Device-based neuromodulation for cardiovascular diseases and patient’ s age

Abstract

The autonomic nervous system plays an important role in the pathogenesis of cardiovascular diseases. With aging, autonomic activity changes, and this impacts the physiological reactions to internal and external signals. Both sympathetic and parasympathetic responses seem to decline, reflecting functional and structural changes in nervous regulation. Although some investigators suggested that both the sympathetic and parasympathetic activities were suppressed, others found that only the parasympathetic activity was suppressed while the sympathetic activity increased. In addition, cardiac innervation progressively diminishes with aging. Therefore, one may suggest that neuromodulation interventions may have different effects, and older age groups can express an attenuated response. This article aims to discuss the effect of device-based neuromodulation in different cardiovascular diseases, depending on the patient’s age. Thus, we cover renal denervation, pulmonary artery denervation, baroreceptor activation therapy, vagus nerve stimulation, spinal cord stimulation, ganglionated plexi ablation for the management of arterial and pulmonary hypertension, heart failure, angina and arrhythmias. The results of many clinical studies appeared to be unconvincing. In view of the low rate of positive findings in clinical studies incorporating neuromodulation approaches, we suggest the underestimation of advanced age as a potential contributing factor to poorer response. Analysis of outcomes between different age groups in clinical trials may shed more light on the true effects of neuromodulation when neutral/ambiguous results are obtained.

The autonomic nervous system (ANS) plays a central role in cardiovascular regulation. In the early stages of cardiovascular diseases, the ANS regulates adaptive resources to unfavorable circumstances and provides a compensatory salvation to functional changes.^[1] However, during further pathological development, the ANS enters into a vicious circle and contributes to cardiovascular deterioration.^[2] Thus, sympathetic overactivity corresponds to the progression of myocardial dysfunction and vascular remodeling. Therefore, approaches to modulate ANS activity, in order to manage many cardiovascular diseases, have been under development for a long time.

With aging, autonomic activity changes, and this impacts the physiological reactions to internal and external signals. Both sympathetic and parasympathetic ANS responses seem to decline, reflecting functional and structural changes in nervous regulation.^[3] Although some investigators suggested that both the sympathetic and parasympathetic activities were suppressed,^[3] others found that only the parasympathetic activity was suppressed while the sympathetic activity increased.^[4] In addition, cardiac innervation progressively diminishes with aging.^[5] Therefore, one may suggest that neuromodulation interventions may have different effects, and older age groups can express an attenuated response. Although there is a lack of direct comparison of ANS interventions in the elderly, some clinical studies addressed this issue in their primary or ancillary analyses.

Over the last decades, devices that modify the ANS have been developed and tested for the management of conditions, such as systemic and pulmonary hypertension, heart failure, ischemic heart disease, arrhythmias, and hypotension (Figure 1).

Figure 1.

Autonomic nervous system modulation in cardiovascular disease and age-related effects.

The Figure was partly generated using Servier Medical Art.

Here, we present a brief overview of the main clinical studies on the use of device-based ANS modulation in different cardiovascular pathologies. This review focuses specifically on comparing the outcomes of these techniques in patients with advanced age to those in younger patients.

HYPERTENSION

Hypertension is the world’s most prevalent cardiovascular disorder affecting > 26% of the human population.^[6] Despite the extensive research in the field of essential hypertension pathophysiology, its precise causes and mechanisms are still elusive. Thus, prehypertension is already characterized by autonomic disbalance, as shown in the normotensive offspring of hypertensive parents, in whom heart rate variability (HRV) analysis demonstrated sympathetic overdrive and parasympathetic activity impairment.^[7] Elevated sympathetic nervous drive has been established in the early stages of hypertension,^[8] suggesting that neurohormonal dysregulation may be essential to its etiology,^[9] and is associated with endothelial dysfunction, subsequent end-organ damage, raised arterial stiffness and left ventricular hypertrophy. Thus, the adrenergic overdrive in hypertensives follows the blood pressure increase and the progression of the disease to complicated stages.^[10,11-16]

In addition to the sympathetic overdrive, baroreflex dysfunction and other mechanisms have been found attributed to the autonomic disbalance in high blood pressure: cardiopulmonary reflex abnormality, the role of arterial chemoreceptors, as well as humoral factors—leptin, insulin, angiotensin II, and other signaling molecules.^[12]

Sympathetic drive with the increased levels of neurotransmitters (adrenaline and norepinephrine) eventually leads to a deterioration of the blood supply to the glomerulus of the nephron, activation of the juxtaglomerular apparatus, and the activation of the renin-angiotensin-aldosterone system, which enhances and prolongs the effect of catecholamines on the artery wall, increasing its tone to an even greater extent.^[17,18]

Evaluating potential differences in ANS activity, HRV analysis has become a simple and widely accepted methodology in cross-sectional and prospective studies. Thus, in a meta-analysis of HRV in different population groups, women had a decreased power in the low-frequency domain when compared to men. With aging, both the low- and high-frequency HRV domains are decreased, but the low/high-frequency indices ratio increases.^[19] This suggests the natural decline in general ANS activity with the prevalence of sympathetic drive with aging and warns about possible differences in response to neuromodulatory interventions for blood pressure regulation in elderly subjects.

