Advertisement

Estimated Kidney Tubular Secretion and Kidney, Cardiovascular, and Mortality Outcomes in CKD: The Systolic Blood Pressure Intervention Trial (SPRINT)

  • Author Footnotes
    ∗ S.B.A. and M.G.S. contributed equally to this work.
    Simon B. Ascher
    Correspondence
    Simon B. Ascher, MD, MPH, Division of Hospital Medicine, Department of Medicine, University of California Davis, 2315 Stockton Blvd, Suite 2P101, Sacramento, CA 95817.
    Footnotes
    ∗ S.B.A. and M.G.S. contributed equally to this work.
    Affiliations
    Kidney Health Research Collaborative, Department of Medicine, San Francisco Veterans Affairs Health Care System and University of California San Francisco, San Francisco, California

    Division of Hospital Medicine, University of California Davis, Sacramento, California
    Search for articles by this author
  • Author Footnotes
    ∗ S.B.A. and M.G.S. contributed equally to this work.
    Michael G. Shlipak
    Footnotes
    ∗ S.B.A. and M.G.S. contributed equally to this work.
    Affiliations
    Kidney Health Research Collaborative, Department of Medicine, San Francisco Veterans Affairs Health Care System and University of California San Francisco, San Francisco, California
    Search for articles by this author
  • Ronit Katz
    Affiliations
    Department of Obstetrics and Gynecology, University of Washington, Seattle, Washington
    Search for articles by this author
  • Alexander L. Bullen
    Affiliations
    Division of Nephrology-Hypertension, University of California San Diego, San Diego, California

    Nephrology Section, Veterans Affairs San Diego Healthcare System, San Diego, California
    Search for articles by this author
  • Rebecca Scherzer
    Affiliations
    Kidney Health Research Collaborative, Department of Medicine, San Francisco Veterans Affairs Health Care System and University of California San Francisco, San Francisco, California
    Search for articles by this author
  • Stein I. Hallan
    Affiliations
    Department of Clinical and Molecular Medicine, Faculty of Medicine, Norwegian University of Science and Technology, Trondheim, Norway

    Department of Nephrology, St Olav University Hospital, Trondheim, Norway
    Search for articles by this author
  • Alfred K. Cheung
    Affiliations
    Division of Nephrology and Hypertension, University of Utah Health, Salt Lake City, Utah

    Medical Service, Veterans Affairs Salt Lake City Health Care System, Salt Lake City, Utah
    Search for articles by this author
  • Kalani L. Raphael
    Affiliations
    Division of Nephrology and Hypertension, Department of Medicine, Oregon Health and Science University and VA Portland Health Care System, Portland, Oregon
    Search for articles by this author
  • Michelle M. Estrella
    Affiliations
    Kidney Health Research Collaborative, Department of Medicine, San Francisco Veterans Affairs Health Care System and University of California San Francisco, San Francisco, California
    Search for articles by this author
  • Vasantha K. Jotwani
    Affiliations
    Kidney Health Research Collaborative, Department of Medicine, San Francisco Veterans Affairs Health Care System and University of California San Francisco, San Francisco, California
    Search for articles by this author
  • Jesse C. Seegmiller
    Affiliations
    Department of Laboratory Medicine and Pathology, University of Minnesota, Minneapolis, Minnesota
    Search for articles by this author
  • Joachim H. Ix
    Affiliations
    Division of Nephrology-Hypertension, University of California San Diego, San Diego, California

    Nephrology Section, Veterans Affairs San Diego Healthcare System, San Diego, California
    Search for articles by this author
  • Pranav S. Garimella
    Correspondence
    Address for Correspondence: Pranav S. Garimella, MBBS, MPH, Division of Nephrology-Hypertension, Department of Medicine, University of California San Diego, 9452 Med Center Dr L3E #204, La Jolla, CA 92037.
    Affiliations
    Division of Nephrology-Hypertension, University of California San Diego, San Diego, California
    Search for articles by this author
  • Author Footnotes
    ∗ S.B.A. and M.G.S. contributed equally to this work.
Open AccessPublished:September 23, 2022DOI:https://doi.org/10.1016/j.xkme.2022.100546

      Rational & Objective

      Many drugs, metabolites, and toxins are cleared by the kidneys via tubular secretion. Whether novel endogenous measures of tubular secretion provide information about kidney, cardiovascular, and mortality risk is uncertain.

      Study Design

      Longitudinal subgroup analysis of clinical trial participants.

      Setting & Participants

      2,089 Systolic Blood Pressure Intervention Trial participants with estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m2 at baseline.

      Exposure

      Summary score incorporating urine-to-plasma ratios of 10 endogenous secretion markers measured in paired urine and plasma samples at baseline.

      Outcome

      The primary outcome was longitudinal changes in eGFR. Secondary outcomes included chronic kidney disease (CKD) progression (≥50% eGFR decline or incident kidney failure requiring dialysis or kidney transplantation), a cardiovascular disease (CVD) composite (myocardial infarction, acute coronary syndrome, stroke, acute decompensated heart failure, or death from cardiovascular causes), and mortality.

      Analytical Approach

      Linear mixed-effect models were used to evaluate the association between secretion score and changes in eGFR, and Cox proportional hazards models were used to evaluate associations with CKD progression, CVD, and mortality.

      Results

      At baseline, mean age was 73 ± 9 years and eGFR was 46 ± 11 mL/min/1.73 m2. During a median follow-up of 3.3 years, mean change in eGFR was −1.44% per year, and 72 CKD progression events, 272 CVD events, and 144 deaths occurred. In multivariable analyses, lower secretion score was associated with faster eGFR decline and greater risk of CKD progression, CVD, and mortality. After further adjustment for baseline eGFR and albuminuria, each 1-standard deviation lower secretion score was associated with faster eGFR decline (−0.65% per year; 95% CI, −0.84% to −0.46%), but not CKD progression (HR, 1.23; 95% CI, 0.96-1.58), CVD (HR, 1.02; 95% CI, 0.89-1.18), or mortality (HR, 0.90; 95% CI, 0.74-1.09). The secretion score association with eGFR decline appeared stronger in participants with baseline eGFR <45 mL/min/1.73 m2 (P for interaction < 0.001).

      Limitations

      Persons with diabetes and proteinuria >1 g/d were excluded.

      Conclusions

      Among SPRINT participants with CKD, lower estimated tubular secretion was associated with faster eGFR decline, independent of baseline eGFR and albuminuria, but not with CKD progression, CVD, or mortality.

