The prevalence of atrial fibrillation (AF) is increased in hyperthyroidism. The degree to which thyroid hormones affect the outcomes of left atrial (LA) ablation is still unclear. From September 2010 to September 2013, 1,095 patients who underwent LA ablation (59.7% paroxysmal AF, 32.3% persistent AF, and 8.0% LA tachycardia) had their serum thyroid-stimulating hormone (TSH) and free thyroxine (FT 4 ) levels measured in the 48 hours before the procedure. Patients were followed until they presented the first AF relapse after a blanking period of 3 months. TSH and FT 4 were assessed as predictors of arrhythmia relapse and were adjusted for possible confounders. During a mean follow-up of 12.5 ± 7.9 months, 28.9% of patients presented an atrial arrhythmia relapse. TSH was not a predictor of relapse. In contrast, after adjustment, FT 4 (median = 11.8 ng/L and interquartile range 10.6 to 14.6 ng/L) remained a predictor of relapse with 15% increase per quartile (hazard ratio 1.15, 95% confidence interval 1.03 to 1.29, p = 0.014). In conclusion, FT 4 levels influence the success rate of LA ablation procedures, even when in the normal range.
Atrial fibrillation (AF) is the most common sustained arrhythmia. Over 6 million Europeans have this arrhythmia, and in the next 50 years, as population ages, its prevalence is estimated to double. Overt hyperthyroidism is a well-known risk factor for incident AF. Over the past few years, free thyroxine (FT 4 ) levels were also identified as a predictor of AF, even in subjects with thyroid-stimulating hormone (TSH) levels in the normal range. A high normal level of thyroid function may be a predictor of recurrence after catheter ablation of paroxysmal AF. However, to the best of our knowledge, the impact of FT 4 levels on persistent AF and left atrial tachycardia (LAT) ablation is still unknown. We aimed to assess the impact of FT 4 levels (procedural success and midterm outcomes) in a large sample of patients undergoing left atrial (LA) ablation procedures.
Methods
Paroxysmal, persistent AF and LAT were defined according to the Heart Rhythm Society/European Heart Rhythm Association/European Cardiac Arrhythmia Society Expert Consensus Statement on catheter and surgical ablation of AF.
Patients admitted in a high-volume private hospital for catheter ablation of symptomatic drug-refractory AF or LAT, from September 2010 to September 2013 and with FT 4 and TSH measurement in the 48 hours before the procedure were prospectively enrolled. Patients with overt clinical hyperthyroidism or overt hypothyroidism were excluded, as this was considered a contraindication for catheter ablation by the preassessment anesthetist. Similarly, patients with LA thrombus were excluded as per consensus recommendations.
In total, 1,095 patients were enrolled, and all patients gave written informed consent before the procedure. The study complied with the Declaration of Helsinki and was approved by the institutional review board. Radiofrequency (RF) and cryoballoon ablation was performed by 4 experienced electrophysiologists (each operator is dedicated exclusively to 1 technique and performed more than 100 AF ablations/year). Patients with paroxysmal AF were referred for RF or cryoballoon ablation according to the electrophysiologist discretion because there is no significant outcome difference between both techniques at our institution. Patients with persistent AF were referred in the vast majority to RF ablation. However, some patients presenting with sinus rhythm (SR; mainly earlier persistent AF–lasting <6 months) were referred for cryoballoon ablation. All patients with LAT underwent RF ablation.
Serum TSH and FT 4 were assessed with a commercial assay (Dimension Vista, Newark; Siemens Healthcare Diagnostic Inc.). Laboratory reference ranges for FT 4 and TSH were 7.6 to 14.6 ng/L and 0.358 to 3.740 mIU/L, respectively. Four quartiles were defined according to the distribution of FT 4 levels in this cohort: quartile 1 <10.6, quartile 2 (10.6 to 11.8), quartile 3 (11.8 to 14.6), and quartile 4 ≥14.6. The known history of thyroid disease was defined as previous history of thyroiditis, thyroidectomy, or ongoing medical treatment for hypothyroidism or hyperthyroidism, independently of the actual levels of FT 4 .
