1Immanuel Klinikum Bernau Herzzentrum Brandenburg Bernau bei Berlin, Deutschland; 2Immanuel Klinikum Bernau Herzzentrum Brandenburg / Kardiologie Bernau bei Berlin, Deutschland; 3Medizinische Hochschule Brandenburg Kardiologie/Angiologie Neuruppin, Deutschland
Background: Left atrial (LA) low voltage areas (LVA) are a controversial target in atrial fibrillation ablation procedures. However, LVA and LA volume are good predictors of arrhythmia recurrence in ablation-naïve patients. Their predictive value in pre-ablated, myopathic atria is uncertain.
Purpose: Patient selection for repeat ablations is sometimes difficult. Predicting factors of rhythm outcome might help with the decision.
Methods: Consecutive patients with recurrent atrial fibrillation (AF) or atrial tachycardia (AT) scheduled for repeat LA ablation were included in the prospective Bernau ablation registry between 2016 and 2020. All patients received a complete LA ultra-high-density map before ablation. Maps were analyzed for LA size, LVA percentage and distribution. The predictive value of demographic, anatomic and mapping variables on AF recurrence was analyzed in univariate logistic regression and integrated in a multivariate Cox regression if the p-value was <0.250.
Results: 160 patients (50.6% male, 1.3 pre-ablations, 60% persistent AF) with complete LA voltage maps were included. Mean follow-up time was 16 ± 11 months. Mean recorded electrograms per map were 9754 ± 5808, mean LA volume was 176.1 ± 35.9 ml and mean rate of LVAs < 0.5 mV was 30.6% ± 23.1%. During follow-up recurrence rate of AF or AT > 30 seconds was 55.6%. Patients with recurrence had a significant higher percentage of LVAs (40.0% vs. 18.8%, p<.001) but no relevant difference in electroanatomic mapping (EAM) derived LA volume (172 vs. 178 ml, p=.299). ROC curves revealed a cut-off of 22% LVAs with highest sensitivity (73.0%) and specificity (71.8%). Based on this cut-off, freedom from recurrence of AF was significantly higher in the LVA group ≤22% (p<0.01), Figure 1. Other relevant clinical, anatomic or laboratory predictors could not be found in multivariate Cox regression analysis, Table 1.
Conclusion: The amount of left atrial LVA is a good predictor for arrhythmia recurrence, even in a cohort of repeat ablations with progressively diseased atria. EAM derived left atrial volume has minor impact on recurrence in this group and should not solely guide the decision in favor or against further invasive treatment.
Table 1 Multivariate Cox Regression for potential predictors of arrhythmia recurrence.
|
Variable |
Exp(B) |
95 % CI |
P-value | |
|
LVA cut-off 22 % |
3.028 |
1.782 |
5.146 |
<0.001 |
|
LAD |
1.0.18 |
0.975 |
1.064 |
0.419 |
|
LA surface area |
1.005 |
0.994 |
1.015 |
0.378 |
|
Hypertension |
1.283 |
0.696 |
2.364 |
0.425 |
|
Years since diagnosis |
0.963 |
0.894 |
1.037 |
0.319 |
|
NT-proBNP |
1.000 |
1.000 |
1.000 |
0.311 |
LVA = low voltage areas, LAD = left atrial diameter, LA = left atrial.