Treatment of acute-onset hypertension in pregnancy: A network meta-analysis of randomized controlled trials comparing anti-hypertensives and route of administration

Background: Consensus on the relative efficacy of available antihypertensive agents used in pregnancy is lacking. Objective: To compare treatment success with antihypertensives and categorize by route of administration. Search strategy: MEDLINE, Embase, PubMed, Web of Science, Scopus, CINAHL, and clinicaltrials.gov were searched without date restriction.Data collection: Peer-reviewed randomized controlled trials (RCTs) comparing pharmacologic agents used to treat hypertension in parturients were included. Evaluated treatment groups included IV-labetalol (BBIV), IV-hydralazine (DIV), oral-nifedipine (CCBPO), sublingual nifedipine (CCBSL), IV-calcium channel blocker (nonspecific)(CCBIV), IV-nitroglycerine (NTG), epoprostenol infusion (PRO), IV-ketanserin (5HT2B), IV-diazoxide (BZO), oral-nifedipine + methyldopa (CCBAG), oral-methyl-dopa (AAG), and oral prazosin (ABPO). Analysis: Seventy-four studies (8324 patients) were eligible post PRISMA guidelines screening. Results were pooled using a Bayesian-approach for success of treatment (study defined target blood pressure), time to achieve target pressure, and neonatal intensive-care admissions.Results: Treatment success (primary outcome, 55 trials with 5518 patients) was analyzed. Surface under the cumulative ranking curve (SUCRA) was categorized for 13 drugs, CCBPO (0.84) followed by CCBSL (0.78) were most likely to be effective in achieving target blood pressure. After sub-grouping by presence/absence of pre-eclampsia, CCB-PO ranked highest for both [(0.82) vs. (0.77), respectively]. Serotonin antagonists (0.99) and nitroglycerin (0.88) ranked highest for time to target pressure. NICU admissions were lowest for alpha-2 agonists (0.89), followed by BB PO (0.82) and hydralazine IV (0.49).Conclusion: Oral calcium-channel blockers ranked highest for treatment success. Ketanserin achieved target blood pressure fastest, warranting additional research. The results should be interpreted with caution as SUCRA values may not indicate whether the differences between interventions have clinically meaningful effect sizes.