HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Sudish C. Murthy Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Murthy, S. C. Articles by Wong, J. Search for Related Content PubMed PubMed Citation Articles by Murthy, S. C. Articles by Wong, J. Related Collections Electrophysiology - arrhythmias

J Thorac Cardiovasc Surg 2003;126:1162-1167
© 2003 The American Association for Thoracic Surgery

General thoracic surgery

Atrial fibrillation after esophagectomy is a marker for postoperative morbidity and mortality

^a Division of Esophageal Surgery, Department of Surgery, University of Hong Kong Medical Centre, Queen Mary Hospital, Hong Kong, China
^b Department of Thoracic and Cardiovascular Surgery, The Cleveland Clinic Foundation, Cleveland, Ohio, USA

Received for publication December 7, 2001; revisions received April 23, 2002; revisions received August 22, 2002; accepted for publication June 22, 2003.

^* Address for reprints: John Wong, PhD, Department of Surgery, University of Hong Kong Medical Centre, Queen Mary Hospital, Hong Kong, China
jwong{at}hku.hk

	Abstract

Top Abstract Methods Results Discussion Appendix 1 Appendix 2 References

 
OBJECTIVE: Postoperative atrial fibrillation complicates recovery in 20% to 25% of patients after esophagectomy for cancer. The purpose of this study is to understand this phenomenon.

METHODS: Between 1982 and 2000, 198 (22% of 921) patients had postoperative atrial fibrillation after esophagectomy. Propensity scoring and the Greedy Match algorithm were used to develop a cohort of control patients for statistical comparisons. One hundred forty-four patients who had postoperative atrial fibrillation were matched.

RESULTS: Pulmonary complications affected 42% of patients in the atrial fibrillation group compared with 17% in the control group (P < .001). Anastomotic leakage was more common in the atrial fibrillation group (6.9% vs 1.4%, P = .035). Surgical sepsis migrated with atrial fibrillation 4 times more frequently (P = .001). Multivariate analysis demonstrated that postoperative pulmonary complications (odds ratio, 2.5; 95% confidence interval, 1.42-4.3) and surgical sepsis (odds ratio, 3.4; 95% confidence interval, 1.2-9.6) were associated with postoperative atrial fibrillation. The mortality rates of the atrial fibrillation and control groups were 23% and 6.3%, respectively (P < .001). Median survival, excluding hospital deaths, was not different at 14.5 months (atrial fibrillation group) and 16.9 months (control group; P = .4).

CONCLUSION: Atrial fibrillation is a surrogate for surgical morbidity and mortality after esophagectomy. The occurrence of atrial fibrillation after esophageal resection should prompt not only the appropriate management of the arrhythmia but also a search for a more ominous underlying cause.

Surgical resection for carcinoma of the esophagus and gastric cardia is arguably one of the most morbid thoracic operations.¹² Although patient selection and improved postoperative care can significantly decrease hospital mortality,^13-15 morbidity remains high.^12,13,16 AF frequently complicates recovery from esophagectomy,^17,18 and its significance remains unclear. In this study risk factors for the development of postoperative AF were identified. Moreover, the effect of AF on postoperative outcome was examined.

	Methods

Top Abstract Methods Results Discussion Appendix 1 Appendix 2 References

 
Between July 1982 and December 2000, 1816 patients with cancer of the esophagus or gastric cardia were managed at the University of Hong Kong Medical Center, Queen Mary Hospital. Of these patients, 1138 underwent surgical resection. Patients who underwent pharyngolaryngoesophagectomy, patients with gastric cardia cancer and an abdominal resection, and patients with staged resections were excluded. Prospectively collected data from the remaining 921 patients were retrospectively reviewed.

Preoperative assessment of all patients included a thorough medical history and complete physical examination, standard blood chemistries and blood counts, and cardiopulmonary evaluation. A history of cardiac disease included at least one of the following: hypertension, atrial or ventricular arrhythmia, ischemic heart disease, congestive heart failure, or valvular disease. A history of pulmonary disease included emphysema, bronchiectasis, or tuberculosis.

