Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Jul 31.
Published in final edited form as: Circ Arrhythm Electrophysiol. 2026 Jul 29;19(8):e014978. doi: 10.1161/CIRCEP.126.014978

Innervation of the Human Atrioventricular Node via the Inferior Pyramidal Space: Characterization Using Intersectional Strategies

Takanori Sato 1, Shumpei Mori 1, Peter Hanna 1, Joseph E Hadaya 1, Yuichiro Miyazaki 1, Logan G Kirkland 2, Donald B Hoover 2,3, Olujimi A Ajijola 1, Kalyanam Shivkumar 1
PMCID: PMC13419474  NIHMSID: NIHMS2197761  PMID: 42522551

Abstract

Background:

The inferior pyramidal space (IPS), the epicardial adipose tissue extending from the inferior crux of the heart, contains the atrioventricular node (AVN) at its apex. However, innervation of the human AVN via the IPS remains unknown.

Methods:

Seven whole-mount samples of the human IPS involving the AVN were obtained from hearts rejected for transplantation (85.7% male, 49.4 ± 12.8 years old). The IPS was further divided into the anterior (behind the floor of the triangle of Koch) and posterior (within the Eustachian ridge) fat pads relative to the coronary sinus. A whole-mount tissue-clearing method with immunohistochemical staining was applied to each fat pad to count and characterize the nerve fascicles ≥ 10 μm and ganglia ≥ 6000 μm2.

Results:

Within the anterior fat pad, sympathetic nerve fascicles ascended towards the AVN along the AVN artery, whereas non-sympathetic nerve fascicles approached the AVN from the superior and posterior aspects without a consistent relationship to the AVN artery. Within the IPS, sympathetic nerve fascicles were thicker than non-sympathetic nerve fascicles (51.7 ± 32.8 vs. 22.1 ± 10.6 μm, p < 0.001). The ganglia, which were mainly non-sympathetic, were found predominantly within the posterior fat pad compared with the anterior fat pad (63.2 ± 40.3 vs. 5.8 ± 5.0, p < 0.01). The compact AVN regions were devoid of thick nerve fascicles and ganglionated plexuses but filled with a fine meshwork of both sympathetic and non-sympathetic nerve fibers.

Conclusion:

The human IPS exhibits a distinct pattern of autonomic innervation and distribution of ganglia. These neural structures are likely to contribute to the physiological control of the human AVN and could be a novel target for neuromodulation to restore normal AVN function.

Keywords: Anatomy, Atrioventricular node, Ganglionated plexus, Inferior pyramidal space, Innervation

Graphical Abstract

graphic file with name nihms-2197761-f0007.jpg

Introduction

The atrioventricular node (AVN) plays a crucial role in the cardiac conduction system, serving as a gatekeeper for atrioventricular (AV) conduction by setting a physiological delay that allows effective AV synchrony to achieve optimal ventricular filling.1 Since such AV conduction needs fine and timely adjustment within a physiological range of variable heart rates, the autonomic nervous system integrates sympathetic and parasympathetic inputs to regulate AV conduction and refractoriness in both animal and human hearts.2–5 Disruption of these autonomic influences can lead to functional AV block and supraventricular, junctional, and ventricular arrhythmias,6–8 highlighting the importance of understanding the innervation of the AVN in both normal and pathological states.

The AVN is located at the apex of the triangle of Koch, inferoposterior to the AV portion of the membranous septum. The AVN is fixed on the right-sided surface of the central fibrous body, also referred to as the right fibrous trigone, and it is covered by the thin fibro-adipose tissue intervening between the right atrium and the left ventricle.9–12 This epicardial fibro-adipose tissue is referred to as the inferior pyramidal space (IPS), extending from the inferior crux of the heart towards the central fibrous body (Figure 1).9–12 Thus, the IPS bears the AVN at its apex, and it allows epicardial vessels, including the AVN artery, and nerves to communicate with the AVN.

Figure 1: Inferior pyramidal space and atrioventricular node sampling.

Figure 1:

Open in a new tab

Multiplanar reconstruction images obtained from clinical cardiac computed tomography datasets demonstrate the inferior pyramidal space (IPS) (yellow dotted line) viewed from the right anterior oblique (A) and inferior (B) directions. The IPS is divided by the coronary sinus (white asterisk) into the anterior and posterior fat pads.

The middle panels (C-E) show the progressive dissection of the floor of the triangle of Koch in a real heart to expose the anterior fat pad of the IPS and atrioventricular node (red asterisk) from the right lateral view. Note that the IPS, epicardial adipose tissue wedging from the inferior crux towards the central fibrous body, is sandwiched between the right atrium and left ventricle, carrying the atrioventricular node artery. A virtual dissection image of the IPS reconstructed from cardiac computed tomography datasets (F) and the corresponding en-bloc IPS specimen from a real dissection (G) shows the three-dimensional feature of the IPS for whole-mount analysis. The IPS specimen is further separated into the anterior and posterior fat pads by the coronary sinus (H). L, left coronary aortic sinus; N, non-coronary aortic sinus; R, right coronary aortic sinus.

Innervation of the AVN region has been investigated in both animal and human hearts.13–25 Tranum-Jensen et al. described a ganglionated plexus located beneath the AVN within the floor of the triangle of Koch in the porcine heart, which was termed the ganglion atrioventriculare14 as confirmed by Crick et al.19 However, the IPS of the human heart, as a potential route for cardiac innervation, has not been studied in detail despite containing the AVN artery and AVN. This study aims to clarify the comprehensive topographic characteristics of human AVN innervation via the IPS using a whole-mount tissue-clearing technique with immunohistochemical staining.

Methods

Data Availability

The data underlying this article will be shared on reasonable request to the corresponding author.

