Human Pulmonary Vein Myocardial Sleeve Autonomic Neural Density and Cardiovascular Mortality

Abstract

Myocardial sleeves around pulmonary veins (PVs) are highly innervated structures with heterogeneous morphological and electrophysiological characteristics. Autonomic nerve dysfunction in the myocardium may be associated with an increased risk of cardiovascular morbidity and mortality. This article studied autonomic neural remodeling in myocardial sleeves around PVs and atrial-PV ostia with immunohistochemical and morphometric methods with clinicopathological correlations. PVs were collected from 37 and atrial-PV ostia from 17 human autopsy hearts. Immunohistochemical analysis was performed using antibodies against tyrosine hydroxylase (TH), choline acetyltransferase (CHAT), and growth-associated protein 43 (GAP43). In the PV cohort, subjects with immediate cardiovascular cause of death had significantly decreased sympathetic nerve density in fibro-fatty tissue vs those with non-cardiovascular cause of death (1624.53 vs 2522.05 µm²/mm², p=0.038). In the atrial-PV ostia cohort, parasympathetic nerve density in myocardial sleeves was significantly increased in subjects with underlying cardiovascular cause of death (19.48 µm²/mm²) than subjects with underlying non-cardiovascular cause of death with no parasympathetic nerves detected (p=0.034). Neural growth regionally varied in sympathetic nerves and was present in most of the parasympathetic nerves. Heterogeneous autonomic nerve distribution and growth around PVs and atrial-PV ostia might play a role in cardiovascular morbidity and mortality. No association in nerve density was found with atrial fibrillation.

Keywords

autonomic nervous system cardiovascular mortality morphometry myocardial sleeves nerve density pulmonary veins

Introduction

Pulmonary veins (PVs) and adjacent atrial-PV ostia participate in the regulation of atrial fibrillation.^1–3 Myocardial sleeves are extensions of the atrial myocardium that reach proximal portions of PVs,⁴ and they are morphologically heterogeneous.^3,5 In addition, PVs are associated with variable responses to autonomic activity and unique electrophysiological characteristics.⁶

The autonomic nervous system (ANS) regulates heart activity by stimulatory and inhibitory neurotransmitters. In pathological conditions, such as congestive heart failure and myocardial ischemia, myocardial remodeling and dysregulation of ANS increase the risk for lethal arrhythmias and sudden cardiac death.^7–13 The process is usually associated with structural, electrophysiological, and neural remodeling of the myocardium.^7,14,15

Uneven regional distribution of the autonomic nerves and various patterns of innervation were studied in caval veins,¹⁶ PVs, and atrial-PV ostia.^17–21 However, the role of myocardial sleeves around PVs and in atrial-PV ostia in the pathogenesis of cardiovascular (CVS) mortality and various heart diseases awaits more studies.

In our recent study, we assessed nerve density in caval veins and provided evidence of sympathetic denervation to be associated with CVS mortality.¹⁶ We hypothesized that similar mechanisms may apply also in the PVs and atrial-PV ostia area. Thus, in this study, we scoped on autonomic neural remodeling in and around myocardial sleeves in PVs and atrial-PV ostia using immunohistochemistry and computerized morphometry to measure the distribution of ANS nerves and ganglia in myocardial sleeves and surrounding fibro-fatty tissue in PVs and atrial-PV ostia of autopsy hearts.

Materials and Methods

Characterization of Pulmonary Vein Cohort

PVs were collected from 37 autopsies [mean age ± standard deviation (SD), 68±9 years; M: F ratio, 22:15; mean heart weight ± SD, 484 ± 140 g]. The most common immediate cause of death was CVS (n=12), followed by respiratory (n=9), infectious (n=6), brain death (n=4), and miscellaneous in the rest of the cases (n=6). Out of 12 subjects with documented immediate CVS cause of death, 8 subjects (66.6%) were diagnosed with heart failure. Underlying CVS cause of death was diagnosed in 17 cases and malignancies in 9 cases. Other underlying causes of death included central nervous system diseases (n=3), chronic obstructive pulmonary disease (n=1), congenital connective tissue disorder (n=1), chronic tubulointerstitial nephritis (n=1), diabetes mellitus (n=1), and anemia (n=1). According to the autopsy referral, 14 subjects were diagnosed with atrial fibrillation (AF) and 23 subjects had documented sinus rhythm (SR). Detailed clinical information of the PV cohort is shown in Tables 1 and 2.

Table 1.

Clinical Data in PV Cohort.

ID No.	Age	Sex	Heart Weight (g)	Heart Rhythm	Underlying Cause of Death		Immediate Cause of Death
1	86	Female	510	SR	ASVD	A	Acute on chronic HF	A
2	70	Female	470	SR	Multiple myeloma	B	HF	A
3	57	Male	540	SR	Liver cholangiocarcinoma	B	Respiratory failure	B
4	62	Female	620	SR	Congenital connective tissue disorder	B	HF	A
5	65	Female	530	SR	ASVD	A	HF	A
6	56	Female	180	SR	Cervical carcinoma	B	Bronchopneumonia	B
7	73	Female	380	SR	Choledocholithiasis	B	Respiratory failure	B
8	70	Female	390	SR	Chronic tubular nephritis	B	Sepsis	B
9	63	Male	620	AF	Mitral stenosis	A	HF	A
10	70	Male	630	AF	ASVD	A	Acute on chronic HF	A
11	57	Female	490	AF	Multiple myeloma	B	Respiratory failure	B
12	42	Male	870	AF	Mixed aortic disease	A	Extreme obesity	B
13	60	Male	600	AF	Abdominal aortic aneurysm	A	Pulmonary artery embolism	B
14	81	Male	560	AF	ASVD	A	Brain death	B
15	74	Female	600	SR	Acute pancreatitis	B	Cardiorespiratory insufficiency	A
16	77	Male	370	AF	Renal cell carcinoma	B	Bronchopneumonia	B
17	61	Male	550	SR	Renal papillary carcinoma	B	Right respiratory failure	B
18	73	Male	460	SR	Diabetes mellitus	B	Cardiorespiratory insufficiency	A
19	70	Male	510	SR	Myelodysplastic syndrome	B	Respiratory failure	B
20	79	Male	400	SR	ASVD	A	Posthemorrhagic shock	B
21	55	Male	430	SR	Anemia	B	Lung edema	B
22	75	Female	380	AF	ASVD	A	Brain death	B
23	66	Male	600	SR	ASVD	A	Myocardial infarction	A
24	72	Female	320	SR	Bronchioalveolar carcinoma	B	Respiratory failure	B
25	77	Male	500	SR	ASVD	A	Acute on chronic HF	A
26	71	Male	770	AF	ASVD	A	Multiorgan failure	B
27	61	Female	620	SR	Infective endocarditis	A	HF	A
28	57	Male	340	SR	Meningomyelitis	B	Brain death	B
29	69	Female	290	SR	Brain hemorrhage	B	Brain death	B
30	69	Male	400	SR	Postchemotherapy status	B	Sepsis	B
31	78	Male	460	AF	ASVD	A	Pulmonary fat embolism	B
32	67	Male	410	AF	Gastric ulcer disease	B	Posthemorrhagic shock	B
33	75	Male	470	SR	ASVD	A	Bronchopneumonia	B
34	73	Male	280	AF	COPD	B	Respiratory failure	B
35	56	Male	550	SR	Arterial hypertension	A	Food aspiration	B
36	76	Female	540	AF	ASVD	A	Sepsis	B
37	71	Female	280	AF	Epilepsy	B	Cardiorespiratory insufficiency	A

