Abstract
Purpose
Our aim was to determine what specific patient and peri-operative factors contribute to major complications after esophagectomy.
Methods
Using the American College of Surgeons National Surgical Quality Improvement Program database, data for esophagectomies between the years 2005 and 2008 were extracted and analyzed. Thirty-day post-operative complications were classified into seven major groups: (1) wound infections, (2) respiratory complications (pneumonia, intubation), (3) cardiac complications, (4) deep venous thrombosis, (5) sepsis/septic shock, (6) re-operation, and (7) death. Univariate analysis and logistic regression modeling were performed to determine if a significant association existed between patient factors or peri-operative factors and these post-operative complications.
Results
One thousand thirty-two patients who underwent esophagectomy were identified. Diabetes was the strongest pre-operative independent predictor of death (odds ratio (OR) 10.98; 95% confidence interval (CI) 1.37–1.15, p < 0.1) or respiratory (OR 1.86; 95% CI 1.03–3.29, p = 0.04) or cardiac (OR 5.14; 95% CI 1.93–13.20, p < 0.01) complications following esophagectomy. Thoracotomy performed during the operation was not associated with an increased risk of respiratory or cardiac complications.
Conclusions
The major predictors of morbidity after an esophagectomy are the patient factors of diabetes, dyspnea, peripheral vascular disease, and cerebrovascular accident while the peri-operative factors are pre-operative international normalized ratio, contaminated wound classification, and American Society of Anesthesiologists class. Similarly, the major predictors of mortality are diabetes, dyspnea, and age for patient factors and contaminated wound classification for peri-operative factors.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Bailey SH, Bull DA, Harpole DH, Rentz JJ, Neumayer LA, Pappas TN, Daley J, Henderson WG, Krasnicka B, Khuri SF. Outcomes after esophagectomy: a ten-year prospective cohort. Ann Thorac Surg. 2003; 75: 217–22.
Wright CD, Kucharczuk JC, O’Brien SM, Grab JD, Allen MS. Predictors of major morbidity and mortality after esophagectomy for esophageal cancer: a Society of Thoracic Surgeons General Thoracic Surgery Database risk adjustment model. J Thorac Cardiovasc Surg. 2009; 137: 587–95.
Bartels H, Stein HJ, Siewert JR. Preoperative risk analysis and postoperative mortality of esophagectomy for resectable esophageal cancer. Br J Surg 1998; 85: 840–4.
Kohn GP, Galanko JA, Meyers MO, Feins HR, Farrell TM. National trends in esophageal surgery—are outcomes as good as we believe? J Gastrointestinal Surg. 2009; 13: 1900–12.
Chang AC, Birkmeyer JD. The volume performance relationship in esophagectomy. Thorac Surg Clin. 2006; 18: 87–94.
Rodgers M, Jobe BA, O’Rourke RW, Sheppard B, Diggs B, Hunter JG. Case volume as a predictor of inpatient mortality after esophagectomy. Arch Surg. 2007; 142: 829–39.
Atkins BZ, Shah AS, Hutcheson KA, Mangum JH, Pappas TN, Harpole DH, D’Amico TA. Reducing hospital morbidity and mortality following esophagectomy. Ann Thorac Surg. 2004; 78: 1170–6.
Karl RC, Schreiber R, Boulware D, Baker S, Coppola D. Factors affecting morbidity, mortality, and survival in patients undergoing Ivor Lewis esophagogastrectomy. Ann Surg. 2000; 231: 635–43.
ACS-NSQIP Program Specifics, Surgical Case Inclusion/Exclusion Overview. Available at: http://acsnsqip.org/main/program_case_inclusion_exclusion.asp Accessed April 15, 2010.
Trivers KF, Sabatino SA, Stewart SL. Trends in esophageal cancer incidence by histology, United States, 1998–2003. Int J Cancer 2008; 123: 1422–1428.
Muller JM, Erasmi H, Stelzner M, Zieren U, Pichlmaier H. Surgical therapy for esophageal carcinoma. Br J Surg 1990; 77: 845–57.
