Targeting Preload in Heart Failure: Splanchnic Nerve Blockade and Beyond

Abstract

Preload augmentation represents a critical mechanism for the cardiovascular system to increase effective circulating blood volume to increase cardiac filling pressures and, subsequently, for the heart to increase cardiac output. The splanchnic vascular compartment is the primary source of vascular capacity and thus the primary target for preload recruitment in humans. Under normal conditions, sympathetic stimulation of these primary venous vessels promotes the shift of blood from the splanchnic to the thoracic compartment and elevates preload and cardiac output. However, in heart failure (HF), since filling pressures may be elevated at rest due to decreased venous capacitance, incremental recruitment of preload to enhance cardiac output may exacerbate congestion and limit exercise capacity. Accordingly, recent attention has focused on therapies designed to regulate splanchnic vascular redistribution to improve cardiac filling pressures and patient-centered outcomes such as quality of life and exercise capacity in patients with HF. In this review, we discuss the relevance of splanchnic circulation as a venous reservoir, the contribution of stressed blood volume to HF pathogenesis, and the implications for pharmacological therapeutic interventions to prevent HF decompensation. Further, we review emerging device-based approaches for cardiac preload reduction such as partial/complete occlusion of the superior vena cava or the inferior vena cava.

Introduction:

Preload reserve refers to the ability of the cardiovascular system to increase effective circulating blood volume in order for the heart to increase its cardiac output (CO). Preload reserve represents a critical mechanism required for rapid adaptation of the body to dramatically changing metabolic demands. For example, during exercise, CO increases about 5-fold in healthy adults and about 8-fold in athletes (1). Both cardiac and extracardiac factors mediate these changes, including increased heart rate, myocardial contractility, and decreased systemic vascular resistance (2). Less well recognized is the role of the venous system in these responses. Veins are not only conduits that return blood to the heart but also serve as functional reservoirs of blood. Veins contain around 70% of the total blood volume (TBV) compared to only 30% contained in arteries (3,4). Additionally, approximately 20–30% of the TBV is contained in the organs perfused heavily by the splanchnic compartment including the liver, spleen, and gut (5). In comparison with central (e.g., vena cava) and peripheral vessels (muscles of extremities), the splanchnic vessels are innervated by more dense adrenergic fibers with the veins receiving most of the innervation (2,6,7). Therefore, the venous pool of the splanchnic vascular compartment is the primary source of venous capacity in humans (5,8). Sympathetic stimulation of these venous vessels promotes the shift of blood from the splanchnic to the thoracic compartment. Under normal conditions, this augmented venous return leads to elevations in preload and CO by the Frank-Starling law of the heart (9). However, in states of acute and chronic heart failure (HF), preload reserve can be exhausted. Rather than increasing CO, such blood shifts primarily result in increased systemic and pulmonary venous pressures.

Thus, in addition to regulation of TBV, regulation of splanchnic tone and its implications for distribution of blood volume throughout the body are critical components of the pathophysiology of HF. Increased understanding of these fundamental concepts has led to the development of therapies targeting the splanchnic vasculature to improve acute and chronic HF outcomes. Accordingly, this review has two primary aims: First, we review recent evidence point to the pivotal role of the splanchnic vascular compartment as a key driver of cardiovascular decompensation in the pathophysiology of HF. Second, we review emerging, novel therapies aimed at reducing cardiac filling pressures independent of changes of TBV in the management of HF. Understanding the concept of stressed blood volume (SBV) is a core aspect of both of these aims.

Fundamentals of blood volume distribution and cardiac preload modulation

Functionally, the intravascular blood pool is divided into the SBV, ~25% and unstressed blood volume (UBV, ~75%) (TBV = SBV + UBV) (Supplemental Figure 1). The UBV is defined as the volume of blood required to fill the vasculature to the point where wall stress and intravascular pressure rise to just above 0 mmHg. The volume within the vasculature above the UBV, which determines wall stress and intravascular pressure, is the SBV, roughly 20–25% of TBV. The SBV determines venous pressure and generates the preload filling pressures seen by the heart. Thus, under normal conditions, SBV serves as a potent regulator of CO. The splanchnic vascular compartment (which accounts for ~ 30% of the total blood pool) contains a large reservoir of volume in its venous bed (10)(11). The sympathetic nervous system modulates the capacitance of the splanchnic venous circulation through the greater splanchnic nerve (GSN), and activation of these fibers results in venoconstriction and rapid functional shifts of blood volume from UBV to SBV which ultimately result in physical translocation of blood from the splanchnic bed to the central circulation. The splanchnic circulation is very sensitive to sympathetic tone because it contains a large concentration of adrenergic receptors in the vessel walls.

