Abstract & Commentary
Diurnal Variation in Troponin T
By Michael H. Crawford, MD, Editor
Klinkenberg LJ, et al. Circulating cardiac troponin T exhibits a diurnal rhythm. J Am Coll Cardiol 2014;63:1788-1795.
The advent of highly sensitive troponin assays has shown detectable levels in apparently healthy subjects. Persistently elevated levels (> 99% percentile) can be observed in patients with a variety of chronic diseases, including coronary artery disease (CAD), making the accurate diagnosis of myocardial infarction (MI) challenging. However, a significant rise or fall in troponin levels should identify those with MI. In order to interpret any observed changes in the troponin level over time, one needs to know the expected biologic variability in troponin. Thus, this study evaluated within day, diurnal, and between week variation in troponin T (TnT) levels in 23 male subjects with long-standing type 2 diabetes without evidence of acute cardiovascular disease (mean age 63 years). Beginning at 8:30 a.m., all subjects exhibited a gradual decrease in TnT levels until 8 p.m. when they began to rise again. The percent decrease was higher in those with higher peak levels. Average TnT peaked at 8:30 a.m. (17 ± 3 ng/L) and bottomed at 8:30 p.m. (12 ± 2 ng/L) in sedentary subjects. This diurnal variation was not affected by the introduction of physical activity and was considerably greater than that observed in the between week variation. The authors concluded that the use of changes in TnT over time to diagnose MI should take this diurnal variation into account, and risk stratification values could benefit from standardized sampling times.
Commentary
This is the kind of information I wish I had known about years ago. It would have helped inform some difficult clinical decisions I’ve had to make regarding acute MI (yes or no). Concern for acute MI arises when symptomatic patients have an elevated Tn, but a normal unchanged ECG. If the next sample is higher or lower, it suggests myocardial injury. However, the circadian variation noted with TnT in this study was 30-40% on average. Also, the higher the TnT, the greater the variation. Prior studies in normal subjects have shown variation in random samples or short time frame serial samples, but normal subjects may not be representative of the type of patient who presents to the emergency room with chest pain. Also, unless very high sensitivity assays are used, normals rarely have elevated Tn levels. In diabetes and metabolic syndrome patients, up to 20% have been shown to have elevated Tn levels in random samples. Thus, the selection of type 2 diabetes patients without symptomatic or overt ischemic heart disease was a relevant population for this study.
There are some limitations to this study. They only studied diabetic men and the number of subjects in the hourly circadian variation substudy was only 7. Also, heart rate and blood pressure were not recorded, so we don’t know how their circadian pattern related to TnT. In addition, all the subjects were either overweight or obese and such individuals have a higher incidence of obstructive sleep apnea. It could be that hypertension and hypoxia could explain the nocturnal rise in TnT. In addition, this study does not shed any light on the mechanism of the observed changes in TnT levels. Finally, only TnT was studied (not TnI), which many hospitals use exclusively.
The clinical implications of this study are that in the grey zone possible MI patient, modest rises in TnT in serial samples after 8 p.m. may be due to diurnal variation. After 8 a.m., levels should be trending downward, so rises observed during the day are more consistent with acute MI.
