LIVER TRANSPLANTATION 20:1306–1316, 2014
ORIGINAL ARTICLE
High Early Cardiovascular Mortality After Liver Transplantation Lisa B. VanWagner,1,2,3 Brittany Lapin,3,4 Josh Levitsky,2,4 John T. Wilkins,1,5 Michael M. Abecassis,3,4 Anton I. Skaro,3,4 and Donald M. Lloyd-Jones1,5 1 Department of Preventive Medicine, 2Division of Gastroenterology and Hepatology, Department of Medicine, 3Northwestern University Transplant Outcomes Research Collaborative, 4Division of Organ Transplantation, Department of Surgery, and 5Division of Cardiology, Department of Medicine, Northwestern University Feinberg School of Medicine, Chicago, IL
Cardiovascular disease (CVD) contributes to excessive long-term mortality after liver transplantation (LT); however, little is known about early postoperative CVD mortality in the current era. In addition, there is no model for predicting early postoperative CVD mortality across centers. We analyzed adult recipients of primary LT in the Organ Procurement and Transplantation Network (OPTN) database between February 2002 and December 2012 to assess the prevalence and predictors of early (30-day) CVD mortality, which was defined as death from arrhythmia, heart failure, myocardial infarction, cardiac arrest, thromboembolism, and/or stroke. We performed logistic regression with stepwise selection to develop a predictive model of early CVD mortality. Sex and center volume were forced into the final model, which was validated with bootstrapping techniques. Among 54,697 LT recipients, there were 1576 deaths (2.9%) within 30 days. CVD death was the leading cause of 30-day mortality (40.2%), and it was followed by infection (27.9%) and graft failure (12.2%). In a multivariate analysis, 9 significant covariates (6 recipient covariates, 2 donor covariates, and 1 operative covariate) were identified: age, preoperative hospitalization, intensive care unit status, ventilator status, calculated Model for End-Stage Liver Disease score, portal vein thrombosis, national organ sharing, donor body mass index, and cold ischemia time. The model showed moderate discrimination (C statistic 5 0.66, 95% confidence interval 5 0.63-0.68). In conclusion, we provide the first multicenter prognostic model for the prediction of early post-LT CVD death, the most common cause of early post-LT mortality in the current transplant era. However, evaluations of additional CVD-related variables not collected by the OPTN are needed in C 2014 AASLD. order to improve the model’s accuracy and potential clinical utility. Liver Transpl 20:1306-1316, 2014. V Received March 10, 2014; accepted June 30, 2014. Cardiovascular disease (CVD), as defined by the American Heart Association,1 includes ischemic heart disease, stroke, heart failure, and thromboembolism and is a leading cause of long-term complications after liver
transplantation (LT).2 However, it is unknown what role CVD plays in early posttransplant mortality or whether certain features may predict early CVD mortality. In addition, little is known regarding the range
Additional Supporting Information may be found in the online version of this article. Abbreviations: BMI, body mass index; CI, confidence interval; CIT, cold ischemia time; CVD, cardiovascular disease; ICU, intensive care unit; LT, liver transplantation; MELD, Model for End-Stage Liver Disease; N/A, not applicable; NASH, nonalcoholic steatohepatitis; OPTN, Organ Procurement and Transplantation Network; OR, odds ratio; PVT, portal vein thrombosis. The data reported here have been supplied by the United Network for Organ Sharing as the contractor for the Organ Procurement and Transplantation Network. The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as an official policy of or interpretation by the Organ Procurement and Transplantation Network or the US Government. Potential conflict of interest: Nothing to report. Lisa B. VanWagner is supported by the National Institutes of Health (1 F32 HL116151-01) and the American Liver Foundation (New York, NY). Address reprint requests to Lisa B. VanWagner, M.D., M.S., Northwestern University Feinberg School of Medicine, 680 North Lakeshore Drive, Suite 1400, Chicago, IL 60611. Telephone: 630-229-3761; FAX: 312-695-3999; E-mail: [email protected] DOI 10.1002/lt.23950 View this article online at wileyonlinelibrary.com. LIVER TRANSPLANTATION.DOI 10.1002/lt. Published on behalf of the American Association for the Study of Liver Diseases
C 2014 American Association for the Study of Liver Diseases. V
LIVER TRANSPLANTATION, Vol. 20, No. 11, 2014
of CVD complications that may occur after LT apart from those related to ischemic heart disease. The specific cardiovascular and hemodynamic responses that occur in end-stage liver disease unrelated to traditional coronary risk factors may contribute to increased CVD complications after transplantation.3,4 Finally, there are liver donor and surgical characteristics, such as the donor age and the cold ischemia time (CIT), that are known to increase early mortality for recipients.5 Because of the limited availability of donor organs, recipient selection and appropriate monitoring for and prevention of early CVD mortality are of paramount importance.6 Historically, prevalent CVD has been considered a relative and, under certain circumstances, absolute contraindication to LT with estimated posttransplant mortality rates as high as 50%.7 Thus, this study was aimed at better defining early CVD-related mortality and aiding with recipient selection and recipient-donor matching in the most recent era of LT. Among the instruments for clinical risk stratification are decision tools such as the Revised Cardiac Risk Index,8 which uses 6 variables (history of ischemic heart disease, heart failure, stroke, insulin-dependent