Combined Algorithm Using a Poor Increase in Inferior P-Wave Amplitude During Sympathetic Stimulation and Sinus Node Recovery Time for the Diagnosis of Sick Sinus Syndrome

Abstract

Background: This study sought to evaluate whether a poor increase in inferior P-wave amplitude during sympathetic stimulation might be a helpful diagnostic tool for sick sinus syndrome (SSS).

Methods and Results: Three-dimensional electroanatomic mapping of the right atrium, inferior P-wave amplitude and conventional corrected sinus node recovery time (CSNRT) were compared in 112 consecutive atrial fibrillation (AF) patients with (n=21) and without SSS (n=91). The significant cranial shift of earliest activation site (EAS) (the distance from the superior vena cava to the EAS: 11.1 vs. 5.9 mm, P<0.001) and the increases of inferior P-wave amplitudes during isoproterenol infusion (all P<0.001) were observed in patients without SSS. However, cranial shift of EAS (16.5 vs. 14.2 mm, P=0.375) and P-wave amplitude increases were not observed in those with SSS. Although CSNRT >550 ms showed a sensitivity of 50% and specificity of 84% for diagnosing SSS, poor increases of P-waves amplitude in lead aVF (<0.1 mV) during isoproterenol infusion showed an improved sensitivity of 71% and specificity of 89%. Finally, the combined algorithm using CSNRT >550 ms and poor increase of P-waves amplitude in lead aVF showed more improved diagnostic accuracy (sensitivity 89%, specificity 75%).

Conclusions: A combined algorithm using inferior P-wave amplitude showed improved performance for the diagnosis of SSS compared with CSNRT >550 ms alone. (Circ J 2015; 79: 2148–2156)

Sinoatrial node (SAN) disease results from impaired impulse generation in the SAN or impaired conduction of the impulse to the atrial tissue.¹ Sick sinus syndrome (SSS) can be diagnosed based on clinical criteria. There are several methods of assessing the degree of SAN dysfunction^2,3 and of them, sinus node recovery time (SNRT) and corrected SNRT (CSNRT) are commonly used,^4,5 usually accompanied by an invasive electrophysiological study (EPS).⁶ However, the accuracy of CSNRT for diagnosing SSS is not ideal,⁷ and an improved diagnostic test for SSS is needed.

The Ca²⁺ clock of the superior SAN is primarily responsible for rate acceleration during sympathetic stimulation.⁸ Unresponsiveness of the Ca²⁺ clock in the superior SAN to sympathetic stimulation is a characteristic finding in dogs with atrial fibrillation (AF) and sinus node dysfunction.⁹ Using 3D electroanatomical mapping techniques, we found that the superior SAN serves as the earliest activation site (EAS) during sympathetic stimulation in normal patients and in most patients with AF without symptomatic bradycardia. Unresponsiveness of the superior SAN to sympathetic stimulation was a characteristic finding in patients with AF and symptomatic bradycardia.⁷

ECG is noninvasive, and a previous study suggested that the amplitude of the filtered P-wave on signal-averaged ECGs may be useful for the detection of SSS.¹⁰ Recently, we found that the P-wave amplitude of the inferior leads correlated well with EAS location.¹¹ The aim of this study was to evaluate whether a combined algorithm using poor increase in inferior P-wave amplitude during sympathetic stimulation and conventional CSNRT might be a helpful diagnostic tool for SSS.

Methods

Study Subjects

Written informed consent was given by all patients. The study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki, and was approved by the Clinical Research and Ethics Committee of the Yonsei University Hospital, Seoul, South Korea. This study enrolled consecutive patients with AF from January 2010 to August 2012 in the Yonsei Ablation AF Cohort. All patients underwent radiofrequency catheter ablation (RFCA) for symptomatic drug-refractory AF. We performed 3D endocardial mapping of the right atrium (RA) at baseline and during isoproterenol infusion in all patients. All antiarrhythmic medications (including β-blockers and calcium blockers) were suspended for >5 half-lives before the study. We excluded patients from the study if they had recent (≤3 months) myocardial infarction, ongoing myocardial ischemia, heart failure, valvular heart disease or had used amiodarone. In total, 112 patients were included in this study. Among them, 21 subjects with either marked sinus bradycardia (<40 beats) (n=5), prolonged sinus pause (>3.0 s) (n=4) or tachycardia-bradycardia syndrome (n=12) were diagnosed as SSS¹² and they had symptoms such as cardiac syncope, presyncope or dizziness. One patient in the AF patient with SSS group had a pacemaker that had been implanted 3 years before the EPS.

