Subjects and study protocol
This is a cross-sectional study with 29 age-matched control subjects and 32 AF patients (paroxysmal AF n=18 and persistent AF n=14) who came to Fukushima Medical University Hospital between June 2020 and October 2020. AF was diagnosed and classified as paroxysmal or persistent according to the duration of AF; paroxysmal, ≤ 7 days; persistent, > 7 days.33 The study subjects underwent echocardiography, carotid artery ultrasonography, laboratory testing, psychological testing and NIRS. The verbal fluency task (VFT), which is commonly used with NIRS analysis, was used to test brain activity.34 35 36 The control subjects had no past history of AF, heart failure or structural cardiac abnormalities detected by echocardiography. The study protocol was approved by the ethical committee of Fukushima Medical University (#823), and the investigation conforms to the principles outlined in the Declaration of Helsinki. All subjects provided written informed consent to participate in the study. Patients with carotid artery stenosis, cerebral infarction, dementia, and those receiving treatment for schizophrenia, depression, or bipolar disorder were excluded. Regarding the psychological testing, the Center for Epidemiologic Studies Depression Scale (CES-D) was used to evaluate depressive symptoms, 21 and the Mini-Mental State Examination (MMSE) was used to evaluate cognitive function.22 We compared the findings of CES-D (depressive symptoms), MMSE (cognitive function) and NIRS (brain activities) between the AF patients and control subjects. Next, we also followed up persistent AF patients who kept SR at 3 months after CA and measured their brain activities using NIRS, CES-D and MMSE after CA to investigate the associations of changes in brain activities with changes in both CES-D and MMSE.
Echocardiography and carotid artery ultrasonography were performed blindly by experienced sonographers using standard techniques.37 38 The left ventricular ejection fraction was calculated using Simpson’s method in a four-chamber view.37 38 All measurements were performed using ultrasound systems (ACUSON Sequoia, Siemens Medical Solutions USA, Inc., Mountain View, CA, USA). Blood samples were obtained from all subjects at Fukushima Medical University Hospital. B-type natriuretic peptide levels were measured using a specific immunoradiometric assay (Shionoria BNP kit, Shionogi, Osaka, Japan).39
Measurement of NIRS
In the current study, oxy-Hb, deoxy-Hb, and total hemoglobin were measured with a 52-channel NIRS machine (Hitachi ETG4000, Hitachi Medical Corp., Tokyo, Japan) using two wavelengths of near-infrared light (695 and 830 nm). Artifacts must be eliminated by having the subject sit in a chair, relax, and move as little as possible. The 52 channels were attached symmetrically around the prefrontal cortex.40 41 The main measured channels were as follows: right temporal lobe (channels 1–3, 11–14, 22–24, 32–35, and 43–45), left temporal lobe (channels 8–10, 18–21, 29–31, 39–42, and 50–52) and frontal region (channels 25–28, 36–38, and 46–49). An increase in cerebral oxy-Hb concentration in response to the VFT is considered as a marker of brain activity.41 Most of the lower and forward channels were placed along the line connecting T3-Fpz-T4, based on the international 10-20 system. This system allows prediction of the measurement sites on the brain surface with relatively high accuracy. The sampling rate of oxy-Hb concentration data was 0.1 s. The obtained data were analyzed using the integral mode: the mean oxy-Hb concentrations over the 10-s p-task period and over the last 5 s of the post-task period were used as the baselines. Oxy-Hb concentrations were measured during a 10-s p-task period, a 60-s VFT period, and a 55-s post-task baseline period. Linear fitting was applied to the data between these two baselines. The average oxy-Hb concentration during the 60-s VFT period was used for the analysis. A previously reported algorithm 35 was used to automatically reject data with artifacts. Data are expressed as waveforms and topographic map. The intraclass correlation coefficient of the mean oxy-Hb concentration during the task period was calculated for the 52 channels. The single measure intraclass correlation coefficient was 0.5309, and the average measure intraclass correlation coefficient was 0.6936, which are both reliable, as previously reported.42 NIRS analyses were performed using MATLAB R2011 (Math Works Inc., Natick, MA, USA), and Prism 6.0 software (GraphPad Software, Inc., San Diego, CA, USA). Changes in hemoglobin oxygenation occur in the subject undergoing the VFT.
Activation task, VFT
The VFT was used as an activating task during NIRS analysis as previously reported.34 35 36 An outline of the VFT procedure is as follows.35 36 The subject is first prompted by a voice saying, “Start /a/, /i/, /u/, /e/, /o/” to repeat the utterance “/a/, /i/, /u/, /e/, /o/” for 30 s. The baseline activity, which is recorded and is used to remove the effect of vocalization on brain activity from the data. The subject is next prompted by a voice to vocalize as many words as possible that start with a certain letter. This is done in three 20-s sets. The subject is verbally prompted to vocalize words starting with a certain letter to increase the difficulty of the task. The task is scored by recording the number of words uttered in every 20 s. Finally, the subject is prompted by a voice saying, “Stop /a/, /i/, /u/, /e/, /o/” to stop the task and repeat “/a/, /i/, /u/, /e/, /o/” for 70 s.
Catheter ablation for AF and follow-up
Detailed methods of CA and follow-up have been reported.43 44 45 CA was performed for AF following the cessation of all antiarrhythmic drugs for over five half-lives. Circumferential pulmonary vein (PV) isolation was performed using a single-lasso technique while the patient was sedated with dexmedetomidine hydrochloride. A 7-Fr decapolar ring catheter (Lasso, Biosense Webster, Inc., Diamond Bar, CA, USA) and a 7.5-Fr irrigation catheter with a 3.5-mm distal electrode and real time contact force monitoring (ThermoCool SmartTouch SF, Biosense Webster, Inc., Diamond Bar, CA, USA) were inserted into the left atrium via the transseptal approach. After selective PV angiography, radiofrequency energy was delivered with a wide antral ablation line around the PVs, guided by a 3-dimensional mapping system (CARTO system, Biosense Webster, Inc., Diamond Bar, CA, USA). Radiofrequency power output was set at 45–50 W, and contact force was maintained between 10 and 15 g. After circumferential PV isolation, the elimination of the PV potential in each of the upper and lower PVs was confirmed. The endpoint of the PV isolation was defined as the creation of a bidirectional conduction block between the left atrium and PVs. Each patient was followed up at 1, 3, and 6 months after the CA, then every 3 months thereafter. At each visit, 12-lead electrocardiogram and 24-hour Holter monitoring were performed. Recurrence was defined as documentation of atrial tachycardia or AF lasting >30 s recorded in 12-lead ELECTROCARDIOGRAM or 24-hour Holter monitoring. Administration of antiarrhythmic drugs after PV isolation depended on the operator’s decision.
Categorical variables are expressed as numbers and percentages. The chi-square test was used for comparisons of categorical variables and followed by Fisher’s exact test when/if appropriate. Normality was confirmed using the Shapiro-Wilk test in each group. Parametric variables are presented as mean ± SD, and non-parametric variables (e.g. BNP, and NIRS findings) are presented as a median and interquartile range. Parametric variables were compared using Student’s t-test, whereas non-parametric variables were compared using the Mann-Whitney U test. To compare continuous variables among the paroxysmal AF, persistent AF and control subjects, Kruskal-Wallis test was used. We performed regression analysis to determine brain activity confounding factors. Correlations of changes in brain activities with changes in CES-D and MMSE before and after CA were assessed using Spearman’s correlation analysis. A P value of <0.05 was considered statistically significant for all comparisons. All analyses were performed using a statistical software package (SPSS ver. 24.0, IBM, Armonk, NY, USA).