Study Design and Population
All 2224 incident PD patients who were followed up at four PD centers from November 1, 2005, to February 28, 2017, were enrolled at this multi-center retrospective cohort study. Inclusion criteria were age ≥18 years at the start of PD and survival for ≥three months from the first PD therapy. Previous studies reported that the increased AST/ALT ratio is due to induction by alcohol consumption and cardio-hepatic interaction [16, 17]. Patients with a history of cardiovascular disease or liver disease have high risk of CVD episodes. Thus, to minimize the effect of a history of CVD or liver disease on the association between AST/ALT ratio and CVD mortality, we excluded those with a history of CVD or liver disease. To further minimize selective bias, we excluded those with an AST/ALT ratio >2, who may have underlying liver disease. Thus, patients were excluded from the study if they had current drinking, had been diagnosed with a history of CVD, chronic liver disease, or AST or AST values more than two times higher normal values. The study is consistent with the ethical principles of the Declaration of Helsinki and was approved by the Human Ethics Committee of each research center. Written informed consent was obtained from all eligible patients.
Baseline demographic data included age, sex, Charlson comorbidity index (CCI), diabetes, hypertension, hyperlipidemia, gastrointestinal bleeding, current smoking, and medication use. Clinical and biochemical data at the initiation of PD included body mass index (BMI), ejection fraction, estimated glomerular filtration rate (eGFR), hemoglobin, serum albumin, AST, ALT, total bilirubin, cholesterol, triglycerides, high-density lipoprotein (HDL), low-density lipoprotein (LDL), high-sensitivity C-reactive protein (hs-CRP), N-terminal -prohormone BNP (NT-proBNP), and 24-hr urine output. All baseline data were obtained during the first month of PD.
The primary and secondary endpoints were CVD and all-cause mortality, respectively. The PD team consisted of two nephrologists at each center who reviewed the details of individual medical records and identified the causes of death. If death had two or more potential causes, we generally ascribed the death to the primary cause for hospitalization or the initial presenting condition. If a patient died within three months of transfer to hemodialysis therapy, he or she was not censored because the early mortality was considered to reflect health status during the period of failing PD treatment. All patients were followed up until cessation of PD, death, or May 31, 2017. The censored data included switching to hemodialysis (for more than three months), renal transplantation, moving to another center, loss to follow-up, or still at our PD centers with a follow-up duration of 5 years. All patients received continuous ambulatory PD treatment. Conventional PD solutions (Dianeal 1.5%, 2.5%, or 4.25% dextrose; Baxter Healthcare, Guangzhou, China), Y sets, and twin bag systems were used in all PD patients.
Definitions
CVD was defined as coronary events, arrhythmias, sudden cardiac death, congestive heart failure, or cerebrovascular events [18]. Chronic liver diseases are defined as alcoholic and non-alcoholic liver disease, autoimmune liver disease, hepatitis B or C [13, 19, 20]. We defined aminotransferase elevation as any value above normal of ALT or AST based on a recent, nationally representative the United States survey (AST >40 IU/L, or ALT >43 IU/L) [21]. The comorbidity score was determined according to the CCI, which is one of the most commonly used comorbidity models [22]. Baseline residual renal function was assessed by eGFR using the Chronic Kidney Disease Epidemiology Collaboration creatinine equation [23].
Statistical analysis
AST and ALT measured according to the standard measurements of Chinese. Data were expressed as mean ± standard deviation, percentages, or median (25th-75th percentile). All eligible patients were divided into high and normal groups according to the AST/ALT ratio cut-off for CVD mortality with the receiver operating characteristic (ROC) curve. Comparisons of baseline parameters between two groups were conducted with t-testsfor continuous normally distributed variables, Mann-Whitney test for continuous non-normally distributed data and χ2 analyses for categorical data. Logistic regression analyses were conducted to evaluate the association between baseline variables and high AST/ALT ratio. Variables clinically considered to be associated with high AST/ALT ratio were picked into a multivariate-adjusted Logistic regression model. Survival was estimated using the Kaplan-Meier curve, and differences were examined using the log-rank test. The associations between the AST/ALT ratio and CVD and all-cause mortality were evaluated by Cox proportional hazards regression. Unadjusted association was first examined, followed by adjustments for age, sex and CCI, current smoking, and medication use, including angiotensin-converting enzyme inhibitor/angiotensin receptor blocker (ACEI/ARB), calcium antagonist, β-blocker, diuretic and statin use. Next, BMI, eGFR, hemoglobin, albumin, total bilirubin, cholesterol, triglycerides, hs-CRP, and 24-hr urine output were added to examine whether the association of the AST/ALT ratio with endpoints was independent of confounding factors. The results of the Cox regression models were presented as the hazard ratio (HR) and the 95% confidence interval (CI). A value of P<0.05 was considered statistically significant. Statistical analyses were performed using GraphPad software 8.0 (GraphPad Prism Software Inc., San Diego, California) and the R package 3.6.0 (https://www.r-project.org/).