Angiographic findings and treatment of coronary sinus spasm during left ventricular lead placement: A case report

Abstract

A 71-year-old man with ischemic cardiomyopathy, a left ventricular (LV) ejection fraction of 23 %, left bundle branch block with a QRS duration of 160 milliseconds, and nonsustained ventricular tachycardia was admitted for cardiac resynchronization therapy combined with an implantable defibrillator. During LV lead placement, the guiding sheath encountered strong resistance during deep coronary sinus (CS) cannulation. CS venography showed a complete occlusion, and we diagnosed venospasm because the occlusion self-resolved after several minutes. After administering intravenous isosorbide dinitrate and waiting several minutes without manipulating the catheters, we could successfully place the LV lead in the target branch. Although CS spasm is considered a rare complication of LV lead placement, in some cases catheter manipulation can trigger it. Therefore, clinicians should recognize its possibility and be aware of the associated angiographic findings and treatment.

Learning objective

In some cases, coronary sinus (CS) spasm can be triggered during left ventricular lead placement. It presents as occlusion with abrupt tapering on CS venography. After administering intravenous nitrates and waiting several minutes without manipulating the catheters, the spasm resolves and the catheter can be advanced.

Introduction

Cardiac resynchronization therapy (CRT) is used for cardiac pacing in patients with left ventricular (LV) systolic dysfunction and dyssynchronous ventricular activation. Because the posterolateral wall is often the last segment to contract in LV dyssynchrony in the presence of left bundle branch block (LBBB), the optimal LV lead position is typically in the lateral or posterolateral branches of the coronary sinus (CS). However, LV lead placement in the target branch can be difficult in patients with variations of the coronary venous anatomy, valves of Thebesian and Vieussens, and venous dissection [1]. Herein, we report a case in which CS spasm occurred during LV lead placement and made it difficult to place the lead in the target branch. We describe the angiographic findings and treatment of CS spasm and discuss similar case reports.

Case report

The patient was a 71-year-old man with ischemic cardiomyopathy who had been on maintenance hemodialysis for end-stage renal disease. He had been taking bisoprolol and amiodarone orally for nonsustained ventricular tachycardia. Although he had undergone catheter ablation for persistent atrial fibrillation (AF) 3 years earlier at our hospital, he was admitted to our hospital 6 months before the current admission because of worsening heart failure with severe functional mitral regurgitation (FMR) and recurrent AF refractory to repeated electrical defibrillation (heart rate, 118 beats per minute). Hemodialysis did not remove sufficient fluid because of low systolic blood pressure (80 mmHg), his heart failure did not improve and he became inotrope dependent (dobutamine 3 γ). Catheterization showed post-capillary pulmonary hypertension (mean pulmonary artery wedge pressure, 35 mmHg; mean pulmonary atrial pressure, 38 mmHg) and low cardiac output (cardiac index measured by the thermodilution method, 1.82 L/min/m²), although de novo lesions were not found in coronary arteries. Because the surgical risk for FMR was high (Society of Thoracic Surgeons score, 29.9 %), we referred the patient for transcatheter mitral valve repair.

At the other hospital, a MitraClip (Abbott, Abbott Park, IL, USA) was implanted for FMR, and a second session of catheter ablation was performed for persistent AF. Postoperatively, mitral regurgitation improved from severe to mild, the patient could be weaned off dobutamine, and the sinus rhythm was maintained. However, despite optimal medical treatment, the patient had residual heart failure symptoms [New York Heart Association (NYHA) functional class III]. Echocardiogram showed severe LV systolic dysfunction, with an ejection fraction of 23 %, and 12‑lead electrocardiogram (ECG) showed a complete LBBB with a QRS duration of 160 milliseconds. Therefore, the patient was readmitted to our hospital for CRT combined with an implantable defibrillator.