Abbreviations: PV, pulmonary vein; SR, sinus rhythm; ASVD, atherosclerotic vascular disease; A, cardiovascular cause of death; HF, heart failure; B, non-cardiovascular cause of death; AF, atrial fibrillation; COPD, chronic obstructive pulmonary disease.

Table 2.

Characteristics of Subjects in PV Cohort.

PV Cohort	Immediate CVS Death	Immediate Non-CVS Death	Underlying CVS Death	Underlying Non-CVS Death
Subjects (n)	12	25	17	20
Age (years ± SD)	70±7	67 ± 10	70 ± 11	67 ± 7
Female (%)	58	32	29	50
Heart weight (g ± SD)	537 ± 103	459 ± 150	565 ± 122	416 ± 116
AF, n (%)	3 (25)	11 (44)	9 (53)	5 (25)

Abbreviations: PV, pulmonary vein; CVS, cardiovascular; AF, atrial fibrillation; SD, standard deviation.

Characterization of Atrial-PV Vein Ostia Cohort

Atrial-PV ostia were collected from 17 autopsies (mean age ± SD, 63±15 years; M:F ratio, 10:7; mean heart weight ± SD, 484±139 g). The most common immediate cause of death was CVS in 8 cases, followed by infectious in 3 cases, respiratory in 2 cases, gastrointestinal and brain death each in 1 case, and miscellaneous in 2 cases. The underlying CVS cause of death was diagnosed in 11, malignant in 1, gastrointestinal in 1, hematological in 1, and miscellaneous in 3 subjects. Out of 11 subjects with documented underlying CVS cause of death, 8 subjects (72%) were diagnosed with systemic atherosclerosis. According to the autopsy referral, 7 subjects had documented AF and 10 subjects had documented SR. More details on the atrial-PV ostia cohort are shown in Tables 3 and 4.

Table 3.

Clinical Data in Atrial-PV Ostia Cohort.

ID No.	Age	Sex	Heart Weight (g)	Heart Rhythm	Underlying Cause of Death		Immediate Cause of Death
1	63	Female	480	AF	Mixed mitral valve disease	A	HF	A
2	57	Male	600	AF	ASVD	A	Respiratory failure	B
3	71	Male	450	SR	ASVD	A	Acute myocardial infarction	A
4	79	Male	520	SR	Aortic stenosis	A	HF	A
5	70	Male	770	SR	ASVD	A	Myocardial infarction	A
6	68	Male	670	AF	Schizophrenia	B	Pulmonary embolism	B
7	79	Male	390	AF	ASVD	A	HF	A
8	64	Male	450	SR	Leukemia	B	Bronchopneumonia	B
9	69	Female	430	AF	ASVD	A	Cardiogenic shock	A
10	76	Female	360	SR	Peptic ulcer	B	Hemorrhagic shock	B
11	54	Male	610	SR	ASVD	A	HF	A
12	39	Female	290	SR	Chronic alcoholism	B	Hemorrhagic shock	B
13	33	Female	290	SR	Chronic alcoholism	B	Respiratory failure	B
14	77	Female	620	AF	ASVD	A	HF	A
15	46	Female	320	SR	Vascular malformation	A	Brain death	B
16	80	Male	410	AF	ASVD	A	Urosepsis	B
17	52	Male	570	SR	Laryngeal carcinoma	B	Bronchopneumonia	B

Abbreviations: PV, pulmonary vein; AF, atrial fibrillation; A, cardiovascular cause of death; HF, heart failure; ASVD, atherosclerotic vascular disease; B, non-cardiovascular cause of death; SR, sinus rhythm.

Table 4.

Characteristics of Subjects in Atrial-PV Ostia Cohort.

Atrial-PV Ostia Cohort	Immediate CVS Death	Immediate Non-CVS Death	Underlying CVS Death	Underlying Non-CVS Death
Subjects (n)	8	9	11	6
Age (years ± SD)	70 ± 9	57 ± 16	68 ± 11	55 ± 17
Female (%)	38	44	36	50
Heart weight (g ± SD)	534 ± 126	440 ± 142	509 ± 130	438 ± 156
AF, n (%)	4 (50)	3 (33)	6 (55)	1 (17)

Abbreviations: PV, pulmonary vein; CVS, cardiovascular; AF, atrial fibrillation; SD, standard deviation.

Sample Collection

PVs and atrial-PV ostia were collected from adult autopsy hearts at the Fingerland Department of Pathology, Charles University Hospital, Hradec Králové, Czech Republic.³

After the excision of the heart, PVs were separated from the left atrium at the level of the atrial-PV ostium, which was macroscopically determined. All PVs were longitudinally cut into pieces, spread, fixed in 10% buffered formalin, and processed into paraffin blocks. The collection technique has been described in detail previously.³ Atrial-PV ostia were collected and processed correspondingly. The samples consisted of 94 PVs (24 right upper, 25 right lower, 24 left upper, 19 left lower, 1 right common, and 1 left common PVs) in the PV cohort. The atrial-PV ostia cohort included ostia samples from 51 PVs (11 right upper, 10 right lower, 9 left upper, 11 left lower, 4 right common, and 6 left common PVs). In total, 125 samples of PV tissue and 90 samples of atrial-PV ostia tissue were included in this study. The presence of myocardial sleeves was microscopically verified in all included samples.