diabetes, chronic kidney disease, and high-risk surgery) to predict cardiac complications after noncardiac surgery. In addition, both noninvasive cardiac tests (dobutamine stress echocardiography) and invasive cardiac tests (coronary angiography) may be used preoperatively to assist with risk stratification. However, these risk algorithms and tests have poor discriminative ability to predict early cardiac mortality after LT.9,10 Current guidelines for preoperative cardiac evaluations before noncardiac surgery are based on studies of patient cohorts not undergoing LT, so the optimal preoperative cardiac evaluation for LT patients remains unknown.9,11,12 Although many studies have reported an increased incidence of longterm CVD complications after transplantation, few have focused on predictors of early CVD outcomes.13,14 In addition, despite the high prevalence of cirrhotic cardiomyopathy, studies of early CVD outcomes after LT have predominantly focused on complications of ischemic heart disease.7,14 Finally, attempts at describing LT-specific CVD risk predictors have been inconsistent in their findings, have been limited by single-center data with inherent variability in candidate selection, and often predate recent advances in surgical technique and anesthesia.2,14-16 Using a large multicenter national database, we hypothesized that there are unique characteristics of end-stage liver disease and LT-specific risk factors that are associated with excessive early CVD mortality. In the long term, the incorporation of these factors into validated risk algorithms may improve patient outcomes and organ utilization.
PATIENTS AND METHODS Study Population Adults (age 18 years) who were listed for LT between February 1, 2002 and December 31, 2012 and who
VANWAGNER ET AL. 1307
underwent transplantation within the same time period were identified from the Organ Procurement and Transplantation Network (OPTN) Standard Transplant Analysis and Research files (created on March 15, 2013; n 5 56,914). Those who were listed as status 1 and those who underwent retransplantation or received simultaneous heart, lung, intestine, or pancreas transplants were excluded (n 5 2217). The institutional review board of Northwestern University approved the study.
Definitions and Outcomes Each recipient’s cause of death was determined by a physician’s review (L.B.V.) of primary and contributory causes of death (including all free text inputs) listed in the OPTN database. Any potential case with a death due to CVD, which was defined as the primary cause of death from arrhythmia, heart failure, myocardial infarction, primary cardiac arrest, thromboembolism, and/or stroke, was then manually reviewed by an independent panel of 3 physicians (2 cardiologists and 1 surgeon) in order to attempt to adjudicate CVD case mortality. The primary study outcome was early (30day) CVD mortality. This standardized time period was chosen because of its use as an outcome-based quality indicator.17 The time period also allows a fair assessment of transplant outcomes across centers and minimizes the degree to which differences in variations in the length of the posttransplant stay affect the measurement. Because some operative factors such as electrolyte flux are not captured within the OPTN data set and may have a differential effect on cardiac events, analyses were also categorized as analyses of perioperative CVD mortality, which was defined as CVD mortality within the first 24 hours of transplantation, or early postoperative CVD mortality, which was defined as CVD mortality occurring between 1 and 30 days. Secondary outcomes included overall patient and graft survival. Patients were censored at the time of death, at the date of the last follow-up, at the time of retransplantation, or at 30 days. Potential risk factors for CVD-related mortality after LT were examined according to a priori clinical hypotheses. Covariates included known traditional CVD risk factors (eg, diabetes status) as well as transplantspecific critical illness indicators known to contribute to competing mortality risk. Evaluated recipient risk factors included the following: age at transplant, sex, race/ethnicity (black, non-Hispanic white, Asian, or Hispanic), socioeconomic status, body mass index (BMI), etiology of liver disease [including diseases known to increase the risk for CVD, such as nonalcoholic steatohepatitis (NASH) and hepatitis C], history of comorbid CVD conditions (diabetes, angina, cerebrovascular disease, hypertension, pulmonary embolism, peripheral vascular disease, and renal failure), laboratory values at the time of transplantation (creatinine, albumin, sodium, international normalized ratio, alanine aminotransferase, and bilirubin), hepatocellular carcinoma, calculated Model for End-Stage Liver Disease (MELD) score at the time of transplantation, wait-
1308 VANWAGNER ET AL.
list time, functional capacity before transplantation, complications of end-stage liver disease [eg, ascites, encephalopathy, and portal vein thrombosis (PVT)], hospitalization, and ventilator status at transplant. Donor risk factors included the following: age, sex, race, BMI, cause of death, donor type (living, deceased, or donation after cardiac death), donor risk index, procurement medications (inotropic support, vasopressin, antihypertensives, steroids, thyroid replacement, and desmopressin), history of CVD comorbidity (diabetes, hypertension, renal disease, and myocardial infarction), and health behaviors (smoking status and alcohol/cocaine/other drug use). Transplant-related variables included the transplant center location and volume, region, organ allocation type, CIT, steroid induction, and use of a calcineurin inhibitor.
