ATS
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Ann Thorac Surg 2008;85:1938-1945. doi:10.1016/j.athoracsur.2008.03.014
© 2008 The Society of Thoracic Surgeons
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Access Nomogram Calculator
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lagarde, S. M.
Right arrow Articles by van Lanschot, J. J. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lagarde, S. M.
Right arrow Articles by van Lanschot, J. J. B.
Related Collections
Right arrow Esophagus - cancer
Right arrowRelated Article

Original Articles: General Thoracic

Preoperative Prediction of the Occurrence and Severity of Complications After Esophagectomy for Cancer With Use of a Nomogram

Sjoerd M. Lagarde, MDa,*, Johannes B. Reitsma, MDb, Anna-Karin D. Maris, MSa, Mark I. van Berge Henegouwen, MDa, Olivier R.C. Busch, MDa, Hugo Obertop, MDa, Aelko H. Zwinderman, AH, PhDb, J. Jan B. van Lanschot, MDa

a Department of Surgery, Academic Medical Center, University of Amsterdam, Amsterdam, the Netherlands
b Department of Clinical Epidemiology, Biostatistics and Bioinformatics, Academic Medical Center, University of Amsterdam, Amsterdam, the Netherlands

Accepted for publication March 5, 2008.

* Address correspondence to Dr Lagarde, St Lucas Andreas Hospital, Jan Tooropstraat 164, Amsterdam, 1061 AE, the Netherlands (Email: soerdlagarde{at}hotmail.com).


The nomogram discussed in this article can be viewed on the Internet at http://ats.ctsnetjournals.org/content/vol85/issue6/images/data/1938/DC1/calculator.xls.

 

   Abstract
Top
 Abstract
Introduction
 Patients and Methods
 Results
 Comment
 Acknowledgments
 References
 
Background: Predicting the severity of complications after esophagectomy may supply important information for both patient and surgeon. The aim of the present study was to develop a nomogram based on preoperative risk factors to predict the severity of complications in patients who undergo esophagectomy for cancer.

Methods: A consecutive series of 663 patients who underwent esophagectomy between January 1993 and August 2005 was used to develop a prognostic model. The model was validated in a second group of patients who were operated between August 2005 and November 2006. Ordinal logistic regression analysis was performed to predict the severity of complications. Diverse simple and conventional preoperative risk factors were evaluated. A nomogram was developed to enhance clinical applicability.

Results: Patients were divided into three complication categories: those who suffered from no complications (n = 197); minor complications (n = 354); and major complications (n = 112). The following predictors remained in the model after multivariate analysis: higher age (p = 0.014); cerebrovascular accident/transient ischemic attack (CVA/TIA) (p = 0.009) or myocardial infarction in the medical history (p = 0.066); lower forced expiratory volume in the first second of expiration (FEV1) (p = 0.030); presence of electrocardiogram-changes (p = 0.008); and more extensive surgery (p < 0.001). A nomogram based on these variables was constructed. Overall agreement between the predicted probabilities and the observed frequencies was good in the development and the validation set.

Conclusions: The nomogram predicts the severity of complications for individual patients and may help in informing the patient before undergoing esophagectomy for cancer and in choosing the optimal extent of surgery. When externally validated, the nomogram may play a role in risk-adjusted audit of morbidity after esophagectomy.

Surgery is the best curative treatment option for esophageal cancer, but is accompanied by a high operative risk [1, 2]. In high volume centers, the operative mortality risk has steadily decreased and is generally around 5% [1, 3–8]. One of the possibilities to measure quality of care between hospitals is to compare the operative mortality rate. However, comparing hospitals by using crude in-hospital mortality rates can be misleading. Ideally this parameter should be corrected for the so-called case mix (age, general health, and comorbidity). Risk-adjusted models, which are based on predicting in-hospital mortality for individual patients, have been developed [9–11]. However, these models all suffer from a low discriminative ability [9–11] or even a lack of fit when externally validated [12]. In-hospital mortality is a relatively rare event in esophageal cancer surgery and therefore a model that includes other serious, but more frequent complications, is preferred.