Sample collection

Seven human donor hearts rejected for transplantation were collected with institutional review board approval (IRB#19–000624) and consent. Each heart underwent pressure-perfusion and fixation in 4% paraformaldehyde at 4°C for 24 hours to preserve the physiological morphology of the heart.26 The IPS was carefully dissected with the AVN on its apex (Figure 1)12 and further divided into two subregions based on the location relative to the coronary sinus: the anterior fat pad behind the floor of the triangle of Koch, and the posterior fat pad within the Eustachian ridge. For comparative and quantitative analysis, five additional samples of the sinus node region, five samples of the AVN artery and sinus node artery, and three samples of the AVN region were obtained from different donor hearts.

Histological preparation

The modified immunolabeling-enabled three-dimensional imaging of solvent-cleared organs (iDISCO+) method was employed to visualize and assess the three-dimensional distribution of the neural components within the structures of interest, as previously described.26,27 Primary antibodies were applied for 7 days at 37 °C: rabbit anti-protein gene product 9.5 (Abcam, ab108986, 1:1000) as a pan-neuronal marker and sheep anti-tyrosine hydroxylase (TH) (Sigma-Aldrich, ab1542, 1:500) to stain for sympathetic neurons. Appropriate secondary antibodies were applied for 7 days at 37 °C: donkey anti-rabbit Cy3 (Jackson ImmunoResearch Laboratories, AB_2307443, 1:200) and donkey anti-sheep 647 (Jackson ImmunoResearch Laboratories, AB_2340750, 1:200). Samples were cleared using the iDISCO+ protocols and were imaged using a confocal laser scanning microscope (ZEISS LSM 880, Carl Zeiss Microscopy, Oberkochen, Germany). Images were captured and analyzed using ZEISS ZEN Blue software (Carl Zeiss Microscopy).

Conventional histological analysis was also performed on 5 μm thick paraffin-embedded sections involving the compact AVN. The sections were stained with Masson’s trichrome using standard techniques.28 The stained slides were imaged using an Aperio AT Turbo scanning machine (Leica Biosystems, Nussloch, Germany) at 20× magnification and analyzed with Aperio ImageScope software (Leica Biosystems). The slides were also evaluated with enzyme-based immunohistochemistry staining for TH and vesicular acetylcholine transporter (VAChT, specific marker for cholinergic neurons), as previously applied to human sinus node tissue.29 Digital images were captured at 20× magnification.

For whole-mount samples treated with iDISCO+ method, cholinergic marker such as VAChT and afferent marker such as calcitonin gene-related peptide did not show reliable and reproducible results. Therefore, ‘non-sympathetic’ fascicles/fibers should involve not only parasympathetic component but also afferent component.

Quantitative analyses

Nerve fascicles measuring 10 μm or more in diameter were identified, and their diameters were measured. Each nerve fascicle was categorized as sympathetic (TH-positive) or non-sympathetic (TH-negative). Ganglia were defined as clusters of neuron cell bodies occupying an area of 6000 μm2 or more on maximum intensity projection images. These ganglia were counted within each region of the IPS (anterior and posterior to the coronary sinus) and within 1 mm external to the tunica media of both the AVN artery and sinus node artery, using ImageJ software (National Institutes of Health, Bethesda, MD, USA). Nerve fiber density was quantified as the percentage of tissue area occupied by sympathetic (TH-positive) or parasympathetic (VAChT-positive) fibers using ImageJ software.

Statistical analysis

Data were analyzed using JMP Pro 18.0.2 (SAS Institute, Cary, USA). Descriptive statistics were computed for all variables. Numerical variables were presented as mean ± standard deviation, and categorical variables were expressed as percentages. Differences in average nerve fascicle diameter and ganglionated plexus distribution were evaluated using one-way analysis of variance (ANOVA). Given the small sample size and sensitivity of ANOVA to minor departure from normality, nonparametric test was also performed to confirm the ANOVA results. The Wilcoxon signed-rank test was used for paired comparisons, and the Mann–Whitney U test was used for unpaired comparisons, as appropriate. A p-value < 0.05 was considered statistically significant.

Results

A total of seven human IPS samples (85.7% male, 49.4 ± 12.8 years) were analyzed. All samples were obtained from hearts that were in sinus rhythm with a left ventricular ejection fraction of 57.7 ± 12.6%. The average body mass index was 27.1 ± 6.3 kg/m2. The anatomical and histological characteristics of the IPS are listed in Table 1. Gross dissection showed the IPS extending from the inferior crux at the diaphragmatic surface towards the central fibrous body. The coronary sinus divides the IPS into the anterior and posterior fat pads, with individual variation in size. The anterior fat pad is approximately twice as large as the posterior fat pad (Table 1). The compact AVN was located at the apex of the anterior fat pads of the IPS. The AVN was found on the right atrial surface of the central fibrous body, the thick and firm connective tissue inferior to the non-coronary aortic sinus. The AVN artery ascended within the IPS to reach the AVN (Figure 1). This square pyramidal-shaped epicardial adipose tissue mass comprising the IPS could also be detected and reconstructed when using cardiac computed tomography datasets of living patients (Figure 1). Volume-rendered images show identical morphology, and multiplanar reconstruction images show how the anterior and posterior fat pads comprise the floor of the triangle of Koch and the Eustachian ridge, respectively (Figure 1).