Immunohistochemical Analysis

The collected tissue samples were cut into 5-µm-thick sections and stained as described previously.¹⁶ In brief, the Ventana Automatic System (Ventana Medical Systems; Tucson, AZ) was used for immunohistochemical staining. Anti-TH antibody (dilution 1:100, AB152, Chemicon, Merck KGaA, Darmstadt, Germany) was used to detect sympathetic nerves and ganglia, anti-CHAT antibody (dilution 1:300, AB143, Chemicon, Merck KGaA) was used to label parasympathetic nerves and ganglia, and GAP43 antibody (dilution 1:100, AB5220, Chemicon, Merck KGaA) was used to detect neural growth.

Morphometric Analysis

The samples were scanned (NanoZoomer-XR; Hamamatsu Photonics, Hamamatsu, Japan) at 40× magnification. Bioimage analysis software was used for histomorphometry (QuPath; Queens University, Belfast, Northern Ireland²²). Nerve density (µm²/mm²) was quantified as in the previous study.¹⁶ Briefly, the nerves were detected in the whole slide images, which were divided into two regions: myocardial sleeves and fibro-fatty tissue outside the sleeves. Total areas (mm²) of the whole section, myocardial sleeves, and fibro-fatty tissue were measured in all samples. The nerve area (µm²) was measured by manually delimiting positive nerves and ganglia at 20× magnification separately for TH-stained, CHAT-stained, and GAP43-stained sections.

To evaluate the subset of nerves showing neural growth (regenerative activity) in TH- and CHAT-positive nerves, the percentages of nerves stained positively with GAP43 antibodies were calculated in TH- and CHAT-stained samples. The nerves stained for both TH and GAP43, and CHAT and GAP43, respectively, were manually selected in all three analyzed areas (whole section, myocardial sleeves, and fibro-fatty tissue) within each PV and atrial-PV ostium.

Statistical Analysis

Data are presented as mean ± SD. Categorical variables were evaluated as the count and percentage. Because of the small cohort size, the Mann–Whitney non-parametric test was used for continuous variables, and the chi-square test was used for categorical analysis. An average percentage value used to express the representation of neural growth (GAP43-positive nerves) in TH- and CHAT-positive nerves was calculated from all the TH- and CHAT-stained PV samples. Statistical analysis was performed by the Statistical Package for the Social Sciences version 24.0 (SPSS Inc.; Chicago, IL). A p value of <0.05 was considered significant for these comparisons.

Ethical Statement

The study was approved by the Pirkanmaa Health Care District Ethical Committee (permission application number R15013). The collection and research use of the samples were approved by the Ethical Committee of the University Hospital, Hradec Králové, Czech Republic. The study was done in accordance with the Declaration of Helsinki.

Results

Autonomic Nervous System Characteristics and Localization

Myocardial sleeves were found in 62.8% of studied PVs (59/94) and in 90.2% of atrial-PV ostia (46/51). TH-, CHAT-, and GAP43-positive nerves were identified both in muscle sleeves and in the surrounding fibro-fatty tissue, with a predominant location in the latter (Fig. 1A and B). TH and GAP43 positivity dominated over CHAT-positive nerves in all PVs and atrial-PV ostia samples. Variably sized and shaped nerve bundles consisted of nerve fibers. Ganglia consisted of large ganglion cells and nerve fibers (Figs. 1 and 2). Nerve bundles with colocalizing TH- and CHAT-positive fibers (i.e., intermediate nerves) were evaluated according to the predominant positivity.

Quantitative Evaluation

Study Cohorts’ Characteristics

No differences were found in age, sex, and heart weight among subjects with CVS mortality and the presence of atrial fibrillation in both studied cohorts (Tables 2 and 4). In addition, no association in nerve density was found among subjects in age, sex, heart weight, heart rhythm, and CVS mortality. Topographically, our study did not show any association in nerve density gradient among individual PV and atrial-PV ostia locations. Results of nerve densities of all pooled PVs and ostia are given below.

Pulmonary Veins

The total PV nerve density of TH-positive nerves was 2090.28 ± 2607.59 µm²/mm² in the whole section, 1248.67 ± 2406.27 µm²/mm² in myocardial sleeves, and 2187.86 ± 2994.48 µm²/mm² in the fibro-fatty tissue. The total CHAT-positive nerve density was 204.56 ± 409.83 µm²/mm² in the whole slide, 65.75 ± 217.21 µm²/mm² in the myocardial sleeves, and 241.30 ± 598.71 µm²/mm² in the fibro-fatty tissue. The GAP43-positive nerve density was 1754.71 ± 2130.90 µm²/mm² in the whole section, 1514.62 ± 2973.23 µm²/mm² in the myocardial sleeves, and 1622.63 ± 2197.77 µm²/mm² in the fibro-fatty tissue.

Although the mean TH-positive nerve density of PV area was decreased in all measured regions (whole slide, myocardial sleeves, fibro-fatty tissue) in subjects with documented immediate CVS death (Table 5, Fig. 1C–F), only the fibro-fatty tissue region showed a statistically significant difference in comparison with subjects with documented immediate non-CVS cause of death (1378.5 ± 2159.68 µm²/mm² vs 2512.53 ± 2771.53 µm²/mm², p=0.064; 633.22 ± 1550.85 µm²/mm² vs 1613.77 ± 2740.45 µm²/mm², p=0.105; 1624.53 ± 3148.36 µm²/mm² vs 2522.05 ± 2874.43 µm²/mm², p=0.038, respectively). The results of nerve distribution are illustrated in Fig. 3.

Table 5.

Pulmonary Vein Cohort Autonomic Nerve Densities.

	Area	Immediate CVS Death	Immediate Non-CVS Death	P Value	Underlying CVS Death	Underlying Non-CVS Death	P Value
TH	Whole section	1378.5 ± 2159.68	2512.53 ± 2771.53	0.064	2398.22 ± 3178.98	1819.3 ± 1968.93	0.428
	Myocardial sleeves	633.22 ± 1550.85	1613.77 ± 2740.45	0.105	1444.65 ± 3183.23	1076.2 ± 1425.62	0.976
	Fibro-fatty tissue	1624.53 ± 3148.36	2522.05 ± 2874.43	0.038*	2428.47 ± 3638.68	1976.13 ± 2302.7	0.761
CHAT	Whole section	217.72 ± 397.39	196.76 ± 420.2	0.948	292.18 ± 533.96	127.46 ± 235.83	0.112
	Myocardial sleeves	20.98 ± 79.72	92.31 ± 264.57	0.053	67.29 ± 174.64	64.4 ± 250.55	0.637
	Fibro-fatty tissue	305.56 ± 664.62	203.18 ± 558.43	0.696	386.19 ± 812.81	113.81 ± 255.19	0.081
GAP43	Whole section	1330.31 ± 1548.63	2006.47 ± 2387.84	0.270	2145.99 ± 2726.03	1410.38 ± 1355.85	0.100
	Myocardial sleeves	962.65 ± 2202.73	1842.06 ± 3322.26	0.330	1847.84 ± 3724.25	1221.38 ± 2102.14	0.807
	Fibro-fatty tissue	1374.62 ± 1769.91	1769.76 ± 2418.56	0.559	1897.13 ± 2780.29	1381.08 ± 1502.27	0.361

Abbreviations: CVS, cardiovascular; TH, tyrosine hydroxylase; CHAT, choline acetyltransferase; GAP43, growth-associated protein 43.