EPS

The EPS was performed while the patient was in the postabsorptive state. Patients were sedated with midazolam and fentanyl. Multipolar catheters were positioned as follows: (1) 20-pole catheter with 2–5–2-mm interelectrode spacing in the coronary sinus with the proximal 10 electrodes positioned at the lateral RA; (2) 10-pole catheter with 2–7–2-mm interelectrode spacing along the lateral RA. SAN function was evaluated as follows: (1) baseline sinus cycle length (CL) was determined over 10 consecutive sinus cycles; and (2) corrected SAN recovery time (CSNRT) was determined after a 30-s drive train at CLs of 600, 500 and 400 ms, correcting for the baseline CL. CSNRT were determined 3 times and the average value was used for subsequent analyses. A CSNRT ≤550 ms was considered normal.

Atrial effective refractory periods (ERPs) were evaluated with S₂ strength at twice the diastolic threshold current (for a pacing threshold <2 mA) after 8 S₁-paced beats at CLs of 600, 500 and 400 ms. An incremental technique was used starting with an S₁–S₂ coupling interval of 150 ms. The coupling interval was then increased in 5-ms increments until S₂ captured the atria. The ERP was defined as the longest coupling interval that failed to capture the atria. ERP was measured from the distal and proximal coronary sinus and from the low and high lateral RA for a total of 3 times. Averaged values were used for analyses.

LA conduction time was measured along the coronary sinus by pacing from the distal bipole (1–2) of the coronary sinus catheter and measuring the activation time at the proximal bipole (9–10). RA conduction time was measured along the lateral RA by pacing from the distal bipole (1–2) of the lateral RA catheter and measuring the time required to activate the proximal bipoles (9–10). At both sites, conduction time was measured at pacing CLs of 600, 500 and 400 ms during stable capture.

Measurement of P-Wave Amplitude

Surface 12-lead ECG P-wave morphology was assessed as previously described.¹³ We measured inferior P-wave amplitude (leads II, III and aVF) before and during isoproterenol infusion at the end of RFCA by 2 physicians who were blinded to all clinical data. Isoproterenol was infused at a dosage of 10 μg/min. In patients who presented to the laboratory with AF, we measured P-wave amplitude after sinus conversion during ablation or after electrical cardioversion. The height or depth of the P-wave was measured from the P-wave peak or nadir to the isoelectric line (TP interval) using the electronic caliper of Pruka (GE Healthcare, Milwaukee, WI, USA). It was filtered from 0.05 to 80 Hz at a sweep speed of 50 mm/s. We selected a P-wave without noise during continuous monitoring and derived the average value from at least 5 P-waves. P-waves were included for analysis only if an isoelectric interval was present and there was no fusion with the preceding QRS or T-wave. There was no limitation to measuring P-waves at heart rates within 150 beats/min. The inter- and intra-observer variabilities for P-wave amplitude were 0.89 and 0.98, respectively, which was evaluated by the inter- and intra-class correlation coefficients.¹⁴

Electroanatomical Mapping

3D mapping of the RA was performed after circumferential pulmonary vein isolation and/or linear ablation in the LA. No ablation was performed in the RA or near the SAN. Electroanatomical maps of the RA were created before and during isoproterenol infusion using Ensite NavX, which records the 12-lead ECG and bipolar electrograms filtered at 30–200 Hz from the mapping catheter and reference electrogram. Fluoroscopy, RA angiography, computed tomography and Ensite NavX merging were used to facilitate mapping of anatomic structures, particularly the crista terminalis and superior vena cava (SVC)-RA junction, and for ensuring endocardial contact when individual points were acquired. High-density mapping was performed along the crista terminalis, septal RA and in areas of low voltage. Points were acquired if the stability criteria in space (≤6 mm) and local activation time (≤5 ms) were met.¹⁵ Editing of points was performed offline by a single technician blinded to the clinical data. Local activation was manually annotated to the beginning of the first rapid deflection from the isoelectric line on bipolar electrograms.

The linear distance from the SVC-RA junction to the most cranial EAS was used as a quantitative measure of EAS location.⁷ Mapping during isoproterenol infusion at a dosage of 10 μg/min was performed at stable heart rate approximately 5 min after the commencement of isoproterenol infusion.

