Skip to main content
Springer logoLink to Springer
. 2026 Sep 7;69(9):3105–3116. doi: 10.1007/s10840-026-02422-7

An optimal output and electrode configuration for evaluating tissue transmurality - a preclinical study

Iwanari Kawamura 1,#, Aiko Fujimaki 2,3,#, Masateru Takigawa 1,4,✉,#, Shoko Kurosawa 2, Junji Yamaguchi 1,5, Ryosuke Kato 1,6, Masaki Honda 1, Hidehiro Iwakawa 6, Ryo Tateishi 1, Miho Negishi 1, Kentaro Goto 1, Kensuke Ihara 1, Takuro Nishimura 1, Kazuya Yamao 1, Susumu Tao 1, Sayaka Suzuki 7, Takehiro Iwanaga 8, Shinsuke Miyazaki 1,4, Masaru Kubota 3, Tetsuo Sasano 1
PMCID: PMC13769730  PMID: 42704576

Abstract

Introduction

Achieving transmural lesions is critical for durable outcomes in catheter ablation. While electrogram amplitude reduction and pacing threshold changes have been reported as useful parameters, the optimal electrode configuration and cutoff values remain unclear. This preclinical study evaluated the efficacy of conventional and microelectrode configurations in predicting transmurality using the QDOT MICRO catheter.

Methods

Radiofrequency ablation was performed in 14 swine, creating 131 atrial lesions. Pacing thresholds and electrogram amplitudes were measured before and after ablation using three configurations: conventional bipolar, microelectrode bipolar, and unipolar. Lesion transmurality was determined histologically. Receiver operating characteristic analyses were conducted to assess predictive accuracy.

Results

Among all configurations, the conventional bipolar (A1–A2) showed the highest predictive value. Post-ablation bipolar amplitudes were lower in transmural lesions (AUC 0.777, 0.86mV), with greater amplitude reduction (64.4% vs. 40.5%, AUC 0.682). Pacing thresholds increased significantly after ablation and were higher in transmural lesions; the A1–A2 configuration had the highest accuracy in predicting transmurality (AUC 0.880, cutoff 3.4 V). Microelectrodes demonstrated inconsistent patterns, and even when the best configurations were selected, they remained less accurate than conventional bipolar measurements. Furthermore, morphological changes, including disappearance of R-waves (bipolar) and S-waves (unipolar), showed limited predictive value (AUCs ≤ 0.77 and ≤ 0.56, respectively).

Conclusion

Conventional bipolar electrodes outperformed microelectrode configurations in assessing transmurality via electrogram amplitude and pacing threshold. Tip-to-ring bipolar pacing with the QDOT-MICRO catheter most effectively assesses atrial lesion transmurality, with 3.4 V as the optimal threshold.

Graphical abstract

graphic file with name 10840_2026_2422_Figa_HTML.webp

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s10840-026-02422-7.

Keywords: Catheter ablation, Preclinical, In vivo, Transmural lesion, Amplitude reduction

Introduction

The formation of transmural lesions is crucial for catheter ablation to create durable lesions. Conventionally, factors including the reduction of bipolar electrogram amplitude, the disappearance of the S-wave in unipolar electrograms, and changes in impedance have been reported as useful indicators [1–3]. While myocardial capture threshold before and after radiofrequency ablation (RFA) is also predicted to serve as an index [4,5], the optimal electrode configuration and output remain unclear in clinical practice.

Compared to the conventional bipolar, bipolar electrograms using microelectrodes have been suggested to be larger, sharper, and less attenuated, making them useful for visualizing local myocardial electrograms [6,7]. However, no studies have investigated the utility of microelectrodes in myocardial capture to evaluate the atrial tissue transmurality. We aimed to elucidate the optimal electrode configuration and pacing output for evaluating tissue transmurality using the QDOT-MICRO (Biosense Webster, Inc., Diamond Bar, CA) catheter.

Methods

Experimental animal model

Fourteen swine (weighing 45–65 kg) were sedated via intramuscular injection of ketamine hydrochloride (10 mg/kg) and xylazine (2 mg/kg). Following inhalational induction with isoflurane, endotracheal intubation was performed, and anesthesia was maintained with 2–3% isoflurane. The animals were mechanically ventilated using a respirator (Carestation/Carescape, GE HealthCare, Chicago, IL) with room air supplemented with oxygen. For drug and fluid administration, an intravenous sheath was placed in the internal jugular vein. Arterial blood gases were periodically measured, and ventilator settings were adjusted to maintain values within physiological ranges.

Radiofrequency ablation lesion formation

A steerable sheath (Agilis; St. Jude Medical, St. Paul, MN) was inserted via the right femoral vein, through which a QDOT MICRO catheter was advanced. A three-dimensional anatomical map was created using a multielectrode mapping catheter (OctaRay; Biosense Webster, Inc., Diamond Bar, CA) and the CARTO mapping system (Biosense Webster, Inc.). To produce a spectrum of lesion depths ranging from non-transmural to transmural, endocardial ablation was performed on the right atrium of beating swine hearts using variable radiofrequency settings (90 W for 1–4 s or 50 W for 8–12 s). The QDOT catheter enables lesion formation using not only conventional power settings but also high-power (90 W) ablation, both of which are applied in clinical practice. Regarding irrigation, the irrigation flow rate was set at 8 mL/min during high-power ablation (90 W). During ablation at 50 W, the irrigation flow rate was automatically adjusted between 4 and 15 mL/min according to the tissue temperature. During all non-ablation periods, including mapping and pacing, the irrigation flow rate was maintained at 2 mL/min. Accordingly, lesions were created using both power settings in this study, as lesions generated under different ablation conditions may have distinct characteristics that could influence the predictive value of pacing-based assessment. Lesions were created at inter-lesion distance intervals of at least 5 mm to allow for the evaluation of transmurality. Atrial capture thresholds were measured before and after RFA. Subsequently, the animals were euthanized, and the ablation lesions were assessed for transmurality. Transmurality was defined based on previously published criteria as the presence of a lesion ≥ 10 mm² on the epicardial surface (Fig. 1A) [8].