Anatomy and innervation of the splanchnic vasculature

The splanchnic circulation supplies blood to the abdominal gastrointestinal organs including the stomach, liver, pancreas, spleen, small intestine, and large intestine. The blood flow through the splanchnic circulation is dynamic and is regulated by various intrinsic and extrinsic mechanisms according to the changing needs of the body. The visceral (sympathetic) nerve fibers provide innervation to the blood vessels in this compartment (12). The greater, lesser, and least thoracic splanchnic nerves arise from the bilateral sympathetic ganglia to provide fibers to the celiac plexus (13,14). Recruitment of blood from the splanchnic blood vessels mediated by sympathetic activation is an integral component of the homeostatic regulatory mechanisms in response to blood loss and strenuous physical activity.

Splanchnic nerve stimulation in regulating volume distribution and hemodynamics: evidence from animal and human studies

Research in both animals and humans suggests that the translocation of blood from the splanchnic to the central compartment leads to increased preload and CO (2,15,16). In preclinical studies, it has been demonstrated that splanchnic nerve stimulation recruits up to 80% of the splanchnic volume (or 20% of the TBV) and drives a modest increase in vascular pressures and CO (17–19). A more recent study in canines has substantiated these findings (20). Bapna and colleagues tested the viability of the implantable cuff system and its effect on the circulation by right-sided unilateral cuff placement in five mongrel hounds. The stimulation of the right GSN led to increases in central and peripheral hemodynamic pressures, with a mean increase of mean arterial pressure (MAP) by 36.9 ± 13.4 mmHg (p < 0.001), of central venous pressure (CVP) by 6.9 ± 1.7 mmHg (p < 0.001), and of pulmonary arterial mean pressure by 6.3 ± 2.0 mmHg (p < 0.0001). The mean increase in estimated CO was 7.4 ± 3.1 L/min (p < 0.001).

Investigations in humans have recently corroborated these findings. The hemodynamic effects of splanchnic nerve stimulation in humans were first detailed in a case series of five patients with pancreatic cancer who underwent treatment with irreversible electroporation (IRE). Data from this analysis demonstrated that cardiac preload was increased by ~ 50% via splanchnic nerve stimulation. This rise in preload was thought to be a consequence of catecholamine release from stimulation of the adrenal glands and splanchnic vasoconstriction. However, the absence of neurohormonal testing and direct evaluation of volume redistribution limited the interpretation of these findings (21).

An increase in cardiac preload via SBV increase is essential to the augmentation of CO in a healthy exercising human (2). Inadequate augmentation of preload with gravitational challenges such as standing or with exercise is termed preload reserve failure. Diseases that can present that way are referred to as dysautonomia (postural tachycardia syndrome, neurogenic orthostatic hypotension, etc.) but can also occur with venous obstruction, liver disease and now also seen with long-COVID syndrome.

Growing evidence suggests that dysregulation of volume redistribution also plays a significant role in acute decompensation of HF. Since the splanchnic vascular compartment holds the largest pool of intravascular blood volume, neurohormonal imbalance with increased sympathetic tone via the GSN causes an effective volume redistribution from the abdominal to the thoracic compartment. This volume distribution represents a functional increase of SBV and concomitant decrease of UBV. While in a healthy adult this might be a means of preload recruitment to increase CO, in patients with HF already having elevated TBV and elevated resting cardiac filling pressures, the diminished capability of the vascular bed to store blood can lead to unintended consequences.