Statistical Analysis

Continuous variables are reported as median values (25–75 percentile range) and compared using the Mann-Whitney test. Categorical variables are reported as count (percentage) and compared using the Chi-square test or Fisher’s exact test. The correlation between the distance from the SVC-RA junction to the most cranial EAS and P-wave amplitude was evaluated with Pearson’s correlation. The means of the paired P-wave amplitudes (baseline and isoproterenol infusion) were evaluated with the Wilcoxon test. Receiver-operating characteristic (ROC) curves were used to compare the performance of CSNRT and inferior P-wave amplitude and pairwise comparison was done using the difference between AUC curves. ANCOVA was used to adjust the means of inferior P-wave amplitudes for heart rate and propensity-score analysis with stabilized inverse probability weight¹⁶ was used to adjust for the unequal size of the 2 groups. SAS software, version 9.2 (SAS Institute Inc, Cary, NC, USA) was used for all analyses. P≤0.05 was considered statistically significant.

Results

Patients Characteristics

Table 1 details the clinical and electrophysiological characteristics of the 2 groups. There were no differences between them in age, percentage of paroxysmal AF, underlying diseases, CHADS2 score (≥2), or echocardiographic parameters such as LA dimension and left ventricular ejection fraction. However, there were more male patients in the AF without SSS group (P=0.03).

Table 1. Clinical and Electrophysiological Characteristics of Study Patients With AF

	AF without SSS (n=91)	AF with SSS (n=21)	P value
Age, years	58 (45–64)	58 (54–65)	0.28
Male, n (%)	77 (85)	13 (62)	0.03
Type of AF, n (%)			0.92
Paroxysmal	64 (70)	15 (71)
Persistent	27 (30)	6 (29)
Underlying disease, n (%)
Heart failure	4 (4)	1 (5)	>0.99
Hypertension	33 (36)	8 (38)	0.88
Diabetes mellitus	1 (1)	1 (5)	0.34
Stroke or TIA	4 (4)	1 (5)	>0.99
CHADS2 score ≥2, n (%)	6 (7)	1 (5)	>0.99
LA dimension, mm	42 (37–44)	41 (37–45)	0.98
LVEF, %	63 (59–67)	62 (58–68)	0.96
CSNRT, ms	345 (239–459)	511 (286–842)	0.04
ERP, ms
High RA	218 (200–235)	234 (224–249)	<0.01
Low RA	218 (198–233)	235 (216–258)	0.03
Proximal CS	230 (213–256)	247 (231–266)	0.07
Distal CS	230 (213–252)	248 (227–263)	0.06
Conduction time, ms
RA	54.3 (42.7–67.9)	67.9 (53.4–82.7)	0.01
LA	28.8 (24.3–33.7)	30.6 (28.8–37.0)	0.03
Procedure time, min	168 (145–194)	164 (138–203)	0.72
Ablation time, s	4,205 (3,120–4,976)	3,511 (3,081–5,523)	0.70

Data are presented as median (25–75% interquartile range). AF, atrial fibrillation; CS, coronary sinus; CSNRT, corrected sinoatrial node recovery time; ERP, effective refractory period; LA, left atrium; LVEF, left ventricular ejection fraction; RA, right atrium; SSS, sick sinus syndrome; TIA, transient ischemic attack.

Comparison of Electrophysiological Characteristics

In the AF patients with SSS, the median CSNRT and ERP of the high and low RA were longer (Table 1; P=0.04, <0.01 and 0.03, respectively) and the conduction times of the RA and LA were more delayed (P=0.01 and 0.03, respectively) than in AF patients without SSS. The distribution of CSNRT is presented in Figure S1.

EAS at Baseline and During Sympathetic Stimulation

Using 3D mapping, we analyzed a mean of 237±30 and 210±25 points per patient during sinus rhythm and isoproterenol infusion, respectively. There were no significant differences in the number of points analyzed among the groups at baseline (P=0.29) and isoproterenol infusion (P=0.14). Figures 1 and 2 show typical examples of EAS and P-wave amplitude responses to isoproterenol infusion in AF patients without and with SSS.

Figure 1.

Response of the EAS of the SAN and inferior P-wave amplitude to isoproterenol infusion in AF patients without sick sinus syndrome. (A) RA activation map at baseline (a) and during isoproterenol infusion (b). Note the cranial shift of the EAS (arrows) during isoproterenol infusion. (B) Twelve-lead ECGs at baseline (a) and during isoproterenol infusion (b). Note the increase in the P-wave amplitude (red arrows) in leads II, III and aVF. AF, atrial fibrillation; EAS, earliest activation site; RA, right atrium; SAN, sinoatrial node.

Figure 2.

