Association of Indoxyl Sulfate with Heart Failure among Patients on Hemodialysis

Study Population

Patients who had been on hemodialysis treatment at least 6 months were enrolled from the Blood Purification Center, Zhongshan Hospital, Fudan University. The enrollment was completed within 6 months from July to December of 2009. Patients with congestive heart failure, angina pectoris, acute myocardial infarction, cerebral infarction, and cerebral hemorrhage within 3 months before the study or those <18 years old were excluded from this study. The study adhered to the Declaration of Helsinki and was approved by the Ethical Committee, Zhongshan Hospital, Fudan University. All participants provided written informed consent.

Patients were treated three times per week (4 hours per session) with standard bicarbonate dialysate (Na⁺: 138.0 mmol/L, HCO₃⁻: 32.0 mmol/L, K⁺: 2.0 mmol/L, Ca²⁺: 1.2 5 mmol/L, Mg²⁺: 0.5 mmol/L) by low-flux hemodialysis using 1.4-m² dialyzers with synthetic membranes (BLS514SD; Sorin Group Italia, Mirandola, Italy and Polyflux 14L; Gambro Dialysatoren GmbH, Hechigen, Germany). The blood flow was 200–280 ml/min, and dialysate flow was 500 ml/min. The water quality conformed to the Association for the Advancement of Medical Instrumentation standard and was examined every month. Dry weight was targeted in every patient to achieve an edema-free state. During the study, dry weight was reevaluated every month to guarantee a proper dry weight in every patient. In our center, all patients on hemodialysis were advised to have a high-protein diet (at least 1.2 g/kg per day with mainly animal protein).

Anthropometric Measurements, Blood Sampling, and Clinical Data Collection

Height and weight were measured while patients were barefoot and wearing light clothes only. Blood sampling was performed on a midweek nondialysis day (Thursday for patients scheduled with a Monday/Wednesday/Friday dialysis cycle and Friday for patients scheduled with a Tuesday/Thursday/Saturday dialysis cycle) from 8:00 to 10:00 a.m. Demographic and clinical data were collected, including age, sex, smoking history, underlying kidney disease, dialysis duration, comorbidity, and history of taking medicine. BP, average interdialytic weight gain, and ultrafiltration volume were obtained by averaging all predialysis BP, interdialytic weight gain, and ultrafiltration volume values, respectively, during 4 weeks (12 times in total) before this study. Urinary volume was measured, and residual renal function (RRF) was estimated by calculating GFR expressed as a mean of urea and creatinine clearance (milliliters per minute per 1.73 m²) (14). Normalized protein nitrogen appearance rate (nPNA) was calculated according to the work by Depner and Daugirdas (15).

Biochemical Measurements

Serum albumin, prealbumin, hemoglobin, BUN, serum creatinine (SCr), uric acid (UA), calcium (Ca), phosphorus (P), lipids, homocysteine (Hcy), iron, ferritin, and transferrin were measured using standard methods in the clinical laboratory. The concentrations of high-sensitivity C-reactive protein (hsCRP) and β₂-microglobulin (β₂M) were determined using immunoturbidimetry assay, and concentration of intact parathyroid hormone (iPTH) was measured using electrochemiluminescence immunoassay.

The standard of IS potassium salt (99.8%) was purchased from Sigma-Aldrich (St. Louis, MO). The internal standard of hydrochlorothiazide (99.5%) was provided by the Shanghai Institute for Drug Control. Use of stable isotope-labeled IS as the internal standard may be one of the best approaches in terms of accuracy of the assay. However, it is not practical, because isotope-labeled IS is not commercially available in many places. The HPLC tandem mass spectrometry method modified according to Wang and Korfmacher (16) was used to detect IS concentration in plasma. Briefly, 100 μl plasma was pipetted into a 2-ml polypropylene tube. Then, 500 μl internal standard/protein precipitation solution (50 ng/ml hydrochlorothiazide in methanol) was added to precipitate the proteins. The contents were vortex mixed for 1 minute. After centrifugation at 12,000×g for 10 minutes, a 100-μl aliquot of clear supernatant was mixed with 100 μl water in a polypropylene tube and transferred to the autosampler. A volume of 5 μl was injected into liquid chromatography tandem mass spectrometry. The chromatographic separation was achieved on a Venusil XBP Phenyl column (100×2.1 mm, 5 μm; Bonna-Agela Technologies Inc., Wilmington, DE). Mobile phase A was 2 mmol/L ammonium acetate in 0.1% (vol/vol) formic acid. Mobile B was methanol. The mobile phase (A:B=30:70) was delivered at a flow rate of 0.3 ml/min. The temperatures of the column and the autosampler were maintained at 40°C and 4°C, respectively. Then, IS and the internal standard were eluted at 1.8 and 1.4 minutes, respectively. The chromatographic run time of each sample was 4 minutes. Mass spectrometric detection was performed on an API 3000 triple quadrupole instrument (Applied Biosystems, Toronto, ON, Canada) in multiple reaction monitoring mode. A TurboIonSpray ionization interface in negative ionization mode was used. Turbo spray voltage was set at −4200 V. Source temperature was maintained at 500°C. The compound parameters of collision energy, declustering potential, entrance potential, and collision exit potential were −25, −40, −10, and −10 V for IS and −30, −58, −10, and −6 V for hydrochlorothiazide. Quadrupoles 1 and 3 were maintained at unit resolution. Dwelling time was set at 200 ms for all of the analytes. Mass transitions m/z 212.1→81.1 for IS and m/z 296.2→268.8 for hydrochlorothiazide were used. Data processing was performed with the Analyst 1.4.1 software package (Applied Biosystems). The standard curves for IS were set at 0.125, 0.25, 0.5, 1, 5, 10, 50, and 100 μg/ml, with an average r value of 0.999 (n=6). The lower limit of quantitation was 0.125 μg/ml. Intra- and interassay mean biases were 3.3% and 4.2%, respectively.