Birkmeyer JD, Siewers AE, Finlayson EVA, Stukel TA, Lucas FL, Batista I, Welch HG, Wennberg DE. Hospital volume and surgical mortality in the United States. N Eng J Med 2002; 346: 1128–37.
Dimick JB, Pronovost PJ, Cowan JA, Lipsett PA. Complications and costs after high-risk surgery: where should we focus quality improvement initiatives. J Am Coll Surg 2003; 196: 671–78.
Ferguson MK, Martin TR, Reeder LB, Olak J. Mortality after esophagectomy: risk factor analysis. World J Surg. 1997; 21: 599–603.
Pultrum BB, Bosch DJ, Nijsten MWN, Rodgers MGG, Groen H, Slaets JPJ, Plukker JTM. Extended esophagectomy in elderly patients with esophageal cancer: minor effect of age alone in determining the postoperative course and survival. Ann Surg Oncol. 2010; 17(6):1572–80.
Orringer MB. Age does not preclude an esophagectomy...if only it were that simple. Ann Surg Oncol. 2010; 17(6):1487–9.
Kelly J, Rudd A, Lewis RR, Coshall C, Moody A, Hunt BJ. Venous thromboembolism after acute ischemic stroke: a prospective study using magnetic resonance direct thrombus imaging, Stroke 2004; 35: 2320–355.
CLOTS Trials Collaboration, Dennis M, Sandercock PA, Reid J, Graham C, Murray G, Venables G, Rudd A, Bowler G. Effectiveness of thigh-length graduated compression stockings to reduce the risk of deep vein thrombosis after stroke (CLOTS trial 1): a multicentre, randomised controlled trial. Lancet 2009; 373:1958–965.
Mangran AJ, Horan TC, Pearlson ML, Silver LC, Jarvis WR. Guideline for prevention of surgical site infection, Infect Control Hosp Epidemiol 1999; 20:247–278.
Culver DH, Horan TC, Gaynes RP, Martone WJ, Jarvism WR, Emori G, Banerjee S, Edwards JR, Tolson JS, Henderson TA, Hughes JM. Surgical wound infection rates by wound class, operative procedure, and patient risk index. Am J Med 1991; 91:152–157.
Author information
Authors and Affiliations
Corresponding author
Additional information
Discussant
Dr. Henry Pitt (Indianapolis, IN): I would like to congratulate the authors on using the ACS-NSQIP database to determine which risk factors lead to bad outcomes after esophagectomy.
The NSQIP database is robust, certainly compared to many single-institution studies, but it may not be mature enough yet to determine exactly which risk factors result in a bad outcome after esophagectomy.
My group has been working with the statisticians at the college over the last year with respect to pancreatectomy. In that analysis, we have over 7,000 operations to analyze. However, the College statisticians are concerned that enough data are not yet available to properly analyze esophagectomy.
I have three questions.
The first question has to do with your choice to use only the 2005, 2006, and 2007 data which include 600-plus operations. In fact, as of last summer, the 2008 data were available, and there were another 400-plus esophagectomies in the database. Including the 2008 data would have given you more than a thousand rather than 600 cases to analyze. Why didn’t you use the most recent Participant Use File data?
The second question has to do with your statistical analysis. You did a univariate analysis of 36 factors, looking at seven different outcomes, which gave you 252 separate univariate analyses. Again, in working with statisticians at the college, they have lumped the analyses into mortality, serious morbidity, and overall morbidity, or just three outcomes. Among the seven outcomes that you chose to analyze, three of them had a very low incidence. DVT was only 5%, mortality only 4%, and cardiac only 2%. Therefore, you were not likely to see differences when the percentage of the complication was so low.
Thirdly, when you went from the univariate to the multivariable analysis, you came up with 14 variables that were significant. In the manuscript, in your conclusion, you only picked three of those 14 as being important on the basis of their hazard ratios. However, the others were just as statistically significant. You chose diabetes, peripheral vascular disease, and dyspnea, which are hard to alter. Why not choose parameters such as low hematocrit or radiation therapy or blood transfusions that you might be able to affect.