For example, the impact of GSN activation in a patient with heart failure with preserved ejection fraction (HFpEF) and atrial fibrillation was recently published. A brief stimulation (40 seconds) of the splanchnic nerves increased the mean CVP from 6 to 8 mmHg, arterial blood pressure (ABP) from 85/45 to 123/75 mmHg, mean left atrial pressure (LAP) from 7 to 14 mmHg, and mean pulmonary artery pressure (PAP) from 14 to 21 mmHg. Unlike in preclinical or clinical studies with healthy animals/patients, the increase in central hemodynamics did not cease after the termination of splanchnic nerve stimulation and, in fact, the intracardiac pressures continued to increase (22). While this was a single experimental case, it powerfully demonstrated how the underlying cardiovascular disease could predispose to cardiac decompensation when exposed to increased SBV. These findings were extended by Kaye et al. who modeled the contribution of SBV to the intracardiac pressures [CVP and pulmonary capillary wedge pressure (PCWP)] as well as CO in normal and in patients with HFpEF in response to supine exercise. In these studies, SBV was found to be the major driver of increased cardiac filling pressures (both CVP and PCWP) without notable gains in CO (23,24) (Figure 1).

Figure 1.

Contribution of cardiovascular parameters to exercise-related changes in central venous pressure, pulmonary capillary wedge pressure, and cardiac output based on a cardiovascular modelling approach. This analysis is based on sequentially adding changes of systemic vascular resistance (SVR), left ventricular (LV) contractility (quantified by end-systolic elastance, LV Ees), changes in diastolic properties (indexed by the end-diastolic pressure-volume relationship, EDPVR), pulmonary vascular resistance, heart rate, right ventricular (RV) contractility (RV Ees) and stressed blood volume (SBV) (24).

Pharmacological approaches to modifying SBV

Wang and colleagues assessed the effects of vasodilator drugs such as hydralazine, enalaprilat, and nitroglycerin (NTG) on the splanchnic venous pressure-volume relation by experimental induction of acute ischemic HF in 19 splenectomized dogs. Radionuclide plethysmography was performed to determine the splanchnic vascular pressure-volume relation during the control stage, post-acute HF, and post-administration of a pharmacologic vasodilator. After induction of acute HF, there was a decrease in splanchnic vascular volume (SVV) from 100% to 86±2% and a parallel increase in left ventricular end-diastolic pressure (LVEDP) from 6±1 to 21±1 mm Hg (P<.001). Administration of hydralazine, enalaprilat, and NTG increased SVV from 86% to 88±3%, 96±3%, and 113±3% (P=NS, P<.01, and P<.001, respectively, versus HF) and lowered LVEDP to 18±2, 16±1, and 13±1 mm Hg (P=NS, P<.05, and P<.001, respectively) (Table 1). These data highlight the association of acute HF with blood volume redistribution to the central compartment as a consequence of splanchnic vasoconstriction. The administration of nitroglycerin and enalaprilat (but not hydralazine) contributed to splanchnic venodilation, thereby modulating left ventricular preload (25). In a recent analysis, Okamoto and colleagues investigated the splanchnic capacitance (pressure-volume relationship, PVR) and compliance before and after administration of 0.6 mg sublingual NTG using radioplethysmography. There was a significant rightward shift of the PVR towards larger volume induced by sublingual NTG indicating increased splanchnic capacitance (Table 2). These data directly link a therapy that induces splanchnic vasodilation to reductions of cardiac filling pressures (26).

Table 1.

Relative Hemodynamic and Splanchnic Vascular Volume Changes produced by acute heart failure, enalaprilat, hydralazine, and nitroglycerin

	Control	Acute Heart Failure	Enalaprilat	Hydralazine	Nitroglycerin
Acute Heart Failure (n=19)
CO, L/min	3.83±0.9	1.87±0.6	-	-	-
SVR, mm Hg · min · L−1	31.6±4.9	54.5±6.6	-	-	-
PVP, mm Hg	6.6±0.7	7.1±0.8	-	-	-
SVV, %	100	86±2	-	-	-
LVEDP, mm Hg	6.2±0.4	21.4±1.3
Enalaprilat (n=7)
CO, L/min	4.3±0.5	2.4±0.3	2.6±0.2	-	-
SVR, mm Hg · min · L−1	27.9±3	45.1±4.3	32.2±3.3	-	-
PVP, mm Hg	7.2±0.9	7.7±0.6	7.0±0.5	-	-
SVV, %	100	85±3	96±3	-	-
LVEDP, mm Hg	7.2±0.4	21.1±1.3	15.8±1.3	-	-
Hydralazine (n=6)
CO, L/min	3.7±0.8	1.5±0.5	-	2.2±0.6	-
SVR, mm Hg · min · L−1	30.3±4.1	53.7±5.9	-	29.4±3.3	-
PVP, mm Hg	6.5±1.0	7.1±0.7	-	6.2±0.8	-
SVV, %	100	86±2	-	88±3	-
LVEDP, mm Hg	6.3±0.5	20.4±1.8	-	17.8±1.9	-
Nitroglycerin (n=6)
CO, L/min	3.5±0.7	1.7±0.5	-	-	1.9±0.5
SVR, mm Hg · min · L−1	36.5±2.4	64.6 ±3.6	-	-	52.5±1.9
PVP, mm Hg	6.2±0.8	6.4±1.2	-	-	5.2 ±0.7
SVV, %	100	86 ±2	-	-	113±3
LVEDP, mm Hg	5.0 ±0.3	22.7±0.9	-	-	13.1±0.4

