Previous studies have reported that increased high-sensitive C-reactive protein (hs-CRP) levels are associated with an inflammatory state. This study investigated the association among hs-CRP, substrate properties, and long-term clinical outcomes after catheter ablation of atrial fibrillation (AF). A total of 137 patients with AF (54 ± 13 years) who underwent mapping and catheter ablation were included. The hs-CRP was measured before the first ablation procedure. The substrate properties (initiating triggers, biatrial mean voltage, and high-frequency sites) of the 2 atria and long-term outcome were investigated in patients in the low hs-CRP group (<75%, 2.92 mg/L) and high hs-CRP group (>75%, 2.92 mg/L). Patients with a higher hs-CRP were associated with an increased number of identified nonpulmonary vein ectopies (34.4% vs 17%, p = 0.034), lower mean left atrial (LA) voltage (1.72 ± 0.73 vs 1.92 ± 0.72 Hz, p = 0.045), and higher-frequency sites in the left atrium (71% vs 37%, p = 0.027). After a median follow-up period of 15 months, the single-procedure success rate (72% vs 53%, p = 0.008) and final success rate after multiple procedures (94% vs 81%, p = 0.02) were higher in the low hs-CRP group. In a multivariable regression model adjusted for other potential covariates, hs-CRP level (p = 0.021) and LA diameter (p = 0.032) were independent predictors of recurrence. In conclusion, baseline CRP levels before the first AF ablation procedure had an independent prognostic value in predicting long-term recurrence. Patients with a high hs-CRP level were associated with an abnormal LA substrate and high incidence of nonpulmonary vein AF sources.
Atrial fibrillation (AF) is the most common type of tachyarrhythmia encountered in clinical practice. Pulmonary vein isolation (PVI) has become the mainstream of nonpharmacologic therapy of AF. However, the recurrence of atrial arrhythmias is frequently observed after the blanking period, and multiple procedures are usually required in some patients. Numerous studies have demonstrated that inflammation may play an important role in the initiation and perpetuation of AF. Increased inflammatory markers correlate with a longer duration of AF and successful cardioversion. However, there is a paucity of data regarding the substrate properties and clinical outcome after PVI in patients with AF and a marked inflammatory burden. The purpose of this study was to investigate the electrophysiologic characteristics in patients with an increased high-sensitivity C-reactive protein (hs-CRP) level and the long-term outcome of catheter ablation in these patients.
Methods
This study enrolled 137 patients with paroxysmal AF (<7 days) and nonparoxysmal AF (mean 53.6 ± 12.6 years of age) who underwent radiofrequency catheter ablation under the guidance of a NavX mapping system (NavX, St Jude Medical, Inc., St. Paul, Minnesota). Each patient underwent an electrophysiological study and catheter ablation in the fasting state, after informed consent was obtained. All antiarrhythmic drugs except for amiodarone were discontinued for ≥5 half-lives before the procedure. No patient received intravenous amiodarone during the electrophysiologic procedure to terminate AF. The method of 3-dimensional electroanatomic mapping has been described previously. Mapping was performed with an irrigated 4-mm tip deflectable catheter (EPT, Boston Scientific Corporation, Natick, Massachusetts) inserted into the left atrium alongside the transseptal sheath without the need for an additional puncture site. A 3-dimensional geometry of the left atrium was then created using the NavX system.
Identification of PV and non-PV triggers was performed in all patients. In patients with paroxysmal AF (n = 107), we attempted to find the spontaneous onset of atrial ectopic beats or repetitive episodes of short runs or sustained AF and to predict the location of the initiating foci at baseline. Regarding the initiating ectopies in patients with nonparoxysmal AF (n = 30), cardioversion was performed first with the identification of the triggers before the PV isolation (n = 12). If sinus rhythm (SR) was difficult to maintain after the electric cardioversion, PVI was performed first. Then, we identified the initiating ectopies after restoration to SR by procedural AF termination or cardioversion (n = 18). Methods of the identification of the PV and non-PV ectopies were described in our previous publications.