Invasive approaches to neuromodulation are emerging together with pharmacological agents aiming at inhibiting neurohumoral activity (beta-adrenoblockers, anti-angiotensin II agents). Although not mainly recommended in clinical practice guidelines, renal denervation (RND) remains one of the most frequently used invasive procedures for suppression of sympathetic activity in hypertension. Following the success of the first clinical trials, Symplicity HTN-1 and HTN-2, the first sham-controlled study, Symplicity HTN-3, failed to demonstrate the benefits of RND in correcting blood pressure in comparison with a sham procedure.^[20] However, the newest technologies are raising hopes for RND in hypertensive subjects.^[21]

In a total of 12 main original studies evaluating the clinical effects of RND in patients with hypertension, including one global registry, eight included a sham procedure in a control group (Table 1). In five studies, a comparison of the RND procedure on blood pressure effects was provided between older (> 65 or > 75 years) and younger patient groups, and four of these studies found no difference in blood pressure effects across age groups. A sub-analysis of the Symplicity HTN-3 study identified that the group under 65 years of age was statistically associated with a reduction in systolic blood pressure at 6 months after RND; however, this result was not reproducible in the multivariate model.^[22] The largest-to-date global RND registry included 2466 patients with a 3-year follow-up: the authors found that RND had similar efficacy in patients aged < 65 years (n = 1407) and those aged ≥ 65 years (n = 1059) in terms of blood pressure reduction.^[23]

Table 1. Clinical studies evaluating device-based neuromodulation therapies for cardiovascular diseases (main original studies included; case reports and duplicating cohorts were excluded, except reference [9]).