While mortality rates of around 5% can be reached in high volume centers, esophageal cancer resection is still associated with substantial morbidity. Early postoperative complication rates vary between 40% and 80%, depending on the applied criteria and depending on the extent of resection [1, 13, 14]. Complications can range from minor complications (eg, urinary tract infection) to major complications (eg, respiratory failure). Several previous studies focused on predisposing factors for complications after esophagectomy for cancer, but this did not result in reliable predictive models [15–19], except for the specific prediction of pulmonary complications [20]. Models that focus on the presence of unspecified complications cannot be used in esophageal surgery due to the large variation in the severity of complications [21]. The severity of complications [22] was only taken into account in one study [9] and was validated in a relatively small external series [23]. These studies describe three different risk groups. These groups have a relation with a normal, prolonged, or severe postoperative course. However, estimation of individual risk is not possible with use of this risk score. Predicting the severity of complications with use of preoperative risk factors may reveal important information for both patient and surgeon. Individualized risk assessment may help in choosing the optimal extent of surgery. So far, the severity of complications was never taken into account in predictive models.

To improve applicability of predictive models with diverse predictive factors nomograms have been developed. A nomogram gives a graphical representation of the predictive strength of individual predictors and enables clinicians to calculate an overall score for individual patients reflecting their personal risk. Nomograms are presently extensively applied to predict cancer recurrence after treatment [24–26], but have recently also been designed to predict the probability of developing major complications after breast reconstruction [27]. The aim of the present study was to design a nomogram that can predict the severity of complications with the use of conventional and widely available preoperative risk factors for patients undergoing potentially curative esophagectomy for cancer.


   Patients and Methods
Top
 Abstract
 Introduction
 Patients and Methods
Results
 Comment
 Acknowledgments
 References
 
Patients
The model was developed with data of a consecutive series of patients who underwent potentially curative esophagectomy for adenocarcinoma or squamous cell carcinoma of the esophagus or gastroesophageal junction (GEJ) in the Academic Medical Center at the University of Amsterdam, the Netherlands, a tertiary referral center with a wide experience in esophageal surgery, between January 1993 and August 2005. Subsequently, the validity of the model was tested in a consecutive series of new patients who underwent potentially curative esophagectomy in the same hospital between August 2005 and November 2006. Extensive preoperative staging had not revealed local irresectability and distant metastases (including tumor positive cervical lymph nodes or irresectable celiac nodes). All patients had a detailed preoperative assessment of their general health status and organ function (eg, lung function tests, electrocardiography).

Surgery was performed/supervised by an experienced surgeon (MIvBH, ORCB, HO, JJBvL). Both transhiatal and transthoracic esophagectomies were performed. Clinicopathologic data from all operated patients were permanently collected in a prospective database. The study was approved by the Medical Ethical Committee of the Academic Medical Center at the University of Amsterdam and the necessity of an individual consent for the study was waived by this committee.

Definition of Complications
The severity of postoperative complications was graded according to the morbidity scale proposed by Dindo and colleagues [22]. This classification system is based on the therapeutic consequences of complications and consists of five grades and two subgrades. Grade I complications do not need any medical or surgical intervention, grade II complications need pharmacologic treatment, grade III complications need a radiologic (grade IIIa) or operative (grade IIIb) reintervention, grade IV complications are life threatening and represent single organ (grade IVa) or multiorgan (grade IVb) dysfunction. Finally, grade V complications are complications leading to death. Grading of complications was performed according to the most severe complication in each patient by a panel of four contributing authors (SML, MIvBH, ORCB, and JJBvL). Subsequently, for the purpose of the present study, three categories of complications were defined by the above mentioned contributing authors (see Table 1); no complications (category 0), minor to moderate complications (category 1, grades I to IIIb), and severe complications (category 2, grades IVa, IVb, and V).


View this table:
[in this window]
[in a new window]

  		Table 1 Categorization of the Severity of Complications Based on the Classification According to Dindo and Colleagues [22] a
 
Statistical Analysis
A proportional odds model (ordinal logistic regression) was used to examine the association between potential predictors and the occurrence of complications classified in three categories of severity. The proportional odds model is an extension of the binary logistic regression model to the case of three or more outcome states that are naturally ordered [28]. Odds ratios with 95% confidence intervals were used to quantify the strength of the association between predictors and severity of complications. Various readily available preoperative potential predictors were selected by a panel of four contributing authors (SML, MIvBH, ORCB, and JJBvL). These factors included general predictors (eg, age, sex, smoking) and predictors related with the medical history of the patients (eg, myocardial infarction, hypertension). Function tests (lung function and electrocardiogram) and therapy-related predictors (eg, location of tumor, neoadjuvant chemoradiation therapy) were also selected.

