Clinical Epidemiology

Urinary Sulfur Metabolites Associate with a Favorable Cardiovascular Risk Profile and Survival Benefit in Renal Transplant Recipients

Else van den Berg, Andreas Pasch, Welmoet H. Westendorp, Gerjan Navis, Elizabeth J. Brink, Reinold O.B. Gans, Harry van Goor and Stephan J.L. Bakker
JASN June 2014, 25 (6) 1303-1312; DOI: https://doi.org/10.1681/ASN.2013050497

Abstract

In post-transplant conditions, sulfur may be protective by intermediate conversion to hydrogen sulfide and thiosulfate. However, sulfate, the end product of sulfur-containing amino acids (SAAs), contributes to metabolic acid load and may adversely influence acid-base homeostasis. We investigated the association of urinary sulfur metabolites with cardiometabolic parameters in renal transplant recipients (RTRs) and analyzed their predictive capacity for mortality. We studied urinary sulfate and thiosulfate excretion in 24-hour urine samples from 707 RTRs at a median 5.4 years (interquartile range, 1.9 to 12.2) after transplantation as well as from 110 controls. Diet was assessed for SAA content and various risk factors were measured. Urinary sulfate was similar, whereas thiosulfate was higher in RTRs versus controls. SAA intake was lower in RTRs compared with controls and correlated with sulfate but not thiosulfate excretion. Sulfate beneficially associated with eGFR, net acid excretion, systolic BP, high-sensitivity C-reactive protein, N-terminal probrain natriuretic peptide, and proteinuria (all P≤0.01). Thiosulfate beneficially associated with eGFR, serum acidity, high-sensitivity C-reactive protein, and N-terminal probrain natriuretic peptide (all P≤0.001). During a median 27 months (interquartile range, 22–36) of follow-up, 47 RTRs died. After adjustment for age, sex, and eGFR, hazard ratios for mortality were 0.87 (95% confidence interval, 0.82 to 0.92; P<0.001) for urinary sulfate and 0.60 (95% confidence interval, 0.41 to 0.59; P=0.01) for thiosulfate. Thus, despite the association of urinary sulfate with metabolic acid load, urinary sulfate and thiosulfate beneficially associated with survival in RTRs, possibly by influencing cardiovascular parameters. Intervention studies with exogenous sulfur are warranted to elucidate mechanisms underlying these promising associations in RTRs.

Renal transplantation is the preferred treatment for patients with ESRD. However, despite improvement of quality of life and life expectancy in renal transplant recipients (RTRs), morbidity and mortality rates remain high compared with the general population.1,2 This likely has a multifactorial origin because many risk factors, including cardiovascular disease, acidosis, low-grade inflammation, and impaired graft function, are highly prevalent after transplantation.2–4 A pressing need exists to gain insight into mechanisms underlying post-transplant morbidity and mortality to provide therapeutic tools for improvement of patient and graft survival. Previously, nutritional factors including sodium and metabolic acid load appeared to be adversely associated with various cardiovascular and metabolic processes in RTRs.5,6 In the past decades, interest in sulfur metabolism in humans has grown considerably, particularly with regard to the synthesis and function of hydrogen sulfide (H2S).7–15 H2S is an endogenously produced gaseous compound and a signaling molecule with substantial biologic potential, suggested to be beneficially involved in various (patho-)physiologic processes including BP regulation, inflammation, angiogenesis, and cytoprotection during hypoxia.7–15 H2S is synthesized by conversion of the amino acid cysteine, which is either directly ingested via food or originates from the essential amino acid methionine. A specific intermediate of the H2S metabolism is thiosulfate (S2O32−), which is physiologically excreted in urine. Indeed, urinary thiosulfate has been reported to be linearly associated with inhaled or intravenously administered H2S, and might therefore, at least in part, reflect systemic H2S levels.16 Because of the protective effects of H2S, it can be hypothesized that increasing intake of sulfur-containing amino acids (SAAs) contributes to the synthesis of H2S and beneficially influences the cardiovascular profile and, consequently, patient survival. On the other hand, the major end product of SAAs is sulfate (SO42−),17,18 which is allegedly adverse for its contribution to metabolic acid load and systemic acidosis, particularly in patients with impaired renal function.6 Therefore, the roles of sulfur metabolism in cardiovascular and metabolic health in RTRs and whether sulfuric compounds should be considered beneficial or harmful are unclear.

We investigated the association of sulfur-containing metabolites with cardiovascular and metabolic parameters in a large, well defined cohort of stable RTRs. Intake of protein, particularly SAAs, and protein sources was assessed with food frequency questionnaires (FFQs). In RTRs and in healthy controls, we measured sulfate and thiosulfate and investigated the associations of these metabolites with cardiovascular and metabolic parameters in RTRs. In addition, we analyzed the association of urinary sulfate and thiosulfate excretion with mortality in RTRs.

