Pulmonary artery denervation for treatment of a patient with pulmonary hypertension secondary to left heart disease

Abstract Abstract

Pulmonary hypertension (PH) predicts poor outcome in patients with left heart disease. A 62-year-old man was referred for heart failure associated with ischemic cardiomyopathy. He received a diagnosis of combined postcapillary and precapillary PH secondary to left heart disease on the basis of hemodynamic parameters. After the pulmonary artery denervation procedure was performed, hemodynamic parameters were markedly improved, which resulted in a significant increase in functional capacity.

Pulmonary hypertension (PH) predicts poor outcome in patients with left heart disease.¹ PH associated with left heart disease (PH-LHD) occurs initially as pulmonary venous hypertension as a result of backward transmission of elevated left ventricular filling pressure. Over a period of time, pulmonary vasoconstriction and vascular remodeling result in sustained elevation of pulmonary vascular resistance.² Currently, no specific therapy is available for PH-LHD. Pulmonary vasodilator drugs found effective for idiopathic pulmonary arterial hypertension have recently been tried in the treatment of PH-LHD, but the results are not very promising.³ Recently, the efficacy and the safety of pulmonary artery denervation (PADN) in treating patients with idiopathic pulmonary arterial hypertension has been reported.⁴ We successfully employed PADN in treating a patient with PH-LHD.

Case Description

A 62-year-old man was referred with a 4-year history of angina pectoris, worsening dyspnea, and easy fatigability. His medical history was significant for hypertension and chronic atrial fibrillation. He denied a history of diabetes mellitus and was a nonsmoker. He underwent coronary angiography, and an 85% stenosis in the ostium of the left anterior descending artery (LAD) was detected on August 21, 2013. A $3.0\times 24$-mm drug-eluting stent (Firebird2; Microport) was successfully deployed in the patient’s LAD. The patient was prescribed aspirin, clopidogrel, atorvastatin, furosemide, spironolactone, metoprolol, and ramipril.

On October 10, 2014, the patient was transferred to our hospital because of progressively worsening dyspnea without typical chest pain. At admission, his blood pressure was 125/68 mmHg, and his pulse rate was 66 bpm. Physical examination findings were remarkable for elevated jugular venous pressure, a positive hepato-jugular reflux, and some crackles at the lung bases. Cardinal signs were an enlarged left and right border of the heart and a grade 3/6 systolic murmur that was heard with maximum intensity at the left parasternal border. A 12-lead electrocardiogram showed atrial fibrillation and left ventricular hypertrophy without axis deviation. Transthoracic echocardiography revealed a marked enlargement of the left atrium at 65 mm, a left ventricle end-diastolic dimension of 83 mm, an ejection fraction of 39%, a markedly enlarged right atrium at $65\times 77$ mm, and a right ventricle with tricuspid regurgitation, with a peak pressure gradient of 78 mmHg. Chronic interstitial lung congestion was found on pulmonary artery enhanced computed tomography without signs of pulmonary embolism. The diameter of the central pulmonary artery is 45 mm (Fig. 1).

Figure 1.

Pulmonary arterial angiography.

The patient received a diagnosis of PH-LHD. He was removed from fluid retention, and his condition improved significantly due to the optimization of drug treatment, including the combination of diuretics (furosemide at a dosage of 20 mg, administered by injection once per day; spironolactone at a dosage of 20 mg twice per day), perindopril (4 mg per day), and metoprolol (zero-order kinetic formulation; 47.5 mg per day). On the fifth day of hospitalization, the dosage of furosemide was decreased to 20 mg per day administered orally, with administration of the other drugs continued at the same dosages. On the same day, the patient underwent another coronary angiography with normal findings. Right heart catheterization (RHC) was performed through the right jugular vein, and the average of five continuous hemodynamic pressures was recorded because this patient had atrial fibrillation. Findings included right atrial pressure (RAP) of 11 mmHg, pulmonary arterial pressure (PAP) of 97.1/38.7 (58.2) mmHg, pulmonary capillary wedge pressure (PCWP) of 18 mmHg, cardiac output (CO) of 2.7 L/min, and pulmonary vascular resistance (PVR) of 15.4 Wood units (Table 1).