Statistical Methods Clinical characteristics and causes of death of primary LT recipients from the 2002-2012 OPTN data set were described with frequency counts and percentages for categorical variables and with means and standard deviations for continuous variables. Logistic regression models were first fit for each variable separately to determine associations with 30-day CVD mortality after LT as the dependent variable. Odds ratios (ORs) with 95% confidence intervals (CIs) as well as their corresponding P values are shown. Twenty-two candidate variables that were significant in the univariate analysis were entered into a multivariate logistic regression model. Stepwise regression was performed with entry and exit criteria set to P 5 0.05. Nine covariates were selected for the final model on the basis of their significance and additive contribution to the model. Further covariates did not improve the model fit. It was determined a priori that center and recipient sex should be forced into the final model. The performance of the logistic regression model was then internally validated with 1000 bootstrap resamples. Bootstrapping is a nonparametric method for assigning measures of accuracy to sample estimates. Bootstrapping essentially resamples (multiple times) from the study population to approximate how precise statistical estimates (eg, C statistics and CIs) are related to the true population of interest. A Kaplan-Meier analysis with a log-rank test was used to assess the time to CVD mortality and all-cause graft and patient survival. All analyses were performed with SAS 9.3 (SAS Institute, Cary, NC).
LIVER TRANSPLANTATION, November 2014
TABLE 1. Clinical Characteristics of First Orthotopic LT Recipients in the OPTN Database, 2002-2012 (n 5 54,697) Characteristic Age (years)* Sex: female [n (%)] Race/ethnicity [n (%)] Non-Hispanic white Black Hispanic Asian Other Socioeconomic status [n (%)] Less than high school education Working for income at time of transplant Etiology of liver disease [n 5 46,375] Hepatitis C Alcohol NASH Other Calculated MELD score at transplant* Wait-list time (days)* Hepatocellular carcinoma [n (%)] Simultaneous liver-kidney transplant [n (%)] BMI at transplant (kg/m2)* Pretransplant comorbid CVD conditions [n (%)] Angina [n 5 23,069] Cerebrovascular disease Diabetes Hypertension Functional status at transplant [n 5 49,168] Independent Partially dependent Totally dependent Hospitalization status at transplant [n 5 54,611] Not hospitalized Hospitalized, not in ICU In ICU Mean Revised Cardiac Risk Index
Value 54.0 6 9.5 17,198 (31.4) 39,747 (72.7) 4820 (8.8) 7165 (13.1) 2435 (4.5) 530 (1.0) 2405 (4.4) 11,463 (21.0)
15,529 (28.4) 11,220 (20.5) 3085 (5.6) 16,545 (30.2) 19.0 6 10.4 273.6 6 493.5 12,694 (23.2) 3206 (5.9) 28.2 6 5.6
790 (3.4) 137 (0.3) 13,902 (25.4) 4399 (8.0)
24,411 (44.6) 12,229 (22.4) 12,528 (22.9)
41,090 (75.1) 9032 (16.5) 4489 (8.2) 1.8 6 1.2
*The data are presented as means and standard deviations. The data for socioeconomonic status, etiology of liver disease, functional status and hospitalization status and angina do not equal 54,697 due to missing data within OPTN.
RESULTS Patient Characteristics The baseline characteristics of 54,697 orthotopic LT recipients who were included in the final analysis are shown in Table 1. The mean age of the study sample was 54.0 6 9.5 years, 31.4% were female, and 72.7% were non-Hispanic white. The all-cause early mortality rate was 2.9% (n 5 1576). CVD accounted for 40.2% of all deaths within 30 days, and it was fol-
lowed by infection (27.9%) and graft failure (12.2%; Fig. 1). The mean time to an early CVD death was 6.2 6 8.2 days, and the median was 2.0 days (range 5 0-30 days). The leading underlying cause of early CVD mortality was cardiac arrest (47.9%), and this was followed by stroke (12.5%), heart failure (12.3%), and pulmonary embolism (9.1%). The prevalence of cardiac arrest as an underlying cause of CVD mortality did not differ significantly between those
LIVER TRANSPLANTATION, Vol. 20, No. 11, 2014
Figure 1.
VANWAGNER ET AL. 1309
Distribution of causes of death for 1576 adult first LT recipients who died within 30 days of LT (OPTN data, 2002-2012).
TABLE 2. Secondary Causes of Death for Patients Who Died of Cardiac Arrest Within 30 Days of LT Secondary Cause of CVD-Related Mortality Infection/sepsis Graft failure Hemorrhage Operative Renal failure Unknown/none Other*
All Early CVD
All Cardiac
Perioperative Cardiac Arrests
Postoperative Cardiac Arrests
Deaths (n 5 633)
Arrests (n 5 303)
(n 5 140 or 46.2%)
(n 5 163 or 53.7%)
46 (7.3) 30 (4.7) 32 (5.1) 14 (2.2) 18 (2.8) 475 (75.0) 18 (2.8)
20 (6.6) 16 (5.3) 19 (6.3) 7 (2.3) 11 (3.6) 225 (74.3) 5 (1.7)
3 (2.1) 4 (2.9) 9 (6.4) 5 (3.6) 3 (2.1) 113 (80.7) 3 (2.1)
17 (10.4) 12 (7.4) 10 (6.1) 2 (1.2) 8 (4.9) 112 (68.7) 2 (1.2)
NOTE: The data are presented as numbers and percentages. *Other includes malignancies (n 5 1), suicide/trauma (n 5 0), drugs/toxins (n 5 0), and other (n 5 4) for cardiac arrest deaths and malignancies (n 5 1), suicide/trauma (n 5 1), drugs/toxins (n 5 0), and other (n 5 16) for CVD deaths.