Details of the periprocedure management and the paroxysmal ablation technique at our institution have been published previously and are described in detail in the Supplementary Data .
Patients with persistent AF submitted to cryoablation underwent only pulmonary vein isolation (PVI).
The vast majority of patients with persistent AF were treated with RF ablation, through a sequential stepwise approach: wide PVI, electrogram-based ablation, and linear ablation. After defining LA geometry, a wide PVI was performed. However, if the patient was in AF in the beginning of the procedure, the Lasso catheter would be used to perform a complex fractionated atrial electrogram (CFAE) map (automatic algorithms on the basis of bipolar recordings with acquisition of −dV/dT >5 seconds and CFAE were defined as a mean cycle length of <120 ms), which would be used for guiding the wide antral circumferential ablation of the pulmonary veins (PVs). The detailed technique for ablating CFAE using the automated mapping software has been described and validated previously. The procedure terminated if conversion to SR occurred. The second step was ablating the remaining CFAE located outside the PVI lines. If the patient remained in AF, ablation of the mitral isthmus line and LA roof line was empirically performed. When AF was converted to a regular atrial arrhythmia, this was mapped and ablated using the following approach: a circumferential mapping catheter was used to collect an activation map using a 3-dimensional electroanatomic mapping system (Carto 3 [Biosense-Webster, Baldwin Park, California] or Velocity [St. Jude Medical, Inc., St. Paul, Minnesota]). According to the observed propagation and the postpacing intervals with entrainment maneuvers, focal or reentrant atrial tachycardias were targeted for ablation. If a macroreentrant circuit in the LA was identified, the LA roof or mitral isthmus line was targeted. Endocardial and epicardial access through the coronary sinus were used for obtaining block across the mitral isthmus, if necessary. In case of a likely focal or a microreentrant mechanism, ablation targeting the culprit location was performed. In the event of failure to restore SR, direct current cardioversion was performed in the end of the procedure. Block across the different ablation lines was verified while in SR, and additional lesions were delivered targeting the gaps if needed until this was achieved.
Patients were evaluated before hospital discharge, as well as at 1, 3, 6, 9, and 12 months after the procedure. After the first year, follow-up was done on an annual basis. A systematic transthoracic echocardiography and 24-hour Holter monitoring were performed before discharge. Information collected during follow-up included a 12-lead electrocardiogram (ECG) and 24-hour Holter monitoring at each visit. Antiarrhythmic drugs (AADs) were prescribed on discharge only for specific indications (namely relapse during the hospitalization, patients with persistent AF, or need for cardioversion). In those instances, AADs were stopped after 3 months in the absence of recurrence. The first 3 months after the procedure were considered as a blanking period. If there was documented recurrence of symptomatic AF during this interval and the patient required AAD, a previously ineffective but tolerated class I or class III (sotalol) drug was the preferred option. Anticoagulation was prescribed at discharge and on the basis of the CHA 2 DS 2 -Vasc score after the 3-month blanking period. AF or LAT recurrence was defined as any symptomatic or asymptomatic atrial arrhythmia lasting >30 seconds after the blanking period. This was actively ascertained by ECGs and Holter monitoring at 1, 3, 6, 9, and 12 months or by 12-lead ECG in case of symptomatic palpitation during clinical appointment. Holter monitoring was also used for documentation of symptomatic relapses after the first year. Patients with relapse during the blanking period with no response to chemical or electrical cardioversion were classified as having a relapse.
Comparisons were performed between: (1) patients with paroxysmal and nonparoxysmal AF (i.e., persistent AF or LAT) and (2) across the different FT 4 quartiles. The chi-square test was used for nominal variables, and the Student t test or analysis of variance was used for comparison of continuous variables, where appropriate; the Levene’s test was used to check the homogeneity of variance; equivalent nonparametric tests were used when Kolmogorov–Smirnov was in favor of the absence of normal distribution. Results with p <0.05 were regarded as significant.
