Title: Results from the German Chronic Kidney Disease (GCKD) study support association of relative telomere length with mortality in a large cohort of patients with moderate chronic kidney disease
Abstract: Telomere length is known to be inversely associated with aging and has been proposed as a marker for aging-related diseases. Telomere attrition can be accelerated by oxidative stress and inflammation, both commonly present in patients with chronic kidney disease. Here, we investigated whether relative telomere length is associated with mortality in a large cohort of patients with chronic kidney disease stage G3 and A1-3 or G1-2 with overt proteinuria (A3) at enrollment. Relative telomere length was quantified in peripheral blood by a quantitative PCR method in 4,955 patients from the GCKD study, an ongoing prospective observational cohort. Complete four-year follow-up was available from 4,926 patients in whom we recorded 354 deaths. Relative telomere length was a strong and independent predictor of all-cause mortality. Each decrease of 0.1 relative telomere length unit was highly associated with a 14% increased risk of death (hazard ratio1.14 [95% confidence interval 1.06-1.22]) in a model adjusted for age, sex, baseline eGFR, urine albumin/creatinine ratio, diabetes mellitus, prevalent cardiovascular disease, LDL-cholesterol, HDL-cholesterol, smoking, body mass index, systolic and diastolic blood pressure, C-reactive protein and serum albumin. This translated to a 75% higher risk for those in the lowest compared to the highest quartile of relative telomere length. The association was mainly driven by 117 cardiovascular deaths (1.20 [1.05-1.35]) as well as 67 deaths due to infections (1.27 [1.07-1.50]). Thus, our findings support an association of shorter telomere length with all-cause mortality, cardiovascular mortality and death due to infections in patients with moderate chronic kidney disease. Telomere length is known to be inversely associated with aging and has been proposed as a marker for aging-related diseases. Telomere attrition can be accelerated by oxidative stress and inflammation, both commonly present in patients with chronic kidney disease. Here, we investigated whether relative telomere length is associated with mortality in a large cohort of patients with chronic kidney disease stage G3 and A1-3 or G1-2 with overt proteinuria (A3) at enrollment. Relative telomere length was quantified in peripheral blood by a quantitative PCR method in 4,955 patients from the GCKD study, an ongoing prospective observational cohort. Complete four-year follow-up was available from 4,926 patients in whom we recorded 354 deaths. Relative telomere length was a strong and independent predictor of all-cause mortality. Each decrease of 0.1 relative telomere length unit was highly associated with a 14% increased risk of death (hazard ratio1.14 [95% confidence interval 1.06-1.22]) in a model adjusted for age, sex, baseline eGFR, urine albumin/creatinine ratio, diabetes mellitus, prevalent cardiovascular disease, LDL-cholesterol, HDL-cholesterol, smoking, body mass index, systolic and diastolic blood pressure, C-reactive protein and serum albumin. This translated to a 75% higher risk for those in the lowest compared to the highest quartile of relative telomere length. The association was mainly driven by 117 cardiovascular deaths (1.20 [1.05-1.35]) as well as 67 deaths due to infections (1.27 [1.07-1.50]). Thus, our findings support an association of shorter telomere length with all-cause mortality, cardiovascular mortality and death due to infections in patients with moderate chronic kidney disease. Telomeres are non-coding, repetitive nucleotide sequences (TTAGGG) ranging from 5 