Is it safe to perform esophagectomy in esophageal cancer patients combined with liver cirrhosis
http://www.100md.com
《血管的通路杂志》
Division of Thoracic Surgery, Chang Gung Memorial Hospital, Chang Gung University, 5 Fu-Hsing Street, Kweishan, Taoyuan, Taiwan 333
Abstract
Patients with liver cirrhosis undergoing gastrointestinal surgery still suffer from high operative morbid-mortality despite advancements in surgical critical care. The objective of this study is to see if this same relationship applies to patients undergoing esophagectomy for cancer. From 1993 to 2003, sixteen esophageal cancer patients with liver cirrhosis were operated on. They were all male with a mean age of 51.5 years. According to the Child-Pugh classification, 10 patients were Child ‘A’, 4 patients Child ‘B’ and Child ‘C’ in 2 patients. The surgical procedure was through an Ivor-Lewis esophagogastrectomy with intra-thoracic anastomosis. Major morbidity included: 4 respiratory failure, 2 acute renal failure, 3 pneumonia, and one in each of the patients with gastrointestinal bleeding and hepatic failure. The mean follow up among the survivors was 19.1 months. The hospital mortality was 25% (4/16). Using the rate according to Child classification, the mortality rates were: A: 1/10 (10%), B: 2/4 (50%) and C: 2/2 (100%). We conclude that patients with liver cirrhosis in Child-Pugh A could tolerate esophagectomy with an acceptable risk. However, patients with a more advanced state of liver dysfunction are at higher risk for esophagogastrectomy. Careful patient selection and meticulous peri-operative care is warranted in those embarking on surgical resection.
Key Words: Esophageal cancer; Esophageal surgery; Perioperative care
1. Introduction
Esophageal cancer is one of the most lethal cancers. Yearly, 400,000 new cases of esophageal cancer are diagnosed worldwide and more than 350,000 of them die from this disease [1]. Different treatment modalities have evolved for the treatment of this lethal disease but up to now, surgery remains the mainstay treatment. Patients with liver cirrhosis undergoing different surgical procedures are at increased risk for post-operative complications. Liver cirrhosis has been listed as the 10th most common cause of mortality in the United States with more than 25,000 deaths in 1998 [2]. It is considered per se as an end-stage of liver disease that precludes a shorter life expectancy for the patient [3]. This study reviews the results of patients with liver cirrhosis and esophageal cancer undergoing esophagectomy and gastric tube reconstruction.
2. Patients and method
We reviewed the clinical records of adults (age >18) who had a diagnosis of liver cirrhosis and underwent esophagectomy for esophageal cancer between January 1993 and December 2003, in the Division of Thoracic Surgery, Chang Gung Memorial Hospital, Taiwan. Seventeen patients were operated on and were matched with the diagnosis of esophageal cancer and cirrhosis of the liver. Cirrhosis was diagnosed by liver biopsy or by a history of liver disease with an impaired liver function test, plus one or more of the following: liver ultrasound or computed tomography scan suggesting the diagnosis of liver cirrhosis, or direct examination of the liver during abdominal examination. Pre-operative data included age, gender, co-existing medical disease, tumor location, cancer stage, history of neoadjuvant therapy, type of liver cirrhosis; and the presence of ascitis and hepatic encephalopathy. Pre-operative chemistry including albumin, total bilirubin, creatinine level and prothrombin time were recorded. Child's classification as modified by Pugh [4] was used as a measure of liver dysfunction.
2.1. Surgical procedure
All patients received an Ivor-Lewis esophagogectomy with intrathoracic anastomosis and two field lymph nodes dissections. Laparotomy was performed through an upper midline incision. The stomach was harvested preserving the right gastroepiploic artery as the main arterial supply. Drainage procedure such as pyloro-plasty or pyloromyotomy was done only in the case of a well-documented outlet obstruction. The esophagectomy was performed through a right transthoracic subtotal esophagectomy. Reconstruction with gastric tube was done above the azygos vein with a mechanical stapler. During lymph nodes dissection, meticulous care was taken to ligate or clip the small lymph duct in an attempt to reduce lymphorrhea. Post-operative care included admission to the intensive care unit, fluid restriction, albumin and fresh frozen plasma infusion, and diuretic agent as needed. Post-operative complications such as pneumonia, sepsis, prolonged ventilator use (>3 days) and renal failure requiring hemodialysis were included for analysis. Surgical mortality was defined as those occurring within 30 days of operation or when this occurred during the same admission.