      Index Words

      Many drugs, metabolites, and toxins are cleared by the kidneys via tubular secretion, but the clinical relevance of measuring tubular secretion has not been established. We used 10 novel markers to evaluate the association between estimated tubular secretion and risk of adverse outcomes in individuals with chronic kidney disease (CKD). Lower estimated tubular secretion was associated with faster longitudinal declines in estimated glomerular filtration rate, independent of baseline estimated glomerular filtration rate and albuminuria, but not with CKD progression, cardiovascular disease, or mortality. These findings suggest a broader assessment of kidney health that incorporates estimated tubular secretion may provide additional information about risk of kidney function decline in individuals with CKD but contributes little additional insight about cardiovascular disease risk or survival.
      Chronic kidney disease (CKD) affects nearly 10% of the global population and confers an increased risk of cardiovascular disease (CVD), kidney failure, and early death.
      GBD Chronic Kidney Disease Collaboration
      Global, regional, and national burden of chronic kidney disease, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017.
      ,
      • Go A.S.
      • Chertow G.M.
      • Fan D.
      • McCulloch C.E.
      • Hsu C.Y.
      Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization.
      Although estimated glomerular filtration rate (eGFR) and albuminuria are established risk markers for these outcomes, they primarily reflect glomerular function and injury and do not fully capture the degree of kidney tubule pathology.
      • Gansevoort R.T.
      • Matsushita K.
      • van der Velde M.
      • et al.
      Lower estimated GFR and higher albuminuria are associated with adverse kidney outcomes. A collaborative meta-analysis of general and high-risk population cohorts.
      • van der Velde M.
      • Matsushita K.
      • Coresh J.
      • et al.
      Lower estimated glomerular filtration rate and higher albuminuria are associated with all-cause and cardiovascular mortality. A collaborative meta-analysis of high-risk population cohorts.
      • Astor B.C.
      • Matsushita K.
      • Gansevoort R.T.
      • et al.
      Lower estimated glomerular filtration rate and higher albuminuria are associated with mortality and end-stage renal disease. A collaborative meta-analysis of kidney disease population cohorts.
      • Levin A.
      • Stevens P.E.
      • Bilous R.W.
      • et al.
      Kidney Disease: improving global outcomes (KDIGO) CKD work group
      KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease.
      The kidney tubules comprise over 90% of the kidney’s cortical mass and have a central role in blood pressure regulation, electrolyte balance, and drug secretion. On kidney biopsy, interstitial fibrosis and tubular atrophy severity is highly prognostic of CKD progression and represents the common final pathway of nearly all forms of CKD, but these pathologic findings are poorly correlated with eGFR and albuminuria.
      • Bohle A.
      • Christ H.
      • Grund K.E.
      • Mackensen S.
      The role of the interstitium of the renal cortex in renal disease.
      • Nath K.A.
      Tubulointerstitial changes as a major determinant in the progression of renal damage.
      • Eddy A.A.
      Molecular insights into renal interstitial fibrosis.
      • Rule A.D.
      • Amer H.
      • Cornell L.D.
      • et al.
      The association between age and nephrosclerosis on renal biopsy among healthy adults.
      Recent studies in ambulatory populations have demonstrated that novel markers reflecting kidney tubule health, including reabsorptive capacity, injury, inflammation, and fibrosis, are associated with greater risk of longitudinal eGFR decline, acute kidney injury, CVD, and mortality.
      • Garimella P.S.
      • Lee A.K.
      • Ambrosius W.T.
      • et al.
      Markers of kidney tubule function and risk of cardiovascular disease events and mortality in the SPRINT trial.
      • Jotwani V.K.
      • Lee A.K.
      • Estrella M.M.
      • et al.
      Urinary biomarkers of tubular damage are associated with mortality but not cardiovascular risk among systolic blood pressure intervention trial participants with chronic kidney disease.
      • Bullen A.L.
      • Katz R.
      • Lee A.K.
      • et al.
      The SPRINT trial suggests that markers of tubule cell function in the urine associate with risk of subsequent acute kidney injury while injury markers elevate after the injury.
      • Jotwani V.
      • Garimella P.S.
      • Katz R.
      • et al.
      Tubular biomarkers and chronic kidney disease progression in SPRINT participants.
      • Malhotra R.
      • Katz R.
      • Jotwani V.
      • et al.
      Urine markers of kidney tubule cell injury and kidney function decline in SPRINT trial participants with CKD.
      These findings suggest that a broader assessment of kidney structure and function, beyond eGFR and albuminuria, could inform prognosis in persons with CKD.
      • Ix J.H.
      • Shlipak M.G.
      The promise of tubule biomarkers in kidney disease: a review.
      Kidney tubular secretion is an essential mechanism for the clearance of many drugs, metabolites, and toxins. In contrast to glomerular filtration, secretion primarily occurs in the proximal tubule, involves extraction of protein-bound solutes directly from the peritubular capillaries, and relies on mitochondrial function for energy.
      • Nigam S.K.
      • Wu W.
      • Bush K.T.
      • Hoenig M.P.
      • Blantz R.C.
      • Bhatnagar V.
      Handling of drugs, metabolites, and uremic toxins by kidney proximal tubule drug transporters.
      ,
      • Wang K.
      • Kestenbaum B.
      Proximal tubular secretory clearance: a neglected partner of kidney function.
      However, the clinical relevance of measuring secretory function has not been established, in part because of a lack of standardized laboratory assays.
      • Nigam S.K.
      • Wu W.
      • Bush K.T.
      • Hoenig M.P.
      • Blantz R.C.
      • Bhatnagar V.
      Handling of drugs, metabolites, and uremic toxins by kidney proximal tubule drug transporters.
      • Wang K.
      • Kestenbaum B.
      Proximal tubular secretory clearance: a neglected partner of kidney function.
      • Marshall E.K.
      Two lectures on renal physiology.
      Emerging evidence from the Chronic Renal Insufficiency Cohort Study suggests that lower estimated clearance of novel endogenous tubular secretion markers is associated with risk of CKD progression and mortality but not CVD. These findings were independent of eGFR, albuminuria, and other CKD risk factors.
      • Chen Y.
      • Zelnick L.R.
      • Wang K.
      • et al.
      Kidney clearance of secretory solutes is associated with progression of CKD: the CRIC study.
      ,
      • Chen Y.
      • Zelnick L.R.
      • Huber M.P.
      • et al.
      Association between kidney clearance of secretory solutes and cardiovascular events: the Chronic Renal Insufficiency Cohort (CRIC) study.
      However, the prognostic significance of tubular secretion has not been confirmed in other studies of persons with CKD.
      We measured concentrations of 10 endogenous secretory solutes suspected to be eliminated primarily by tubular secretion in paired plasma and urine collected at the baseline visit in Systolic Blood Pressure Intervention Trial (SPRINT) participants with eGFR <60 mL/min/1.73 m2. The biomarkers included: apidic acid, cinnamoylglycine, p-cresol sulfate, 1,7-dimethyluric acid, 2-furoylglycine, hippuric acid, m-hydroxy hippurate, indoxyl sulfate, phenylacetylglutamine, and tiglylglycine. These solutes were selected based on the following criteria: established specificity for organic anion transporters 1 and 3 (OAT1 and OAT3), which are 2 of the primary transporters in the proximal tubule for many common drugs, metabolites, and toxins; an increase in circulating concentrations after OAT3-transporter knockout in experimental models; a high reported protein-binding percentage; and/or kidney clearances that substantially exceed glomerular filtration rate or creatinine clearance.
      • Sirich T.L.
      • Aronov P.A.
      • Plummer N.S.
      • Hostetter T.H.
      • Meyer T.W.
      Numerous protein-bound solutes are cleared by the kidney with high efficiency.
      ,
      • Suchy-Dicey A.M.
      • Laha T.
      • Hoofnagle A.
      • et al.
      Tubular secretion in CKD.
      Additional information about each secretory solute is available in the Human Metabolome Database (www.hmdb.ca). For example, indoxyl sulfate is a gut bacteria metabolite that is highly protein-bound, undergoes elimination primarily via tubular secretion mediated by OAT1/3, accumulates in persons with CKD, and has been identified as a uremic toxin according to the European Toxin Working Group (www.uremic-toxins.org). We hypothesized that baseline estimated tubular secretory function, assessed by relative urine-to-plasma concentrations of these biomarkers, would be associated with faster eGFR decline and greater risk of CKD progression, CVD, and all-cause mortality, independent of baseline eGFR and albuminuria.

      Methods

      Study Design

      The design and protocol of SPRINT have been reported previously.
      • Wright J.T.
      • Williamson J.D.
      • et al.
      SPRINT Research Group
      A randomized trial of intensive versus standard blood-pressure control.
      ,
      • Ambrosius W.T.
      • Sink K.M.
      • Foy C.G.
      • et al.
      The design and rationale of a multicenter clinical trial comparing two strategies for control of systolic blood pressure: the Systolic Blood Pressure Intervention Trial (SPRINT).
      In brief, SPRINT was an open-label clinical trial that randomized participants with hypertension and high risk for CVD to an “intensive” systolic blood pressure (BP) target of <120 mm Hg versus a “standard” BP target of <140 mm Hg. Inclusion criteria were age greater than or equal to 50 years; systolic BP 130-180 mm Hg; and high CVD risk defined as prior clinical or subclinical CVD other than stroke, CKD (eGFR 20-59 mL/min/1.73 m2), age greater than or equal to 75 years, or 10-year CVD risk > 15% based on the Framingham risk score. Key exclusion criteria included diabetes mellitus, eGFR <20 mL/min/1.73 m2, and proteinuria > 1 g/day. A total of 9,361 participants were enrolled between November 2010 and March 2013 across 102 sites in the United States and Puerto Rico. The SPRINT protocol comprised a baseline visit and follow-up visits monthly for the first 3 months, then every 3 months thereafter. All participants provided written informed consent and the institutional review boards of all participating institutions approved the study. The trial was stopped early after a median follow-up of 3.26 years based on the recommendations of the data and safety monitoring board owing to interim CVD and mortality results that favored the intensive arm.
      For this analysis, we measured 10 endogenous secretory solutes in plasma and urine at baseline among 2,514 SPRINT participants with CKD, defined as a baseline eGFR <60 mL/min/1.73 m2 according to the 2012 Chronic Kidney Disease Epidemiology Collaboration combined creatinine and cystatin C estimating equation.
      • Inker L.A.
      • Schmid C.H.
      • Tighiouart H.
      • et al.
      Estimating glomerular filtration rate from serum creatinine and cystatin C.
      We excluded 374 participants because of unavailable specimens for both plasma and urine biomarker measurements and 51 participants because of missing covariate data, resulting in a final study sample of 2,089 participants. The present study was approved by the committees on human research at the University of California, San Francisco, the San Francisco Veterans Affairs Health Care System, and the Veterans Affairs San Diego Healthcare System.