Predictors with a p value of 0.10 or less in univariate analysis were all entered in a multivariate model. Possible mathematical linkage of parameters (eg, the presence of myocardial infarction [MI] and Q-waves and ST-T changes) was checked for. A nomogram was developed to visualize the prognostic strength of the different factors in a single figure and to calculate the expected distribution across the severity of complications based on a specific profile of a patient. In a backward elimination procedure (p < 0.10 to stay in the model) the number of predictors was reduced to increase the practical applicability of the nomogram. The number of points for each predictor was based on the coefficient from the nomogram model by multiplying it by 10 and rounding it to the lowest whole number. The total number of points derived by specifying values for all predictors was used to calculate the expected probability to develop no complications, minor to moderate complications, and severe complications. This was visualized by a series of stacked bar-graphs for the three outcome categories across the range of predicted risks.

Because missing data result in loss of statistical power and can lead to possible bias, multiple imputation techniques were applied [29, 30]. All predictors as well as the observed outcome together were used to impute missing values based on multivariate normal distributions using the Markov chain Monte Carlo method. The coefficients of 10 rounds of imputations were combined to obtain the final estimates and their 95% confidence intervals of the multivariate model.

The fit and validity of models were checked in the following ways. The discriminatory properties of the model were examined by visualizing the distribution and degree of overlap in risk scores of individual patients within and between the three outcome categories. Subsequently, the discriminative ability was quantified by using the concordance (c) statistic. The c-statistic is a measure that can be interpreted as the probability among all possible pairs between patients from different outcome categories that the patient with the more severe complication also has the higher risk score. Values can range from 0.5 (due to chance; no discrimination) to 1.0 (perfect discrimination). The proportional odds assumption was evaluated by performing the test for parallel lines. Calibration was checked by comparing expected and observed number of patients in each of the three outcome categories across deciles of expected risk and tested for significance by using an extension of the Hosmer-Lemeshow goodness-of-fit statistic [31]. All analyses were performed using SAS software version 9.1 (SAS Institute Inc, Cary, NC).


   Results
Top
 Abstract
 Introduction
 Patients and Methods
 Results
Comment
 Acknowledgments
 References
 
Between January 1993 and August 2005, a consecutive series of 663 patients underwent a macroscopically radical esophagectomy for squamous cell carcinoma (187 patients, 28%) or adenocarcinoma (476 patients, 72%) of the esophagus or GEJ. Within this series, 466 patients (70%) experienced at least one complication during their hospital stay. Patients were classified according to the severity of complications into three complication categories; no complications (category 0, n = 197), minor to moderate complications (category 1, n = 354), and major complications (category 2, n = 112) (Table 1). In-hospital mortality was observed in 24 patients (3.6%). Eighteen patients died within 30 days (2.7%) and 31 patients (4.7%) died within 90 days after operation. Patients in the more severe complication category had significantly longer intensive care stays (p < 0.001) as well as longer hospital stays (p < 0.001) (Table 2). The diverse preoperative predictors that were tested for are given in Table 3. Descriptive data on the distribution of the various preoperative predictors in each of the three outcome categories are given in Table 3. Interestingly, American Society for Anesthesiologists (ASA) classification was not related with the severity of complications (p = 0.294). Age (p = 0.003, forced expiratory volume in the first second (FEV1) (p = 0.003), the occurrence of cerebrovascular accident/transient ischemic attack CVA/TIA (0.008) or myocardial infarction (p = 0.007) in the medical history, the presence of Q-waves and ST-T changes on the electrocardiogram (p = 0.001), and a more extended transthoracic esophagectomy (p = 0.001) were all associated with more severe complications in the univariate analysis. The presence of dyspnea or hypertension showed a trend toward significance (p = 0.066 and p = 0.072, respectively).