to 15 kilobase pairs in length that are located at the end of eukaryotic chromosomes.1Moyzis R.K. Buckingham J.M. Cram L.S. et al.A highly conserved repetitive DNA sequence, (TTAGGG)(n), present at the telomeres of human chromosomes.Proc Natl Acad Sci U S A. 1988; 85: 6622-6626Crossref PubMed Scopus (1867) Google Scholar Their functions include protection of the DNA and maintenance of chromosomal integrity. Telomeres shorten at each cycle of cell division due to the incapacity of DNA polymerase to replicate the very ends of linear chromosomes.2Baird D.M. Telomere dynamics in human cells.Biochimie. 2008; 90: 116-121Crossref PubMed Scopus (44) Google Scholar Approximately 50–200 base pairs are lost during each cell division, and when a critical telomere length is reached, cells undergo replicative senescence or apoptosis. Consequently, telomere length (TL) has been proposed as a marker of biological age,3von Zglinicki T. Martin-Ruiz C.M. Telomeres as biomarkers for ageing and age-related diseases.Curr Mol Med. 2005; 5: 197-203Crossref PubMed Scopus (320) Google Scholar and its predictive role in aging-related disease has been investigated in many epidemiologic studies.4Forero D.A. González-Giraldo Y. López-Quintero C. et al.Telomere length in Parkinson's disease: a meta-analysis.Exp Gerontol. 2016; 75: 53-55Crossref PubMed Scopus (46) Google Scholar, 5Mons U. Müezzinler A. Schöttker B. et al.Leukocyte telomere length and all-cause, cardiovascular disease, and cancer mortality: results from individual-participant-data meta-analysis of 2 large prospective cohort studies.Am J Epidemiol. 2017; 185: 1317-1326Crossref PubMed Scopus (77) Google Scholar, 6Wang Q. Zhan Y. Pedersen N.L. et al.Telomere length and all-cause mortality: a meta-analysis.Ageing Res Rev. 2018; 48: 11-20Crossref PubMed Scopus (137) Google Scholar Telomere attrition, accelerated by oxidative stress and inflammation, leads to cell senescence, which compromises regeneration and functionality of vital organs, including the kidneys.7Wills L.P. Schnellmann R.G. Telomeres and telomerase in renal health.J Am Soc Nephrol. 2011; 22: 39-41Crossref PubMed Scopus (36) Google Scholar In particular, it has been shown that chronic inflammation leads to lymphocyte telomere attrition, cell senescence, and finally impairment of the immune response.8Kordinas V. Tsirpanlis G. Nicolaou C. et al.Is there a connection between inflammation, telomerase activity and the transcriptional status of telomerase reverse transcriptase in renal failure?.Cell Mol Biol Lett. 2015; 20: 222-236Crossref PubMed Scopus (6) Google Scholar This T-cell dysfunction can contribute to increased susceptibility to kidney infections and injury.7Wills L.P. Schnellmann R.G. Telomeres and telomerase in renal health.J Am Soc Nephrol. 2011; 22: 39-41Crossref PubMed Scopus (36) Google Scholar Chronic kidney disease (CKD) is a complex disease, and its heritability has been estimated to be 30%–70%.9Cañadas-Garre M. Anderson K. Cappa R. et al.Genetic susceptibility to chronic kidney disease—some more pieces for the heritability puzzle.Front Genet. 2019; 10: 453Crossref PubMed Scopus (40) Google Scholar, 10Satko S.G. Freedman B.I. The familial clustering of renal disease and related phenotypes.Med Clin North Am. 2005; 89: 447-456Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar, 11Regele F. Jelencsics K. Shiffman D. et al.Genome-wide studies to identify risk factors for kidney disease with a focus on patients with diabetes.Nephrol Dial Transplant. 2015; 30: iv26-iv34Crossref PubMed Scopus (40) Google Scholar, 12Wuttke M. Köttgen A. Insights into kidney diseases from genome-wide association studies.Nat Rev Nephrol. 