      Secretion

      All paired blood and urine specimens were processed immediately, shipped overnight on dry ice, and stored at -80°C until biomarker measurement without prior thaw. Plasma and urine biomarker concentrations were measured at the SPRINT Central Laboratory (University of Minnesota, Minneapolis, Minnesota) by liquid chromatography-tandem mass spectrometry as previously described.
      • Seegmiller J.C.
      • Wolfe B.J.
      • Albtoush N.
      • et al.
      Tubular secretion markers, glomerular filtration rate, effective renal plasma flow, and filtration fraction in healthy adolescents.
      Biomarker analytic ranges and interassay coefficients of variation are shown in Table S1. All plasma and urine solute interassay coefficients of variation were <6%. Stored urine specimens in SPRINT were from spot samples; we previously demonstrated in a pilot study that the fractional excretion of tubular secretion markers using spot urine specimens was similar to using 24-hour urine collections.
      • Garimella P.S.
      • Li K.
      • Naviaux J.C.
      • et al.
      Utility of spot urine specimens to assess tubular secretion.
      We assessed tubular secretion by calculating the urine-to-plasma ratio (UPR) of each solute and creating a summary secretion score that provided a single, overall assessment of tubular secretion. We first natural log-transformed the UPR of each secretion marker, and then standardized each secretion marker to a common 0 to 100 scale based on the minimum and maximum level of log-transformed UPR of that specific marker: standardized UPR = {[ln(UPR) – min(ln[UPR])] / range(ln[UPR])} × 100, where ln(UPR) represents the natural log-transformed UPR, min(ln[UPR])] represents the minimum value in the distribution of ln(UPR), and range(ln[UPR]) represents the difference between the minimum and maximum ln(UPR) values. For each participant, we calculated the average of the 10 secretion markers’ standardized UPR ratios to create the summary secretion score, consistent with prior studies.
      • Garimella P.S.
      • Katz R.
      • Waikar S.S.
      • et al.
      Kidney tubulointerstitial fibrosis and tubular secretion.

      Outcomes

      The primary outcome of interest was annualized eGFR slope, based on serial serum creatinine measurements collected every 3 months and measured at the SPRINT Central Laboratory. Estimated glomerular filtration rate was calculated by the 2009 Chronic Kidney Disease Epidemiology Collaboration equation for creatinine.
      • Levey A.S.
      • Stevens L.A.
      • Schmid C.H.
      • et al.
      A new equation to estimate glomerular filtration rate.
      Serum creatinine was measured with assays using an enzymatic creatinine method traceable to isotope dilute mass spectrometry (Roche).
      The binary CKD progression endpoint was assessed as a secondary endpoint because few participants experienced this outcome in SPRINT.
      • Wright J.T.
      • Williamson J.D.
      • et al.
      SPRINT Research Group
      A randomized trial of intensive versus standard blood-pressure control.
      CKD progression was defined according to SPRINT’s primary composite kidney outcome, which required either a ≥50% eGFR decline (confirmed by repeat testing ≥ 90 days) or incident kidney failure requiring dialysis or kidney transplantation. We also evaluated associations of the summary secretion score with SPRINT’s primary composite CVD outcome and all-cause mortality. SPRINT’s primary composite CVD outcome included myocardial infarction, acute coronary syndrome not resulting in myocardial infarction, stroke, acute decompensated heart failure, or death from cardiovascular causes, all of which were centrally adjudicated. Clinical events occurring during follow-up were ascertained primarily through surveillance of self-reported events obtained via structured interviews every 3 months, and through laboratory and electrocardiogram data collected by the study, and were adjudicated by members of the Morbidity and Mortality subcommittee masked to treatment assignment.

      Covariates

      Age, sex, race, past medical history, and smoking status were obtained by questionnaire. Trained study coordinators measured BP using a standardized protocol, and recorded BP as the mean of 3 seated BP measurements taken 1 minute apart after a 5-minute rest period using an automated oscillometric device (Model 907; Omron Healthcare).
      • Johnson K.C.
      • Whelton P.K.
      • Cushman W.C.
      • et al.
      Blood pressure measurement in SPRINT (Systolic Blood Pressure Intervention Trial).
      Body mass index was calculated as weight in kilograms divided by height in meters squared. Fasting serum total cholesterol, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, and triglycerides were measured at the SPRINT Central Laboratory. Urine albumin was measured by a nephelometric method using the Siemens ProSpec nephelometer (Siemens).

      Statistical analyses

      Spearman correlation coefficients (r) were used to evaluate correlations among the secretion score, eGFR, and albuminuria. Linear mixed-effect models with random intercepts, random slopes, and an exchangeable covariance structure were used to evaluate the association of baseline secretion score with annualized eGFR slope. To interpret the slope as annualized percent change, eGFR was log-transformed. Cox proportional hazards models were used to evaluate the associations of baseline secretion score with risk of CKD progression, CVD, and all-cause mortality. We first evaluated the functional form of the secretion score association with each outcome using restricted cubic splines, adjusted for age, sex, race, and randomization arm. We modeled the secretion score both as a continuous, linear predictor (per 1-standard deviation [SD]) and categorized into quartiles. Models were sequentially adjusted for age, sex, race, intervention arm; smoking, body mass index, systolic BP, number of antihypertensive medications, history of CVD, history of heart failure, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, triglycerides, and statin use; and baseline eGFR and urine albumin-to-creatinine ratio. Participants were followed until death or the last available study visit when the trial stopped in August 2015. There was no evidence that the proportional hazards assumption was violated. Because BP targets may alter urine biomarker levels and exert hemodynamic effects on eGFR, we evaluated whether the secretion score association with each outcome varied by randomized treatment arm using a likelihood ratio test.
      • Cheung A.K.
      • Rahman M.
      • Reboussin D.M.
      • et al.
      Effects of intensive BP control in CKD.
      We also evaluated whether secretion score associations varied by baseline eGFR <45 versus ≥45 mL/min/1.73 m2. As a sensitivity analysis, we used the Lunn-McNeil extension to the Cox model to account for the competing risk of death.
      • Lunn M.
      • McNeil D.
      Applying Cox regression to competing risks.
      All analyses were conducted using Stata (Stata Statistical Software, release 13; StataCorp LP) and SPSS (released 2016, IBM SPSS Statistics for Windows, Version 24.0, IBM Corp).