View this table:
[in this window]
[in a new window]

  		Table 2 Intensive Care Stay and Hospital Stay a
 

View this table:
[in this window]
[in a new window]

  		Table 3 Univariate Analysis of Potential Preoperative Predictors for the Severity of Complications a
 
All these variables were selected for multivariate analysis. Using backward elimination the following predictors remained in the model: age (0.014); the presence of CVA/TIA (p = 0.009) or myocardial infarction in the medical history (p = 0.066); FEV1 (p = 0.030); the presence of Q-waves and ST-T changes (p = 0.008); and the operation type (p = 0.001). The strength of the association using odds ratios and 95% confidence intervals are shown in Table 4. A nomogram based on these variables was constructed and is shown in Figure 1. The value of each predictor corresponds to a score. The scores for all predictors are summed to a total score, which is then translated into a probability for each of the three outcome states visualized by stacked bar graphs. A calculator to calculate the individual risk for severity of complications can be found online at http://ats.ctsnetjournals.org/content/vol85/issue6/images/data/1938/DC1/calculator.xls.


View this table:
[in this window]
[in a new window]

  		Table 4 Remaining Predictors of the Nomogram Model After Backward Elimination for the Prediction of Severity of Complications a
 

Figure 1
View larger version (27K):
[in this window]
[in a new window]

  		Fig 1. Nomogram for prediction of severity of complications with use of preoperative risk-factors. Complication categories are graded according to severity; no complications (category 0), minor to moderate complications (category 1), and major complications (category 2). Instruction: Locate the Age on the axis. Determine how many points the patient receives. Repeat this for each axis. Sum the points for all predictors and locate the sum on the Total points axis. Draw a line straight down to the bar graphs. Bar graphs represent the chance for an individual patient after esophagectomy for cancer to develop major, minor-to-moderate, or no complications. (FEV1 = forced expiratory volume in the first second; MI = myocardial infarction; TIA = transient ischemic attack.)

 
A graphical impression of the discriminative ability of the model is shown in Figure 2. This figure shows that although the mean risk score is significantly different between all three complication categories (p < 0.05) there is substantial overlap in the risk scores between patients from different categories of severity. This moderate discrimination is confirmed by a c-statistic of 0.65. Calibration plots for the prediction of the three outcome categories are shown in Figure 3. The goodness-of-fit test for ordinal models (p = 0.366) indicated that the overall agreement between the predicted and actual probabilities was good.


Figure 2
View larger version (29K):
[in this window]
[in a new window]

  		Fig 2. Distribution of risk score (by means of the total points derived by the nomogram) for the prediction of the severity of complications with use of preoperative risk factors. Complication categories are graded according to severity; no complications ({blacksquare}; category 0), minor to moderate complications ({blacklozenge}; category 1), and severe complications ({blacktriangledown}; category 2). No complications: n = 197, mean = 15.4, SD = 5.7, SE = 0.41; CI = 14.6 to 16.2. Minor complications: n = 354, mean = 18.2, SD = 6.0, SE = 0.32; CI = 17.6 to 18.8. Major complications: n = 112, mean = 21.6, SD = 7.4, SE = 0.70; CI = 20.2 to 23.0. (CI = confidence interval; SD = standard deviation; SE = standard error of the mean.)

 

Figure 3
View larger version (11K):
[in this window]
[in a new window]

  		Fig 3. Observed versus predicted plots for the prediction of the severity of complications in the nomogram development set. The dotted line represents an optimal prediction of the predictive model. The three outcome categories are represented by the three lines which indicate no complications ({blacksquare}), minor to moderate complications ({blacktriangledown}), and major complications (bullet). Observed proportions for each of the three categories are calculated for quartiles of the risk score and given with their 95% confidence interval.

 
The validity of the model was tested in an independent consecutive series of new patients (n = 95) who underwent potentially curative esophagectomy for adenocarcinoma or squamous cell carcinoma of the esophagus or GEJ in the period immediately after the developmental set. The development and test set were comparable for case-mix. As in the development set, a great overlap in distribution was seen, but again the mean risk score was significantly different between all three complication categories (p < 0.05). Calibration plots for the prediction of the three outcome categories are shown in Figure 4. The Hosmer-Lemeshow goodness-of-fit test indicated that the differences between the probabilities predicted by the model and the actual probabilities were small and nonsignificant (p = 0.626). Again a moderate discrimination was confirmed by a c-statistic of 0.66.