deaths that occurred perioperatively and those that occurred in the early postoperative period (46.2% versus 53.7%, P 5 0.08). Because a significant proportion of the events were coded as cardiac arrest, we also examined secondary causes of death listed in the OPTN files that may have significantly contributed to the cardiac arrest (Table 2). There were no significant differences in the secondary causes of death between those recipients with cardiac arrest and those recipients with other causes of early CVD mortality among the 25.1% (n 5 159) of recipients who had a secondary cause of death listed within OPTN.
Predictors of Early CVD Mortality Univariate predictors of early CVD mortality are shown in Table 3. Recipients who died from CVD
within 30 days of LT were slightly older (55.3 6 9.7 versus 54.0 6 9.5 years) and had higher calculated MELD scores (22.8 versus 19.0) and a higher prevalence of medical comorbidities (eg, renal and pulmonary disease) than those without early CVD mortality (P < 0.05 for all). There were no statistically significant differences in recipient race, ethnicity, sex, or pretransplant cerebrovascular disease between those recipients with either perioperative or early postoperative CVD mortality and those recipients without early CVD mortality. However, an older age, the presence of NASH (versus hepatitis C), and pretransplant diabetes, hypertension, and chronic obstructive pulmonary disease were all more prevalent in recipients with perioperative CVD mortality versus those without early CVD mortality (P < 0.05 for all). Donor factors related to early perioperative CVD mortality included a higher
174 (74.0) 21 (8.9) 28 (11.9) 12 (5.1) 0 (0.0) 9 (3.8) 28 (11.9) 47 (20) 42 (17.9) 20 (8.5) 93 (39.6) 22.3 6 11.9 342.7 6 637.8 157 (66.8) 54 (23.0) 24 (10.2) 48 (20.4) 0 (0.0) 28.9 6 5.8 1.9 6 1.4 136.0 6 5.8
197 (83.8) 15 (6.4) 177 (75.3) 34 (14.5) 36 (15.3) 11 (4.7)
10 (4.3)
455 (71.9) 53 (8.4) 86 (13.6) 34 (5.4) 5 (0.8) 36 (5.7) 72 (11.4) 154 (24.3) 128 (20.2) 39 (6.2) 233 (36.8) 22.8 6 11.5 288.5 6 548.8 456 (72.0) 127 (20.1) 50 (7.9) 110 (17.4) 26 (4.1) 28.5 6 5.8 1.9 6 1.5 136.3 6 5.8
530 (83.7) 34 (5.4) 485 (76.6) 78 (12.3) 87 (13.7) 32 (5.1)
22 (3.5)
(n 5 235) 56.2 6 9.7 87 (37.0)
(n 5 633) 55.3 6 9.7 221 (34.9)
(n 5 54,064)
Characteristic
54.0 6 9.5 Recipient age (years)† Recipient sex: female [n (%)] 16,977 (31.4) Recipient race/ethnicity [n (%)] Non-Hispanic white 39,292 (72.7) Black 4767 (8.8) Hispanic 7079 (13.1) Asian 2401 (4.4) Other 525 (1.0) Recipient socioeconomic status [n (%)] Less than high school education 2369 (4.4) Working for income 11,391 (26.1) Etiology of liver disease [n (%)] Hepatitis C 15,375 (28.4) Alcohol 11,092 (20.5) NASH 3046 (5.6) Other 16,312 (30.2) 19.0 6 10.4 Calculated MELD score at transplant† Wait-list time (days)† 273.4 6 492.8 Organ allocation type [n (%)] Local 41,223 (76.2) Regional 10,001 (18.5) National 2840 (5.3) Hepatocellular carcinoma [n (%)] 12,584 (23.3) Simultaneous liver-kidney 3180 (5.9) transplant [n (%)] Recipient BMI at transplant (kg/m2)† 28.2 6 5.6 Laboratory values at transplant† Creatinine (mg/dL) 1.5 6 1.3 Sodium (mEq/L) 135.9 6 5.1 Recipient complications of end-stage liver disease [n (%)] Ascites at transplant 42,289 (78.2) Spontaneous bacterial peritonitis 1779 (3.3) Encephalopathy 36,219 (67.0) PVT 4117 (7.6) Transjugular intrahepatic 5576 (10.3) portosystemic shunt Variceal bleed 1300 (2.4) Recipient comorbid CVD conditions [n (%)] Angina 768 (1.4)
No CVD Any Early Perioperative Mortality CVD Mortality CVD Mortality OR (95% CI)*
ORIGINAL ARTICLE
High Early Cardiovascular Mortality After Liver Transplantation Lisa B. VanWagner,1,2,3 Brittany Lapin,3,4 Josh Levitsky,2,4 John T. Wilkins,1,5 Michael M. Abecassis,3,4 Anton I. Skaro,3,4 and Donald M. Lloyd-Jones1,5 1 Department of Preventive Medicine, 2Division of Gastroenterology and Hepatology, Department of Medicine, 3Northwestern University Transplant Outcomes Research Collaborative, 4Division of Organ Transplantation, Department of Surgery, and 5Division of Cardiology, Department of Medicine, Northwestern University Feinberg School of Medicine, Chicago, IL
Cardiovascular disease (CVD) contributes to excessive long-term mortality after liver transplantation (LT); however, little is known about early postoperative CVD mortality in the current era. In addition, there is no model for predicting early postoperative CVD mortality across centers. We analyzed adult recipients of primary LT in the Organ Procurement and Transplantation Network (OPTN) database between February 2002 and December 2012 to assess the prevalence and predictors of early (30-day) CVD mortality, which was defined as death from arrhythmia, heart failure, myocardial infarction, cardiac arrest, thromboembolism, and/or stroke. We performed logistic regression with stepwise selection to develop a predictive model of early CVD mortality. Sex and center volume were forced into the final model, which was validated with bootstrapping techniques. Among 54,697 LT recipients, there were 1576 deaths (2.9%) within 30 days. CVD death was the leading cause of 30-day mortality (40.2%), and it was followed by infection (27.9%) and graft failure (12.2%). In a multivariate analysis, 9 significant covariates (6 recipient covariates, 2 donor covariates, and 1 operative covariate) were identified: age, preoperative hospitalization, intensive care unit status, ventilator status, calculated Model for