Left ventricular reverse remodeling after transcatheter aortic valve implantation complicated by paroxysmal complete atrioventricular block

Abstract

An 86-year-old man with unremarkable clinical history complaining of asthenia and dyspnea was diagnosed with low-flow low-gradient aortic stenosis [LFLG-AS; left ventricular ejection fraction (LVEF) 40% and transaortic mean gradient 37 mmHg, increasing to 52% and 55 mmHg after dobutamine infusion]. The patient underwent transcatheter aortic valve implantation (TAVI; Edwards CENTERA™ 29, Irvine, CA, USA). The procedure and the following hospital stay were free from complications, with no changes on electrocardiography (ECG).

Six months later, few syncopal episodes occurred. No signs of orthostatic hypotension or neurologic disorders were present. Echocardiography showed normal functioning of the prosthetic valve and recovery of LV systolic function (LVEF 55%). Baseline ECG and 24-h Holter monitoring were unremarkable. An implantable loop recorder (ILR) was implanted to verify the occurrence of paroxysmal conduction disturbances. One month later, during a syncopal episode, ILR interrogation showed a complete atrioventricular (AV) block. Therefore, a dual chamber, single lead pacemaker was implanted.

We are providing the first report of complete AV block occurring months after TAVI, possibly because of reverse LV remodeling following TAVI, with ensuing relative oversizing of the prosthetic valve. This possibility should be considered in patients with syncope not otherwise explained, and previous TAVI, especially in cases of LFLG-AS.

<Learning objective: Complete atrioventricular block can occur even months after transcatheter aortic valve implantation (TAVI), possibly because of left ventricular reverse remodeling following valve replacement, with ensuing relative valve oversizing. This possibility should be considered in patients with syncope not otherwise explained, and previous TAVI, especially in cases of low flow low gradient aortic stenosis. Loop recorder implantation should be considered in this group of patients.>

Introduction

Transcatheter aortic valve implantation (TAVI) is a novel, minimally invasive treatment option for patients with severe aortic stenosis especially in patients who are at high risk for conventional surgery. It can lead to an improvement in symptoms and quality of life but is burdened by risk of complications, most notably complete atrioventricular (AV) block [1], [2]. AV block occurs mostly during or immediately after the valve implantation with the incidence being variable according to the valve characteristics [3], [4]. Herein, we provide the first report of paroxysmal AV block occurring more than six months after TAVI, after a significant recovery from left ventricular (LV) dysfunction with resulting changes in LV outflow tract geometry.

Case report

An 86-year-old man with unremarkable clinical history came to medical attention complaining of asthenia and worsening dyspnea. He was in sinus rhythm with no conduction disturbances. He was diagnosed with low-flow low-gradient aortic stenosis (LF/LG AS) (Fig. 1, panel A and Supplementary Video S1): indeed, baseline LV ejection fraction (LVEF) was 40%, and during low-dose dobutamine infusion (10 μg/kg/min) peak transaortic jet velocity increased from 3.9 to 4.7 m/s, and mean pressure gradient from 37 to 55 mmHg, while LVEF became 52%. Coronary angiography showed normal coronary vessels, and the patient underwent TAVI (Edwards CENTERA™ 29, Irvine, CA, USA) (Supplementary Video S2). The procedure and the following hospital stay were free from complications, with unchanged PR interval (200 ms). The patient was discharged and resumed his normal activities, free from symptoms. Serial transthoracic examinations demonstrated normal functioning of the prosthetic aortic valve, and a progressive recovery of LV systolic function, LVEF becoming 55% six months after TAVI (Fig. 1, panel B and Supplementary Video S3). A few days later, the patient experienced a syncopal episode. Similar episodes followed, with a frequency of around one per month. They had no clear relationship with exercise or body posture, although most of the episodes occurred during mild physical activity. The episodes were not heralded by prodromal symptoms, and ended with a slow recovery of consciousness. No signs of orthostatic hypotension were present and tilt-test was negative. There were no clues of an epileptic disorder at electroencephalography recordings, and no lesions were found at brain computed tomography and magnetic resonance imaging. Carotid artery ultrasound displayed no significant stenosis. Baseline electrocardiogram (ECG) was unchanged (Fig. 2), and 24-h ECG Holter monitoring did not disclose any conduction disturbance, arrhythmias, or pauses. Nonetheless, a loop recorder (Saint Jude Medical Confirm™ DM2100, St Paul, MN, USA) was implanted one year after TAVI to verify the occurrence of a paroxysmal conduction disturbance. After one month, the patient collapsed again. The ECG tracing showed only mild (220 ms) PR prolongation (Supplementary Fig. S1), but loop recorder interrogation disclosed a complete AV block during the syncopal episode (Fig. 3). Therefore, a dual chamber single lead pacemaker (Biotronik Philos II SLR, VDD mode, 60–130 i.p.m., AV delay 200 ms, Berlin, Germany) was implanted. At six-month follow up, the patient was well and had not developed any further syncopal episodes. Pacemaker interrogation revealed normal device functioning with a progressively increasing burden of ventricular pacing (from 14% after one month to 25% four months after the implantation).