      Results

      Among the 2,089 SPRINT participants with baseline eGFR <60 mL/min/1.73 m2 included in this analysis, mean ± SD age was 73 ± 9 years, 41% were women, and median (interquartile range [IQR]) baseline eGFR and albuminuria were 48 mL/min/1.73m 2 (IQR, 38-55) and 15 mg/g (IQR, 7-48), respectively. Table 1 shows the baseline characteristics stratified by the summary secretion score quartiles. Compared to participants in the lowest secretion score quartile, those in the highest quartile had higher eGFR and lower albuminuria. Baseline systolic BP and diastolic BP, the proportion randomized to each treatment arm, and the number of antihypertensive medications were similar across quartiles. Compared with SPRINT participants with baseline eGFR <60 mL/min/1.73 m2 not included in this analysis, those included did not have significantly different baseline characteristics (P ≥ 0.10 for all).
      Table 1Baseline Characteristics of SPRINT Participants with Baseline CKD Stratified by Summary Secretion Score Quartiles
      CharacteristicQuartile 1 (N = 517)Quartile 2 (N = 530)Quartile 3 (N = 535)Quartile 4 (N = 507)All (N = 2089)
      Secretion score50 [46, 53]58 [56, 59]62 [61, 63]68 [66, 70]60 [55, 64]
      Age, y73 (10)74 (9)74 (9)72 (8)73 (9)
      Female238 (46)197 (37)201 (38)218 (43)854 (41)
      Race
       Non-Hispanic White333 (64)356 (67)384 (72)328 (65)1401 (67)
       African American137 (27)133 (25)108 (20)131 (26)509 (24)
       Hispanic and other47 (9)41 (8)43 (8)48 (10)179 (9)
      BMI, kg/m229.3 (5.8)29.3 (6.0)29.9 (5.9)30.0 (5.7)29.6 (5.9)
      Intensive BP arm266 (51)284 (54)281 (53)245 (48)1076 (52)
      Prevalent CVD or HF146 (28)173 (33)137 (26)132 (26)588 (28)
      Current smoker40 (8)55 (10)40 (8)45 (9)180 (9)
      eGFR, mL/min/1.73 m239 (12)44 (10)48 (9)51 (7)46 (11)
      Urine ACR, mg/g31 [10, 144]15 [8, 56]12 [6, 29]10 [6, 27]15 [7, 48]
      Systolic BP, mm Hg142 (17)139 (16)139 (16)138 (17)140 (16)
      Diastolic BP, mm Hg74 (12)74 (12)74 (12)75 (13)74 (12)
      No. of antihypertensive meds2.35 (1.07)2.19 (1.00)2.10 (1.00)2.03 (0.95)2.17 (1.01)
      Total cholesterol, mg/dL183 (41)182 (39)184 (41)185 (42)183 (41)
      HDL cholesterol, mg/dL53 (15)52 (15)52 (14)52 (14)52 (14)
      Triglycerides, mg/dL111 [79, 158]112 [80, 149]109 [81, 158]112 [85, 150]112 [82, 154]
      Statin use77 (15)66 (13)76 (14)72 (14)291 (14)
      Note: Data displayed are mean (standard deviation), n (%), or median (interquartile range).
      Abbreviations: ACR, albumin-to-creatinine ratio; BMI, body mass index; BP, blood pressure; CKD, chronic kidney disease; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate by creatinine and cystatin C; HDL, high-density lipoprotein cholesterol; HF, heart failure; SPRINT, Systolic Blood Pressure Intervention Trial.
      The mean ± SDsecretion score was 59.3 ± 7.6. The secretion score was moderately, positively correlated with baseline eGFR (r = 0.39) and inversely correlated with albuminuria (r = -0.30). The average UPR of each secretion marker varied significantly across baseline eGFR categories, such that the UPRs among participants with eGFR 45-59 mL/min/1.73 m2 were approximately 2 to 3 times higher than the UPRs among those with eGFR <30 mL/min/1.73 m2 (P < 0.001 for each tubular secretion marker, Table 2).
      Table 2Urine-to-Plasma Ratios of Tubular Secretion Markers Stratified by Baseline eGFR
      BiomarkereGFR, mL/min/1.73 m2P
      45-59 (N = 1238)30-44 (N = 641)<30 (N = 210)
      Adipic Acid47 [26, 76]35 [18, 64]22 [13, 36]< 0.001
      Cinnamoylglycine160 [101, 239]110 [70, 162]67 [42, 102]< 0.001
      p-Cresol sulfate18 [12, 26]13 [8, 19]7 [5, 11]< 0.001
      1,7-Dimethyluric acid264 [172, 385]195 [126, 295]110 [64, 175]< 0.001
      2-Furoylglycine406 [246, 644]195 [177, 461]160 [96, 277]< 0.001
      Hippuric acid378 [248, 555]280 [186, 415]170 [106, 264]< 0.001
      m-Hydroxy hippurate408 [265, 650]307 [188, 453]176 [102, 308]< 0.001
      Indoxyl sulfate49 [31, 72]34 [22, 50]18 [11, 29]< 0.001
      Phenylacetylglutamine268 [176, 383]190 [128, 279]106 [63, 169]< 0.001
      Tiglylglycine336 [221, 484]236 [156, 336]137 [81, 210]< 0.001
      Abbreviation: eGFR, estimated glomerular filtration rate by creatinine and cystatin C.
      The mean annualized eGFR slope during the median 3.3 years of follow-up was −1.44% per year (95% confidence interval [CI], −1.60% to −1.27%). Annualized eGFR decline was progressively faster from the highest to lowest secretion score quartile (Fig 1). In unadjusted analyses, higher UPRs of 9 of the 10 individual secretion markers were associated with faster eGFR decline (Table S2). After multivariable adjustment, higher UPRs of hippuric acid and m-hydroxy hippurate and lower UPR of adipic acid were associated with faster eGFR decline. Table 3 shows the association of the summary secretion score with change in eGFR. In multivariable models adjusting for demographics, clinical characteristics, intervention arm, baseline eGFR, and albuminuria, lower secretion score was significantly associated with faster annualized eGFR decline (−0.65% less per year per 1-SD lower secretion score; 95% CI, −0.84% to −0.46%). When analyzed using secretion score quartiles, the lowest quartile was significantly associated with faster eGFR decline compared with the highest quartile (−0.77% per year; 95% CI, −1.29% to −0.26%).
      Figure thumbnail gr1
      Figure 1Percent annualized change in eGFR in SPRINT participants with CKD stratified by summary secretion score quartiles. Bars represent unadjusted estimated annual change in eGFR with 95% CIs displayed. Estimates are derived from linear mixed-effect models. Abbreviations: CI, confidence interval; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; SPRINT, Systolic Blood Pressure Intervention Trial.
      Table 3Associations of Summary Secretion Score with Annualized eGFR Change and CKD Progression in Persons with CKD in SPRINT
      % Annualized eGFR ChangeMean eGFR Decline (%/y) (95% CI)Model 1
      Model 1 adjusts for baseline age, sex, race, and intervention arm.


      β
      β corresponds to the difference in annualized percentage change in eGFR.
      (95% CI)
      Model 2
      Model 2 adjusts for Model 1 + smoking, body mass index, systolic blood pressure, number of antihypertensive medications, prevalent cardiovascular disease, low-density lipoprotein cholesterol level, high-density lipoprotein cholesterol level, triglyceride level, statin use.


      β (95% CI)
      Model 3
      Model 3 adjusts for Model 2 + baseline eGFR and urine albumin-to-creatinine ratio.


      β (95% CI)
       Per 1-SD lower score−1.44 (−1.60 to −1.27)−0.72 (−0.88 to −0.55)−0.65 (−0.81 to −0.48)−0.65 (−0.84 to −0.46)
       Quartile 1
      Median (IQR) secretion score was 60 (55-64) overall, 50 (46-53) in quartile 1, 58 (56-59) in quartile 2, 62 (61-63) in quartile 3, and 68 (66-70) in quartile 4.
      −2.18 (−2.56 to −1.81)−1.22 (−1.68 to −0.76)−1.01 (−1.47 to −0.54)−0.77 (−1.29 to −0.26)
       Quartile 2−1.71 (−2.03 to −1.39)−0.64 (−1.11 to −0.18)−0.57 (−1.03 to −0.11)−0.39 (−0.87 to 0.08)
       Quartile 3−0.98 (−1.28 to −0.68)0.04 (−0.41 to 0.50)0.11 (−0.34 to 0.57)0.04 (−0.42 to 0.49)
       Quartile 4−0.97 (−1.26 to −0.69)ReferenceReferenceReference
      CKD progressionEvents/N (%)Model 1
      Model 1 adjusts for baseline age, sex, race, and intervention arm.


      HR (95% CI)
      Model 2
      Model 2 adjusts for Model 1 + smoking, body mass index, systolic blood pressure, number of antihypertensive medications, prevalent cardiovascular disease, low-density lipoprotein cholesterol level, high-density lipoprotein cholesterol level, triglyceride level, statin use.


      HR (95% CI)
      Model 3
      Model 3 adjusts for Model 2 + baseline eGFR and urine albumin-to-creatinine ratio.


