- Email: [email protected]
Morbidity and mortality after operation in nonbleeding cirrhotic patients
Morbidity and Mortality After Operation in Nonbleeding Cirrhotic Patients
Raymond C. Doberneck, MD, PhD, Albuquerque, New Mexico William A. Sterling, Jr., MD, Albuquerque, New Mexico David C. Allison, MD, PhD, Albuquerque, New Mexico
Recent reports by Aranha et al [1] and Schwartz [2] clearly illustrate the high mortality rate for biliary tract operations in cirrhotic patients and their findings are in contrast to those in the report by McSherry and Glenn [3] of such operations in noncirrhotic patients. The most common causes of postoperative death in cirrhotic patients include bleeding, liver failure, and sepsis. Additional reports by Smith et al [4], Zeppa et al [5], and Martin et al [S] illustrate impressively low mortality rates for seemingly high-risk elective portosystemic shunt operations in cirrhotic patients. The reason for the discrepancy in mortality rates in cirrhotic patients having biliary tract operations and portosystemic shunt operations is not obvious, nor are data available regarding postoperative mortality rates for cirrhotic patients who undergo operations other than those on the biliary tract or portosystemic shunts. Thus, the surgeon is hard pressed to advise cirrhotic patients as to the risk of operation. The purpose of this report is to provide the surgeon with data relative to operative risk for nonbleeding cirrhotic patients and to identify factors that portend a grave prognosis for operations in such patients. Material
and Methods
The charts of all cirrhotic patients who underwent operation at the University of New Mexico Hospital between January 21, 1972 and June 30, 1982 were reviewed. Included for initial analysis were cirrhotic patients who underwent elective and emergency operations and had unequivocal evidence of cirrhosis by biopsy or inspection at operation. Excluded were those cirrhotic patients who underwent operation for active gastrointestinal bleeding
From the University of New Mexico School of Medicine, Department of Surgery, Albuquerque, New Mexico. Requests for reprints should be addressed to Raymond C. Dobemeck, MD, University of New Mexico School of Medicine, Department of Surgery, Albuquerque, New Mexico 87131.
306
and diagnostic endoscopy. A total of 102 charts were ultimately reviewed in detail, and the following data were obtained: (1) demographic data: age, sex, type of operation, degree of urgency of operation, and concomitant medical condition; (2) liver function: concentrations of bilirubin, albumin, serum glutamic oxalacetic transaminase (SGOT), alkaline phosphatase, and prothrombin and partial thromboplastin times; (3) history and physical examination: previous gastrointestinal bleeding, hepatomegaly, and ascites; (4) operative and postoperative course: blood loss, complications, and outcome. Each of the preceding factors was reviewed in relation to outcome, and the additive effect of significant risk factors was related to outcome. This method of analysis was previously used by Pitt et al [7].
