Table 1. Postoperative Ventricular Arrhythmias

Transcription

1 Unanticipated Postoperative Ventricular Tachyarrhythmias Irving L. Kron, M.D., John P. DiMarco, M.D., Ph.D., P. Kent Harman, M.D., Ivan K. Crosby, M.D., Robert M. Mentzer, Jr., M.D., Stanton P. lan, M.D., and Harry A. Wellons, Jr., M.D. ABSTRACT Eighteen (1.4%) of 1,251 patients who underwent cardiac operations during a three-year period had new sustained ventricular tachycardia (12 patients) or ventricular fibrillation (6 patients) not caused by but resulting in hemodynamic compromise. In 13 patients, the initial arrhythmia occurred in the first 48 hours postoperatively. Lidocaine was being administered to 10 of these patients for suppression of previously noted ventricular ectopy, but it did not prevent the occurrence of the arrhythmia. The initial episode was fatal for 5 patients. Two of these deaths were directly related to the adverse effects of the antiarrhythmic agents used to suppress ventricular tachycardia or fibrillation. Five of 10 survivors underwent electrophysiological studies after initial resuscitation. In all 5, programmed ventricular stimulation reproduced the clinical arrhythmia. There have been 2 late sudden deaths in patients who either did not undergo or remained uncontrolled at electrophysiological study during serial drug trials. Our experience suggests that a cardiac operation may unmask or induce potentially lethal arrhythmias that previously had not been apparent. Pharmacological suppression of ventricular ectopy does not necessarily prevent ventricular tachycardia or ventricular fibrillation in the early postoperative period. Electrophysiological study may be helpful in determining the appropriate prophylactic therapy in such patients. The development of ventricular tachycardia or ventricular fibrillation during the postoperative period is a dreaded, but poorly understood, complication of cardiac operations. In many cases, the occurrence of lifethreatening arrhythmias appears to be secondary to refractory shock from congestive heart failure. In some patients, the hemodynamic variables are normal prior to the episode. The prevalence, the short-term and longterm effects on prognosis, and the mechanisms of these arrhythmias have not been well defined [l]. In this report, we review our three-year experience with unexpected postoperative ventricular tachyarrhythmias. From the Departments of Surgery and Internal Medicine, University of Virginia Medical Center, Charlottesville, VA. Presented at the Thirtieth Annual Meeting of the Southern Thoraac Surgical Association, Marco Island, FL, v 3-5, Address reprint requests to Dr. Kron, Box 181, Department of Surgery, University of Virginia Medical Center, Charlottesville, VA Material and Methods Between April, 1980, and April, 1983, 1,251 patients underwent open-heart operations at the University of Virginia Hospital. Retrospective chart review identified 18 patients in whom unanticipated ventricular tachycardia or ventricular fibrillation developed in the first six weeks after operation. The arrhythmia had to result in but was not caused by hemodynamic deterioration. All patients with known preoperative ventricular tachycardia or ventricular fibrillation were excluded from the study. Subsequent intracardiac electrophysiological studies were performed on 5 patients within a week of the initial episode of postoperative ventricular tachycardia or ventricular fibrillation using techniques previously described [2, 31. In these patients, long-term antiarrhythmic drug therapy was selected on the basis of the results of serial electrophysiological testing [4]. For the remaining hospital survivors, long-term antiarrhythmic therapy was selected empirically. Follow-up of all patients was continued over a mean postoperative period of 18 months. Results lnitial Arrhythmia and Predisposing Factors Eighteen (1.4%) of 1,251 patients who underwent cardiac operations during the three-year period had unanticipated, sustained ventricular tachycardia or ventricular fibrillation postoperatively (Table 1). The initial episode occurred within 48 hours of operation in 13 patients, within 2 to 6 days after operation in 3 patients, and four weeks and six weeks after operation in the remaining 2 patients. In 6 patients, the initial