Supplementary Video S1

Ecochardiography before transcatheter aortic valve implantation. Four-chamber view.

Supplementary Video S2

Transcatheter aortic valve implantation implantation. The prosthetic valve (Edwards CENTERA™ 29) is implanted under radioscopic guidance in a left anterior oblique 30° projection.

Supplementary Video S3

Recovery from low-flow, low-gradient aortic stenosis. Follow-up echocardiogram six months after transcatheter aortic valve implantation. Four-chamber view.

Supplementary Fig. S1

Electrocardiogram (ECG) after the syncopal episode. ECG obtained soon after the syncopal episode, showing mild (220 ms) PR prolongation.

Fig. 1.

(A) Echocardiographic examination before transcatheter aortic valve implantation (TAVI). Above: Time-velocity graph at Doppler echocardiography denoting severe aortic stenosis. Below: The left ventricular outflow tract (LVOT) in parasternal long-axis view. (B) Control echocardiographic examination 6 months after TAVI. Above: a normal profile of the time-velocity curve is registered. Below: The LVOT diameter is reduced compared to baseline (from 2.2 cm to 1.6 cm).

Fig. 2.

Baseline electrocardiographic recording (sinus rhythm, PR interval 200 ms, non-specific repolarization abnormalities).

Fig. 3.

Implantable loop recorder registration during syncope, demonstrating paroxysmal complete atrioventricular block and wide QRS escape rhythm. Arrows indicate evident P waves.

Discussion

TAVI can lead to a recovery in LV geometry and function (reverse remodeling, RR), and improved quality of life, but is burdened by a risk of AV block [1], [2]. The mechanism of the AV block has been ascribed to the pressure applied by the device on the conducting system in the LV outflow tract [2]. Known risk factors are prostheses extending farther into the LV, pre-existing conduction abnormalities, and valve oversizing [2], [3]. The susceptibility of AV block after TAVI is also device specific, with the self-expanding valves being associated with a higher burden of heart block compared to the balloon-expanded valves. The higher rigidity of and the major extension in the LV outflow tract of self-expanding valves is thought to apply a significant degree of persistent force to the aortic annulus, which is adjacent to the LV outflow tract and may contribute to a higher incidence of post-implant AV block [4]. However, post-implant AV block occurs mostly during or in the hours following the procedure. We are reporting for the first time that complete AV block can occur even months after TAVI. This possibility should be considered when assessing patients with syncope, not otherwise explained, and previous TAVI. Furthermore, the demonstration of RR in our patient came shortly before his first syncopal episode, and it is intriguing to postulate a cause-effect relationship between the two events. Indeed, the implanted valve is a relatively rigid structure, and cannot probably adapt to a reduction in LV outflow tract dimensions; this mechanism could produce a relative valve oversizing, and then a greater compression on the conducting tissues.

Future studies on LF/LG AS patients should verify the correlation between the extent of RR and risk of AV block over the long term, and the predictive value of a greater contractile reserve at dobutamine stress echocardiography, which points to a greater potential for RR. Loop recorder implantation should be considered in patients with syncope not otherwise explained, and previous TAVI, especially in cases of LF/LG AS.

Conflict of interest

None to be declared.