Statistical Analysis
A Paired T-Test was performed to test for a significant difference in
group means of preprocedural LVEF and postprocedural LVEF. This section
of analysis was done to determine if the mean ejection fraction was
significantly different after the procedure. A Wilcoxon Signed-Rank Test
was performed to see if the distribution of the before and after LVEF
were symmetric about zero.
We then set out to determine if the location of the Micra had any
significance in EF performance. Before performing our two-way ANOVA to
test the significance of the group (implantation location) means in each
of the EF states (before and after), we had to confirm homoscedasticity
(homogeneity of variance), which would confirm that the variance around
the mean was roughly equal in all groups. For this purpose, we employed
the Bartlett Test, which confirmed that there was roughly equal variance
in the before and after LVEF groups. We were then able to perform a
two-way ANOVA test, an ANCOVA test, and a Fixed Effects Model to
determine if the Micra implantation location had any significant effect
on the LVEF.
Basic statistical analysis was also used to evaluate the impact of
device implantation on the QRS, TR severity, and occurrence of PICM.