	Author, year	Primary condition	Age category and groups	N patients treated using a neuromodulation approach	Mean patient age (years)	% of males	Control group	Sham- controlled?	Advanced age associated with less efficacy	Comments	Primary outcome	Result
*Only studies with a control group with a standard PV isolation. AF: atrial fibrillation; BAT: baroreceptor activation therapy; BNP: brain natriuretic peptide; BP: blood pressure; CTEPH: chronic thromboembolic pulmonary hypertension; GP: ganglionated plexi; HF: heart failure; HFpEF: heart failure with preserved left ventricle ejection fraction; HFrEF: heart failure with reduced left ventricle ejection fraction; HTN: hypertension; LVEDV: left ventricular end diastolic volume; LVEF: left ventricle ejection fraction; LVESV: left ventricular end systolic volume; MIBG: meta-iodobenzylguanidine; MLWHFQ: Minnesota Living with Heart Failure Questionnaire; NT-pro–BNP: N-terminal fragment of the brain natriuretic peptide; NYHA: New York Heart Failure Association; PA: pulmonary artery; PAH: pulmonary arterial hypertension; PH: pulmonary hypertension; PV: pulmonary vein; PVR: pulmonary vascular resistance; 6MWT: six-minute walk test; QoL: quality of life; RND: renal denervation.
Renal denervation
1.	Ziegler, et al., 2015[91]	HTN	Only patients ≥ 75 years old were included	24	● 75 (75–89)	46%	No	No	Not analyzed	A significant BP reduction following RND was found	Change in mean systolic office BP at 6-months	Positive
2.	Krum, et al., 2011 (Symplicity HTN-1 trial)[92]	HTN	≤ 65 years old and > 65 years old	88	57 ± 11	61%	No	No	No	No difference in BP change between groups	Change in office BP	Positive
3.	Esler, et al., 2010 (Symplicity HTN-2 trial)[93]	HTN	None	53	58 ± 12	65%	54	No	Not analyzed	-	Change in average office systolic BP at 6 months	Positive
4.	Bhatt, et al., 2014 (Symplicity HTN-3 trial)[20]	HTN	None	535 246 patients < 65 years old 104 patients ≥ 65 years old	57.9 ± 10.4	59.1%	171 128 patients < 65 years old 104 patients ≥65 years old	Yes	Yes	In patients ≥ 65 years old, no difference in BP change following either an RND or sham procedure. In patients <65 years old RND produced better BP lowering	Difference in office systolic blood pressure change at 6 months	Neutral
5.	Kandzari, et al., 2015[22] Same population as above	HTN	None	535	57.9 ± 10.4	59.1%	171	Same population as above	Maybe	A subanalysis of Symplicity HTN 3. Age < 65 years was not an independent factor associated with the effect	Difference in office systolic blood pressure change at 6 months	The subgroup of patients <65 years in age was associated with SBP change in the RDN group in univariable analysis but not in the multivariable model
6.	Desch, et al., 2015[94]	HTN	None	35	64.5 ± 7.6	77%	36	Yes	Not analyzed	-	Change in 24-hour systolic BP at 6 months	Neutral
7.	Mathiassen, et al., 2016 (ReSET Study)[95]	HTN	None	36	54.3 ± 7.8	75%	33	Yes	Not analyzed	-	Mean change in daytime systolic ambulatory BP monitoring from baseline to 3 months	Neutral
8.	Townsend, et al., 2017 (SPYRAL OFF MED)[96]	HTN	None	166	55.8 ± 10.1	68.4%	165	Yes	No	Treatment differences suggested efficacy of renal denervation for patients of different age	Change in 24-h blood pressure at 3 months	Positive (although not powered for efficacy endpoints)
9.	Schmeider, et al., 2018 (WAVE IV Study)[97]	HTN	None	42	60.3 ± 11.2	81.4%	39	Yes	Not analyzed	-	Difference in office systolic BP at 24 weeks	Neutral
10.	Azizi, et al., 2018 (RADIANCE HTN SOLO)[98]	HTN	None	74	54.4 ± 10.2	62%	72	Yes	No	The blood pressure lowering effect of renal denervation was consistent across ages	Mean change in daytime ambulatory systolic BP at 2 months	Positive
11.	Kandzari, et al., 2018 (SPYRAL ON MED)[99]	HTN	None	38	53.9 ± 8,7	87%	42	Yes	Not analyzed	-	24h BP reduction at 6 months	Positive
12.	Mahfoud, et al., 2020[23]	HTN	< 65 years and > 65 years	2466 patients with 3-years follow-up: 1059 ≥65 years old, 1407 <65 years old	NA	NA	No	No	No	Patients aged < 65 and ≥ 65 years had similar efficacy of RND. Patients ≥65 years old had a higher mortality rate during follow-up	Document long-term safety and effectiveness of RDN in a real-world patient population	Positive
13.	Gao, et al., 2019[52]	HFrEF	None	30	60.2 ± 11.6	78.3%	30	No	Not analyzed	Results for different age groups were not reported	NT-proBNP, LVEF, NYHA class, 6MWT at 6 months	Positive for all outcome measures
14.	Chen, et al., 2016[51]	HFrEF	Patients > 75 years old were excluded	30	48.5 ± 8.4	73.3%	30	No	Not analyzed	Results for different age groups were not reported	Change in LVEF at 6 months	Positive
15.	Spadaro, et al., 2019[54]	HFrEF (Chagas’ disease)	Patients > 70 years old were excluded	11	52.6 ± 8.7	90.9%	6	No	Not analyzed	Results for different age groups were not reported	Composite of all-cause death, myocardial infarction, stroke, need for renal artery invasive treatment, or worsening renal function	Neutral
16.	Feyz, et al., 2022 (IMPROVE-HF-I)[53]	HFrEF	None	24	60 ± 9	86%	25	No	Not analyzed	Results for different age groups were not reported	The change in 123I-MIBG heart-to-mediastinum ratio at 6 months	Neutral
17.	Patel, et al., 2016 (RDT-PEF)[57]	HFpEF	None	17	74.3 ± 6.1	60%	8	No	Not analyzed	Results for different age groups were not reported	Improvement in at least one parameter: MLWHFQ; VO2 peak; BNP; E/e′; left atrial volume index; LV mass index	Neutral
18.	Kresoja, et al., 2021[56]	HTN + HFpEF	Patients with HF were older	154 (99 with HF and 65 without HF)	66 (61–73)	66%	No	No; retrospective study	Not analyzed	Results for different age groups were not reported	None predefined	Improvements in LV filling
19.	Turagam, et al., 2021 (HFIB studies)[87]	AF + HTN	Patients in the RND group were significantly younger (59 ± 10 vs 68 ± 9 years); p=0.01)	31 (PV isolation + RND)	59.1 ± 10.4; 64.2 ± 6.8	62%	39 (PV isolation only)	No	Not analyzed	RND group patients were younger that in the control group	AF-freedom	Neutral
20.	Pokushalov, et al., 2012[85]	AF + HTN	None	13 (PV isolation+RND)	57 ± 8	84%	14 (PV isolation only)	No	Not analyzed	Results for different age groups were not reported	AF recurrence at 12 months	Positive
21.	Pokushalov, et al., 2014[86]	AF + HTN	None	41 (PV isolation + RND)	56 ± 6	76%	39 (PV isolation only)	No	Not analyzed	Results for different age groups were not reported	AF recurrence	Positive
22.	Kiuchi, et al., 2017[100]	AF + kidney disease	None	39 (PV isolation + RND)	60 ± 14	62%	197 (PV isolation only)	No	Not analyzed	Results for different age groups were not reported	AF recurrence	Neutral (Positive in chronic kidney disease stage 4 only – 13 patients)
23.	Steinberg, et al., 2020 (ERADICATE-AF study)[101]	AF+HTN	None	154 (PV isolation + RND)	59 [54; 65]	59.1%	148 (PV isolation only)	No	Not analyzed	Results for different age groups were not reported	AF freedom at 12 months	Positive
Pulmonary artery denervation
24.	Chen, et al., 2013[102]	PAH	None	13	40 ± 16	69%	8	No	Not analyzed	Results for different age groups were not reported	Improvement of functional capacity by the 6MWT and mean PA pressure at 3 months	Positive
25.	Zhang, et al., 2019[103]	PAH + PVH	None	48	63.7 ± 11.8	62.5%	50	Yes	Not analyzed	Results for different age groups were not reported	Change in the 6MWD at 6-month follow-up evaluation at 6 months	Positive
26.	Romanov, et al., 2020[104]	PAH (CTEPH)	None	25	48 ± 14	48%	25	No	Not analyzed	Results in different age groups were not reported	Change in PVR at 12 months	Positive
27.	Chen, et al., 2015[105]	PAH + PH due to LV dysfunction	None	66	52 ± 16	41%	No	No	Not analyzed	Results for different age groups were not reported	Changes in hemodynamic, functional, and clinical responses within 1-year	Positive
28.	Rothman, et al., 2020 (THROPHY1)[106]	PAH	None	23	60.0 ± 11.4	22%	No	No	Not analyzed	Results for different age groups were not reported	Secondary endpoints: PH worsening, death at 12 months, change in PVR, mean pulmonary artery pressure, right atrial pressure, 6MWD, QoL, NT-pro–BNP, disease-specific medication at 4–6 months	Positive
Baroreflex activation therapy
29.	Gronda, et al., 2014 (BAT in HF)[43]	HFrEF	None	11	67 ± 9	72.7%	No	No	Not analyzed	Results for different age groups were not reported	Muscle sympathetic nerve traffic	Positive
30.	Abraham, et al., 2015 (HOPE4HF)[44]	HFrEF	None	71	64 ± 11	81.7%	69	No	Not analyzed	Results for different age groups were not reported	Three primary efficacy endpoints: changes in NYHA functional class; QoL, and 6MHW	Positive
31.	Zile, et al., 2020 (BeAT-HF)[45]	HFrEF	42% of BAT subjects ≥65 years old	130	62 ± 11	82%	134 controls in cohort “D”	No	Not analyzed	Difference in effects in patients ≥65 years old was not reported	Three primary effectiveness endpoints: 6MHW, QoL, and NT-proBNP	(1) Positive (6MHW, QoL, NT-proBNP). (2) Neutral in patients with baseline NT-proBNP > 1600 pg/mL (except QoL)