      HR (95% CI)
       Per 1-SD lower score72/2089 (3.4%)1.89 (1.56-2.31)1.87 (1.54-2.28)1.23 (0.96-1.58)
       Quartile 134/517 (6.6%)4.18 (2.00-8.73)3.99 (1.90-8.38)0.98 (0.40-2.38)
       Quartile 217/530 (3.2%)1.99 (0.89-4.48)1.84 (0.81-4.15)0.96 (0.42-2.24)
       Quartile 312/535 (2.2%)1.46 (0.61-3.47)1.45 (0.61-3.46)1.11 (0.46-2.67)
       Quartile 49/507 (1.8%)ReferenceReferenceReference
      Abbreviations: CI, confidence interval; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; HR, hazard ratio; SD, standard deviation SPRINT, Systolic Blood Pressure Intervention Trial.
      a β corresponds to the difference in annualized percentage change in eGFR.
      b Model 1 adjusts for baseline age, sex, race, and intervention arm.
      c Model 2 adjusts for Model 1 + smoking, body mass index, systolic blood pressure, number of antihypertensive medications, prevalent cardiovascular disease, low-density lipoprotein cholesterol level, high-density lipoprotein cholesterol level, triglyceride level, statin use.
      d Model 3 adjusts for Model 2 + baseline eGFR and urine albumin-to-creatinine ratio.
      e Median (IQR) secretion score was 60 (55-64) overall, 50 (46-53) in quartile 1, 58 (56-59) in quartile 2, 62 (61-63) in quartile 3, and 68 (66-70) in quartile 4.
      In subgroup analyses, lower secretion score was more strongly associated with faster eGFR decline among participants with baseline eGFR <45 mL/min/1.73 m2 (−1.17% per year; 95% CI, −1.49% to −0.86%), compared with participants with baseline eGFR ≥45 mL/min/1.73 m2 (−0.07% per year; 95% CI, −0.28% to 0.14%; P for interaction < 0.001; Fig 2), independent of baseline eGFR and albuminuria. Secretion score associations with eGFR decline did not vary by randomized treatment arm (P for interaction = 0.59, Fig 2).
      Figure thumbnail gr2
      Figure 2Forest plot of summary secretion score associations with difference in annualized eGFR slope in SPRINT participants with CKD stratified by intervention arm and baseline eGFR. Beta coefficients with 95% confidence intervals correspond to the difference in percent annualized eGFR slope and were obtained from linear mixed-effect models. Hazard ratios (per 1-standard deviation lower secretion score) with 95% confidence intervals obtained from multivariable Cox proportional hazards models. All models included demographics (age, sex, and race), intervention arm, clinical characteristics (smoking, body mass index, systolic blood pressure, diastolic blood pressure, number of antihypertensive medications at baseline, prevalent cardiovascular disease, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, triglycerides, and statin use), baseline eGFR, and urine albumin-to-creatinine ratio. Abbreviations: CI, confidence interval; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; SPRINT, Systolic Blood Pressure Intervention Trial.
      The binary CKD progression endpoint occurred among 72 participants (3.4%). This endpoint occurred among 1.8% of participants in the highest secretion score quartile compared with 6.6% in the lowest quartile. In unadjusted analyses, higher UPRs of all 10 individual secretion markers were associated with greater risk of CKD progression (Table S2). After adjusting for clinical characteristics, eGFR, and albuminuria, only p-cresol sulfate and m-hydroxy hippurate remained associated with CKD progression. Lower secretion score (per 1-SD) was associated with greater risk of CKD progression in multivariable-adjusted analyses (hazard ratio [HR], 1.87; 95% CI, 1.54-2.28). However, after further adjustment for baseline eGFR and albuminuria, the association with CKD progression attenuated and was no longer statistically significant (HR, 1.23; 95% CI, 0.96-1.58). In quartile analyses, the secretion score relationship with CKD progression was relatively flat and did not reach statistical significance in fully adjusted models (Table 3).
      There were 272 participants who experienced the composite CVD event (13.0%), and 144 deaths (6.9%) in the study sample. More participants in the lowest secretion score quartile experienced a composite CVD event or death (16.1% and 8.9%, respectively), compared with the highest quartile (10.7% and 6.7%, respectively). In multivariable analyses, none of the individual secretion markers were associated with risk of CVD events or all-cause mortality (Table S3). In multivariable models that adjusted for demographics, clinical characteristics, and intervention arm, lower secretion score (per 1-SD) was independently associated with greater risk of the composite CVD event (HR, 1.20; 95% CI, 1.07-1.36) and all-cause mortality (HR, 1.18; 95% CI, 1.00-1.39). However, after further adjustment for baseline eGFR and albuminuria, secretion score associations with both endpoints were considerably attenuated and no longer statistically significant (Table 4). Across secretion score quartiles, the relationships between secretion score and CVD and mortality were relatively flat and did not reach statistical significance in the fully adjusted model. Secretion score associations with CKD progression, CVD, and mortality did not vary by baseline eGFR <45 versus ≥45 mL/min/1.73 m2 or by randomized treatment arm (P > 0.20 for all interaction tests; Fig 3).
      Table 4Associations of Composite Secretion Score with CVD Events and All-Cause Mortality in Persons with CKD in SPRINT
      Events/N (%)Model 1
      Model 1 adjusts for baseline age, sex, race, and intervention arm.
      HR (95% CI)
      Model 2
      Model 2 adjusts for Model 1 + smoking, body mass index, systolic blood pressure, number of antihypertensive medications, prevalent cardiovascular disease, low-density lipoprotein cholesterol level, high-density lipoprotein cholesterol level, triglyceride level, statin use.
      HR (95% CI)
      Model 3
      Model 3 adjusts for Model 2 + baseline eGFR and urine albumin-to-creatinine ratio.
      HR (95% CI)
      CVD events
       Per 1-SD lower score272/2089 (13.0%)1.23 (1.09-1.39)1.20 (1.07-1.36)1.02 (0.89-1.18)
       Quartile 1
      Median (IQR) secretion score was 60 (55-64) overall, 50 (46-53) in quartile 1, 58 (56-59) in quartile 2, 62 (61-63) in quartile 3, and 68 (66-70) in quartile 4.
      83/517 (16.1%)1.64 (1.16-2.31)1.57 (1.11-2.23)1.03 (0.70-1.52)
       Quartile 272/530 (13.6%)1.23 (0.86-1.75)1.17 (0.82-1.67)0.96 (0.66-1.38)
       Quartile 363/535 (11.8%)1.07 (0.74-1.53)1.07 (0.75-1.55)1.00 (0.69-1.44)
       Quartile 454/507 (10.7%)ReferenceReferenceReference
      All-cause mortality
       Per 1-SD lower score144/2089 (6.9%)1.22 (1.04-1.44)1.18 (1.00-1.39)0.90 (0.74-1.09)
       Quartile 146/517 (8.9%)1.42 (0.91-2.22)1.33 (0.85-2.09)0.66 (0.39-1.13)
       Quartile 238/530 (7.2%)1.03 (0.85-1.64)0.96 (0.60-1.53)0.68 (0.42-1.11)
       Quartile 326/535 (4.9%)0.71 (0.43-1.19)0.71 (0.43-1.19)0.63 (0.37-1.05)
       Quartile 434/507 (6.7%)ReferenceReferenceReference
      Abbreviations: CI, confidence interval; CKD, chronic kidney disease; CVD, cardiovascular disease; HR, hazard ratio; SD, standard deviation; SPRINT, Systolic Blood Pressure Intervention Trial.
      a Model 1 adjusts for baseline age, sex, race, and intervention arm.
      b Model 2 adjusts for Model 1 + smoking, body mass index, systolic blood pressure, number of antihypertensive medications, prevalent cardiovascular disease, low-density lipoprotein cholesterol level, high-density lipoprotein cholesterol level, triglyceride level, statin use.
      c Model 3 adjusts for Model 2 + baseline eGFR and urine albumin-to-creatinine ratio.
      d Median (IQR) secretion score was 60 (55-64) overall, 50 (46-53) in quartile 1, 58 (56-59) in quartile 2, 62 (61-63) in quartile 3, and 68 (66-70) in quartile 4.
      Figure thumbnail gr3
      Figure 3Forest plot of summary secretion score associations with risk of CKD progression, CVD, and all-cause mortality in SPRINT participants with CKD stratified by intervention arm and baseline eGFR. Hazard ratios (per 1-standard deviation lower secretion score) with 95% confidence intervals obtained from multivariable Cox proportional hazards models that included demographics (age, sex, and race), intervention arm, clinical characteristics (smoking, body mass index, systolic blood pressure, diastolic blood pressure, number of antihypertensive medications at baseline, prevalent cardiovascular disease, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, triglycerides, and statin use), baseline eGFR, and urine albumin-to-creatinine ratio. Abbreviations: CI, confidence interval; CKD, chronic kidney disease; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate; SPRINT, Systolic Blood Pressure Intervention Trial.
      Summary secretion score associations with CVD and CKD progression did not differ when accounting for the competing risk of death using Lunn-McNeil models.