The Chi-square (2) test and Student's t-test were used for comparative analysis. The results were considered to be significant if the value was <0.05 (P<0.05).
3. Results
Among 389 consecutive patients with primary esophageal cancer who were referred for esophagectomy, seventeen patients matched for the diagnosis of liver cirrhosis and esophageal cancer. One patient was excluded because of insufficient clinical records. Sixteen patients were included for analysis. All 16 patients were male and had a history of cigarette smoking. The mean age was 51.5 (range 43–66) years. The mean follow up period was 19.1 months. The pathology was squamous cell carcinoma in all cases, 1 patient at upper thoracic esophagus, 13 at middle third and 2 patients in the lower third. One patient had hepatitis C, 2 patients had hepatitis B and 13 patients were alcohol related. Classified as Child ‘A’ were 10 patients, Child ‘B’ 4 patients and Child ‘C’ 2 patients (Table 1).
3.1. Postoperative morbidity and mortality
Five patients experienced 11 major complications (5/16, 31.25%); pneumonia, 3; acute renal failure, 2 (one patient required transient dialysis); acute respiratory failure with prolonged mechanical ventilation support, 4; and gastrointestinal bleeding and hepatic failure, one in each case.
Four surgical mortalities were registered. Three cases were related to pneumonia and septic complications and one liver failure related. Three additional patients died during the follow-up; pneumonia and septic complication, hepatoma and hepatic failure, one in each case. Among the survivors, two patients presented with recurrence at an average of 27 months after the operation. The surgical mortality rate was 25% (4/16). According to Child classification the mortality rates were Child A 1/10 (10%), Child B 2/4 (50%) and Child C 2/2 (100%).
3.2. Comparison between non-surgical mortality group (NSM) and surgical mortality group (SM)
The mean age for SM was 52.1 years against 49.5 years in the NSM (P=0.55). The tumor location (P=0.75) and the etiology of liver cirrhosis (P=0.65) did not differ significantly among the NSM and SM groups. However, Child-Pugh classification did show a significant difference between NSM and SM (P=0.012). The analysis of components of Child-Pugh classification revealed that albumin concentration (2.8 mg/dl, P=0.028), prothrombin time (>4 increases in seconds, P=0.022) and the presence of ascites (P=0.000), were associated with increased risk for surgical mortality (Table 2).
4. Comment
The frequency of complication and eventual postoperative death after abdominal operation in patients with liver cirrhosis has been clearly confirmed by several authors [5–7] but only a few reports deal with cirrhotic patients associated with esophageal cancer [8–10]. The overall surgical mortality for esophageal cancer has been estimated at 9.8% [11], but it could be as low as 1.4% when it is performed in a high volume center [12]. However, this overall result could not be reflected in the selected group of patients with liver cirrhosis; it may reach as high as 26% [8].
Child-Pugh classification has been widely used for the evaluation of hepatic functional reserve. In this report, patients with Child ‘A’ liver cirrhosis presented with a surgical mortality of 10%, that is, comparable with general non-cirrhotic patients. Nevertheless, the risk may overwhelm the potential benefit in patients with a more advanced hepatic dysfunction.
Analysis of components of the Child-Pugh classification revealed that albumin level (<2.8 mg/dl), prothrombin time (>4 s) and the presence of ascites were associated with increased risk for surgical mortality. Patients in the NSM presented with a higher serum level of albumin and a lesser increase in prothrombin time compared to the SM group (P=0.028 and P=0.022). In the mortality group, all deceased patients registered the presence of ascites from mild to moderate, whereas only two patients had mild ascites in the non-surgical group. The management of ascites may be especially troublesome in the postoperative period. It could be treated by sodium restriction, with or without diuretics, or paracentesis and drainage. All our patients were managed with conservative treatment, and no drain was placed after the completion of laparotomy. Their use may increase complications such as ascite leaks through the drainage site, infection of ascitis fluid [5] and metabolic disorder. Further analysis of other components of Child-Pugh classification showed that the bilirubin level did not achieve statistical difference among the NSM and SM group. Patients with a neurologic disorder were excluded for operation in the present report.