All patients had a barium contrast study, flexible fiberoptic upper endoscopy, and bronchoscopy. Computed tomographic scans and endoscopic ultrasonography were also routinely performed in recent years. The operative approach was individualized for each patient and was governed by the location and stage of the tumor, the history of previous abdominal surgery, intraoperative findings, and preoperative comorbid disease. Specific surgical techniques have been described previously.^19,20 Patients with cardia cancer were resected through an abdominal-right thoracic approach (sixth or seventh intercostal space). In these patients the esophagogastric anastomosis was below the level of the carina, and consequently, the extent of intrathoracic dissection was restricted. The operation was considered curative when the tumor was completely resected with microscopically negative margins.

Outcomes
AF was documented by means of electrocardiography, with the majority of patients requiring medical intervention for rate control, rhythm control, or both. Treatment included cardioversion, digoxin, calcium channel or ß-blockade, and intravenous amiodarone.

Study end points included several possible postoperative outcomes. Cardiovascular complications were ventricular arrhythmia, congestive heart failure and circulatory overload, myocardial infarction, and pulmonary embolus. Aspiration pneumonia, bronchopneumonia, pulmonary edema, respiratory failure, and shock lung were included as pulmonary complications. Nonanastomotic leakage occurred from the lesser curve gastric staple line, ischemic necrosis of the esophageal substitute, or any other nonanastomotic site and was distinguishable from anastomotic leak in all cases. Endoscopy, barium contrast study, or computed tomographic scans were used to document leak. Sepsis was defined as hemodynamic instability and end-organ (pulmonary, renal, hepatic, and central nervous system) dysfunction. The cause of sepsis was both medical (eg, pneumonia) and surgical (eg, leak related). Unless otherwise specified, mortality was until the day of discharge (hospital mortality). Thirty-day mortality was also recorded for all patients.

Statistical method and generation of the control group
Propensity scoring and the Greedy Match algorithm^21-24 were used to develop a control population suitable for statistical comparison. Twenty-one factors were analyzed by using multivariate logistic regression to identify variables that correlated with postoperative AF (Appendix 1). The 4 variables found to be predictive of postoperative AF were age (P < .0001), history of cardiac disease (P = .007), amount of blood loss (P = .015), and limited intrathoracic dissection (P = .002). These 4 factors were then used to generate a matched control group by using the Greedy Match algorithm. In brief, a propensity score for development of postoperative AF was generated for each patient in the entire population by means of logistic modeling with the 4 variables identified above. Patients with postoperative AF were matched to patients without AF on the basis of propensity score. The C statistic was 0.7, and the Hosmer-Lemeshow goodness-of-fit test had a P value of .1. One hundred forty-four patients with postoperative AF (study group) were matched with 144 patients who did not have AF (control group) by means of this methodology. The remaining 54 unmatched patients with postoperative AF were compared with the matched patients with AF (Appendix 2). Unmatched patients had a higher frequency of preoperative pulmonary disease and a higher preoperative serum hemoglobin level. Transhiatal esophagectomy was more prevalent in the unmatched group.

All continuous variables are expressed as means ± SD unless otherwise stated. Univariate analyses were performed by means of analysis of variance for continuous variables and ² testing for categoric variables. The Fisher exact test was used if any expected cell value in a 2 x 2 table was less than 5. Stepwise logistic regression analysis was used for multivariate analysis. Survival curves were generated by means of Kaplan-Meier methodology. All calculations were conducted with biostatistical programs of SPSS (Chicago, Ill) and Statistical Analysis System software (version 6.12; SAS, Cary, NC).

	Results

Top Abstract Methods Results Discussion Appendix 1 Appendix 2 References

 
Of the 921 patients in the population, 198 (22%) who had postoperative AF were identified. Appropriate control subjects were identified for 144 patients with AF. The preoperative demographics of the matched groups are shown in Table 1. Operative factors are shown in Table 2. These data demonstrate that the 2 groups have been successfully matched on the basis of the methodology applied. Notably, significant variables identified in the initial analysis (age, cardiac disease, blood loss, and limited thoracic dissection) were well matched after propensity scoring and application of the Greedy Match algorithm.