Fig. 1.

Fig. 1

Lesion Formation, QDOT MICRO Catheter, and Morphological Assessment of Electrograms before and after Ablation. A Cross-sectional images of transmural and non-transmural lesions created on the posterior wall of the right atrium. B The QDOT MICRO catheter is equipped with three microelectrodes in addition to the conventional tip and ring electrodes. These microelectrodes are capable of pacing and are characterized by a diameter of 0.33 mm, a surface area of 0.086 mm², and an interelectrode distance of 1.5 mm. C An example of morphological assessment of electrograms is presented. Post-ablation evaluation involved assessing the disappearance of the R wave in bipolar electrograms and the S wave in unipolar electrograms

Parameter assessment for transmurality

The QDOT MICRO catheter incorporates three microelectrodes (diameter: 0.33 mm, area: 0.086 mm², interelectrode spacing: 1.5 mm) capable of pacing, in addition to a conventional distal electrode (Fig. 1B). Atrial myocardial capture thresholds were measured under three configurations both before and after RFA: (1) conventional bipolar configuration (tip-to-ring electrodes: A1–A2), (2) bipolar configuration using microelectrodes (M1–M2, M2–M3, and M3–M1), and (3) unipolar configuration (A1u, M1u, M2u, and M3u). In addition to threshold measurements, the following parameters were evaluated: (1) bipolar and unipolar electrogram amplitude post-ablation, (2) the percentage reduction in amplitude before and after ablation (calculated as [pre-amplitude – post-amplitude] / pre-amplitude), (3) Threshold increase ratio (calculated as post-ablation threshold/pre-ablation threshold), (4) disappearance of R-waves in bipolar electrograms, and (5) disappearance of S-waves in unipolar electrograms [3]. During threshold measurements, the contact force was adjusted to approximately 10 g both before and after ablation. Furthermore, since the parameters of the microelectrode depend on tissue contact, we added an analysis in which the configuration with the highest amplitude prior to RFA was used for EGM amplitude analysis, and the configuration with the lowest pacing threshold was used for pacing threshold analysis [9]. The microelectrode configuration that yielded the highest bipolar amplitude before ablation, or lowest pacing threshold before ablation was defined as the “Best” configuration for each analysis. In addition, before measuring the parameters, we confirmed that the electrogram amplitude varied by less than 10% over a 3-second window.

Electrogram amplitudes for both bipolar and unipolar recordings were automatically calculated using the peak-to-peak function of the CARTO system. Capture thresholds were determined using a cardiac stimulator (BC-1100; Fukuda Denshi, Tokyo, Japan), by decrementally reducing the output voltage from a maximum of 20 V (pulse width: 1.0 ms) until the lowest value capable of achieving atrial capture was identified. The disappearance of R-waves (bipolar) and S-waves (unipolar) was independently assessed, blinded to the ablation setting, by three electrophysiologists (I.K., A.F., and M.T.) (Fig. 1C).

Statistical analysis

Continuous variables with a normal distribution were compared using the Student’s t-test, while non-normally distributed continuous variables were analyzed using the Wilcoxon rank-sum test. Categorical variables were compared using χ² test or Fisher’s exact test. Paired comparisons of pre- and post-ablation parameters were conducted using paired t-tests for normally distributed data and the Wilcoxon signed-rank test for non-parametric data. Optimal cut-off values in significant predictors to discriminate transmural lesions from non-transmural lesions were measured based on a receiver operator characteristic (ROC) curve analysis and Youden’s Index, and the area under the ROC curve (AUC) was calculated to assess the diagnostic accuracy of the parameters. A P value of < 0.05 was considered statistically significant. All statistical analyses were performed using JMP version 12.0 (SAS Institute Inc., Cary, NC, USA). Regarding sample size, this study was designed as an exploratory preclinical investigation, and therefore, no formal a priori power calculation was performed. The number of animals was based on prior preclinical studies [10,11].

Ethical approval

The study protocol of the animal experiment received approval from the Institutional Animal Care and Use Committee of Institute of SCIENCE TOKYO (A2023-148 C).

Results

Lesions located in trabeculated tissue (n = 17) were excluded from analysis due to the difficulty in accurately assessing transmurality. After excluding those lesions, a total of 131 ablation lesions were evaluated for myocardial transmurality, of which 101 (77.1%) were determined to be transmural. No gross thrombus was observed in any of the lesions. The contact force before and after ablation was 11.6 ± 4.2 g and 12.0 ± 4.7 g, respectively, with no significant difference (p = 0.234). For each parameter, measurements were obtained only after confirming that the electrogram amplitude varied by less than 10% over a 3-second window.