Measurements of lipid levels, hs-CRP protein levels, blood glucose levels, and glycated hemoglobin values were performed in a central laboratory. Plasma levels of hs-CRP were determined by the immunoprecipitation method using an in vitro diagnostic assay (Beckman Coulter, Inc., Fullerton, California). Biochemical parameters were measured before catheter ablation of AF. Procedures were carried out according to the instructions of the manufacturers. Median level of hs-CRP level was 1.09 mg/L. We further divided the study population into 2 groups: the low hs-CRP group (<75%, 2.92 mg/L) and high hs-CRP group (>75%, 2.92 mg/L).
Bipolar atrial electrograms were collected from all recording sites in the left atrium with a point-by-point approach sequentially during SR under the guidance of a NavX system and conventional mapping system. Peak-to-peak bipolar voltage of each site was measured. Spectral analysis was performed on single discrete bipolar electrograms during SR (unrectified, Hanning window, 30 to 300 Hz, 1 second in duration). Frequency resolution was 0.54 Hz. Time-domain bipolar signals were sampled at 1,200 Hz and filtered with a bandwidth from 32 to 300 Hz. In patients with SR at the beginning of the procedure, voltage analysis and spectral analysis were performed before catheter ablation. In patients with incessant AF, voltage analysis and spectral analysis were performed after cardioversion to SR. If cardioversion was difficult before ablation or the duration of SR was too short in patients with nonparoxysmal AF, voltage and frequency analyses were obtained after restoration to SR after catheter ablation. In those patients, sites of a previous ablation, including sites with a previous PVI, linear ablation, and/or complex fractionation ablation, were excluded from analyses. The mean dominant frequency of each site in the left atrium and right atrium was measured, and the distribution of the high dominant frequency sites (AF nest: defined as a local dominant frequency >70 Hz) was investigated. The average numbers of mapping sites were 238 ± 97 (range 145 to 436) and 156 ± 67 (range 101 to 461) in the left atrium and right atrium, respectively.
The stepwise procedure of catheter ablation involved the following steps.
Step 1 (isolation of PVs)
After a successful transseptal procedure, continuous circumferential lesions were created encircling the right and left PV ostia guided by the NavX system using an irrigated-tip 3. 5-mm ablation catheter. The intention was to place the radiofrequency lesions ≥1 to 2 cm away from the angiographically defined ostia. Successful circumferential PVI was demonstrated by the absence of any PV activity or dissociated PV activity.
Step 2 (linear ablation by anatomic approach)
After successful isolation of all 4 PVs, additional linear ablation was performed at the anterior roof and lateral mitral isthmus in patients with positive AF inducibility of paroxysmal AF and in all patients with nonparoxysmal AF if AF did not stop after PVI. Linear ablation was guided by the NavX system with the creation of split potentials or an electrographic voltage decrease of >50% after each application of radiofrequency energy. In the right atrium, cavotricuspid isthmus ablation was performed with an 8-mm–tip ablation catheter. Bidirectional conduction block of the cavotricuspid isthmus was confirmed after restoration to SR.
Step 3 (complex fractionated electrographic ablation)
If AF with nonparoxysmal AF did not stop after steps 1 to 2 of the ablation procedure, an additional complex fractionated electrographically guided substrate ablation was performed sequentially based on complex fractionated electrographic maps after PVI. Complex fractionated electrographic ablation was confined to the continuous complex fractionated electrograms (>5 seconds) in the left atrium and proximal coronary sinus (CS). The end point of complex fractionated electrographic site ablation was to obtain a prolongation of cycle length, eliminate complex fractionated electrograms, or abolish local fractionated potentials (bipolar voltage <0.05 mV). The end point of the step 3 procedure was the elimination of all continuous complex fractionated electrograms in the left atrium and CS. If AF terminated during linear ablation through the complex fractionated electrogram, complete linear ablation to an anatomic obstacle or nearest ablation line was performed to prevent proarrhythmias.