A striking disparity existed between hospital mortality rates of patients with and without AF (Figure 1). Overall, the mortality was 3.7 times greater when AF was observed during the postoperative course (P < .001). Although the ability to recover patients from esophagectomy improved with time, postoperative AF continued to correlate with mortality regardless of early (1982-1990) or recent experience (1991-2000). The median survival was shorter for patients with postoperative AF compared with control subjects (11.5 months vs 14.5 months, P = .02). However, when hospital mortality was excluded from the analysis, survival was not different (14.5 months vs 16.9 months, P = .4).

View larger version (14K): [in this window] [in a new window]   Figure 1. Hospital mortality rates after esophagectomy of patients with and without postoperative AF. Although there is a decreased mortality rate during the most recent experience, AF remains associated with poor postoperative outcome throughout the study period.

	Discussion

Top Abstract Methods Results Discussion Appendix 1 Appendix 2 References

Our preliminary analysis demonstrated that advancing age and a history of cardiac disease predisposed to postoperative AF. This is not surprising because similar associations have been demonstrated for pulmonary^8-11 and cardiac^25,26 surgery. Hypertension has also been shown to be a risk factor for AF after cancer surgery.²⁷

Intraoperative blood loss and extensive thoracic dissection (required for a high-chest or cervical anastomosis) also migrated with postoperative AF. From previous work, we found no appreciable difference in the incidence of postoperative AF when transhiatal and Lewis-Tanner techniques were contrasted.^17,18 However, included in the current study was a cohort of patients whose thoracic dissection was restricted below the level of the carina and anastomosis placed in the low chest. Most of these patients had cardia cancer, and this more limited resection does not appear to compromise tumor clearance.²⁸ It is unclear as to why this lesser resection is associated with a lower incidence of postoperative AF. Clearly, less of the posterior mediastinum is dissected, and perhaps less single-lung ventilation time is required compared with the Lewis-Tanner approach. In addition, vagal nerves cephalad to the inferior pulmonary veins are left largely unperturbed in these resections, and this might be important. Ultimately, preservation of lung function might be the most important beneficiary of the limited thoracic dissection. Regardless, this should be considered when planning the surgical approach for cardia cancers. It would be interesting to evaluate the incidence of AF after video-assisted esophageal resections.²⁹ For lung resections, video-assisted procedures boast similar rates of postoperative AF compared with those of open resections.³⁰

Major pulmonary complications and surgical sepsis were frequent morbidities in patients with AF, as documented by means of multivariate analysis. We suspect that in such patients postoperative AF is merely a systemic manifestation of these serious illnesses, and in almost 50% of patients, the onset of AF should be considered a surrogate for these grave complications. We attempted to study the temporal relationship between the development of AF and the diagnosis of either morbidity. Unfortunately, we had insufficient data on many of the early patients in the study to properly address this query. The results of a limited review of patients who had both AF and surgical sepsis after esophagectomy suggested that when AF was noted between 3 and 10 days postoperatively, it heralded the onset of sepsis by 1 day. AF occurring at 0 to 3 days after resection demonstrated no temporal connection to the diagnosis of sepsis (unpublished data). If true, this might define a group of patients whose AF is self-limiting and attributable to the trauma and electrolyte imbalances that routinely accompany esophagectomy.

An alarming mortality rate was observed for patients who had AF after esophagectomy (>20%). The association of postoperative AF with mortality has also been documented after lung surgery^3,4,6,8 and, more recently, after major noncardiac and nonthoracic operations.³¹ Although improvements in intensive care and refinement of surgical techniques have decreased the postoperative death rate after esophagectomy,³² the striking association of AF with hospital mortality remained throughout the study period.