Response of the EAS of the SAN and inferior P-wave amplitude to isoproterenol infusion in AF patients with sick sinus syndrome. (A) RA activation map at baseline (a) and during isoproterenol infusion (b). Note the caudal shift of the EAS (arrows) during isoproterenol infusion. (B) Twelve-lead ECGs at baseline (a) and during isoproterenol infusion (b). Note no increase in P-wave amplitude (red arrows) in leads II, III and aVF. AF, atrial fibrillation; EAS, earliest activation site; RA, right atrium; SAN, sinoatrial node.

The EAS values at baseline and during isoproterenol infusion are presented in Table 2. Baseline median heart rates were 64 beats/min (95% confidence interval (CI) 53–80 beats/min) and 81 beats/min (95% CI 71–90 beats/min in AF patients with and without SSS, respectively (P=0.002). The EAS was located significantly inferiorly in AF patients with SSS (16.5 mm from SVC-RA junction, 95% CI 10.0–31.4) compared with those without SSS (11.1 mm from SVC-RA junction, 95% CI 2.7–22.8, P=0.025).

Table 2. EAS and Inferior P-Wave Amplitude at Baseline and During Isoproterenol Infusion in Study Patients With AF

	AF without SSS (n=91)	AF with SSS (n=21)	P value
Baseline
Heart rate, beats/min	81 (71–90)	64 (53–80)	0.002
EAS location, mm*	11.1 (2.7–22.8)	16.5 (10.0–31.4)	0.025
P-wave amplitude, mV
Lead II	0.14 (0.11–0.17)	0.11 (0.07–0.13)	0.001
Lead III	0.11 (0.07–0.12)	0.06 (0.04–0.08)	<0.001
Lead aVF	0.12 (0.09–0.14)	0.07 (0.06–0.10)	<0.001
During isoproterenol infusion
Heart rate, beats/min	128 (117–140)	109 (100–124)	0.002
EAS location, mm*	5.9 (−0.7 to 12.1)	14.2 (5.2–30.5)	0.009
P-wave amplitude, mV
Lead II	0.18 (0.15–0.22)	0.11 (0.09–0.12)	<0.001
Lead III	0.13 (0.10–0.16)	0.06 (0.05–0.09)	<0.001
Lead aVF	0.15 (0.12–0.18)	0.08 (0.07–0.12)	<0.001

Data are presented as median (25–75% interquartile range). *EAS location refers to the distance from the superior vena cava and right atrial junction to the most cranial EAS. EAS, earliest activation site. Other abbreviations as in Table 1.

During isoproterenol infusion, the median heart rates were significantly lower in AF patients with SSS (109 beats/min, 95% CI 100–124 beats/min) than in those without SSS (128 beats/min, 95% CI 117–140 beats/min, P=0.002). The EAS was located significantly inferiorly in AF patients with SSS (14.2 mm from SVC-RA junction, 95% CI 5.2–30.5) than in those without SSS (5.9 mm from SVC–RA junction, 95% CI −0.7–12.1, P=0.009). The significant cranial shift of the EAS during isoproterenol infusion was noted in AF patients without SSS (Figure S2, P<0.001), but not in those with SSS.

Poor Increase in Inferior P-Wave Amplitude During Sympathetic Stimulation in SSS

In Table 2, compared with AF patients without SSS, those with SSS had significantly lower inferior P-wave amplitudes at baseline (all P≤0.001) and during isoproterenol infusion (all P<0.001). In subgroup analysis according to initially presenting rhythm (sinus rhythm or AF), these findings were consistently observed (Table S1).

Inferior P-wave amplitudes were significantly increased in AF patients without SSS during isoproterenol infusion (leads II, III and aVF, all P<0.001) (Figure 3A). However, there were no significant changes in the inferior P-wave amplitude in AF patients with SSS (leads II, III and aVF, P=0.605, 0.095 and 0.055, respectively).

Figure 3.

(A) Changes in inferior P-wave amplitude during isoproterenol infusion in AF patients with and without sick sinus syndrome. The dashed line shows P-wave amplitude of 0.1 mV. (B) Correlation between the distance from the SVC-RA junction to the EAS and inferior P-wave amplitude. Unfilled and filled circles represent P-wave amplitude at baseline and during isoproterenol infusion, respectively. Red and blue circles represent patients with and without sick sinus syndrome, respectively. AF, atrial fibrillation; EAS, earliest activation site; SVC-RA junction, junction of superior vena cava and right atrium.