Results Eighty patients were men who ranged in age from 24 to 89 years (mean 45.3 years), and 22 were women
who ranged in age from 35 to 75 years (mean 47.6 years). Twenty patients (19.6 percent) died, of whom 4 (18.2 percent) were women and 16 (20 percent) were men. The most common operation was that performed on the alimentary tract, with operations on the extremities and for hernia the next most common (Table I). The following factors were associated with significantly increased postoperative mortality rates (Table II): a bilirubin concentration greater than 3.5 mg/dl, an alkaline phosphatase concentration greater than 70 U/dl, an increase in prothrombin time greater than 2 seconds, an increase on partial thromboplastin time greater than 2 seconds, emergency operation, alimentary tract operation, ascites present, operative blood loss of more than 1,000 ml, and postoperative complications. The following factors were not associated with significantly increased postoperative mortality rates (Table II): sex, albumin concentration less than 3 g/dl, serum glutamic oxalacetic transaminase concentration greater than 40 units, hepatomegaly, or previous upper gastrointestinal bleeding. TheAmerican
Journal of Surgery
Operation in Nonbleeding Cirrhotic Patients
Forty-eight (47.1 percent) patients had 75 postoperative complications, and 19 of these patients died (39.6 percent). Liver failure occurred in 43 (42.2 percent), sepsis in 19 (18.6 percent), bleeding in 9 (8.8 percent), hepatorenal syndrome in 7 (6.9 percent), and cardiac complications in 4 (3.9 percent). A variety of additional complications occurred in 12 patients (11.8 percent). The sum of the unfavorable factors present in each patient was significantly associated with the mortality rate and ranged from 5.1 percent for zero or one factor present to 66.7 percent for six or more factors present (Table III). Comments Reports by Aranha et al [1] and Schwartz [2] have clearly established the high risk of cholecystectomy in cirrhotic patients. Our study points out that the same high risk holds for cirrhotic patients who undergo a wide variety of operations. The overall mortality rate of 19.6 percent in our study is diluted by cirrhotic patients who underwent relatively low-risk operations such as genitourinary, gynecologic, and orthopedic operations, and would have been much
TABLE II
TABLE I
Operation and Mortality Rate Deaths Oo
Operation
Patients (n)
Elective portosystemic shunt Alimentary, intraperitoneal Genitourinary synecologic Hernia, extremities Thorax, central nervous system
12 46 12 23 9
0 16 1 2 1
0 34.8 8.3 8.7 11.1
102
20
19.6
Total
n
higher had patients who underwent such operations been excluded. The only general surgical operation with a relatively low mortality rate was herniorrhaphy, although the risk for herniorrhaphy in cirrhotic patients is far greater than for noncirrhotic patients. The patient complication rate of 47.1 percent is horrendous and further points out the grave risk of all operations on cirrhotic patients. As in previously published reports [1,2], the most common complications in our series were hepatic failure, sepsis, and bleeding, all of which are related to the state of liver function. Certain tests of liver function are of major
Mortality Rate Per Factor Factor
History and physical examination Ascites Yes No Hepatomegaly Yes No Previous gastrointestinal bleeding Yes No Liver function Total bilirubin >3.5 mg/dl Total bilirubin <3.5 mg/dl Alkaline phosphatase >70 U/d1 Alkaline phosphatase <70 U/dl PT increase >2 s PT increase <2 s PTT increase >2 s PTT increase <2 s Serum albumin >3 g Serum albumin (3 g SGDT >40 units SGOT <40 units Operation Emergency Elective Alimentary Nonalimentary Blood loss Cl,000 ml >l,OOO ml Postoperative complications Yes No NS = not significant.
Patients Who Died/Total Patients
Percent of Patients Who Died
p Value
9124 ii/78
37.5 14.1
‘co.05
13166 7136
19.7 19.4
NS
7134 13168
20.6 19.1
NS
a/la lll80 9122 9168 13136 6161 9118 at58 9132 10759 5/31 8758
44.4 13.8 40.9 13.2 36.1 9.8 50 13.8 28.1 16.9 16.1 13.8
<0.05
1 II24 9178 16158 4144
45.8 11.8 27.6 9.1
10178 8724
12.8 33.3
<0.05
19148 l/54
39.6 1.9
PT = prothrombin time; PTT = partial thromboplastin
Volume 148, September 1983
time; SGOT = serum glutamic oxalacetic
‘co.05
<0.05
transaminase.
