| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
 
|
|
|
||
|
|
Anesth Analg 2003;97:550-554 © 2003 International Anesthesia Research Society
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
, |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
The impact of the hepatitis C virus (HCV) infection on the postoperative complication rate is unknown. We identified a population of surgical patients (n = 2457) for whom the HCV antibody (anti-HCV) had been measured and compared after surgical complications and mortality between those who were positive (17.9%) versus negative. The complication rates were 10% in the anti-HCV positive and 13% in the negative group (P = 0.125), whereas the mortality rates were 0.7% and 2.5%, respectively (P = 0.017). The anti-HCV positive patients were younger, had lower ASA physical status, and underwent shorter procedures. In the univariate analysis, emergent surgery and high ASA physical status but not anti-HCV positivity were associated with a more frequent complication. In the multivariate analysis, the urgency of surgery, age, ASA physical status, length of surgery, and preoperative hematocrit (but not platelet count) were associated with complications. Anti-HCV positivity was associated with an odds ratio for having a complication of 1.08 (95% confidence interval, 0.90–1.30), which was not statistically significant (P = 0.405). In conclusion, we were unable to show HCV antibody status to be an independent risk factor for postoperative complications when other co-factors were considered.
IMPLICATIONS: In this large study at a Veterans Administration medical center, the urgency of surgery, age, ASA physical status, length of surgery, and preoperative hematocrit were all independently associated with postoperative complications. However, hepatitis C infection was not an independent risk factor for postoperative complications.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The initial studies on surgical and anesthesia risk in patients with liver disease were all performed before the discovery of hepatitis C virus (HCV). Chronic HCV infection is now the most common cause of chronic liver disease in the United States. The incidence of HCV is 1.8% in the general population (1) and is as frequent as 17.7% in the veteran population (2,3). Because patients with chronic HCV infection often present for surgery, it is important to know if these patients are at increased risk for postoperative complications. Unfortunately, most of what is known about the risk of surgery in patients with liver disease is based on retrospective studies of cirrhotic patients who underwent abdominal surgery (4–13). These early studies demonstrated that patients with severe liver disease are at high risk. For example, patients with well-compensated Child A cirrhosis have a perioperative mortality of 10% and up to 82% for Child C cirrhosis (8,14).
These numbers may not be applicable to most of the patients with HCV because these patients undergo a broader range of surgical procedures, and relatively few have clinical evidence of severe liver dysfunction. Assessing the degree of dysfunction in a patient with HCV can be difficult because the symptoms, physical examination, and laboratory tests are all relatively insensitive for determining the severity of liver disease. Liver biopsy is the most accurate means of determining the severity of liver disease, but it is not used as a routine preoperative test for these patients because the risk and expense cannot be justified in asymptomatic patients. Because patients with HCV typically proceed to surgery without histologic confirmation, there is a risk that a subset of these patients will have undiagnosed cirrhosis (15). Unsuspected cirrhosis by other causes has been associated with increased mortality in patients undergoing bariatric surgery (16). Even without cirrhosis, patients with chronic HCV may be at increased risk for complications because of other concomitant medical problems.
Unfortunately, determining the risk of surgery in HCV-infected patients is a problem that is not amenable to prospective controlled clinical trials, so a large epidemiologic study is required. The frequent prevalence of HCV in the veteran population makes this group suitable for this type of study. This combined with access to centralized databases that include demographic data and key health and medical history information make it possible to identify large numbers of patients who have been tested for HCV and underwent surgery. Determining if HCV infection has a measurable impact on postoperative morbidity and mortality is relevant not only to veterans but may be applicable to other patient populations as well. As the population ages, HCV is likely to become increasingly common in patients undergoing surgical procedures. If so, the Veterans Affairs Health Care System experience may help predict what will happen to the complication rate as more of these infected patients undergo surgery.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This cross-sectional study of postoperative complications was approved by the Stanford University IRB and the Research Administration of the Veterans Affairs Palo Alto Health Care System (VAPAHCS). All patients who had surgical procedures performed at the VAPAHCS during a 5-yr period (1995–1999) with known HCV status were included in this study. The VAPAHCS is a tertiary academic referral center in Northern California and is a teaching hospital for Stanford University. All surgeries were performed at the Palo Alto Division by surgical and anesthesiology staff with the assistance of trainees. The HCV status was determined from a database that tracked all anti-HCV tests performed at VAPAHCS since July, 1992, in the clinical laboratory using a second or third generation enzyme-linked immunosorbent assay (Abbott Lab, Abbott Park, IL).