Step 4 (non-PV ectopic ablation)
After SR was restored from AF by procedural AF termination or electric cardioversion, mapping and ablation were applied only to spontaneously initiating focal atrial tachycardias and non-PV ectopy that initiated AF. If any non-PV ectopy initiating AF from the superior vena cava was identified, isolation of the superior vena cava was guided by circular catheter recordings from the superior vena cava–atrial junction.
In this study, the end point of catheter ablation of paroxysmal AF was the noninducibility of AF. If AF remained inducible after LA linear ablation, cardioversion was performed to restore SR from AF. Then, only non-PV ectopies were targeted. In patients with nonparoxysmal AF, the end point of catheter ablation was procedural AF termination. An AF inducibility test was not performed routinely in patients with persistent AF or long-lasting persistent AF.
After discharge, patients underwent follow-up (2 weeks after ablation, then every 1 month to 3 months thereafter) at our cardiology clinic or with the referring physicians where routine electrocardiograms were obtained during each follow-up, and oral amiodarone was prescribed for 8 weeks to prevent any early recurrence of AF. All patients received oral amiodarone for 8 weeks. If patients could not tolerate amiodarone, propafenone or flecainide was used in those patients. When patients developed symptoms suggestive of a tachycardia after ablation, 24-hour Holter monitoring and/or cardiac event recording with a recording duration of 1 week were performed to define the cause of clinical symptoms. Recurrence of an atrial arrhythmia was defined as an episode lasting >1 minute and that was confirmed by electrocardiograms 2 months after ablation (blanking period). The end point for follow-up was the clinically documented recurrence of atrial arrhythmias or repeat ablation procedures.
Data were presented as mean ± SD if normally distributed. Because the distribution of hs-CRP levels was not normally distributed, CRP levels were presented as median values with an interquartile range. Chi-square test with Fisher’s exact test was used for categorical data. Normally distributed continuous variables were compared using Student’s t test, whereas abnormally distributed variables were compared using the Mann-Whitney U test. Various clinical and electrophysiologic factors were used to assess predictors of a recurrence of atrial arrhythmias after the first procedure. Variables selected to be tested in the multivariate analysis were those with a p value <0.2 in univariate models. Logistic regression was applied for multivariate analysis. Freedom from atrial arrhythmias after the first procedure was also determined and compared using Kaplan-Meier analysis and log-rank test. Statistical significance was considered when the 2-sided p value was <0.05.
Results
A total of 107 patients (78%) with paroxysmal AF and 30 (22%) with nonparoxysmal AF were included. LA diameter was larger in patients with nonparoxysmal AF (43.7 ± 8.3 vs 37.8 ± 5.0 mm, p <0.001). However, hs-CRP level was similar in patients with paroxysmal AF (median 0.89 mg/L, interquartile range 0.56 to 2.17) and nonparoxysmal AF (median 1.21 mg/L, interquartile range 0.49 to 2.90). Clinical characteristics, left ventricular ejection fraction, LA size, previous medications, and proportion of structural heart disease were similar between patients with high and low hs-CRP levels ( Table 1 ). Patients with a high hs-CRP level had higher plasma glucose and glycated hemoglobin levels compared to those with a lower hs-CRP level.
