Immunoglobulin Light Chains Ameliorate Survival of Hemodialysis Patients

Background/Aims: Impaired immune function is common in patients with chronic renal failure. Now, we determined whether serum levels of free immunoglobulin light chains predict mortality in patients with chronic kidney disease stage 5 on hemodialysis. Methods: We performed a prospective cohort study of 160 hemodialysis patients with a median follow-up of 15 months (interquartile range, 3–44 months). Serum levels of free  and immunoglobulin light chains were measured at the start of the study. The primary end point was mortality from any cause. Results: In survivors, median serum levels of free  plus  immunoglobulin light chains were significantly higher compared with nonsurvivors (p ! 0.05). Survival was significantly longer in those patients who had serum levels of free  plus immunoglobulin light chains above the median compared with patients with serum levels below the median of 210 mg/l ( 2 = 5.91; p = 0.015 by log-rank, Mantel-Cox, test). We performed univariate and multivariate regression analysis showing that older age and lower serum levels of free plus immunoReceived: January 14, 2011 Accepted: April 1, 2011 Published online: June 3, 2011 Dr. Martin Tepel Department of Nephrology, Odense University Hospital, and Cardiovascular and Renal Research, Institute of Molecular Medicine, University of Southern Denmark Winslowparken 21.3, DK–5000 Odense C (Denmark) E-Mail mtepel @ health.sdu.dk © 2011 S. Karger AG, Basel 1420–4096/11/0345–0344$38.00/0 Accessible online at: www.karger.com/kbr D ow nl oa de d by : 54 .7 0. 40 .1 1 11 /6 /2 01 7 7: 08 :0 4 P M Free Light Chains and Survival Kidney Blood Press Res 2011;34:344–349 345 and afterwards reabsorbed by the proximal tubuli mediated by megalin-cubulin receptor endocytosis and metabolized [3] . Only if the serum concentration of free light chains is greatly increased, the absorption maximum of the kidney is exceeded and free light chains are excreted by the urine. Due to reduced free light chain clearances, patients with chronic kidney disease have increased levels of serum free immunoglobulin light chains. Previous studies showed that the increased serum concentrations of free immunoglobulin light chains correlate with parameters of kidney function including creatinine and cystatin C [3, 4] . Several factors have been established that predict mortality in hemodialysis patients [5, 6] . However, currently the consequences of increased serum concentrations of free immunoglobulin light chains for life expectancy in patients with chronic kidney disease remain unknown. Therefore, we prospectively evaluated the role of the serum concentrations of free and  immunoglobulin light chains as an independent predictor of mortality in 160 patients with chronic kidney disease stage 5 on hemodialysis. Subjects and Methods Patients We prospectively studied 160 patients with chronic kidney disease stage 5 on hemodialysis treatment. All subjects gave written informed consent, and the study was approved by the local ethics committee. 112 were male and 48 were female, median age was 67 years (interquartile range 56–76 years). Dialyses were performed using standardized techniques with bicarbonate-based dialysates and controlled ultrafiltration rate. Blood flow rates were 250–300 ml/min, dialysate flow rates were 500 ml/min, and dialysate conductivity was 135 mS/cm. Patient dry weight was defined as the body weight below which the patient experienced hypotension or muscle cramps, and postural hypotension was clinically manifest. Patient history was assessed by trained personnel using medical records and a standardized questionnaire, and comprised personal history and previous history of cardiovascular disease, including coronary artery disease, angina pectoris and peripheral vascular and cerebrovascular disease. Blood pressure was measured before dialysis with a sphygmomanometer after 10 min of recumbency. Phases I and V of the Korotkoff sounds were taken as systolic and diastolic blood pressure, respectively. Pulse pressure was calculated as the difference between systolic and diastolic blood pressure. Blood samples were taken before the hemodialysis session. Kt/V values, i.e. the amount of plasma cleared of urea