In summary, AF after esophageal resection is a marker for poor short-term outcome and frequently acts as a surrogate for pulmonary complications and surgical sepsis. In this regard strategies designed to prevent postoperative AF³³ should be met with tempered optimism because we found that postoperative AF is not particularly difficult to manage. A controlled trial of prophylaxis against postoperative AF might be necessary to help validate this contention. That AF after esophagectomy is accompanied by a 6-fold increased leak rate, as well as the serious morbidities described above, should alert the surgeon to not simply treat the rhythm disturbance but to more thoroughly evaluate the postoperative patient with new-onset AF. This strategy might identify an underlying life-threatening but potentially treatable complication.

	Appendix 1

Top Abstract Methods Results Discussion Appendix 1 Appendix 2 References

 

Appendix 1

List of variables used for multivariate linear regression to define elements used in the Greedy Match algorithm

Variables P value Date of operation – Age <.0001 Sex – Histology of tumor – Location of tumor – Stage of disease – Preoperative Cardiac disease .007 Pulmonary disease – Serum albumin – Hemoglobin – Smoking history – FEV1 – FVC – Po2 – Pco2 – Intraoperative Thoracotomy – Extent of thoracic dissection .002 Intent of resection – Estimated blood loss .015 Length of operation –

FEV₁, Forced expiratory volume in 1 second; FVC, forced vital capacity.

	Appendix 2

Top Abstract Methods Results Discussion Appendix 1 Appendix 2 References

 

Appendix 2

Analysis of matched and unmatched patients with postoperative AF

Variable Matched patients (n = 144) Unmatched patients (n = 54) P value Male sex 116 47 .2 Smoking history 108 46 .09 Squamous cancer 128 48 .4 History of pulmonary disease 27 20 .007 History of cardiac disease 31 14 .6 FEV1 (% predicted) 91 90 .7 Preoperative albumin (g/L) 40.5 40 .7 Preoperative hemoglobin (g/dL) 12.6 13.2 .05 Thoracotomy 127 40 .02

Figures represent numbers of patients unless otherwise stated. FEV₁, Forced expiratory flow in 1 second.

	Footnotes

 
S.C.M. was supported in part by a Warren-Whitman-Richardson Fellowship from Harvard Medical School and travel grant from the Division of Thoracic Surgery, Brigham and Women's Hospital, Boston, Mass.

	References

Top Abstract Methods Results Discussion Appendix 1 Appendix 2 References

 

1. Narayan SM, Cain ME, Smith JM. Atrial fibrillation. Lancet. 1997;350(9082):943–950[Medline]
   
2. Mathew JP, Parks R, Savino JS, Friedman AS, Koch C, Mangano DT, et al. Atrial fibrillation following coronary artery bypass graft surgery: predictors, outcomes, and resource utilization. MultiCenter Study of Perioperative Ischemia Research Group. JAMA. 1996;276(4):300–306[Abstract]
   
3. Bailey C, Betts R. Cardia arrhythmias following pneumonectomy. N Engl J Med. 1943;229:356–359
   
4. Currens J, White P, Churchill E. Cardiac arrhythmias following thoracic surgery. N Engl J Med. 1943;229:360–364
   
5. Beck-Nielsen J, Raehbeck-Soresen H, Alstrup P. Atrial fibrillation following thoracotomy for non-cardiac diseases, in particular cancer of the lung. Acta Med Scand. 1973;193:425–429[Medline]
   
6. Krowka MJ, Pairolero PC, Trastek VF, Payne WS, Bernatz PE. Cardiac dysrhythmia following pneumonectomy. Clinical correlates and prognostic significance. Chest. 1987;91(4):490–495[Abstract/Free Full Text]
   
7. Wahi R, McMurtrey MJ, DeCaro LF, Mountain CF, Ali MK, Smith TL, et al. Determinants of perioperative morbidity and mortality after pneumonectomy. Ann Thorac Surg. 1989;48(1):33–37[Abstract]
   
8. von Knorring J, Lepantalo M, Lindgren L, Lindfors O. Cardiac arrhythmias and myocardial ischemia after thoracotomy for lung cancer. Ann Thorac Surg. 1992;53(4):642–647[Abstract]
   