Electrogram amplitudes

Post-ablation electrogram amplitudes were significantly reduced compared to pre-ablation values across all bipolar and unipolar configurations (Table 1). In transmural lesions, electrogram amplitudes were lower in all bipolar and unipolar configurations compared to non-transmural lesions (Table 2, Supplement Figure S1). Compared to conventional bipolar and unipolar configurations, the amplitude was numerically lower in the bipolar and unipolar configurations using microelectrodes. Similar results were observed even when using the microelectrode configuration that yielded the highest amplitude before ablation. Among all configurations, the bipolar electrogram amplitude from the A1–A2 electrode pair demonstrated the highest predictive value for transmurality, with an area under the curve (AUC) of 0.777 at a cutoff value of 0.86 mV (sensitivity: 0.908, specificity: 0.519) (Fig. 2). In comparison, the amplitude of the microelectrode showed a lower AUC for predicting transmurality, even when using the configuration with the highest amplitude before ablation, which yielded an AUC of 0.731 with a cutoff value of 0.49 mV (Fig. 3).

Table 1.

Difference of electrogram amplitudes and pacing thresholds between pre and post ablation

pre RFA (N = 131) post RFA (N = 131) p value
Amplitude
 A1-2, mV 2.11 ± 1.36 0.67 ± 0.70 < 0.01
 M1-2, mV 1.59 ± 1.75 0.43 ± 0.69 < 0.01
 M2-3, mV 1.64 ± 2.01 0.50 ± 0.68 < 0.01
 M3-1, mV 1.45 ± 1.53 0.53 ± 0.76 < 0.01
 Best, mV 1.99 ± 2.10 0.47 ± 0.56 < 0.01
 A1u, mV 1.99 ± 1.24 1.37 ± 0.69 < 0.01
 M1u, mV 1.83 ± 1.37 1.19 ± 0.77 < 0.01
 M2u, mV 1.96 ± 1.83 1.12 ± 0.70 < 0.01
 M3u, mV 1.75 ± 1.33 1.20 ± 0.82 < 0.01
 Best, mV 2.60 ± 1.99 1.24 ± 0.75 < 0.01
Pacing threshold
 A1-2, V 1.1 ± 1.9 7.9 ± 5.9 < 0.01
 M1-2, V 8.2 ± 6.5 18.4 ± 4.2 < 0.01
 M2-3, V 9.6 ± 7.2 18.2 ± 4.6 < 0.01
 M3-1, V 9.4 ± 6.2 18.6 ± 3.8 < 0.01
 Best, V 5.8 ± 5.4 17.9 ± 4.7 < 0.01
 A1u, V 6.0 ± 6.7 10.3 ± 6.3 < 0.01
 M1u, V 11.7 ± 6.8 17.8 ± 5.3 < 0.01
 M2u, V 12.7 ± 6.6 18.0 ± 5.2 < 0.01
 M3u, V 12.2 ± 7.9 17.9 ± 5.0 < 0.01
 Best, V 8.6 ± 6.6 17.4 ± 5.7 < 0.01

Best = the microelectrode configuration that yielded the highest bipolar amplitude before ablation, or lowest pacing threshold before ablation, RFA=radiofrequency ablation

Table 2.

Difference of electrogram amplitudes between transmural and non-transmural lesion

transmural lesions (N = 101) non-transmural lesions (N = 30) P values
Amplitude
 A1-2, mV 0.49 ± 0.36 1.28 ± 1.13 < 0.01
 M1-2, mV 0.35 ± 0.64 0.72 ± 0.81 0.01
 M2-3, mV 0.35 ± 0.29 1.00 ± 1.21 < 0.01
 M3-1, mV 0.44 ± 0.64 0.88 ± 1.00 0.01
 Best, mV 0.49 ± 0.43 1.12 ± 1.51 < 0.01
 A1u, mV 1.22 ± 0.54 1.90 ± 0.90 < 0.01
 M1u, mV 1.13 ± 0.74 1.42 ± 0.87 0.06
 M2u, mV 0.98 ± 0.50 1.61 ± 1.00 < 0.01
 M3u, mV 1.09 ± 0.55 1.63 ± 1.34 0.02
 Best, mV 1.15 ± 0.67 1.71 ± 2.27 0.02
Amplitude Reduction rate
 A1-2, % 64.4 ± 31.5 40.5 ± 43.4 < 0.01
 M1-2, % 52.9 ± 117 54.6 ± 35.2 0.45
 M2-3, % 50.8 ± 63.5 20.9 ± 66.7 0.01
 M3-1, % 43.4 ± 111 28.1 ± 57.6 0.16
 Best, % 68.2 ± 22.0 59.2 ± 11.8 0.13
 A1u, % 24.6 ± 44.6 8.5 ± 31.4 0.02
 M1u, % 15.8 ± 66.7 21.1 ± 32.5 0.29
 M2u, % 23.8 ± 56.1 16.3 ± 36.7 0.21
 M3u, % 15.9 ± 66.7 9.5 ± 34.9 0.26
 Best, % 39.1 ± 12.6 30.5 ± 9.5 0.10

Best = the microelectrode configuration that yielded the highest bipolar amplitude before ablation

Fig. 2.

Fig. 2

ROC curve analysis for transmural lesions and EGM amplitude. The receiver operating characteristic curves for post-ablation electrogram amplitudes and lesion transmurality in both bipolar and unipolar configurations are presented. The conventional bipolar configuration (A1-A2) yielded the highest diagnostic performance, with an area under the curve of 0.777 at a cutoff value of 0.86 mV

Fig. 3.