307
Doberneck et al
TABLE Ill
Association of Grave Factors and Mortality Rate
Number Of Grave Factors Present
Number of Patients
o-1 2-3 4-5 >6
39 36 15 9
Number of Patients Who Died 2 7
Percent Who Died
P Value
5.1 19.4 33.3 66.7
importance in the preoperative evaluation of the risk of operation on cirrhotic patients. Prothrombin time and partial thromboplastin time in excess of 2 seconds longer than control values predict a mortality rate greater than 36 percent. Aranha et al [I] have reported a mortality rate of 83.3 percent for cholecystectomy in cirrhotic patients whose prothrombin times were more than 2.5 seconds longer than the control value and 9.3 percent for cholecystectomy in cirrhotic patients whose prothrombin times were less than 2.5 seconds longer than the control value. Schwartz [Z] has pointed out that excessive blood loss may occur during biliary tract operations on cirrhotic patients whose coagulation tests are within the normal range, and that excessive operative blood loss is associated with an increased mortality rate. He reported no deaths in 9 cirrhotic patients with moderate blood loss as opposed to two deaths in 12 cirrhotic patients with excessive (greater than 1,500 ml) blood loss during cholecystectomy. He recommended the intraoperative use of vasopressin to reduce portal flow and thereby reduce intraoperative blood loss. In our series, the mortality rate was 41.2 percent for cirrhotic patients with a bilirubin concentration greater than 3.5 mg/dl and 13.7 percent for those with concentrations less than 3.5 mg/dl. Child [8], many years ago, also made a similar observation in cirrhotic patients undergoing portosystemic shunt operations. The present report differs from those of Aranha et al [1] and Schwartz [Z] who report no association of alkaline phosphatase and mortality rate. In the present series, an alkaline phosphatase concentration greater than 70 units is associated with a mortality rate of 40.9 percent, and less than 70 units, 13.2 percent. Aranha et aI [I], however, reported that in patients undergoing cholecystectomy, a low albumin concentration is associated with a significantly higher mortality rate than is a normal albumin concentration. This observation is similar to that of Child [S] for cirrhotic patients undergoing portosystemic shunt operations. Our report fails to confirm an association of mortality rate with albumin and serum glutamic oxalacetic transaminase concentrations. In our report, the finding of ascites was associated with a mortality rate of 37.5 percent, in contrast to 14.1 percent in the absence of ascites. The association of
308
ascites and a high-mortality rate has also been reported by Child [8], but was not mentioned by Schwartz [2] and Aranha et al [I]. Hepatomegaly and previous upper gastrointestinal bleeding failed to provide discriminatory information as to the risk of operation on cirrhotic patients. In our study, the mortality rate for biliary tract operations was 35 percent, whereas Schwartz [2] reported a 26.9 percent mortality rate and Aranha et al [I] a 25.5 percent rate. Excluding such operations in our study, the mortality rate was 15.9 percent. Our study, moreover, provides data for additional types of intraabdominal operations with a mortality rate of 83.3 percent for operations on the spleen and pancreas and 22.7 percent for resection or repair of the gastrointestinal tract. The foregoing data are dismal and have prompted Aranha et al [1] and Schwartz [2] to recommend operation only on those cirrhotic patients who have a clear indication for operation. A corollary of that recommendation is that the simplest and lowest risk procedure that will suffice is best. The decision to operate and the choice of procedure are often difficult and require mature judgement for which an established body of data is not available. One must wonder why elective portosystemic shunt operations can be performed in cirrhotic patients with a comparatively low mortality rate. The most likely explanation is that the liver function of such patients is optimum, although not entirely normal, because of intensive preoperative treatment. Such a luxury is impossible in cirrhotic patients who truly require an emergency operation. A corollary is that no cirrhotic patient should undergo elective operation without the same intensive preoperative preparation that a cirrhotic patient who is to undergo an elective portosystemic shunt operation receives, and that only true emergency operations should be performed without such preparation, since the mortality rate for emergency operation was 45.8 percent in our study. The problem of the unknown cirrhotic patient remains, but a history of alcohol abuse should alert the surgeon to the possibility of preexisting cirrhosis and signal the need for further investigation of liver function. Summary The purpose of this study has been to provide information on the mortality and morbidity rates for operation on nonbleeding cirrhotic patients and to identify factors that portend a grave prognosis. A review of 102 cirrhotic patients who underwent a variety of major therapeutic operations revealed a mortality rate of 19.6 percent. Mortality rates were significantly increased (p <0.05) by emergency operation (45.8 percent), gastrointestinal related operation (27.6 percent), ascites (37.5 percent), a bilirubin concentration greater than 3.5 mg (44.4 percent), a prothrombin time increase greater than 2
The American Journal of Surgev
Operation in Nonbleeding Cirrhotic Patients
seconds (36.1 percent), a partial thromboplastin time increase greater than 2 seconds (50 percent), an alkaline phosphatase concentration greater than 70 units (40.9 percent), an operative blood loss greater than 1,000 ml (33.3 percent), and the presence of one or more postoperative complications (39.6 percent). Mortality rates were not increased after extremity, genitourinary, or gynecologic operations, an albumin concentration less than 3 g, a serum glutamic oxalacetic transaminase concentration greater than 40 units, hepatomegaly, and a history of previous gastrointestinal bleeding. When significant risk factors were added, mortality rates were significantly associated (p
simple and expeditious procedure must be performed to avoid excessive blood loss and postoperative complications, References 1. Aranha 2. 3.