      Discussion

      Tubular secretion is an essential kidney function, but its prognostic implications in persons with CKD have yet to be fully characterized. In this ancillary study of SPRINT participants with CKD at baseline, worse estimated tubular secretion was associated with faster eGFR decline independent of baseline eGFR and albuminuria. This association was stronger among the subgroup of participants with baseline eGFR <45 mL/min/1.73 m2. In contrast, estimated tubular secretion was not associated with risk of CVD or all-cause mortality independent of eGFR and albuminuria.
      Our findings are consistent with work using novel secretion measures in the Chronic Renal Insufficiency Cohort Study, Seattle Kidney Study, and the Modified Diet in Renal Disease Study. These studies found that worse estimated tubular secretion is associated with faster declines in eGFR longitudinally and greater risk of CKD progression, independent of baseline eGFR and albuminuria.
      • Chen Y.
      • Zelnick L.R.
      • Wang K.
      • et al.
      Kidney clearance of secretory solutes is associated with progression of CKD: the CRIC study.
      ,
      • Suchy-Dicey A.M.
      • Laha T.
      • Hoofnagle A.
      • et al.
      Tubular secretion in CKD.
      ,
      • Garimella P.S.
      • Tighiouart H.
      • Sarnak M.J.
      • Levey A.S.
      • Ix J.H.
      Tubular secretion of creatinine and risk of kidney failure: the modification of diet in renal disease (MDRD) study.
      Similar to our findings, studies of the Chronic Renal Insufficiency Cohort observed no association between estimated tubular secretion and CVD risk but found that worse estimated tubular secretion was independently associated with all-cause mortality.
      • Chen Y.
      • Zelnick L.R.
      • Wang K.
      • et al.
      Kidney clearance of secretory solutes is associated with progression of CKD: the CRIC study.
      ,
      • Chen Y.
      • Zelnick L.R.
      • Huber M.P.
      • et al.
      Association between kidney clearance of secretory solutes and cardiovascular events: the Chronic Renal Insufficiency Cohort (CRIC) study.
      In the Modified Diet in Renal Disease Study, worse secretion of creatinine was associated with kidney failure risk but not CVD or all-cause mortality.
      • Garimella P.S.
      • Tighiouart H.
      • Sarnak M.J.
      • Levey A.S.
      • Ix J.H.
      Tubular secretion of creatinine and risk of kidney failure: the modification of diet in renal disease (MDRD) study.
      An association between lower estimated tubular secretion and greater risk of CKD progression also appeared compatible with our data, although the finding did not reach statistical significance. This may have been because of insufficient power for this binary endpoint, which only occurred in 72 participants. SPRINT was designed as a CVD endpoint trial, and few CKD progression events accrued because the trial excluded individuals with baseline eGFR <20 mL/min/1.73 m2 or severe proteinuria, and there was a relatively short follow-up period. For this reason, we selected longitudinal eGFR decline as our primary endpoint a priori for this analysis. Collectively, these findings support and reaffirm previous findings that measurement of tubule secretion identifies individuals at higher risk of loss of kidney function, independent of eGFR and albuminuria, and that the association with loss of kidney function appears much stronger than that for CVD or all-cause mortality.
      We hypothesized that measures of tubular secretion could contribute additional prognostic information about kidney outcomes, CVD, and all-cause mortality above and beyond measurements of glomerular health for several reasons. First, recent work by our group demonstrated that lower estimated tubular secretion is associated with biopsy-proven interstitial fibrosis and tubular atrophy independent of eGFR and albuminuria,
      • Garimella P.S.
      • Katz R.
      • Waikar S.S.
      • et al.
      Kidney tubulointerstitial fibrosis and tubular secretion.
      and prior studies consistently demonstrated that interstitial fibrosis and tubular atrophy on biopsy are strongly prognostic of CKD progression.
      • Nath K.A.
      Tubulointerstitial changes as a major determinant in the progression of renal damage.
      ,
      • Rule A.D.
      • Amer H.
      • Cornell L.D.
      • et al.
      The association between age and nephrosclerosis on renal biopsy among healthy adults.
      Second, tubular secretion occurs primarily in the proximal tubules, and novel biomarkers reflecting proximal tubule damage and dysfunction are independently associated with longitudinal eGFR decline, CKD progression, CVD, and all-cause mortality in SPRINT CKD participants.
      • Garimella P.S.
      • Lee A.K.
      • Ambrosius W.T.
      • et al.
      Markers of kidney tubule function and risk of cardiovascular disease events and mortality in the SPRINT trial.
      ,
      • Jotwani V.K.
      • Lee A.K.
      • Estrella M.M.
      • et al.
      Urinary biomarkers of tubular damage are associated with mortality but not cardiovascular risk among systolic blood pressure intervention trial participants with chronic kidney disease.
      ,
      • Jotwani V.
      • Garimella P.S.
      • Katz R.
      • et al.
      Tubular biomarkers and chronic kidney disease progression in SPRINT participants.
      ,
      • Malhotra R.
      • Katz R.
      • Jotwani V.
      • et al.
      Urine markers of kidney tubule cell injury and kidney function decline in SPRINT trial participants with CKD.
      Third, lower tubular secretion suggests a reduced ability to clear uremic toxins, which have been implicated as a pathophysiologic driver of vascular calcification, endothelial dysfunction, inflammation, and fibrosis.
      • Duranton F.
      • Cohen G.
      • De Smet R.
      • et al.
      Normal and pathologic concentrations of uremic toxins.
      ,
      • Ravid J.D.
      • Kamel M.H.
      • Chitalia V.C.
      Uraemic solutes as therapeutic targets in CKD-associated cardiovascular disease.
      We found that the relationship between lower estimated tubular secretion and faster eGFR decline was stronger in those with baseline eGFR < 45 mL/min/1.73 m2 relative to those with milder CKD. Previous studies have shown that even though estimated tubular secretion can vary considerably at any given level of glomerular filtration rate, average secretory function decreases in parallel with declining filtration function.
      • Suchy-Dicey A.M.
      • Laha T.
      • Hoofnagle A.
      • et al.
      Tubular secretion in CKD.
      ,
      • Chen Y.
      • Zelnick L.R.
      • Wang K.
      • et al.
      Association of tubular solute clearances with the glomerular filtration rate and complications of chronic kidney disease: the chronic renal insufficiency cohort study.
      Similarly, we observed progressively lower estimated tubular secretion across lower eGFR categories uniformly for each of the tubular secretion markers. We hypothesize that tubular secretion may become increasingly important to maintain homeostasis and carry out the kidney’s important biological functions as glomerular filtration rate declines. Further investigations are warranted to better understand how novel measures of tubular secretion relate to adverse outcomes at different CKD stages.
      As an ancillary study of SPRINT, this analysis benefited from the inclusion of a large CKD population, frequent and protocol driven eGFR assessments during follow-up, and clinically adjudicated outcomes. We also used a broad panel of candidate secretory solutes, and were able to quantify the concentrations of 10 metabolites from a single liquid chromatography-tandem mass spectrometry run. Future analyses could evaluate whether fewer markers may achieve effective characterization of estimated tubular secretory function. There are also important limitations. First, because of the SPRINT design, our findings may not generalize to persons with CKD who have severe proteinuria, eGFR <20 mL/min/1.73 m2, or diabetes mellitus. In addition, the summary secretion score was internally derived from tubular secretion measures from SPRINT participants. However, the distribution of summary secretion scores was similar to the distribution of scores observed in Chronic Renal Insufficiency Cohort Study using similar tubular secretion markers, suggesting the present study captures a common range of estimated tubular secretory function. In addition, our findings were similar when we evaluated the secretion markers individually. Second, because direct measurements of tubular secretion are not available, we estimated tubular secretory function using relative urine-to-plasma concentrations of endogenous secretory solutes that are cleared primarily by tubular secretion.
      • Sirich T.L.
      • Aronov P.A.
      • Plummer N.S.
      • Hostetter T.H.
      • Meyer T.W.
      Numerous protein-bound solutes are cleared by the kidney with high efficiency.
      ,
      • Suchy-Dicey A.M.
      • Laha T.
      • Hoofnagle A.
      • et al.
      Tubular secretion in CKD.
      We previously demonstrated that spot urine-to-plasma concentrations of the secreted solutes are associated with interstitial fibrosis and tubular atrophy severity on kidney biopsy.
      • Garimella P.S.
      • Katz R.
      • Waikar S.S.
      • et al.
      Kidney tubulointerstitial fibrosis and tubular secretion.
      Third, stored urine specimens in SPRINT were spot samples, and the secretion of individual solutes may be subject to intraindividual variability.
      • Rivara M.B.
      • Zelnick L.R.
      • Hoofnagle A.N.
      • et al.
      Diurnal and long-term variation in plasma concentrations and renal clearances of circulating markers of kidney proximal tubular secretion.
      However, any misclassification due to missed variability would have biased our results toward the null. Future studies evaluating how using spot urine measurements to estimate tubular secretion compare with 24-hour urine collections are needed. Finally, we did not have information on certain medications such as antibiotics or antacids that may affect tubular secretion.
      In summary, among persons with hypertension and CKD, lower estimated tubular secretion was associated with faster eGFR decline, independent of baseline eGFR and albuminuria. This relationship was stronger in persons with more advanced CKD at baseline. In contrast, lower estimated tubular secretion was not independently associated with a binary CKD progression endpoint, CVD composite endpoint, or all-cause mortality after adjusting for baseline eGFR and albuminuria. Overall, our findings suggest that a broader assessment of kidney health that incorporates estimates of tubular secretion may provide additional information about subsequent risk of kidney function decline but contributes little additional insight about CVD risk or survival. Additional studies are needed to validate these findings in other CKD cohorts, to determine the importance of monitoring tubular secretory capacity for predicting long-term changes in kidney function and for assessing treatment efficacy and safety, and to investigate the prognostic contributions of tubular secretion measures for other adverse outcomes in persons with CKD.