Correlation Between the Distance From the SVC–RA Junction to the EAS and Inferior P-Wave Amplitude

Distance from the SVC–RA junction to the EAS had a negative correlation with inferior P-wave amplitude at baseline and during isoproterenol infusion (lead II, r=−0.38, P<0.001; lead III, r=−0.41, P<0.001; and lead aVF, r=−0.44, P<0.001, respectively) (Figure 3B).

Comparison of Inferior P-Wave Amplitudes

Table 3 shows the adjusted means of inferior P-wave amplitudes for heart rate in the 2 groups. Using ANCOVA analysis, P-wave amplitude in lead aVF was significantly lower in AF patients with SSS (0.089 mV, 95% CI, 0.076–0.102 mV) than in those without SSS (0.133 mV, 95% CI, 0.127–0.140 mV, P<0.001) after adjustment for heart rate. In the propensity score-adjusted model, AF patients with SSS had significantly lower P-wave amplitude in lead aVF (0.090 mV, 95% CI, 0.078–0.102 mV) than those without SSS (0.134 mV, 95% CI, 0.128–0.140 mV, P<0.001). Subgroup analysis according to initially presenting rhythm also showed similar results (Table S2).

Table 3. Adjusted Mean of Inferior P-Wave Amplitude in Study Patients With AF

P-wave amplitude, mV	AF without SSS (n=91)	AF with SSS (n=21)	P value
Crude model†
Lead II	0.163 (0.157–0.170)	0.115 (0.101–0.129)	<0.001
Lead III	0.112 (0.106–0.118)	0.073 (0.059–0.086)	<0.001
Lead aVF	0.133 (0.127–0.140)	0.089 (0.076–0.102)	<0.001
Propensity score-adjusted model using stabilized IPW
Lead II	0.163 (0.157–0.170)	0.120 (0.107–0.132)	<0.001
Lead III	0.113 (0.106–0.119)	0.074 (0.061–0.086)	<0.001
Lead aVF	0.134 (0.128–0.140)	0.090 (0.078–0.102)	<0.001

^†Mean (95% CI) was adjusted for heart rate (101 beats/min) using ANCOVA only. CI, confidence interval; IPW, inverse probability weight. Other abbreviations as in Table 1.

Inferior P-Wave Amplitude for the Diagnosis of SSS

Figure 4 shows the ROC curves for the diagnosis of SSS using inferior P-wave amplitudes and CSNRT. The areas under the curves (AUCs) of the inferior P-wave amplitudes tended to be superior to those for CSNRT. However, there were no statistically significant differences at baseline. During isoproterenol infusion, the AUCs of the inferior P-wave amplitude were further increased, which resulted in significantly enhanced diagnostic performance compared with CSNRT (difference between areas=0.239, 95% CI 0.082–0.397, P=0.003 in lead II; difference between areas=0.181, 95% CI 0.023–0.339, P=0.025 in lead III; difference between areas=0.225, 95% CI 0.065–0.384, P=0.006 in lead aVF).

Figure 4.

Receiver-operating characteristic curves of CSNRT and inferior P-wave amplitudes for diagnosis of sick sinus syndrome at baseline (A) and during isoproterenol infusion (B). AUC, area under the curve; CI, confidence interval; CSNRT, corrected sinoatrial node recovery time.

Table 4 summarizes the sensitivities and specificities using conventional CSNRT >550 ms and inferior P-wave amplitudes <0.1 mV. CSNRT >550 ms had a sensitivity of 50.0% and specificity of 84.0% for detection of SSS. At baseline, using a cutoff value <0.1 mV, the noninvasive test of P-wave amplitudes in leads II, III and aVF had a sensitivity of 42.9, 90.5 and 76.2% and specificity of 90.1, 52.8 and 69.2%, respectively. However, sympathetic stimulation during isoproterenol infusion increased the specificity to 96.7%, 78.0% and 89.0%, respectively. Using the combined algorithm (Figure S3), the diagnostic accuracy for SSS was improved (sensitivity 88.9%, specificity 75.3% in lead aVF).