4. 5. 6. 7. 8.
GV, Sontag SJ, Greenlee HB. Cholecystectomy in cirrhotic patients: a formidable operation. Am J Surg 1982; 14355-60. Schwartz SI. Biliary tract surgery and cirrhosis: a critical combination. Surgery 1981;90:577-83. McSherry CK, Glenn F. The incidence and causes of death following surgery for nonmalignant biliary tract disease. Ann Surg 1980;191:271-5. Smith RB, Warren WD, Salam AA, et al. Dacron interposition shunts for portal hypertension. Ann Surg 1980; 192:9- 17. Zeppa R, Hutson DG, Bergstresser PR, et al. Survival after distal splenorenal shunt. Surg Gynecol Obstet 1977; 145: 12-6. Martin EW Jr, Molnar J, Cooperman M, et al. Observations on 50 distal splenorenal shunts. Surgery 1978;84:379-83. Pitt HA, Cameron JL, Postur FIG, Gadacz TR. Factors affecting mortality in biliary surgery. Am J Surg 1981;141:66-72. Child CG Ill. The liver and portal hypertension. Philadelphia: WB Saunders, 198450.
See page 417 for a corresponding CME test and page 415 for a related editorial comment
Volume 146, September 1983
309
Raymond C. Doberneck, MD, PhD, Albuquerque, New Mexico William A. Sterling, Jr., MD, Albuquerque, New Mexico David C. Allison, MD, PhD, Albuquerque, New Mexico
Recent reports by Aranha et al [1] and Schwartz [2] clearly illustrate the high mortality rate for biliary tract operations in cirrhotic patients and their findings are in contrast to those in the report by McSherry and Glenn [3] of such operations in noncirrhotic patients. The most common causes of postoperative death in cirrhotic patients include bleeding, liver failure, and sepsis. Additional reports by Smith et al [4], Zeppa et al [5], and Martin et al [S] illustrate impressively low mortality rates for seemingly high-risk elective portosystemic shunt operations in cirrhotic patients. The reason for the discrepancy in mortality rates in cirrhotic patients having biliary tract operations and portosystemic shunt operations is not obvious, nor are data available regarding postoperative mortality rates for cirrhotic patients who undergo operations other than those on the biliary tract or portosystemic shunts. Thus, the surgeon is hard pressed to advise cirrhotic patients as to the risk of operation. The purpose of this report is to provide the surgeon with data relative to operative risk for nonbleeding cirrhotic patients and to identify factors that portend a grave prognosis for operations in such patients. Material
and Methods
The charts of all cirrhotic patients who underwent operation at the University of New Mexico Hospital between January 21, 1972 and June 30, 1982 were reviewed. Included for initial analysis were cirrhotic patients who underwent elective and emergency operations and had unequivocal evidence of cirrhosis by biopsy or inspection at operation. Excluded were those cirrhotic patients who underwent operation for active gastrointestinal bleeding
From the University of New Mexico School of Medicine, Department of Surgery, Albuquerque, New Mexico. Requests for reprints should be addressed to Raymond C. Dobemeck, MD, University of New Mexico School of Medicine, Department of Surgery, Albuquerque, New Mexico 87131.