Fig. 3

ROC curve analysis for transmural lesions with the best microelectrode. The receiver operating characteristic curves for lesion transmurality and post-ablation electrogram amplitudes in the best microelectrode configuration (A) and for lesion transmurality and electrogram amplitudes reduction rate in the best microelectrode configuration (B). The Receiver Operating Characteristic curves for lesion transmurality and pacing threshold in the best microelectrode configuration (C) and for lesion transmurality and pacing threshold increase in the best microelectrode configuration (D)

Percentage reduction in amplitude

When comparing the percentage reduction in amplitude between transmural and non-transmural lesions, conventional bipolar showed a statistically significant greater reduction in transmural lesions (64.4% vs. 40.5%). Similarly, conventional unipolar (A1u) also demonstrated a statistically significant amplitude reduction in transmural lesions (24.6% vs. 8.5%, p = 0.02). However, no consistent trend in amplitude reduction rate was observed with the microelectrode configurations, and no significant difference was observed between transmural and non-transmural lesions even when using the best configuration (P = 0.13, Table 2). The A1–A2 bipolar configuration exhibited the highest AUC among all configurations when evaluating the percentage reduction in amplitude (AUC: 0.683, cutoff value: 68.8%) (Supplement Figure S2).

Pacing thresholds

Pacing thresholds significantly increased after ablation compared to pre-ablation values across all configurations (Table 1). In addition, transmural lesions exhibited significantly higher pacing thresholds compared to non-transmural lesions (Table 3). In both transmural and non-transmural lesions, the pacing threshold was elevated when using microelectrodes compared to the conventional configuration. Among the various configurations, the bipolar configuration using the A1–A2 electrode pair demonstrated the highest predictive accuracy for transmurality, with an AUC of 0.880 at a cutoff value of 3.40 V (sensitivity: 0.940, specificity: 0.714) (Fig. 4). Using the bipolar configuration of the microelectrode with the lowest pre-ablation pacing threshold to predict transmurality of pacing thresholds resulted in an AUC of 0.784 (cutoff value: 19.0 V) (Fig. 3).

Table 3.

Difference of pacing threshold between transmural and non-transmural lesion

transmural lesions (N = 101) non-transmural lesions (N = 30) P values
Pacing threshold
 A1-2, V 9.4 ± 5.8 2.8 ± 2.4 < 0.01
 M1-2, V 19.0 ± 2.2 15.0 ± 6.6 < 0.01
 M2-3, V 19.4 ± 2.5 14.1 ± 7.3 < 0.01
 M3-1, V 19.5 ± 2.1 15.3 ± 6.0 < 0.01
 Best, V 19.0 ± 1.1 12.3 ± 5.0 < 0.01
 A1u, V 11.0 ± 6.5 7.9 ± 5.2 < 0.01
 M1u, V 18.4 ± 4.8 15.7 ± 6.4 0.02
 M2u, V 18.6 ± 4.7 15.8 ± 6.3 0.01
 M3u, V 18.1 ± 4.9 16.6 ± 5.5 0.07
 Best, V 17.2 ± 3.6 12.4 ± 5.4 < 0.01
Pacing threshold Increase Ratio
 A1-2 37.96 ± 47.18 13.74 ± 19.15 < 0.01
 M1-2 12.21 ± 35.45 6.75 ± 14.15 0.107
 M2-3 9.75 ± 29.55 13.48 ± 39.07 0.315
 M3-1 8.57 ± 28.44 8.98 ± 36.13 0.478
 Best 19.52 ± 44.91 20.97 ± 51.40 0.444
 A1u 5.87 ± 14.42 7.77 ± 21.71 0.330
 M1u 7.55 ± 29.62 1.60 ± 1.04 0.024
 M2u 2.36 ± 3.51 9.67 ± 35.31 0.133
 M3u 3.69 ± 10.31 2.57 ± 3.55 0.182
 Best 10.09 ± 31.04 11.15 ± 35.77 0.366

Minimum = the microelectrode configuration that yielded the lowest pacing threshold before ablation

Fig. 4.

Fig. 4

ROC curve analysis for transmural lesions and post-ablation pacing threshold. The receiver operating characteristic curves for post-ablation pacing thresholds and lesion transmurality in both bipolar and unipolar configurations are shown. The conventional bipolar configuration (A1–A2) demonstrated the highest diagnostic accuracy, with an area under the curve of 0.880 at a cutoff value of 3.40 V

Pacing threshold increase ratio

Regarding the increase in pacing threshold, a significantly higher increase was observed in transmural lesions compared to non-transmural lesions with conventional bipolar configuration (A1-A2) (Table 3). However, the AUC was low at 0.690. For the microelectrode, as with the amplitude reduction, no consistent trend was observed (Supplement Figure S3).

Morphological analysis of electrograms

The AUC values for the disappearance of R-waves in bipolar electrograms were 0.612 (A1–A2), 0.670 (M1–M2), 0.767 (M2–M3), and 0.711 (M3–M1), indicating limited predictive value. Similarly, the AUC values for the disappearance of S-waves in unipolar electrograms were 0.557 (A1u), 0.539 (M1u), 0.551 (M2u), and 0.556 (M3u), also demonstrating suboptimal predictive performance (Table 4).

Table 4.