32.	Bisognano, et al., 2011 (Rheos Pivotal Trial)[107]	HTN	None	181	53.7 ± 10.5	60%	84	No	Not analyzed	Difference in effects in patients ≥65 years old was not reported	Two co-primary efficacy endpoints: 1) acute efficacy; 2) sustained efficacy	(1) Neutral; (2) Positive
33.	Hoppe, et al., 2012 (Barostim neo trial)[108]	HTN	None	30	57 ± 12	46.7%	No	No	Not analyzed	Difference in effects in patients ≥65 years old was not reported	Office systolic blood pressure	Positive
34.	de Leeuw, et al., 2017 (US Rheos Feasibility, DEBuT-HT, Rheos Pivotal)[25]	HTN	None	383	53 ± 10	60%	No	No	Yes	Diastolic BP was reduced to a lesser extent in patients > 60 years of age, but these patients already had a lower diastolic pressure at the start of the study	Blood pressure reduction at 6 years	Positive
35.	Beige, et al., 2017[109]	HTN	None	16 patients with therapy withdrawal after a period of BAT therapy	56.5 ± 14.4	76%	No	No	Not analyzed	Difference in effects in patients ≥65 years old was not reported	Automated office blood pressure	Positive
Vagus nerve invasive stimulation
36.	Schwartz, et al., 2008[36]	HFrEF	None	8	54.1 ±  11.5 (31–70)	100%	No	No	Not analyzed	Difference in effects in age groups was not reported	Changes in NYHA functional class, QoL, exercise capacity, LVESV, LVEDV, LVEF	Positive
37.	De Ferrari, et al., 2011 (CardioFit trial)[37]	HFrEF	None	32	56 ±  11	94%	No	No	Not analyzed	Difference in effects in age groups was not reported	NYHA class, QoL, 6MWT, LVEF, LVEDV, LVESV	Positive
38.	Dicarlo, et al., 2013[38]; Premchand, et al., 2015[39] (ANTHEM-HF)	HFrEF	None	60	51.5 ±  12.2	87%	No	No	Not analyzed	Difference in effects in age groups was not reported	Changes in LVEF and LVESV	No statistical data presented
39.	Zannad, et al., 2015; De Ferrari, et al., 2017 (NECTAR-HF)[40,41]	HFrEF	None	63	59.8 ±  12.2	89%	32	No	Not analyzed	Difference in effects in age groups was not reported	LVESD change	Neutral
40.	Gold, et al., 2016 (INOVATE-HF)[42]	HFrEF	None	436	61.7 ±  10.5	77.8%	271	No	Not analyzed	Difference in effects in age groups was not reported	Death or heart failure events	Neutral
Vagus nerve non-invasive stimulation
41.	Stavrakis, et al., 2015[88]	AF	None	20	60.9 ±  7.8	75%	20	Yes	Not analyzed	Difference in effects in age groups was not reported	Pacing-induced AF duration	Positive
42.	Stavrakis, et al., 2020 (TREAT AF)[73]	AF	None	26	65.2 ±  14.5	46%	27	Yes	Not analyzed	Difference in effects in age groups was not reported	AF burden	Positive
43.	Yu, et al., 2017[89]	Ischemia and reperfusion injury	None	47	59 ±  11	78.7%	48	Yes	Not analyzed	Difference in effects in age groups was not reported	Myocardial ischemia-reperfusion injury	Positive
44.	Bretherton, et al., 2019[90]	Heart rate variability, QoL, mood and sleep in subjects > 55 years	None	26 patients in Study-3	64.12 (1.02)	35%	No	No	Not analyzed	Difference in effects in age groups was not reported	Autonomic function, QoL, mood, sleep	Positive in subjects with baseline parameters deviation
Splanchnic nerve
45.	Fudim, et al., 2018[48]	Acute decompensated heart failure	None	11	64  ±  13	80%	No	No	Not analyzed	Difference in effects in age groups was not reported	Hemodynamic changes	Positive
46.	Fudim, et al., 2018[49]	Acute heart failure	None	5	56	55%	No	No	Not analyzed	Difference in effects in age groups was not reported	Hemodynamic changes	Positive
47.	Fudim, et al., 2020[50]	Chronic heart failure	None	15	58 ±  13	53%	No	No	Not analyzed	Difference in effects in age groups was not reported	Cardiac output and exercise capacity	Positive
Spinal cord invasive stimulation
48.	Bondesson, et al., 2008[110]	Angina	None	44	69 (54–87)	82%	No	No	Not analyzed	Difference in effects in age groups was not reported	Glyceryl trinitrate usage and Canadian Cardiovascular Society classification	Positive
49.	DeJongste, et al., 1994[111]	Angina	None	17 (8 treatment; 9 control)	62.5 ± 2.6	88%	No	No	Not analyzed	Difference in effects in age groups was not reported	Exercise tolerance test and QoL	Positive
50.	Eddicks, et al., 2006[112]	Angina	None	12 (patients were randomly assigned to different study phases, including a placebo phase)	65 ± 8	66.7%	No	Yes	Not analyzed	Difference in effects in age groups was not reported	Functional status and symptoms	Positive in conventional or subthreshold stimulation
51.	Greco, et al., 1999[113]	Angina	None	23	69 ± 11 (38–83)	NA	No	No	Not analyzed	Difference in effects in age groups was not reported	Number of angina episodes	Positive
52.	Hautvast, et al., 1998[114]	Angina	Only ≤ 75 years old were included	13	62 ± 8	46%	12	Yes	Not analyzed	Difference in effects in age groups was not reported	Exercise duration, time to angina, anginal attacks, nitrate consumption, number of ischemic episodes	Positive
53.	Lanza, et al., 2010[115]	Angina	None	10 with high-voltage stimulation; 7 with low-voltage stimulation	67.5 ± 13	78%	8	Yes	Not analyzed	Difference in effects in age groups was not reported	Angina episodes, nitroglycerin use, angina class, quality of life	Positive in the treatment “paresthetic” arm
54.	Mannheimer, et al., 1988[116]	Angina	None	10	51–74 years	80%	No	No	Not analyzed	Difference in effects in age groups was not reported	Exercise tolerance test	Positive
55.	McNab, et al., 2006[117]	Angina	None	32	64.2 ± 7.3	85%	33 (laser revascularization)	No	Not analyzed	Difference in effects in age groups was not reported	Time to angina during Exercise tolerance test	Positive
56.	Jesserun, et al., 1999[118]	Angina	None	26	61.3 ± 7	50%	No	No	Not analyzed	Difference in effects in age groups was not reported	QoL	Positive
57.	Zipes, et al., 2012 (STARTSTIM)[119]	Angina	None	32 (high stimulation)	61.3 ± 9.1 (high stimulation) and 60.9 ± 12.1 (low stimulation)	75%	36 (low stimulation)	Yes (high- and low-stimulation groups)	Not analyzed	Difference in effects in age groups was not reported	Number of angina attacks	Neutral
58.	Tse, et al., 2015 (SCS HEART study)[46]	Heart failure (HFrEF)	None	17	62.9 ± 10.1	100%	4	No	Not analyzed	Difference in effects in age groups was not reported	Heart failure symptoms, functional status, and LV function and remodeling	Positive
59.	Zipes, et al., 2016 (DEFEAT-HF study)[47]	Heart failure (HFrEF)	Treatment arm was younger that controls	42	58 ± 11	76.2%	24	Yes	Not analyzed	Difference in effects in age groups was not reported	Heart failure metrics: heart size, biomarkers, functional capacity, and symptoms	Neutral
GP ablation*
60.	Barta, et al., 2017[79]	AF (surgical epicardial ablation)	None	35 (PV isolation+GP ablation)	69 ± 6.4	51%	65 (PV isolation only)	No	Not analyzed	Difference in effects in age groups was not reported	Duration of sinus rhythm during 1 year	Neutral
61.	Driessen, et al., 2016 (AFACT study)[80]	AF (surgical epicardial ablation)	Age ≥65 years – 28% of patients	117 (PV isolation+GP ablation)	60.2 ± 8.2	73%	123 (PV isolation only)	No	Analyzed effects in patients <65 years and > 65 years old	No difference in effects between age groups	AF recurrence at 1 year	Neutral. More complications in the treatment arm.
62.	Gelsomino, et al., 2016[81]	AF (surgical MAZE IV with or without epicardial GP ablation)	None	213 (PV isolation+GP ablation)	63 ± 7	64%	306 (PV isolation only)	No	Not analyzed	Difference in effects in age groups was not reported	AF-freedom (median follow-up 36.7 months)	Neutral
63.	Katritsis, et al., 2013[83]	AF (catheter ablation)	None	82 (PV isolation+GP ablation)	56 ± 8.1	60%, 70%	78 (PV isolation only)	No	Not analyzed	Difference in effects in age groups was not reported	AF-freedom during 2 years of follow-up	Positive
64.	Mikhaylov, et al., 2011[82]	AF (catheter ablation)	Age <65 years only	35 (GP ablation only)	56.9 ± 10.1	51%	35 (PV isolation only)	No	Only patients <65 years were included	Difference in effects in age groups was not reported	AF-freedom at 3 years	Negative
65.	Onorati, et al., 2008[84]	AF (surgical ablation)	Age > 65 years – 47.7%	31 (mini-Maze procedure+GP ablation)	64.2 ± 9.8	75%	44 (mini-Maze procedure only)	No	Not analyzed	Difference in effects in age groups was not reported	AF-freedom at 3 months following amiodarone stop	Positive
Spinal cord non-invasive stimulation
66.	Mikhaylov, et al., 2020[61]	Blood pressure control (no structural heart disease)	None	9	58 ± 12	55%	No	No	Not analyzed	Difference in effects in age groups was not reported	BP elevation; atrioventricular refractoriness change	Positive
67.	Phillips, et al., 2018[60]	Orthostatic hypotension (spinal trauma)	None	5	23–32 years old	75%	No	No	Not analyzed	Difference in effects in age groups was not reported	BP correction	Positive