306
and diagnostic endoscopy. A total of 102 charts were ultimately reviewed in detail, and the following data were obtained: (1) demographic data: age, sex, type of operation, degree of urgency of operation, and concomitant medical condition; (2) liver function: concentrations of bilirubin, albumin, serum glutamic oxalacetic transaminase (SGOT), alkaline phosphatase, and prothrombin and partial thromboplastin times; (3) history and physical examination: previous gastrointestinal bleeding, hepatomegaly, and ascites; (4) operative and postoperative course: blood loss, complications, and outcome. Each of the preceding factors was reviewed in relation to outcome, and the additive effect of significant risk factors was related to outcome. This method of analysis was previously used by Pitt et al [7].
Results Eighty patients were men who ranged in age from 24 to 89 years (mean 45.3 years), and 22 were women
who ranged in age from 35 to 75 years (mean 47.6 years). Twenty patients (19.6 percent) died, of whom 4 (18.2 percent) were women and 16 (20 percent) were men. The most common operation was that performed on the alimentary tract, with operations on the extremities and for hernia the next most common (Table I). The following factors were associated with significantly increased postoperative mortality rates (Table II): a bilirubin concentration greater than 3.5 mg/dl, an alkaline phosphatase concentration greater than 70 U/dl, an increase in prothrombin time greater than 2 seconds, an increase on partial thromboplastin time greater than 2 seconds, emergency operation, alimentary tract operation, ascites present, operative blood loss of more than 1,000 ml, and postoperative complications. The following factors were not associated with significantly increased postoperative mortality rates (Table II): sex, albumin concentration less than 3 g/dl, serum glutamic oxalacetic transaminase concentration greater than 40 units, hepatomegaly, or previous upper gastrointestinal bleeding. TheAmerican
Journal of Surgery
Operation in Nonbleeding Cirrhotic Patients
Forty-eight (47.1 percent) patients had 75 postoperative complications, and 19 of these patients died (39.6 percent). Liver failure occurred in 43 (42.2 percent), sepsis in 19 (18.6 percent), bleeding in 9 (8.8 percent), hepatorenal syndrome in 7 (6.9 percent), and cardiac complications in 4 (3.9 percent). A variety of additional complications occurred in 12 patients (11.8 percent). The sum of the unfavorable factors present in each patient was significantly associated with the mortality rate and ranged from 5.1 percent for zero or one factor present to 66.7 percent for six or more factors present (Table III). Comments Reports by Aranha et al [1] and Schwartz [2] have clearly established the high risk of cholecystectomy in cirrhotic patients. Our study points out that the same high risk holds for cirrhotic patients who undergo a wide variety of operations. The overall mortality rate of 19.6 percent in our study is diluted by cirrhotic patients who underwent relatively low-risk operations such as genitourinary, gynecologic, and orthopedic operations, and would have been much
TABLE II
TABLE I
Operation and Mortality Rate Deaths Oo
Operation
Patients (n)
Elective portosystemic shunt Alimentary, intraperitoneal Genitourinary synecologic Hernia, extremities Thorax, central nervous system
12 46 12 23 9
0 16 1 2 1
0 34.8 8.3 8.7 11.1
102
20
19.6
Total
n
higher had patients who underwent such operations been excluded. The only general surgical operation with a relatively low mortality rate was herniorrhaphy, although the risk for herniorrhaphy in cirrhotic patients is far greater than for noncirrhotic patients. The patient complication rate of 47.1 percent is horrendous and further points out the grave risk of all operations on cirrhotic patients. As in previously published reports [1,2], the most common complications in our series were hepatic failure, sepsis, and bleeding, all of which are related to the state of liver function. Certain tests of liver function are of major
Mortality Rate Per Factor Factor
History and physical examination Ascites Yes No Hepatomegaly Yes No Previous gastrointestinal bleeding Yes No Liver function Total bilirubin >3.5 mg/dl Total bilirubin <3.5 mg/dl Alkaline phosphatase >70 U/d1 Alkaline phosphatase <70 U/dl PT increase >2 s PT increase <2 s PTT increase >2 s PTT increase <2 s Serum albumin >3 g Serum albumin (3 g SGDT >40 units SGOT <40 units Operation Emergency Elective Alimentary Nonalimentary Blood loss Cl,000 ml >l,OOO ml Postoperative complications Yes No NS = not significant.