Values are mean ± SD. CO indicates cardiac output; SVR, systemic vascular resistance; PVP, mean portal vein pressure; SVV, splanchnic vascular volume.

Table 2.

Baseline characteristics and measurement of SBV pre-intervention and post-intervention

Study Title	Publication Year	Number of Participants (N)	Primary Outcome	Age (years)	BMI (kg/m2)	LVEF (%)	Intervention	Resting SBV Pre-intervention (ml)	Resting SBV Post-intervention (ml)	Pre-intervention SBV at peak load	Post-Intervention SBV at  Peak Load (ml)
Kaye et al. (23)	2020	10	Left ventricular end-systolic, end-diastolic pressure–volume relations, stressed blood volume, heart rate, and arterial mechanics	68 ± 2	31 ± 1	64 ± 2	Milrinone	1539	1066	-	-
Brener et al. (28)	2021	35	Mean pulmonary capillary wedge pressure at rest and during exercise	68 ± 9	34 ± 8	58 ± 8	Levosimendan	2750	2449*	-	-
Fudim et al.(10)	2021	14	Exercise capacity measured by peak oxygen uptake, mean pulmonary arterial pressure, and pulmonary capillary wedge pressure	58 ± 13	31 ± 16	21 ± 12	Splanchnic nerve blockade	2,664 ± 488	2,132 ± 570	3,243 ± 352	2662 ± 656
Okamoto et al. (26)	2021	85	Effect of SL NTG for increasing splanchnic capacitance and compliance	37 ± 44	31 ± 38	-	Nitroglycerin	100 ± 1.81 *	104.6 ± 8.81*	-	-
Omar et al. (30)	2021	35	Change in estimated SBV after 12 weeks of empagliflozin treatment	59 ± 8	29 ± 6	-	Empaglifozin	1697 ± 312	1601 ± 337	3269 ± 486	3105 ± 509

Values are mean ± SD or n or cm H20/ relative splanchnic volume*.

SBV indicates stressed blood volume; BMI indicates body mass index; LVEF indicates left ventricular ejection fraction; SL NTG indicates sublingual nitroglycerin.

Since SBV can actually only be directly measured accurately by inducing circulatory arrest, we employ an indirect method of estimating this quantity based on computer simulations relying on widely used models of the cardiovascular system. Estimated SBV (eSBV) is simulated using heart rate, CO, CVP, PCWP, systolic and diastolic systemic arterial pressures, PAP, and left ventricular ejection fraction. To account for differences in patient sizes, eSBV values are presented as ml/70 kg body weight (27). In a series of recent studies, investigators explored the effects of different pharmacologic therapies on hemodynamics and eSBV. First, Kaye et al. measured the effects of milrinone on resting and exercise hemodynamics in HFpEF patients. Milrinone decreased eSBV from 1539 ml to 1066 ml (Table 2) (23). Similarly, Brener et al studied the effects of levosimendan in patients with HFpEF patients with pulmonary hypertension (28,29) showing an ~500 ml decrease of eSBV (28). Omar et al. reported the effect of inhibition of sodium-glucose cotransporter-2 (SGLT2) on eSBV. Importantly, these effects on hemodynamics and eSBV were observed without any detected effect on myocardial, pulmonary, or systemic arterial effects. Empagliflozin (10 mg) was administered in patients with HFrEF and the change in eSBV was observed after 12 weeks of treatment over the full range of exercise. The results showed a statistically significant ~200 ml reduction (9%) of SBV (Table 2) (30). In addition, Sorimachi and colleagues performed invasive hemodynamic exercise testing in 62 HFpEF patients and concluded that eSBV was significantly higher in obese compared to non-obese HFpEF patients (Table 3). A modest increase in eSBV was observed due to decreased venous capacitance, abnormal right ventricular-pulmonary artery interaction, and obesity (31). Altogether, these findings support the underlying hypothesis that resting and exercise hemodynamics in patients with HF can be improved by decreasing SBV. Whether this decrease is specifically due to dilation of the splanchnic venous system remains unclear.