Baroreceptor electrical stimulation, or baroreceptor activation therapy (BAT), is another approach to reducing blood pressure. Natural physiological activation of the baroreceptors by increased blood pressure results in suppression of central sympathetic drive and reduction in blood pressure and heart rate. Direct electrical irritation of the carotid baroreceptors (bypassing mechanotransduction within the carotid body) provides afferent impulsation into the brain and consequently may suppress central sympathetic outflow and lead to vasodilatation. With aging, changes in baroreflex function are associated with an impaired ability to buffer changes in blood pressure.^[24]

BAT was evaluated in four multicenter clinical trials. One study analyzed combined data from three trials (US Rheos feasibility, DEBuT-HT, Rheos pivotal) and counted 383 patients in total. It demonstrated that patients aged > 60 years had less reduced diastolic blood pressure at 6-year follow-up than younger subjects.^[25]

PULMONARY HYPERTENSION

The increased sympathetic drive has been suggested as one of the major factors for pulmonary arterial hypertension development.^[26] Pulmonary vasculature is innervated by vagal branches, branches of the stellate ganglion, and cervicothoracic sympathetic fibers. The latter branches include adrenergic, cholinergic, and sensory fibers.^[27] The perivascular neural net is formed by fibers from the hilum of the lung and is located in the pulmonary artery adventitia; the maximum nerve concentration is decreased from proximal to distal PA. Among the abundant artery neural net, sympathetic fibers prevail.^[28]

Considering specific innervation of the pulmonary artery and the role of ANS activity in the pathophysiology of pulmonary hypertension, proximal pulmonary artery denervation was developed, either transcatheter or surgical.^[29] The procedure aims at ablations of the artery wall from the endothelial or adventitial surface to damage the perivascular nerves, resulting in the abolishment or significant decrease of the efferent sympathetic nervous supply of pulmonary vasculature and vasodilatation.