The analysis of the SM group revealed that 75% (3/4) of cases were related to pneumonia and sepsis. Pneumonia and respiratory failure have been cited as the most frequent complications after esophagectomy [11]. Law et al. reported pneumonia as being responsible for 55% of hospital deaths in patients with liver cirrhosis [14]. The inability to wean patients from mechanical ventilation for more than three days in this review was associated with a mortality rate of 100%. An increased risk for respiratory failure and death, as well as an increased risk for hospitalization associated with sepsis and death from sepsis, has been reported for patients with liver cirrhosis [15].
Liver associated disease was responsible for death only in our SM group. This is somehow different from gastrectomy for gastric cancer in cirrhotic patients where a cirrhosis related complication was the main reason [13]. Tachibana et al. [10] have explained this discrepancy. First, esophagectomy implied a radical esophagogastric devascularization. Second, the loss of appetite after esophagectomy may render these patients unable for alcohol consumption. Thus, the patients might have some liver function recovery.
Anastomotic leakage has been cited as a disastrous complication. We have not encountered such a complication. Most thoracic surgeons perform the esophagogram on post-operative 5th to 7th days. We routinely keep a nothing per mouth and parenteral nutrition policy for two weeks before the water-soluble esophagogram is performed. Perhaps these two weeks allow any possible ‘subclinical leakage’ to heal before oral intake is resumed.
Despite the small number of cases and retrospective analysis, our analysis suggests that patients with esophageal cancer and liver cirrhosis Child-Pugh A can undergo esophagectomy with an acceptable risk. Nevertheless, very careful patient selection is required for patients in a more advanced stage of liver disease and meticulous peri-operative care is warranted in those embarking on surgical resection.
References
Parkin DM, Bray FI, Devesa SS. Cancer burden in the year 2000.The global picture. Eur J Cancer 2001;37:S4–66.
Vital Health Stat 13, No.145 . National Center for Health Statistics.
Gines P, Quintero E, Arroyo V, Teres J, Bruguera M, Rimola A, Caballeria J, Rodes J, Rozman C. Compensated cirrhosis: Natural history and prognosis factors. Hepatology 1987;7:122–128.
Pugh RNH, Murray-Lyon IM, Dawson JL, Pietroni MC, Williams R. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 1973;60:646–649.
Garrison RN, Cryer HM, Howard DA, Polk HC. Clarification of risk factors for abdominal operations in patients with hepatic cirrhosis. Ann Surg 1984;199:648–655.
Aranha GV, Greenlee HB. Intra-abdominal surgery in patients with advanced cirrhosis. Arch Surg 1986;121:275–277.
Mansour A, Watson W, Shayani V, Pickleman J. Abdominal operations in patients with liver cirrhosis: Still a major surgical challenge. Surgery 1997;122:730–736.
Fekete F, Belghiti J, Cherqui D, Langonnet F, Gayet B. Results of esophagogastrectomy for carcinoma in cirrhotic patients: a series of 23 consecutive patients. Ann Surg 1987;206:74–78.
Belghiti J, Cherqui D, Langonnet F, Fekete F. Esophagogastrectomy for carcinoma in cirrhotic patients. Hepatogastroenterol 1990;37:388–391.
Tachibana M, Kotoh T, Kinugasa S, Dhar DK, Shibakita M, Ohno S, Masunaga R, Kubota H, Kohno H, Nagasue N. Esophageal cancer with cirrhosis of the liver: Results of esophagectomy in 18 consecutive patients. Ann Surg Oncol 2000;7:758–763.
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;75:217–222.
Visbal AL, Allen MS, Miller DL, Deschamps C, Trastek VF, Pairolero PC. Ivor Lewis esophagogastrectomy for esophageal cancer. Ann Thorac Surg 2001;71:1803–1808.
Isozaki H, Okajima K, Ichinona T, Fujii K, Nomura E, Izumi N. Surgery for gastric cancer in patients with cirrhosis. Surg Today 1997;27:17–21.
Law S, Wong KH, Kwok KF, Chu KM, Wong J. Predictive factors for postoperative pulmonary complications and mortality after esophagectomy for cancer. Ann Surg 240:791–800.
