Discussion
Summary of findings
Overall hypertonic saline was associated with a reduced
length of stay, with a pooled difference of −0.36, [95 %
CI −0.50 to −0.22]. The evidence for this result is of
moderate varies quality (GRADE summary of findings in
Table 2). However, our confidence in this estimate is
undermined by the excessive heterogeneity of the results
(I2 = 78 %) and inconsistency between the main and
sensitivity analysis in general. Of particular note is that
the overall pooled estimate is contradicted by the four
largest component trials [65, 70, 71, 73], all of which
found no benefit of hypertonic saline. The true impact
of hypertonic saline therefore appears to depend greatly
on the patient population within which each trial was
conducted, and in particular the level of standard care.