Table 4. Sensitivities and Specificities for the Diagnosis of SSS in Patients With AF

	Sensitivity, %	Specificity, %	PPV, %	NPV, %
CSNRT >550 ms	50.0	84.0	40.9	88.3
P-wave amplitude
Baseline
<0.1 mV in lead II	42.9	90.1	50.0	87.2
<0.1 mV in lead III	90.5	52.8	30.7	96.0
<0.1 mV in lead aVF	76.2	69.2	36.4	92.7
During isoproterenol infusion
<0.1 mV in lead II	33.3	96.7	70.0	86.3
<0.1 mV in lead III	81.0	78.0	46.0	94.7
<0.1 mV in lead aVF	71.4	89.0	60.0	93.1
Combined algorithm‡
<0.1 mV in lead II	55.6	81.5	40.0	89.2
<0.1 mV in lead III	94.4	63.0	36.2	98.1
<0.1 mV in lead aVF	88.9	75.3	44.4	96.8

^‡Figure S3 shows combined algorithm using CSNRT >550 ms and poor increase in inferior P-wave amplitude. NPV, negative predictive value; PPV, positive predictive value. Other abbreviations as in Table 1.

Discussion

The main finding of this study was that the EAS was located more caudally in AF patients with SSS than in those without SSS at baseline and during isoproterenol infusion. Second, the P-wave amplitude in the inferior leads negatively correlated with a caudal shift of EAS. Interestingly, the poor increase in the P-waves in the inferior leads during sympathetic stimulation was more accurate in diagnosing SSS than conventional CSNRT. Our results suggest that analysis of P-wave morphology with sympathetic stimulation may have additive diagnostic value for determining the function of the SAN.

Poor Cranial Shift of the EAS in the Setting of SSS

We documented an upward shift of the EAS during isoproterenol infusion in a canine RA perfusion model with optical mapping, and in humans with 3D endocardial mapping.^7,8 A superior shift of the EAS is a normal response to isoproterenol infusion in patients without AF and in the majority of AF patients without symptomatic bradycardia. However, unresponsiveness of the superior EAS during sympathetic stimulation is a common finding among patients with both AF and symptomatic bradycardia. This finding has also been reported in a canine model⁹ and these findings are consistent with previous studies that showed Ca²⁺ clock malfunction may underlie abnormal physiological responses to sympathetic stimulation.^9,17,18

Poor Increase of Inferior P-Waves in SSS

Recently, we found that an abnormal response of the inferior P-wave amplitude to sympathetic stimulation may be a valuable test of SAN dysfunction after amiodarone use.¹¹ In this present study, we found that the P-wave amplitude in the inferior leads correlates negatively with a caudal shift of the EAS in patients with SSS.

We also analyzed the inferior P-wave amplitudes using a cutoff of 0.1 mV for convenient clinical application. An advantage of this method is that it does not require invasive EPS, thus assisting in the noninvasive diagnosis of a patient with possible SSS.

Diagnosis of SAN Dysfunction

SAN dysfunction is a major reason for pacemaker implantation. Generally, SSS is diagnosed by symptoms and the documentation of sinus pause and sinus bradycardia during ECG or Holter monitoring. The diagnosis of SSS is not easy if the patient does not have an ECG abnormality during the recording. Therefore, when more evidence is needed, invasive EPS for SANRT and CSNRT measurements may be required. However, the CSNRT has limited diagnostic accuracy.⁷ The results from the present study also showed that its sensitivity as just 50% and the CSNRT had poor predictive value in determining sinus node disease (Table 4, Figure S1).

Yamada et al¹⁰ suggest that low P-waves on signal-averaged ECGs may be characteristic of SSS. They explain that low amplitude potentials early in signal-averaged P-waves might reflect conduction abnormalities in the perinodal atrial muscle. The pathologic findings of SSS are degeneration and fibrosis not only in the SAN, but also the atrial muscle, especially the perinodal portion.^19,20 It was demonstrated that SAN dysfunction is associated with diffuse atrial remodeling characterized by conduction abnormalities and increased RA refractoriness.¹⁵ These findings could be consistent with our results, in which there were significant differences in the RA ERP and conduction time between AF patients with and without SSS. RA remodeling including myocardial damage and concomitant fibrosis are expected in AF patients with SSS, because the ERP and conduction time in the RA are longer in these patients than in those without SSS. Therefore, atrial electroanatomical remodeling may result in the low amplitudes observed in the inferior leads and in the unresponsiveness to sympathetic stimulation. However, the evaluation of electroanatomical remodeling caused by SSS is difficult without electrophysiological testing.

Bhandari et al studied the value of phamacological tests to predict SSS, in which a significantly higher increase in heart rate following isoprenaline adminstration was observed in patients without sinus node dysfunction.²¹ Our study also showed a significant difference in heart rate between patients with and without SSS at baseline (Table 2; 64 beats/min, (53–80) vs. 81 beats/min (71–90), P=0.002) and during isoproterenol (109 beats/min (100–124) vs. 128 beats/min (117–140), P=0.002). Therefore, a poor increase in heart rate could be a characteristic of patients with SSS. However, after adjustment for heart rate, the P-wave amplitudes in leads II, III and aVF were significantly lower in patients with SSS (Table 3).