The assessment of electrogram morphology change

transmural lesions (N = 101) non-transmural lesions (N = 30) AUC sensitivity specificity
R wave disappearance
 A1-2, N (%) 40 (39.6) 4 (13.3) 0.612 0.356 0.876
 M1-2, N (%) 75 (74.3) 11 (36.7) 0.670 0.703 0.657
 M2-3, N (%) 78 (77.2) 6 (20.0) 0.767 0.925 0.529
 M3-1, N (%) 70 (69.3) 7 (23.3) 0.711 0.904 0.603
S wave disappearance
 A1u, N (%) 88 (87.1) 22 (73.3) 0.557 0.84 0.267
 M1u, N (%) 91 (90.1) 24 (80.0) 0.539 0.87 0.200
 M2u, N (%) 90 (89.1) 23 (76.7) 0.551 0.86 0.233
 M3u, N (%) 91 (90.1) 23 (76.7) 0.556 0.87 0.233

AUC=area under the curve

Discussion

Modern microelectrode-equipped catheters offer high-resolution local signal analysis, yet the optimal strategy to confirm lesion transmurality remains a critical unmet need. This proof-of-concept study was designed to answer a direct clinical question: which electrophysiological parameter, from which electrode configuration, most reliably predicts macroscopic lesion transmurality? Our principal findings provide a clear and actionable answer to this question: First, the conventional tip-ring bipolar configuration (A1–A2) consistently outperformed all microelectrode configurations in predicting transmurality. Second, the post-ablation pacing threshold was the single most robust predictor, demonstrating superior diagnostic accuracy (AUC 0.880) compared to electrogram amplitude (AUC 0.777) or morphological changes. Finally, we identified an optimal and clinically practical cutoff value of 3.4 V (at 1.0 ms) for predicting transmurality with high sensitivity and specificity using this configuration.

Taken together, myocardial injury, the broader field of view offered by conventional electrodes is more suitable than the narrow, superficial view of microelectrodes. Consistent with this interpretation, a secondary analysis including lesions formed in trabeculated regions suggested a trend toward reduced diagnostic performance of pacing- and electrogram-based criteria. This likely reflects diminished signal specificity in anatomically complex regions, where far-field capture and mixed near- and far-field activation are more prevalent. Importantly, this observation should be viewed not only as a technical limitation of the present study, but also as a fundamental biophysical constraint inherent to pacing-based transmurality assessment. Accordingly, while the underlying biophysical principle remains valid, the direct application of specific cutoff values to heavily trabeculated atrial regions should be approached with caution in the absence of dedicated validation.

EGM amplitude in microelectrodes and conventional electrodes

The key to understanding the differences in pre- and post-ablation electrogram changes between microelectrodes and conventional electrodes lies in considering the differences in the field of view, which is determined by the size of the electrodes and the interelectrode spacing of the catheter. Microelectrodes are considered more sensitive in detecting post-ablation amplitude reduction compared to conventional electrodes, with reported reduction rates ranging from 40% to 80%, which is consistent with the findings of our study [10,12]. Furthermore, when comparing the voltage amplitudes recorded with microelectrodes and conventional tip-ring configurations, microelectrodes demonstrated higher amplitudes at non-ablation sites and lower amplitudes at ablation sites [13]. More recently, it has been reported that the low-voltage area after ablation also varies depending on electrode size and interelectrode spacing, with smaller electrodes and shorter interelectrode distances resulting in smaller low-voltage areas [14,15].

An interesting observation has been reported in a previous study [11]: when RFA was performed under identical settings in both the atria and ventricles, conventional electrodes showed a pronounced reduction in electrogram amplitude in the atria but only a modest change in the ventricles. In contrast, microelectrodes exhibited comparable amplitude reductions in both chambers, and the degree of reduction was consistently greater than that observed with conventional electrodes. This discrepancy is likely attributable to the smaller field of view of microelectrodes. That is, as long as the lesion fully encompasses the microelectrode’s limited sensing area—regardless of atrial or ventricular location—a marked attenuation in amplitude can be observed. On the other hand, conventional electrodes, which have a broader field of view, are more susceptible to far-field signals originating from unablated, remote myocardium. This effect is particularly relevant in the ventricles, where thicker myocardial walls may result in far-field contributions that mask true lesion-induced EGM reductions (Fig. 5).

Fig. 5.

Fig. 5

Illustration of different field of view and tissue capturing by microelectrodes and conventional electrodes. Differences in EGM amplitudes and pacing capture threshold between using microelectrodes and conventional electrodes are demonstrated. Since microelectrodes have a small field of view, the recorded amplitude tends to be similarly low in both non-transmural and transmural lesions. In contrast, conventional bipolar electrodes have a larger field of view, resulting in a difference in recorded amplitude between non-transmural and transmural lesions (Left Panel). Despite using the same output, variations in electrode size and interelectrode distance result in differences in the amount of myocardium captured, with microelectrodes—due to their smaller size and shorter interelectrode spacing—capturing a more localized region of tissue (Right Panel)

These findings suggest that for evaluating lesion formation in thicker myocardial tissue, electrodes with a wider field of view may be more suitable, thereby providing a rationale for the utility and potential superiority of conventional electrodes in such settings. In our study, we observed no significant difference in amplitude reduction between transmural and non-transmural lesions when using the microelectrode configuration. However, with the conventional bipolar configuration, the amplitude reduction rate was greater in transmural lesions compared to non-transmural lesions. Indeed, a previous study using conventional electrodes reported that conventional bipolar EGMs show a reduction of more than 50% in transmural lesions, whereas such reductions are less common in non-transmural lesions [16]. This is consistent with our findings, where the bipolar EGM amplitude was reduced by 64.4% in transmural lesions and 40.5% in non-transmural lesions. These results also reflect the differences in the field of view between microelectrodes and conventional bipolar electrodes.