End-Stage Liver Disease score, portal vein thrombosis, national organ sharing, donor body mass index, and cold ischemia time. The model showed moderate discrimination (C statistic 5 0.66, 95% confidence interval 5 0.63-0.68). In conclusion, we provide the first multicenter prognostic model for the prediction of early post-LT CVD death, the most common cause of early post-LT mortality in the current transplant era. However, evaluations of additional CVD-related variables not collected by the OPTN are needed in C 2014 AASLD. order to improve the model’s accuracy and potential clinical utility. Liver Transpl 20:1306-1316, 2014. V Received March 10, 2014; accepted June 30, 2014. Cardiovascular disease (CVD), as defined by the American Heart Association,1 includes ischemic heart disease, stroke, heart failure, and thromboembolism and is a leading cause of long-term complications after liver
transplantation (LT).2 However, it is unknown what role CVD plays in early posttransplant mortality or whether certain features may predict early CVD mortality. In addition, little is known regarding the range
Additional Supporting Information may be found in the online version of this article. Abbreviations: BMI, body mass index; CI, confidence interval; CIT, cold ischemia time; CVD, cardiovascular disease; ICU, intensive care unit; LT, liver transplantation; MELD, Model for End-Stage Liver Disease; N/A, not applicable; NASH, nonalcoholic steatohepatitis; OPTN, Organ Procurement and Transplantation Network; OR, odds ratio; PVT, portal vein thrombosis. The data reported here have been supplied by the United Network for Organ Sharing as the contractor for the Organ Procurement and Transplantation Network. The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as an official policy of or interpretation by the Organ Procurement and Transplantation Network or the US Government. Potential conflict of interest: Nothing to report. Lisa B. VanWagner is supported by the National Institutes of Health (1 F32 HL116151-01) and the American Liver Foundation (New York, NY). Address reprint requests to Lisa B. VanWagner, M.D., M.S., Northwestern University Feinberg School of Medicine, 680 North Lakeshore Drive, Suite 1400, Chicago, IL 60611. Telephone: 630-229-3761; FAX: 312-695-3999; E-mail: [email protected] DOI 10.1002/lt.23950 View this article online at wileyonlinelibrary.com. LIVER TRANSPLANTATION.DOI 10.1002/lt. Published on behalf of the American Association for the Study of Liver Diseases
C 2014 American Association for the Study of Liver Diseases. V
LIVER TRANSPLANTATION, Vol. 20, No. 11, 2014
of CVD complications that may occur after LT apart from those related to ischemic heart disease. The specific cardiovascular and hemodynamic responses that occur in end-stage liver disease unrelated to traditional coronary risk factors may contribute to increased CVD complications after transplantation.3,4 Finally, there are liver donor and surgical characteristics, such as the donor age and the cold ischemia time (CIT), that are known to increase early mortality for recipients.5 Because of the limited availability of donor organs, recipient selection and appropriate monitoring for and prevention of early CVD mortality are of paramount importance.6 Historically, prevalent CVD has been considered a relative and, under certain circumstances, absolute contraindication to LT with estimated posttransplant mortality rates as high as 50%.7 Thus, this study was aimed at better defining early CVD-related mortality and aiding with recipient selection and recipient-donor matching in the most recent era of LT. Among the instruments for clinical risk stratification are decision tools such as the Revised Cardiac Risk Index,8 which uses 6 variables (history of ischemic heart disease, heart failure, stroke, insulin-dependent diabetes, chronic kidney disease, and high-risk surgery) to predict cardiac complications after noncardiac surgery. In addition, both noninvasive cardiac tests (dobutamine stress echocardiography) and invasive cardiac tests (coronary angiography) may be used preoperatively to assist with risk stratification. However, these risk algorithms and tests have poor discriminative ability to predict early cardiac mortality after LT.9,10 Current guidelines for preoperative cardiac evaluations before noncardiac surgery are based on studies of patient cohorts not undergoing LT, so the optimal preoperative cardiac evaluation for LT patients remains unknown.9,11,12 Although many studies have reported an increased incidence of longterm CVD complications after transplantation, few have focused on predictors of early CVD outcomes.13,14 In addition, despite the high prevalence of cirrhotic cardiomyopathy, studies of early CVD outcomes after LT have predominantly focused on complications of ischemic heart disease.7,14 Finally, attempts at describing LT-specific CVD risk predictors have been inconsistent in their findings, have been limited by single-center data with inherent variability in candidate selection, and often predate recent advances in surgical technique and anesthesia.2,14-16 Using a large multicenter national database, we hypothesized that there are unique characteristics of end-stage liver disease and LT-specific risk factors that are associated with excessive early CVD mortality. In the long term, the incorporation of these factors into validated risk algorithms may improve patient outcomes and organ utilization.