Study Limitations

The strength of this study is it being the first trial evaluating the role of inferior P-wave amplitude to determine sinus node dysfunction using isoproterenol infusion. Taking into account that CSNRT is invasive, this method has easy applicability to daily practice. However, there are several limitations. Our study was at a single center and we used retrospective analysis, so a selection bias may exist. Further prospective studies will be needed to confirm these findings, which should be evaluated in clinically relevant subjects (ie, patients with unexplained syncope). We included only patients with AF, who may have some degree of sinus node dysfunction.²² Therefore, we could not extrapolate this result to all SSS patient without AF. Further prospective studies are needed to apply this finding to these patients. Electrophysiological studies were performed under sedation with midazolam and fentanyl in the same session as PV isolation. Therefore, it is possible that some patients were sedated more deeply than others. However, there was no difference in the procedure time between the 2 groups in this study. PV isolation can include inadvertent GP ablation, which is expected to affect autonomic tone. Therefore, sinus function might be influenced by PV isolation. However, RA function was measured after RF ablation in both groups. Moreover, the patients with sinus dysfunction were diagnosed by ECG and Holter data before ablation. The responsiveness to sympathetic stimulation using isoproterenol infusion may have limitations in patients who are intolerant to β-agonists. Additionally, because we acquired activation maps using point-by-point contact mapping, a spatiotemporally homogeneous distribution is not reflected.

Conclusions

The superior SAN serves as the EAS during sympathetic stimulation in AF patients without symptomatic bradycardia. The P-wave amplitudes in the inferior leads negatively correlated with a caudal shift of the EAS, and were significantly smaller in AF patients with SSS than in those without SSS. Interestingly, a poor increase in the P-waves in the inferior leads during sympathetic stimulation was more accurate in diagnosing SSS than conventional CSNRT >550 ms. Moreover, the combined algorithm using a poor increase in inferior P-wave amplitude during sympathetic stimulation and conventional CSNRT showed improved performance for the diagnosis of SSS compared with CSNRT >550 ms alone.

Acknowledgments

This study was supported in part by research grants from the Basic Science Research Program through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology (NRF-2010-0021993, NRF-2012R1A2A2A02045367), and by a grant from the Korean Healthcare Technology R&D project funded by the Ministry of Health & Welfare (HI12C1552).

Disclosures

None of the authors has any conflicts of interest to declare.

Authors’ Contribution

J.-K.P. and J.P. participated in the study design. J.-K.P. performed the analysis. J.-K.P., J.P., J.-S.U. and B.J. drafted the manuscript. B.J., M.-H.L. and H.-N.P. critically revised the manuscript. All authors participated in the revision and final approval of the manuscript. B.J. acts as the guarantor of the manuscript.

Competing Interests

We certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.

Supplementary Files

Supplementary File 1

Figure S1. Dot plots of the CSNRT from AF patients with and without SSS at PCL of 600 ms, showing large overlaps.

Figure S2. The distance from the SVC-RA junction to the EAS at baseline and during isoproterenol infusion.

Figure S3. Combined algorithm using inferior P-wave amplitude during sympathetic stimulation and corrected sinus node recovery time for the diagnosis of sick sinus syndrome.

Table S1. Subgroup analysis of the EAS and inferior P-wave amplitude at baseline and during isoproterenol infusion according to initial presenting rhythm of SR

Table S2. Adjusted mean of inferior P-wave amplitude according to initial presenting rhythm of SR or AF