Statistical significance.

The mean density of CHAT-positive nerves did not differ in the PV area between subjects with documented immediate CVS and non-CVS cause of death (whole section: 217.72 ± 397.39 µm²/mm² vs 196.76 ± 420.2 µm²/mm², p=0.948; myocardial sleeves: 20.98 ± 79.72 µm²/mm² vs 92.31 ± 264.57 µm²/mm², p=0.053; fibro-fatty tissue: 305.56 ± 664.62 µm²/mm² vs 203.18 ± 558.43 µm²/mm², p=0.696, respectively) (Table 5, Fig. 1G–J).

Table 5 and Fig. 1K to N show a slight decrease in GAP43-positive nerve densities of the whole section, myocardial sleeves, and fibro-fatty tissue regions between subjects with documented immediate CVS vs non-CVS cause of death, yet with no statistically significant difference (1330.31 ± 1548.63 µm²/mm² vs 2006.47 ± 2387.84 µm²/mm², p=0.270; 962.65 ± 2202.73 µm²/mm² vs 1842.06 ± 3322.26 µm²/mm², p=0.330; 1374.62 ± 1769.91 µm²/mm² vs 1769.76 ± 2418.56 µm²/mm², p=0.559, respectively).

On average, 51% of TH-positive (sympathetic) nerves in the PV whole section were also stained with anti-GAP43 antibodies, a marker of neural growth. These TH-positive nerves represented 24.26% of nerves in the myocardial sleeves and 52.67% of nerves in the fibro-fatty tissue. Similarly, CHAT-positive nerves with GAP43 activity constituted 72.19% of nerves in the whole section, 63.64% of nerves in the myocardial sleeves, and 75.45% of nerves in the fibro-fatty tissue (Fig. 4).

Atrial-PV Vein Ostia

The total mean TH-positive nerve density in the atrial-PV ostia was 1972.77 ± 2660.69 µm²/mm² in the whole section, 1302.49 ± 2107.76 µm²/mm² in the myocardial sleeves, and 1940.02 ± 2475.84 µm²/mm² in the fibro-fatty tissue. The CHAT-positive nerve density was 35.13 ± 68.14 µm²/mm² in the whole section, 11.84 ± 36.62 µm²/mm² in the myocardial sleeves, and 51.15 ± 108.03 µm²/mm² in the fibro-fatty tissue. The total GAP43-positive nerve density was 2015.67 ± 3086.62 µm²/mm² in the whole section, 1404.77 ± 2018.74 µm²/mm² in the myocardial sleeves, and 1881.56 ± 2979.04 µm²/mm² in the fibro-fatty tissue.

In the atrial-PV ostia area, CHAT-positive nerve density in the myocardial sleeves was significantly higher in subjects with documented underlying CVS cause of death (19.48 ± 45.62 µm²/mm²) vs non-CVS cause of death with no parasympathetic nerves detected (0 ± 0 µm²/mm²) (p=0.034). CHAT-positive nerve density in the whole slide and fibro-fatty tissue regions did not show a difference between subjects with documented underlying CVS and non-CVS cause of death as shown in Table 6 and Fig. 2C to F (33.87 ± 57.95 µm²/mm² vs 37.08 ± 83.12 µm²/mm², p=0.613; 41.39 ± 80.95 µm²/mm² vs 66.27±141.3 µm²/mm², p=0.839, respectively) (Fig. 5).

Table 6.

Atrial-Pulmonary Vein Ostia Cohort Nerve Densities.

	Area	Immediate CVS Death	Immediate Non-CVS Death	P Value	Underlying CVS Death	Underlying Non-CVS Death	P Value
TH	Whole section	2656.02 ± 3924.35	1531.96 ± 1238.17	0.501	2331.47 ± 3305.04	1416.78 ± 920.3	0.421
	Myocardial sleeves	1679.73 ± 2814.97	1059.11 ± 1493.24	0.630	1714.17 ± 2581.65	664.38 ± 680.05	0.421
	Fibro-fatty tissue	2425.29 ± 3584.95	1626.94 ± 1348.53	0.700	2136.82 ± 3010.26	1634.98 ± 1289.21	0.546
CHAT	Whole section	34.74 ± 50.28	35.38 ± 78.34	0.287	33.87 ± 57.95	37.08 ± 83.12	0.613
	Myocardial sleeves	16.43 ± 42.59	8.87 ± 32.6	0.072	19.48 ± 45.62	0	0.034*
	Fibro-fatty tissue	39.34 ± 59.5	58.77 ± 130.59	0.770	41.39 ± 80.95	66.27 ± 141.3	0.839
GAP43	Whole section	2521.52 ± 4665.85	1689.31 ± 1345.47	0.923	2392.16 ± 3863.55	1432.1 ± 955.47	0.546
	Myocardial sleeves	1325.42 ± 2540.51	1455.96 ± 1642.44	0.386	1718.89 ± 2469.42	917.89 ± 827.95	0.763
	Fibro-fatty tissue	2375.27 ± 4440.04	1563.03 ± 1427.68	0.847	2150.66 ± 3672.5	1464.45 ± 1325.67	0.763

Abbreviations: CVS, cardiovascular; TH, tyrosine hydroxylase; CHAT, choline acetyltransferase; GAP43, growth-associated protein 43.

Statistical significance.

Table 6 and Fig. 2G to J compare the densities of TH-positive nerves between subjects with documented underlying CVS and non-CVS cause of death showing an increase within all three regions, however without statistical significance (2331.47 ± 3305.04 µm²/mm² vs 1416.78 ± 920.3 µm²/mm², p=0.421; 1714.17 ± 2581.65 µm²/mm² vs 664.38 ± 680.05 µm²/mm², p=0.421; and 2136.82 ± 3010.26 µm²/mm² vs 1634.98 ± 1289.21 µm²/mm², p=0.546, respectively).

