The quality of the included studies was generally moderate. Five studies (listed in the table; reported as four studies in text) reported adequate allocation concealment. Most of the studies had moderate to high risk of bias or did not report the trial in sufficient detail for the reviewers to assess this.
Thirteen studies (1,059 participants) were included in the review: nine studies (673 participants) were randomised controlled trials (RCTs) and four studies (386 participants) were case-matched or propensity-matched studies. All studies were small, ranging in size from 12 to 108 patients.
Fenoldopam was associated with a reduced risk of the need for RRT (seven studies, 869 participants), pooled OR 0.37 (95% CI 0.23, 0.59, P<0.001). There was no significant heterogeneity, P=0.53, I2 = 0%.
Fenoldopam was associated with a reduced risk of hospital death (seven studies, 770 participants), pooled OR 0.46 (95% CI 0.29, 0.75, P=0.002). There was no significant heterogeneity, P=0.66, I2 = 0%.
Fenoldopam was associated with a reduction in time on mechanical ventilation, WMD -0.93 hours (95% CI -1.57, -0.28, P=0.005), although there was significant statistical heterogeneity, P=0.001, I2 = 73%, and a shorter time in intensive care, WMD -0.93 days (95% CI -1.27, 0.58, P<0.001), although there was again a significant statistical heterogeneity, P=0.003, I2 = 67%. See CRD commentary for comment regarding this and other inconsistencies in the results.
Fenoldopam was associated with a higher rate of hypotensive episodes or use of vasopressin, pooled OR 1.94 (95% CI 1.19, 3.16, P=0.008). There was no significant heterogeneity, P=0.76, I2 = 0%.
Subgroup and sensitivity analyses, including restricting the analysis to only RCTs, did not substantially alter any of the results other than that fenoldopam was beneficial in patients undergoing cardiac surgery, but not those undergoing vascular surgery.
There was no evidence of publication bias.