The right ventricular pacing/defibrillator lead was first implanted in the right ventricular apex. Next, a hook-shaped guiding sheath (Selectra Extended Hook-45; Biotronik, Berlin, Germany) with a 5-Fr quadripolar electrophysiological (EP) catheter (Abbott Laboratories, St. Paul, MN, USA) was introduced via the left subclavian vein and inserted into the CS ostium. The EP catheter was advanced into the great cardiac vein (GCV) without difficulty, but the guiding sheath encountered strong resistance when it was inserted over the catheter and could not be advanced to more distal portions. It was exchanged for another guiding sheath (Selectra BIO2-45; Biotronik) with a more gentle curved shape, which also encountered resistance at deep CS cannulation. CS venography with a balloon-tipped occlusion catheter showed a complete occlusion with abrupt tapering (Fig. 1A).

Fig. 1.

(A) Coronary sinus (CS) venography performed with a balloon-tipped occlusion catheter showed a complete occlusion with abrupt tapering (white arrow). (B, C) After 9 min of CS occlusion, venography from the middle cardiac vein showed a patent CS lumen (black arrows) and tributary veins suitable for left ventricular lead placement (red arrows).

CS, coronary sinus; ICD, implantable cardioverter defibrillator; LAO, left anterior oblique projection; MCV, middle cardiac vein; RAO, right anterior oblique projection.

Although the cause was unknown, we pursued LV lead placement from the middle cardiac vein (MCV) because no venous dissection was detected. After 9 min of CS occlusion, venography from the MCV showed a patent CS lumen and tributary veins suitable for LV lead placement (Fig. 1B, C). Because the occlusion resolved naturally, we diagnosed venospasm. We were concerned that the approach from the MCV might not allow the LV lead to pass through the distal end of the tributary veins to which the MCV and CS were connected, so the previous guiding sheath was reinserted into the CS ostium. A 0.014-in. guidewire (SION blue; Asahi Intecc Co., Ltd., Aichi, Japan) was successfully passed through the target mid-lateral branch. However, when we continued to advance the quadripolar LV lead (Sentus ProMRI OTW QP L-85/49; Biotronik) over its guidewire, the lead could not be advanced beyond the previous occlusive site with strong resistance (Fig. 2A, B). We had determined recurrent spasm, we administered 500 μg isosorbide dinitrate intravenously and, at the same time waited 5 min without manipulating the catheters. Subsequently, we could advance the tip of the lead to the proximal portion of the mid-lateral branch (Fig. 2 1 → 2). However, the lead became stuck again and we considered that the spasm was repeated. Because isosorbide dinitrate deceased his systolic blood pressure to 60 mmHg, we could not administer another dose. Therefore, we simply waited without manipulating the catheters and then advanced the lead as far as possible. We repeated this procedure and could gradually move the tip of the lead forward (Fig. 2 2 → 3 → 4). Approximately 15 min after passage of a micro-guidewire, we successfully placed the lead in the target branch and obtained the optimal pacing threshold without phrenic nerve stimulation. Finally, we implanted the atrial lead in the right atrial appendage. The CRT programming was DDD 70. Atrial-ventricular (AV) timing was adjusted by CRT AutoAdapt (Biotronik), which automatically and continuously adjusts the AV delay and sets the ventricular pacing configuration to biventricular or LV pacing, ventricle-ventricle timing was set with a fast LV activation of 20 milliseconds, and multipole pacing was added.

Fig. 2.

(A, B) The left ventricular lead could not pass through the distal coronary sinus (CS) by the recurrent spasm (black arrows).

The tip of the lead was advanced to the proximal portion of the mid-lateral branch after intravenous administration of isosorbide dinitrate and waiting 5 min without manipulating the catheters at the same time (1 → 2). The tip of the lead was gradually advanced after simply waiting without manipulating the catheters (2 → 3 → 4) and placed in the target mid-lateral branch.

LAO, left anterior oblique projection; RAO, right anterior oblique projection.

