Angiotensin II exerts proinflammatory effects leading to atrial fibrosis that is associated with persistence of atrial fibrillation (AF). Renal function plays a major role in activation of the renin–angiotensin–aldosterone system. We examined whether the level of impaired renal function, defined by glomerular filtration rate (GFR), would influence the maintenance of sinus rhythm after successful external electric cardioversion (ECV). One hundred two consecutive patients with persistent AF underwent successful ECV. Patients were prospectively followed for recurrence of AF by telephone interviews, Holter electrocardiograms, and electrocardiograms sent by primary care providers. Repeated GFR assays were performed before and 1 month after ECV. Patients were divided into 4 groups according to baseline GFR (I >90 ml/min, II 60 to 90 ml/min, III 30 to 59 ml/min, IV <30 ml/min). AF recurrence rate was significantly higher in patients with moderately or severely decreased renal function (GFR <60 ml/min, p = 0.003). Patients with moderately (GFR 30 to 59 ml/min, p = 0.02) or only mildly (GFR 60 to 90 ml/min, p = 0.01) decreased renal function showed an increase in GFR if sinus rhythm was maintained at 1 month follow-up. In conclusion, impaired renal function was associated with an increased risk of AF recurrence after successful ECV.
The prevalence of atrial fibrillation (AF) is associated with renal impairment and decreasing glomerular filtration rate (GFR). In patients with end-stage renal disease, prevalence rates from 13% to 23.3% have been reported. However, no previous study has assessed the role of renal impairment in predicting the risk of recurrence after electrical cardioversion (ECV) in patients with persistent AF. We prospectively studied the association of GFR and recurrence of AF after successful ECV in consecutive patients with persistent AF.
Methods
One hundred two consecutive patients with persistent AF (lasting >7 days) who underwent successful elective ECV from August 2007 to July 2008 were included in the study. Transesophageal echocardiography was performed in all patients to exclude thrombus formation in the left atrial appendage before ECV. Successful ECV was defined as conversion to stable sinus rhythm (SR) for ≥15 minutes after ECV. Exclusion criteria were inflammatory or neoplastic diseases, acute coronary syndromes within 4 weeks before ECV, previously performed AF ablation procedures, left ventricular dysfunction with ejection fraction <45%, left ventricular hypertrophy (left ventricular wall thickness >12 mm, maximal wall thickness in parasternal short-axis view at end-diastole) or enlarged left atrial diameter (>50 mm, parasternal long-axis view by echocardiography), and thrombus formation in the left atrial appendage.
To exclude reversible acute renal failure causing decreased GFR, all patients with renal dysfunction were given saline 24 hours before ECV. GFR analysis was performed ≥2 times within 3 days before ECV.
R-wave–synchronized ECV using an external biphasic defibrillator (LIFEPAK 20, Medtronic, Inc., Minneapolis, Minnesota) was performed in all patients. Energy requirements were titrated from 100 to 360 J according to clinical response. Anticoagulation was maintained according to current guidelines and to the individual risk of stroke. All patients were prospectively followed to 1 month for recurrence of AF by telephone interviews and by electrocardiograms sent by primary care providers. Additional office visits with electrocardiographic monitoring were performed in case of symptom recurrence. After 1-month follow-up, all patients underwent an outpatient visit including a clinical examination, analysis of GFR, 48-hour Holter monitoring, and echocardiography evaluating left atrial size and left ventricular ejection fraction. Recurrent AF was defined as any symptomatic or asymptomatic detected episode lasting >30 seconds. Antiarrhythmic therapy was not modified during the study period. Patients were treated with β blockers or amiodarone according to the discretion of the physician. The end point of the study was the presence of recurrence of AF according to GFR levels. This study was approved by the ethics committee of the University of Freiburg (Freiburg, Germany).
Blood samples for analysis of GFR were drawn immediately before ECV. Citrated plasma was stored at −70°C until analysis and kidney function was then assessed by calculation of the GFR using the abbreviated Modification of Diet in Renal Disease Study (MDRD) equation: GFR (milliliters per minute per 1.73 m 2 ) = 186.3 × (serum creatinine [milligrams per deciliter] −1.154 ) × (age −0.203 ) × (0.742 if a woman).
Continuous variables are reported as means ± SDs. Categorical variables are reported as numbers and percentages. Comparison of continuous variables with analysis of variance was performed. Multiple comparisons between groups (GFR quartiles) were performed using Kruskal-Wallis analysis of variance. Proportions were compared using Yates-corrected chi-square test.
