| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2003;75:1340-1348
© 2003 The Society of Thoracic Surgeons
a Department of Intensive Care Medicine, Edegem, Belgium
b Department of Thoracic and Vascular Surgery, Antwerp University Hospital, Edegem, Belgium
* Address reprint requests to Dr De Decker, Department of Intensive Care Medicine, Antwerp University Hospital, Wilrijkstraat 10, B-2650 Edegem, Belgium
e-mail: koen.de.decker{at}uza.be
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
Most reviews in this population involve analgesic or respiratory complications, or both. Although arrhythmias after thoracic surgery have been studied extensively, the quest for reviews concerning postoperative cardiac complications arose, but with very little results [3, 5]. The purpose of this review is to briefly describe the available peer-reviewed publications on cardiac complications after thoracic surgery, with the emphasis on prevention of these complications.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
Study selection
Initially the abstracts of all articles were scanned for relevance by a first reviewer. The inclusion and exclusion criteria were applied by the first reviewer and cross checked by a second. We excluded patients undergoing cardiac surgery, lung transplantations, and pediatric patients as these form special subpopulations with specific complications that we believe should be subject for a separate review. All reports were validated by two reviewers separately using the following aspects: randomization, blinding, controls used, clarity of results, and population size. The criteria used for labeling the obtained studies are summarized in Table 1
and were based on two reference articles [6, 7] and the levels of recommendation of the National Health Science Centre for Evidence-Based Medicine (http://www.minervation.com/cebm). We retrieved 72 studies that were relevant for this review. After exclusion of level C and D publications, 51 reports were retained (Table 2).
|
|
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
The incidence, prophylaxis, and onset of arrhythmias, as well as their mortality, are summarized in Table 3. Although these occurrences are recognized, the pathophysiology in noncardiac surgery remains poorly defined [8–12]. In some studies, serially performed echocardiograms suggest that increased right heart pressure and increased pulmonary vascular resistance, but not fluid overload or right heart enlargement, predispose to clinically significant supraventricular tachycardias after pulmonary resection [12, 13]; another study contradicts this finding [14]. Besides, prolonged oxygen supply (causing dilatation of the pulmonary vessels) has failed to demonstrate any benefit to the prevention of postoperative arrhythmias [15]. The prognostic significance of arrhythmias is difficult to interpret as it can be associated or induced by other complications, such as heart failure associated with pulmonary edema. No difference was observed between the AF group and the non-AF group regarding short-term or long-term mortality or regarding long-term atrial fibrillation recurrences [8, 16]. However, other authors have reported an increase of arrhythmia-related mortality [9, 12, 17–19]. Tachydysrhythmias occur more frequently after intrapericardial dissection and in patients who have postoperative interstitial or perihilar pulmonary edema develop [18].
|
Class I antiarrhythmic agents
Flecainide was studied in two randomized trials by Borgeat and coworkers [23, 24]. A constant rate infusion was compared with a placebo in 30 noncardiac thoracic surgery patients [23], whereas in the other report flecainide was compared with digoxin in 30 pulmonary surgery patients [24]. Flecainide prevented arrhythmias completely in the first trial (p < 0.01) and reduced its incidence by 40% in the second trial (p < 0.05). No toxic drug levels were measured and mortality was absent. Flecainide was also very successful (100%) in converting the failures in the digoxin group.
ß-blockade
The surgical stress response may cause important hemodynamic effects. Patients treated with metoprolol showed less changes in heart rate and cardiac index [25]. In addition, their oxygen consumption was reduced. Only two randomized controlled trials studied the prophylactic antiarrhythmic effects of ß-blocking agents [26, 27]. Metoprolol, initiated preoperatively and continued once daily postoperatively, reduced the incidence of atrial fibrillation from 40% to 67% (p < 0.05) after elective lung resection [26]. No difference was found between the metoprolol and the placebo group in the incidence of other arrhythmias or complications. The average time for having atrial fibrillation develop was 2.9 days. Bayliff and coworkers [27] included 99 patients who were undergoing major thoracic surgery in a randomized, double-blind, placebo-controlled trial of propranolol (10 mg qid). Although the drug induced a 70% relative risk reduction in treatable arrhythmias, statistical significance was not achieved (p = 0.071), probably because of the lower than expected event rate in the placebo group. Atrial flutter (AF) was present in 9 of 13 arrhythmia episodes. The mean onset time was less than 24 hours in both groups. However, there were side effects associated with the use of propranolol, in particular hypotension and bradycardia, although without clinical importance.