Table 1:

Anatomical and histological characteristics of the inferior pyramidal space

Variable Inferior pyramidal space (IPS)
(n=7)
Age (years) 49.4 ± 12.8
Men, n (%) 6 (85.7)
Body height (cm) 177.7 ± 9.0
Body weight (kg) 85.8 ± 23.1
Body mass index (kg/m2) 27.1 ± 6.3
Anterior fat pad of the IPS
(n = 7)		 Posterior fat pad of the IPS
(n = 7)		 p value
Area of the each region (mm2) 355.0 ± 124.6 175.7 ± 101.1 0.02
Diamter of the sympathetic nerve fascicles (μm) 51.0 ± 32.7 52.4 ± 33.0 0.69
Diamter of the non-sympathetic nerve fascicles (μm) 18.2 ± 8.2 23.7 ± 11.1 <0.01
Number of the ganglia (>6000 μm2) 5.8 ± 5.0 63.2 ± 40.3 <0.01
Density of the ganglia (/cm2) 1.9 ± 1.8 41.2 ± 25.1 <0.01
Open in a new tab

Whole-mount tissue-clearing and immunohistochemistry staining revealed the distribution of sympathetic and non-sympathetic nerve fascicles and ganglia within the IPS (Figures 2 and 3). Across the IPS, sympathetic nerve fascicles were significantly thicker than non-sympathetic nerve fascicles (51.7 ± 32.8 μm vs. 22.1 ± 10.6 μm, p < 0.001). In the anterior fat pad, thick sympathetic fascicles extended towards the AVN along the AVN artery. In contrast, non-sympathetic nerve fascicles were distributed independently of the AVN artery and converged to the AVN, with fascicles approaching the AVN from the superior and posterior directions (Figure 2). In the posterior fat pad, thick sympathetic fascicles were positioned superficially along the epicardial side, whereas thinner non-sympathetic fascicles were more diffusely distributed and predominated in the superior aspect (Figure 3). Ganglia, predominantly non-sympathetic, were located mainly in the posterior fat pad rather than in the anterior fat pad (Figure 4). The diameter of non-sympathetic nerve fascicles in the posterior fat pad was greater than that in the anterior fat pad (Table 1). These findings suggest a predominant role of the posterior fat pad as a region carrying a local control center via postganglionic parasympathetic neurons. Due to limited sample size, in part, size of each fat pad, nerve fascicles, and number of the ganglia exhibited individual variations as noted by large standard deviations (Table 1).

Figure 2: Whole-mount staining of the anterior fat pad of the inferior pyramidal space.

Figure 2:

Open in a new tab

The anterior fat pad of the inferior pyramidal space (A) is processed using immunolabeling-enabled three-dimensional imaging of solvent-cleared organs (B, C). The maximum intensity projection image (C) highlights thick sympathetic nerve fascicles surrounding the right coronary artery (G). The magnified images (D-G) confirm thick sympathetic nerve fascicles (green arrowheads) surrounding the right coronary artery (G) and extending towards the atrioventricular node (AVN) along the AVN artery (E). In contrast, non-sympathetic fascicles (white arrowheads) are distributed independently of the AVN artery and converged on the AVN (F), with additional nerve fascicles approaching from the superior and posterior aspects (D). PGP, protein gene product; TH, tyrosine hydroxylase. Scale bars are 5 mm (B, C), 2 mm (D-G).

Figure 3: Whole-mount staining of the posterior fat pad of the inferior pyramidal space.

Figure 3:

Open in a new tab

The posterior fat pad of the inferior pyramidal space (A) is processed using immunolabeling-enabled three-dimensional imaging of solvent-cleared organs (B, C). Thick sympathetic nerve fascicles are located superficially along the epicardial side. Non-sympathetic nerve fascicles, thinner than sympathetic fascicles (Table 1), are more diffusely distributed, extending from the epicardium towards the superior aspect (D). Therefore, non-sympathetic nerve fascicles predominate within the superior region (E). PGP, protein gene product; TH, tyrosine hydroxylase. Scale bars are 5 mm (B, C) and 500 μm (D, E). Based on the present findings (Figures 1–4), innervation of the human atrioventricular node is illustrated (F). As we separated the anterior and posterior fat pads, further research is required to discern connection or communication between the anterior and posterior fat pads (question marks), presumably around the coronary sinus orifice.

Figure 4: Distribution of ganglia in the inferior pyramidal space.

Figure 4:

Open in a new tab

Multiple ganglia within the inferior pyramidal space are digitally highlighted in sky-blue by using a whole-mount staining sample labeled with protein gene product (PGP) 9.5 (A). The posterior fat pad of the inferior pyramidal space contains significantly more ganglionated plexuses, predominantly non-sympathetic, than the anterior fat pad (Table 1). Representative maximum intensity projection confocal images obtained from the posterior (B) and anterior (C) fat pads of the inferior pyramidal space show that the number of ganglia and their individual sizes are larger in the posterior fat pad. Scale bar is 5 mm (A) and 200 μm (B, C). TH, tyrosine hydroxylase.

Examination of the AVN artery in the additional five isolated artery specimens further confirmed thick perivascular sympathetic nerve fascicles (Figure 5, Table 2). Non-sympathetic nerve fascicles were present in smaller numbers and rarely formed sizable ganglia in this area. The sinus node artery also showed sympathetic predominance, but with more non-sympathetic nerve fascicles combined with occasional small non-sympathetic ganglia (Figure 5, Table 2). Statistical significance of each variable listed in Tables 1 and 2 detected using one-way ANOVA was also confirmed using nonparametric test.

Figure 5: Whole-mount staining of the atrioventricular and sinus node arteries.