Table 3.

Baseline characteristics and distribution of blood volume at rest and during exercise

	HFpEF (n=62)	HFrEF (n=14)
Age (years)	68 ± 12	58 ± 13
BMI (kg/m2)	33.3 ± 6.9	31 ± 16
LVEF, %	61 ± 6	21 ± 12
Hypertension n (%)	47 (76)	4 (29)
Atrial fibrillation, n (%)	21 (34)	7 (50)
Diabetes mellitus, n (%)	14 (23)	4 (29)
Creatinine (mg/dl)	1.0 ± 0.4	1.2 ± 0.4
Beta blockers n (%)	29 (47)	14 (100)
ACE-I/ARB	20 (32)	10 (71)
Baseline Stressed Volume (ml)	2494 ± 3162	2664±488
20W Stressed Volume (ml)	3504 ± 2764	3260±362
Peak Stressed Volume (ml)	3795 ± 3610	3243±352

Values are mean ± SD or n (%). BMI indicates body mass index; ACE-I is angiotensin-converting enzyme inhibitor; ARB is angiotensin receptor blocker; HFpEF indicates heart failure with preserved ejection fraction; HFrEF indicates heart failure with reduced ejection fraction. Source: Sorimachi et al. and Fudim et al.

Although both approaches may have similar hemodynamic effects on preload, modulation of GSN to increase venous capacitance has potential advantages over pharmacologic approaches to venodilation since GSN modulation has lesser effects on the systemic circulation, limiting undesirable adverse effects such as headaches and hypotension. Accordingly, though clinical trials of nitrates in HFpEF have not shown benefit with regard to amelioration of HF symptoms or exercise capacity (NEAT-HFpEF, INDIE-HFpEF) (32,33), these data may not inform the benefits of GSN modulation as a strategy for HF management.

Effects of splanchnic nerve blockade

Given that increases of SBV in a functionally impaired ventricle can result in increased filling pressures without a meaningful increase in CO, the concept of splanchnic nerve blockade (SNB) was introduced (22). Investigations into SNB in humans date back to the 1930s where surgical splanchnectomy (denervation) of splanchnic nerves was used to treat uncontrolled hypertension (26,27). Currently, bilateral SNB is used for several clinical conditions, including severe abdominal pain in patients with chronic pancreatitis or carcinoma. Evidence indicates that permanent blockade of the splanchnic nerve is well-tolerated. Reported side-effects of bilateral SNB are minimal and limited to transient diarrhea or abdominal colic and transient orthostatic hypotension (drop in systolic blood pressure of >20mmHg with upright posture) without clinical sequelae. In a recent review of 1,511 published cases, longer-term orthostatic hypotension events were largely limited to <48h with reports of five patients experiencing hypotension for weeks after SNB (34). Transient orthostatic hypotension can be prevented by the intravenous administration of 500 to 1000 mL of crystalloid prior to or immediately following SNB. This is achieved initially by expanding the TBV pool thereby overcoming the increase in splanchnic capacitance. Additionally, long-term pharmacological SNB via botulinum toxin has shown to be efficacious and safe in an instance of refractory hypertension (34). Despite the reported efficacy of SNB in patients without HF, to date, the long-term effects of SNB in patients with HF are unknown.