Amiodarone
A prospective randomized trial comparing the administration of amiodarone, verapamil, and a placebo as prophylactic treatment for supraventricular dysrhythmias after pulmonary surgery [28] was interrupted after 64 patients, because severe life-threatening side effects (adult respiratory distress syndrome; ARDS) occurred in 3 postpneumonectomy patients in the amiodarone group. Subsequently, a retrospective analysis of all cases of pulmonary resection (n = 552) was done. The overall incidence of ARDS was 11% in the patients treated with amiodarone and 1.8% in the nonamiodarone group. Despite the apparently high incidence of ARDS after the use of intravenous amiodarone after pneumonectomy, the favorable results of a pilot study on the use of prophylactic oral amiodarone after pulmonary resection supports the demand for a prospective randomized trial [29].
Calcium antagonists
After interruption of the amiodarone trial, the study design was rewritten and the administration of the placebo was compared with verapamil, administered as a 10 mg bolus followed by a 30-minute infusion of 0.375 mg/min and then 0.125 mg/min for 3 days [30]. AF occurred in 15% of the patients receiving a placebo and in 8% of the patients receiving verapamil (not significant, p not mentioned). Most episodes occurred on the second or third postoperative day. More important was the interruption of the infusion in 9% of the patients caused by bradycardia and in 14% of patients caused by related hypotension. Three patients died in this study, of whom only 1 had AF.
A smaller trial investigated the effects of verapamil on right ventricular pressure and supraventricular dysrhythmia (SVD) [13]. No cardiac arrhythmias occurred in the verapamil group, whereas 6 patients (50%) in the placebo group suffered from supraventricular arrhythmias (p < 0.05), all of which occurred on the second postoperative day. The increase in both end-diastolic right ventricular pressure and central venous pressure in 30% of the control patients was associated with atrial tachyarrhythmia, whereas this was not observed in the verapamil group. Only one randomized controlled trial studied diltiazem in thoracic surgery patients [31]. A loading dose of 0.25 mg/kg was given on admission to the postanesthesia care unit, followed by a 0.1 mg/kg/h intravenous infusion for 18 to 24 hours. The controls received intravenous placebo-loading doses, followed by a placebo infusion at 0.1 mL/kg/h. Starting in the morning on postoperative day 1, patients received either diltiazem slow release (SR) 120 mg or a placebo orally for 14 days. Postoperative SVD occurred in 15% of patients treated with diltiazem and 25% of the placebo group (p = 0.03). No differences were observed in major postoperative complications, overall duration, or costs of hospitalization. Because of mild hypotension, diltiazem treatment was stopped in 6 patients. Three patients in the placebo group and 1 in the diltiazem group died.
Magnesium sulfate
There is only one randomized controlled trial investigating the prophylactic antiarrhythmic role of magnesium sulfate in noncardiac surgery [32]. Two hundred patients were randomized to receive either an MgSO4 infusion, no treatment, or digoxin (if >70 years old, in cases of pneumonectomy or an intrapericardial procedure). The incidence of atrial tachyarrhythmias, mainly AF (90% in the magnesium and 85% in the control group), was reduced from 26.7% in the control group to 10.7% in the magnesium group (p = 0.008). About 75% of arrhythmias occurred within the first 2 days postoperative. The operative mortality was 1.03% and again was not related to arrhythmias.
Ischemia
In a series of 598 patients undergoing thoracic surgery for lung cancer [17], transient ischemic electrocardiographic changes were documented in 23 patients (3.8%) and myocardial infarction (MI) in 7 (1.2%). Abnormal exercise testing and intraoperative hypotension were the strongest predictors for ischemic events [17].
Herrington and Shumway [5] reviewed the literature on MI in the postthoracotomy period. Mortality rates ranged from 2.1% to 21%. However, the incidence of perioperative MI was low (0.13%) in patients with no previous cardiac history to moderate (2.8% to 17%) in patients with a prior history of infarction. No association with present anesthetic techniques or duration of operative procedure with perioperative ischemia or MI could be found, as long as patients were monitored invasively, base line medication was continued, and perioperative fluid administration was minimalized. The authors recommend continuous monitoring for at least 3 days, certainly in high-risk patients, as postoperative MI was associated with a mortality as high as 32% to 70%.