9. Dyszkiewicz W, Skrzypczak M. Atrial fibrillation after surgery of the lung: clinical analysis of risk factors. Eur J Cardiothorac Surg. 1998;13:625–628
   
10. Curtis JJ, Parker BM, McKenney CA, Wagner-Mann CC, Walls JT, Demmy TL, et al. Incidence and predictors of supraventricular dysrhythmias after pulmonary resection. Ann Thorac Surg. 1998;66:1766–1771[Abstract/Free Full Text]
    
11. Rena O, Papalia E, Oliaro A, Casadio C, Ruffini E, Filosso PL, et al. Supraventricular arrhythmias after resection surgery of the lung. Eur J Cardiothorac Surg. 2001;20:688–693[Abstract/Free Full Text]
    
12. Earlam R, Cunha-Melo JR. Oesophageal squamous cell carcinoma. Br J Surg. 1980;67:457–461[Medline]
    
13. Bartels H, Stein HJ, Siewert JR. Preoperative risk analysis and postoperative mortality of oesophagectomy for resectable oesophageal cancer. Br J Surg. 1998;85(6):840–844[Medline]
    
14. Law SYK, Wong J. Complications: prevention and management. Daly JM, Hennessy TPJ, Reynolds JV. Management of upper gastrointestinal cancer. London: W.B. Saunders; 1999. p. 240–262
    
15. Law SY, Fok M, Wong J. Risk analysis in resection of squamous cell carcinoma of the esophagus. World J Surg. 1994;18(3):339–346[Medline]
    
16. Poon RT, Law SY, Chu KM, Branicki FJ, Wong J. Esophagectomy for carcinoma of the esophagus in the elderly: results of current surgical management. Ann Surg. 1998;227(3):357–364[Medline]
    
17. Chu KM, Law SY, Fok M, Wong J. A prospective randomized comparison of transhiatal and transthoracic resection for lower-third esophageal carcinoma. Am J Surg. 1997;174(3):320–324[Medline]
    
18. Fok M, Wong J. Cancer of the oesophagus and gastric cardia. Standard oesophagectomy and anastomotic technique. Ann Chir Gynaecol. 1995;84(2):179–183[Medline]
    
19. Law S, Wong J. Esophagogastrectomy for carcinoma of the esophagus and cardia, and the esophageal anastomosis. Baker RJ, Fischer JE. Mastery of surgery. Philadelphia: Lippincott Williams & Wilkins; 2001. p. 813–827
    
20. Law S, Wong J. Esophagectomy without thoracotomy. Baker RJ, Fischer JE. Mastery of surgery. Philadelphia: Lippincott Williams & Wilkins; 2001. p. 828–836
    
21. Parsons L. Using SAS software to perform a case-controlled method for bias reduction in an observational study. In: Proceedings of the Twenty-Fifth Annual SAS Users Group International Conference. Cary, NC: SAS Institute, Inc; 2000. p. 1166-71
    
22. Blackstone EH. Comparing apples and oranges. J Thorac Cardiovasc Surg. 2002;123:8–15[Free Full Text]
    
23. Rosenbaum PR, Rubin DB. Reducing bias in observational studies using subclassification on the propensity score. J Am Stat Assoc. 1984;79:516–524
    
24. Rosenbaum PR. From association to causation in observational studies: the role of tests of strongly ignorable treatment assignment. J Am Stat Assoc. 1984;79:41–48
    
25. Yousif H, Davies G, Oakley CM. Peri-operative supraventricular arrhythmias in coronary bypass surgery. Int J Cardiol. 1990;26(3):313–318[Medline]
    
26. Crosby LH, Pifalo WB, Woll KR, Burkholder JA. Risk factors for atrial fibrillation after coronary artery bypass grafting. Am J Cardiol. 1990;66(20):1520–1522[Medline]
    