Foreman MG, Mannino DM, Moss M. Cirrhosis as a risk factor for sepsis and death. Analysis of the national hospital discharge survey. Chest 2003;124:1016–1020.(Ming-Shian Lu, Yun-Hen Li)
Abstract
Patients with liver cirrhosis undergoing gastrointestinal surgery still suffer from high operative morbid-mortality despite advancements in surgical critical care. The objective of this study is to see if this same relationship applies to patients undergoing esophagectomy for cancer. From 1993 to 2003, sixteen esophageal cancer patients with liver cirrhosis were operated on. They were all male with a mean age of 51.5 years. According to the Child-Pugh classification, 10 patients were Child ‘A’, 4 patients Child ‘B’ and Child ‘C’ in 2 patients. The surgical procedure was through an Ivor-Lewis esophagogastrectomy with intra-thoracic anastomosis. Major morbidity included: 4 respiratory failure, 2 acute renal failure, 3 pneumonia, and one in each of the patients with gastrointestinal bleeding and hepatic failure. The mean follow up among the survivors was 19.1 months. The hospital mortality was 25% (4/16). Using the rate according to Child classification, the mortality rates were: A: 1/10 (10%), B: 2/4 (50%) and C: 2/2 (100%). We conclude that patients with liver cirrhosis in Child-Pugh A could tolerate esophagectomy with an acceptable risk. However, patients with a more advanced state of liver dysfunction are at higher risk for esophagogastrectomy. Careful patient selection and meticulous peri-operative care is warranted in those embarking on surgical resection.
Key Words: Esophageal cancer; Esophageal surgery; Perioperative care
1. Introduction
Esophageal cancer is one of the most lethal cancers. Yearly, 400,000 new cases of esophageal cancer are diagnosed worldwide and more than 350,000 of them die from this disease [1]. Different treatment modalities have evolved for the treatment of this lethal disease but up to now, surgery remains the mainstay treatment. Patients with liver cirrhosis undergoing different surgical procedures are at increased risk for post-operative complications. Liver cirrhosis has been listed as the 10th most common cause of mortality in the United States with more than 25,000 deaths in 1998 [2]. It is considered per se as an end-stage of liver disease that precludes a shorter life expectancy for the patient [3]. This study reviews the results of patients with liver cirrhosis and esophageal cancer undergoing esophagectomy and gastric tube reconstruction.
2. Patients and method
We reviewed the clinical records of adults (age >18) who had a diagnosis of liver cirrhosis and underwent esophagectomy for esophageal cancer between January 1993 and December 2003, in the Division of Thoracic Surgery, Chang Gung Memorial Hospital, Taiwan. Seventeen patients were operated on and were matched with the diagnosis of esophageal cancer and cirrhosis of the liver. Cirrhosis was diagnosed by liver biopsy or by a history of liver disease with an impaired liver function test, plus one or more of the following: liver ultrasound or computed tomography scan suggesting the diagnosis of liver cirrhosis, or direct examination of the liver during abdominal examination. Pre-operative data included age, gender, co-existing medical disease, tumor location, cancer stage, history of neoadjuvant therapy, type of liver cirrhosis; and the presence of ascitis and hepatic encephalopathy. Pre-operative chemistry including albumin, total bilirubin, creatinine level and prothrombin time were recorded. Child's classification as modified by Pugh [4] was used as a measure of liver dysfunction.
2.1. Surgical procedure
All patients received an Ivor-Lewis esophagogectomy with intrathoracic anastomosis and two field lymph nodes dissections. Laparotomy was performed through an upper midline incision. The stomach was harvested preserving the right gastroepiploic artery as the main arterial supply. Drainage procedure such as pyloro-plasty or pyloromyotomy was done only in the case of a well-documented outlet obstruction. The esophagectomy was performed through a right transthoracic subtotal esophagectomy. Reconstruction with gastric tube was done above the azygos vein with a mechanical stapler. During lymph nodes dissection, meticulous care was taken to ligate or clip the small lymph duct in an attempt to reduce lymphorrhea. Post-operative care included admission to the intensive care unit, fluid restriction, albumin and fresh frozen plasma infusion, and diuretic agent as needed. Post-operative complications such as pneumonia, sepsis, prolonged ventilator use (>3 days) and renal failure requiring hemodialysis were included for analysis. Surgical mortality was defined as those occurring within 30 days of operation or when this occurred during the same admission.