arrhythmia was ventricular fibrillation; in the other 12, uniform morphology ventricular tachycardia was documented. All patients were normotensive immediately prior to the onset of the initial episode of ventricular tachyarrhythmia. Ten patients were still in the surgical intensive care unit at the time of the arrhythmia. In these 10 patients, the mean cardiac index prior to the arrhythmia was 1.9 Wmin/m2. Three of them were receiving dobutamine hydrochloride for inotropic support. The 8 remaining patients were hemodynamically stable and were not being monitored invasively immediately prior to the initial episode. Three patients had experienced intraoperative myocardial infarctions. In 2 of them, the acute onset of chest pain precipitated urgent myocardial revascularization. In 2 of the 3 patients, ventricular fibrillation developed 1 hour after operation; the other experienced ventricular fibrillation during a treadmill exercise test six weeks 317

2 318 The Annals of Thoracic Surgery Vol 38 4 October 1984 Table 1. Postoperative Ventricular Arrhythmias Patient Timing of Initial Drugs before Long-term. Operation Previous MI Arrhythmia Arrhythmia Outcome" Therapy CABG x 1 CABG x 1 CABG x 1 + aneurysm resection CABG x 5 VF, 2 hr ne Early death... VF, 6 hr ne and well ne + evolving VF, 1 hr Lidocaine Early death... new one Np; evolving new VF, 1 hr Lidocaine Early death CABG x 2 one + intraopera- VF. 6 wk ne and well EPS, then mextive MI ilitene + pro- AVR + MVR cainamide VT, 2 hr Nitroprusside + and well lidocaine AVR VT, 4 hr Nitroprusside + Early death fol-... lidocaine lowing brety- AVR VT. 1 hr Lidocaine lium Early death following Prones- AVR VT, 2 hr Dobutamine tylb Severe hypotension following bretylium CABG x 3 VT. 18 hr Dobutamine + Late death lidocaine CABG x 5 VT, 24 hr Dobutamine + Early death lidocaine Lidocaine Late death CABG x 3 + aneurysm resection CABG X 3 CABG X 3 CABG X 3 VF, 4 d; VT on procainamide EPS, then amiodarone VT, 6 d... Early death... VT, 4 wk... and well EPS, then pro- VT, 48 hr '... and well cainamide EPS, the quinidine VT, 1 hr Lidocaine Early death... CABG x 1 + aneurysm resection CABG x 3 VT, 2 hr Lidocaine Severe hypoten- sion following CABG X 3 + AVR bretylium VT, 5 d... and well EPS, then amiodarone 'Early death signifies death during initial hospitalization. bi'rocainamide hydrochloride. MI = myocardial infarction; CABG = (coronary artery bypass grafting; VF = ventricular fibrillation; AVR = aortic valve replacement; MVR = mitral valve replacement; VT = ventricular tachycardia; EI'S = electrophysiological study. postoperatively. Two other patients with ventricular fibrillation had transient ST segment changes suggestive of coronary spasm immediately preceding the onset of the arrhythmia, but myocardial necrosis could not be documented by subsequent myocardial enzyme determination. One other patient, without intraoperative myocardial infarction, collapsed on the fourth postoperative day and was found to be in ventricular fibrillation. This patient subsequently experienced multiple episodes of uniform morphology ventricular tachycardia while on a regimen of antiarrhythmic drug therapy during the hospital course. Eight of 12 patients with uniform morphology ventricular tachycardia had previous transmural myocardial infarctions. Most of them had depressed preoperative ejection fractions (38.5 f 15%) (standard deviation). The

3 319 Kron et al: Unanticipated Postoperative Ventricular Tachyarrhythmias Table 2. Recurrent Arrhythmia in Hospital Despite Antiarrhythmic Therapy in 9 Patients Surviving First Episode of Arrhythmia Patient. Antiarrhythmic Therapy" Result EPS flidocaine Lidocainehretylium Lidocaine /amiodarone hretylium flidocaine Early death Early death Late death Late death Early death 'Administered prior to recurrent arrhythmia. EPS = electrophysiological study. remaining 4 patients had valvular heart disease with marked left ventricular hypertrophy. ne of these patients had enzymatic or electrocardiographic evidence of intraoperative or postoperative myocardial infarction. Medications prior to the onset of ventricular tachycardia or fibrillation included dobutamine in 3 patients, sodium nitroprusside in 2 patients, and digoxin given preoperatively in 8 patients. Digoxin levels measured preoperatively