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-15-0561

References

• 1.    Ferrer  MI. The sick sinus syndrome in atrial disease. JAMA 1968; 206: 645–646.
• 2.    Mandel  WJ,  Hayakawa  H,  Allen  HN,  Danzig  R,  Kermaier  AI. Assessment of sinus node function in patients with the sick sinus syndrome. Circulation 1972; 46: 761–769.
• 3.    Burnett  D,  Abi-Samra  F,  Vacek  JL. Use of intravenous adenosine as a noninvasive diagnostic test for sick sinus syndrome. Am Heart J 1999; 137: 435–438.
• 4.    Benditt  D,  Sakaguchi  S,  Goldstein  M,  Lurie  K,  Gornick  C,  Adler  S. Sinus node dysfunction: Pathophysiology, clinical features, evaluation, and treatment. Cardiac electrophysiology: from cell to bedside 1995: 1215–1247.
• 5.   JCS Joint Working Group. Guidelines for clinical cardiac electrophysiologic studies (JCS 2011): Digest version. Circ J 2013; 77: 497–518.
• 6.   JCS Joint Working Group. Guidelines for non-pharmacotherapy of cardiac arrhythmias (JCS 2011): Digest version. Circ J 2013; 77: 249–274.
• 7.    Joung  B,  Hwang  HJ,  Pak  HN,  Lee  MH,  Shen  C,  Lin  SF, et al. Abnormal response of superior sinoatrial node to sympathetic stimulation is a characteristic finding in patients with atrial fibrillation and symptomatic bradycardia. Circ Arrhythm Electrophysiol 2011; 4: 799–807.
• 8.    Joung  B,  Tang  L,  Maruyama  M,  Han  S,  Chen  Z,  Stucky  M, et al. Intracellular calcium dynamics and acceleration of sinus rhythm by beta-adrenergic stimulation. Circulation 2009; 119: 788–796.
• 9.    Joung  B,  Lin  SF,  Chen  Z,  Antoun  PS,  Maruyama  M,  Han  S, et al. Mechanisms of sinoatrial node dysfunction in a canine model of pacing-induced atrial fibrillation. Heart Rhythm 2010; 7: 88–95.
• 10.    Yamada  T,  Fukunami  M,  Kumagai  K,  Abe  Y,  Kim  J,  Sanada  S, et al. Detection of patients with sick sinus syndrome by use of low amplitude potentials early in filtered P wave. J Am Coll Cardiol 1996; 28: 738–744.
• 11.    Mun  HS,  Shen  C,  Pak  HN,  Lee  MH,  Lin  SF,  Chen  PS, et al. Chronic amiodarone therapy impairs the function of the superior sinoatrial node in patients with atrial fibrillation. Circ J 2013; 77: 2255–2263.
• 12.    Zipes  DP,  Jalife  J. Cardiac electrophysiology: From cell to bedside. New York: Saunders, 2009.
• 13.    Tang  CW,  Scheinman  MM,  Van Hare  GF,  Epstein  LM,  Fitzpatrick  AP,  Lee  RJ, et al. Use of P wave configuration during atrial tachycardia to predict site of origin. J Am Coll Cardiol 1995; 26: 1315–1324.
• 14.    Posner  KL,  Sampson  PD,  Caplan  RA,  Ward  RJ,  Cheney  FW. Measuring interrater reliability among multiple raters: An example of methods for nominal data. Stat Med 1990; 9: 1103–1115.
• 15.    Sanders  P,  Kistler  PM,  Morton  JB,  Spence  SJ,  Kalman  JM. Remodeling of sinus node function in patients with congestive heart failure: Reduction in sinus node reserve. Circulation 2004; 110: 897–903.
• 16.    Cole  SR,  Hernan  MA. Constructing inverse probability weights for marginal structural models. Am J Epidemiol 2008; 168: 656–664.
• 17.    Joung  B,  Ogawa  M,  Lin  SF,  Chen  PS. The calcium and voltage clocks in sinoatrial node automaticity. Korean Circ J 2009; 39: 217–222.
• 18.    Shinohara  T,  Joung  B,  Kim  D,  Maruyama  M,  Luk  HN,  Chen  PS, et al. Induction of atrial ectopic beats with calcium release inhibition: Local hierarchy of automaticity in the right atrium. Heart Rhythm 2010; 7: 110–116.
• 19.    Bharati  S,  Nordenberg  A,  Bauernfiend  R,  Varghese  JP,  Carvalho  AG,  Rosen  K, et al. The anatomic substrate for the sick sinus syndrome in adolescence. Am J Cardiol 1980; 46: 163–172.
• 20.    Sugiura  M,  Ohkawa  S. A clinicopathologic study on sick sinus syndrome with histological approach to the sinoatrial node. Jpn Circ J 1980; 44: 497–504.
• 21.    Bhandari  S,  Talwar  KK,  Kaul  U,  Bhatia  ML. Value of physical and pharmacological tests in predicting intrinsic and extrinsic sick sinus syndrome. Int J Cardiol 1986; 12: 203–212.
• 22.    Elvan  A,  Wylie  K,  Zipes  DP. Pacing-induced chronic atrial fibrillation impairs sinus node function in dogs: Electrophysiological remodeling. Circulation 1996; 94: 2953–2960.