Figure 4
View larger version (13K):
[in this window]
[in a new window]

  		Fig 4. Observed versus predicted plots for the prediction of the severity of complications in the nomogram validation set. The dotted line represents an optimal performance of the predictive model. The three outcome categories are represented by the three lines which indicate no complications ({blacksquare}), minor to moderate complications ({blacktriangledown}), and major complications (bullet). Observed proportions for each of the three categories are calculated for quartiles of the risk score and given with their 95% confidence interval.

 

   Comment
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Comment
Acknowledgments
 References
 
Esophagectomy is accompanied by a high operative risk [1, 2]. On the basis of the present series of patients who underwent potentially curative resection for cancer of the esophagus or gastroesophageal junction, an ordinal regression model was developed to define a nomogram in order to predict the severity of complications for individual patients. Because individualized risk-assessment may help in informing the patient before surgery and in choosing the optimal extent of surgery, only preoperative risk factors were included in the model. The nomogram model showed moderate discrimination but good calibration in both the development and validation set.

Age, lung function, and comorbidity were all significantly related with the severity of complications, but the discriminative ability of the nomogram risk score when combining these preoperative factors was only moderate. Partially, this could be expected because other predictive models in esophageal surgery suffer from a low discriminative ability [10, 12, 21]. Apparently, in esophageal cancer surgery, patient-related factors are not the only factors responsible for developing complications. In this complex and extensive type of surgery many more factors may all contribute to an increased risk for the severity of complications. First, unexpected complexities during surgery resulting in a longer operation time or more blood loss are not included. Also, complexities during the early postoperative period are not included. For example, failing epidural analgesia after transthoracic esophagectomy leads to more complications, as was described recently [32]. The inclusion of those factors might increase the discriminatory effect of the nomogram, but cannot be used in the preoperative setting, when treatment-related decisions have to be made by the treating surgeon and the patient. Furthermore, our knowledge concerning the development of complications is still incomplete. Immune function and genetic alterations are known to be associated with increased susceptibility to infectious complications [33]. A recent study showed that preoperative interferon-{gamma} determination helps in the prediction of postoperative major infectious complications after esophagectomy [34]. Another study found a predictive effect in the levels of secretory leukocyte protease inhibitor levels in bronchoalveolar lavage fluid [35]. An angiotensin-converting enzyme (ACE) insertion or deletion may play a role in affecting individual susceptibility to pulmonary complications after pulmonary injury [36]. Genetic polymorphisms of mannan-binding lectin (MBL) have recently been associated with infectious complications [37]. Probably, genetic and immunologic knowledge will rapidly increase and if proven useful, a predictive model could incorporate these molecular predictors. Finally, the grouping of complications also might have an effect in the moderate discriminatory ability. This grouping might result in a large heterogeneity within each group of complications. However, for both doctor and patient, the prediction of these three categories represents important information that can be helpful in clinical decision making.

Calibration of the nomogram model was good in both the development and the validation data set, which means that predictions for groups of patients with similar risk profile match the observed probabilities. For individual patients the model can provide an indication, but certainly not a certitude, that complications will occur [22]. The nomogram based on an ordinal logistic regression model not only predicts whether or not a complication will develop, it also provides information about the severity of complications.

In esophageal cancer surgery, only one study [9] developed a risk score that has a relation with the postoperative course. In a prospective analysis it could be demonstrated that a consequent selection of patients based on this risk score could even reduce the postoperative mortality. Recently, another study confirmed that this individualized risk analysis may help to select patients for esophageal surgery based on this prediction of postoperative outcome [23]. These studies showed the importance of age and pulmonary status, which is in line with the present study. However, instead of a risk score, the present study presents a nomogram. This nomogram gives a graphical representation of the predictive strength of individual predictors and enables clinicians to calculate an overall score for individual patients reflecting their personal risk. Applying the nomogram only requires drawing lines and adding points for each individual predictor. This simplicity allows easy day-to-day clinical use for risk assessment in esophageal surgery. Moreover, a calculator can be found online at http://ats.ctsnetjournals.org/content/vol85/issue6/images/data/1938/DC1/calculator.xls. It should be realized that because the present study is not validated in an independent series, the use of this calculator should be seen as experimental.