27. Gibbs HR, Swafford J, Nguyen HD, Ewer MS, Ali MK. Postoperative atrial fibrillation in cancer surgery: preoperative risks and clinical outcome. J Surg Oncol. 1992;50(4):224–227[Medline]
    
28. Siewert JR, Stein HJ, Sendler A, Fink U. Surgical resection for cancer of the cardia. Semin Surg Oncol. 1999;17:125–131[Medline]
    
29. Law S, Fok M, Chu KM, Wong J. Thoracoscopic esophagectomy for esophageal cancer. Surgery. 1997;122(1):8–14[Medline]
    
30. Neustein SM, Kahn P, Krellenstein DJ, Cohen E. Incidence of arrhythmias after thoracic surgery: thoracotomy versus video-assisted thoracoscopy. J Cardiothorac Vasc Anesth. 1998;12(6):659–661[Medline]
    
31. Brathwaite D, Weissman C. The new onset of atrial arrhythmias following major noncardiothoracic surgery is associated with increased mortality. Chest. 1998;114:462–468[Abstract/Free Full Text]
    
32. Whooley BP, Law S, Murthy SC, Alexandrou A, Wong J. Analysis of reduced death and complication rates after esophageal resection. Ann Surg. 2001;233:338–344[Medline]
    
33. Shimada M, Namai H, Morisaki H, Ochiai R, Takeda J, Fukushima K. [Preventive use of diltiazem to suppress supraventricular tachyarrhythmia in the patients after esophagectomy]. Masui. 1997;46(5):658–663[Medline]
    

This article has been cited by other articles:

				S. Salman, A. Bajwa, O. Gajic, and B. Afessa Review of A Large Clinical Series: Paroxysmal Atrial Fibrillation in Critically Ill Patients With Sepsis J Intensive Care Med, May 1, 2008; 23(3): 178 - 183. [Abstract] [PDF]

				S. Asbach, M. Olschewski, T. S. Faber, M. Zehender, C. Bode, and M. Brunner Mortality in patients with atrial fibrillation has significantly decreased during the last three decades: 35 years of follow-up in 1627 pacemaker patients Europace, April 1, 2008; 10(4): 391 - 394. [Abstract] [Full Text] [PDF]

				S. Goodman, T. Shirov, and C. Weissman Supraventricular Arrhythmias in Intensive Care Unit Patients: Short and Long-Term Consequences Anesth. Analg., April 1, 2007; 104(4): 880 - 886. [Abstract] [Full Text] [PDF]

				S. K. Malhotra, R. P. Kaur, N. M. Gupta, A. Grover, K. Ramprabu, and D. Nakra Incidence and types of arrhythmias after mediastinal manipulation during transhiatal esophagectomy. Ann. Thorac. Surg., July 1, 2006; 82(1): 298 - 302. [Abstract] [Full Text] [PDF]

				E. E. Roselli, S. C. Murthy, T. W. Rice, P. L. Houghtaling, C. D. Pierce, D. P. Karchmer, and E. H. Blackstone Atrial fibrillation complicating lung cancer resection J. Thorac. Cardiovasc. Surg., August 1, 2005; 130(2): 438 - 444. [Abstract] [Full Text] [PDF]

				H. J. Ahn, W. S. Sim, Y. M. Shim, and J. A. Kim Thoracic epidural anesthesia does not improve the incidence of arrhythmias after transthoracic esophagectomy Eur. J. Cardiothorac. Surg., July 1, 2005; 28(1): 19 - 21. [Abstract] [Full Text] [PDF]

				S. Law, D. T. K. Suen, K.-H. Wong, K.-F. Kwok, and J. Wong A Single-Layer, Continuous, Hand-Sewn Method for Esophageal Anastomosis: Prospective Evaluation in 218 Patients Arch Surg, January 1, 2005; 140(1): 33 - 39. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Sudish C. Murthy Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Murthy, S. C. Articles by Wong, J. Search for Related Content PubMed PubMed Citation Articles by Murthy, S. C. Articles by Wong, J. Related Collections Electrophysiology - arrhythmias