Pacing threshold with microelectrodes and conventional electrodes

It has also been reported that microelectrodes show higher post-ablation pacing thresholds compared to conventional electrodes [9]. This is considered to be due to the reduced influence of far-field potentials, and in our study as well, pacing thresholds were higher with microelectrode configurations. However, the ROC curve using pacing thresholds from microelectrodes to predict transmural lesions was inferior to that of conventional electrodes. The best predictive performance for identifying transmural lesions was observed with conventional bipolar electrodes, with an ROC of 0.880 at a cutoff value of 3.4 V. In a previous study that used pacing to identify lesion gaps during pulmonary vein isolation, pacing at 10 V/2 ms was performed. If capture was observed, the lesion was considered non-transmural, and additional ablation was applied, leading to favorable outcomes in pulmonary vein isolation [17]. In contrast, our study proposes a cutoff value of 3.4 V/1.0 ms to identify non-transmural lesions. Applying this cutoff may help avoid unnecessary ablation.

The microelectrode, with a smaller electrode size of less than 1 mm compared to conventional electrodes, is capable of reflecting the myocardial potentials of the contacted myocardial tissue without being influenced by far-field potentials. However, the present study revealed that microelectrodes, both in terms of electrogram amplitude and pacing threshold, demonstrated lower accuracy in predicting lesion transmurality compared to the conventional tip-ring bipolar configuration. One possible reason why the microelectrode was not useful for predicting transmurality is that, due to its small electrode size, it primarily records the potential from the surface of the tissue. As shown in Fig. 6, with the bipolar configuration using the microelectrode, the potential disappears immediately after the ablation for both M1–2 and M3–1, although the lesion created was non-transmural. In contrast, A1–2 shows only a reduction in amplitude due to the influence of far-field potentials. This suggests that the microelectrode records only superficial potentials, which likely explains why post-ablation electrogram amplitude did not accurately reflect lesion transmurality. Additionally, theoretically, the bipolar configuration of the microelectrode, due to its smaller electrode size and shorter interelectrode distance, may capture a smaller range of myocardial tissue than the conventional bipolar configuration, even with the same output (Fig. 5). In other words, the smaller area captured by pacing may have limited its utility in evaluating transmurality.

Fig. 6.

Fig. 6

Intracardiac electrograms during RFA with M1 microelectrode in tissue contact. Intracardiac electrograms recorded before and after ablation are shown, with microelectrode M1 in contact with the tissue. When microelectrode M1 is in contact, sharp bipolar electrograms are recorded from M1–M2 and M3–M1, while the bipolar electrogram recorded from M2–M3 reflects only far-field potentials. Following high-power short-duration ablation (90 W for 2 s), the bipolar electrograms recorded by the microelectrodes (M1–M2 and M3–M1) are abolished, whereas the conventional bipolar electrogram (A1–A2) demonstrates only attenuation

This is further suggested by the higher pacing thresholds observed with the microelectrode bipolar configuration even before ablation. The presence of sharp and large local electrograms recorded with microelectrodes, along with low pacing thresholds, may more specifically indicate that a small superficial area of tissue remains unablated, suggesting a non-transmural lesion in the atrium. However, the inverse is not necessarily true; that is, small local electrograms or high pacing thresholds observed with microelectrodes do not necessarily indicate the presence of a transmural lesion. The atrial wall thickness is predominantly between 3 and 5 mm, with thicker regions reaching up to approximately 10 mm [18]. To assess the transmurality of such atrial walls, a certain field of view is required, and in this context, conventional electrodes are superior to microelectrodes. These biophysical principles, however, were established in a controlled preclinical model. In our study, lesions located in trabeculated regions were excluded from the primary analysis. In these regions, complex anatomy and prominent trabeculation substantially increase the likelihood of far-field capture and mixed near- and far-field potential, which inherently reduces the accuracy of pacing-based assessment of transmurality. When we secondarily included these lesions, the predictive performance of the 3.4-V cutoff and the 0.86-mV amplitude threshold decreased slightly (ROC 0.865 and 0.751, respectively), supporting the notion that trabeculation degrades signal specificity. Taken together, our data suggest that while the underlying biophysical principle may still hold when lesions are homogeneously formed, the specific cutoff values proposed in this study may not be directly extrapolated to the human left atrium, which is thicker, more heterogeneous, and frequently fibrotic.

Effect of microelectrode contact on tissue

From a mechanistic standpoint, the sensitivity of microelectrodes to subtle variations in tissue contact may further explain their inferior performance in predicting lesion transmurality. Because microelectrodes sample a very limited and superficial myocardial volume, incomplete or asymmetric contact can disproportionately affect both recorded electrogram amplitudes and pacing thresholds. As a result, even small differences in contact geometry may lead to apparent signal attenuation despite the persistence of deeper viable myocardium. This contact-dependent behavior is therefore more likely to confound transmurality assessment with microelectrodes than with conventional bipolar electrodes, which integrate signals over a broader myocardial field of view.

Another possible explanation for the superior predictive performance of the conventional electrode configuration over the microelectrode configuration in identifying transmural lesions is that microelectrodes may not always be in direct contact with the myocardial tissue. While contact force can be evaluated at the catheter tip, the extent to which each microelectrode is in contact with the myocardial tissue cannot be directly assessed. Low or insufficient contact between the microelectrode and the myocardial tissue could influence the ROC curve. As shown in the example in Fig. 6, M1 is in contact with the tissue, which is reflected in sharp bipolar electrograms observed between M1–M2 and M3–M1. However, no such sharp potential is seen with M2–M3. In contrast, the conventional A1–A2 configuration reliably demonstrates a sharp electrogram.