The ASA classification, which is used by anesthesiologists all around the world to assess operative risks, did not have any relation with the severity of complications; probably because ASA was originally developed for intraoperative problems of anesthesia.

In the present study, the nomogram was validated on a relatively small validation set of comparable patients who underwent esophagectomy in our hospital after the period of model development. Although calibration was good it should be realized that this type of validation does not address the wider issue of generalizability, in which the performance of the model is examined in other settings (hospitals) with possible differences in case mix and procedures. If adequate model performance can be achieved in other settings the nomogram might be used to adjust for case mix when comparing hospital performance. High volume centers who specialize in esophageal surgery report decreasing mortality rates and report rates of less than 5% [2–6, 13]. The association between hospital volume and operative mortality has been investigated since the 1990s and is well-documented in the literature. Nearly all the studies demonstrate that high-volume medical centers in esophageal cancer surgery consistently achieve lower mortality rates, which are around 5% [2–6, 38–40]. Therefore, esophageal surgery for cancer should be regionalized in specialized high-volume hospitals [41] and case-mix models should be based on numbers achieved in these hospitals. While mortality rates decrease, morbidity is still significant and, consequently, case mix models should preferably not focus on only a rare event such as mortality but rather on the severity of complications (which also includes mortality). In this way it is probably easier to make a comparison among hospitals.

Cost efficiency also seems related with specialized high-volume care [42]. Recently a case-mix model was used to analyze cost efficiency in pancreatic surgery [43]. Where the observed number of complications was lower than expected (as measured with patient-operated selector mechanism [POSSUM]) in that study, costs decreased with more than one third. Cost reductions were primarily achieved not through decreasing duration of hospital stay or intensive care unit stay, but rather through substantial decreases in pharmacy, radiology, and laboratory costs.

In conclusion, a statistical model was developed to define a nomogram that predicts the incidence and severity of complications for individual patients after esophagectomy for cancer. The nomogram model uses only preoperative predictors and showed moderate discrimination but good calibration. The nomogram may help in informing the patient before surgery and in choosing the optimal extent of surgery. When externally validated, the nomogram may play a role in risk-adjusted audit of morbidity after esophagectomy.


   Acknowledgments
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Comment
 Acknowledgments
References
 
Dr Lagarde is supported by a grant (04-77) from the Maag Lever Darm Stichting (Dutch Digestive Foundation).


   References
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Comment
 Acknowledgments
 References
 

