Background: Telomere attrition is a novel risk factor for cardiovascular disease. Studies of telomere length in relation to kidney function are limited. We explored the association of kidney function with telomere length and telomere shortening. Methods: The Heart and Soul Study is a longitudinal study of patients with stable coronary heart disease. Measures of baseline kidney function included: serum creatinine, creatinine-derived estimated glomerular filtration rate (eGFR(CKD-EPI)), 24-hour urine measured creatinine clearance, cystatin C, cystatin C-derived estimated glomerular filtration rate (eGFRcys) and urine albumin to creatinine ratio. Telomere length was measured from peripheral blood leukocytes at baseline (n = 954) and 5 years later (n = 608). Linear regression models were used to test the association of kidney function with (i) baseline telomere length and (ii) change in telomere length over 5 years. Results: At baseline, mean eGFR(CKD-EPI) was 72.6 (±21.5) ml/min/1.73 m(2), eGFRcys was 71.0 (±23.1) ml/min/1.73 m(2) and ACR was 8.6 (±12.3) mg/g. Only lower baseline eGFR(CKD-EPI) was associated with shorter baseline telomere length (9.1 (95% CI 1.2-16.9) fewer base pairs for every 5 ml/min/1.73 m(2) lower eGFR(CKD-EPI)). Lower baseline eGFR(CKD-EPI) (and all other measures of kidney function) predicted more rapid telomere shortening (10.8 (95% CI 4.3-17.3) decrease in base pairs over 5 years for every 5 ml/min/1.73 m(2) lower eGFR(CKD-EPI)). After adjustment for age, these associations were no longer statistically significant. Conclusions: In patients with coronary heart disease, reduced kidney function is associated with (i) shorter baseline telomere length and (ii) more rapid telomere shortening over 5 years; however, these associations are entirely explained by older age.