Results

The characteristics of RTRs and controls are shown in Table 1. For RTRs, the median time between transplantation and baseline measurements was 5.4 years (interquartile range [IQR], 1.9–12.2). The two groups were similar with respect to age, body mass index (BMI), and body surface area (BSA). Men were overrepresented in the RTR group compared with the control group. Urinary sulfate excretion was similar in both groups (18.0±5.6 in controls versus 17.6±6.4 in RTRs; P=0.29), but thiosulfate excretion was significantly higher in RTRs compared with controls (7.0 [IQR, 3.9–11.9] versus 2.1 [IQR, 0.29–10.1]; P<0.001). As anticipated, creatinine clearance was significantly lower in RTRs than in healthy participants (66±26 versus 132±41 ml/min; P<0.001). Correspondingly, higher BP values were observed in RTRs than in controls (systolic BP [SBP], 136±18 versus 125±15 mmHg; diastolic BP [DBP], 83±11 versus 75±9 mmHg; both P<0.001), despite 89% of RTRs using one or more antihypertensive drugs. Compared with controls, RTRs had significantly higher serum levels of triglycerides, glycosylated hemoglobin (HbA1c), high-sensitivity C-reactive protein (hs-CRP), and N-terminal probrain natriuretic peptide (NT-proBNP) (all P<0.001). Serum albumin levels, serum pH, and serum bicarbonate levels were lower and RTRs were more likely to have metabolic acidosis (blood HCO3−<24 mmol/L; P<0.001). Energy intake, as well as intake of total protein, plant protein, SAA, and total fat, was significantly lower in RTRs than in controls.

View this table:
Table 1.

Characteristics of 110 healthy controls and 707 RTRs at the day of their visit to the outpatient clinic

Patient characteristics according to sex-stratified tertiles of urinary sulfate and urinary thiosulfate excretion are shown in Supplemental Tables 1 and 2. There were no differences across tertiles regarding preexisting disease, diabetes, and time since transplantation. The regression coefficients for the associations of urinary sulfate and thiosulfate with relevant cardiovascular parameters are presented in Table 2. After adjustment for potential confounders (model 3), urinary sulfate excretion was significantly inversely associated with SBP (β=−0.34; P=0.004), pulse pressure (β=−0.33; P<0.001), serum NT-proBNP (β=−0.03; P<0.001), serum hs-CRP (β=−0.04; P<0.001), serum HbA1c (β=−0.02; P=0.01), and proteinuria (β=−0.04; P=0.002). A direct association was observed between urinary sulfate excretion and eGFR (β=0.48; P<0.001). All observed associations remained significant after additional adjustment for urinary thiosulfate excretion. After adjustment for sulfate excretion (model 4), thiosulfate excretion was also significantly inversely associated with NT-proBNP (β=−0.03; P<0.001) and hs-CRP (β=−0.02; P<0.001). A direct association was observed between thiosulfate and eGFR (β=0.34; P<0.001). Associations of sulfate and thiosulfate with metabolic parameters are shown in Table 3. Sulfate was positively associated with net acid excretion (NAE) (β=1.33; P<0.001), reflecting metabolic acid load, but no association was seen with serum HCO3− (β=1.03; P=0.13) or serum pH (β=0.001; P=0.55). For thiosulfate, no association was observed with NAE; however, direct associations were observed with venous HCO3− (β=0.03; P<0.001) and venous pH (β=0.01; P<0.001). Again, these associations remained significant after additional adjustment for urinary sulfate excretion. Figure 1 depicts the standardized β values of the associations of sulfur-containing metabolites with cardiovascular and metabolic parameters (model 3), allowing for mutual comparison of the strengths of associations.

View this table:
Table 2.

Regression coefficients for the association of urinary sulfate and thiosulfate with cardiovascular parameters in 707 RTRs

View this table:
Table 3.

Regression coefficients for the association of urinary sulfate and thiosulfate with metabolic parameters in 707 RTRs

Figure 1.
Figure 1.

Regression coefficients for the association of urinary sulfate and thiosulfate excretion with cardiovascular and metabolic parameters. (A) Associations are shown for cardiovascular parameters and are adjusted for age, sex, BSA, eGFR, time since transplantation, urinary sodium excretion, and medication use (antihypertensives, calcineurin inhibitors, proliferation inhibitors). (B) Associations are shown for metabolic parameters and are adjusted for age, sex, BSA, eGFR, time since transplantation, urinary sodium excretion (mmol/24 h), and medication use (antihypertensives, calcineurin inhibitors, proliferation inhibitors). Regression coefficients are given as standardized β values, referring to the number of SDs the cardiovascular or metabolic parameter changes per SD increase of either sulfate or thiosulfate. MAP, mean arterial pressure.

The regression coefficients for the association of dietary protein and urinary sulfate and thiosulfate excretion are shown in Table 4. After adjustment for age, sex, BSA, and renal function, significant positive associations were observed between urinary sulfate and intake of total protein (standardized β=0.22; P<0.001), animal protein (standardized β=0.23; P<0.001), vegetable protein (standardized β=0.08; P=0.03), and intake of SAA (standardized β=0.20–0.24; P<0.001). No such associations were observed between dietary protein and urinary thiosulfate excretion.

View this table:
Table 4.

Regression coefficients for the association of protein intake with urinary sulfate and thiosulfate excretion in 637 RTRs