2016; 12: 549-562Crossref PubMed Scopus (71) Google Scholar In past years, genome-wide association studies have identified many genetic loci associated with kidney function and CKD.13Böger C.A. Gorski M. Li M. et al.Association of eGFR-related loci identified by GWAS with incident CKD and ESRD.PLoS Genet. 2011; 7: e1002292Crossref PubMed Scopus (153) Google Scholar, 14Chambers J.C. Zhang W. Lord G.M. et al.Genetic loci influencing kidney function and chronic kidney disease.Nat Genet. 2010; 42: 373-375Crossref PubMed Scopus (219) Google Scholar, 15Köttgen A. Pattaro C. Böger C.A. et al.New loci associated with kidney function and chronic kidney disease.Nat Genet. 2010; 42: 376-384Crossref PubMed Scopus (626) Google Scholar, 16Pattaro C. Teumer A. Gorski M. et al.Genetic associations at 53 loci highlight cell types and biological pathways relevant for kidney function.Nat Commun. 2016; 7: 10023Crossref PubMed Scopus (300) Google Scholar, 17Wuttke M. Li Y. Li M. et al.A catalog of genetic loci associated with kidney function from analyses of a million individuals.Nat Genet. 2019; 51: 957-972Crossref PubMed Scopus (270) Google Scholar However, index single nucleotide polymorphisms at the identified loci explain only a minor part of the heritability, and additional genetic contributors might be missing. To date, only a few small studies have investigated the association between TL and kidney disease. Some studies found that short TL correlates with impaired kidney function in the general population,18Bansal N. Whooley M.A. Regan M. et al.Association between kidney function and telomere length: the heart and soul study.Am J Nephrol. 2012; 36: 405-411Crossref PubMed Scopus (28) Google Scholar,19Eguchi K. Honig L.S. Lee J.H. et al.Short telomere length is associated with renal impairment in Japanese subjects with cardiovascular risk.PLoS One. 2017; 12e0176138Crossref PubMed Scopus (11) Google Scholar as well as in heart failure patients.20Wong L.S.M. Van Der Harst P. De Boer R.A. et al.Renal dysfunction is associated with shorter telomere length in heart failure.Clin Res Cardiol. 2009; 98: 629-634Crossref PubMed Scopus (29) Google Scholar We recently described significantly shorter relative TL (RTL) in patients with moderately severe CKD who have prevalent cardiovascular disease (CVD),21Raschenberger J. Kollerits B. Titze S. et al.Association of relative telomere length with cardiovascular disease in a large chronic kidney disease cohort: the GCKD study.Atherosclerosis. 2015; 242: 529-534Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar as well as an association with duration22Raschenberger J. Kollerits B. Titze S. et al.Do telomeres have a higher plasticity than thought? Results from the German Chronic Kidney Disease (GCKD) study as a high-risk population.Exp Gerontol. 2015; 72: 162-166Crossref PubMed Scopus (16) Google Scholar and progression of CKD.23Raschenberger J. Kollerits B. Ritchie J. et al.Association of relative telomere length with progression of chronic kidney disease in two cohorts: effect modification by smoking and diabetes.Sci Rep. 2015; 5: 1-8Crossref Scopus (41) Google Scholar Patients who have reached kidney failure treated by hemodialysis are described as having reduced TL in comparison with healthy controls,24Betjes M.G.H. Langerak A.W. Van Der Spek A. et al.Premature aging of circulating T cells in patients with end-stage renal disease.Kidney Int. 2011; 80: 208-217Abstract Full Text Full Text PDF PubMed Scopus (141) Google Scholar, 25Hirashio S. Nakashima A. Doi S. et al.Telomeric G-tail length and hospitalization for cardiovascular events in hemodialysis patients.Clin J Am Soc Nephrol. 