According to a recent review [33], the American College of Cardiology and the American Heart Association guidelines [34] for perioperative cardiovascular evaluation for noncardiac surgery remain the best available method for risk assessment in noncardiac thoracic surgery. Thoracic surgery is categorized as a high-risk surgical procedure in this matter. Coronary angiography is advocated in case of major clinical predictors such as unstable angina, decompensated heart failure, significant arrhythmias, or severe valvular disease. In cases of intermediate or minor clinical predictors the decision whether to perform an angiography is based on noninvasive testing. As adenosine and dipyridamole should be avoided in patients with clinical bronchospasms, dobutamine stress echocardiography is the evaluation of choice for patients with cardiac ischemia referred for thoracic surgery [33]. In general the indications for coronary angiography are similar to those in the nonoperative setting. No prospective randomized data exists on the role of prophylactic coronary bypass surgery. Whether percutaneous coronary intervention is superior to bypass surgery is uncertain, but in cases of angioplasty with stenting it is probably safer to postpone surgery for 2 to 4 weeks. In conclusion, the preoperative cardiac assessment of thoracic surgery patients is of great importance, although prospectively controlled data for this type of surgery are lacking.
Heart failure
Four studies using thermodilution and one study investigating the right ventricle (RV) by echocardiograph [35–39] addressed the problem of postoperative right ventricular dysfunction. Alterations in RV contractile performance and changes in RV afterload are the presumed mechanisms of RV dysfunction. Whereas right ventricular end-diastolic volume remains stable in the early postoperative hours, significant increases may be observed on the first and second postoperative day. Pulmonary artery pressure and pulmonary vascular resistance only rose modestly in 15 patients studied [35], suggesting that the rise in afterload is not the only causing factor. In a two-staged study including only 12 patients, RV dysfunction was seen as early as the second postoperative day. In the second phase of the study, prostaglandin E1 was given through a right atrial catheter, but this intervention was unable to correct the postoperative RV dilatation and dysfunction [36]. Other authors [12, 37–39] contradict this finding and claim that afterload alteration is the major determinant of RV dysfunction. In 10 thoracotomy patients [37] the preload recruitable stroke work relation, which is the relationship between RV stroke work index and RV end-diastolic volume, was used to assess RV contractility. Despite changes in loading conditions, RV end-diastolic volume and RV stroke work index were within the limits of the preload recruitable stroke work relation. Another argument favoring afterload augmentation as the cause of RV dysfunction, is the fact that postoperative pulmonary artery pressure, pulmonary vascular resistance, and central venous pressure only change significantly during exercise [38]. Changes in RV function are able to compensate for the increased RV end-diastolic volume at rest, but not during exercise, with a resultant increase in pulmonary artery pressure and pulmonary vascular resistance. One study used serially performed transthoracic echocardiography to assess the effects of pulmonary resection [39]. Only pneumonectomy patients had mild postoperative pulmonary hypertension without significant RV systolic dysfunction. Most likely this is caused by the decrease in volume of the pulmonary vascular bed. Both pulmonary embolism, occurring in 1%, and cardiac herniation are rare mechanisms that may cause RV dysfunction; especially the latter, which presents with a high mortality rate (40% to 50%). Although right heart failure is a substantial concern in thoracic surgery, no predictive technique or test exists at present [40].
Left heart decompensation [12] is usually a consequence of impaired right heart function, either by decreasing left ventricular preload or by shifting the intraventricular septum resulting in a decreased left ventricular volume. Other causes of left ventricular dysfunction are acute myocardial infarction, preexisting valvular disorders [33] or cardiac herniation. The latter is a very rare complication, the most common cause being intrapericardial pneumonectomy with high negative pressures in the pleural cavity [12]. Symptoms usually develop within 24 hours, but late onset has been reported. In right-sided pneumonectomy the heart rotates counterclockwise around the axis of the vena cava, inducing a vena cava superior syndrome. In left-sided pneumonectomy the heart is strangulated by the pericardial sac, possibly causing superimposed myocardial ischemia.
Pulmonary edema
Pulmonary edema is an uncommon but serious complication after major resection of the lung, usually after pneumonectomy. Its incidence is approximately 2.5% to 4% [12, 41–46]. It probably arises from an increased filtration gradient across the pulmonary microcirculation together with hyperpermeability, rather than from cardiac dysfunction [12, 41–46]. Possible contributing factors in its pathogenesis are positive fluid balance, impaired lymphatic drainage, and surgical manipulation, although postpneumonectomy pulmonary edema more probably represents panendothelial injury caused by inflammatory processes from repetitive collapse and reexpansion maneuvers, ie, ischemia and reperfusion [45]. Conflicting results are found concerning the need for postoperative fluid administration [46–48]. Slinger [48] proposed the following recommendations and statements concerning fluid administration: (1) there is no "third space" in the thorax, (2) total positive fluid balance in the first 24 hours should not exceed 20 mL/kg, (3) a urinary output greater than 0.5 mL/kg/h is unnecessary, (4) the use of invasive monitoring techniques is advisable if increased tissue perfusion is necessary postoperatively, (5) factors that contribute to increased pulmonary venous pressures should be minimized postoperatively, (6) hyperinflation of the residual lung should be avoided, (7) regular chest roentgenograms should be obtained postoperatively, (8) prolonged periods with the residual lung in the dependent position should be avoided, and (9) prophylactic digitalization has not been shown to reduce the incidence of postresection supraventricular arrhythmias or of postpneumonectomy pulmonary edema.