Figure 5:

Open in a new tab

A whole-mount image of the atrioventricular node artery (AVNA) with an accompanying venous branch shows thick sympathetic nerve fascicles running in parallel with the artery, whereas fewer non-sympathetic fascicles are found within the surrounding adipose tissue (A). Similarly, the sinus node artery (SNA) displays predominant sympathetic nerve fascicles parallel to the artery, though more non-sympathetic ganglia (yellow arrow) appear than around the AVNA (B). The SNA shows a larger vessel diameter (C), and in both arteries, nerve fascicles (> 10 μm) are predominantly sympathetic (blue) (D) (Table 2). These sympathetic nerve fascicles are thicker than non-sympathetic ones (orange) (E) (Table 2). PGP9.5, protein gene product 9.5. SMA, smooth muscle actin. TH, tyrosine hydroxylase. Scale bars are 1 mm (A, B)

Table 2:

Histological characteristics of the atrioventricular and sinus node arteries

Variable Atrioventricular node artery
(n = 5)		 Sinus node artery
(n = 5)		 p value
Age (years) 41.6 ± 6.4 44.8 ± 4.4 0.38
Men, n (%) 4 (80) 4 (80) -
Body height (cm) 170.6 ± 10.6 168.7 ± 7.1 0.75
Body weight (kg) 103.0 ± 23.2 100.3 ± 52.6 0.92
Body mass index (kg/m2) 35.7 ± 9.8 34.8 ± 15.9 0.92
Left ventricular ejection fraction (%) 42.5 ± 24.4 50.0 ± 21.7 0.64
Vessel diameter (μm) 658.4 ± 190.5 1159.5 ± 175.5 < 0.01
Number of the nerve fascicles (> 10 μm) 5.5 ± 3.1 13.5 ± 4.2 < 0.01
Number of the sympathetic nerve fascicles (> 10 μm) 5.1 ± 3.2 11.2 ± 3.2 0.01
Number of the non-sympathetic nerve fascicles (> 10 μm) 0.4 ± 0.7 2.4 ± 1.0 < 0.01
Sympathetic/non-sympathetic nerve fascicle ratio (%) 92.6 ± 12.7 83.3 ± 3.3 0.12
Diamter of the nerve fascicles (μm) 28.8 ± 10.9 42.9 ± 8.9 0.04
Diamter of the sympathetic nerve fascicles (μm) 30.5 ± 12.7 47.5 ± 10.3 0.04
Diamter of the non-sympathetic nerve fascicles (μm) 15.5 ± 0.9 18.9 ± 2.9 0.17
Number of the ganglionated plexuses (> 6000 μm2) 0.8 ± 1.3 6.5 ± 4.4 0.02
Open in a new tab

The whole-mount staining of the compact AVN revealed a fine meshwork of both sympathetic and non-sympathetic nerve fibers (axons) intermingled with the specialized nodal myocardium and connective tissue arranged in a disarrayed fashion (Figure 6, Supplementary movie 1). This fine meshwork within the AVN was consistent across all seven IPS samples and exhibited identical morphological features to those observed in the sinus node (Figure 6). However, further quantitative analysis using histological sections confirmed a modest sympathetic predominance within the fine meshwork of the AVN, in contrast to the parasympathetic predominance found in the sinus node (Figure 6).29

Figure 6: Histological characteristics of the atrioventricular and sinus node.

Figure 6:

Open in a new tab

A fine, dense meshwork pattern of thin nerve fibers within both the atrioventricular (A, B, Supplementary Movie 1) and sinus (C, D) nodes was visualized using tissue-cleared whole-mount samples labeled with protein gene product (PGP) 9.5. Notably, no thick sympathetic nerve fascicles are observed. The fine meshwork pattern is consistent with the randomly aligned specialized thin nodal myocardial cells and collagen fibers within both atrioventricular (E) and sinus (H) nodes (Masson’s trichrome staining). Standard slides of the atrioventricular (F, G) and sinus nodes (I, J) are also stained using enzyme-based immunohistochemistry for vesicular acetylcholine transporter (VAChT) (F, I) and tyrosine hydroxylase (TH) (G, J). Consistent with findings from whole-mount samples (B, D), both nodal regions exhibit rich nerve fibers. Quantitative analysis demonstrates that the sinus node shows parasympathetic predominance, whereas the atrioventricular node shows sympathetic predominance (K, L). Panel L is adapted and modified from a previously published figure.29 Scale bars are 500 μm (B, D) and 50 μm (E-J).

Discussion

Previous anatomical descriptions of cardiac ganglionated plexuses related to the AVN, were not always obtained from the human heart, often being described with simple two-dimensional schematic illustrations.14,20,30 The three-dimensional anatomy of the human AVN in association with its surrounding structures, including the IPS, coronary sinus, and central fibrous body is complex.9–12 Therefore, it has been difficult to understand the three-dimensional distribution of the ganglia in the setting of the intricate gross anatomy of the human AVN, which is necessary to improve/develop a neuromodulatory approaches to the AVN. This study, for the first time by using whole-mount tissue clearing method of the IPS and AVN, provided comprehensive topographic characteristics of innervation of the AVN via the IPS. We identified four key features; (1) in the anterior fat pad of the IPS, thick sympathetic fascicles preferentially coursed towards the AVN along the AVN artery, whereas non-sympathetic fascicles approached the AVN without a consistent association with the AVN artery, and these fascicles approached the AVN from the superior and posterior aspects, (2) the ganglion atrioventriculare described in porcine hearts14 was barely observed in our human hearts in the identical location at the floor of the triangle of Koch, (3) non-sympathetic ganglia and nerve fascicles predominated in the posterior fat pad of the IPS, and (4) the compact AVN itself was filled with a fine mixed three-dimensional meshwork of sympathetic and non-sympathetic fibers.