In summary, I applaud your efforts, but I think that this report is a little premature and needs better statistical support.
Closing Discussant
Dr. Birat Dhungel: For the first question regarding the use of data from 2005 to 2007 only, when we started this project, ACS-NSQIP had not published their 2008 data. But now since it is available, we will definitely look into it to yield a more statistically robust analysis.
For your second question, where we used only 36 variables for univariate analysis, actually we had used more than 36. I think we had about 58 variables that we looked into but the ones I showed in those two tables in the PowerPoint are the ones with a p value less than 0.05, that is, significant ones in univariate analysis. I think this is also related to low number of patients with those factors reported in the NSQIP database.
For example, we also looked at something like race, which other studies have found to be significant. But in our review, it was not significant. So we dropped that off, too.
And for your third question about focusing on those three factors, dyspnea, diabetes, and peripheral vascular disease, you’re right; I focused on those because of their strong association with increased risk for morbidity and mortality. These factors that I listed, I think, are chronic issues and can also be addressed and more so at the primary care level.
Discussant
Dr. Steven Demeester (Los Angeles, CA): One of the leading causes of morbidity and mortality after esophagectomy is anastomotic complications and graft ischemia. We previously looked at that and found that factors that correlated with anastomotic complications were diabetes, peripheral vascular disease, and neoadjuvant therapy.
Did you look at what caused the morbidity and mortality in your analysis here of global morbidity? Because I suspect that what you are finding then is those same factors are causing anastomotic complications, leaks, graft ischemia, and subsequent complications.
Secondly, did you analyze it based on squamous versus adenocarcinoma to see differences in the different tumor histologies?
Closing Discussant
Dr. Birat Dhungel: For your first question about anastomotic leaks, one of the limitations with the use of NSQIP database is that it does not report on procedure-specific complications. So I think anastomotic leaks here are likely included in the deep wound infections. But they do not list it separately, so we could not analyze it.
No, we did not analyze cases for adenocarcinoma versus squamous cell carcinoma separately. That’s definitely something that can be done using this database.
Discussion of Paper #1044 (40)
Title of Paper: Patient and Peri-operative Predictors of Morbidity and Mortality After Esophagectomy: American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP), 2005–2007.
Discussant
Dr. Robert Rout (Gainesville, FL): In my experience, patients with cancer have severe problems with nutrition. And did you look at the albumin and the prealbumin in these patients?
Closing Discussant
Dr. Birat Dhungel: We did and we did not find it to be significant in this study. This can be looked at again in the future when we have more patients added to the data set.
Discussant
Dr. Henry Pitt (Indianapolis, IN): Additional work has been done to develop procedure-specific outcomes for esophagectomy, pancreatectomy, hepatectomy, and other procedures. Going forward, your hospitals will be able to keep track of new preoperative, intraoperative, and post-operative variables that are procedure specific. Therefore, when the new “procedure targeted” module becomes available, I would recommend that your hospitals switch to this option.
Discussant
Dr. Margo Shoup (Maywood, IL): I noticed that you included patients who had ASA classes 4 and 5 undergoing esophagectomy. Those had to have been emergent cases. I would encourage you to exclude those patients and just look at the patients that are undergoing elective esophagectomy.
Closing Discussant
Dr. Birat Dhungel: That’s a very good suggestion. I think we had less than 10% of patients with ASA classes IV and V combined. Thank you.
Appendix
Appendix
Rights and permissions
About this article
Cite this article
Dhungel, B., Diggs, B.S., Hunter, J.G. et al. Patient and Peri-operative Predictors of Morbidity and Mortality After Esophagectomy: American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP), 2005–2008. J Gastrointest Surg 14, 1492–1501 (2010). https://doi.org/10.1007/s11605-010-1328-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11605-010-1328-2