PATIENTS AND METHODS Study Population Adults (age 18 years) who were listed for LT between February 1, 2002 and December 31, 2012 and who
VANWAGNER ET AL. 1307
underwent transplantation within the same time period were identified from the Organ Procurement and Transplantation Network (OPTN) Standard Transplant Analysis and Research files (created on March 15, 2013; n 5 56,914). Those who were listed as status 1 and those who underwent retransplantation or received simultaneous heart, lung, intestine, or pancreas transplants were excluded (n 5 2217). The institutional review board of Northwestern University approved the study.
Definitions and Outcomes Each recipient’s cause of death was determined by a physician’s review (L.B.V.) of primary and contributory causes of death (including all free text inputs) listed in the OPTN database. Any potential case with a death due to CVD, which was defined as the primary cause of death from arrhythmia, heart failure, myocardial infarction, primary cardiac arrest, thromboembolism, and/or stroke, was then manually reviewed by an independent panel of 3 physicians (2 cardiologists and 1 surgeon) in order to attempt to adjudicate CVD case mortality. The primary study outcome was early (30day) CVD mortality. This standardized time period was chosen because of its use as an outcome-based quality indicator.17 The time period also allows a fair assessment of transplant outcomes across centers and minimizes the degree to which differences in variations in the length of the posttransplant stay affect the measurement. Because some operative factors such as electrolyte flux are not captured within the OPTN data set and may have a differential effect on cardiac events, analyses were also categorized as analyses of perioperative CVD mortality, which was defined as CVD mortality within the first 24 hours of transplantation, or early postoperative CVD mortality, which was defined as CVD mortality occurring between 1 and 30 days. Secondary outcomes included overall patient and graft survival. Patients were censored at the time of death, at the date of the last follow-up, at the time of retransplantation, or at 30 days. Potential risk factors for CVD-related mortality after LT were examined according to a priori clinical hypotheses. Covariates included known traditional CVD risk factors (eg, diabetes status) as well as transplantspecific critical illness indicators known to contribute to competing mortality risk. Evaluated recipient risk factors included the following: age at transplant, sex, race/ethnicity (black, non-Hispanic white, Asian, or Hispanic), socioeconomic status, body mass index (BMI), etiology of liver disease [including diseases known to increase the risk for CVD, such as nonalcoholic steatohepatitis (NASH) and hepatitis C], history of comorbid CVD conditions (diabetes, angina, cerebrovascular disease, hypertension, pulmonary embolism, peripheral vascular disease, and renal failure), laboratory values at the time of transplantation (creatinine, albumin, sodium, international normalized ratio, alanine aminotransferase, and bilirubin), hepatocellular carcinoma, calculated Model for End-Stage Liver Disease (MELD) score at the time of transplantation, wait-
1308 VANWAGNER ET AL.
list time, functional capacity before transplantation, complications of end-stage liver disease [eg, ascites, encephalopathy, and portal vein thrombosis (PVT)], hospitalization, and ventilator status at transplant. Donor risk factors included the following: age, sex, race, BMI, cause of death, donor type (living, deceased, or donation after cardiac death), donor risk index, procurement medications (inotropic support, vasopressin, antihypertensives, steroids, thyroid replacement, and desmopressin), history of CVD comorbidity (diabetes, hypertension, renal disease, and myocardial infarction), and health behaviors (smoking status and alcohol/cocaine/other drug use). Transplant-related variables included the transplant center location and volume, region, organ allocation type, CIT, steroid induction, and use of a calcineurin inhibitor.