Pulmonary artery innervation and autonomic activity have been shown to differ in patients with pulmonary hypertension of different age groups. Previously, a predominant expression of vasoconstrictor nerves in pre-capillary vessels in infants has been reported, a condition that may lead to and explain the susceptibility to hypertensive crisis.^[30] However, beyond the general decline in HRV and physiologic autonomic response with age,^[31] there is a lack of information regarding possible differences in pulmonary circulation autonomic regulation in young adults versus elderly patients.

Four clinical studies and one prospective registry evaluated transcatheter pulmonary artery denervation for the reduction of pulmonary artery pressure. Generally, pulmonary arterial hypertension develops at a younger age, while left-heart disease-related pulmonary hypertension may develop at any age. Not surprisingly, patients in those studies were relatively young (40–63 years). Despite the variety of etiologies of pulmonary hypertension, all reviewed studies demonstrated consistent positive effects of pulmonary artery denervation. Table 1 summarizes the primary outcomes of the studies and the number of subjects included.

HEART FAILURE WITH REDUCED LEFT VENTRICULAR EJECTION FRACTION

One of the key physiologic responses to myocardial alterations is the activation of the sympathetic drive, resulting in increased release and simultaneously decreased uptake of norepinephrine at nerve endings. The role of ANS disbalance in heart failure has been obtained mainly from studies on subjects with dilated ventricles and depressed left ventricular systolic function.^[32] Chronic sympathetic nerve drive in heart failure leads to chronically elevated stimulation of the cardiac β-adrenoreceptors, which has detrimental effects on the failing heart. Further consequences lead to cardiac β1- and β2-adrenoreceptor desensitization and β1-adrenoreceptor downregulation and the progressive loss of the inotropic reserves of the heart.^[33] Less data exists about the role of the parasympathetic nervous activity in heart failure, but it is mainly reduced leading to an increase in heart rate and decreased HRV, both of which are correlated with increased mortality.^[34,35]

Considering an increase in sympathetic nerve activity during aging in general, heart failure patients may exhibit age-related changes as well. Moreover, the loss of cardiac and vascular adrenal transmitters sensitivity appears not uniform and depends on many other factors beyond functional decline and aging, requiring further investigation.

As an approach to evaluating parasympathetic activity increase for heart failure, vagus nerve stimulation was suggested to increase parasympathetic nervous tone in heart failure and demonstrated promising results in preclinical investigations. The first-in-man study included eight patients with heart failure with reduced left ventricular ejection fraction (HFrEF) aged 31–70 years. Significant changes in heart failure functional class, quality of life, exercise capacity, and echocardiography data were shown.^[36] Following that, the Cardiofit trial enrolled 32 patients, although with no comparator group, repeated the positive results seen in vagal stimulation.^[37] Three larger clinical trials (ANTHEM-HF, NECTAR-HF, INOVATE-HF) reported neutral effects of vagal stimulation in relation to left ventricular ejection fraction and dimensions. In addition, other cardiovascular events or death were reported.^[38–42] None of the discussed trials assessed possible differences in therapeutic effects in older and younger patients.

Experimental results suggested that BAT was beneficial in patients with HFrEF because, along with the suppression of sympathetic drive, it activates the parasympathetic nervous activity. Three clinical BAT studies in HFrEF patients were published between 2014 and 2020 and evaluated the physiological or clinical outcomes, but none of them reported on the effects of BAT in different age groups. In 2014, the BAT in HF study results were published and showed the suppressive impact of BAT on muscle sympathetic nerve traffic in 11 patients with heart failure.^[43] Later, the HOPE4HF study, released in 2015,^[44] evaluating BAT in patients with HFrEF, demonstrated that there was a significant difference in three primary outcome measures (NYHA functional class, quality of life assessment, and walk distance during 6-min walk test [6MWT]) between the 71 subjects with primarily activated BAT (mean age: 64 ± 11 years), and the control group. The N-terminal prohormone of brain natriuretic peptide (NT-proBNP) is an important heart failure marker and a surrogate parameter in many heart failure studies. In the BeAT-HF study, the authors assessed the 6MWT distance, quality of life, and NT-proBNP level in 130 subjects with active BAT and in 134 control patients. BAT was associated with better quality of life and longer 6 MWT distance in patients with NT-proBNP levels less than 1600 pg/mL, while patients with a higher biomarker level responded to therapy only with the moderate quality of life improvement.^[45]

Spinal cord stimulation (SCS) is a therapeutic approach that has travelled a long way from its use in pain management to application in refractory angina and heart failure. It has been proposed that SCS at low cervical and high thoracic levels can improve the balance between myocardial oxygen supply and demand during ischemia and modulate the reflex activation of the parasympathetic and sympathetic nervous system by modulating intrinsic afferent sensory cardiac neurons.^[46] The earlier non-randomized study of SCS in HFrEF has provided a positive approach for the improvement of left ventricle systolic function and heart failure functional class.^[46] However, the largest multicentre randomized study of SCS in HFrEF (DEFEAT-HF), which included 42 patients in the treatment arm and 24 patients in the control arm (device was implanted but not activated), showed a neutral effect in heart size, biomarkers levels, functional capacity, and symptoms.^[47] It should be noted that patients were relatively young (58 ± 11 years), and the treatment arm was younger than the control arm. The effects of age and impact of the treatment have not been reported.