Patients Who Died/Total Patients
Percent of Patients Who Died
p Value
9124 ii/78
37.5 14.1
‘co.05
13166 7136
19.7 19.4
NS
7134 13168
20.6 19.1
NS
a/la lll80 9122 9168 13136 6161 9118 at58 9132 10759 5/31 8758
44.4 13.8 40.9 13.2 36.1 9.8 50 13.8 28.1 16.9 16.1 13.8
<0.05
1 II24 9178 16158 4144
45.8 11.8 27.6 9.1
10178 8724
12.8 33.3
<0.05
19148 l/54
39.6 1.9
PT = prothrombin time; PTT = partial thromboplastin
Volume 148, September 1983
time; SGOT = serum glutamic oxalacetic
‘co.05
<0.05
transaminase.
307
Doberneck et al
TABLE Ill
Association of Grave Factors and Mortality Rate
Number Of Grave Factors Present
Number of Patients
o-1 2-3 4-5 >6
39 36 15 9
Number of Patients Who Died 2 7
Percent Who Died
P Value
5.1 19.4 33.3 66.7
importance in the preoperative evaluation of the risk of operation on cirrhotic patients. Prothrombin time and partial thromboplastin time in excess of 2 seconds longer than control values predict a mortality rate greater than 36 percent. Aranha et al [I] have reported a mortality rate of 83.3 percent for cholecystectomy in cirrhotic patients whose prothrombin times were more than 2.5 seconds longer than the control value and 9.3 percent for cholecystectomy in cirrhotic patients whose prothrombin times were less than 2.5 seconds longer than the control value. Schwartz [Z] has pointed out that excessive blood loss may occur during biliary tract operations on cirrhotic patients whose coagulation tests are within the normal range, and that excessive operative blood loss is associated with an increased mortality rate. He reported no deaths in 9 cirrhotic patients with moderate blood loss as opposed to two deaths in 12 cirrhotic patients with excessive (greater than 1,500 ml) blood loss during cholecystectomy. He recommended the intraoperative use of vasopressin to reduce portal flow and thereby reduce intraoperative blood loss. In our series, the mortality rate was 41.2 percent for cirrhotic patients with a bilirubin concentration greater than 3.5 mg/dl and 13.7 percent for those with concentrations less than 3.5 mg/dl. Child [8], many years ago, also made a similar observation in cirrhotic patients undergoing portosystemic shunt operations. The present report differs from those of Aranha et al [1] and Schwartz [Z] who report no association of alkaline phosphatase and mortality rate. In the present series, an alkaline phosphatase concentration greater than 70 units is associated with a mortality rate of 40.9 percent, and less than 70 units, 13.2 percent. Aranha et aI [I], however, reported that in patients undergoing cholecystectomy, a low albumin concentration is associated with a significantly higher mortality rate than is a normal albumin concentration. This observation is similar to that of Child [S] for cirrhotic patients undergoing portosystemic shunt operations. Our report fails to confirm an association of mortality rate with albumin and serum glutamic oxalacetic transaminase concentrations. In our report, the finding of ascites was associated with a mortality rate of 37.5 percent, in contrast to 14.1 percent in the absence of ascites. The association of
308
ascites and a high-mortality rate has also been reported by Child [8], but was not mentioned by Schwartz [2] and Aranha et al [I]. Hepatomegaly and previous upper gastrointestinal bleeding failed to provide discriminatory information as to the risk of operation on cirrhotic patients. In our study, the mortality rate for biliary tract operations was 35 percent, whereas Schwartz [2] reported a 26.9 percent mortality rate and Aranha et al [I] a 25.5 percent rate. Excluding such operations in our study, the mortality rate was 15.9 percent. Our study, moreover, provides data for additional types of intraabdominal operations with a mortality rate of 83.3 percent for operations on the spleen and pancreas and 22.7 percent for resection or repair of the gastrointestinal tract. The foregoing data are dismal and have