      Article Information

      Authors’ Full Names and Academic Degrees

      Simon B. Ascher, MD, MPH, Michael G. Shlipak, MD, MPH, Ronit Katz, DPhil, Alexander L. Bullen, MD, MAS, Rebecca Scherzer, PhD, Stein I. Hallan, MD, PhD, Alfred K. Cheung, MD, Kalani L. Raphael, MD, MS, Michelle M. Estrella, MD, MHS, Vasantha K. Jotwani, MD, Jesse C. Seegmiller, PhD, Joachim H. Ix, MD, MAS, and Pranav S. Garimella, MBBS, MPH

      Authors’ Contributions

      Research idea and study design: SBA, MGS, RK, JHI, PSG; data acquisition: MGS, JHI; data analysis: RK; data interpretation: SBA, MGS, RK, ALB, RS, SIH, AKC, KLR, MME, VKJ, JCS, JHI, PSG; supervision or mentorship: MGS, JHI, PSG. Co-senior authors: JHI, PSG. Each author contributed important intellectual content during manuscript drafting or revision and accepts accountability for the overall work by ensuring that questions pertaining to the accuracy or integrity of any portion of the work are appropriately investigated and resolved.

      Support

      This ancillary study was supported by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) ( R01DK098234 for MGS/JHI and K24DK110427 for JHI). PSG was supported by a career development grant from the NIDDK ( K23DK114556 ). The funders had no role in study design, data collection, analysis, reporting, or the decision to submit for publication. The authors thank the participants and staff members of the Systolic Blood Pressure Intervention Trial, which sponsored by the National Institutes of Health (NIH), including the National Heart, Lung, and Blood Institute (NHLBI), the NIDDK, the National Institute on Aging (NIA), and the National Institute of Neurological Disorders and Stroke ( NINDS ), under Contract Numbers HHSN268200900040C , HHSN268200900046C , HHSN268200900047C , HHSN268200900048C , HHSN268200900049C , and Inter-Agency Agreement Number A-HL-13–002-001. It was also supported in part with resources and use of facilities through the Department of Veterans Affairs. The SPRINT investigators acknowledge the contribution of study medications (azilsartan and azilsartan combined with chlorthalidone) from Takeda Pharmaceuticals International, Inc. All components of the SPRINT study protocol were designed and implemented by the investigators. The investigative team collected, analyzed, and interpreted the data. All aspects of manuscript writing and revision were carried out by the coauthors. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH, the US Department of Veterans Affairs, or the US Government. For a full list of contributors to SPRINT, please see the supplementary acknowledgement list: www.sprinttrial.org/public/dspScience.cfm. We also acknowledge the support from the following CTSAs funded by NCATS: CWRU: UL1TR000439, OSU: UL1RR025755, U Penn: UL1RR024134& UL1TR000003, Boston: UL1RR025771, Stanford: UL1TR000093, Tufts: UL1RR025752, UL1TR000073 & UL1TR001064, University of Illinois: UL1TR000050, University of Pittsburgh: UL1TR000005, UT Southwestern: 9U54TR000017–06, University of Utah: UL1TR000105–05, Vanderbilt University: UL1 TR000445, George Washington University: UL1TR000075, University of CA, Davis: UL1 TR000002, University of Florida: UL1 TR000064, University of Michigan: UL1TR000433, Tulane University: P30GM103337 COBRE Award NIGMS, Wake Forest University: UL1TR001420.

      Financial Disclosure

      MGS has received consulting income from Cricket Health, Inc and Intercept Pharmaceuticals. JHI holds an investigator initiated research grant from Baxter International Inc, serves as a member of a data safety monitoring board for Sanifit Therapeutics, is a member of the scientific advisory board for Alpha Young, and has served on advisory boards for AstraZeneca and Ardelyx. The remaining authors declare that they have no relevant financial interests.

      Prior Presentation

      The results presented in this paper have not been published previously in whole or part, except in abstract format.

      Peer Review

      Received June 7, 2022 as a submission to the expedited consideration track with 3 external peer reviews. Direct editorial input from the Statistical Editor and the Editor-in-Chief. Accepted in revised form August 28, 2022.

      Supplementary Material

      Table S1: Laboratory Characteristics of Endogenous Secretory Solutes Measured in SPRINT Participants with CKD
      Table S2: Associations of Individual Tubular Secretion Markers with Annualized eGFR Change and CKD Progression in Persons with CKD in SPRINT
      Table S3: Associations of Individual Tubular Secretion Markers with CVD Events and All-Cause Mortality in Persons with CKD in SPRINT