Table 1.

Hemodynamic parameters

Parameter	Baseline	Immediately after PADN	24 h after PADN	6 months after PADN
HR, bpm	61	67	61	63
BP, mmHg	96/61	104/67	100/60	105/65
Weight, kg	67	67	67	67
RAP, mmHg	11.0	11	9	8
RVSP, mmHg	98.3	74.2	75.6	48.0
RVDP, mmHg	9.1	10.7	6.4	8.1
PASP, mmHg	97.1	73.9	74.4	47.8
PADP, mmHg	38.7	34.9	38.9	23.0
mPAP, mmHg	58.2	48.3	50.7	31.3
PCWP, mmHg	18.0	16.0	15	12.0
LVEDP, mmHg	16.7	15.1	…	12.3
CO, L/min	2.7	3.2	3.3	4.3
CI, L/min/m2	1.5	1.7	1.8	2.4
PVR, Wood units	15.4	10.4	10.8	4.3

^NoteBP: blood pressure; CI: cardiac index; CO: cardiac output; HR: heart rate; LVEDP: left ventricular end-diastolic pressure; mPAP: mean pulmonary arterial pressure; PADN: pulmonary artery denervation; PADP: pulmonary arterial diastolic pressure; PASP: pulmonary arterial systolic pressure; PCWP: pulmonary capillary wedge pressure; PVR: pulmonary vascular resistance; RAP: right atrial pressure; RVDP: right ventricular diastolic pressure; RVSP: right ventricular systolic pressure.

PADN was performed for the patient with the “out-of-proportion” pulmonary arterial hypertension and increased PVR immediately after RHC was finished and informed consent was signed, as approved by the Nanjing First Hospital ethical committee. A baseline pulmonary artery angiography was performed first (Fig. 1), and then an 8-French long sheath was inserted through the femoral vein and advanced to the main pulmonary artery. In accordance with the details of the PADN procedure and the device, as described elsewhere, the special PADN catheter with 10 electrodes on a circular tip was advanced along this long sheath and was connected with ablation equipment.⁴ The circular tip was positioned between the distal main trunk and the ostial left branch pulmonary artery, where the radiofrequency ablation targets were located. (Fig. 2). The ablation parameters (temperature, 50°C; energy, 10 W; time, 120 seconds) were programmed at each point. The patient reported slight, tolerable chest pain throughout the procedure.

Figure 2.

Pulmonary artery denervation circular catheter and the target location, indicated by red arrows.

Immediately after the PADN procedure, RHC parameters showed improvement, with PAP decreased to 73.9/34.9 (48.3) mmHg, CO increased to 3.2 L/min, and PVR decreased to 10.4 Wood units. The patient was monitored in the cardiac care unit. The hemodynamic parameters were retested after 24 hours, and the improvement in mean PAP, CO, and PVR remained. These improvements were maintained through 6 months of follow-up. RHC was performed again at 6 months after PADN. RAP was 8 mmHg, PAP was 47.8/23.0 (31.3) mmHg, and PCWP was 12 mmHg. CO was 4.33 L/min. PVR was 4.33 Wood units.

Echocardiographic measurements were improved accordingly. The left heart kept the same size, and the ejection fraction remained 40% at 6 months. The right atrium and right ventricle showed no dilatation. The tricuspid excursion index was significantly improved from 0.55 at baseline to 0.36 at 24 hours and 0.24 at 6 months. The peak pressure gradient of tricuspid regurgitation was 63 mmHg, indicative of a systolic PAP of 78 mmHg at baseline, and it was decreased to 58 mmHg at 6 months. Thus, systolic PAP decreased 20 mmHg in this 6-month period. The mean PAP approximated this, with peak early diastolic pressure between the RV and pulmonary artery decreased from 51 mmHg at baseline to 37 mmHg at 24 hours and 35 mmHg at 6 months (Fig. 3).