Potential predictors for recurrence of AF entered a univariate screening. Those with a p value <0.20 were added to the multivariate stepwise logistic regression model with backward variable selection. Because statins and antiarrhythmic drugs are known risk factors for recurrence of AF, these values were also forced into the model. The resulting hazard ratios with 95% confidence intervals and p values were reported.
To define the optimal cut-off value for GFR as a predictor for AF recurrence, receiver operating characteristics analysis was performed. For exploratory purposes, the optimal cut-off value for GFR was defined as the value that maximized the sum of sensitivity and specificity.
Statistical analysis was done with SPSS 16.0 (SPSS, Inc., Chicago, Illinois).
Results
One hundred two consecutive patients were included in the analysis. Demographics and baseline characteristics of the population are presented in Table 1 .
| Variable | No | Yes | p Value |
|---|---|---|---|
| (n = 58) | (n = 44) | ||
| Men | 28 (48%) | 32 (72%) | 0.08 |
| Age (years) | 68 ± 11 | 71 ± 9 | 0.09 |
| Body mass index (kg/m 2 ) | 30 ± 8 | 28 ± 4 | 0.82 |
| AF duration (months) | 11 ± 26 | 12 ± 24 | 0.76 |
| Glomerular filtration rate at baseline (ml/min) | 73 ± 21 | 52 ± 25 | 0.001 |
| Glomerular filtration at 1 month (ml/min) | 82 ± 20 | 52 ± 24 | <0.001 |
| C-reactive protein (mg/dl) | 1.0 ± 1.5 | 1.1 ± 2.0 | 0.29 |
| Diabetes mellitus | 11 (19%) | 10 (22%) | 0.47 |
| Hypertension | 41 (71%) | 33 (75%) | 0.85 |
| Coronary artery disease | 22 (38%) | 18 (40%) | 0.53 |
| Ejection fraction (%) | 60 ± 8 | 59 ± 8 | 0.37 |
| Left atrial diameter (mm) | 47 ± 4 | 46 ± 4 | 0.61 |
| Left ventricular wall thickness (mm) | 10.3 ± 2.3 | 10.1 ± 1.7 | 0.72 |
| Cardioversion (J) | 213 ± 44 | 237 ± 67 | 0.23 |
| β blockers | 49 (84%) | 39 (88%) | 0.33 |
| Amiodarone | 3 (3%) | 3 (3%) | 0.80 |
| Class I antiarrhythmics | 4 (7%) | 2 (5%) | 0.84 |
| Angiotensin-converting enzyme inhibitors | 25 (43%) | 25 (56%) | 0.98 |
| Angiotensin receptor blockers | 5 (8%) | 3 (7%) | 0.31 |
| Statins | 22 (38%) | 19 (43%) | 0.67 |
Biphasic ECV was successfully performed in all patients (criterion for inclusion in the study) and 1-month follow-up was completed in all patients. There were no deaths or major cardiovascular events (myocardial infarction or stroke) during follow-up. Estimated durations of AF before ECV were up to 1 month in 42 patients (41%), 1 month to 1 year in 43 patients (42%), and ≥1 year in 17 patients (17%).
Patients were divided into 4 groups according to stages of chronic kidney diseases based on their GFRs (I >90 ml/min, II 60 to 90 ml/min, III 30 to 59 ml/min, IV <30 ml/min). Group IV included 5 patients with end-stage renal disease on long-term hemodialysis. Forty-four patients (43%) developed recurrence of AF at 1-month follow-up, 31 evidenced persistent AF, and 13 paroxysmal AF. Clinical features according to GFR are listed in Table 2 . The 4 groups of patients were similar in age, gender, hypertension, diabetes, use of antiarrhythmic drugs, left atrial size, left ventricular ejection fraction, and duration of AF before ECV ( Table 2 ).