2014; 9: 2117-2122Crossref PubMed Scopus (14) Google Scholar, 26Ramírez R. Carracedo J. Soriano S. et al.Stress-induced premature senescence in mononuclear cells from patients on long-term hemodialysis.Am J Kidney Dis. 2005; 45: 353-359Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar, 27Tsirpanlis G. Chatzipanagiotou S. Boufidou F. et al.Telomerase activity is decreased in peripheral blood mononuclear cells of hemodialysis patients.Am J Nephrol. 2006; 26: 91-96Crossref PubMed Scopus (38) Google Scholar and reduced TL is inversely associated with mortality.28Carrero J.J. Stenvinkel P. Fellström B. et al.Telomere attrition is associated with inflammation, low fetuin-A levels and high mortality in prevalent haemodialysis patients.J Intern Med. 2008; 263: 302-312Crossref PubMed Scopus (155) Google Scholar Only a few investigations have been conducted in non-dialysis-dependent kidney patients.21Raschenberger J. Kollerits B. Titze S. et al.Association of relative telomere length with cardiovascular disease in a large chronic kidney disease cohort: the GCKD study.Atherosclerosis. 2015; 242: 529-534Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar, 22Raschenberger J. Kollerits B. Titze S. et al.Do telomeres have a higher plasticity than thought? Results from the German Chronic Kidney Disease (GCKD) study as a high-risk population.Exp Gerontol. 2015; 72: 162-166Crossref PubMed Scopus (16) Google Scholar, 23Raschenberger J. Kollerits B. Ritchie J. et al.Association of relative telomere length with progression of chronic kidney disease in two cohorts: effect modification by smoking and diabetes.Sci Rep. 2015; 5: 1-8Crossref Scopus (41) Google Scholar,29Mazidi M. Rezaie P. Covic A. et al.Telomere attrition, kidney function, and prevalent chronic kidney disease in the United States.Oncotarget. 2017; 8: 80175-80181Crossref PubMed Scopus (27) Google Scholar,30Kidir V. Aynali A. Altuntas A. et al.Telomerase activity in patients with stage 2–5D chronic kidney disease.Nefrologia. 2017; 37: 592-597Crossref PubMed Scopus (7) Google Scholar To our knowledge, the current study is the first prospective study that investigates the association between leukocyte RTL and causes of mortality in a large non-dialysis-dependent CKD cohort. RTL was quantified in peripheral blood by a quantitative polymerase chain reaction method in 4955 patients from the German Chronic Kidney Disease study. Complete 4-year follow-up was available from 4926 patients. Baseline characteristics of these 4926 patients according to quartiles of the RTL are provided in Table 1. RTL ranged from a minimum of 0.40 to a maximum of 2.31 (Supplementary Figure S1), with a mean ± SD of 0.95 ± 0.19 and a median of 0.92 (1st quartile = 0.82; 3rd quartile = 1.05). RTL was negatively correlated with age (r = –0.36, P < 0.001) and positively correlated with estimated glomerular filtration rate (eGFR; r = 0.17, P < 0.001) and urine albumin–creatinine ratio (r = 0.05, P < 0.001). When we adjusted RTL for age and sex, we no longer observed a significant correlation with eGFR and urine albumin–creatinine ratio.Table 1Characteristics of all patients available (n = 4926) for analysis grouped by relative telomere length quartilesRTL quartilesTotalQuartile 1Quartile 2Quartile 3Quartile 4P valueRTL rangeRTL mean ± SDRTL median [25th, 75th percentile]0.4–2.310.95 ± 0.190.92 [0.82, 1.05]0.40–0.820.73 ± 0.070.75 [0.69, 0.78]0.82–0.920.87 ± 0.030.87 [0.84, 0.90]0.92–1.050.98 ± 0.040.98 [0.95, 1.01]1.05–2.311.20 ± 0.141.16 [1.10, 1.26]N49261232123212321230Age, yr60.2 ± 11.963 [53, 70]64.8 ± 8.768 [61, 71]62.4 ± 10.165 [57, 70]59.5 ± 11.763 [52, 69]54.1 ± 13.656 [45, 65]8.9e-110Sex (female)1 959 (39.8)384 (31.2)445 (36.1)540 (43.8)590 (48)6.2e-19Body mass index, kg/m229.8 ± 6.028.9 [25.7, 33.2]30.4 ± 6.029.5 [26.2, 34.1]29.9 ± 5.829.1 [26.1, 33.2]29.9 ± 5.829.0 [25.9, 