ECG signs mimicking acute inferior wall myocardial infarction are associated with elevated myocardial enzymes during isolation of pulmonary vein for focal atrial fibrillation

Abstract

In this study, we report an intraprocedural incident in patients undergoing ablation for atrial fibrillation. During left atrial manipulation our patients suffered from acute chest pain, showed ECG signs of an acute inferior wall myocardial infarction, and increased levels of cardiac Troponin I (cTnI). We strongly recommend being aware of unexpected reactions during isolating pulmonary veins for focal atrial fibrillation, especially when passing the dorsal part of the left atrium. If pericardial effusion is ruled out and ECG signs as well as symptoms disappear, the ablation procedure should proceed. We think patients undergoing pulmonary vein ablation for atrial fibrillation should be informed of this threatening complication.

pulmonary vein isolation, ablation, atrial fibrillation, complete heart block, myocardial infarction

Introduction

Treating patients with paroxysmal or intermittent atrial fibrillation (AF) always has been challenging for physicians. The catheter ablation of AF by isolating pulmonary veins (PVs) has become a promising approach by the use of an innovative technique. In addition to PV stenosis, thromboembolism and pericardial tamponade are potential complications during or following the procedure [1–,3] .

Methods

We investigated 13 consecutive patients undergoing ablation of focal AF regarding differences in periprocedural ECG tracings and myocardial cell damage. The ablation procedure was performed with a specially designed catheter system (REVELATION Helix, Cardima Inc.) [4] . This system provides circumferential mapping and radiofrequency delivery capabilities in one single catheter. This catheter was inserted in a deflectable catheter sheath (NAVIPORT, Cardima Inc.). The sheath is 8 French in diameter and provides continuous purging. All patients had failed two antiarrhythmic drug treatments at least including class I or III drugs. Before starting the ablation procedure, all patients were in sinus rhythm and treated by the same ablation protocol. The anticoagulation therapy consisted of 70 IE/kg weight i.v. after transseptal puncture. Activated clotting time was held between 250 and 300 s throughout the whole procedure. To assess possible myocardial cell damage, levels of cardiac Troponin I (cTnI, fluorometric enzyme immunoassay) were measured just before ablation, 30 min, 60 min, 90 min, 6 h, and 24 h after successful ablation. During our study we were able to identify two patients who experienced dramatic changes in the ECG, decrease in blood pressure, and were highly symptomatic with chest pain and dizziness.

Results

The first patient was a 42-year-old woman with the primary onset of AF 9 years before ablation. The diameter of the left atrium was 47 mm and no structural heart disease was present. We started the PV ablation with the left upper pulmonary vein (LUPV). The intervention was performed with the catheter held in an ostial position. The guiding sheath with a deflectable tip enabled us to guide the system very easily and smoothly to the intended position. No repositioning before and during ablation occurred. Both left-sided PVs were successfully isolated, i.e. PV potentials vanished and were absent for more than 10 min ( Fig. 1 ). After drawing back the catheter into the sheath, it was moved towards the right upper pulmonary vein (RUPV). While the guiding sheath was passing along the dorsal wall of the left atrium, suddenly complete heart block occurred for 5 beats ( Fig. 2 ). After that, ST-segment elevation in leads II, III and aVF, associated with angina pectoris, occurred. ST-segment elevations were observed for about 5 min. The ablation procedure was stopped immediately and three doses of nitroglycerin using a pump spray were administered. Thereafter, symptoms improved and disappeared after 10 min. An immediate echocardiogram ruled out pericardial effusion. Even 1 h post-ablation no pericardial effusion was detected. Due to signs of acute inferior myocardial infarction coronary angiography was performed. The acquired images excluded significant coronary artery stenosis. Subsequently, the ablation procedure was continued and terminated successfully after ablation of the RUPV. Thereafter, the patient was transferred to the intensive care unit. Over a 48 h period no signs of rhythm disturbances, pericardial effusion, or ST-segment changes were observed.

Figure 1

Pulmonary vein potential elimination in the left upper pulmonary vein. Radiofrequency application using a maximum power output of 30 W at a target temperature of 50 °C to persistently abolish the pulmonary vein (PV) potential (+) from the left upper PV. After PV elimination, the far-field left atrial signal (∗) remained detectable inside the PV during coronary sinus pacing. (Surface ECG lead II, DCS: distal coronary sinus and eight bipolar circumferential recordings from the left upper pulmonary vein: PV 1/2–8/1, paper speed: 25 mm/s.)