Fudim and co-workers conducted a series of pilot studies on splanchnic nerve block in HFrEF. The splanchnic nerve participates in blood volume redistribution to the central compartment, modifying cardiac preload. Patients with acute heart failure, decompensated heart failure, and chronic heart failure were evaluated before, during, and after transcutaneous pharmacologic splanchnic nerve block.^[48-50] In these studies, relatively young patients with HFrEF were included, the mean age ranged from 56 to 64 years. The authors reported on short-term hemodynamic changes following the intervention, such as increase in blood pressure, cardiac output, and exercise capacity. No comparison between age groups was attempted, and limited sample sizes of the pilot studies should be taken into account.

RND was tested in small clinical studies in patients with HFrEF. Two studies analyzed left ventricle ejection fraction among other outcome measures at 6 months following RND and found positive impact of RND when compared with the control group.^[51,52] In a recent study by Feyez, et al.,^[53] the authors evaluated cardiac sympathetic innervation by iodine-123 meta-iodobenzylguanidine heart-to-mediastinum ratio following multielectrode bipolar ablation of the renal arteries. The results were neutral in comparison with the control group. Another small study (17 patients in total) assessed RND in addition to HF treatment in patients with Chagas’ disease, and found no additional benefit.^[54] None of the studies reported on possible effect differences between age groups.

HEART FAILURE WITH PRESERVED LEFT VENTRICULAR EJECTION FRACTION

Patients with heart failure with preserved left ventricular ejection fraction (HFpEF) represent approximately one-half of the total population with heart failure.^[55] Subjects with HFpEF are usually older and characterized by impaired ventricular relaxation. ANS in HFpEF has been poorly investigated, however, there are suggestions for higher sympathetic drive in its pathophysiology.^[56]

Several studies evaluated RND in HFpEF patients. One retrospective analysis found an improvement in left ventricle filling and pressures following RND (no control group without RND),^[56] while another small randomized study was neutral in relation to patient’s functional status, exercise capacity, brain natriuretic peptide level, and echocardiography parameters.^[57] No age-associated effects were evaluated.

HYPOTENSION

Hypotension might be associated to endocrine, central and peripheral neurological disorders. A 30% hypotension prevalence is present in the elderly population, in contrast with young patients where prevalence is lower.^[58] In patients with trauma, cardiovascular dysfunction resulting in hypotension, may be due to spinal cord injury and is characterized by general autonomic impairment. Neurohormonal dysregulation and nerve structural lesions above the T6 segment level are suggested to be responsible for symptoms attributable to spinal cord injury, although its mechanisms can be multifactorial. In other clinical settings, aging is linked to physiological changes predisposing to hypotension as reported by Robertson et al., where the mean age of people with orthostatic hypotension was 74 years.^[59]

Electrical spinal cord stimulation may correct autonomic regulation of blood pressure. A suggested mechanism of this effect is thought to be associated with sympathetic pre-ganglionic neuron excitation or propriospinal preganglionic neurons excitation directly through the stimulation reaching the spinal cord, leading to increased vascular tone. ^[60]

Phillips and co-workers extensively worked on transcutaneous spinal cord stimulation (t-SCS) in patients undergoing rehabilitation after spinal cord injury. The authors noted that patients experiencing orthostatic hypotension had correction of blood pressure with t-SCS activation.^[60] Following this report, an evaluation of t-SCS in overall healthy subjects undergoing electrophysiological testing was performed with the assessment of hemodynamic and cardiac electrophysiology parameters. In this pilot study, t-SCS was performed in nine patients at levels T1, T7, and T11.^[61] During T1 and T7 stimulation, transient blood pressure elevation and atrioventricular nodal refractory period shortening were noted. Interestingly, in three patients aged 63–74 years, blood pressure elevation was ≤ 5 mm Hg, while in six patients aged 29–64 years, the elevation was > 10 mm Hg, suggesting a better response of the latter group to non-invasive stimulation.

ANGINA

The development of ischemia is a dynamic process, and the imbalance between oxygen supply and myocardial demand has a non-linear association and can be modulated by a number of factors.^[62] The sensation of angina pectoris seems to be the result of activity in neural circuits with the potential for modulation of the message at different levels.^[63,64] Mechanistically, the “peripheral” and “central” mechanisms are considered in the pathogenesis of pain and are firmly related to synaptic transduction and central nervous system regulation, the system acts as a fully integrated whole.^[63]

The theoretic background for the treatment of pain, including angina, with electrical stimulation, is largely based on the classic "gate control theory",^[65] which resulted in the development of the modulation of the nociceptive input to the dorsal horn of the spinal cord.^[66] Stimulation of large, non-nociceptive A-fibers afferents reduces activity in small, nociceptive C-fiber afferents and reduces tire transmission of "pain" impulses.^[66]

However, it is hardly believed that anti-anginal effects of neural stimulation is being achieved by only nociceptive modulation, since documented ischemia develops later and with higher exercise level than without stimulation. Therefore, general anti-sympathetic effects and changes in myocardial perfusion due to neuromodulation interventions have been suggested and confirmed in experimental studies and in patients with angina.^[67]