| Characteristic | GFR (ml/min) | p Value | |||
|---|---|---|---|---|---|
| <30 | 30–59 | 60–90 | >90 | ||
| (n = 12) | (n = 31) | (n = 36) | (n = 23) | ||
| Men | 8 (75%) | 16 (51%) | 21 (58%) | 15 (65%) | 0.33 |
| Age (years) | 73 ± 7 | 73 ± 8 | 68 ± 9 | 66 ± 12 | 0.06 |
| Body mass index (kg/m 2 ) | 28 ± 3 | 29 ± 4 | 31 ± 10 | 28 ± 3 | 0.54 |
| AF duration (months) | 6 ± 7 | 16 ± 36 | 12 ± 22 | 5 ± 6 | 0.55 |
| Creatinine (mg/dl) | 3.4 ± 2.0 | 1.3 ± 0.2 | 1.0 ± 0.2 | 0.8 ± 0.2 | 0.001 |
| C-reactive protein (mg/dl) | 0.9 ± 0.9 | 1.6 ± 2.4 | 0.8 ± 1.0 | 0.5 ± 0.4 | 0.65 |
| Potassium (mmol/L) | 4.4 ± 0.5 | 4.3 ± 0.5 | 4.1 ± 0.4 | 4.0 ± 0.3 | 0.03 |
| Diabetes mellitus | 2 (17%) | 6 (19%) | 9 (25%) | 4 (17%) | 0.42 |
| Hypertension | 9 (75%) | 22 (71%) | 29 (78%) | 15 (65%) | 0.82 |
| Coronary artery disease | 6 (50%) | 12 (39%) | 14 (39%) | 10 (43%) | 0.62 |
| Ejection fraction (%) | 57 ± 9 | 59 ± 7 | 60 ± 8 | 61 ± 7 | 0.11 |
| Left atrial diameter (mm) | 47 ± 3 | 47 ± 4 | 46 ± 3 | 45 ± 5 | 0.11 |
| Left ventricular wall thickness | 10.3 ± 2.0 | 10.0 ± 2.8 | 9.8 ± 1.8 | 10.2 ± 2.5 | 0.29 |
| Cardioversion (J) | 226 ± 62 | 231 ± 64 | 222 ± 54 | 218 ± 54 | 0.82 |
| β blockers | 11 (92%) | 28 (90%) | 31 (86%) | 18 (78%) | 0.86 |
| Amiodarone | 0 (0%) | 2 (6%) | 3 (8%) | 1 (4%) | 0.75 |
| Class I antiarrhythmics | 0 (0%) | 2 (6%) | 1 (3%) | 3 (13%) | 0.16 |
| Angiotensin-converting enzyme inhibitors | 7 (58%) | 16 (52%) | 17 (47%) | 10 (43%) | 0.94 |
| Angiotensin receptor blockers | 2 (17%) | 1 (3%) | 3 (8%) | 2 (9%) | 0.78 |
| Statins | 6 (50%) | 12 (39%) | 16 (44%) | 8 (35%) | 0.57 |
Univariate screening of potentially relevant parameters returned age, used energy, cardioversion attempts, GFR and potassium, and use of angiotensin-converting enzyme inhibitors ( Table 3 ). These parameters then entered the multivariate logistic regression analysis.
| Variable | p Value |
|---|---|
| Age | 0.17 |
| Atrial fibrillation duration | 0.79 |
| Body mass index | 0.28 |
| Energy (J) | 0.04 |
| Cardioversion (attempts) | 0.14 |
| C-reactive protein | 0.91 |
| Ejection fraction | 0.41 |
| Glomerular filtration rate | <0.001 |
| Left atrial size | 0.25 |
| Potassium | 0.001 |
| Diabetes | 0.56 |
| Amiodarone | 0.62 |
| Class I antiarrhythmics | 0.38 |
| β blocker | 0.39 |
| Statins | 0.40 |
| Angiotensin-converting enzyme inhibitors | 0.15 |
Because uses of statins and of antiarrhythmic drugs have been previously reported to be predictive for recurrence of AF, these parameters were forced into the model, although not meeting the inclusion criteria by univariate screening. Using logistic stepwise regression analysis with backward selection, GFR and potassium remained the only independent predictors of AF recurrence ( Table 4 ).
| Variable | Hazard Ratio | 95% Confidence Interval | p Value |
|---|---|---|---|
| Age (years) | 1.01 | 0.96–1.06 | 0.73 |
| Cardioversion (attempts) | 0.75 | 0.13–4.50 | 0.75 |
| Energy (J) | 1.00 | 0.99–1.01 | 0.21 |
| Amiodarone | 0.58 | 0.59–5.70 | 0.64 |
| β blocker | 2.65 | 0.48–14.56 | 0.26 |
| Class I antiarrhythmics | 3.53 | 0.27–46.17 | 0.37 |
| Angiotensin-converting enzyme inhibitors | 0.92 | 0.33–2.56 | 0.88 |
| Statins | 0.85 | 0.31–2.34 | 0.76 |
| Glomerular filtration rate | 0.97 | 0.95–0.99 | 0.004 |
| Potassium | 3.68 | 1.17–11.59 | 0.03 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
