To investigate the relationship between radiation dose to the heart and cardiac biomarker levels; and to investigate the temporal relationship between RT delivery and anticipated elevation of cardiac biomarkers.
We will test whether the correlation between the change from baseline in the enzyme level and the heart V30 (volume of heart receiving > 30 Gy) is significantly greater than 0. We will also estimate the correlation between the change from baseline in the enzyme level and the heart V30 with a 95% confidence interval. We will perform the analysis described here for each of the 2 enzymes and each of the 3 time points (beyond baseline) described above.
We will use descriptive statistics (mean, median, quartiles) to summarize the change from baseline in the enzyme levels at each time point (beyond baseline). We will use side-by-side boxplots to illustrate the distribution of the change from baseline in the enzyme levels over time. We will use mixed effects regression with second order terms to model the change from baseline in the enzyme levels as a function of time, and we will graph the regression line for each enzyme over a scatter plot of the data.
We will tabulate the incidence of adverse cardiac outcomes including myocardial infarction, heart failure, arrhythmias, all-cause and cardiac-specific mortality at the same time points used to analyze cardiac biomarker levels (i.e., within 48 hours prior to RT, 12 to 24 hours following the first fraction, within 48 hours of the last fraction of RT, and 1 to 2 months following completion of RT)
The results of this study will provide the basis for long-term evaluation of cardiac outcomes in a larger cohort of patients receiving radiation therapy to the heart / mediastinal region.