Similarly, the densities of GAP43-positive nerves within atrial-PV ostia were higher in the whole section, myocardial sleeves, and fibro-fatty tissue in subjects with documented underlying CVS cause of death, but the differences did not reach a statistical significance (2392.16 ± 3863.55 µm²/mm² vs 1432.1 ± 955.47 µm²/mm², p=0.546; 1718.89 ± 2469.42 µm²/mm² vs 917.89 ± 827.95 µm²/mm², p=0.763; 2150.66 ± 3672.5 µm²/mm² vs 1464.45 ± 1325.67 µm²/mm², p=0.763, respectively) (Table 6, Fig. 2K–N).

In the atrial-PV ostia area, GAP43-positive neural growth in the TH-positive nerves was detected in 45.56% in the whole section, 27.93% in the myocardial sleeves, and 55.03% in the fibro-fatty tissue. Finally, 70.85% of CHAT-positive nerves in the whole section, 78.79% of CHAT-positive nerves in the myocardial sleeves, and 77.43% of CHAT-positive nerves in the fibro-fatty tissue were concomitantly stained with anti-GAP43 antibodies.

Discussion

Main Findings

The most significant finding of this study is that myocardial sleeves around PVs and atrial-PV ostia contain heterogeneous autonomic innervation. Furthermore, autonomic nerves were associated with CVS mortality in our autopsy cohort. Sympathetic nerve density was significantly decreased in the fibro-fatty tissue surrounding myocardial sleeves of the PVs in subjects with documented immediate CVS cause of death compared with subjects with documented immediate non-CVS cause of death (p=0.038). Parasympathetic nerve density was significantly increased in the myocardial sleeves of the atrial-PV ostia in subjects with documented underlying CVS cause of death compared with subjects with documented underlying non-CVS cause of death (p=0.034). No association was found between TH-, CHAT-, and GAP43-positive nerve densities and a history of atrial fibrillation. In addition, no differences in nerve density topography among individual PVs were found when related to age, sex, heart weight, heart rhythm, and CVS mortality. The proportion of GAP43-positive neural growth in the ANS showed a similar pattern in both PV and atrial-PV ostia cohorts. GAP43-positive neural growth was observed in nearly one-fourth of the TH-positive nerves in the myocardial sleeves and in roughly half of the TH-positive nerves in the surrounding fibro-fatty tissue. Surprisingly, the majority of CHAT-positive nerves was immunoreactive to GAP43 in all the studied regions.

Autonomic Nervous System and Heart Failure

Dysregulation of ANS in heart failure has been investigated due to the impact on morbidity and mortality. As shown previously, heart failure is associated with decreased neuronal density in both atrial²³ and ventricular myocardium.^24–28 Conversely to the decreased sympathetic nerve density, sympathetic activity is increased to preserve the cardiac output in systolic heart failure as a part of the neurohumoral compensatory mechanism.^24,29 These sympathetic nerve alterations in heart failure may be associated with increased oxidative stress^28,30 and can further impair cardiac function and promote arrhythmogenesis.³¹ In addition, the plasticity of ANS may play a role in heart failure. It has been reported that sympathetic nerve fibers may become cholinergic through a transdifferentiation process, which could explain sympathetic innervation decrease in heart failure.^31,32

In our recent study, we showed decreased TH- and GAP43-positive nerve densities in the superior vena cava myocardial sleeves in subjects with CVS mortality.¹⁶ Furthermore, in agreement with these results, this study shows that sympathetic innervation was decreased in the myocardial sleeves and the fibro-fatty tissue around PV in patients with documented immediate CVS cause of death, of which 66.6% of subjects were diagnosed with heart failure. However, the difference reached the statistical difference only in fibro-fatty tissue. Interestingly, Parisi et al.³³ demonstrated a significant correlation between the thickness of epicardial adipose tissue and the extent of myocardial sympathetic denervation in subjects with systolic heart failure. The epicardial adipose tissue thickness can be used as a clinical predictor for myocardial sympathetic dysfunction. However, parasympathetic nerve density around PVs and atrial-PV ostia did not differ in subjects with documented immediate CVS cause of death. This finding is in agreement with a previous canine study observing atrial neural remodeling in congestive heart failure.³⁴

Autonomic Nervous System and Myocardial Ischemia

Altered autonomic innervation has been associated with increased atherosclerosis,³⁵ which is the most common underlying cause of CVS death in our cohort of PVs. Previous studies provided strong evidence of myocardial sympathetic denervation in ischemic heart disease,^36,37 causing hypersensitivity to adrenergic stimulation.³⁸ This can lead to lethal arrhythmias and sudden cardiac death.¹² In contrast to earlier findings, in our study, there was a trend of increased sympathetic innervation in atrial-PV ostia in subjects with documented underlying CVS cause of death. However, these results did not reach statistical significance.

Interestingly, we found that parasympathetic nerve density in atrial-PV ostia myocardial sleeves was significantly increased in subjects with documented underlying CVS cause of death compared with subjects with documented underlying non-CVS cause of death (p=0.034). In the atrial-PV ostia cohort, 72% of subjects with documented underlying CVS cause of death were diagnosed with systemic atherosclerosis. Increased parasympathetic innervation may be explained by the cardioprotective mechanism of the parasympathetic nervous system against myocardial ischemia to reduce myocardial injury in subjects with ischemic heart disease.³⁹

Pulmonary Veins, Atria, and Atrial Fibrillation

Myocardial sleeves around PVs were identified as the most common morphological substrate of atrial fibrillation,⁴⁰ especially in the atrial-PV ostia region.^17,41 Previous studies suggested a relationship between the myocardial sleeves’ length and the presence of atrial fibrillation, reporting that subjects with diagnosed atrial fibrillation had longer myocardial sleeves than subjects with sinus rhythm.^3,42 Also, the association of atrial fibrillation with scaring and amyloid deposits in the myocardial sleeves of PVs was shown previously.^2,43 Sympathetic innervation was found to be heterogeneously increased in the right atrial appendage compared with the left atrial appendage in persistent atrial fibrillation.⁴⁴ Moreover, Chang et al.⁴⁵ reported a significant increase in sympathetic innervation and nerve sprouting in canine models with sustained atrial fibrillation, with higher immunohistochemical positivity in the right atrial tissue than in the left atrial tissue.⁴⁵ Similarly, animal studies found a significant increase in the parasympathetic density and atrial parasympathetic innervation heterogeneity in chronically rapidly paced atria.^46,47 Interestingly, Nguyen et al.²⁰ described a substrate of chronic atrial fibrillation in PV myocardial sleeves characterized by the presence of periodic acid-Schiff–positive cells, interstitial Cajal-like cells, fibrotic tissue, inflammatory cells, and sympathetic nerves. They found increased sympathetic nerve density in the atria and the PV myocardial sleeves in subjects with atrial fibrillation compared with subjects with sinus rhythm, but without statistical significance.