33.1]29.0 ± 6.227.8 [24.5, 32.4]3.5e-10Current smoker785 (16)170 (13.9)175 (14.2)208 (16.9)232 (18.9)1.3e-3Diabetes mellitus1768 (35.9)530 (43)474 (38.5)405 (32.9)359 (29.2)6.9e-13Prevalent CVD1261 (25.6)408 (33.1)346 (28.1)302 (24.5)305 (16.7)4.7e-20eGFR (CKD-EPI formula), ml/min per 1.73 m249.5 ± 18.246 [37, 58]46.4 ± 15.844 [35, 54]47.6 ± 17.245 [36, 55]50.1 ± 18.447 [38, 58]53.8 ± 20.450 [40, 62]6.0e-22UACR, mg/g430 ± 96951 [9, 383]413 ± 1052140 [9, 304]350 ± 78444 [9, 294]440 ± 91258 [10, 441]519 ± 109065 [10, 522]1.6e-04Serum albumin, mg/l38.3 ± 4.338.7 [36.2, 40.8]37.9 ± 4.238.3 [36.0, 40.5]38.4 ± 3.938.7 [36.2, 40.7]38.5 ± 4.138.9 [36.4, 41]38.5 ± 4.839 [36.2, 41.1]4.3e-04Systolic blood pressure, mm Hg139.5 ± 20.2138 [126, 152]141.1 ± 20.0140 [128, 154]139.7 ± 20.2139 [126, 152]140.1 ± 20.9138 [125, 152]137.2 ± 19.8135 [124, 149]2.1e-06Diastolic blood pressure, mm Hg79.2 ± 11.779 [71, 87]77.4 ± 11.577 [70, 85]78.6 ± 11.778 [71, 86]79.7 ± 11.779 [72, 87]81.1 ± 11.681 [73, 88]3.5e-14High-sensitivity C-reactive protein, mg/l4.77 ± 8.482.27 [1.02, 5.01]5.25 ± 8.042.53 [1.22, 5.57]4.80 ± 7.822.34 [1.06, 4.97]4.73 ± 8.512.27 [1.01, 5.10]4.31 ± 9.411.91 [0.85, 4.4]2.0e-07HDL-cholesterol, mg/dl51.8 ± 1848.3 [39.2, 61.3]50.2 ± 17.246.9 [38.2, 58.9]51.1 ± 17.847.7 [39.0, 60.3]52.4 ± 18.349.2 [39.8, 61.4]53.5 ± 18.450.1 [40.1, 64.1]1.5e-05LDL-cholesterol, mg/dl118.2 ± 43.5113.6 [89.1, 142.7]113.1 ± 41.1108.8 [84.9, 138.9]115.8 ± 40.4112.2 [87.3, 140.5]120.7 ± 42.5117.9 [90.7, 146.3]123.3 ± 48.8117.0 [92.3, 147.2]5.6e-08CKD-EPI, Chronic Kidney Disease Epidemiology Collaboration; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate; HDL, high-density lipoprotein; LDL, low-density lipoprotein; RTL, relative telomere length; UACR, urine albumin–creatinine ratio.Data are given as mean ± SD, with median [25th, 75th percentile], or n (%), unless otherwise indicated. Open table in a new tab CKD-EPI, Chronic Kidney Disease Epidemiology Collaboration; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate; HDL, high-density lipoprotein; LDL, low-density lipoprotein; RTL, relative telomere length; UACR, urine albumin–creatinine ratio. Data are given as mean ± SD, with median [25th, 75th percentile], or n (%), unless otherwise indicated. A total of 354 deaths occurred during a median follow-up period of 4 years (1483 days). The causes of death were CVD including myocardial infarction, coronary heart disease, sudden cardiac death, congestive heart failure, pulmonary embolism, cardiac valve disease and ischemic stroke (117 patients, 33.1%), infections (67 patients, 18.9%), non-ischemic cerebrovascular causes (9 patients, 2.5%), peripheral vascular disease (7 patients, 2.0%), kidney failure (8 patients, 2.3%), various other causes (103 patients, 29.1%) and unknown causes (43 patients, 12.1%). Cumulative incidence plots show that incidence of all-cause mortality (Figure 1a) increases with shorter RTL, with the highest incidence with lowest RTL quartile. In the cumulative incidence function curves of cardiovascular (Figure 1b) and infection mortality (Figure 1c), the difference between quartiles was less pronounced, but the order of the quartiles was the same. Results of Cox regression models applying different adjustments are provided in Table 2 and showed a significant association between shorter RTL and the risk of all-cause mortality. Evaluated continuously, each decrease of 0.1 RTL units was associated with a 16% increased risk of death in a model adjusted for age and sex (hazard ratio [HR], 1.16; 95% confidence interval [CI], 1.08–1.24; P = 1.7e-05). The association remained significant after an extended adjustment for eGFR, urine albumin–creatinine ratio, diabetes mellitus, and