divided by the urea distribution volume, was measured according to the formula Kt/V = –ln (R – 0.03) + (4 – 3.5 ! R) ! UF/W, where R = post-/pre-plasma urea nitrogen ratio, UF = ultrafiltrate volume (liters) removed, and W = postdialysis weight (kg) [7] . To be included in the study, patients needed to have chronic kidney disease stage 5 and to be routinely dialyzed for 4–5 h three times weekly using biocompatible polysulfone low flux membrane (F8, Fresenius Medical Care; ultrafiltration coefficient, 7 ml/h/mm Hg; surface area, 1.0 m 2 ) with no dialyzer reuse. Patients were ambulatory and free of acute intercurrent illness. Both male and female patients were included. Patients who did not give written consent were excluded from the study. Patients with known or already treated monoclonal gammopathies were excluded. Measurement of Serum Free Immunoglobulin Light Chains Serum levels of free and immunoglobulin light chains were measured by nephelometry as described [8] . We used a particleenhanced, high-specificity, homogenous immunoassay according to the recommendations of the manufacturer. The assay sensitivity was less than 1 mg/l [9] . The manufacturer of the immunoassay (Freelite ; The Binding Site, Birmingham, UK) provided the assay for determination of serum levels of free and immunoglobulin light chains. The company had no role in the design of the study or the collection, analysis, or interpretation of the data or the writing of the manuscript. The authors vouch for the accuracy and completeness of the reported data. Statistics Data were expressed as the median and interquartile range. Statistical analyses were performed using GraphPad Prism 5.0 (GraphPad Software, San Diego, Calif., USA) or SPSS for Windows (version 15.0; SPSS, Chicago, Ill., USA). Data on free immunoglobulin light chains were analyzed using the D’AgostinoPearson normality test. Because data on free light chains did not show Gaussian distribution (K2 = 234, p ! 0.0001), data were transformed logarithmically. Between groups, comparisons were made using the nonparametric Mann-Whitney test. The relationships between serum concentrations of free  plus  immunoglobulin light chains and other variables were analyzed by nonparametric Spearman correlation. Categorical variables comprising gender (male = 1, female = 2) and diabetes (yes = 1, no = 0) were compared using Fisher’s exact test. A p value less than 0.05 was taken to indicate statistical significance. All statistical tests were two sided. The outcome event investigated was all-cause mortality. This outcome event was specified a priori. Survival curves (time-toevent analysis) were generated using the Kaplan-Meier product limit method and compared by the log-rank (Mantel-Cox) test. For the Kaplan-Meier analysis, the cohort was divided into two groups according to serum levels of free  plus  immunoglobulin concentrations above or below the median value. The analysis used the unadjusted Cox proportional hazards analysis. Multivariate Cox proportional hazard analysis was then applied to determine the independence of the relationship of all covariates. Covariates were selected on the basis of prior research results, clinical experience and classical clinical performance measures, and included age, Kt/V urea value, gender, smoking, diabetes mellitus, previous cardiovascular disease and pulse pressure. In a stepwise forward regression analysis, variables with a p value less than 0.05 were retained. Adjusted relative risks were calculated as the antilogarithm of the  -coefficient of the Cox-proportional hazards regression analysis. The 95% confidence interval for the relative risk estimates was obtained using the antilogarithm ( 8 1.96 ! standard error of ). Each significant predictor identified by that analysis was subsequently tested in a backward selection process with all covariates forced into the model. D ow nl oa de d by : 54 .7 0. 