Short term splanchnic nerve blockade in patients with heart failure

Several studies have examined the alteration of SBV and cardiac filling pressures following SNB (Table 4). The initial studies that investigated GSN blockade in HF used short-term pharmacological blockade. Two small proof-of-concept studies were conducted in patients with acute decompensated HFrEF (Splanchnic HF-1 (NCT02669407; n=11; resting hemodynamics) and with chronic HFrEF (ambulatory HF) (Splanchnic HF-2 (NCT03453151; n=15; exercise hemodynamics). In both studies, patients with known coagulopathies, stage V kidney disease, and those on oral anticoagulants or oral antiplatelet agents were excluded to ensure safety. The primary aim of the Splanchnic HF-1 study was to investigate the impact of short-term SNB on CVP, mean pulmonary arterial pressure (mPAP), and PCWP. The eligible subjects with acute decompensated HFrEF underwent bilateral temporary, percutaneous SNB with lidocaine (duration of action <90 min). The SNB-induced reduction of estimated SBV and systemic vascular resistance lowered resting right- and left-sided filling pressures, yet CO was increased in both ADHF and chronic HF patients. The increase of CO despite reduced cardiac filling pressures in patients with HFrEF may be attributed partially to the reduction of afterload resistance; however, since the average PCWP in both groups of patients was ~27–29 mmHg, it is also plausible that reduction in filling pressure may have driven the heart to a more favorable point on the Frank-Starling curve (i.e., off the descending limb of the Starling curve).

Table 4.

Baseline characteristics and timeline of splanchnic nerve block evidence

Study Title	Publication Year	Primary Endpoint	Age (years)	BMI (kg/m2)	Hypertension (%)	Atrial Fibrillation (%)	Diabetes (%)	LVEF (%)	Patient Population	Intervention
Splanchnic HF-1 (n=11) (10)	2019	Central venous pressure, pulmonary arterial mean pressure, and pulmonary capillary wedge pressure	64 ± 13	32 (22–43)	6 (55)	7 (64)	7 (64)	18 ± 11	Decompensated HFrEF	Splanchnic nerve blockade
Splanchnic HF-2 (n=14) (10)	2020	Exercise capacity measured by peak oxygen uptake, mean pulmonary arterial pressure, pulmonary capillary wedge pressure	58 ± 13	31 (22–56)	4 (29)	7 (50)	4 (29)	21 ± 12	Ambulatory HF	Splanchnic nerve blockade
Malek et al. (n=10) (35)	2020	Reduction in exercise pulmonary capillary wedge pressure	70 ± 10	31 (29–35)	80	90	60	58 ± 10	HFpEF	Thoracoscopic surgery
Shah et al. (n=11) (36)	2020	Quality of life, exercise capacity, diastolic function, and heart failure severity	70 ± 8	-	-	-	-	≥	HFpEF	Endovascular catheter-based ablation
Reddy et al. (n=11) (37)	2021	Quality of life, exercise capacity, heart failure severity	60 ± 10	-	-	-	-	34 ± 4	HFrEF	Endovascular catheter-based ablation
Rebalance-HF (n=80*) (38)	Ongoing	Change in mean pulmonary capillary wedge pressure and device or procedure-related serious adverse events	-	-	-	-	-	-	HFpEF	Endovascular catheter-based ablation

Values are mean ± SD, n (%) or median (Interquartile range).

HF indicates heart failure; HFrEF, heart failure with reduced ejection fraction; HFpEF, heart failure with preserved ejection fraction. Estimated recruitment*

Splanchnic HF-2 aimed to evaluate the effects of percutaneous SNB on exercise capacity, mPAP, and PCWP in ambulatory patients with chronic HF. Injection of ropivacaine (duration of action <24 hours) in this population was associated with a decrease in resting systemic vascular resistance from 1676 ± 692 to 1306 ± 584 dynes/s/cm5, resting PCWP from 27.5 ± 7.3 to 19.1 ± 8.4 mmHg (P < 0.001) and peak exercise PCWP from 34.3 ± 10.1 to 24.4 ± 10.7 mmHg (P < 0.001).At the end of 3 months, the primary endpoint of reduction in PCWP was met. A mean reduction of PCWP of 6.2 mmHg (95% CI −12.2 to −0.2; p<0.05) was seen at 20 W exercise with a similar reduction of −5.1 mmHg (95% CI −10.1 to −0.1; p<0.05) observed at peak exercise. In addition, an improvement was seen in NYHA class and quality of life at 12 months.