in all of these patients were within the therapeutic range. electrolyte abnormalities were identified. Ten of the 18 patients were receiving lidocaine hydrochloride because of ventricular ectopy noted prior to the first episode of sustained ventricular tachycardia or ventricular fibrillation. Early Mortality, Morbidity, and Recurrence of Arrhythmia The initial episode of ventricular tachycardia or ventricular fibrillation was fatal for 5 patients. Nine of the 13 patients who survived the first episode had recurrence subsequently. All 9 were receiving antiarrhythmic therapy at the time of the second episode. Three of these repeat episodes were fatal (Table 2). Only 4 patients who survived the initial arrhythmia had no further recurrences. The overall early mortality for this group of 18 patients was 44% (8 patients). Two patients who were in stable condition at hospital discharge died suddenly six months postoperatively. One was receiving procainamide hydrochloride, and one was receiving no antiarrhythmic therapy after side effects developed from amiodarone (Patients 10 and 12; see Table 1). Four patients experienced severe adverse reactions to the antiarrhythmic drug therapy begun after the initial episode of ventricular tachycardia or ventricular fibrillation. Three patients became severely hypotensive after receiving a loading dose of bretylium tosylate. One died, and the other 2 survived after prolonged resuscitation. Refractory hypotension developed in an additional pa- tient, who died after a rapid intravenous loading dose of procainamide. Electrophysiological Studies and Late Results Five of the 10 hospital survivors underwent electrophysiological studies. In each, programmed ventricular stimulation reproduced the same arrhythmia observed clinically (Figs 1, 2): sustained ventricular tachycardia in 3 patients and nonsustained polymorphic ventricular tachycardia in 2 patients with prior clinical ventricular fibrillation (Table 3). The 5 patients who underwent electrophysiological testing had underlying coronary artery disease and had had a previous myocardial infarction. For these 5 patients, long-term antiarrhythmic drug therapy was prescribed on the basis of serial electrophysiological testing. Three patients remain alive and well. One patient experienced an episode of welltolerated ventricular tachycardia after reduction in the dose of amiodarone to 400 mg daily. Reinstitution of a 600 mg daily dose has prevented further episodes. One other patient died suddenly after the discontinuation of amiodarone because of drug toxicity. The remaining 5 hospital survivors did not undergo electrophysiological testing. Two had valvular heart disease. All 5 had experienced the initial episode of ventricular tachyarrhythmia in the first 24 hours after operation. In contrast, the patients who underwent electrophysiological testing experienced the tachyarrhythmia 2, 5, 6, 28, and 42 days postoperatively. Four of the 5 patients who did not undergo electrophysiological testing were discharged on a regimen of procainamide. One of these 4 died suddenly 6 months postoperatively. The other 4 remain alive and well. Comment The data presented in this report demonstrate that the development of new ventricular tachyarrhythmias in patients in otherwise stable condition after cardiac opera-

4 320 The Annals of Thoracic Surgery Vol 38 4 October 1984 Fig I. Results of electrophysiological study in a patient with ventricular fibrillation (VF) six weeks after coronary artery bypass operation. The tracings shown are surface electrocardiographic leads I, 11, and V,, and intracardiac recordingsfrom the right atrium (RA), bundle of His (HB), and right ventricle (RV). Turn premature stimuli (S2-SJ) during ventricular pacing (S,-S1) initizte polymorphic ventricular tachycardia (VT), which rapidly degenerates to ventricular fibrillation. (A = atrial; V = ventricular; H = His bundle.) tion is a relatively rare but frequently catastrophic occurrence. The effectiveness of antiarrhythmic drug therapy in preventing the development of new ventricular tachycardia or ventricular fibrillation remains uncertain. In this retrospective study, the use of prophylactic antiarrhythmic drugs was not randomized, and we cannot estimate the effect of such therapy from our data. However, 