The surgical database consisted of the date of surgery, patient’s age, type of surgical procedure, ASA physical status, type of anesthesia (general, epidural, or spinal), urgency of surgery (nonemergent or emergent), and duration of surgery in terms of minutes in the operating room (OR). Other data included patients’ preoperative hematocrit (HCT) and platelet (PLT) count, if these were obtained in the 30 days before surgery.
As part of the quality assurance process for surgery and anesthesia, all major postoperative complications were recorded in a computerized database. In addition to our policy to call every patient after outpatient surgery and to follow-up every surgery with a surgery clinic visit, postoperative complications were also tracked as part of the National Veterans Affairs Surgical Quality Improvement Program database. All inhospital mortality during the admission for surgery was automatically registered into this database. If death occurred after a patient was discharged from the hospital, the database was updated by the quality management nurse reviewer based on information obtained from our decedent affairs office. This office was informed about deaths directly or by the Department of Veterans Affairs when updating the Veterans Affairs Beneficiary Identification and Record Locator System to track payments to veteran’s dependents applying for burial benefits or surviving spouse pensions. This system is highly accurate and is comparable with the National Death Index (17).
Postoperative complications were identified and reviewed by a qualified registered nurse reviewer and entered into the database. Any complication or death that occurred within 30 days after surgery was considered a postoperative complication in the current study. Complications were then grouped into seven major categories: infection (e.g., systemic sepsis, urinary tract infection, or wound infection), respiratory (e.g., unplanned intubation, ventilator use for more than 48 h, or iatrogenic pneumothorax), bleeding (e.g., re-operation for bleeding or postoperative bleeding requiring transfusion), renal (e.g., acute renal failure or renal insufficiency), cardiac (e.g., cardiac arrest or acute myocardial infarction), neurologic (e.g., cerebral vascular accident or seizure), and miscellaneous. The complications were grouped in this manner to help determine whether there was an increase of a certain type of complications such as bleeding in the HCV positive group. Systemic sepsis was defined as fever and increased white blood cell count often accompanied by hypotension or bacteremia that was treated with broad-spectrum antibiotics. Urinary tract infection was identified by a temperature more than 38.0°C, urinary symptoms, and a culture with more than 105 colonies/mL of urine. Wound infection included either superficial or deep wound infections that occurred at an incision site with purulent drainage or a positive culture of fluid from the wound. Acute renal failure was defined as worsening of renal function after surgery requiring dialysis in a patient who did not require dialysis before surgery. Renal insufficiency was evidenced by an increase in creatinine of more than 2 mg/dL that did not lead to dialysis. Cardiac arrest was identified by asystole requiring cardiopulmonary resuscitation of any duration. Myocardial infarction was defined as a transmural myocardial infarction manifested by new Q-waves on the electrocardiogram or increases in the cardiac enzymes consistent with a myocardial infarction. Cerebral vascular accidents were identified by an embolic, thrombotic, or hemorrhagic stroke with motor, sensory, or cognitive dysfunction that persists for 24 h.
The following categorical factors were included in the analysis: anti-HCV status (positive or negative), urgency of surgery (emergent or nonemergent, with emergent procedures defined as those that are expected to prevent the immediate threat of loss of life or limb), type of anesthesia (epidural, spinal, or general anesthesia), operative risk as defined by ASA physical status (low, Class I–II; high, Class III–V). The continuous variables included in the analysis were age of patient, length of surgery, and preoperative HCT and PLT count.