  1. Hulscher JB, van Sandick JW, de Boer AG, et al. Extended transthoracic resection compared with limited transhiatal resection for adenocarcinoma of the esophagus N Engl J Med 2002;347:1662-1669.[Abstract/Free Full Text]
  2. 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]
  3. Ferri LE, Law S, Wong KH, Kwok KF, Wong J. The influence of technical complications on postoperative outcome and survival after esophagectomy Ann Surg Oncol 2006;13:557-564.[Medline]
  4. Lerut T, Nafteux P, Moons J, et al. Three-field lymphadenectomy for carcinoma of the esophagus and gastroesophageal junction in 174 R0 resections: impact on staging, disease-free survival, and outcome: a plea for adaptation of TNM classification in upper-half esophageal carcinoma Ann Surg 2004;240:962-972.[Medline]
  5. Mariette C, Taillier G, Van Seuningen I, Triboulet JP. Factors affecting postoperative course and survival after en bloc resection for esophageal carcinoma Ann Thorac Surg 2004;78:1177-1183.[Abstract/Free Full Text]
  6. Portale G, Hagen JA, Peters JH, et al. Modern 5-year survival of resectable esophageal adenocarcinoma: single institution experience with 263 patients J Am Coll Surg 2006;202:588-596.[Medline]
  7. Siewert JR, Feith M, Werner M, Stein HJ. Adenocarcinoma of the esophagogastric junction: results of surgical therapy based on anatomical/topographic classification in 1,002 consecutive patients Ann Surg 2000;232:353-361.[Medline]
  8. Metzger R, Bollschweiler E, Vallböhmer D, Maish M, DeMeester TR, Hölscher AH. High volume centers for esophagectomy: what is the number needed to achieve low postoperative mortality? Dis Esophagus 2004;17:310-314.[Medline]
  9. Bartels H, Stein HJ, Siewert JR. Preoperative risk analysis and postoperative mortality of oesophagectomy for resectable oesophageal cancer Br J Surg 1998;85:840-844.[Medline]
  10. Steyerberg EW, Neville BA, Koppert LB, et al. Surgical mortality in patients with esophageal cancer: development and validation of a simple risk score J Clin Oncol 2006;24:4277-4284.[Abstract/Free Full Text]
  11. Tekkis PP, McCulloch P, Poloniecki JD, Prytherch DR, Kessaris N, Steger AC. Risk-adjusted prediction of operative mortality in oesophagogastric surgery with O-POSSUM Br J Surg 2004;91:288-295.[Medline]
  12. Lagarde SM, Maris AKD, de Castro SMM, Busch ORC, Obertop H, van Lanschot JJB. Evaluation of O-POSSUM in predicting in-hospital mortality after resection for oesophageal cancer Br J Surg 2007;94:1521-1526.[Medline]
  13. Hulscher JB, Tijssen JG, Obertop H, van Lanschot JJ. Transthoracic versus transhiatal resection for carcinoma of the esophagus: a meta-analysis Ann Thorac Surg 2001;72:306-313.[Abstract/Free Full Text]
  14. Swisher SG, Deford L, Merriman KW, et al. Effect of operative volume on morbidity, mortality, and hospital use after esophagectomy for cancer J Thorac Cardiovasc Surg 2000;119:1126-1132.[Abstract/Free Full Text]
  15. Avendano CE, Flume PA, Silvestri GA, King LB, Reed CE. Pulmonary complications after esophagectomy Ann Thorac Surg 2002;73:922-926.[Abstract/Free Full Text]
  16. Law S, Wong KH, Kwok KF, Chu KM, Wong J. Predictive factors for postoperative pulmonary complications and mortality after esophagectomy for cancer Ann Surg 2004;240:791-800.[Medline]
  17. Sauvanet A, Mariette C, Thomas P, et al. Mortality and morbidity after resection for adenocarcinoma of the gastroesophageal junction: predictive factors J Am Coll Surg 2005;201:253-262.[Medline]
  18. Gockel I, Exner C, Junginger T. Morbidity and mortality after esophagectomy for esophageal carcinoma: a risk analysis World J Surg Oncol 2005;3:37.[Medline]
  19. Abunasra H, Lewis S, Beggs L, Duffy J, Beggs D, Morgan E. Predictors of operative death after oesophagectomy for carcinoma Br J Surg 2005;92:1029-1033.[Medline]
  20. Ferguson MK, Durkin AE. Preoperative prediction of the risk of pulmonary complications after esophagectomy for cancer J Thorac Cardiovasc Surg 2002;123:661-669.[Abstract/Free Full Text]
  21. Zafirellis KD, Fountoulakis A, Dolan K, Dexter SP, Martin IG, Sue-Ling HM. Evaluation of POSSUM in patients with oesophageal cancer undergoing resection Br J Surg 2002;89:1150-1155.[Medline]
  22. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey Ann Surg 2004;240:205-213.[Medline]
  23. Schröder W, Bollschweiler E, Kossow C, Hölscher AH. Preoperative risk analysis–a reliable predictor of postoperative outcome after transthoracic esophagectomy? Langenbecks Arch Surg 2006;391:455-460.[Medline]