prevalent cardiovascular disease (model 2: HR, 1.16; 95% CI, 1.08–1.24) as well the additional CVD risk factors low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, smoking, body mass index, systolic blood pressure, diastolic blood pressure, C-reactive protein, and serum albumin at baseline (model 3: HR, 1.14; 95% CI, 1.06–1.22; P = 3.5e-04). Nonlinear P spline analyses are given in Figure 2 and revealed an almost linear association of RTL with all-cause mortality. Patients with the shortest RTL (1st quartile) had a 75% higher risk for all-cause mortality compared to those in the quartile with the longest RTL (Supplementary Table S1, fully adjusted model: HR, 1.75; 95% CI, 1.22–2.50; P = 0.0024).Table 2Results of Cox model on all-cause mortality, death due to cardiovascular disease (cause-specific hazard ratios [HRs] are given), and death due to infections (cause-specific HRs are given) for each decrease in 0.1 units of relative telomere length (RTL) as well as the first quartile of RTL versus quartiles 2 to 4 (combined as reference category)Adjustment modelaModel 1: adjusted for age and sex; model 2: adjusted for age, sex, estimated glomerular filtration rate, urine albumin–creatinine ratio, diabetes mellitus, prevalent cardiovascular disease; model 3: adjustment as in model 2 plus low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, smoking, body mass index, systolic blood pressure, diastolic blood pressure, C-reactive protein, and serum albumin.,bDue to missing values, not all models include the same number of events.For each decrease of 0.1 RTLHR [95% CI]P valueAll-cause mortality Model 1 (354 events)1.16 [1.08–1.24]1.7e-05 Model 2 (343 events)1.16 [1.08–1.24]4.4e-05 Model 3 (333 events)1.14 [1.06–1.22]3.5e-04Cardiovascular death Model 1 (117 events)1.22 [1.08–1.38]0.0014 Model 2 (116 events)1.21 [1.07–1.36]0.0026 Model 3 (113 events)1.20 [1.05–1.35]0.0052Death due to infections Model 1 (67 events)1.26 [1.07–1.48]0.005 Model 2 (65 events)1.28 [1.08–1.51]0.0024 Model 3 (63 events)1.27 [1.07–1.50]0.0051CI, confidence interval.a Model 1: adjusted for age and sex; model 2: adjusted for age, sex, estimated glomerular filtration rate, urine albumin–creatinine ratio, diabetes mellitus, prevalent cardiovascular disease; model 3: adjustment as in model 2 plus low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, smoking, body mass index, systolic blood pressure, diastolic blood pressure, C-reactive protein, and serum albumin.b Due to missing values, not all models include the same number of events. Open table in a new tab CI, confidence interval. Next, we analyzed what is driving the association of RTL with all-cause mortality (Table 2). We evaluated the 2 most frequent specific causes of death and observed that each decrease of 0.1 RTL units was associated with a 20% increased risk of CVD death in the fully adjusted model (HR, 1.20; 95% CI, 1.05–1.35; P = 0.0052). Reduced RTL was also significantly inversely associated with death due to infections. Each 0.1 unit decrease of RTL was associated with a 1.27-fold higher risk for death due to infections (HR, 1.27; 95% CI, 1.07–1.50; P = 0.0051). Looking at the estimates for the various quartiles in Supplementary Table S1 revealed for death due to infections that the estimates for each of the quartiles 1, 2, and 3 were similarly elevated compared to that for quartile 4. The analysis with other causes of death as well as unknown causes of death was obviously too heterogeneous and did not reveal any association with RTL (data not shown). The graph of the scaled Schoenfeld residuals and test on proportional hazard assumptions did not suggest any time-varying effects for RTL on any of the investigated outcomes. The subdistribution