Since SBV can actually only be directly measured accurately by inducing circulatory arrest, we employ an indirect method of estimating this quantity based on computer simulations relying on widely used models of the cardiovascular system. eSBV is simulated using heart rate, CO, CVP, PCWP, systolic and diastolic systemic arterial pressure, PAP, and left ventricular ejection fraction. To account for differences in patient sizes, eSBV values are presented as ml/70 kg body weight (27). eSBV is higher in decompensated patients with HFrEF (10) than ambulatory patients with chronic HF and reduced EF (p=0.019). Additionally, in ambulatory patients with HFrEF, SNB decreased eSBV by 532 ± 264 min/70kg (p<0.001) at rest, and with peak exercise, the eSBV decreased from 3243 ± 352 to 2662 ± 656 min/70kg (p<0.001). Therefore, SNB appears to be effective in reducing SBV and may have the potential the outcomes of HF. Of note, four out of five patients who underwent bilateral GSN block reported episodes of symptomatic orthostatic hypotension, but no such association was observed with unilateral SNB (10). The hemodynamic effects of unilateral SNB were comparable to bilateral SNB, however appeared to be attenuated.

Permanent splanchnic nerve blockade in patients with heart failure

Permanent blockade of the splanchnic nerves in patients with HF was investigated by Malik et al. using a minimally invasive thoracoscopic approach (35). Ten patients with HFpEF who had elevated natriuretic peptides and elevated left and right-sided pressures underwent right-sided GSN ablation via thoracoscopic surgery. A trend to lower resting CVP from 10.5 [5,11] mmHg to 6.0 [2,7] mmHg at follow-up of 12 months was observed. Additionally, exercise performance improved with a +1.6 ml/kg/min (95% CI −0.3 to 5.7; p=0.050) increment in peak oxygen consumption. Altogether, unilateral disruption of the right GSN showed favorable outcomes, but over a 12-month follow-up, adverse events related to thoracoscopic GSN ablation included several post-surgical complications such as hematoma, surgical site infection, and also prolonged postoperative length of stay. Concerns regarding the procedure-related morbidity of surgical GSN ablation have fueled interest in less invasive, catheter-based approaches.

Recently, the Satera Ablation system (Axon Therapies) has been examined as a method to achieve permanent nerve blockade using a minimally invasive approach. In 11 patients with HFpEF (3 male, 8 female, 70 ± 8 years), Shah and colleagues reported the feasibility of endovascular catheter-based ablation of the right-sided GSN using the Satera system for treatment. Significant improvements in quality of life, exercise capacity (6-minute walk test, 6MWT), diastolic function, and HF severity were noted without detrimental cardiac effects. Although there was no statistically significant improvement in 6MWT from baseline, a considerable improvement in the distance of at least 30 meters compared to baseline was observed in all patients at follow-up of 1 and 3 months (36). Reddy and colleagues further examined the safety and feasibility of splanchnic ablation in 6 patients with HFrEF (mean age 60 ± 10 years). As in the HFpEF study, the 3-month follow-up data showed notable improvements in quality of life, exercise capacity, and HF severity in patients with HFrEF (37). Following on these favorable preliminary data, a randomized, sham-controlled clinical trial of this approach in HFpEF is underway (Rebalance-HF; ClinicalTrials.gov Identifier: NCT04592445) and is anticipated to reach completion in 2023. The purpose of this trial is to evaluate the safety and initial effectiveness of catheter-based unilateral ablation of the right GSN using the Satera ablation system as a strategy for controlling volume redistribution through targeted venodilation and improved venous capacitance. The primary efficacy endpoint is the change in mean PCWP and any device or procedure-related side effects (38).

Long-term data of GSN blockade in HF are not well understood and limited to pilot studies (28–29). Additionally, it must be noted that preload-sensitive states such as severe right-ventricular dysfunction, and pulmonary arterial hypertension require special consideration when applying preload-reducing therapies. Furthermore, it is conceivable that advanced forms of restrictive cardiomyopathy (e.g., amyloid heart disease) might not tolerate a sudden and/or significant reduction in the ability to recruit preload volume following GSN ablation.

