseases in 47 (12.4%) and 140 (37.0%) patients had "lone" atrial fibrillation. Left atrial enlargement was noted in 224 (59.3%) patients, and reduced left ventricular ejection fraction in 82 (21.7%). Atrial fibrillation lasted 48 hours to 9 years, mean 8.5 +/- 18.14 months before cardioversion. While 43 patients had previous episodes of persistent AF for last 1-30 years, mean 10.5 +/- 7.3,335 patients never experienced AF before. There was a statistically significant difference in percent of diabetic patients (18.6%/43 vs. 5.7%/335, value of Chi-square test = 7.759, p < 0.01) in these two groups. We analysed the impact of diabetes mellitus on outcome of attempted cardioversion and on presence of previous episodes of AF reflecting the success in maintaining sinus rhythm. Multiple logistic regression models for all of 378 patients, with dependent variable being present in previous recurrent atrial fibrillations and independent variables of clinical and echocardiographic parameters as listed, identified diabetes mellitus to be an independent predictor of repeated atrial fibrillations with relative risk of 4.6 (CI 95%). When dependent variable in the same model was outcome of cardioversion (sinus rhythm is restored in 281/378 patients--74%) diabetes mellitus was not among independent predictors of successful cardioversion.
Discussion: The relationship between atrial fibrillation and diabetes mellitus is not completely understood, including the impact of known complications of diabetes mellitus on electrophysiological properties of atrial myocardium and development of atrial fibrillation . Besides being the independent risk factor for occurrence of atrial fibrillation, diabetes mellitus, according to our results, appears to influence the possibilities of maintaining sinus rhythm after cardioversion of permanent atrial fibrillation in diabetic patients. We found that patients with diabetes mellitus and persistent atrial fibrillation may be successfully converted to sinus rhythm like any other group of patients, but the presence of diabetes mellitus increases the risk of arrhythmia recurrence for 4.6 times compared to patients without diabetes mellitus. Obviously, diabetic patients need to be treated with more efficacious antiarrhythmics from the very beginning, including amiodarone, which successfully prevents recurrent atrial fibrillation in the majority of patients [7, 8].
Conclusion: We concluded that diabetes mellitus independently predicts the presence of recurrent atrial fibrillation but does not influence the possibility of sinus rhythm restoration. The relationship between atrial fibrillation and diabetes mellitus needs further investigation.