The Chi-square (2) test and Student's t-test were used for comparative analysis. The results were considered to be significant if the value was <0.05 (P<0.05).
3. Results
Among 389 consecutive patients with primary esophageal cancer who were referred for esophagectomy, seventeen patients matched for the diagnosis of liver cirrhosis and esophageal cancer. One patient was excluded because of insufficient clinical records. Sixteen patients were included for analysis. All 16 patients were male and had a history of cigarette smoking. The mean age was 51.5 (range 43–66) years. The mean follow up period was 19.1 months. The pathology was squamous cell carcinoma in all cases, 1 patient at upper thoracic esophagus, 13 at middle third and 2 patients in the lower third. One patient had hepatitis C, 2 patients had hepatitis B and 13 patients were alcohol related. Classified as Child ‘A’ were 10 patients, Child ‘B’ 4 patients and Child ‘C’ 2 patients (Table 1).
3.1. Postoperative morbidity and mortality
Five patients experienced 11 major complications (5/16, 31.25%); pneumonia, 3; acute renal failure, 2 (one patient required transient dialysis); acute respiratory failure with prolonged mechanical ventilation support, 4; and gastrointestinal bleeding and hepatic failure, one in each case.
Four surgical mortalities were registered. Three cases were related to pneumonia and septic complications and one liver failure related. Three additional patients died during the follow-up; pneumonia and septic complication, hepatoma and hepatic failure, one in each case. Among the survivors, two patients presented with recurrence at an average of 27 months after the operation. The surgical mortality rate was 25% (4/16). According to Child classification the mortality rates were Child A 1/10 (10%), Child B 2/4 (50%) and Child C 2/2 (100%).
3.2. Comparison between non-surgical mortality group (NSM) and surgical mortality group (SM)
The mean age for SM was 52.1 years against 49.5 years in the NSM (P=0.55). The tumor location (P=0.75) and the etiology of liver cirrhosis (P=0.65) did not differ significantly among the NSM and SM groups. However, Child-Pugh classification did show a significant difference between NSM and SM (P=0.012). The analysis of components of Child-Pugh classification revealed that albumin concentration (2.8 mg/dl, P=0.028), prothrombin time (>4 increases in seconds, P=0.022) and the presence of ascites (P=0.000), were associated with increased risk for surgical mortality (Table 2).
4. Comment
The frequency of complication and eventual postoperative death after abdominal operation in patients with liver cirrhosis has been clearly confirmed by several authors [5–7] but only a few reports deal with cirrhotic patients associated with esophageal cancer [8–10]. The overall surgical mortality for esophageal cancer has been estimated at 9.8% [11], but it could be as low as 1.4% when it is performed in a high volume center [12]. However, this overall result could not be reflected in the selected group of patients with liver cirrhosis; it may reach as high as 26% [8].
Child-Pugh classification has been widely used for the evaluation of hepatic functional reserve. In this report, patients with Child ‘A’ liver cirrhosis presented with a surgical mortality of 10%, that is, comparable with general non-cirrhotic patients. Nevertheless, the risk may overwhelm the potential benefit in patients with a more advanced hepatic dysfunction.
Analysis of components of the Child-Pugh classification revealed that albumin level (<2.8 mg/dl), prothrombin time (>4 s) and the presence of ascites were associated with increased risk for surgical mortality. Patients in the NSM presented with a higher serum level of albumin and a lesser increase in prothrombin time compared to the SM group (P=0.028 and P=0.022). In the mortality group, all deceased patients registered the presence of ascites from mild to moderate, whereas only two patients had mild ascites in the non-surgical group. The management of ascites may be especially troublesome in the postoperative period. It could be treated by sodium restriction, with or without diuretics, or paracentesis and drainage. All our patients were managed with conservative treatment, and no drain was placed after the completion of laparotomy. Their use may increase complications such as ascite leaks through the drainage site, infection of ascitis fluid [5] and metabolic disorder. Further analysis of other components of Child-Pugh classification showed that the bilirubin level did not achieve statistical difference among the NSM and SM group. Patients with a neurologic disorder were excluded for operation in the present report.