Sulfate, Thiosulfate, and Mortality

Median follow-up from baseline was 27 months (IQR, 22–36). During this prospective follow-up, 47 RTRs (7%) died. RTRs who died had significantly lower urinary excretion of sulfate (13.2±5.5 versus 17.9±6.3 mmol/24 h; P<0.001) and thiosulfate (5.4 [IQR, 2.8–8.5] versus 7.2 [IQR, 4.0–12.1] μmol/24 h; P=0.01) than RTRs who survived during follow-up. According to sex-stratified tertiles of sulfate, incidence of mortality during follow-up was 31 of 203 (15%) for the lowest tertile, whereas it was 10 of 226 (4%) and 6 of 229 (3%) in the middle and highest tertiles, respectively (log-rank test P<0.001; Figure 2A). According to increasing sex-stratified tertiles of thiosulfate, these values were 22 (10%), 17 (7%), and 8 (3%), respectively (log-rank test P=0.02; Figure 2B). Results of Cox regression analyses for mortality in RTRs are shown in Table 5. Sulfate and thiosulfate were significantly associated with mortality, with hazard ratios of 0.87 (95% confidence interval [95% CI], 0.82 to 0.92; P<0.001) and 0.60 (95% CI, 0.41 to 0.89; P=0.01), respectively. Regarding tertiles of urinary sulfate and thiosulfate excretion (with the lowest tertile as reference category), the hazard ratios of the highest tertiles of urinary sulfate and urinary thiosulfate excretion were 0.20 (95% CI, 0.08 to 0.49; P<0.001) and 0.30 (95% CI, 0.12 to 0.78; P=0.01), respectively, after adjustment for age, sex, and renal function (model 3). Slight but significant differences were observed regarding several factors that might be considered confounders of the association between urinary sulfate or thiosulfate excretion and mortality in RTRs. These include BMI, proteinuria, use of immunosuppressive agents, and use of antihypertensive medication. However, secondary Cox regression analyses with additional adjustment for these factors did not materially change our findings (Table 5).

Figure 2.
Figure 2.

Higher urinary sulfate and thiosulfate levels were associated with significant survival benefit in renal transplant recipients. Kaplan–Meier curves for patient survival according to urinary sulfate (A) and thiosulfate (B) excretions.

View this table:
Table 5.

Cox regression analyses for prediction of patient mortality based on urinary excretion of sulfate and thiosulfate

Discussion

The cardinal finding of this study is the significant association of urinary sulfur metabolites with a favorable cardiovascular risk profile and with improved survival in RTRs. These associations were found despite the observed direct association of urinary sulfate with the allegedly adverse metabolic acid load. In addition, RTRs had a markedly elevated excretion of urinary thiosulfate, but not sulfate, compared with healthy controls. Our findings suggest a protective role for sulfur metabolites in cardiovascular changes after renal transplantation and in long-term patient survival after transplantation.

Humans depend on exogenous sources for their sulfur supplies because sulfur enters the body mainly as a constituent of the SAAs cysteine and methionine. In line with previous studies, we observed a significant direct association between intake of SAAs and urinary sulfate excretion.17,18 SAAs are eventually converted to sulfate and are allegedly adverse for their contribution to metabolic acid load, which has previously been shown to be associated with the extent of acidosis in RTRs.6 In accordance, a significant association between intake of SAAs and NAE was found, the latter reflecting metabolic acid load.19 Nevertheless, regarding systemic metabolic parameters, we observed no significant association of sulfate with blood pH and HCO3−. The opposite held true for thiosulfate, showing significant direct associations with both blood pH and HCO3−, which could suggest a favorable role for thiosulfate in acid-base homeostasis, given the large body of RTRs experiencing systemic acidosis.6 The significance of and mechanism underlying this association are not yet clear and need further elucidation.

Regarding cardiovascular parameters, we observed beneficial associations of thiosulfate with NT-proBNP, hs-CRP, and eGFR. Because thiosulfate is a central metabolite in the metabolic pathway of H2S, a gaseous transmitter known for its protective role in several pathologic processes such as inflammation and hypertension, a potential explanation for our findings might be involvement of H2S. H2S is synthesized endogenously by the activity of three major enzymes, cystathionine-β-synthetase (CBS), cystathionine-γ-lyase (CSE), and 3-mercaptopyruvate sulfurtransferase, which all use the amino acid L-cysteine as their main substrate.20 CSE, the most predominant enzyme in cardiovascular tissue, is expressed in endothelial and smooth muscle cells. In the vasculature, H2S is known to function as an endothelial cell–relaxing factor. Accordingly, CSE−/− mice have a slightly higher BP than wild-type mice.21 Endothelial-derived H2S also functions as a promoter of angiogenesis and as an inhibitor of leukocyte adhesion, thereby reducing inflammation.7,13 Tokuda et al. recently reported that inhalation of H2S prevented endotoxin-induced systemic inflammation and even prolonged survival in endotoxemic mice.22 H2S eventually gets degraded to thiosulfate, which is partly reused as a sulfur donor to form a new H2S molecule, partly converted to sulfite, the latter being oxidized to sulfate and physiologically excreted in the urine. A certain amount of thiosulfate, however, is excreted directly in the urine, possibly as “spillover” of H2S synthesis and might therefore at least in part reflect systemic levels of H2S.23 Another potential explanation for our findings might be that thiosulfate itself underlies the favorable associations with cardiovascular parameters in RTRs. Thiosulfate has been reported to exert protective activities in several pathophysiologic processes such as oxidative stress.24 In addition, thiosulfate has recently been shown to completely prevent vascular calcifications in uremic rats.25 Moreover, thiosulfate can be safely administered to humans and is well tolerated.26

Unexpectedly, similar beneficial associations were observed for sulfate, with even additional favorable associations with SBP, pulse pressure, HbA1c, and proteinuria. This was in contrast with our hypothesis that urinary sulfate, reflecting intake of acidifying SAA and metabolic acid load, contributes to systemic acidosis and related complications such as hypertension and insulin resistance, particularly in patients with impaired renal function.6 An explanation might be that part of the excreted sulfate has been incorporated in the H2S and/or thiosulfate synthesis as well, which is in line with previous studies showing that systemically applied H2S is excreted as either thiosulfate or sulfate.23,27 This could also explain our findings of survival benefit in RTRs with either higher urinary sulfate or higher thiosulfate excretion.