The statistical analysis was performed using the SAS software package (Cary, NC). 
2 was performed using Statistical Package for Social Science for Windows 10.0 (SPSS Inc, Chicago, IL). Preliminary data inspection revealed a small proportion (approximately 25%) of missing data in HCT or PLT count because these were not required for all patients before surgery. For missing data, we used the statistical technique of multiple imputation, which assigns values for each missing observation so that the completed data sets could be analyzed. We assumed that the missing variables were distributed as multivariate normal and their data were missed at random. Software procedure SAS PROC MI was used to perform multiple imputations to generate complete data sets for subsequent multivariate analysis. For each of the imputed data sets, a logistic regression model was fitted. To complete the imputation process, procedure SAS PROC MIANALYSIS was used to summarize the results from the logistic regression models.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
During the 5-year study period, a total of 2457 male veterans with complete information on age, anti-HCV status, ASA class, type of anesthesia, urgency of the procedure, and length of surgery were included in the study. Halothane was not administered during any of the surgical procedures in this study. For patients who underwent multiple surgical procedures during the study period, only the first surgical procedure was included in the final analysis. Outcome of the first surgery for these 2457 patients with known anti-HCV status constituted the study group. In this cohort, 440 (17.9%) patients were anti-HCV positive. The patients who were anti-HCV positive were on average almost 10 yr younger and were less likely to be classified as ASA III or higher compared with the anti-HCV negative group (Table 1).
|
The mortality rate was less frequent in the anti-HCV positive group (0.7%) than in the anti-HCV negative group (2.5%) (P = 0.017; Table 1). The patients who died after surgery compared with those who did not were older (68 versus 60 yr old; P < 0.0001) and were more likely to be classified as ASA class III–V (99% versus 50%; P < 0.0001). This group also had longer procedures (mean time, 181 min) compared with 130 min for the entire group.
There were 305 complications, resulting in an overall complication rate of 12%. The complications rates were comparable between the two groups (Table 1). Bleeding and renal complications seemed more common in the anti-HCV positive group, but this did not reach statistical significance. The correlation of risk factors and complication rate controlled for anti-HCV status is summarized in Table 2.
|
Veterans with complications when compared with the group without complications were significantly (P < 0.0001) older (64 ± 12 yr versus 59 ± 12 yr), more often classified as ASA physical status III, IV, or V (74% versus 48%), required more OR time (191 ± 127 versus 121 ± 90 min), and had lower preoperative HCT (40% ± 6% versus 42% ± 5%) but no difference in the PLT count (226 ± 83 versus 233 ± 85; P = 0.229). The type of anesthesia was not a significant risk factor for complications. Therefore, the type of anesthesia was not included in the final logistic regression model.
A multivariate analysis was performed with logistic regression model using HCV antibody status, urgency of surgery, age, ASA class, OR time, and preoperative HCT and PLT count. HCT and PLT count were missing in 618 (25%) and 632 (26%) of the patients, respectively. Imputed data were used to replace missing values of HCT and PLT for the multivariate logistic regression analysis, as described previously. In the multivariate analysis, anti-HCV status and PLT did not have an impact on the complication rate (Table 3). Surgery lasting over 100 min was most predictive of postoperative complications with an odds ratio of 2.87 (95% confidence interval, 2.16–3.81; P < 0.0001).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Others have demonstrated that, in the absence of cirrhosis, elective surgery is relatively safe in asymptomatic patients with mild liver disease (18). In contrast, subclinical cirrhosis at the time of surgery has been associated with increased mortality (16). Among veterans with chronic HCV evaluated for treatment at our liver clinic, 40% have advanced fibrosis on liver biopsy (data not shown). Therefore, we postulated that a number of our anti-HCV positive patients would have unsuspected cirrhosis (15) and hence a frequent rate of complications. Instead, our study revealed a similar complication profile for patients, regardless of their HCV status. Because the risk associated with surgery in patients with liver disease correlates with the presence and severity of cirrhosis, our findings could potentially be explained by postulating an infrequent incidence of cirrhosis in the study group. Severe thrombocytopenia may be used as a surrogate marker for cirrhosis, and <3% of the patients in this study had platelet counts <100,000/mm3 before surgery; only 5% of these thrombocytopenic patients had emergency surgery. The smaller mortality rate for anti-HCV positive patients may also be attributable to the observation that these patients were younger, had lower ASA physical status classifications, and underwent shorter procedures than the anti-HCV negative patients.