  24. Brennan MF, Kattan MW, Klimstra D, Conlon K. Prognostic nomogram for patients undergoing resection for adenocarcinoma of the pancreas Ann Surg 2004;240:293-298.[Medline]
  25. DiBlasio CJ, Kattan MW. Use of nomograms to predict the risk of disease recurrence after definitive local therapy for prostate cancer Urology 2003;62(suppl 1):9-18.[Medline]
  26. Kattan MW, Reuter V, Motzer RJ, Katz J, Russo P. A postoperative prognostic nomogram for renal cell carcinoma J Urol 2001;166:63-67.[Medline]
  27. Lin KY, Johns FR, Gibson J, Long M, Drake DB, Moore MM. An outcome study of breast reconstruction: presurgical identification of risk factors for complications Ann Surg Oncol 2001;8:586-591.[Medline]
  28. Hosmer DW, Lemeshow S. Applied logistic regressionNew York: Wiley; 2000.
  29. Lttle RJ, Rubin DB. Statistical analysis with missing dataNew York, NY: Wiley; 1989.
  30. Arnold AM, Kronmal RA. Multiple imputation of baseline data in the cardiovascular health study Am J Epidemiol 2003;157:74-84.[Abstract/Free Full Text]
  31. Pulkstenis E, Robinson TJ. Goodness-of-fit tests for ordinal response regression models Stat Med 2004;23:999-1014.[Medline]
  32. Cense HA, Lagarde SM, de JK, et al. Association of no epidural analgesia with postoperative morbidity and mortality after transthoracic esophageal cancer resection J Am Coll Surg 2006;202:395-400.[Medline]
  33. Hensler T, Heidecke CD, Hecker H, et al. Increased susceptibility to postoperative sepsis in patients with impaired monocyte IL-12 production J Immunol 1998;161:2655-2659.[Abstract/Free Full Text]
  34. van Sandick JW, Gisbertz SS, ten Berge IJ, et al. Immune responses and prediction of major infection in patients undergoing transhiatal or transthoracic esophagectomy for cancer Ann Surg 2003;237:35-43.[Medline]
  35. Tsukada K, Miyazaki T, Katoh H, et al. Relationship between secretory leukocyte protease inhibitor levels in bronchoalveolar lavage fluid and postoperative pulmonary complications in patients with esophageal cancer Am J Surg 2005;189:441-445.[Medline]
  36. Lee JM, Lo AC, Yang SY, Tsau HS, Chen RJ, Lee YC. Association of angiotensin-converting enzyme insertion/deletion polymorphism with serum level and development of pulmonary complications following esophagectomy Ann Surg 2005;241:659-665.[Medline]
  37. Van Till JW, Boermeester MA, Modderman PW, et al. Variable mannose-binding lectin expression during postoperative acute-phase response Surg Infect (Larchmt) 2006;7:443-452.[Medline]
  38. van Lanschot JJ, Hulscher JB, Buskens CJ, Tilanus HW, ten Kate FJ, Obertop H. Hospital volume and hospital mortality for esophagectomy Cancer 2001;91:1574-1578.[Medline]
  39. Birkmeyer JD, Sun Y, Goldfaden A, Birkmeyer NJ, Stukel TA. Volume and process of care in high-risk cancer surgery Cancer 2006;106:2476-2481.[Medline]
  40. Metzger R, Bollschweiler E, Vallböhmer D, Maish M, DeMeester TR, Hölscher AH. High volume centers for esophagectomy: what is the number needed to achieve low postoperative mortality? Dis Esophagus 2004;17:310-314.[Medline]
  41. Casson AG, van Lanschot JJ. Improving outcomes after esophagectomy: the impact of operative volume J Surg Oncol 2005;92:262-266.[Medline]
  42. Gordon TA, Bowman HM, Bass EB, et al. Complex gastrointestinal surgery: impact of provider experience on clinical and economic outcomes J Am Coll Surg 1999;189:46-56.[Medline]
  43. Vollmer Jr CM, Pratt W, Vanounou T, Maithel SK, Callery MP. Quality assessment in high-acuity surgery: volume and mortality are not enough Arch Surg 2007;142:371-380.[Abstract/Free Full Text]

Related Article

Invited Commentary
Joseph LoCicero, III
Ann. Thorac. Surg. 2008 85: 1946. [Extract] [Full Text] [PDF]



This article has been cited by other articles:


Home page
 Ann. Thorac. Surg.Home page
J. LoCicero III
Invited Commentary
Ann. Thorac. Surg., June 1, 2008; 85(6): 1946 - 1946.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Access Nomogram Calculator
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lagarde, S. M.
Right arrow Articles by van Lanschot, J. J. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lagarde, S. M.
Right arrow Articles by van Lanschot, J. J. B.
Related Collections
Right arrow Esophagus - cancer
Right arrowRelated Article

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