At the 6-month follow-up, chest X-ray showed appropriate positioning of the leads, and 12‑lead ECG showed a shortened QRS duration of 140 milliseconds. His symptoms were also improved to NYHA functional class II.

Discussion

Excessive manipulation of wires or leads during LV lead placement can precipitate CS spasm, thereby making it difficult to position the LV lead in the target branch. The literature includes reports of 5 such cases [2], [3], [4], [5], [6] and the demographic, clinical, and treatment characteristics are summarized in Table 1.

Table 1.

Characteristics of 6 cases of coronary sinus spasm.

Reference	Patient characteristics Age (y), sex Underlying cardiac disease	Location of spasm	Treatment
Woollett et al., 2005 [2]	48, M Idiopathic dilated cardiomyopathy	Distal CS	Nitroglycerin into CS (did not relieve spasm); venography balloon dilatation
Russo et al., 2007 [3]	60, M Idiopathic dilated cardiomyopathy	Distal CS	Nitroglycerin into CS
Heist et al., 2007 [4]	48, M Ischemic cardiomyopathy	Distal CS	Nitroglycerin into CS and intravenously
Ellis et al., 2010 [5]	62, M Nonischemic cardiomyopathy	Lateral branch	Nitroglycerin 100 μg into CS
Kapoor et al., 2011 [6]	48, M Idiopathic dilated cardiomyopathy	Posterolateral vein	Nitroglycerin injection (failed to restore flow); balloon angioplasty
Our case	71, M Ischemic cardiomyopathy	Distal CS	Isosorbide dinitrate 500 μg intravenously and waiting several minutes without manipulating catheters

CS, coronary sinus; M, male.

A review of these cases highlights several notable considerations. First, CS spasm presents as occlusion with abrupt tapering on CS venography. Angiographic findings were similar in all 4 cases, with spasm at the distal CS. The abrupt tapering and occlusion likely reflect contraction of both the smooth muscle layer of the venous wall and a cuff of striated muscle extending from the CS ostium to the GCV [7]. When difficulties occur during LV lead placement, clinicians need to presume various conditions including variations of coronary venous anatomy, valves of Thebesian and Vieussens, venous dissection, and this CS spasm. Therefore, it is important to recognize its characteristic angiographic findings in their differential diagnosis.

Second, administering nitrates directly into the CS resolved the spasm in 3 of the 5 earlier cases, but angioplasty was necessary in the 2 cases that were refractory to nitrates. In our case, we initially considered administering isosorbide dinitrate directly into the CS, but instead administered it intravenously because, as a result of the spasm, we could not confirm backflow of blood from the guiding sheath. The spasm was temporarily relieved after administering nitrate, but it was repeated afterward. As it had the characteristics with self-resolution after several minutes, by waiting without manipulating the catheters, we could gradually advance the LV lead and place it in the target branch. Balloon angioplasty is reported to be effective for CS spasm, although the procedure is associated with venous rupture as a severe complication [8]. Therefore, simply waiting without manipulating the catheters is considered to be the most appropriate approach for CS spasm refractory to nitrates.

Last, although CS spasm occurred during LV lead placement, we did not encounter any problems during the initial passage of the EP catheter before insertion of the guiding sheath or during electrode placement into the CS during previous AF ablation. At first, we thought that a large guiding sheath inserted into the CS caused the spasm by physical contact, but it also occurred after passage of a micro-guidewire. Therefore, the proposed mechanism was that the irritation of the initial passage of the first catheter caused the spasm and the second catheter could not pass through the CS. Considering the patient's clinical course, we believe that administration of nitrates and waiting without manipulating the catheters relieve the spasm even in the presence of a catheter in the CS.

There are several reports of CS spasm, but few explain its characteristics and underlying mechanism. We emphasize it because these understandings are important for the management of CS spasm.

Declaration of competing interest

The authors have no conflict of interest to declare regarding this literature.