prompted Aranha et al [1] and Schwartz [2] to recommend operation only on those cirrhotic patients who have a clear indication for operation. A corollary of that recommendation is that the simplest and lowest risk procedure that will suffice is best. The decision to operate and the choice of procedure are often difficult and require mature judgement for which an established body of data is not available. One must wonder why elective portosystemic shunt operations can be performed in cirrhotic patients with a comparatively low mortality rate. The most likely explanation is that the liver function of such patients is optimum, although not entirely normal, because of intensive preoperative treatment. Such a luxury is impossible in cirrhotic patients who truly require an emergency operation. A corollary is that no cirrhotic patient should undergo elective operation without the same intensive preoperative preparation that a cirrhotic patient who is to undergo an elective portosystemic shunt operation receives, and that only true emergency operations should be performed without such preparation, since the mortality rate for emergency operation was 45.8 percent in our study. The problem of the unknown cirrhotic patient remains, but a history of alcohol abuse should alert the surgeon to the possibility of preexisting cirrhosis and signal the need for further investigation of liver function. Summary The purpose of this study has been to provide information on the mortality and morbidity rates for operation on nonbleeding cirrhotic patients and to identify factors that portend a grave prognosis. A review of 102 cirrhotic patients who underwent a variety of major therapeutic operations revealed a mortality rate of 19.6 percent. Mortality rates were significantly increased (p <0.05) by emergency operation (45.8 percent), gastrointestinal related operation (27.6 percent), ascites (37.5 percent), a bilirubin concentration greater than 3.5 mg (44.4 percent), a prothrombin time increase greater than 2
The American Journal of Surgev
Operation in Nonbleeding Cirrhotic Patients
seconds (36.1 percent), a partial thromboplastin time increase greater than 2 seconds (50 percent), an alkaline phosphatase concentration greater than 70 units (40.9 percent), an operative blood loss greater than 1,000 ml (33.3 percent), and the presence of one or more postoperative complications (39.6 percent). Mortality rates were not increased after extremity, genitourinary, or gynecologic operations, an albumin concentration less than 3 g, a serum glutamic oxalacetic transaminase concentration greater than 40 units, hepatomegaly, and a history of previous gastrointestinal bleeding. When significant risk factors were added, mortality rates were significantly associated (p
simple and expeditious procedure must be performed to avoid excessive blood loss and postoperative complications, References 1. Aranha 2. 3.
4. 5. 6. 7. 8.
GV, Sontag SJ, Greenlee HB. Cholecystectomy in cirrhotic patients: a formidable operation. Am J Surg 1982; 14355-60. Schwartz SI. Biliary tract surgery and cirrhosis: a critical combination. Surgery 1981;90:577-83. McSherry CK, Glenn F. The incidence and causes of death following surgery for nonmalignant biliary tract disease. Ann Surg 1980;191:271-5. Smith RB, Warren WD, Salam AA, et al. Dacron interposition shunts for portal hypertension. Ann Surg 1980; 192:9- 17. Zeppa R, Hutson DG, Bergstresser PR, et al. Survival after distal splenorenal shunt. Surg Gynecol Obstet 1977; 145: 12-6. Martin EW Jr, Molnar J, Cooperman M, et al. Observations on 50 distal splenorenal shunts. Surgery 1978;84:379-83. Pitt HA, Cameron JL, Postur FIG, Gadacz TR. Factors affecting mortality in biliary surgery. Am J Surg 1981;141:66-72. Child CG Ill. The liver and portal hypertension. Philadelphia: WB Saunders, 198450.
See page 417 for a corresponding CME test and page 415 for a related editorial comment
Volume 146, September 1983
309