In our study, unlike sulfate, thiosulfate was not associated with dietary sulfur intake. Although the synthesis of H2S, and consequently thiosulfate, depends on the intake of SAAs, particularly methionine, the rate-limiting step is the presence of specific enzymes such as CBS and CSE. The absence of any association of thiosulfate with dietary intake could be explained by an excess presence of SAAs. We observed significantly higher urinary excretion of thiosulfate in RTRs compared with healthy controls. This was quite surprising because it might be expected that RTRs, known to be at high cardiovascular risk compared with the general population, excrete less of an indicator of the allegedly favorable H2S metabolism. It seems unlikely that differences in dietary protein intake between both groups underlie the differences in urinary thiosulfate excretion, because intake was lower in RTRs compared with controls. Possibly, the higher urinary thiosulfate excretion reflects a compensatory response to meet the increased demand for cardiovascular protection in RTRs. It could also be that prednisolone treatment in RTRs explains the difference, because it is known that glucocorticoids induce CBS activity in the liver.28

Little is known regarding renal handling of sulfate and thiosulfate. Regarding thiosulfate, Newman et al. reported that the clearance rates of inulin and thiosulfate are nearly identical.29 In the proximal tubulus, there seems to be some tubular handling, both secretion and reabsorption, as evidenced from studies in rats, rabbits, cats, and dogs.30–32 Sulfate is freely filtered in the glomerulus, and approximately 80% is reabsorbed in the proximal tubules.33 It is known that at least two transporters are involved in this process (NaSi-1 and Sat-1). Both of these transporters can also transport thiosulfate.34

A finding that puzzled us was the observation that a single transplanted kidney seems to excrete more thiosulfate than two normal kidneys do. An explanation might be that two normal kidneys can more completely perform proximal tubular reabsorption of filtered thiosulfate than one transplanted kidney. Another possibility is that a certain degree of renal insufficiency in one way or another leads to a compensatory increase in endogenous thiosulfate synthesis and subsequent excretion of it. However, to date, there are no data available to support or disprove either of these lines of reasoning or both.

Strengths of our study include the large sample size of this specific patient group consisting of well defined, stable RTRs. Extensive data collection, including data from 24-hour urine samples and dietary intake, allowed for adjustment for many confounders. We acknowledge, however, that there are also limitations. This study is an observational epidemiologic study, which makes it difficult to draw conclusions on causality. In general, statistical significance in observational studies suggests, but does not confirm, biologic significance. Whether the significant relation between urinary sulfate and thiosulfate and mortality in RTRs is a causal or an associative relation remains to be determined. In addition, it is not known whether thiosulfate excretion truly represents increased H2S production. However, it has repeatedly been shown that urinary thiosulfate excretion increases during H2S inhalation.16,35

Because little is known regarding the role of sulfur in cardiovascular health in RTRs and the sulfur handling in the kidney, a third factor might underlie the observed associations. For example, it can be hypothesized that higher urinary excretion of sulfur compounds is caused by decreased tubular reabsorption due to renal damage, which itself is well known to be associated with cardiovascular risk. However, the association between urinary sulfur metabolites and cardiovascular parameters remained significant in multivariate regression analyses, including renal function, suggesting that this association is independent of confounding factors. Nevertheless, residual confounding could have remained, because it is hard to completely adjust for the severity of each risk factor. In addition, our study population consisted predominantly of Caucasian individuals, which calls prudence to extrapolation of our results to populations of other ethnicities. The detailed metabolic determinants influencing urinary excretions of both sulfate and thiosulfate are not clear thus far and require further elucidation. Additional studies could focus on the H2S and thiosulfate metabolism and the involved enzymes, to reveal the mechanism underlying the observed associations and to investigate the potentially beneficial effects of sulfur suppletion.

Post-transplant conditions are associated with markedly increased risk for cardiovascular diseases compared with healthy individuals. The observed beneficial associations of urinary sulfur metabolites with cardiovascular parameters and patient survival in our transplant cohort might point toward a protective role of components in the sulfur metabolism. Long-term intervention studies with exogenous sulfur, either from the diet or pharmacologic agents such as sodium thiosulfate, are warranted to clarify the exact role and underlying mechanisms of sulfur metabolism in cardiovascular health and survival in RTRs.

Concise Methods

Study Population

All RTRs (aged≥18 years) with a functioning graft for at least 1 year who visited our outpatient clinic between 2008 and 2010 were invited to participate in this study. RTRs were all transplanted at the University Medical Center Groningen and had no history of drug or alcohol addiction. Of 817 initially invited RTRs, 707 (87%) signed written informed consent to participate in this study. As a reference group, we included 110 participants who were evaluated and approved for living kidney donation in our center. None had a history of kidney disease, diabetes, or cardiovascular events. Hypertension, if present, was treated with a maximum of one antihypertensive drug. For analyses regarding dietary intake, we excluded patients with missing dietary data, resulting in 637 RTRs eligible for analyses. Our institutional review board approved the study protocol (METc 2008/186), which adhered to the Declaration of Helsinki.