Emerging non-pharmacological preload reduction therapies

In contrast to the therapies reviewed above which decrease SBV, an alternate approach to decreasing cardiac filling pressures is to redistribute blood from the pulmonary circulation to the peripheral circulation. The rate of venous return to the heart can be reduced mechanically through partial or complete occlusion of the superior vena cava (SVC) or the inferior vena cava (IVC). Kaiser and colleagues were the first to conduct a proof-of-concept study to provide initial safety and feasibility of modulating preload through this method in HFpEF patients during exercise. Balloon inflation in the IVC was performed during exercise in 6 HFpEF patients to reduce and maintain pulmonary artery pressures. Titrated partial occlusion of the IVC during exercise significantly reduced pulmonary artery pressures by 25% (68± 7 mm Hg to 51±7 mm Hg) with no significant reduction in CO (14.4 ± 5.9 l/min to 12.8 ±2.9 l/min). Mechanical preload control also led to longer exercise times and reduced respiratory rate (39).

Following on these preliminary data, caval occlusion approaches have been designed for clinical application (40). The PreCARDIA device adopts a minimally invasive approach by using a balloon catheter and pump controller to intermittently occlude the SVC. This device is technically easy to deploy and has been shown to improve urine output (and therefore TBV) by reducing CVP without compromising CO. Improvement in urine output could in parts be driven by a reduction in CVP allowing for a greater renal perfusion gradient. The safety and performance of this device in patients with acute decompensated HF was recently confirmed and expanded on in the VENUS-HF trial (NCT03836079)(41–43). In an alternate though related approach, a clinical trial was conducted to investigate the safety and effectiveness of the Doraya catheter, which provides partial infra-renal IVC occlusion, in patients admitted with acute HF showing poor response to diuretic treatment and continual volume overload (NCT03234647) (44). Similar to PreCARDIA, Doraya is intended for short-term use to decrease CVP and improve diuretic response in hospitalized patients with acute HF.

Pharmacological therapy and device-based therapy to reduce SBV or restrict preload are likely to be complementary approaches to preload modulation in HF. Many pharmacological therapies that modulate SBV are already part of the routine armamentarium of acute and chronic HF management (such as angiotensin II inhibitors, sacubitril-valsartan, nitrates, morphine, and, as most recently demonstrated, levosimendan). Device-based management is likely to be an appropriate adjunct to pharmacological treatment for those patients in whom these therapies alone are inadequate to prevent/reverse decompensation or are contraindicated e.g., due to renal function. Since the effects of pharmacological and device-based therapies on preload reduction may be additive, further study and clinical experience is needed to refine patient selection and optimize strategies to limit side effects (particularly over the long-term) from excessive preload reduction.

In summary, caval occlusion approaches aim to ameliorate HF by reducing cardiac filling pressures through a shift of volume out of the thoracic compartment by creating an obstacle to venous return, without augmenting splanchnic reservoir capacity (Figure 2). These primary reductions of CVP due to partial mechanical venous obstructions enhance diuretic responsiveness and urine output (45) thus providing indirect means of reducing TBV and SBV. While these approaches are primarily being developed for short-term applications, longer-term applications may ultimately be investigated with particular attention to whether there could be adverse effects on other organs such as kidneys and liver due to reduced organs perfusion. If to be used chronically, additional complications can be anticipated such as device-based thrombosis and revascularization.

Figure 2.

The targets and mechanisms of various interventions for cardiac preload reduction. The figure depicts known pharmacological and device strategies to reduce cardiac preload. Notably, the mechanism by which a preload reduction is achieved differs between the device-based solutions and the pharmacological/splanchnic nerve block solution.

ACEI indicates angiotensin-converting enzyme inhibitor; SGLT-2i indicates sodium-glucose cotransporter 2 inhibitors.

Conclusions

Translocation of blood from the splanchnic vascular compartment to the systemic circulation plays a fundamental role in the pathophysiology of acute and chronic HF. Targeting preload via modulation of splanchnic venous capacitance, and through other emerging pharmacological and non-pharmacological therapies has the potential to offer significant advances in the management of HF. Ongoing clinical trials will provide further insights to help guide future clinical use, and establish long-term safety and efficacy of these therapies.

Supplementary Material

Supplemental Publication Material

Supplemental Figure 1. Distribution of the intravascular blood volume between the functional and anatomical compartments. Function: Stressed blood volume (SBV) vs unstressed blood volume (UBV) and anatomical: splanchnic vs non splanchnic