Anterior fat pad of the IPS

In the porcine heart, the ganglion atrioventriculare, located immediately beneath the AVN at the floor of the triangle of Koch, has been shown to send nerve branches along the AVN artery towards the AVN and the bundle of His.14 With the whole-mount tissue-clearing technique, we confirmed the rare existence of such ganglia at the floor of the triangle of Koch of the human heart as indicated by Crick et al.17,19 with histological sections. In the anterior fat pad of the IPS in the human, thicker sympathetic fascicles ascend towards the AVN along the AVN artery (Figure 2). In contrast, non-sympathetic fascicles converge on the AVN without a consistent relationship to the AVN artery, and they included fascicles extending from the superior and posterior aspects of the anterior fat pads (Figure 2). It suggests that these non-sympathetic components originate not from the anterior fat pads but from the posterior fat pads behind the coronary sinus. Although it requires further investigation, these components are likely to travel around the coronary sinus to reach the anterior fat pad towards the AVN. Prior human histology has reported autonomic fibers around the AVN predominantly in perivascular regions, with extensions among adjacent nodal cells.16,21 Our whole-mount approach extended these observations by providing meso-scale, three-dimensional topography of innervation to the AVN within the IPS. This clear difference between porcine and human hearts, regarding the existence/absence of the ganglion atrioventriculare within the floor of the triangle of Koch, may be associated with the difference that the bundle of His in ungulate hearts is accompanied by highly recognizable thicker nerve fascicles.31 On the other hand, in human hearts, any recognizable nerve fascicles do not accompany the bundle of His, except for thin nerve fibers.31 These characteristic features suggest increased innervation of the ungulate AV conduction system,19 which will need further functional and comparative investigations.

Posterior fat pad of the IPS

The posterior fat pad of the IPS, which is embedded within the Eustachian ridge posterior to the coronary sinus orifice (Figure 1), contains abundant non-sympathetic ganglia and more diffuse nerve fascicles that are thicker than those in the anterior pad (Figure 4, Table 1). Identical to the ganglion atrioventriculare in porcine hearts, these non-sympathetic ganglionated plexuses found within the posterior fat pad potentially work as a local parasympathetic control hub adjusting AVN function. The epicardial side of the posterior fat pad investigated in this study is identical to the region that has been referred to as the posteromedial left atrial ganglionated plexuses in humans,32 or the inferior vena cava–inferior atrial ganglionated plexus in animals,33–35 defined as the epicardial fat located at the inferior vena cava–left atrial junction, posterior to the coronary sinus orifice. The previous studies focusing on these ganglionated plexuses provide a functional counterpart to the current morphological finding. Stimulation or ablation of these intrinsic ganglionated plexuses in dogs and humans has been shown to be associated with exacerbation or recovery of functional AV block, respectively.36–38 In addition to these ganglionated plexuses in the epicardial surface of the posterior fat pad, multiple non-sympathetic ganglia were found within the body of the fat pad (Figure 4). This finding will be clinically relevant for finding additional targets for neuromodulatory intervention of the AVN by using an endocardial approach targeting the Eustachian ridge and its attachment to the anterior limbus of the fossa ovalis.

Comparison between the sinus node and AVN innervation

Both the sinus node and AVN exhibit a fine meshwork of autonomic nerve fibers (Figure 6). However, the sinus node showed non-sympathetic predominance and the AVN demonstrated sympathetic predominance (Figure 6). This is consistent with previous findings. The human sinus node shows 5-fold higher cholinergic nerve density than noradrenergic density,27 supported by the cholinergic fibers from the right atrial ganglionated plexuses posterior to the sinus node,20 also referred to as the ganglion sinuatriale in animals14 located within the epicardial fat at the posterior interatrial groove between the superior vena cava and the right superior pulmonary vein. Innervation of the human AVN has not been as clear as the innervation of the sinus node,27 presumably because comprehensive sampling of the human AVN deep within the heart can be more challenging. The sympathetic predominance of the AVN found in this study is consistent with the functional findings demonstrating greater susceptibility of the AVN to isoproterenol than that of the sinus node.39,40

Clinical implications

Slow pathway modification for AV nodal reentrant tachycardia commonly targets the septal isthmus, which is part of the anterior fat pad of the IPS. Given that thick sympathetic and thin non-sympathetic nerve fascicles are distributed within the anterior fat pad, slow pathway modification may actually ablate or stimulate these autonomic nerve components, or it can inadvertently damage the AVN artery itself.12 These factors could be related to the junctional beats or AV block observed during and after the procedure.41,42

Although the ganglion atrioventriculare was not identified in humans, in rare settings, Bezold–Jarisch–like responses have been reported during ablation at the anterior fat pad of the IPS, including the left posteroseptal accessory pathway ablation via the coronary sinus43 and ablation targeting the premature ventricular contraction originating from the inferoseptal process of the left ventricle.44 The mechanism was considered an activation of afferent fibers that could trigger a vagal reflex. Therefore, a sensory component may occasionally reside within the anterior fat pad, which was not assessed in detail in the present study.

On the other hand, radiofrequency catheter ablation within the triangle of Koch has been known to induce an immediate increase in sinus rate likely due to parasympathetic denervation controlling the sinus node function.45 Further studies of human samples are necessary to delineate potential complicated efferent/afferent communications between the ganglionated plexuses within the posterior fat pad and fat pad around the sinus node.27

The posterior fat pad within the Eustachian ridge attaching to the anterior limbus of the fossa ovalis contains abundant non-sympathetic components within the relatively limited amount of epicardial fat (Figure 4). In addition to the epicardial approach,46,47 development of ablation in this region via the endocardial approach could potentially achieve selective parasympathetic denervation in cases of vagally mediated functional AV block.