HRs for both cardiovascular and infection death, reported in Supplementary Table S2, are only slightly attenuated compared to the cause-specific HRs. We also evaluated whether the effect of RTL on the 3 different outcomes differed between men and women and for patients with and without diabetes mellitus, but we did not detect a significant interaction for these variables, or for age (all P values of interaction >0.1 in the fully adjusted models). Given that we recently observed a U-shaped association between duration of CKD and RTL,22Raschenberger J. Kollerits B. Titze S. et al.Do telomeres have a higher plasticity than thought? Results from the German Chronic Kidney Disease (GCKD) study as a high-risk population.Exp Gerontol. 2015; 72: 162-166Crossref PubMed Scopus (16) Google Scholar we performed a sensitivity analysis additionally adjusting for the duration of CKD at baseline defined as less than 6 months, between 6 month and 5 years, and more than 5 years. This additional adjustment resulted in only marginal changes of the HRs obtained for all 3 endpoints (Supplementary Table S3). The results of this study showed a significant association of RTL with all-cause mortality in a non-dialysis-dependent CKD cohort. Shorter RTL was associated with higher risk of mortality independently from kidney function and traditional CVD risk factors. This association was driven by death due to CVD as well as death due to infections. Prior studies,5Mons U. Müezzinler A. Schöttker B. et al.Leukocyte telomere length and all-cause, cardiovascular disease, and cancer mortality: results from individual-participant-data meta-analysis of 2 large prospective cohort studies.Am J Epidemiol. 2017; 185: 1317-1326Crossref PubMed Scopus (77) Google Scholar,6Wang Q. Zhan Y. Pedersen N.L. et al.Telomere length and all-cause mortality: a meta-analysis.Ageing Res Rev. 2018; 48: 11-20Crossref PubMed Scopus (137) Google Scholar,31Batsis J.A. Mackenzie T.A. Vasquez E. et al.Association of adiposity, telomere length and mortality: data from the NHANES 1999-2002.Int J Obes. 2018; 42: 198-204Crossref Scopus (29) Google Scholar, 32Fitzpatrick A.L. Kronmal R.A. Kimura M. et al.Leukocyte telomere length and mortality in the cardiovascular health study.J Gerontol A Biol Sci Med Sci. 2011; 66A: 421-429Crossref Scopus (225) Google Scholar, 33Needham B.L. Rehkopf D. Leukocyte telomere length and mortality in the National Health and Nutrition Examination Survey, 1999–2002.Epidemiology. 2015; 26: 528-535Crossref PubMed Scopus (114) Google Scholar, 34Pusceddu I. Kleber M. Delgado G. et al.Telomere length and mortality in the Ludwigshafen risk and cardiovascular health study.PLoS One. 2018; 13e0198373Crossref PubMed Scopus (26) Google Scholar, 35Rode L. Nordestgaard B.G. Bojesen S.E. Peripheral blood leukocyte telomere length and mortality among 64 637 individuals from the general population.J Natl Cancer Inst. 2015;107:djv074; Crossref PubMed Scopus (209) Google Scholar with few exceptions,36Loprinzi P.D. Loenneke J.P. Leukocyte telomere length and mortality among U.S. adults: effect modification by physical activity behaviour.J Sports Sci. 2018; 36: 213-219Crossref PubMed Scopus (10) Google Scholar,37Gao X. Zhang Y. Mons U. et al.Leukocyte telomere length and epigenetic-based mortality risk score: associations with all-cause mortality among older adults.Epigenetics. 