As a cross-sectional study, there were a number of limitations to our study. We could not control who was tested for anti-HCV, but presumably the test was ordered for patients at high risk for HCV infection or those with known or suspected liver disease. We found an overall anti-HCV seroprevalence of 17.9% in this cohort, similar to the findings of HCV screening of unselected veterans in an outpatient clinic in northern California (2). This suggested that the study cohort was representative of our entire surgical population. Multiple studies have also demonstrated that most of these anti-HCV positive veterans are true positives (2,3). We intentionally limited our analysis to the outcome of the first surgery among patients who had multiple surgeries. This was to prevent introducing selection bias such as those from reoperation because of surgical complications or bias related to multiple surgeries on sicker patients. Finally, this study was limited to male veterans who were the large majority of our patient population.
Despite these limitations, our study identified the same risk factors that have been described by others (19,20). In the multivariate analysis, we found that postoperative risk is associated with high ASA physical status, emergent surgery, advanced age, prolonged surgery, and low HCT. The most important risk factor was the length of surgery. However, it is not clear whether prolonged surgery was a reflection of the complexity of surgery itself, attributable to the patient’s co-morbidities, or if it was the direct result of intraoperative complication(s). The HCT can be used as a surrogate marker of the general health of the patient, and in this study, a low preoperative HCT was associated with increased postoperative complications. There were too few patients with a low PLT count to assess this potential risk factor.
In contrast with our findings, a study of patients with hepatocellular carcinoma revealed increased postoperative complication and mortality rates after hepatic resection in patients with HCV compared with similar patients with hepatitis B (21). This might be because the mean age of patients with HCV in the study was significantly older and had more active hepatitis than those with hepatitis B. Another possible explanation of these contradictory findings is that the ASA physical status is a global assessment of risk independent of etiology. For patients with findings of advanced liver disease or cirrhosis from chronic hepatitis C, their increased risk would already be incorporated into a higher ASA physical status. Previous studies have found high ASA physical status to be the strongest predictor of postoperative outcome (19,20).
The current study is a good example of "Simpson’s paradox (22)." It demonstrates the dangers of ignoring covariates that are correlated to an end-point (e.g., complication) and a risk factor (e.g., HCV status). From this study, a simple tabulation (Table 1) showed that complication rates were 13% and 10% for the anti-HCV negative and positive group, respectively. After adjusting for other risk factors (or covariates), logistic regression analysis revealed that anti-HCV positivity was associated with an increased complication rate but did not achieve statistical significance.
Although we did not find anti-HCV status to be an independent risk factor for postoperative complications statistically, special attention should be paid to patients with chronic liver disease. As the group of patients who are HCV positive age and their liver disease progresses, the long-term effects of chronic liver disease could have a more profound effect on postoperative complications. At present, prolonged surgery, emergency surgery, and high ASA physical status are more predictive of postoperative risk than HCV infection.
|
Acknowledgments |
|---|
The authors wish to acknowledge Richard Mazze, MD, and Kevin Fish, MD, for their advice and review of the manuscript.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
- Alter MJ, Kruszon-Moran D, Nainan OV, et al. The prevalence of hepatitis C infection in the United States, 1988 through 1994. N Engl J Med 1999; 341: 556–62.
