Limitations
Several limitations should be addressed when interpreting our results.
First, our analysis consists of observational, non-randomized studies,
which have inherent flaws in terms of selection bias and unmeasured
confounders. Second, the current meta-analysis is based on study-level
data; lack of patient level data precludes further subgroup analysis.
Third, the lack of clinical outcomes limits conclusions regarding the
long-term implications of this association. Fourth, there was
significant heterogeneity among studies for many end point definitions
such as OSA definition, severity, echocardiographic definition of DD.
Fifth, the number of studies examining the correlation between DD and
severity of sleep apnea were limited therefore definitive conclusions
with regards to the association between DD and OSA severity would be
precarious. In spite these limitations, the results of the present study
can be used to inform future studies.