      References

        • GBD Chronic Kidney Disease Collaboration
        Global, regional, and national burden of chronic kidney disease, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017.
        Lancet. 2020; 395: 709-733https://doi.org/10.1016/S0140-6736(20)30045-3
        • Go A.S.
        • Chertow G.M.
        • Fan D.
        • McCulloch C.E.
        • Hsu C.Y.
        Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization.
        N Engl J Med. 2004; 351: 1296-1305https://doi.org/10.1056/NEJMoa041031
        • Gansevoort R.T.
        • Matsushita K.
        • van der Velde M.
        • et al.
        Lower estimated GFR and higher albuminuria are associated with adverse kidney outcomes. A collaborative meta-analysis of general and high-risk population cohorts.
        Kidney Int. 2011; 80: 93-104https://doi.org/10.1038/ki.2010.531
        • van der Velde M.
        • Matsushita K.
        • Coresh J.
        • et al.
        Lower estimated glomerular filtration rate and higher albuminuria are associated with all-cause and cardiovascular mortality. A collaborative meta-analysis of high-risk population cohorts.
        Kidney Int. 2011; 79: 1341-1352https://doi.org/10.1038/ki.2010.536
        • Astor B.C.
        • Matsushita K.
        • Gansevoort R.T.
        • et al.
        Lower estimated glomerular filtration rate and higher albuminuria are associated with mortality and end-stage renal disease. A collaborative meta-analysis of kidney disease population cohorts.
        Kidney Int. 2011; 79: 1331-1340https://doi.org/10.1038/ki.2010.550
        • Levin A.
        • Stevens P.E.
        • Bilous R.W.
        • et al.
        • Kidney Disease: improving global outcomes (KDIGO) CKD work group
        KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease.
        Kidney Int Suppl. 2013; 3: 1-150
        • Bohle A.
        • Christ H.
        • Grund K.E.
        • Mackensen S.
        The role of the interstitium of the renal cortex in renal disease.
        Contrib Nephrol. 1979; 16: 109-114https://doi.org/10.1159/000402883
        • Nath K.A.
        Tubulointerstitial changes as a major determinant in the progression of renal damage.
        Am J Kidney Dis. 1992; 20: 1-17
        • Eddy A.A.
        Molecular insights into renal interstitial fibrosis.
        J Am Soc Nephrol. 1996; 7: 2495-2508https://doi.org/10.1681/ASN.V7122495
        • Rule A.D.
        • Amer H.
        • Cornell L.D.
        • et al.
        The association between age and nephrosclerosis on renal biopsy among healthy adults.
        Ann Intern Med. 2010; 152: 561-567https://doi.org/10.7326/0003-4819-152-9-201005040-00006
        • Garimella P.S.
        • Lee A.K.
        • Ambrosius W.T.
        • et al.
        Markers of kidney tubule function and risk of cardiovascular disease events and mortality in the SPRINT trial.
        Eur Heart J. 2019; 40: 3486-3493https://doi.org/10.1093/eurheartj/ehz392
        • Jotwani V.K.
        • Lee A.K.
        • Estrella M.M.
        • et al.
        Urinary biomarkers of tubular damage are associated with mortality but not cardiovascular risk among systolic blood pressure intervention trial participants with chronic kidney disease.
        Am J Nephrol. 2019; 49: 346-355https://doi.org/10.1159/000499531
        • Bullen A.L.
        • Katz R.
        • Lee A.K.
        • et al.
        The SPRINT trial suggests that markers of tubule cell function in the urine associate with risk of subsequent acute kidney injury while injury markers elevate after the injury.
        Kidney Int. 2019; 96: 470-479https://doi.org/10.1016/j.kint.2019.03.024
        • Jotwani V.
        • Garimella P.S.
        • Katz R.
        • et al.
        Tubular biomarkers and chronic kidney disease progression in SPRINT participants.
        Am J Nephrol. 2020; 51: 797-805https://doi.org/10.1159/000509978
        • Malhotra R.
        • Katz R.
        • Jotwani V.
        • et al.
        Urine markers of kidney tubule cell injury and kidney function decline in SPRINT trial participants with CKD.
        Clin J Am Soc Nephrol. 2020; 15: 349-358https://doi.org/10.2215/CJN.02780319
        • Ix J.H.
        • Shlipak M.G.
        The promise of tubule biomarkers in kidney disease: a review.
        Am J Kidney Dis. 2021; 78: 719-727https://doi.org/10.1053/j.ajkd.2021.03.026
        • Nigam S.K.
        • Wu W.
        • Bush K.T.
        • Hoenig M.P.
        • Blantz R.C.
        • Bhatnagar V.
        Handling of drugs, metabolites, and uremic toxins by kidney proximal tubule drug transporters.
        Clin J Am Soc Nephrol. 2015; 10: 2039-2049https://doi.org/10.2215/CJN.02440314
        • Wang K.
        • Kestenbaum B.
        Proximal tubular secretory clearance: a neglected partner of kidney function.
        Clin J Am Soc Nephrol. 2018; 13: 1291-1296https://doi.org/10.2215/CJN.12001017
        • Marshall E.K.
        Two lectures on renal physiology.
        Physiologist. 1966; 9: 367-384
        • Chen Y.
        • Zelnick L.R.
        • Wang K.
        • et al.
        Kidney clearance of secretory solutes is associated with progression of CKD: the CRIC study.
        J Am Soc Nephrol. 2020; 31: 817-827https://doi.org/10.1681/ASN.2019080811
        • Chen Y.
        • Zelnick L.R.
        • Huber M.P.
        • et al.
        Association between kidney clearance of secretory solutes and cardiovascular events: the Chronic Renal Insufficiency Cohort (CRIC) study.
        Am J Kidney Dis. 2021; 78: 226-235.e1https://doi.org/10.1053/j.ajkd.2020.12.005
        • Sirich T.L.
        • Aronov P.A.
        • Plummer N.S.
        • Hostetter T.H.
        • Meyer T.W.
        Numerous protein-bound solutes are cleared by the kidney with high efficiency.
        Kidney Int. 2013; 84: 585-590https://doi.org/10.1038/ki.2013.154
        • Suchy-Dicey A.M.
        • Laha T.
        • Hoofnagle A.
        • et al.
        Tubular secretion in CKD.
        J Am Soc Nephrol. 2016; 27: 2148-2155https://doi.org/10.1681/ASN.2014121193
        • Wright J.T.
        • Williamson J.D.
        • et al.
        • SPRINT Research Group
        A randomized trial of intensive versus standard blood-pressure control.
        N Engl J Med. 2015; 373: 2103-2116https://doi.org/10.1056/NEJMoa1511939
        • Ambrosius W.T.
        • Sink K.M.
        • Foy C.G.
        • et al.
        The design and rationale of a multicenter clinical trial comparing two strategies for control of systolic blood pressure: the Systolic Blood Pressure Intervention Trial (SPRINT).
        Clin Trials. 2014; 11: 532-546https://doi.org/10.1177/1740774514537404
        • Inker L.A.
        • Schmid C.H.
        • Tighiouart H.
        • et al.
        Estimating glomerular filtration rate from serum creatinine and cystatin C.
        N Engl J Med. 2012; 367: 20-29https://doi.org/10.1056/NEJMoa1114248
        • Seegmiller J.C.
        • Wolfe B.J.
        • Albtoush N.
        • et al.
        Tubular secretion markers, glomerular filtration rate, effective renal plasma flow, and filtration fraction in healthy adolescents.
        Kidney Med. 2020; 2: 670-672https://doi.org/10.1016/j.xkme.2020.05.013
        • Garimella P.S.
        • Li K.
        • Naviaux J.C.
        • et al.
        Utility of spot urine specimens to assess tubular secretion.
        Am J Kidney Dis. 2017; 69: 709-711https://doi.org/10.1053/j.ajkd.2016.12.016
        • Garimella P.S.
        • Katz R.
        • Waikar S.S.
        • et al.
        Kidney tubulointerstitial fibrosis and tubular secretion.
        Am J Kidney Dis. 2022; 79: 709-716https://doi.org/10.1053/j.ajkd.2021.08.015
        • Levey A.S.
        • Stevens L.A.
        • Schmid C.H.
        • et al.
        A new equation to estimate glomerular filtration rate.
        Ann Intern Med. 2009; 150: 604-612
        • Johnson K.C.
        • Whelton P.K.
        • Cushman W.C.
        • et al.
        Blood pressure measurement in SPRINT (Systolic Blood Pressure Intervention Trial).
        Hypertension. 2018; 71: 848-857https://doi.org/10.1161/HYPERTENSIONAHA.117.10479
        • Cheung A.K.
        • Rahman M.
        • Reboussin D.M.
        • et al.
        Effects of intensive BP control in CKD.
        J Am Soc Nephrol. 2017; 28: 2812-2823https://doi.org/10.1681/ASN.2017020148
        • Lunn M.
        • McNeil D.
        Applying Cox regression to competing risks.
        Biometrics. 1995; 51: 524-532
        • Garimella P.S.
        • Tighiouart H.
        • Sarnak M.J.
        • Levey A.S.
        • Ix J.H.
        Tubular secretion of creatinine and risk of kidney failure: the modification of diet in renal disease (MDRD) study.
        Am J Kidney Dis. 2021; 77: 992-994https://doi.org/10.1053/j.ajkd.2020.09.017
        • Duranton F.
        • Cohen G.
        • De Smet R.
        • et al.
        Normal and pathologic concentrations of uremic toxins.
        J Am Soc Nephrol. 2012; 23: 1258-1270https://doi.org/10.1681/ASN.2011121175
        • Ravid J.D.
        • Kamel M.H.
        • Chitalia V.C.
        Uraemic solutes as therapeutic targets in CKD-associated cardiovascular disease.
        Nat Rev Nephrol. 2021; 17: 402-416https://doi.org/10.1038/s41581-021-00408-4
        • Chen Y.
        • Zelnick L.R.
        • Wang K.
        • et al.
        Association of tubular solute clearances with the glomerular filtration rate and complications of chronic kidney disease: the chronic renal insufficiency cohort study.
        Nephrol Dial Transplant. 2020; 36: 1271-1281https://doi.org/10.1093/ndt/gfaa057
        • Rivara M.B.
        • Zelnick L.R.
        • Hoofnagle A.N.
        • et al.
        Diurnal and long-term variation in plasma concentrations and renal clearances of circulating markers of kidney proximal tubular secretion.
        Clin Chem. 2017; 63: 915-923https://doi.org/10.1373/clinchem.2016.260117