Statistical Methods Clinical characteristics and causes of death of primary LT recipients from the 2002-2012 OPTN data set were described with frequency counts and percentages for categorical variables and with means and standard deviations for continuous variables. Logistic regression models were first fit for each variable separately to determine associations with 30-day CVD mortality after LT as the dependent variable. Odds ratios (ORs) with 95% confidence intervals (CIs) as well as their corresponding P values are shown. Twenty-two candidate variables that were significant in the univariate analysis were entered into a multivariate logistic regression model. Stepwise regression was performed with entry and exit criteria set to P 5 0.05. Nine covariates were selected for the final model on the basis of their significance and additive contribution to the model. Further covariates did not improve the model fit. It was determined a priori that center and recipient sex should be forced into the final model. The performance of the logistic regression model was then internally validated with 1000 bootstrap resamples. Bootstrapping is a nonparametric method for assigning measures of accuracy to sample estimates. Bootstrapping essentially resamples (multiple times) from the study population to approximate how precise statistical estimates (eg, C statistics and CIs) are related to the true population of interest. A Kaplan-Meier analysis with a log-rank test was used to assess the time to CVD mortality and all-cause graft and patient survival. All analyses were performed with SAS 9.3 (SAS Institute, Cary, NC).
LIVER TRANSPLANTATION, November 2014
TABLE 1. Clinical Characteristics of First Orthotopic LT Recipients in the OPTN Database, 2002-2012 (n 5 54,697) Characteristic Age (years)* Sex: female [n (%)] Race/ethnicity [n (%)] Non-Hispanic white Black Hispanic Asian Other Socioeconomic status [n (%)] Less than high school education Working for income at time of transplant Etiology of liver disease [n 5 46,375] Hepatitis C Alcohol NASH Other Calculated MELD score at transplant* Wait-list time (days)* Hepatocellular carcinoma [n (%)] Simultaneous liver-kidney transplant [n (%)] BMI at transplant (kg/m2)* Pretransplant comorbid CVD conditions [n (%)] Angina [n 5 23,069] Cerebrovascular disease Diabetes Hypertension Functional status at transplant [n 5 49,168] Independent Partially dependent Totally dependent Hospitalization status at transplant [n 5 54,611] Not hospitalized Hospitalized, not in ICU In ICU Mean Revised Cardiac Risk Index
Value 54.0 6 9.5 17,198 (31.4) 39,747 (72.7) 4820 (8.8) 7165 (13.1) 2435 (4.5) 530 (1.0) 2405 (4.4) 11,463 (21.0)
15,529 (28.4) 11,220 (20.5) 3085 (5.6) 16,545 (30.2) 19.0 6 10.4 273.6 6 493.5 12,694 (23.2) 3206 (5.9) 28.2 6 5.6
790 (3.4) 137 (0.3) 13,902 (25.4) 4399 (8.0)
24,411 (44.6) 12,229 (22.4) 12,528 (22.9)
41,090 (75.1) 9032 (16.5) 4489 (8.2) 1.8 6 1.2
*The data are presented as means and standard deviations. The data for socioeconomonic status, etiology of liver disease, functional status and hospitalization status and angina do not equal 54,697 due to missing data within OPTN.
RESULTS Patient Characteristics The baseline characteristics of 54,697 orthotopic LT recipients who were included in the final analysis are shown in Table 1. The mean age of the study sample was 54.0 6 9.5 years, 31.4% were female, and 72.7% were non-Hispanic white. The all-cause early mortality rate was 2.9% (n 5 1576). CVD accounted for 40.2% of all deaths within 30 days, and it was fol-
lowed by infection (27.9%) and graft failure (12.2%; Fig. 1). The mean time to an early CVD death was 6.2 6 8.2 days, and the median was 2.0 days (range 5 0-30 days). The leading underlying cause of early CVD mortality was cardiac arrest (47.9%), and this was followed by stroke (12.5%), heart failure (12.3%), and pulmonary embolism (9.1%). The prevalence of cardiac arrest as an underlying cause of CVD mortality did not differ significantly between those
LIVER TRANSPLANTATION, Vol. 20, No. 11, 2014
Figure 1.
VANWAGNER ET AL. 1309
Distribution of causes of death for 1576 adult first LT recipients who died within 30 days of LT (OPTN data, 2002-2012).
TABLE 2. Secondary Causes of Death for Patients Who Died of Cardiac Arrest Within 30 Days of LT Secondary Cause of CVD-Related Mortality Infection/sepsis Graft failure Hemorrhage Operative Renal failure Unknown/none Other*
All Early CVD
All Cardiac
Perioperative Cardiac Arrests
Postoperative Cardiac Arrests
Deaths (n 5 633)
Arrests (n 5 303)
(n 5 140 or 46.2%)
(n 5 163 or 53.7%)
46 (7.3) 30 (4.7) 32 (5.1) 14 (2.2) 18 (2.8) 475 (75.0) 18 (2.8)
20 (6.6) 16 (5.3) 19 (6.3) 7 (2.3) 11 (3.6) 225 (74.3) 5 (1.7)
3 (2.1) 4 (2.9) 9 (6.4) 5 (3.6) 3 (2.1) 113 (80.7) 3 (2.1)
17 (10.4) 12 (7.4) 10 (6.1) 2 (1.2) 8 (4.9) 112 (68.7) 2 (1.2)
NOTE: The data are presented as numbers and percentages. *Other includes malignancies (n 5 1), suicide/trauma (n 5 0), drugs/toxins (n 5 0), and other (n 5 4) for cardiac arrest deaths and malignancies (n 5 1), suicide/trauma (n 5 1), drugs/toxins (n 5 0), and other (n 5 16) for CVD deaths.