10 of the 18 patients were receiving lidocaine at the time of the initial episode of ventricular tachycardia or ventricular fibrillation, a finding that indicates that lidocaine does not always provide protection from these arrhythmias. It is certain that once the arrhythmias occur, they carry ominous short-term and long-term prognostic significance. Eight of our 18 patients died during the initial hospitalization, and 2 others died suddenly after discharge, for an overall oine-year mortality of 56%. The high recurrence rate of arrhythmias made imperative the prevention of future rh.ythm disturbances. Electrophysiological studies were performed in 5 patients in whom ventricular tachycardia or ventricular fibrillation developed 48 hours or more after operation. In each, programmed ventricular stimulation reproduced the clinical arrhythmia. Serial electrophysiological testing identified effective long-term antiarrhythmic drug therapy for 4 of these 5 patients. Electrophysiological studies were not performed in 5 patients who experienced ventricular tachycardia or fibrillation less than 48 hours after operation. Originally we hypothesized that the early onset of postoperative ventricular tachyarrhythmias was related to a perioperative event. However, 1 of these patients subsequently died suddenly. Electrophysiological studies may also be useful in this group of patients. The etiology of the arrhythmias observed in our patients is speculative. Four of the 6 patients with ventricular fibrillation had experienced an acute ischemic event, and in 1 patient this arrhythmia developed during postoperative exercise testing. The other patient in whom ventricular fibrillation developed had uniform morphology ventricular tachycardia at electrophysiological study; however, due to its fast rate, it deteriorated to ventricular fibrillation. Our observations agree with the hypothesis that acute ischemia produces the dispersion of refractoriness and repolarization necessary for the propagation of ventricular fibrillation [5, 61. The 12 patients with uniform morphology ventricular tachycardia can be classified into two groups. Nine had coronary artery disease with extensive ventricular scarring, and 3 had undergone aneurysm resection without intraoperative electrophysiological mapping to guide the extent of the resection. Several possibilities exist for the genesis of these arrhythmias. The reperfusion of previously ischemic regions may have produced the proper electrophysiological substrate for abnormal conduction, and this could have given rise to reentrant arrhythmias. The stresses of anesthesia and operation may have led to the release of endogenous catecholamines, which can produce these arrhythmias [7]. For those patients with sustained ventricular tachycardia occurring in the absence of perioperative myocardial infarction, the former hypothesis seems more likely. Repeat electrophysiological studies of our 5 patients revealed chronic sustained ventricular arrhythmias probably related to reentry pathways. These were not clinically apparent prior to coronary revascularization. The etiology may be different for those patients with postoperative ventricular tachycardia occurring in the presence of aortic valve disease. Ventricular arrhythmias are present preoperatively in many patients with aortic valve disease and are not related to the type or severity of the valvular lesion [8]. Decreased left ventricular ejection fraction and intraventricular conduction abnormalities in patients with aortic valve disease are associated with a higher frequency of ventricular arrhythmias

5 321 Kron et al: Unanticipated Postoperative Ventricular Tachyarrhythmias A. POST- OP DAY 4 - PROCAINAMIDE t LlDOCAlNE C. PROCAINAMIM: - EPS Fig 2. Correlation between spontaneous and stimulation-induced arrhythmias. All tracings were obtained in Patient 18 and represent ventricular tachycardia (VT) that occurred spontaneously on the fourth postoperative day. (A) The patient had been receiving procainamide and lidocaine prior do the development of the sustained arrhythmia. (B)Sustained ventricular tachycardia was initiated at a shorter cycle length (CL) with two premature stimuli (S1-SJ) during fixed-rate ventricular pacing (SI-S,) during a control electrophysiological study (EPS).The