Figure 2

ECG and intracardiac recordings during the movement towards the right upper pulmonary vein illustrate complete heart block for 5 beats. At the top leads I, II, III, and V1 are shown. Intracardiac leads reveal an activation of the high right atrium (HRA; paper speed: 25 mm/s). P-waves without succeeding QRS complexes can be easily identified.

Our second patient, a 45-year-old man, had experienced paroxysmal AF for 44 months. The left atrium was 40 mm in diameter and no structural heart disease was present. During ablation the left-sided PVs were isolated successfully. The sheathed catheter was turned towards the RUPV. Throughout this manoeuvre, angina pectoris with ST-segment elevation in leads II, III, and aVF occurred ( Fig. 3 ). Acute i.v. administration of a calcium channel blocker resolved anginal symptoms within 15 min. Cardiac ultrasound ruled out an acute pericardial effusion. Due to the suspected acute myocardial infarction coronary angiography was performed. No significant stenosis of any coronary artery could be detected. After cessation of angina the ablation procedure was continued and terminated as planned. Subsequently, the patient was sent to the intermediate care unit where no changes in the ECG or echocardiography were observed.

Fig. 4 shows the readings of cardiac enzymes prior to and post-ablation. It clearly demonstrates the difference between control patients (control Pts.) and patients presenting ECG changes during the ablation (Pt. #1, Pt. #2). At our laboratory a cTnI level less than 0.1 ng/ml is considered normal.

Figure 3

ECG tracings demonstrate ST-segment elevations (∗) in the inferior leads during intubation of RUPV. Records are performed while pacing the low right atrium with a cycle length of 870 ms (HRA: high right atrium, CS: coronary sinus, paper speed: 50 mm/s).

Figure 4

Levels of cardiac Troponin I (cTnI, ng/ml) prior to and after ablation. No change in the control patients (control Pts., solid triangles) is shown. Patients #1 (Pt. #1, solid circles) and #2 (Pt. #2, solid squares) demonstrate a maximum 6 h after the beginning of the procedure.

Discussion

The isolation of pulmonary veins for the ablation of focal AF has become very promising and successful. Recurrence rates differ from 20 to 45% during a follow-up period of 5–12 months [5–,7] . Complications which may occur include PV stenosis, thrombus formation, thromboembolic events, and coagulum on the catheter. Our two patients were identified with a yet undescribed combination of ECG changes and elevation of cTnI levels. Only few reports on ST-segment elevations during percutaneous interventions exist. One study reports on eight patients who suffered from transient ST-segment changes during mitral valve valvuloplasty [8] . A recent article by Simon and Gill describes three patients with ST-segment changes during radiofrequency ablation for tachycardias arising in the left atrium [9] . In both studies multiple manipulations inside the left atrium had to be performed to achieve an optimal intervention. In our study, intraatrial movements were performed with the ablation catheter inside a steerable guiding catheter. While moving along the dorsal part of the left atrium we observed ECG signs of an acute inferior wall myocardial infarction with ST-segment elevation, complete heart block, and anginal symptoms. The complete heart block lasted a few beats, but ST-segment elevation persisted for 5–10 min, so did angina. The immediately performed angiography excluded significant coronary artery disease. The documented ECG signs, as well as the symptomatic angina, therefore may reflect reactions to coronary spasm. The relief of symptoms after i.v. administration of a calcium channel blocker supports this explanation. Also, a strong vagal reaction due to the movement of the catheter might be responsible. This induced hypervagotonia results in a release of acetylcholine (ACh). The powerful substance ACh is known to cause spasm in the coronary circulation [10] . In our setting levels of ACh were not measured.

The levels of cardiac enzymes, i.e. cardiac Troponin I levels, were elevated with a maximum of 0.3 ng/ml in patient #1 and 0.29 ng/ml in patient #2, 6 h after successful ablation. The control patients showed a slight elevation over the threshold value (0.12 vs. 0.1 ng/ml).

Air embolism might be another reason for the reported findings. Different authors reported typical complications of air embolism in the right coronary artery during ablation in the left atrium with a transseptal approach [11,,12] . However, the continuous irrigation of the sheath with heparinized saline and a slow removal of the ablation catheter from its sheath should prevent air bubbles from reaching the tip of the sheath. Therefore, air embolism seems unlikely, but we cannot entirely exclude this mechanism. In our view, the difference in cTnI levels in the two-presented patients reflects cardiac cellular damage during coronary spasm for which a sudden increase in vagal tone is responsible.

We are grateful to Andrea Wolfertz for carefully reading the manuscript.