- Mishra, Rakesh K;
- Li, Yongmei;
- DeFilippi, Christopher;
- Fischer, Michael J;
- Yang, Wei;
- Keane, Martin;
- Chen, Jing;
- He, Jiang;
- Kallem, Radhakrishna;
- Horwitz, Edward J;
- Rafey, Mohammad;
- Raj, Dominic S;
- Go, Alan S;
- Shlipak, Michael G;
- Investigators, CRIC Study
Background
Serum cardiac troponin T (cTnT) is associated with increased risk of heart failure and cardiovascular death in several population settings. We evaluated associations of cTnT levels with cardiac structural and functional abnormalities in a cohort of patients with chronic kidney disease (CKD) without heart failure.Study design
Cross-sectional.Setting & participants
Chronic Renal Insufficiency Cohort (CRIC; N=3,243).Predictor
The primary predictor was cTnT level. Secondary predictors included demographic and clinical characteristics, hemoglobin level, high-sensitivity C-reactive protein level, and estimated glomerular filtration rate using cystatin C.Outcomes
Echocardiography was used to determine left ventricular (LV) mass and LV systolic and diastolic function.Measurements
Circulating cTnT was measured in stored sera using the highly sensitive assay. Logistic and linear regression models were used to examine associations of cTnT level with each echocardiographic outcome.Results
cTnT was detectable in 2,735 (84%) persons; median level was 13.3 (IQR, 7.7-23.8) pg/mL. Compared with undetectable cTnT (<3.0 pg/mL), the highest quartile (23.9-738.7 pg/mL) was approximately 2 times as likely to have LV hypertrophy (OR, 2.43; 95% CI, 1.44-4.09) in the fully adjusted model. cTnT level had a more modest association with LV systolic dysfunction; as a log-linear variable, a significant association was present in the fully adjusted model (OR of 1.4 [95% CI, 1.2-1.7] per 1-log unit; P < 0.001). There was no significant independent association between cTnT level and LV diastolic dysfunction. When evaluated as a screening test, cTnT level functioned only modestly for LV hypertrophy and concentric hypertrophy detection (area under the curve, 0.64 for both), with weaker areas under the curve for the other outcomes.Limitations
The presence of coronary artery disease was not formally assessed using either noninvasive or angiographic techniques in this study.Conclusions
In this large CKD cohort without heart failure, detectable cTnT had a strong association with LV hypertrophy, a more modest association with LV systolic dysfunction, and no association with diastolic dysfunction. These findings indicate that circulating cTnT levels in patients with CKD are predominantly an indicator of pathologic LV hypertrophy.