deaths that occurred perioperatively and those that occurred in the early postoperative period (46.2% versus 53.7%, P 5 0.08). Because a significant proportion of the events were coded as cardiac arrest, we also examined secondary causes of death listed in the OPTN files that may have significantly contributed to the cardiac arrest (Table 2). There were no significant differences in the secondary causes of death between those recipients with cardiac arrest and those recipients with other causes of early CVD mortality among the 25.1% (n 5 159) of recipients who had a secondary cause of death listed within OPTN.
Predictors of Early CVD Mortality Univariate predictors of early CVD mortality are shown in Table 3. Recipients who died from CVD
within 30 days of LT were slightly older (55.3 6 9.7 versus 54.0 6 9.5 years) and had higher calculated MELD scores (22.8 versus 19.0) and a higher prevalence of medical comorbidities (eg, renal and pulmonary disease) than those without early CVD mortality (P < 0.05 for all). There were no statistically significant differences in recipient race, ethnicity, sex, or pretransplant cerebrovascular disease between those recipients with either perioperative or early postoperative CVD mortality and those recipients without early CVD mortality. However, an older age, the presence of NASH (versus hepatitis C), and pretransplant diabetes, hypertension, and chronic obstructive pulmonary disease were all more prevalent in recipients with perioperative CVD mortality versus those without early CVD mortality (P < 0.05 for all). Donor factors related to early perioperative CVD mortality included a higher
174 (74.0) 21 (8.9) 28 (11.9) 12 (5.1) 0 (0.0) 9 (3.8) 28 (11.9) 47 (20) 42 (17.9) 20 (8.5) 93 (39.6) 22.3 6 11.9 342.7 6 637.8 157 (66.8) 54 (23.0) 24 (10.2) 48 (20.4) 0 (0.0) 28.9 6 5.8 1.9 6 1.4 136.0 6 5.8
197 (83.8) 15 (6.4) 177 (75.3) 34 (14.5) 36 (15.3) 11 (4.7)
10 (4.3)
455 (71.9) 53 (8.4) 86 (13.6) 34 (5.4) 5 (0.8) 36 (5.7) 72 (11.4) 154 (24.3) 128 (20.2) 39 (6.2) 233 (36.8) 22.8 6 11.5 288.5 6 548.8 456 (72.0) 127 (20.1) 50 (7.9) 110 (17.4) 26 (4.1) 28.5 6 5.8 1.9 6 1.5 136.3 6 5.8
530 (83.7) 34 (5.4) 485 (76.6) 78 (12.3) 87 (13.7) 32 (5.1)
22 (3.5)
(n 5 235) 56.2 6 9.7 87 (37.0)
(n 5 633) 55.3 6 9.7 221 (34.9)
(n 5 54,064)
Characteristic
54.0 6 9.5 Recipient age (years)† Recipient sex: female [n (%)] 16,977 (31.4) Recipient race/ethnicity [n (%)] Non-Hispanic white 39,292 (72.7) Black 4767 (8.8) Hispanic 7079 (13.1) Asian 2401 (4.4) Other 525 (1.0) Recipient socioeconomic status [n (%)] Less than high school education 2369 (4.4) Working for income 11,391 (26.1) Etiology of liver disease [n (%)] Hepatitis C 15,375 (28.4) Alcohol 11,092 (20.5) NASH 3046 (5.6) Other 16,312 (30.2) 19.0 6 10.4 Calculated MELD score at transplant† Wait-list time (days)† 273.4 6 492.8 Organ allocation type [n (%)] Local 41,223 (76.2) Regional 10,001 (18.5) National 2840 (5.3) Hepatocellular carcinoma [n (%)] 12,584 (23.3) Simultaneous liver-kidney 3180 (5.9) transplant [n (%)] Recipient BMI at transplant (kg/m2)† 28.2 6 5.6 Laboratory values at transplant† Creatinine (mg/dL) 1.5 6 1.3 Sodium (mEq/L) 135.9 6 5.1 Recipient complications of end-stage liver disease [n (%)] Ascites at transplant 42,289 (78.2) Spontaneous bacterial peritonitis 1779 (3.3) Encephalopathy 36,219 (67.0) PVT 4117 (7.6) Transjugular intrahepatic 5576 (10.3) portosystemic shunt Variceal bleed 1300 (2.4) Recipient comorbid CVD conditions [n (%)] Angina 768 (1.4)
No CVD Any Early Perioperative Mortality CVD Mortality CVD Mortality OR (95% CI)*