Assessment of Dietary Intake

Dietary intake was assessed with a validated semiquantitative FFQ that was developed at Wageningen University36 and updated several times. For our study, the FFQ was slightly modified for accurate assessment of protein intake, including types and sources of protein. Validity of the FFQ in RTRs was assessed as previously described.37 The FFQ inquired about intake of 177 food items during the last month, taking seasonal variations into account. For each item, the frequency was recorded in times per day, week, or month. The number of servings was expressed in natural units (e.g., slice of bread or apple) or household measures (e.g., cup or spoon). The questionnaire was self-administered and filled out at home. All FFQs were checked for completeness by a trained researcher and inconsistent answers were verified with the patients. Dietary data were converted into daily nutrient intake using the Dutch Food Composition Table of 2006.38

Urine and Plasma Parameters

All participants were instructed to collect a 24-hour urine sample according to a strict protocol at the day before their visit to the outpatient clinic. Urine was collected under oil and chlorhexidine was added as an antiseptic agent. Urinary sulfate was measured chromatographically (type 861; Metrohm, Herisau, Switzerland). Intra- and inter-assay variations were 2.0% and 4.3%, respectively. Urinary thiosulfate was determined by a specific HPLC method as previously described.26,39 Briefly, 25 μl of urine was derivatized with 5 μl of 46 mM monobromobimane, 25 μl of acetonitrile, and 25 μl of 160 mM HEPES/16 mM EDTA pH 8 buffer (Invitrogen, Carlsbad, CA) for 30 minutes in the dark. Derivatization of thiol groups was stopped by 50 μl of 65 mM methanosulfonic acid (Fluka, Buchs, Switzerland) and proteins were removed by recentrifugation. Intra- and inter-assay variations were 8.6% and 9.3%, respectively.26 NAE, the gold standard for assessing metabolic acid load, was calculated as titratable acid (TA)+ammonium (NH4+) – bicarbonate (HCO3−).6,19 TA was measured with an automated titrator (855 Robotic Titrosampler; Metrohm). NH4+ (Alliance HT 2795; Waters, Milford, MA) and HCO3− (type 861; Metrohm) were measured chromatographically. Blood was drawn in the morning after completion of the 24-hour urine collection. Venous blood gas analyses were assessed photometrically immediately after collection of blood samples. Plasma and urinary concentrations of electrolytes, phosphate, albumin, and urea were measured using routine laboratory methods, as were serum cholesterol, HbA1c, hs-CRP, and NT-proBNP levels. Serum creatinine was determined using a modified version of the Jaffé method (MEGA AU 510; Merck Diagnostica, Darmstadt, Germany). Renal function was assessed by estimating GFR applying the Chronic Kidney Disease Epidemiology Collaboration equation.40 Urinary albumin concentration was determined by nephelometry (Dade Behring Diagnostic, Marburg, Germany). Total urinary protein concentration was analyzed using the Biuret reaction (MEGA AU 510; Merck Diagnostica). Proteinuria was defined as urinary protein excretion ≥0.5 g/24 h.

Clinical Parameters

All measurements were performed during a morning visit to the outpatient clinic after an 8- to 12-hour overnight fasting period. BP was measured (in millimeters of mercury) according to a strict protocol as previously described.5 Participants were left alone in a room in half-sitting position while SBP, DBP, mean arterial pressure, and heart rate were measured with a semiautomatic device (Dinamap 1846; Critikon, Tampa, FL). Measurements were performed every minute for 15 minutes and the last three values were averaged. Information on participants’ health status, medical history, and medication use was obtained from patient records. Information on smoking behavior was obtained by using a questionnaire. Participants were classified as current, former, or never smokers. Body weight and height were measured with participants wearing indoor clothing without shoes. BMI was calculated as weight in kilograms divided by height in meters squared and BSA was estimated applying the universally adopted formula of Du Bois and Du Bois.41 Diabetes was defined as use of antidiabetic medication or fasting plasma glucose≥7.0 mmol/L according to American Diabetes Association criteria as previously described by our group.42

Statistical Analyses

Data analysis was performed using SPSS 18.0 software (SPSS, Inc., Chicago, IL). Normality was tested with the Kolmogorov–Smirnov test and skewed data were normalized for analyses by logarithmic transformation (albuminuria, thiosulfate excretion, hs-CRP, NT-proBNP, and protein excretion). Data are presented as the mean±SD, unless stated otherwise. Differences between RTRs and healthy controls were tested with the t test for independent samples, the Mann–Whitney U test, or the chi-squared test. The associations of sulfate and thiosulfate excretion with various potentially relevant cardiovascular factors were analyzed with linear regression analysis, with adjustment for age, sex, and BSA (model 1) and cumulative additional adjustments for eGFR (model 2) and urinary sodium excretion and use of medication (antihypertensives, calcineurin inhibitors, proliferation inhibitors; model 3). Cardiovascular factors were considered potentially relevant if the P value for association in sex-stratified analyses was <0.10 (see Supplemental Tables 1 and 2). To study the potential interdependence of both sulfur metabolites, we also adjusted for either thiosulfate or sulfate (model 4). The association of dietary intake with urinary sulfur metabolites was investigated with linear regression analyses adjusting for age and sex (model 1), BSA (model 2), and eGFR (model 3). Regression coefficients are given as β or standardized β values, the latter referring to the number of SDs a dependent variable changes, per SD increase of the independent variable (either sulfate or thiosulfate), allowing for comparison of the strength of the associations. In prospective analyses, we investigated associations of urinary sulfate and thiosulfate excretion with mortality in RTRs. We performed crude Cox regression analyses (model 1) and analyses with adjustment for age, sex (model 2), and renal function (model 3). Although the rule of thumb that no more than one potentially explanatory variable can be included in Cox regression analyses for each outcome 10 events has been debated,43 it is obvious that because of the limited number of outcome events in our study, not all variables that differed significantly between sex-stratified tertiles of urinary sulfate and urinary thiosulfate excretion can be included in one model. Therefore, in further models (models 4–6), we additionally adjusted model 3 for potential confounders, which were selected on the basis of a P value <0.10 in sex-stratified analyses (see Supplemental Tables 1 and 2). Other variables that could lie in the causal pathway of urinary sulfate and urinary thiosulfate excretion with outcome were not adjusted for, because this could unintentionally obscure otherwise existing associations.44 In all analyses, a two-sided P value <0.05 was considered statistically significant.