40 .1 1 11 /6 /2 01 7 7: 08 :0 4 P M Thilo /Caspari /Scholze /Tepel Kidney Blood Press Res 2011;34:344–349 346 Results We investigated the outcome of patients with chronic kidney disease stage 5 on hemodialysis treatment according to the serum concentrations of free and  immunoglobulin light chains in a prospective cohort study. A total of 160 patients with chronic kidney disease stage 5 (112 males, 48 females; median age, 67 years; interquartile range, 56–76 years) who were receiving hemodialysis for at least 1 month were enrolled in the study. The cause of end-stage renal disease was diabetic nephropathy in 29 cases (18%), nephrosclerosis in 62 cases (39%), chronic glomerulonephritis in 17 cases (11%), polycystic kidney disease in 4 cases (3%), and other/unknown in 48 cases (30%). Angiotensin-converting enzyme inhibitors were prescribed in 47 cases (29%), beta-blockers in 83 cases (52%), calcium channel blockers in 48 cases (30%) and erythropoietin therapy in 95 cases (59%). The median duration of hemodialysis at inclusion was 7 months (interquartile range, 1–38 months). 144 previous cardiovascular events were reported in 160 patients (90%), including a history of coronary artery disease (72 cases, 45%), peripheral vascular disease (35 cases, 22%) and cerebrovascular disease (27 cases, 17%). Patients with known or already treated monoclonal gammopathies were excluded from the study. The clinical and biochemical characteristics of the patients are summarized in table 1 . Median Kt/V was 1.1 (interquartile range, 1.0–1.2). In 160 patients, the median serum level of free  light chains was 107 mg/l (interquartile range, 64–182 mg/l), and for free  light chains it was 72 mg/l (interquartile range, 42–139 mg/l). The median serum level of free plus immunoglobulin light chains was 210 mg/l (interquartile range, 116–316 mg/l). Next, we assessed factors that may influence serum levels of free immunoglobulin light chain in hemodialysis patients. We found that serum levels of free plus immunoglobulin light chains were significantly correlated with age (Spearman r = –0.17; 95% CI –0.32 to –0.01; p ! 0.05) but not with Kt/V (Spearman r = 0.07; 95% CI –0.10 to 0.23; p = 0.42) or pulse pressure (Spearman r = 0.10; 95% CI –0.07 to 0.27; p = 0.22). Now, we evaluated whether the serum levels of free and  immunoglobulin light chains predicted mortality in hemodialysis patients. Data on mortality were obtained for the entire cohort. During follow-up (median, 15 months; interquartile range, 3–44 months), 87 patients (54%) died. The cause of death was cardiovascular disease in 44 patients (51%), infections in 31 patients (36%), cancer in 9 patients (10%), and other or unknown in 3 patients (3%). Nine patients (6%) underwent kidney transplantation during follow-up. Those patients were censored on the day of transplantation. In survivors, serum levels of free  plus immunoglobulin light chains were significantly higher (median, 230 mg/l; interquartile range, 141–369 mg/l; n = 73) compared with nonsurvivors (median, 183 mg/l; interquartile range, 86–293 mg/l; n = 87; p ! 0.05). These data indicated that lower serum levels of  plus free immunoglobulin light chains were associated with increased mortality. Figure 1 a shows the probabilities of survival of patients with chronic kidney disease stage 5 on hemodialysis according to the serum level of free  plus  immunoglobulin light chains. Survival was significantly longer in patients with serum levels of free  plus immunoglobulin light chains above the median value (i.e. above 210 mg/l) compared to patients with serum levels below the median ( 2 = 5.91; p = 0.015 by log-rank, Mantel-Cox, test). Median survival in the group with serum levels of free plus immunoglobulin light chains above 210 mg/l was 43 months, whereas median survival in the group with serum levels below 210 mg/l was 27 months. As indicated in figure 1 b, c, patients with  immunoglobulin light chains above median or immunoglobulin light chains above median showed marginally significant improved survival, p = 0.066 or p = 0.054 by log-rank (Mantel-Cox) test, respectively. Table 1. Clinical and biochemical characteristics of 160 patients with chronic kidney disease stage 5 on hemodialysis Characteristic Median (interquartile range) Age, years 67 (56–76) Male/female 112/48 BMI 24.1 (20.9–26.5) Duration of hemodialysis, months 7 (1–38) Systolic blood pressure, mm Hg 130 (113–150) Diastolic blood pressure, mm Hg 69 (57–80) Pulse pressure, mm Hg 61 (50–71) Hemoglobin, g/dl 10.1 (9.2–11.6) Serum albumin, g/dl 3.3 (2.9–3.8) Serum cholesterol, mg/dl 154 (129–186) Serum triglyceride, mg/dl 159 (106–230)