The analysis of the SM group revealed that 75% (3/4) of cases were related to pneumonia and sepsis. Pneumonia and respiratory failure have been cited as the most frequent complications after esophagectomy [11]. Law et al. reported pneumonia as being responsible for 55% of hospital deaths in patients with liver cirrhosis [14]. The inability to wean patients from mechanical ventilation for more than three days in this review was associated with a mortality rate of 100%. An increased risk for respiratory failure and death, as well as an increased risk for hospitalization associated with sepsis and death from sepsis, has been reported for patients with liver cirrhosis [15].
Liver associated disease was responsible for death only in our SM group. This is somehow different from gastrectomy for gastric cancer in cirrhotic patients where a cirrhosis related complication was the main reason [13]. Tachibana et al. [10] have explained this discrepancy. First, esophagectomy implied a radical esophagogastric devascularization. Second, the loss of appetite after esophagectomy may render these patients unable for alcohol consumption. Thus, the patients might have some liver function recovery.
Anastomotic leakage has been cited as a disastrous complication. We have not encountered such a complication. Most thoracic surgeons perform the esophagogram on post-operative 5th to 7th days. We routinely keep a nothing per mouth and parenteral nutrition policy for two weeks before the water-soluble esophagogram is performed. Perhaps these two weeks allow any possible ‘subclinical leakage’ to heal before oral intake is resumed.
Despite the small number of cases and retrospective analysis, our analysis suggests that patients with esophageal cancer and liver cirrhosis Child-Pugh A can undergo esophagectomy with an acceptable risk. Nevertheless, very careful patient selection is required for patients in a more advanced stage of liver disease and meticulous peri-operative care is warranted in those embarking on surgical resection.
References
Parkin DM, Bray FI, Devesa SS. Cancer burden in the year 2000.The global picture. Eur J Cancer 2001;37:S4–66.
Vital Health Stat 13, No.145 . National Center for Health Statistics.
Gines P, Quintero E, Arroyo V, Teres J, Bruguera M, Rimola A, Caballeria J, Rodes J, Rozman C. Compensated cirrhosis: Natural history and prognosis factors. Hepatology 1987;7:122–128.
Pugh RNH, Murray-Lyon IM, Dawson JL, Pietroni MC, Williams R. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 1973;60:646–649.
Garrison RN, Cryer HM, Howard DA, Polk HC. Clarification of risk factors for abdominal operations in patients with hepatic cirrhosis. Ann Surg 1984;199:648–655.
Aranha GV, Greenlee HB. Intra-abdominal surgery in patients with advanced cirrhosis. Arch Surg 1986;121:275–277.
Mansour A, Watson W, Shayani V, Pickleman J. Abdominal operations in patients with liver cirrhosis: Still a major surgical challenge. Surgery 1997;122:730–736.
Fekete F, Belghiti J, Cherqui D, Langonnet F, Gayet B. Results of esophagogastrectomy for carcinoma in cirrhotic patients: a series of 23 consecutive patients. Ann Surg 1987;206:74–78.
Belghiti J, Cherqui D, Langonnet F, Fekete F. Esophagogastrectomy for carcinoma in cirrhotic patients. Hepatogastroenterol 1990;37:388–391.
Tachibana M, Kotoh T, Kinugasa S, Dhar DK, Shibakita M, Ohno S, Masunaga R, Kubota H, Kohno H, Nagasue N. Esophageal cancer with cirrhosis of the liver: Results of esophagectomy in 18 consecutive patients. Ann Surg Oncol 2000;7:758–763.
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;75:217–222.
Visbal AL, Allen MS, Miller DL, Deschamps C, Trastek VF, Pairolero PC. Ivor Lewis esophagogastrectomy for esophageal cancer. Ann Thorac Surg 2001;71:1803–1808.
Isozaki H, Okajima K, Ichinona T, Fujii K, Nomura E, Izumi N. Surgery for gastric cancer in patients with cirrhosis. Surg Today 1997;27:17–21.
Law S, Wong KH, Kwok KF, Chu KM, Wong J. Predictive factors for postoperative pulmonary complications and mortality after esophagectomy for cancer. Ann Surg 240:791–800.
Foreman MG, Mannino DM, Moss M. Cirrhosis as a risk factor for sepsis and death. Analysis of the national hospital discharge survey. Chest 2003;124:1016–1020.(Ming-Shian Lu, Yun-Hen Li)