The impact of FT 4 quartile levels on SR maintenance, after adjustment to baseline differences, was assessed through Cox regression (using the forward stepwise method likelihood ratio; probability for stepwise = 0.05). Kaplan–Meier curves were traced for illustrating SR maintenance in patients in different FT 4 quartiles, and the log-rank test was used for assessing existing differences. A sensitivity analysis was performed after excluding patients with previous use of amiodarone and/or previously known thyroid disease, using a Kaplan–Meier to illustrate freedom from AF. PASW Statistics version 18.0 was used for descriptive and inferential statistical analysis.
Results
The population’s baseline characteristics are detailed in Table 1 . Of the 1,095 patients who underwent LA ablation, 654 patients (59.7%) presented paroxysmal AF, 353 patients (32.3%) presented persistent AF, and 88 patients (8%) presented LAT. Previously known thyroid disease was present in 19.8% and 8.8% (96 patients) of the population that had been previously treated with thyroid hormone. All patients had previously undergone treatment with at least 1 antiarrhythmic agent: 1 agent only was used in 26.8% of patients (a class Ic drug alone was used in 8.1% of patients, sotalol was used in 1.5%, amiodarone in 16.4%, and a β blocker in 0.8%) and 2 or more AADs (in 57.9% amiodarone and another drug and in 15.3% 2 antiarrhythmic agents other than amiodarone) were tried before referral to ablation in the remaining. In sum, amiodarone had been used in 74.3% of population before ablation.
| Overall (n=1095) | Paroxysmal AF (n=654) | Non-Paroxysmal AF (n=441) | P | |
|---|---|---|---|---|
| Age, years | 60.9±9.8 | 60.6±10.2 | 61.4±9.2 | 0.195 |
| Female Gender, % | 27.2 (298) | 29.8 (195) | 23.4 (103) | 0.018 |
| BMI, Kg/m 2 | 27.8±4.5 | 27.2±4.3 | 28.6±4.6 | <0.001 |
| AF duration, years | 5.0±4.5 | 4.6±4.4 | 5.8±4.9 | <0.001 |
| Heart Failure, % | 5.9 (65) | 3.7 (24) | 9.3 (41) | <0.001 |
| Hypertension, % | 46.6 (510) | 43.4 (284) | 51.2 (226) | 0.011 |
| DM, % | 8.9 (97) | 6.7 (44) | 12.0 (53) | 0.003 |
| Stroke/TIA, % | 8.0 (87) | 6.6 (43) | 10.0 (44) | 0.040 |
| Vascular Disease, % | 10.9 (119) | 9.6 (63) | 12.7 (56) | 0.110 |
| CHADS 2 | 0.8±0.9 | 0.7±0.9 | 1.0±1.0 | <0.001 |
| CHA 2 DS 2 -VASc | 1.6±1.3 | 1.5±1.4 | 1.7±1.3 | 0.019 |
| Sleep apnea, % | 9.8 (107) | 8.3 (54) | 12.0 (53) | 0.040 |
| Previously known thyroid disease, % | 19.8 (217) | 19.3 (126) | 20.6 (91) | 0.577 |
| Previously treated with amiodarone, % | 74.3 (687) | 65.9 (361) | 86.5 (326) | <0.001 |
| Clearence (Cockroft-Gault), mL/min | 70.9±23.1 | 71.1±22.9 | 70.7±23.4 | 0.795 |
| C-reactive protein, mg/L | 4.3±8.1 | 3.9±8.6 | 4.8±7.3 | 0.090 |
| Indexed LA volume, mL/m 2 | 44.2±16.8 | 40.5±15.5 | 49.6±17.2 | <0.001 |
| LVEF, % | 63±8 | 64±7 | 61±10 | <0.001 |
Several differences were found between the paroxysmal AF population and the nonparoxysmal AF population, with the latter group presenting a higher CHADS 2 and CHA 2 DS 2 -VASc scores. The nonparoxysmal AF group also presented a higher prevalence of patients previously treated with amiodarone (86.5% vs 65.9%, p <0.001); however, there were no significant differences regarding the existence of known thyroid disease (20.6% vs 19.3%, p = 0.577).