Taking into account which microelectrode had better contact with the myocardial tissue, the bipolar configuration of the microelectrode showing the largest pre-ablation electrogram amplitude was used to predict transmurality in post-ablation amplitude. This yielded an AUC of 0.731 with a cutoff value of 0.49 mV. Similarly, using the bipolar configuration of the microelectrode with the lowest pre-ablation pacing threshold to predict transmurality of pacing thresholds resulted in an AUC of 0.784 (cutoff value: 19.0 V). Although a slight improvement in AUC was observed, the results were still inferior to those of the conventional bipolar configuration. These findings suggest that the superior utility of conventional bipolar electrodes over microelectrodes in predicting transmural lesions is primarily attributable to differences in electrode size and interelectrode spacing, which in turn affect the amount of myocardium captured and the resulting field of view.

Morphology of electrograms

A previous study has reported that when the ablation catheter is applied vertically, the R-wave of the bipolar electrogram disappears [3]. In the current study, however, the catheter was often angled using a steerable sheath, as is commonly done in clinical practice. This may explain why the disappearance of the R-wave was not useful in predicting transmural lesions. Regarding the disappearance of S wave in unipolar electrograms, it was nearly absent in almost all lesions, regardless of whether they were transmural or non-transmural, when assessed from the microelectrodes. Furthermore, it was difficult to determine which microelectrode was in contact with the tissue, making the assessment of transmurality based on S-wave disappearance challenging. In this preclinical study using the QDOT-MICRO catheter, pacing capture threshold with the conventional bipolar configuration was found to be more useful for predicting lesion transmurality than the electrogram morphology.

Taken together, pacing threshold measured using the conventional bipolar configuration represents the most robust functional indicator of lesion transmurality in this preclinical model. Despite the absence of a formal analysis of the relationship between ablation settings and lesion metrics, the consistent superiority of the conventional bipolar configuration across the entire lesion set suggests that the predictive value of pacing threshold is not critically dependent on the specific ablation strategy.

Limitations

This study has several important limitations. First, the findings are derived from an in vivo porcine right atrium model and may not be directly extrapolated to the thicker and more heterogeneous human left atrium, or to ventricular tissue. Second, our analysis was intentionally limited to smooth endocardial surfaces to ensure accurate histological assessment of transmurality. A stratified analysis comparing lesion transmurality prediction in regions with varying atrial wall thickness was not performed. The applicability of the proposed cutoff values to complex atrial anatomy requires further investigation. Third, the results and the optimal cutoff value of 3.4 V are specific to the catheter platform (e.g. electrode size and spacing) similar to the QDOT MICRO. The pulse width was set at 1.0 ms, and validation using other pulse widths has not been performed. Other catheters with different electrode sizes and spacing may yield different absolute values. At present, we are conducting clinical validation of the optimal cutoff value, and are also performing validation using other ablation catheters. Fourth, catheter orientation was not standardized or quantitatively recorded. Although a secondary analysis using the best pre-ablation microelectrode configuration was performed to partially account for variability in microelectrode–tissue contact, residual confounding by catheter orientation cannot be excluded. In addition, time-resolved impedance data were unavailable, precluding evaluation of the relationships of baseline impedance and impedance drop with pacing thresholds and lesion characteristics. Fifth, the findings of the present study are limited to measurements obtained with a contact force of approximately 10 g. Therefore, the changes in electrogram amplitude and pacing threshold observed in this study may not be directly applicable to measurements obtained under very low (< 5 g) or high (> 20 g) contact force conditions, as measurements are highly dependent on catheter-tissue contact. Sixth, lesion width and surface dimensions were not systematically quantified; therefore, their relationships with electrogram changes and pacing thresholds could not be assessed. Accordingly, the present findings pertain specifically to focal lesion transmurality and should not be extrapolated to lesion continuity or overall lesion quality. Despite these limitations, this study provides a critical biophysical principle: for assessing deep myocardial injury, the broader field of view offered by a conventional tip-ring bipolar configuration is superior to the narrow, superficial view of microelectrodes. This principle is likely to be broadly applicable, even if the specific cutoff values require further validation in different clinical contexts. From a clinical perspective, pacing threshold may serve as a practical functional endpoint in relatively smooth atrial myocardium, while its interpretation in anatomically complex or trabeculated regions should be approached with caution.

Conclusions

The evaluation of atrial RFA lesions using the QDOT-MICRO catheter demonstrates that the bipolar configuration between the tip and ring electrodes is effective for assessing the lesion transmurality. A pacing threshold of 3.40 V was identified as the optimal predictor of atrial tissue transmurality.

Supplementary Information

Below is the link to the electronic supplementary material.

Acknowledgements

We thank Mr. Shunto Yamashita, Mr. Takanori Tanabe, and Ms. Megumi Toda at the Institute of Science Tokyo for their technical assistance during the animal experiments.

Abbreviations

AUC

Area under the curve

RFA

Radiofrequency ablation

ROC

Receiver Operating Characteristic

Author contributions

Study conception and design: M.T.Fundings: M.TAnimal Care: S.S., T.I. Data collection: I.K, A.F. , M.T., J.Y., R.K., M.H., H.I., R.T., M.N., K.G., K.I., T.N., K.Y., S.T.Data analysis: A.F. , M.T.Data interpretation: A.F. , I.K, M.T.Drafting the manuscript: A.F. , I.K, M.T.Revising the manuscript: A.F. , I.K, M.T.Critical revision: A.F. , I.K, M.T., S.M., T.S.Final approval of the manuscript: All authors.

Funding

This work was supported by Biosense Webster through the investigator-initiated study program IIS-648 and by the Japan Society for the Promotion of Science (grant/award number 22K16068).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

This work was supported by Biosense Webster through the investigator-initiated study program IIS-648 and by the Japan Society for the Promotion of Science (grant/award number 22K16068).Drs Goto, Takigawa, and Miyazaki received endowments from Medtronic Japan, Boston Scientific, Japan Lifeline, and Win International. No other authors have conflict of interest to declare.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Iwanari Kawamura, Aiko Fujimaki and Masateru Takigawa are contributed equally to this work.