| Clinical and Substrate Factors | Low hs-CRP (n = 105) | High hs-CRP (n = 32) | p Value |
|---|---|---|---|
| Age (years) | 53.2 ± 12.4 | 54.1 ± 12.7 | 0.70 |
| Men | 77 (73%) | 24 (75%) | 0.85 |
| Persistent atrial fibrillation | 25 (24%) | 5 (16%) | 0.24 |
| Underlying heart disease | 29 (28%) | 13 (41%) | 0.10 |
| Hypertension | 29 (28%) | 14 (44%) | 0.08 |
| Diabetes mellitus | 5 (4.9%) | 3 (9.4%) | 0.34 |
| Left atrial diameter (mm) | 38.5 ± 5.15 | 38.4 ± 5.16 | 0.63 |
| Let ventricular ejection fraction (%) | 58.4 ± 7.38 | 58.6 ± 6.02 | 0.920 |
| Body mass index (kg/m 2 ) | 25.3 ± 2.82 | 25.6 ± 3.01 | 0.71 |
| Angiotensin-converting enzyme inhibitor | 7 (6.9%) | 0 | 0.13 |
| Angiotensin II receptor blockers | 16 (16%) | 7 (23%) | 0.39 |
| Statin | 10 (10%) | 3 (10%) | 0.99 |
| Steroid | 0 (0%) | 0 (0%) | 1.00 |
| Fasting glucose (mg/dl) | 95.5 ± 22.7 | 110 ± 71 | 0.04 |
| Glycated hemoglobin (%) | 5.65 ± 0.74 | 6.10 ± 0.87 | 0.04 |
| Triglyceride (mg/dl) | 133 ± 101 | 107 ± 41 | 0.17 |
| Total cholesterol (mg/dl) | 176 ± 36 | 172 ± 31 | 0.53 |
| High-density cholesterol (mg/dl) | 49 ± 13 | 50.0 ± 12 | 0.67 |
| Low-density cholesterol (mg/dl) | 112 ± 31 | 109 ± 28 | 0.68 |
| hs-CRP (mg/L) | 0.8 (0.5–1.3) | 5.6 (3.7–7.65) | <0.001 |
In the first ablation procedure, all 137 patients (100%) underwent successful PVI ablation of all 4 PVs with elimination of PV potentials. In patients with paroxysmal AF (n = 107), AF inducibility was positive in 21 patients (20%) after successful PVI; therefore, an LA linear ablation was performed in those patients (LA roof lines in 15 patients, 14%, and LA lateral mitral lines in 11, 10%). All patients with nonparoxysmal AF (n = 30) underwent LA liner ablation (LA roof lines and lateral mitral lines), and 21 (70%) underwent a complex fractionated electrogram ablation. Non-PV ectopies were observed in 21 of 107 patients (20%) and 8 of 30 patients (27%) with paroxysmal and nonparoxysmal AF, respectively ( Table 2 ). Regarding the regional distribution of the non-PV ectopies, more non-PV ectopies were observed in the LA septum in patients with nonparoxysmal AF (3 of 107, 2.8%, vs 4 of 30, 13.3%, in paroxysmal AF and nonparoxysmal AF, respectively, p = 0.04).
| Clinical and Substrate Factors | Low hs-CRP Group | High hs-CRP Group | ||||
|---|---|---|---|---|---|---|
| Total | Paroxysmal AF | Nonparoxysmal AF | Total | Paroxysmal AF | Nonparoxysmal AF | |
| (n = 105) | (n = 80) | (n = 25) | (n = 32) | (n = 27) | (n = 5) | |
| Patients with PV-initiating triggers | 89 (84%) | 71 (89%) | 18 (72%) | 25 (82%) | 22 (82%) | 4 (80%) |
| Averaged number of PV triggers | 1.12 ± 0.70 | 1.21 ± 0.65 | 0.92 ± 0.78 | 1.00 ± 0.66 | 0.96 ± 0.59 | 1.17 ± 0.98 |
| Patients with non–PV-initiating triggers | 18 (17%) | 12 (15%) | 6 (24%) | 11 (34%) ⁎ | 9 (30%) | 2 (40%) |
| Mean bipolar voltage in left atrium during SR (mV) | 1.92 ± 0.72 | 2.04 ± 0.89 | 1.57 ± 1.01 ‡ | 1.72 ± 0.73 ⁎ | 1.81 ± 0.68 | 0.96 ± 0.41 ‡ |
| Mean bipolar voltage in right atrium during SR (mV) | 1.74 ± 0.77 | 2.01 ± 1.00 | 1.70 ± 0.66 | 2.09 ± 0.89 | 2.33 ± 0.91 | 1.70 ± 0.71 ‡ |
| Presence of low-voltage zone (<0.5 mV) in left atrium/right atrium (%) | 32.9/3.8 | 25.0/5.4 | 52.4/3.9 ‡ | 47.6/9.5 | 41/5.9 | 60/60 |