Therapy consists of administration of diuretics, restriction of fluid, nutritional support, and maintenance of adequate oxygenation, even with mechanical ventilation if necessary [12]. The efficacy of steroids remains unproven [12, 49]. Despite aggressive treatment, the clinical outcome is poor with mortality exceeding 50% [45] and almost reaching 100% in some reports [42]. Nitric oxide ventilation and extracorporeal membrane oxygenation were tried as possible therapies [43, 49–50]. In particular the use of inhaled nitric oxide, in doses of 10 to 20 ppm, was able to lower mortality rates to 30% in a small series of patients [49]. In the same report early intubation (at first signs of ARDS), aspiration, bronchoscopy, and postural changes (to change ventilaton and perfusion matching) are also advocated. Therefore early recognition to identify risk factors (eg, right pneumonectomy, a large perioperative fluid load, increased cardiac output, and high or low urinary output) is of utmost importance for this lethal complication [12].
Shunting
The development of an atrial right-to-left shunt through a persistent foramen ovale may occur and can be the cause of dyspnea or the so-called plathypnea-orthodeoxia syndrome; its incidence is low and seems to occur more frequently after right-sided pneumonectomy [51, 52]. The overall prevalence of permeable foramen ovale is estimated at 20% to 35% in the general population [51]. The contributing factors are (1) anatomical factors, such as mediastinal shifting (caused by modification of the relationship between the right and left atrium, distortion of the foramen ovale, and cardiac rotation), preferential flow of the inferior caval vein into the left atrium and compression of the right atrium by pleural fluid; (2) hemodynamic factors, such as reversal of the interatrial pressure gradient (caused by a decrease in right ventricular compliance and the hydrostatic pressure in the left lateral decubitus) or orthostatic increase of the shunt; and (3) several other factors, such as pulmonary emboli, right ventricular infarction, increased intrathoracic pressure, chronic obstructive pulmonary disease, and positive pressure ventilation.
Typical clinical features of this syndrome include a relative symptomless interval of a few months followed by the development of posture dependent (worse in the upright position) and volume dependent (greater shunt in dehydrated patients) dyspnea. The diagnosis is made with arterial blood gas analysis, nuclear lung perfusion scanning, echocardiography, magnetic resonance imaging, or cardiac catheterization, or a combination thereof. Standard of treatment is surgical repair, although several cases of successful intravascular occlusion of the septal defect have been published in recent years [52].
Thromboembolism
There is only one prospective nonrandomized study [53] that determined the frequency and significance of thromboembolism after pulmonary resection. Thromboembolism may lead to serious cardiac complications such as pulmonary hypertension after pulmonary embolism. Seventy-seven patients were prospectively followed for 30 days or less postoperatively. Incidence was higher in bronchogenic carcinoma than in metastatic cancer or benign disease, in adenocarcinoma as compared with other carcinoma types, in large cancers as compared with smaller lesions, and in pneumonectomy or lobectomy as compared with segmentectomy and wedge resection. The overall incidence of thromboembolic diseases was 26%, of which 19% occurred postoperatively. Eleven patients had deep venous thrombosis develop, and 4 patients had pulmonary embolism that was detected, in which 1 patient caused fatality. Ljungström [54] found a deep venous thrombosis incidence of 18% in 45 consecutive patients, particularly in the most extended procedures. Deep venous thrombosis was detected on average 3.3 days postoperatively.