Compared to the AVN artery, the sinus node artery is thicker and is accompanied by more and recognizable nerve fascicles. Reports of transient sinus node dysfunction after ablation around the sinoatrial node artery during atrial fibrillation ablation procedures may reflect inadvertent damage to perivascular sympathetic nerves or stimulation of non-sympathetic nerves and/or thermal injury to the sinus node artery itself.48,49 These periarterial autonomic nerve components can be explored as potentially attractive targets for neuromodulation of sinus node and AVN function if we could develop specific devices/approaches to selectively affect these neural components without damaging the arteries.

Limitations

This study has several limitations. First, the sample size was limited to seven cases. However, the principal anatomical features of AVN innervation were consistently observed across all specimens. Second, we were not able to trace nerve fibers originating from the non-sympathetic ganglionated plexuses within the posterior fat pad to the AVN, as we separated the anterior and posterior fat pads. Future research of the human AVN is required to discern such direct connection or communication between the anterior and posterior fat pads, presumably around the coronary sinus orifice. This will be feasible by sampling the posterior fat pad and the AVN en-bloc. Third, future studies incorporating parasympathetic and afferent markers on the whole-mount sample could further refine three-dimensional neurochemical mapping of the inferior pyramidal space. Last, functional validation of the present anatomical findings, including stimulation or ablation studies of the human inferior pyramidal space is beyond the scope of the present work. Therefore, at this point, clinical implication of the present anatomical findings to develop potential target for more selective neuromodulation is only ‘hypothesis generating’. Future human electrophysiological studies are required to assess these hypotheses and to prove the physiological significance of our current findings.

Conclusions

This study provides the first comprehensive meso-scale mapping of human AVN innervation within the IPS. The region-specific innervation pattern of the human AVN elaborated in this study will provide a structural basis to further understand autonomic regulation of the human AVN and may help guide selective neuromodulation strategies for human conduction system disorders.

Supplementary Material

014978_-_Supplemental_Material
014978_-_Video_1

Supplemental Video 1

What is Known

  • The human atrioventricular (AV) node is more densely innervated than the working myocardium with sympathetic, parasympathetic, and afferent fibers with regional variation.

  • Fine and timely adjustment of the human AV conduction is regulated by the autonomic nervous system.

What the Study Adds

  • In the anterior inferior pyramidal space (IPS), the thick sympathetic fascicles distribute to the AV node along the AV node artery, whereas the non-sympathetic fascicles approach the AV node without a consistent relationship with the AV node artery.

  • Non-sympathetic ganglia are predominantly found in the posterior IPS within the Eustachian ridge compared to the anterior IPS extending at the floor of the triangle of Koch, and the compact AV node is devoid of any ganglia.

  • The compact AV node contains a fine and dense three-dimensional meshwork of thin sympathetic and non-sympathetic fibers.

Acknowledgments:

We hereby express our thanks for the cooperation of OneLegacy and all the organ and tissue donors and their families for giving the gift of life and the gift of knowledge by their generous donations. We also thank the Amara Yad Project for supporting this work. We greatly appreciate all the staff members in Translational Pathology Core Laboratory at UCLA for their assistance in histological preparation. We deeply appreciate our Research Operations Managers, Ms. Hunter N. Strause and Ms. Amiksha S. Gandhi for their dedication and support for our projects.

Sources of Funding:

This work was made possible by support from NIH grant P01 HL164311 and Leducq Foundation Translational Network Award 23CVD04 to KS, American Heart Association Career Development Award 23CDA1039446 to PH, and the Amara Yad Project: https://www.uclahealth.org/medical-services/heart/arrhythmia/about-us/amara-yad-project

Nonstandard Abbreviations and Acronyms

ANOVA

analysis of variance

AV

atrioventricular

AVN

atrioventricular node

iDISCO+

modified immunolabeling-enabled three-dimensional imaging of solvent-cleared organs

IPS

inferior pyramidal space

TH

tyrosine hydroxylase

VAChT

vesicular acetylcholine transporter

Footnotes

Disclosures: None

References:

  • 1.Billette J, Tadros R. An integrated overview of AV node physiology. Pacing Clin Electrophysiol. 2019;42:805–820. [DOI] [PubMed] [Google Scholar]
  • 2.Rinkema LE, Thomas JX Jr., Randall WC. Effects of individual cardiac nerve stimulation on atrioventricular conduction. J Auton Nerv Syst. 1982;5:357–371. [DOI] [PubMed] [Google Scholar]
  • 3.Ardell JL, Randall WC. Selective vagal innervation of sinoatrial and atrioventricular nodes in canine heart. Am J Physiol. 1986;251:H764–773. [DOI] [PubMed] [Google Scholar]
  • 4.Prystowsky EN. The effects of slow channel blockers and beta blockers on atrioventricular nodal conduction. J Clin Pharmacol. 1988;28:6–21. [DOI] [PubMed] [Google Scholar]
  • 5.Armour JA. Cardiac neuronal hierarchy in health and disease. Am J Physiol Regul Integr Comp Physiol. 2004;287:R262–271. [DOI] [PubMed] [Google Scholar]
  • 6.Dohadwala M, Kamili F, Estes NMR, Homoud M. Atrioventricular block and pause-dependent torsade de pointes. HeartRhythm Case Rep. 2017;3:115–119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Tompkins JD, Buckley U, Salavatian S, Shivkumar K, Ardell JL. Vagally-mediated heart block after myocardial infarction associated with plasticity of epicardial neurons controlling the atrioventricular node. Front Synaptic Neurosci. 2022;14:960458. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Ryzhii M, Ryzhii E. Atrioventricular nodal reentrant tachycardia onset, sustainability, and spontaneous termination in rabbit atrioventricular node model with autonomic nervous system control. Front Physiol. 2024;15:1529426. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Mori S, Fukuzawa K, Takaya T, Takamine S, Ito T, Fujiwara S, Nishii T, Kono AK, Yoshida A, Hirata K. Clinical Structural Anatomy of the Inferior Pyramidal Space Reconstructed Within the Cardiac Contour Using Multidetector-Row Computed Tomography. J Cardiovasc Electrophysiol. 2015;26:705–712. [DOI] [PubMed] [Google Scholar]
  • 10.Mori S, Nishii T, Takaya T, Kashio K, Kasamatsu A, Takamine S, Ito T, Fujiwara S, Kono AK, Hirata K. Clinical structural anatomy of the inferior pyramidal space reconstructed from the living heart: Three-dimensional visualization using multidetector-row computed tomography. Clin Anat. 2015;28:878–887. [DOI] [PubMed] [Google Scholar]
  • 11.Mori S, Hanna P, Bhatt RV, Shivkumar K. The Atrioventricular Bundle: A Sesquicentennial Tribute to Professor Sunao Tawara. J Am Coll Cardiol Clin Electrophysiol. 2023;9:444–447. [DOI] [PubMed] [Google Scholar]
  • 12.Esrailian AJ, Mori S, Sato T, Izawa Y, Shivkumar K. Understanding Cardiac Anatomy and Imaging to Improve Safety of Procedures: The Atrioventricular Node Artery. J Am Coll Cardiol Case Rep. 2025;30:102753. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Thaemert JC. Atrioventricular node innervation in ultrastructural three dimensions. Am J Anat. 1970;128:239–263. [DOI] [PubMed] [Google Scholar]
  • 14.Bojsen-Moller F, Tranum-Jensen J. Whole-mount demonstration of cholinesterase-containing nerves in the right atrial wall, nodal tissue, and atrioventricular bundle of the pig heart. J Anat. 1971;108:375–86. [PMC free article] [PubMed] [Google Scholar]
  • 15.Kent KM, Epstein SE, Cooper T, Jacobowitz DM. Cholinergic innervation of the canine and human ventricular conducting system. Anatomic and electrophysiologic correlations. Circulation. 1974;50:948–955. [DOI] [PubMed] [Google Scholar]
  • 16.Chow LT, Chow SS, Anderson RH, Gosling JA. Innervation of the human cardiac conduction system at birth. Br Heart J. 1993;69:430–435. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Crick SJ, Wharton J, Sheppard MN, Royston D, Yacoub MH, Anderson RH, Polak JM. Innervation of the human cardiac conduction system. A quantitative immunohistochemical and histochemical study. Circulation. 1994;89:1697–1708. [DOI] [PubMed] [Google Scholar]
  • 18.Swindle M, Smith AC. Comparative anatomy and physiology of the pig. Scandinavian Journal of Laboratory Animal Science. 1998;25:11–21. [Google Scholar]
  • 19.Crick SJ, Sheppard MN, Ho SY, Anderson RH. Localisation and quantitation of autonomic innervation in the porcine heart I: conduction system. J Anat. 1999;195:341–357. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Pauza DH, Skripka V, Pauziene N, Stropus R. Morphology, distribution, and variability of the epicardiac neural ganglionated subplexuses in the human heart. Anat Rec. 2000;259:353–382. [DOI] [PubMed] [Google Scholar]
  • 21.Chow LT, Chow SS, Anderson RH, Gosling JA. Autonomic innervation of the human cardiac conduction system: changes from infancy to senility--an immunohistochemical and histochemical analysis. Anat Rec. 2001;264:169–182. [DOI] [PubMed] [Google Scholar]
  • 22.Saburkina I, Gukauskiene L, Rysevaite K, Brack KE, Pauza AG, Pauziene N, Pauza DH. Morphological pattern of intrinsic nerve plexus distributed on the rabbit heart and interatrial septum. J Anat. 2014;224:583–593. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Lelovas PP, Kostomitsopoulos NG, Xanthos TT. A comparative anatomic and physiologic overview of the porcine heart. J Am Assoc Lab Anim Sci. 2014;53:432–438. [PMC free article] [PubMed] [Google Scholar]
  • 24.Aksu T, Gupta D, Pauza DH. Anatomy and Physiology of Intrinsic Cardiac Autonomic Nervous System: Da Vinci Anatomy Card #2. J Am Coll Cardiol Case Rep. 2021;3:625–629. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Aksu T, Gopinathannair R, Gupta D, Pauza DH. Intrinsic cardiac autonomic nervous system: What do clinical electrophysiologists need to know about the “heart brain”? J Cardiovasc Electrophysiol. 2021;32:1737–1747. [DOI] [PubMed] [Google Scholar]
  • 26.Hanna P, Mori S, Sato T, Xu S. Pipeline for Multi-Scale Three-Dimensional Anatomic Study of the Human Heart. J Vis Exp. 2024;208. [DOI] [PubMed] [Google Scholar]