| Mean dominant frequency value during SR in left atrium (Hz) | 42.4 ± 9.3 | 43.7 ± 6.9 | 39.3 ± 14.9 | 47.6 ± 8.9 ⁎ | 47.4 ± 7.8 | 51.4 ± 13.8 ‡ |
| Mean dominant frequency value during SR in right atrium (Hz) | 43.8 ± 13.7 | 45.9 ± 11.1 | 35.1 ± 17.3 | 48.1 ± 6.6 | 45.7 ± 11.6 | 53.9 ± 9.0 |
| Distribution of AF nest in the PV/left atrium/PV + left atrium (%) | 63%/20%/17% | 88%/13%/8% | 28%/29%/43% ‡ | 29%/36%/35% | 33%/42%/25% | 0%/40%/60% |
| Long-term success rate (single procedure) | 76 (72%) | 59 (73%) | 17 (68%) | 16 (50%) † | 13 (48%) | 3 (60%) |
| Long-term success rate (multiple procedures) | 99 (94%) | 76 (96%) | 22 (88%) | 26 (81%) ⁎ | 22 (81%) | 4 (80%) |
† p <0.01 compared to patients with low hs-CRP.
‡ p <0.05 compared to corresponding patients with paroxysmal AF.
The number of patients with PV-initiated ectopy and the average number of PV ectopies were similar in patients with high and low hs-CRP levels ( Table 2 ). The number of identified non-PV ectopies was significantly higher in patients with a high hs-CRP level (34% vs 17%, p = 0.034). They were mainly located in the nonpulmonary thoracic veins (80% and 58% in patients with high hs-CRP and low hs-CRP levels, respectively). The incidence of non-PV ectopies from the superior vena cava was significantly higher in patients with a high hs-CRP level ( Table 3 ).
| Location | First Procedure | Repeat Procedures | ||
|---|---|---|---|---|
| High hs-CRP Group | Low hs-CRP Group | High hs-CRP Group | Low hs-CRP Group | |
| (n = 32) | (n = 105) | (n = 18) | (n = 20) | |
| Superior vena cava | 8 (25%) | 9 (8.6%) ⁎ | 3 (17%) | 2 (10%) |
| Left atrial septal wall | 2 (6.3%) | 5 (4.8%) | 2 (11%) | 2 (10%) |
| Ligament of Marshall | 3 (9.3%) | 3 (2.8%) | 1 (6%) | 0 |
| Left atrial posterior wall | 1 (3.1%) | 3 (2.8%) | 2 (11%) | 1 (5%) |
| Coronary sinus | 1 (3.1%) | 2 (1.9%) | 0 | 0 |
| Left atrial appendage and anterior wall | 0 | 1 (0.9%) | 0 | 0 |
| Right atrial crista terminalis | 0 | 1 (0.9%) | 0 | 0 |
Voltage mapping revealed that the mean peak-to-peak bipolar voltage in the left atrium during SR was lower in patients with high hs-CRP compared to those with low hs-CRP (p = 0.045; Table 2 ). In contrast, mean peak-to-peak bipolar voltage in the right atrium during SR was similar between patients with high and low hs-CRP levels (p = 0.1). Patients with nonparoxysmal AF had a trend toward lower mean LA and right atrial bipolar voltages in each patients group of high and low hs-CRP.
Spectral analysis during SR revealed that the mean dominant frequency value in the left atrium during SR was higher in patients with a high hs-CRP level, indicating that there were more regions of high-frequency AF nests in the left atrium in these patients ( Table 2 ). Regional analysis based on the 3-dimensional geometry revealed that patients with high hs-CRP had more high-frequency sites in the left atrium (71% vs 37%, p = 0.03; Figure 1 ). In the low hs-CRP group, patients with nonparoxysmal AF also had more high-frequency sites in the left atrium compared to those with paroxysmal AF. In contrast, the incidence of high-frequency AF nest sites in the right atrium was similar between patients with high and low hs-CRP levels ( Table 2 ).