[Abstract/Free Full Text] - Briggs ME, Baker C, Hall R, et al. Prevalence and risk factors for hepatitis C virus infection at an urban Veterans Administration medical center. Hepatology 2001; 34: 1200–5.[Web of Science][Medline]
- Cheung RC. Epidemiology of hepatitis C virus infection in American veterans. Am J Gastroenterol 2000; 95: 740–7.[Web of Science][Medline]
- Doberneck RC, Sterling WA, Allison DC. Morbidity and mortality after operation in nonbleeding cirrhotic patients. Am J Surg 1983; 146: 306–9.[Web of Science][Medline]
- Bloch RS, Allaben RD, Walt AJ. Cholecystectomy in patients with cirrhosis: a surgical challenge. Arch Surg 1985; 120: 669–72.
[Abstract/Free Full Text] - Metcalf AMT, Dozois RR, Wolff BG, Beart RW. The surgical risk of colectomy in patients with cirrhosis. Dis Colon Rectum 1987; 30: 529–31.[Web of Science][Medline]
- Lehnert T, Herfarth C. Peptic ulcer surgery in patients with liver cirrhosis. Ann Surg 1993; 217: 338–46.[Web of Science][Medline]
- Mansour A, Watson W, Shayani V, Pickleman J. Abdominal operations in patients with cirrhosis: still a major surgical challenge. Surgery 1997; 122: 730–6.[Web of Science][Medline]
- Rice HE, O’Keefe GE, Helton WS, Johansen K. Morbid prognostic features in patients with chronic liver failure undergoing nonhepatic surgery. Arch Surg 1997; 132: 880–5.
[Abstract/Free Full Text] - Sleeman D, Namias N, Levi D, et al. Laparoscopic cholecystectomy in cirrhotic patients. J Am Coll Surg 1998; 187: 400–3.[Web of Science][Medline]
- Child CG, Turcotte JG. Surgery and portal hypertension. In: Child CG, ed. The liver and portal hypertension. Philadelphia: Saunders, 1964; 50–2.
- Pugh RN, Murray-Lyon IM, Dawson JL, et al. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 1973; 60: 646–9.[Web of Science][Medline]
- Ziser A, Plevak DJ, Wiesner RH, et al. Morbidity and mortality in cirrhotic patients undergoing anesthesia and surgery. Anesthesiology 1999; 90: 42–53.[Web of Science][Medline]
- Garrison RN, Cryer HM, Howard DA, Polk HC Jr. Clarification of risk factors for abdominal operations in patients with hepatic cirrhosis. Ann Surg 1984; 199: 648–55.[Web of Science][Medline]
- Poynard T, Bedossa P, Opolon P. Natural history of liver fibrosis progression in patients with chronic hepatitis C. Lancet 1997; 349: 825–32.[Web of Science][Medline]
- Brolin RE, Bradley LJ, Taliwal RV. Unsuspected cirrhosis discovered during elective obesity operations. Arch Surg 1998; 133: 84–8.
[Abstract/Free Full Text] - Fisher SG, Weber L, Goldberg J, Davis F. Mortality ascertainment in the veteran population: alternatives to the National Death Index. Am J Epidemiol 1995; 141: 242–50.
[Abstract/Free Full Text] - Runyon BA. Surgical procedures are well tolerated by patients with asymptomatic chronic hepatitis. J Clin Gastroenterol 1986; 8: 542–4.[Web of Science][Medline]
- Pedersen T, Eliasen K, Ravnborg M, et al. Risk factors, complications and outcome in anesthesia: a pilot study. Eur J Anaesthesiol 1986; 3: 225–39.[Web of Science][Medline]
- Wolters U, Wolf T, Stutzer H, Schroder T. ASA classification and perioperative variables as predictors of postoperative outcome. Br J Anaesth 1996; 77: 217–22.
[Abstract/Free Full Text] - Higashi H, Matsumata T, Adachi E, et al. Influence of viral hepatitis status on operative morbidity and mortality in patients with primary hepatocellular carcinoma. Br J Surg 1994; 81: 1342–5.[Web of Science][Medline]
- Simpson EH. The interpretation of interaction in contingency tables. J R Stat Soc [Ser B] 1951; 13: 238–41.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