Although abnormal autonomic innervation has been shown as a contributing factor to the initiation and maintenance of atrial fibrillation, neither autonomic nerve density nor neural growth in PV myocardial sleeves and atrial-PV ostia were associated with the presence of atrial fibrillation in our study. This also accords with our earlier observations on myocardial sleeves of caval veins, which showed no association of atrial fibrillation with autonomic nerve densities and neural growth.¹⁶ In addition, the present study material was previously tested for Leu-7 neural marker with no association with a history of atrial fibrillation.⁴⁸ The immunohistochemistry was underlined by electron microscopy with the detection of neuroendocrine granules corresponding to Leu-7 intensity.⁴⁸

Neural Growth Detected by GAP43 in TH- and CHAT-positive Nerves

GAP43 is a polypeptide expressed in growing axons,⁴⁹ and measuring the density of GAP43-positive nerves can be used to approach the growing activity of nerves rather than stable ANS innervation.⁵⁰ This study found that the positive GAP34 staining patterns in the TH-positive nerves differed between the myocardial sleeves and the surrounding fibro-fatty tissue in the study cohorts. The GAP43 neural growth activity was about two times decreased in the myocardial sleeves compared with the fibro-fatty tissue in the PV cohort (24% vs 53%) and the atrial-PV ostia cohort (28% vs 55%). Thus, the increased sympathetic growth in the surrounding fibro-fatty tissue might be involved in pathological processes, causing the formation of more mature synapses than in the myocardial sleeves.⁵⁰ Another interesting finding is that between 64% and 79% of CHAT-positive nerves show GAP43-positive neural growth activity within the studied regions in both study cohorts. Such colocalization implies that parasympathetic nerve growth was enhanced despite its sparse nerve distribution. These notable differences in the TH, CHAT, and GAP43 immunoreactivity suggest an interindividual variance in nerve density and neural growth genetic regulation. The combination of our findings provides possible support for future therapeutic interventions in targeting not only the distribution, but also the neural growth of autonomic nerves and their topography.

Study Limitations

This is a morphometric study with samples obtained postmortem; therefore, we did not perform functional electrophysiological measurements. Clinical data from the autopsy referrals were only available in this study. Our study showed substantial variability in nerve densities among subjects in both study cohorts, which possibly resulted in wide data distribution and high SDs. Such variation may be also attributable to the underlying health conditions influencing autonomic innervation, small series size, and tissue sample sizes with an uneven innervation pattern.

To conclude, variations in autonomic nerve density may be associated with CVS mortality as we observed sympathetic denervation in PVs in patients with documented immediate CVS death and increased parasympathetic innervation in atrial-PV ostia in patients with documented underlying CVS death. Sympathetic neural growth was found to be enhanced in the fibro-fatty tissue rather than in the myocardial sleeves, and neural growth activity was present in the majority of detected parasympathetic nerves. Knowledge of cardiac ANS remodeling in various disorders may provide further light on pathophysiology and treatment development in CVS diseases.

Footnotes

Acknowledgements

The authors thank Sari Toivola and Eini Eskola (Tampere University, Tampere, Finland) for their expert laboratory work. They also thank Heini Huhtala (Tampere University, Tampere, Finland) for her valuable comments on the statistical analysis.

Author Contributions

DD designed the study, performed the morphometric analysis, and drafted the manuscript. AM performed the statistical analysis of data and interpreted the results. RV performed part of the morphometric analysis. TP designed the study and interpreted the results. IK designed the study, interpreted the results, and drafted the manuscript. All authors have read and approved the final manuscript.

Competing Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by grants from the Competitive Research Funding of the Pirkanmaa Hospital District (to I.K., A.M., T.P.), Tampere University Hospital Support Foundation (to I.K.), Aarne Koskelo Foundation (to I.K., D.D.), Emil Aaltonen Foundation (to I.K.), Maire Taponen Foundation (to D.D.), Onni and Hilja Tuovinen Foundation (to D.D.), and the Research Foundation for Laboratory Medicine (to D.D.).

ORCID iDs

Denis Depes

Ari Mennander

Ivana Kholová

References

Sanchez-Quintana

Ramon Lopez-Mínguez

Pizarro

Murillo

Angel Cabrera

. Triggers and anatomical substrates in the genesis and perpetuation of atrial fibrillation. Curr Cardiol Rev. 2012 Nov 5;8(4):310–26. doi:10.2174/157340312803760721.

Steiner

Hájková

Kvasnička

Kholová

. Myocardial sleeves of pulmonary veins and atrial fibrillation: a postmortem histopathological study of 100 subjects. Virchows Arch. 2006 Apr 13;449(1):88–95. doi:10.1007/S00428-006-0197-2.

Kholová

Kautzner

. Anatomic characteristics of extensions of atrial myocardium into the pulmonary veins in subjects with and without atrial fibrillation. Pacing Clin Electrophysiol. 2003 Jun 1;26(6):1348–55. doi:10.1046/j.1460-9592.2003.t01-1-00193.x.

Hassani

Saremi

. Comprehensive cross-sectional imaging of the pulmonary veins. Radiographics. 2017 Nov 1;37(7):1928–54. doi:10.1148/RG.2017170050.

Hocini

Kawara

Linnenbank

Potse

Shah

Jaïs

Janse

Haïssaguerre

De Bakker

JMT

. Electrical conduction in canine pulmonary veins. Circulation. 2002 May 21;105(20):2442–8. doi:10.1161/01.CIR.0000016062.80020.11.

Arora

Ulphani

Mylonas

Subacius

Shade

Gordon

Morris

Belin

Goldberger

Kadish

. Unique autonomic profile of the pulmonary veins and posterior left atrium. J Am Coll Cardiol. 2007 Mar 27;49(12):1340–8. doi:10.1016/J.JACC.2006.10.075.

Ogawa

Zhou

Tan

Song

Gholmieh

Fishbein

Luo

Siegel

Karagueuzian

Chen

Lin

Chen

PS.