Echocardiography

Transthoracic echocardiographic examinations were performed using a Philips echocardiographic machine (Philips IE33; Philips, Eindhoven, The Netherlands) with a 3.5-MHz multiphase array probe by a single experienced cardiologist on a midweek nondialysis day within 2 hours after blood sampling. Measurements of the left ventricular internal dimension, interventricular septal thickness, and posterior wall thickness were determined at end diastole in M-mode echocardiography according to the recommendations of the Penn Convention. Meanwhile, left ventricular ejection fraction (LVEF) was calculated from two-dimensional apical images according to Simpson’s method. Left ventricular mass was calculated with the Devereux Equation (17), and left ventricular mass index (LVMI) was obtained by dividing left ventricular mass by height in meters raised to the power of 2.7. The height-based indexing was specifically chosen to minimize any potential distortion attributable to extracellular volume expansion.

End Point Evaluation

Heart failure was defined as dyspnea in addition to two of the following conditions: (1) raised jugular pressure, (2) bibasilar crackles, (3) pulmonary venous hypertension or interstitial edema on chest x-ray requiring hospitalization or extra ultrafiltration, or (4) LVEF≤40% (18,19). Follow-up was extended until July 31, 2013. Events, including heart failure, angina pectoris, acute myocardial infarction, arrhythmias, cerebral infarction, cerebral hemorrhage, and death (cardiac or noncardiac), were identified by a group of physicians who were blinded to the IS data. The primary end point was first heart failure event during follow-up. All patients were assessed every month for clinical events. If a patient was admitted to the hospital/emergency department or needed an unscheduled/urgent hemodialysis treatment, a record would be made by the doctor on duty and discussed on Friday afternoon of that week.

Statistical Analyses

All data were expressed as means±SDs, medians (interquartile ranges), or frequencies as appropriate. To compare two groups of normal data, independent samples t tests were used, whereas for skewed and categorical data, Mann–Whitney U and chi-squared tests were performed, respectively. The Kaplan–Meier method and Cox proportional hazard model were used to assess the association of IS and first heart failure event. To adjust for confounding risk factors for first heart failure event, we constructed model 1 (age, sex, and body mass index), model 2 (history of primary hypertension, coronary heart disease, and diabetes), model 3 (urinary volume, ultrafiltration volume, average interdialytic weight gain, RRF, and single-pool Kt/V), model 4 (albumin, prealbumin, BUN, SCr, UA, and nPNA), model 5 (LDL-cholesterol, HDL-cholesterol, apoprotein A (Apo-A), Apo-B, and Hcy), model 6 (P, Ca×P, iPTH, and 25-hydroxy vitamin D), model 7 (hsCRP and β₂M), model 8 (LVMI and LVEF), and model 9 (history of taking calcium channel blocker, angiotensin conversion enzyme inhibitor, angiotensin receptor blocker, β-blocker, α-blocker, aspirin, and statin). In model 10, hierarchical selection procedures were used. The inclusion criterion for model selection in a covariate set was predetermined as P<0.10 in univariate Cox proportional hazard models. The covariates with P<0.10 for predicting first heart failure event included age, systolic BP, history of primary hypertension, history of coronary heart disease, history of diabetes, prealbumin, 25-hydroxy vitamin D, LDL-cholesterol, Apo-B, LVMI, and LVEF. In model 10, the backward conditional method was used. IS entered the model either as a continuous variable or a dichotomous variable. A two-tailed P value <0.05 was considered statistically significant. All analyses were performed using SPSS 17.0 (SPSS Inc., Chicago, IL).