2018; 13: 846-857Crossref PubMed Scopus (16) Google Scholar have demonstrated a negative association between RTL and all-cause mortality in the general population. The largest study so far (n = 64,637) was performed by Rode et al., with an adjusted HR for mortality of 1.40 for the decile with the shortest versus the decile with the longest RTL.35Rode L. Nordestgaard B.G. Bojesen S.E. Peripheral blood leukocyte telomere length and mortality among 64 637 individuals from the general population.J Natl Cancer Inst. 2015;107:djv074; Crossref PubMed Scopus (209) Google Scholar In accordance with these results, our study showed with each decrease of 0.1 RTL units a 14% higher risk for all-cause mortality, which translates to a 75% higher risk for those in the lowest compared to the highest quartile of RTL. To our knowledge, only Carrero et al.28Carrero J.J. Stenvinkel P. Fellström B. et al.Telomere attrition is associated with inflammation, low fetuin-A levels and high mortality in prevalent haemodialysis patients.J Intern Med. 2008; 263: 302-312Crossref PubMed Scopus (155) Google Scholar have investigated the relationship between RTL and mortality risk in CKD patients. They studied 175 patients with end-stage kidney disease treated by hemodialysis, of whom 70 died during a median of 31 months of observation. The authors observed that TL independently predicted patient survival after additional adjustment for age, sex, and inflammation. The current study extends these observations to the much larger group of individuals with CKD who do not require dialysis. No studies have investigated the association of RTL with CVD mortality in CKD patients so far, although this is the major cause of death in these patients. Depending on the investigated ethnicity and on the data adjustment models, some, but not all, studies in the general population reported an association between low RTL and CVD outcomes.33Needham B.L. Rehkopf D. Leukocyte telomere length and mortality in the National Health and Nutrition Examination Survey, 1999–2002.Epidemiology. 2015; 26: 528-535Crossref PubMed Scopus (114) Google Scholar,38D'Mello M.J.J. Ross S.A. Briel M. et al.Association between shortened leukocyte telomere length and cardiometabolic outcomes: systematic review and meta-analysis.Circ Cardiovasc Genet. 2015; 8: 82-90Crossref PubMed Scopus (248) Google Scholar, 39Haycock P.C. Heydon E.E. Kaptoge S. et al.Leucocyte telomere length and risk of cardiovascular disease: systematic review and meta-analysis.BMJ. 2014; 349: g4227Crossref PubMed Scopus (564) Google Scholar, 40Madrid A.S. Rode L. Nordestgaard B.G. et al.Short telomere length and ischemic heart disease: observational and genetic studies in 290,022 individuals.Clin Chem. 2016; 62: 1140-1149Crossref PubMed Scopus (77) Google Scholar, 41Mwasongwe S. Gao Y. Griswold M. et al.Leukocyte telomere length and cardiovascular disease in African Americans: The Jackson Heart Study.Atherosclerosis. 2017; 266: 41-47Abstract Full Text Full Text PDF PubMed Scopus (22) Google Scholar Strong support for a causal association came from a Mendelian randomization study in which genetic variants associated with shorter RTL were found to be associated with ischemic heart disease.40Madrid A.S. Rode L. Nordestgaard B.G. et al.Short telomere length and ischemic heart disease: observational and genetic studies in 290,022 individuals.Clin Chem. 2016; 62: 1140-1149Crossref PubMed Scopus (77) Google Scholar In the present study, we identified a significant association of RTL with cardiovascular deaths, with a 20% higher risk with each decrease of RTL by 0.1 units, or a 75% higher risk for those patients in the quartile with the shortest TL compared to the quartile with the longest TLs. This finding is in line with our earlier report of an association with prevalent cardiovascular events in this patient population: each decrease of RTL by 0.1 units was significantly associated with a 6% higher odds for prevalent CVD in a model adjusting for age, sex, current smoking, hypertension, diabetes status, low-density lipoprotein cholesterol, hi