Left stellate ganglion and vagal nerve activity and cardiac arrhythmias in ambulatory dogs with pacing-induced congestive heart failure. J Am Coll Cardiol. 2007 Jul 24;50(4):335–43. doi:10.1016/j.jacc.2007.03.045.

Tomaselli

Marbán

Electrophysiological remodeling in hypertrophy and heart failure. Cardiovasc Res. 1999 May 1;42(2):270–83. doi:10.1016/S0008-6363(99)00017-6.

Shen

Zipes

DP.

Role of the autonomic nervous system in modulating cardiac arrhythmias. Circ Res. 2014;114:1004–21. doi:10.1161/CIRCRESAHA.113.302549.

10.

Wake

Brack

Characterization of the intrinsic cardiac nervous system. Auton Neurosci. 2016;199:3–16. doi:10.1016/j.autneu.2016.08.006.

11.

Florea

Cohn

JN.

The autonomic nervous system and heart failure. Circ Res. 2014;114:1815–26. doi:10.1161/CIRCRESAHA.114.302589.

12.

Fukuda

Kanazawa

Aizawa

Ardell

Shivkumar

Cardiac innervation and sudden cardiac death. Circ Res. 2015 Jun 5;116(12):2005–19. doi:10.1161/CIRCRESAHA.116.304679.

13.

Hadaya

Ardell

. Autonomic modulation for cardiovascular disease. Front Physiol. 2020 Dec 22;11:617459. doi:10.3389/FPHYS.2020.617459.

14.

Linz

Ukena

Mahfoud

Neuberger

Böhm

. Atrial autonomic innervation: a target for interventional antiarrhythmic therapy? J Am Coll Cardiol. 2014 Jan 28;63(3):215–24. doi:10.1016/J.JACC.2013.09.020.

15.

Villuendas

Cokic

Schliamser

Gordon

Koduri

Benefield

Simon

Murthy

SNP

Lomasney

Wasserstrom

Goldberger

Aistrup

Arora

. Autonomic remodeling in the left atrium and pulmonary veins in heart failure creation of a dynamic substrate for atrial fibrillation. Circ Arrhythmia Electrophysiol. 2011 Jun;4(3):388–96. doi:10.1161/CIRCEP.110.959650.

16.

Depes

Mennander

Paavonen

Kholová

Autonomic nerves in myocardial sleeves around caval veins: potential role in cardiovascular mortality?

Cardiovasc Pathol. 2022 Jul 1;59:107426. doi:10.1016/J.CARPATH.2022.107426.

17.

Tan

Wachsmann-Hogiu

Chen

Fishbein

. Autonomic innervation and segmental muscular disconnections at the human pulmonary vein-atrial junction. implications for catheter ablation of atrial-pulmonary vein junction. J Am Coll Cardiol. 2006 Jul 4;48(1):132–43. doi:10.1016/j.jacc.2006.02.054.

18.

Chevalier

Tabib

Meyronnet

Chalabreysse

Restier

Ludman

Aliès

Adeleine

Thivolet

Burri

Loire

François

Fanton

Quantitative study of nerves of the human left atrium. Hear Rhythm. 2005 May 1;2(5):518–22. doi:10.1016/j.hrthm.2005.01.022.

19.

Liu

Chen

Wang

Zhao

Zheng

Cardiac autonomic nerve distribution and arrhythmia. Neural Regen Res. 2012;7(35):2834–41. doi:10.3969/j.Issn.1673-5374.2012.35.012.

20.

Nguyen

Fishbein

Chen

Masroor

. Histopathological substrate for chronic atrial fibrillation in humans. Hear Rhythm. 2009 Apr 1;6(4):454–60. doi:10.1016/j.hrthm.2009.01.010.

21.

Petraitiene

Pauza

Benetis

. Distribution of adrenergic and cholinergic nerve fibres within intrinsic nerves at the level of the human heart hilum. Eur J Cardiothorac Surg. 2014 Jun 1;45(6):1097–105. doi:10.1093/ejcts/ezt575.

22.

Bankhead

Loughrey

Fernández

Dombrowski

McArt

Dunne

McQuaid

Gray

Murray

Coleman

James

Salto-Tellez

Hamilton

. QuPath: open source software for digital pathology image analysis. Sci Rep. 2017 Dec 1;7(1):1–7. doi:10.1038/s41598-017-17204-5.

23.

Kyösola

Partanen

Korkala

Merikallio

Penttilä

Siltanen

. Fluorescence histochemical and electron-microscopical observations on the innervation of the atrial myocardium of the adult human heart. Virchows Arch A Pathol Anat Histol. 1976 Jun;371(2):101–19. doi:10.1007/BF00444927.

24.

Triposkiadis

Karayannis

Giamouzis

Skoularigis

Louridas

Butler

. The sympathetic nervous system in heart failure: physiology, pathophysiology, and clinical implications. J Am Coll Cardiol. 2009 Nov 3;54(19):1747–62. doi:10.1016/J.JACC.2009.05.015.

25.

Himura

Felten

Kashiki

Lewandowski

Delehanty

Liang

CS.

Cardiac noradrenergic nerve terminal abnormalities in dogs with experimental congestive heart failure. Circulation. 1993;88(3):1299–309. doi:10.1161/01.CIR.88.3.1299.

26.

Siltanen

Penttilä

Merikallio

Kyösola

Klinge

Pispa

Myocardial catecholamines and their biosynthetic enzymes in various human heart diseases. Acta Med Scand. 1982 Jan 12;211(Suppl 660):24–33. doi:10.1111/j.0954-6820.1982.tb00357.x.

27.

Kreusser

Buss

Krebs

Kinscherf

Metz

Katus

Haass

Backs

. Differential expression of cardiac neurotrophic factors and sympathetic nerve ending abnormalities within the failing heart. J Mol Cell Cardiol. 2008 Feb 1;44(2):380–7. doi:10.1016/J.YJMCC.2007.10.019.

28.

Wang

Zhu

Chi

Yang

Jie

Han

Wang

Qin

Fan

. The NADPH oxidase inhibitor apocynin improves cardiac sympathetic nerve terminal innervation and function in heart failure. Exp Physiol. 2019 Nov 1;104(11):1638–49. doi:10.1113/EP087552.

29.

Durães Campos

Pinto

Sousa

Pereira

. A brain within the heart: a review on the intracardiac nervous system. J Mol Cellul Cardiol. 2018;119:1–9. doi:10.1016/j.yjmcc.2018.04.005.