References

  • 1.Habibi M, Berger RD, Calkins H. Radiofrequency ablation: technological trends, challenges, and opportunities. Europace. 2021;23(4):511–9. [DOI] [PubMed] [Google Scholar]
  • 2.Reichlin T, Knecht S, Lane C, et al. Initial impedance decrease as an indicator of good catheter contact: insights from radiofrequency ablation with force sensing catheters. Heart Rhythm. 2014;11(2):194–201. [DOI] [PubMed] [Google Scholar]
  • 3.Otomo K, Uno K, Fujiwara H, Isobe M, Iesaka Y. Local unipolar and bipolar electrogram criteria for evaluating the transmurality of atrial ablation lesions at different catheter orientations relative to the endocardial surface. Heart Rhythm. 2010;7(9):1291–300. [DOI] [PubMed] [Google Scholar]
  • 4.Hamner CE, Potter DD Jr, Cho KR, et al. Irrigated radiofrequency ablation with transmurality feedback reliably produces Cox maze lesions in vivo. Ann Thorac Surg. 2005;80(6):2263–70. [DOI] [PubMed] [Google Scholar]
  • 5.Delacretaz E, Soejima K, Brunckhorst CB, Maisel WH, Friedman PL, Stevenson WG. Assessment of radiofrequency ablation effect from unipolar pacing threshold. Pacing Clin Electrophysiol. 2003;26(10):1993–6. [DOI] [PubMed] [Google Scholar]
  • 6.Reddy VY, Grimaldi M, De Potter T, et al. Pulmonary vein isolation with very high power, short duration, temperature-controlled lesions: The QDOT-FAST trial. JACC Clin Electrophysiol. 2019;5(7):778–86. [DOI] [PubMed] [Google Scholar]
  • 7.Berte B, Zeppenfeld K, Tung R. Impact of micro-, mini- and multi-electrode mapping on ventricular substrate characterisation. arrhythm electrophysiol Rev. 2020;9(3):128–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Yavin HD, Leshem E, Shapira-Daniels A, et al. Impact of high-power short-duration radiofrequency ablation on long-term lesion durability for atrial fibrillation ablation. JACC Clin Electrophysiol. 2020;6:973–85. [DOI] [PubMed] [Google Scholar]
  • 9.Bhaskaran A, Chik W, Pouliopoulos J, Nalliah C, et al. Five seconds of 50–60 W radio frequency atrial ablations were transmural and safe: an in vitro mechanistic assessment and force-controlled in vivo validation. Europace. 2017;19(5):874–80. [DOI] [PubMed] [Google Scholar]
  • 10.Avitall B, Horbal P, Vance D, Koblish J, Kalinski A. Maximal electrogram attenuation recorded from mini electrodes embedded on 4.5-mm irrigated and 8-mm nonirrigated catheters signifies lesion maturation. J Cardiovasc Electrophysiol. 2015;26(2):192–202. [DOI] [PubMed] [Google Scholar]
  • 11.Lázaro C, Barrio-López T, Castellanos E, Ortiz M, Arceluz M, Almendral J. Electrogram voltage and pacing threshold before ablation, measured by mini-electrodes, predict parameters indicative of transmural lesions in the human atrium. J Interv Card Electrophysiol. 2020;57(3):443–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Avitall B, Kalinski A, Horbal P, Koblish J. Relationship between lesion formation and electrophysiological responses using catheters equipped with mini-electrodes in chronic atrial fibrillation. Heart Rhythm. 2017;14(6):902–9. [DOI] [PubMed] [Google Scholar]
  • 13.Lloyd MS, Hoskins MH, Shah AD, Langberg JJ. Electrogram characteristics of ablated and non-ablated myocardium in Humans: A comparison of miniaturized embedded electrodes and conventional ablation electrodes. J Cardiovasc Electrophysiol. 2016;27(7):820–4. [DOI] [PubMed] [Google Scholar]
  • 14.Knecht S, Schlageter V, Badertscher P, et al. Atrial substrate characterization based on bipolar voltage electrograms acquired with multipolar, focal and mini-electrode catheters. Europace. 2023;25(5):euad127. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Yamaguchi J, Takigawa M, Goya M, et al. Impact of 1 mm difference in interelectrode spacing on local electrograms and voltage/activation maps. Europace. 2025;27(7):euaf135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Avitall B, Helms RW, Koblish JB, Sieben W, Kotov AV, Gupta GN. The creation of linear contiguous lesions in the atria with an expandable loop catheter. J Am Coll Cardiol. 1999;33(4):972–84. [DOI] [PubMed] [Google Scholar]
  • 17.Eitel C, Hindricks G, Sommer P, et al. Circumferential pulmonary vein isolation and linear left atrial ablation as a single-catheter technique to achieve bidirectional conduction block: the pace-and-ablate approach. Heart Rhythm. 2010;7(2):157–64. [DOI] [PubMed] [Google Scholar]
  • 18.Koruth JS, Kuroki K, Kawamura I, et al. Focal pulsed field ablation for pulmonary vein isolation and linear atrial lesions: A preclinical assessment of safety and durability. Circ Arrhythm Electrophysiol. 2020;13(6):e008716. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Data Availability Statement

No datasets were generated or analysed during the current study.


Articles from Journal of Interventional Cardiac Electrophysiology are provided here courtesy of Springer

RESOURCES