Figure 3.

Echocardiographic images. A, Pulmonary arterial systolic pressure (PASP) at baseline. B, Mean pulmonary arterial pressure (mPAP) at baseline. C, Tricuspid excursion index (Tei) at baseline. D, PASP at 1 month. E, mPAP at 1 month. F, Tei at 1 month. G, PASP at 6 months. H, mPAP at 6 months. I, Tei at 6 months.

N-terminal brain natriuretic peptide level was decreased from a baseline of 2,005 pg/mL to 522 pg/mL at 6 months after the procedure. Six-minute walk distance was increased from a baseline of 245 m to 401 m at 1 month, 452 m at 3 months, and 483 m at 6 months.

Discussion

PH is most common among patients with left heart disease with preserved or reduced ejection fraction. PH-LHD, or group 2, is defined as an increase in mean PAP of >25 mmHg at rest secondary to an elevation in PCWP of ≥15 mmHg. PH develops in LHD in response to a passive backward transmission of filling pressures. In some patients, these purely mechanical components of venous congestion may trigger a superimposed component, combining pulmonary vasoconstriction, decreased nitric oxide availability, increased endothelin expression, desensitization to natriuretic peptide–induced vasodilation, and vascular remodeling. This results in a further increase in mean PAP in excess of the elevation of pulmonary artery wedge pressure and leads to PVR > 3 Wood units or transpulmonary pressure gradient > 15 mmHg or diastolic pressure difference > 7 mmHg.^2,5-6

There are currently no consensus therapeutic strategies for the treatment of PH-LHD. Current pharmacological therapies for heart failure, such as vasodilators and diuretics, may improve “out-of-proportion” PH through a reduction in filling pressures and functional mitral insufficiency.

The use of prostaglandins in systolic heart failure in the Flolan International Randomized Survival Trial showed a strong trend toward decreased survival with intravenous epoprostenol and led to the trial being prematurely terminated.⁷ According to the present evidence, including findings from the ENABLE (Endothelin Antagonist Bosentan for Lowering Cardiac Events in Heart Failure) study, HEAT (Heart Failure Endothelin A Receptor Blockade) study, EARTH (Endothelin A Receptor Antagonist Trial in Heart Failure) study, and VERITAS (Value of Endothelin Receptor Inhibition with Tezosentan in Acute Heart Failure) study, there is no clinical support for the use of nonselective and selective endothelin-1 antagonists for the treatment of PH-LHD.⁸ Small sample trials of phosphodiesterase type 5 inhibition with sildenafil in PH-LHD have proved the potential benefit of this treatment, but large morbidity and mortality trials are still required to assure the role of these agents.⁹ Therefore, pharmacological studies with traditional pulmonary vasodilators failed to demonstrate a benefit in patients with PH-LHD.

The effects of neurohormonal antagonists on PH have not been systematically evaluated. Anatomic and pathological studies have confirmed that baroreceptor structures exist in the pulmonary artery and that the efferent limb of the reflex is predominantly mediated via the adrenergic nervous system.¹⁰ An animal experiment showed that, during occlusion of the left interlobar pulmonary artery, PAP and PVR gradually increased and reached peak values after 5 minutes, and it showed that this pressure response was completely abolished after PADN at the main pulmonary artery bifurcation level.¹¹ Rothman et al.¹² showed that PADN reduces PAP and induces histological changes in an acute porcine model of PH. In two articles, Chen et al.^4,13 reported for the first time the effect of PADN on functional capacity and hemodynamic characteristics in patients with PAH not responding optimally to medical therapy in the first PADN-1 study involving human subjects and reported phase II results from the PADN-1 study.

In conclusion, we provide a possible method to improve the hemodynamic parameters and functional capacity of patients with PH secondary to heart failure. Additional studies are required to confirm the safety and efficacy of PADN for PH-LHD.

Source of Support: This work was supported by projects of the Nanjing Special Program of Science (201402045) and Jiangsu Provincial Special Program of Medical Science (BL2013001).

Conflict of Interest: None declared.