30.

Shite

Qin

Mao

Kawai

Stevens

Liang

Antioxidant vitamins attenuate oxidative stress and cardiac dysfunction in tachycardia-induced cardiomyopathy. J Am Coll Cardiol. 2001 Nov 15;38(6):1734–40. doi:10.1016/S0735-1097(01)01596-0.

31.

Pinkham

Loftus

Amirapu

Guild

Quill

Woodward

Habecker

Barrett

CJ.

Cardiovascular and renal integration: renal denervation in male rats with heart failure improves ventricular sympathetic nerve innervation and function. Am J Physiol Regul Integr Comp Physiol. 2017 Mar 10;312(3):R368. doi:10.1152/AJPREGU.00313.2016.

32.

Kimura

Ieda

Fukuda

. Development, maturation, and transdifferentiation of cardiac sympathetic nerves. Circ Res. 2012 Jan 20;110(2):325–36. doi:10.1161/CIRCRESAHA.111.257253.

33.

Parisi

Rengo

Perrone-Filardi

Pagano

Femminella

Paolillo

Petraglia

Gambino

Caruso

Grimaldi

Baldascino

Nolano

Elia

Cannavo

De Bellis

Coscioni

Pellegrino

Cuocolo

Ferrara

Leosco

Increased epicardial adipose tissue volume correlates with cardiac sympathetic denervation in patients with heart failure. Circ Res. 2016;118(8):1244–53. doi:10.1161/CIRCRESAHA.115.307765.

34.

Villuendas

Cokic

Schliamser

Gordon

Koduri

Benefield

Simon

Murthy

SNP

Lomasney

Wasserstrom

Goldberger

Aistrup

Arora

. Autonomic remodeling in the left atrium and pulmonary veins in heart failure. Circ Arrhythmia Electrophysiol. 2011 Jun;4(3):388–96. doi:10.1161/CIRCEP.110.959650.

35.

Ulleryd

Prahl

Börsbo

Schmidt

Nilsson

Bergström

Johansson

. The association between autonomic dysfunction, inflammation and atherosclerosis in men under investigation for carotid plaques. PLoS ONE. 2017 Apr 1;12(4):e0174974. doi:10.1371/JOURNAL.PONE.0174974.

36.

Hartikainen

Mustonen

Kuikka

Vanninen

Kettunen

Cardiac sympathetic denervation in patients with coronary artery disease without previous myocardial infarction. Am J Cardiol. 1997 Aug 1;80(3):273–7. doi:10.1016/S0002-9149(97)00345-7.

37.

Fernandez

Ovchinnikov

Canty

Jr Fallavollita

JA.

Hibernating myocardium results in partial sympathetic denervation and nerve sprouting. Am J Physiol Hear Circ Physiol. 2013 Jan 15;304(2):H318. doi:10.1152/AJPHEART.00810.2011.

38.

Tapa

Wang

Stuart

SDF

Wang

Jiang

Habecker

Ripplinger

. Adrenergic supersensitivity and impaired neural control of cardiac electrophysiology following regional cardiac sympathetic nerve loss. Sci Rep. 2020 Dec 1;10(1):18801. doi:10.1038/S41598-020-75903-Y.

39.

Gourine

. Neural mechanisms of cardioprotection. Physiology. 2014 Mar 1;29(2):133. doi:10.1152/PHYSIOL.00037.2013.

40.

Haïssaguerre

Jaïs

Shah

Takahashi

Hocini

Quiniou

Garrigue

Le Mouroux

Le Métayer

Clémenty

. Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J Med. 1998 Sep 3;339(10):659–66. doi:10.1056/NEJM199809033391003.

41.

Chen

Tan

AY.

Autonomic nerve activity and atrial fibrillation. Hear Rhythm. 2007 Mar;4(3 Suppl):S61. doi:10.1016/j.hrthm.2006.12.006.

42.

Szczepanek

Bolechała

Koziej

Jasińska

Hołda

. Morphometric characteristics of myocardial sleeves of the pulmonary veins. J Cardiovasc Electrophysiol. 2020 Sep 1;31(9):2455–61. doi:10.1111/JCE.14651.

43.

Steiner

Hájková

. Patterns of isolated atrial amyloid: a study of 100 hearts on autopsy. Cardiovasc Pathol. 2006 Sep 1;15(5):287–90. doi:10.1016/J.CARPATH.2006.01.005.

44.

Gould

Yii

Mclean

Finch

Marshall

Lambert

Kaye

. Evidence for increased atrial sympathetic innervation in persistent human atrial fibrillation. Pacing Clin Electrophysiol. 2006 Aug;29(8):821–9. doi:10.1111/j.1540-8159.2006.00447.x.

45.

Chang

Zhou

Doshi

Lee

Ohara

Fishbein

Karagueuzian

Chen

. Nerve sprouting and sympathetic hyperinnervation in a canine model of atrial fibrillation produced by prolonged right atrial pacing. Circulation. 2001 Jan 9;103(1):22–5. doi:10.1161/01.CIR.103.1.22.

46.

Liu

Geng

Zhao

Yun

Wang

Yan

Ding

Wang

Pan

Gong

Tan

Valsartan reduced atrial fibrillation susceptibility by inhibiting atrial parasympathetic remodeling through MAPKs/neurturin pathway. Cell Physiol Biochem. 2015 Jul 25;36(5):2039–50. doi:10.1159/000430171.

47.

Zhang

Anderson

AS.

The sympathetic nervous system and heart failure. Cardiol Clin. 2014;32:33–45. doi:10.1016/j.ccl.2013.09.010.

48.

Kholová

Niessen

HWM

Kautzner

. Expression of Leu-7 in myocardial sleeves around human pulmonary veins. Cardiovasc Pathol. 2003 Sep 1;12(5):263–6. doi:10.1016/S1054-8807(03)00078-4.

49.

Meiri

Pfenninger

Willard

MB.

Growth-associated protein, GAP-43, a polypeptide that is induced when neurons extend axons, is a component of growth cones and corresponds to pp46, a major polypeptide of a subcellular fraction enriched in growth cones. Proc Natl Acad Sci USA. 1986;83(10):3537. doi:10.1073/PNAS.83.10.3537.

50.

Jong

Kim

Wang

Zemljic-Harpf

Ross

Fishbein

Chen

. Spatial distribution of nerve sprouting after myocardial infarction in mice. Heart Rhythm. 2006 Jun 1;3(6):728–36. doi:10.1016/j.hrthm.2006.02.005.