Disclosures

None.

Acknowledgments

We thank Beatrix Blanchard for her valuable help in measuring urinary thiosulfate concentrations.

This article was supported by Top Institute Food and Nutrition, which is a public–private partnership that generates vision on scientific breakthroughs in food and nutrition, resulting in the development of innovative products and technologies. Partners are major Dutch food companies and research organizations. This work was supported by a Grant from the Dutch Kidney Foundation (C08-2254).

Footnotes

References

  1. ↵
    1. Aakhus S,
    2. Dahl K,
    3. Widerøe TE
    : Cardiovascular morbidity and risk factors in renal transplant patients. Nephrol Dial Transplant 14: 648–654, 1999pmid:10193814
    OpenUrlCrossRefPubMed
  2. ↵
    1. Kreis HA,
    2. Ponticelli C
    : Causes of late renal allograft loss: Chronic allograft dysfunction, death, and other factors. Transplantation 71[Suppl]: SS5–SS9, 2001pmid:11583489
    OpenUrlPubMed
    1. Lentine KL,
    2. Schnitzler MA,
    3. Abbott KC,
    4. Li L,
    5. Burroughs TE,
    6. Irish W,
    7. Brennan DC
    : De novo congestive heart failure after kidney transplantation: A common condition with poor prognostic implications. Am J Kidney Dis 46: 720–733, 2005pmid:16183428
    OpenUrlCrossRefPubMed
  4. ↵
    1. Yakupoglu HY,
    2. Corsenca A,
    3. Wahl P,
    4. Wüthrich RP,
    5. Ambühl PM
    : Posttransplant acidosis and associated disorders of mineral metabolism in patients with a renal graft. Transplantation 84: 1151–1157, 2007pmid:17998871
    OpenUrlCrossRefPubMed
  5. ↵
    1. van den Berg E,
    2. Geleijnse JM,
    3. Brink EJ,
    4. van Baak MA,
    5. Homan van der Heide JJ,
    6. Gans RO,
    7. Navis G,
    8. Bakker SJ
    : Sodium intake and blood pressure in renal transplant recipients. Nephrol Dial Transplant 27: 3352–3359, 2012pmid:22499024
    OpenUrlCrossRefPubMed
  6. ↵
    1. van den Berg E,
    2. Engberink MF,
    3. Brink EJ,
    4. van Baak MA,
    5. Joosten MM,
    6. Gans RO,
    7. Navis G,
    8. Bakker SJ
    : Dietary acid load and metabolic acidosis in renal transplant recipients. Clin J Am Soc Nephrol 7: 1811–1818, 2012pmid:22935845
    OpenUrlAbstract/FREE Full Text
  7. ↵
    1. Szabó C,
    2. Papapetropoulos A
    : Hydrogen sulphide and angiogenesis: Mechanisms and applications. Br J Pharmacol 164: 853–865, 2011pmid:21198548
    OpenUrlCrossRefPubMed
    1. Distrutti E,
    2. Sediari L,
    3. Mencarelli A,
    4. Renga B,
    5. Orlandi S,
    6. Russo G,
    7. Caliendo G,
    8. Santagada V,
    9. Cirino G,
    10. Wallace JL,
    11. Fiorucci S
    : 5-Amino-2-hydroxybenzoic acid 4-(5-thioxo-5H-[1,2]dithiol-3yl)-phenyl ester (ATB-429), a hydrogen sulfide-releasing derivative of mesalamine, exerts antinociceptive effects in a model of postinflammatory hypersensitivity. J Pharmacol Exp Ther 319: 447–458, 2006pmid:16855178
    OpenUrlAbstract/FREE Full Text
    1. Chen CQ,
    2. Xin H,
    3. Zhu YZ
    : Hydrogen sulfide: Third gaseous transmitter, but with great pharmacological potential. Acta Pharmacol Sin 28: 1709–1716, 2007pmid:17959020
    OpenUrlCrossRefPubMed
    1. Blackstone E,
    2. Morrison M,
    3. Roth MB
    : H2S induces a suspended animation-like state in mice. Science 308: 518, 2005pmid:15845845
    OpenUrlAbstract/FREE Full Text
    1. Blackstone E,
    2. Roth MB
    : Suspended animation-like state protects mice from lethal hypoxia. Shock 27: 370–372, 2007pmid:17414418
    OpenUrlCrossRefPubMed
    1. Mustafa AK,
    2. Sikka G,
    3. Gazi SK,
    4. Steppan J,
    5. Jung SM,
    6. Bhunia AK,
    7. Barodka VM,
    8. Gazi FK,
    9. Barrow RK,
    10. Wang R,
    11. Amzel LM,
    12. Berkowitz DE,
    13. Snyder SH
    : Hydrogen sulfide as endothelium-derived hyperpolarizing factor sulfhydrates potassium channels. Circ Res 109: 1259–1268, 2011pmid:21980127
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Zanardo RC,
    2. Brancaleone V,
    3. Distrutti E,
    4. Fiorucci S,
    5. Cirino G,
    6. Wallace JL
    : Hydrogen sulfide is an endogenous modulator of leukocyte-mediated inflammation. FASEB J 20: 2118–2120, 2006pmid:16912151
    OpenUrlCrossRefPubMed
    1. Bos EM,
    2. Leuvenink HG,
    3. Snijder PM,
    4. Kloosterhuis NJ,
