Three databases (PubMed, Cochrane, and Embase) were used to conduct a systematic search to find eligible studies comparing desmopressin (DDAVP) with controls in patients with AA-ICH. The Mantel-Haenszel statistic was used to determine an overall effect estimate for each outcome by calculating the risk ratios and 95% Confidence Intervals using Review Manager 5.4.1. The I2 test was used to measure heterogeneity, and the ROBINS-I was used to measure the risk of bias.

Six studies, containing a total of 603 patients, were included in the analysis. DDAVP was associated with a statistically insignificant decrease in hematoma expansion risk (RR=0.9, 95% CI= 0.59-1.37; P= 0.63, I2=52%). Thrombotic events also showed a statistically insignificant decrease in the DDAVP group (RR=0.69; 95% CI=0.26-1.81; P= 0.33, I2=14%). However, the risk of poor neurological outcomes showed a statistically significant increase in risk in patients on DDAVP (RR=1.36; 95% CI=1.06-1.76; P= 0.026, I2=25%). The studies showed a moderate to low level of risk of bias.

Evidence indicates that DDAVP does not provide any significant advantage in preventing hematoma expansion. However, it could lead to poorer neurological outcome, which underlines the cautious use of this drug in clinical practice. Since most of the studies were retrospective in nature, future research should focus on multicenter prospective designs or well-conducted propensity-matched cohort studies.