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
The widespread use of magnesium sulfate in cardiac surgery has not been equally studied in the noncardiac thoracic surgery population; however its use is supported by the excellent results of the trial of Terzi and coworkers [32]. Although caution is warranted in renal failure, magnesium has a relatively safe profile and deserves further investigation. Because of the absence of major side effects in controlled use, we believe magnesium sulfate is superior to many other antiarrhythmic drugs as most have significant side effects, in particular hypotension and bradycardia, limiting their use in the postthoracotomy patient. A large well-conducted trial with metoprolol, a drug frequently used after cardiac surgery, deserves special attention [26]. Because metoprolol has a limited hemodynamic derangement and capacity to convert arrhythmias to sinus rhythm, amiodarone is one of the preferred drugs for treating both supraventricular and ventricular arrhythmias. However in other settings, because of the high incidence of ARDS after its use in pneumonectomy patients [28], the use of amiodarone after pulmonary surgery is debatable, although a recent pilot trial using oral amiodarone pointed to an overall benefit [29]. Finally, an indication for the use of sotalol as a prophylaxis in thoracic surgery may be retained. Sotalol has only been studied for the prevention of atrial fibrillation in cardiac surgery, but never in thoracic surgery.
Few authors [5, 17] have studied the incidence and outcome of myocardial ischemia in noncardiac thoracic surgery. Mortality rates in this population vary between 2% and 21%, making an adequate preoperative workup essential. Only limited data are available, and we refer to the American Heart Association guidelines for the preoperative evaluation of patients [33]. No prospective data are provided on the beneficial effects of prophylactic angioplasty. The indications for angioplasty are probably the same as in the nonsurgery population, but when performed, a surgery free interval of 2 to 4 weeks is recommended.
Development of postoperative heart failure is most frequently the consequence of right ventricular pathology. Apart from rather rare conditions, such as pulmonary embolism or cardiac herniation, postoperative heart failure is usually caused by an increase in right ventricular afterload. Whether other pathophysiologic mechanisms are involved is uncertain yet and has been the subject of much debate. Without a doubt, it is very difficult to predict the development of postoperative heart failure preoperatively [40].
Pulmonary edema occurs in 2.5% to 4% of thoracic surgery patients, mainly after pneumonectomy. Although probably caused by an increased filtration gradient rather than a cardiac problem [12, 41–46], this postoperative complication has an important impact on morbidity and mortality, with reports of mortality rates of 50% to 100% [42, 45]. The introduction of nitric oxide in the management of postpneumonectomy pulmonary edema has resulted in reduced mortality [49]. The development of shunting after pulmonary operations is a problem that probably occurs more frequently than previously believed, and in rare cases, may be the cause of the platypnea-orthodeoxia syndrome, which may require surgical repair. The introduction of intravascular occlusion will probably reduce the need for surgical reintervention in the future [52].
Studies of incidence, prophylaxis, and treatment of thromboembolic complications in the thoracic surgical population are sparse. An overall incidence of 19% to 26% was reported [53, 54], with the highest risk in very extended procedures. All patients undergoing thoracic surgery should be treated with low molecular weight heparins and antithrombotic stockings to prevent deep venous thrombosis.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. K. Mishra, R. Pandey, M. J. Shackcloth, J. McShane, A. D. Grayson, M. H. Carr, and R. D. Page Cardiac comorbidity is not a risk factor for mortality and morbidity following surgery for primary non-small cell lung cancer Eur. J. Cardiothorac. Surg., March 1, 2009; 35(3): 439 - 443. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Cardinale, A. Colombo, M. T. Sandri, G. Lamantia, N. Colombo, M. Civelli, M. Salvatici, G. Veronesi, F. Veglia, C. Fiorentini, et al. Increased Perioperative N-Terminal Pro-B-Type Natriuretic Peptide Levels Predict Atrial Fibrillation After Thoracic Surgery for Lung Cancer Circulation, March 20, 2007; 115(11): 1339 - 1344. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. F. Bruzzi, M. Remy-Jardin, D. Delhaye, A. Teisseire, C. Khalil, and J. Remy When, Why, and How to Examine the Heart During Thoracic CT: Part 2, Clinical Applications Am. J. Roentgenol., February 1, 2006; 186(2): 333 - 341. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Hendriks, P. Lauwers, and P. Van Schil Extrapericardial pneumonectomy MMCTS, June 28, 2005; 2005(0628): 83. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Sedrakyan, T. Treasure, J. Browne, H. Krumholz, C. Sharpin, and J. van der Meulen Pharmacologic prophylaxis for postoperative atrial tachyarrhythmia in general thoracic surgery: Evidence from randomized clinical trials J. Thorac. Cardiovasc. Surg., May 1, 2005; 129(5): 997 - 1005. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Shrivastava, B. Nyawo, J. Dunning, and G. Morritt Is there a role for prophylaxis against atrial fibrillation for patients undergoing lung surgery? Interactive CardioVascular and Thoracic Surgery, December 1, 2004; 3(4): 656 - 662. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