    5. Hillebrands JL,
    6. Leemans JC,
    7. Florquin S,
    8. van Goor H
    : Hydrogen sulfide-induced hypometabolism prevents renal ischemia/reperfusion injury. J Am Soc Nephrol 20: 1901–1905, 2009pmid:19628669
    OpenUrlAbstract/FREE Full Text
  15. ↵
    1. Bos EM,
    2. Snijder PM,
    3. Jekel H,
    4. Weij M,
    5. Leemans JC,
    6. van Dijk MC,
    7. Hillebrands JL,
    8. Lisman T,
    9. van Goor H,
    10. Leuvenink HG
    : Beneficial effects of gaseous hydrogen sulfide in hepatic ischemia/reperfusion injury. Transpl Int 25: 897–908, 2012pmid:22716165
    OpenUrlCrossRefPubMed
  16. ↵
    1. Kangas J,
    2. Savolainen H
    : Urinary thiosulphate as an indicator of exposure to hydrogen sulphide vapour. Clin Chim Acta 164: 7–10, 1987pmid:3581481
    OpenUrlCrossRefPubMed
  17. ↵
    1. Sabry ZI,
    2. Shadarevian SB,
    3. Cowan JW,
    4. Campbell JA
    : Relationship of dietary intake of sulphur amino-acids to urinary excretion of inorganic sulphate in man. Nature 206: 931–933, 1965pmid:5839851
    OpenUrlCrossRefPubMed
  18. ↵
    1. Wilson HE
    : Studies in nitrogen and sulphur metabolism. Biochem J 20: 76–83, 1926pmid:16743643
    OpenUrlFREE Full Text
  19. ↵
    1. Frassetto LA,
    2. Lanham-New SA,
    3. Macdonald HM,
    4. Remer T,
    5. Sebastian A,
    6. Tucker KL,
    7. Tylavsky FA
    : Standardizing terminology for estimating the diet-dependent net acid load to the metabolic system. J Nutr 137: 1491–1492, 2007pmid:17513412
    OpenUrlFREE Full Text
  20. ↵
    1. Szabó C
    : Hydrogen sulphide and its therapeutic potential. Nat Rev Drug Discov 6: 917–935, 2007pmid:17948022
    OpenUrlCrossRefPubMed
  21. ↵
    1. Yang G,
    2. Wu L,
    3. Jiang B,
    4. Yang W,
    5. Qi J,
    6. Cao K,
    7. Meng Q,
    8. Mustafa AK,
    9. Mu W,
    10. Zhang S,
    11. Snyder SH,
    12. Wang R
    : H2S as a physiologic vasorelaxant: Hypertension in mice with deletion of cystathionine gamma-lyase. Science 322: 587–590, 2008pmid:18948540
    OpenUrlAbstract/FREE Full Text
  22. ↵
    1. Tokuda K,
    2. Kida K,
    3. Marutani E,
    4. Crimi E,
    5. Bougaki M,
    6. Khatri A,
    7. Kimura H,
    8. Ichinose F
    : Inhaled hydrogen sulfide prevents endotoxin-induced systemic inflammation and improves survival by altering sulfide metabolism in mice. Antioxid Redox Signal 17: 11–21, 2012pmid:22221071
    OpenUrlCrossRefPubMed
  23. ↵
    1. Levitt MD,
    2. Furne J,
    3. Springfield J,
    4. Suarez F,
    5. DeMaster E
    : Detoxification of hydrogen sulfide and methanethiol in the cecal mucosa. J Clin Invest 104: 1107–1114, 1999pmid:10525049
    OpenUrlCrossRefPubMed
  24. ↵
    1. Hayden MR,
    2. Goldsmith DJ
    : Sodium thiosulfate: New hope for the treatment of calciphylaxis. Semin Dial 23: 258–262, 2010pmid:20636917
    OpenUrlCrossRefPubMed
  25. ↵
    1. Pasch A,
    2. Schaffner T,
    3. Huynh-Do U,
    4. Frey BM,
    5. Frey FJ,
    6. Farese S
    : Sodium thiosulfate prevents vascular calcifications in uremic rats. Kidney Int 74: 1444–1453, 2008pmid:18818688
    OpenUrlCrossRefPubMed
  26. ↵
    1. Farese S,
    2. Stauffer E,
    3. Kalicki R,
    4. Hildebrandt T,
    5. Frey BM,
    6. Frey FJ,
    7. Uehlinger DE,
    8. Pasch A
    : Sodium thiosulfate pharmacokinetics in hemodialysis patients and healthy volunteers. Clin J Am Soc Nephrol 6: 1447–1455, 2011pmid:21566113
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. Whitfield NL,
    2. Kreimier EL,
    3. Verdial FC,
    4. Skovgaard N,
    5. Olson KR
    : Reappraisal of H2S/sulfide concentration in vertebrate blood and its potential significance in ischemic preconditioning and vascular signaling. Am J Physiol Regul Integr Comp Physiol 294: R1930–R1937, 2008pmid:18417642
    OpenUrlCrossRefPubMed
  28. ↵
    1. Ratnam S,
    2. Maclean KN,
    3. Jacobs RL,
    4. Brosnan ME,
    5. Kraus JP,
    6. Brosnan JT
    : Hormonal regulation of cystathionine beta-synthase expression in liver. J Biol Chem 277: 42912–42918, 2002pmid:12198128
    OpenUrlAbstract/FREE Full Text
  29. ↵
    1. Newman EV,
    2. Gilman A,
    3. Philips FS
    : The renal clearance of thiosulfate in man. Bull Johns Hopkins Hosp 79: 229–242, 1946pmid:20280872
    OpenUrlPubMed
  30. ↵
    1. Ullrich KJ,
    2. Rumrich G,