patient was taking no antiarrhythmic drugs. (C) When procainamide was readministered, ventricular tachycardia similar to that observed clinically could be initiated with programmed stimulation. [9]. Electrophysiological studies have not been performed postoperatively on our patients who have undergone aortic valve replacement and then had ventricular tachycardias. We are not certain if they are at risk for late arrhythmias. We can only conclude that ventricular tachycardia can occur unexpectedly after aortic valve replacement and can be unrelated to postoperative hemodynamics. The ability to prevent ventricular tachycardia or ventricular fibrillation after operation is uncertain. Lidocaine was infused in 10 of these patients; it suppressed prema- Table 3. Electrophysiolo,qical Study and Results Patient. Clinical Arrhythmia (cycle length) Arrhythmia at EPS (cycle length) Mode of Initiation of Arrhythmia 12 VF; VT Polymorphic VT + VF (230 msec) VT (210 msec) Polymorphic VT + VF (200 msec) VT (370 msec) VT (420 msec) EPS when on procainamide (300 msec) VT (210 msec) VF VT (380 msec) VT (420 msec) = electrophysiological study; VF ventricular pacing. = ventricular fibrillation; VT = 1-VPD-VP ventricular tachycardia; VPD = ventricular premature depolarization; VP =

6 322 The Annals of Thoracic Surgery Vol 38 4 October 1984 ture ventricular complexes, but it did not prevent the ventricular tachyarrhythmia. It has been shown that the suppression of premature ventricular contractions in patients with recurrent ventricular tachycardia does not prevent tachyarrhythmia [101. In some cases, the treatment may be worse than the arrhythmia: 2 deaths and 2 major complications were clue to the antiarrhythmic drug used, particularly bretylium. The potential dangers of bretylium in the early postoperative period were described in a previous report Ventricular tachycardia or ventricular fibrillation occurring in the postoperative period is an infrequent but catastrophic event. The cause of the ventricular fibrillation is usually myocardial ischemia, whereas ventricular tachycardia may be due to reperfusion of a previously ischemic area in patients with coronary artery disease. Electrophysiological testing in the long-term follow-up of patients with postoperative ventricular tachyarrhythmias is helpful in defining the specific arrhythmia and preventing late sudden death by providing a test of the effectiveness of antiarrhythmic drugs. References 1. Angelini P, Feldman MI, Lufchanowsk R, Leachman R D Cardiac arrhythmias during and after heart surgery: diagnosis and management. Prog Cardiovasc Dis 16:469, McGovern B, DiMarco JP, Garan H, Ruskin JN: New concepts in the management of ventricular arrhythmias and sudden death. Curr Prob Cardiol71, DiMarco JP, Garan H, Ruskin J: Quinidine for ventricular arrhythmias: value of electrophysiologic testing. Am J Cardiol 51:90, Horowitz LN, Josephson ME, Kastor JD: Intracardiac electrophysiologic studies as a method for the optimization of drug therapy in chronic ventricular arrhythmia. Prog Cardiovasc Dis 23:81, Salerno TA, Stefaniszyn HJ: Spontaneous ventricular fibrillation occurring immediately after institution of cardiopulmonary bypass: possible clinical implications. J Thorac Cardiovasc Surg 86306, Fyke FE, Uliestra RE, Danielson GK, Beynen FM: Verapamil for refractory ventricular fibrillation during cardiac operations in patients with cardiac hypertrophy. J Thorac Cardiovasc Surg 86108, Lubbe WF, Podzuweit T, Danes PS, Opie LH The role of cyclic adenosine monophosphate in adrenergic effects on ventricular vulnerability to fibrillation in the isolated perfused rat heart. J Clin Invest 61:1260, Olshausen K, Schwarz F, Apfelbach J, et al: Determinants of the incidence and severity of ventricular arrhythmias in aortic valve disease. Am J Cardiol51:1104, Santinga JT, Kirsh MM, Brady TJ, et al: Left ventricular function in patients with ventricular arrhythmias and aortic valve disease. Ann Thorac Surg 35:152, Herling IM, Horowitz LN, Josephson ME: Ventricular ectopic activity after medical and surgical treatment for recurrent sustained ventricular tachycardia. Am J Cardiol45:633, Kron IL, lan SP: Severe hypotension due to the use of bretylium for postcardiotomy ventricular arrhythmias. Ann Thorac Surg 35:271, 1983