Median FT 4 level was 11.8 ng/L (95% confidence interval 8.8 to 17.0). Patient’s characteristics according to FT 4 quartile are listed in Table 2 . Significant differences were found among the different FT 4 quartiles, regarding gender distribution (p <0.001), the use of amiodarone before ablation (p <0.001), glomerular filtration rate (p <0.001), C-reactive protein (<0.001), and LA volume (p <0.001). Also, patients in the third and fourth quartiles were older (p <0.001), presented higher prevalence of heart failure (p = 0.006) and hypertension (p = 0.007), a higher CHADS 2 (p = 0.001) and CHA 2 DS 2 -VASc scores (p <0.001), and more previously known thyroid disease (p <0.001).
| Quartile 1 <10.6 (n=271) | Quartile 2 [10.6 – 11.8] (n=258) | Quartile 3 [11.8 – 14.6] (n=291) | Quartile 4 ≥14.6 (n=275) | P | |
|---|---|---|---|---|---|
| Age, years | 60.2±9.4 | 60.1±10.5 | 60.2±10.6 | 63.3±8.4 | <0.001 |
| Female Gender, % | 19.6 (53) | 20.2 (52) | 28.5 (83) | 40.0 (110) | <0.001 |
| BMI, Kg/m 2 | 27.8±4.6 | 27.3±4.3 | 27.7±4.4 | 28.3±4.6 | 0.094 |
| AF duration, years | 5.8±5.2 | 4.8±4.4 | 4.7±4.1 | 4.9±4.9 | 0.040 |
| Heart Failure, % | 3.7 (10) | 3.1 (8) | 7.6 (22) | 9.1 (25) | 0.006 |
| Hypertension, % | 46.9 (127) | 39.1 (101) | 45.7 (133) | 54.2 (149) | 0.007 |
| DM, % | 9.2 (25) | 5.4 (14) | 8.9 (26) | 11.6 (32) | 0.092 |
| Stroke/TIA, % | 8.1 (22) | 5.8 (15) | 9.6 (28) | 8.0 (22) | 0.442 |
| Vascular Disease, % | 9.6 (26) | 10.5 (27) | 12.4 (36) | 10.9 (30) | 0.759 |
| CHADS 2 | 0.8±1.0 | 0.6±0.8 | 0.9±1.0 | 1.0±0.9 | 0.001 |
| CHA 2 DS 2 -VASc | 1.4±1.3 | 1.3±1.2 | 1.6±1.4 | 1.9±0.8 | <0.001 |
| Sleep apnoea, % | 10.0 (27) | 9.3 (24) | 10.3 (30) | 9.5 (26) | 0.977 |
| Previously known thyroid disease, % | 18.8 (51) | 14.7 (38) | 15.1 (44) | 30.5 (84) | <0.001 |
| Previously treated with amiodarone, % | 62.9 (141) | 68.3 (151) | 75.1 (184) | 89.8 (211) | <0.001 |
| Haemoglobin, g/dL | 14.6±1.4 | 14.8±1.3 | 15.2±0.8 | 14.6±1.4 | 0.279 |
| Clearence (Cockroft-Gault), mL/min | 76.8±23.1 | 72.8±22.9 | 70.0±24.6 | 64.3±19.6 | <0.001 |
| C-reactive protein, mg/mL | 3.5±4.9 | 3.5±4.3 | 4.1±6.7 | 6.0±13.0 | <0.001 |
| Indexed LA volume, mL/m 2 | 41.4±15.2 | 42.1±17.7 | 45.1±15.5 | 47.9±18.1 | <0.001 |
| LVEF, % | 63±8 | 64±8 | 62±8 | 62±9 | 0.195 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