    3. Klöss S
    : Bidirectional active transport of thiosulfate in the proximal convolution of the rat kidney. Pflugers Arch 387: 127–132, 1980pmid:7191976
    OpenUrlCrossRefPubMed
    1. Foulks J,
    2. Brazeau P,
    3. Koelle ES,
    4. Gilman A
    : Renal secretion of thiosulfate in the dog. Am J Physiol 168: 77–85, 1952pmid:14903111
    OpenUrlPubMed
  32. ↵
    1. Bucht H
    : On the tubular excretion of thiosulfate and creatinine under the influence of caronamide. Scand J Clin Lab Invest 1: 270–276, 1949
    OpenUrlCrossRef
  33. ↵
    1. Silve C
    : Tubular handling and regulation of sulphate. Nephrol Dial Transplant 15[Suppl 6]: 34–35, 2000pmid:11143983
    OpenUrlCrossRefPubMed
  34. ↵
    1. Morris ME,
    2. Murer H
    : Molecular mechanisms in renal and intestinal sulfate (re)absorption. J Membr Biol 181: 1–9, 2001pmid:11331932
    OpenUrlCrossRefPubMed
  35. ↵
    1. Durand M,
    2. Weinstein P
    : Thiosulfate in human urine following minor exposure to hydrogen sulfide: Implications for forensic analysis of poisoning. Forensic Toxicol 25: 92–95, 2007
    OpenUrlCrossRef
  36. ↵
    1. Feunekes GI,
    2. Van Staveren WA,
    3. De Vries JH,
    4. Burema J,
    5. Hautvast JG
    : Relative and biomarker-based validity of a food-frequency questionnaire estimating intake of fats and cholesterol. Am J Clin Nutr 58: 489–496, 1993pmid:8379504
    OpenUrlAbstract/FREE Full Text
  37. ↵
    1. van den Berg E,
    2. Engberink MF,
    3. Brink EJ,
    4. van Baak MA,
    5. Gans RO,
    6. Navis G,
    7. Bakker SJ
    : Dietary protein, blood pressure and renal function in renal transplant recipients. Br J Nutr 109: 1463–1470, 2013pmid:22906209
    OpenUrlCrossRefPubMed
  38. ↵
    National Institute for Public Health and the Environment: Dutch Food Composition Table of 2006. Available at: http://www.rivm.nl/en/Topics/D/Dutch_Food_Composition_Database/Publications. Accessed December 1, 2013
  39. ↵
    1. Newton GL,
    2. Dorian R,
    3. Fahey RC
    : Analysis of biological thiols: Derivatization with monobromobimane and separation by reverse-phase high-performance liquid chromatography. Anal Biochem 114: 383–387, 1981pmid:7304929
    OpenUrlCrossRefPubMed
  40. ↵
    1. Levey AS,
    2. Stevens LA,
    3. Schmid CH,
    4. Zhang YL,
    5. Castro AF 3rd,
    6. Feldman HI,
    7. Kusek JW,
    8. Eggers P,
    9. Van Lente F,
    10. Greene T,
    11. Coresh J,
    12. CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration)
    : A new equation to estimate glomerular filtration rate. Ann Intern Med 150: 604–612, 2009pmid:19414839
    OpenUrlCrossRefPubMed
  41. ↵
    1. Du Bois D,
    2. Du Bois EF
    : Clinical calorimetry. A formula to estimate the approximate surface area if height and weight be known. Arch Intern Med 17: 863–871, 1916
  42. ↵
    1. Abbasi A,
    2. Peelen LM,
    3. Corpeleijn E,
    4. van der Schouw YT,
    5. Stolk RP,
    6. Spijkerman AM,
    7. van der A DL,
    8. Moons KG,
    9. Navis G,
    10. Bakker SJ,
    11. Beulens JW
    : Prediction models for risk of developing type 2 diabetes: Systematic literature search and independent external validation study. BMJ 345: e5900, 2012pmid:22990994
    OpenUrlAbstract/FREE Full Text
  43. ↵
    1. Vittinghoff E,
    2. McCulloch CE
    : Relaxing the rule of ten events per variable in logistic and Cox regression. Am J Epidemiol 165: 710–718, 2007pmid:17182981
    OpenUrlCrossRefPubMed
  44. ↵
    1. Schisterman EF,
    2. Cole SR,
    3. Platt RW
    : Overadjustment bias and unnecessary adjustment in epidemiologic studies. Epidemiology 20: 488–495, 2009pmid:19525685
    OpenUrlCrossRefPubMed
View Abstract
Back to top

In this issue

View Selected Citations (0)
Print
Download PDF
Email Article
Citation Tools
Request Permissions
Urinary Sulfur Metabolites Associate with a Favorable Cardiovascular Risk Profile and Survival Benefit in Renal Transplant Recipients
Else van den Berg, Andreas Pasch, Welmoet H. Westendorp, Gerjan Navis, Elizabeth J. Brink, Reinold O.B. Gans, Harry van Goor, Stephan J.L. Bakker
JASN Jun 2014, 25 (6) 1303-1312; DOI: 10.1681/ASN.2013050497
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Jump to section

More in this TOC Section

Cited By...

Similar Articles

Related Articles