  • 27.Hanna P, Dacey MJ, Brennan J, Moss A, Robbins S, Achanta S, Biscola NP, Swid MA, Rajendran PS, Mori S, et al. Innervation and Neuronal Control of the Mammalian Sinoatrial Node a Comprehensive Atlas. Circ Res. 2021;128:1279–1296. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Munro BH. Manual of Histologic Staining Methods of the Armed Forces Institute of Pathology. Pathology. 1971;3:249. [Google Scholar]
  • 29.Hanna P, Hoover DB, Kirkland LG, Smith EH, Poston MD, Peirce SG, Garbe CG, Phillips TK, Cha S, Mori S, et al. Noradrenergic and cholinergic innervation of the normal human heart and changes associated with cardiomyopathy. Anat Rec (Hoboken). 2026;309:417–450. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Armour JA, Murphy DA, Yuan BX, Macdonald S, Hopkins DA. Gross and microscopic anatomy of the human intrinsic cardiac nervous system. Anat Rec. 1997;247:289–298. [DOI] [PubMed] [Google Scholar]
  • 31.Tawara S. Das Reizleitungssystem des Säugetierherzens: eine anatomisch-histologische Studie über das Atrioventrikularbündel und die Purkinjeschen Fäden. Jena, Germany: Gustav Fischer; 1906. (English translation sponsored by Dr. Robert H. Anderson and translated by Dr Kozo Suma and Dr Munehiro Shimada. Imperial College Press; 2000). [Google Scholar]
  • 32.Smith RB. The occurrence and location of intrinsic cardiac ganglia and nerve plexuses in the human neonate. Anat Rec. 1971;169:33–40. [DOI] [PubMed] [Google Scholar]
  • 33.Armour JA. Potential clinical relevance of the ‘little brain’ on the mammalian heart. Exp Physiol. 2008;93:165–176. [DOI] [PubMed] [Google Scholar]
  • 34.Cardinal R, Pagé P, Vermeulen M, Ardell JL, Armour JA. Spatially divergent cardiac responses to nicotinic stimulation of ganglionated plexus neurons in the canine heart. Auton Neurosci. 2009;145:55–62. [DOI] [PubMed] [Google Scholar]
  • 35.Chung WH, Do D, Erbay MI, Ajijola OA. Cardioneural ablation: Toward achieving uniformity in nomenclature, procedural approaches, and outcome measures. Heart Rhythm. 2026;23:e75–e83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Chiou CW, Eble JN, Zipes DP. Efferent vagal innervation of the canine atria and sinus and atrioventricular nodes. The third fat pad. Circulation. 1997;95:2573–2584. [DOI] [PubMed] [Google Scholar]
  • 37.Ascione C, Benabou L, Hocini M, Jaïs P, Haïssaguerre M, Duchateau J. Cardioneuroablation: Don’t underestimate the posteromedial left atrial ganglionated plexus. HeartRhythm Case Rep. 2023;9:67–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Yoneda F, Shizuta S, Makiyama T, Masunaga N, Hoshida S, Kimura T. Selective cardioneuroablation of the posteromedial left ganglionated plexus for drug-resistant swallow syncope with functional atrioventricular block. HeartRhythm Case Rep. 2023;9:513–517. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Vargas G, Akhtar M, Damato AN. Electrophysiologic effects of isoproterenol on cardiac conduction system in man. Am Heart J. 1975;90:25–34. [DOI] [PubMed] [Google Scholar]
  • 40.Cossú SF, Rothman SA, Chmielewski IL, Hsia HH, Vogel RL, Miller JM, Buxton AE. The effects of isoproterenol on the cardiac conduction system: site-specific dose dependence. J Cardiovasc Electrophysiol. 1997;8:847–853. [DOI] [PubMed] [Google Scholar]
  • 41.Pelargonio G, Fogel RI, Knilans TK, Prystowsky EN. Late occurrence of heart block after radiofrequency catheter ablation of the septal region: clinical follow-up and outcome. J Cardiovasc Electrophysiol. 2001;12:56–60. [DOI] [PubMed] [Google Scholar]
  • 42.Dorotan-Guevara MM, Crapanzano MS, Snyder CS. Late occurrence of transient advanced second degree atrioventricular block after successful transcatheter cryoablation of atrioventricular nodal reentry tachycardia. Case Rep Cardiol. 2012;2012:752956. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Mircea AA, Mori S, Ajijola OA, Khakpour H. Sinus arrest during radiofrequency ablation from the inferoseptal process of the left ventricle: Proposed mechanisms of an uncommon finding. HeartRhythm Case Rep. 2023;9:291–295. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Schläpfer J, Kappenberger L, Fromer M. Bezold-Jarisch-like phenomenon induced by radiofrequency ablation of a left posteroseptal accessory pathway via the coronary sinus. J Cardiovasc Electrophysiol. 1996;7:445–449. [DOI] [PubMed] [Google Scholar]
  • 45.Kocovic DZ, Harada T, Shea JB, Soroff D, Friedman PL. Alterations of heart rate and of heart rate variability after radiofrequency catheter ablation of supraventricular tachycardia. Delineation of parasympathetic pathways in the human heart. Circulation. 1993;88:1671–1681. [DOI] [PubMed] [Google Scholar]
  • 46.Alboni P, Holz A, Brignole M. Vagally mediated atrioventricular block: pathophysiology and diagnosis. Heart. 2013;99:904–908. [DOI] [PubMed] [Google Scholar]
  • 47.Aksu T, Guler TE, Bozyel S, Yalin K. Potential usage of cardioneuroablation in vagally mediated functional atrioventricular block. SAGE Open Med. 2019;7:2050312119836308. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Kitamura T, Fukamizu S, Arai K, Hojo R, Aoyama Y, Komiyama K, Sakurada H, Hiraoka M. Transient sinus node dysfunction following sinus node artery occlusion due to radiofrequency catheter ablation of the septal superior vena cava-right atrium junction. J Electrocardiol. 2016;49:18–22. [DOI] [PubMed] [Google Scholar]
  • 49.Aerts L, den Uijl DW, Luermans JGL, Maesen B. Temporary sinus node dysfunction following ablation of atrial fibrillation in patients with aberrancy of the sinus node artery: a case series. Interdiscip Cardiovasc Thorac Surg. 2024;39:ivae135. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

014978_-_Supplemental_Material
NIHMS2197761-supplement-014978_-_Supplemental_Material.pdf (146KB, pdf)
014978_-_Video_1
NIHMS2197761-supplement-014978_-_Video_1.mpg (10.5MB, mpg)

Data Availability Statement

The data underlying this article will be shared on reasonable request to the corresponding author.

RESOURCES