Invasive SCS in patients with angina at T1 and T2 segment levels has shown to suppress the pain associated with myocardial ischemia, the effect that is explained by decreasing oxygen consumption, pain impulse inhibition of nociceptive signals, and an enhancement in myocardial perfusion, oxygen supply and myocardial lactate metabolism leading to a quality-of-life improvement and symptoms reduction.^[68,69] Whether the previous mechanisms for clinical improvement are related to modulation of the ANS remains debatable since it has been observed a beneficial effect in patients who underwent sympathectomy. The pathophysiological role of the autonomous sympathetic system in vasospastic angina and coronary artery disease has not been well elucidated and it has been suggested not only an increase in sympathetic tone but also the participation of the parasympathetic nervous system as a trigger in alpha receptors in coronary arteries.^[70]

We identified 10 main studies on invasive SCS in patients with intractable angina pectoris, all of them included a relatively limited number of patients (12–68), four of them were sham-controlled, and nine reported positive results of SCS in terms of clinical improvement and angina episodes reduction, and one reported neutral effect in the number of angina attacks when low energy stimulation was applied. In general, SCS for angina has shown good results, but no study suggested (or analyzed) age-related differences in SCS effects.

ATRIAL FIBRILLATION

The development and perpetuation of cardiac arrhythmias, and atrial fibrillation (AF) in particular, are dependent on autonomic nervous activity. Paradoxically, AF may be suppressed by either, parasympathetic nervous activity reduction or potentiation.^[71–73] This contradiction might be related to the complexity of neural impact on AF triggers and substrate characteristics, suggesting that general ANS disbalance rather than its single-arm overactivity is the key. One of the supposed reasons for autonomic imbalance developing with age might be well-documented β-adrenoreceptor desensitization of the myocardium,^[74] which is a possible consequence of adaptive response against the age-related increase in the levels of catecholamines. Less is known about the impact of aging on parasympathetic nervous activity.^[75] Interestingly, the density and function of the M2 receptor seem to decline with age, leading to a decrease in cardiac parasympathetic response in baroreflex activity.^[76,77] These findings demonstrate some simple manifestations of substrate-based differences in autonomic regulation of cardiac electrophysiology that patients face with age. When age-related neural sensitivity changes join the existent gradients in myocardial innervation,^[78] the resultant neural regulation dispersion becomes individual and less predictable.

Transcutaneous or surgical ablation of cardiac nervous ganglia is one of the approaches that unequivocally may attenuate the effects of invasive AF treatment. Several clinical studies have reported a positive role of ganglionated plexi (GP) ablation in sinus rhythm maintenance when performed either isolated or with pulmonary vein (PV) isolation. However, there are only six studies incorporating a control group with standard PV isolation. When comparing GP ablation to PV isolation outcomes, in four of these clinical investigations, neither surgical epicardial GP ablation nor catheter endocardial GP ablation resulted in better AF freedom after the procedure.^[79–81] Moreover, one study reported a higher rate of major complications when additional GP ablation was performed during surgery.^[81] Another study reported a statistically higher rate of AF recurrence following left atrial GP ablation only versus PV isolation.^[82] However, two studies described a higher AF freedom rate following additional catheter or surgical GP ablation, when compared to PV isolation only.^[83,84]

On the other hand, sympathetic overactivity suppression may result in better AF control. In a series of studies of RND performed in addition to PV isolation, a reduction in AF recurrence rate has been noted.^[85,86] However, in a recent prospective multicenter study, no such benefit of RND was found.^[87] No possible age-specific response to the treatment was evaluated in any of the above studies.

A non-invasive approach to activation of the parasympathetic activity for AF suppression has been proposed by the Oklahoma research group. They reported on transcutaneous tragus stimulation (activating the auricular tragus nerve, a subcutaneous branch of the vagus nerve) as an approach to modifying the frequency and length of AF episodes. Two clinical trials with a sham group were published with positive results and without a sub-analysis of effects in advanced age patients.^[81,88]

OTHER CONDITIONS

One study compared invasive vagus nerve stimulation on ischemia and reperfusion injury in 47 patients versus a sham intervention in 48 patients.^[89] The authors found that stimulation was associated with less myocardial injury during ischemia and reperfusion.

In another study, the authors assessed Vagus nerve stimulation on HRV, quality of life, mood, and sleep in subjects > 55 years. Twenty-six patients were assessed in the so-called study-3 phase and a positive influence of stimulation on the assessed parameters was found.^[90] No age-related data were provided.

POTENTIAL AGE-RELATED DIFFERENCE IN NEUROMODULATION

Interestingly, almost every aforementioned neuromodulation approach has shown promising results in experimental preclinical studies. However, when a technology/approach is being tested in clinical settings, many of them fail to provide the benefits observed with conventional therapy and/or a sham procedure. Thus, among the 65 original studies listed in Table 1, 47 (72%) achieved predefined objectives, demonstrating improvements in primary outcomes. This discrepancy between expected benefits and unexpected failures requires detailed analysis and research for the key factors that lead to such a difference between “ideal” circumstances in experimental settings and “real” patients in clinical investigations. We cannot claim universal definitive factors in every case, because the crucial target of technique application is different in every field discussed. However, patient age may be one of the underrated contributing factors defining therapeutic application results. We should emphasize that our assumption is speculative at the moment, but we suggest that younger patients treated with neuromodulation approaches may benefit better. Advanced age has reportedly been associated with poorer parasympathetic responses and overall lower peripheral nervous activity.^[3] Therefore, the less the ANS has an impact on disease pathophysiology, the less neuromodulation effect is to be expected.

CONCLUSIONS

In summary, given the low rate of positive findings in clinical studies incorporating neuromodulation approaches, we suggest the underestimation of advanced age as a potential contributing factor to the poorer response. Analysis of outcomes between different age groups in clinical trials may shed more light on the true effects of neuromodulation when neutral/ambiguous results are obtained.

CONFLICT OF INTERESTS

None.