Effects of Immediate Postpartum Diuretic Treatment on Postpartum Blood Pressure among Individuals with Hypertensive Disorders of Pregnancy: A Systematic Review and Meta-Analysis

Abstract

Background:

Hypertensive disorders of pregnancy (HDP) are associated with ongoing postpartum hypertension (HTN) and increased morbidity. Extravascular water and sodium mobilization is implicated in postpartum blood pressure (BP) elevation, however trials of postpartum diuretics in HDP have had mixed results. Our meta-analysis aimed to analyze the impact of postpartum diuretics on postpartum hypertension following HDP.

Methods:

Systematic review was performed to identify observational cohort studies and randomized controlled trials (RCTs) studying the efficacy of diuretics in the treatment of postpartum BP. Meta-analysis outcomes included persistent HTN up to 10 days postpartum, mean postpartum systolic and diastolic BPs, and use of additional antihypertensive medications.

Results

From 10 included RCTs and one prospective cohort study with moderate level of bias, 1,624 subjects were included in the meta-analysis. Postpartum diuretic use was associated with lower systolic BP (SMD standard mean difference]: −0.44; 95% confidence interval [CI]: −0.66; −0.21) without a difference in diastolic BP (SMD: −0.15; 95% CI: −0.47; 0.16) compared with controls. There was no difference in rates of persistent HTN between the postpartum diuretics group versus controls (OR: 0.69; 95% CI: 0.44; 1.08) or in antihypertensive medication use (OR: 0.68; 95% CI: 0.46; 1.03). There was significant heterogeneity in diuretic effect on outcomes.

Conclusion:

Postpartum diuretic use was associated with no difference in persistent hypertension, diastolic blood pressure, or need for hypertensive therapy. A modest reduction in systolic blood pressure was observed, though of uncertain clinical significance. Larger, high-quality studies are needed to clarify whether postpartum diuretic use may be of clinical benefit.

Graphical Abstract

INTRODUCTION

Hypertensive disorders of pregnancy (HDP) impact 10–20% of pregnant individuals in the United States and are associated with increased morbidity in the peripartum period including increased risk of hospital readmission, severe hypertension (HTN), heart failure, and stroke.^1–4 Normal pregnancy is characterized by a 40–50% expansion of the circulating blood volume which is associated with retention of sodium and water in the interstitial tissues and a gradual rise in systolic and diastolic blood pressure (BP) towards the end of pregnancy^5,6 and peaking at 3 to 10 days postpartum.^7,8 This postpartum BP elevation has been attributed to extravascular water and sodium mobilization into the intravascular space and may be compounded by iatrogenic administration of intravenous fluid during labor or cesarean section.⁹ Signs and symptoms of volume overload are prevalent among individuals with HDP, including dyspnea, pulmonary edema , and peripheral edema.⁴ Therefore, diuretics have been proposed to theoretically accelerate postpartum BP recovery among individuals with HDP through urinary sodium and water excretion, thereby decreasing intravascular volume.^10,11

While several clinical guidelines have put forth recommendations for postpartum BP monitoring there is a paucity of specific recommendations for postpartum BP management that accelerates BP recovery. According to the American College of Obstetricians and Gynecologists (ACOG), treatment of postpartum HTN is recommended for systolic BP ≥150 mm Hg or diastolic BP ≥100 mm Hg.¹² Current guidance focuses on the use of calcium channel blockers and nonselective beta blockers and lacks specific recommendations for diuretic use which is typically reserved for postpartum management of pulmonary edema or systemic volume overload. Given a proposed pathophysiology of HDP through intravascular volume expansion and increased sodium body sodium content, there may be an opportunity to improve BP control with diuretics.

Randomized controlled trials (RCTs) evaluating the effectiveness of postpartum diuretic treatment for individuals with HDP have been small and have reported mixed results regarding BP effects. The latest meta-analyses of postpartum loop diuretics have reported conflicting findings, with one reporting no difference in persistent hypertension at 1 or 6-weeks postpartum or need for antihypertensive therapy at discharge and the other reporting a small improvement in systolic blood pressure at postpartum day 3.^6,13 With the recent publication of new relevant studies as well as the lack of inclusion of all diuretic classes in prior meta-analyses, there is a need for a contemporary investigation on the role of postpartum diuretics in HDP. We performed an updated systematic review and meta-analysis to compare the effect of postpartum diuretic treatment to placebo or alternative therapies on reducing persistent postpartum HTN, systolic and diastolic BP, need for additional antihypertensive medication, and postpartum complications among individuals with HDP.

METHODS

Data Availability

The protocol for this review was registered in the PROSPERO database (CRD42021253740). We made two protocol modifications from the initial protocol registered on PROSPERO early in the review process, prior to abstract screening. First, we expanded study eligibility to include cohort studies in addition to randomized controlled trial. Second, we revised our primary outcome definition to include persistent hypertension within 0–10 days postpartum (rather than up to 7 days), to better reflect reporting variability and clinical practice and align with ACOG guidance recommending blood pressure monitoring through 7–10 days postpartum¹⁴. The data that support the findings of this study are available from the corresponding author upon reasonable request.

Data Sources and Searches

Searches were developed by a health sciences librarian and performed in the Ovid Medline, Embase.com, Web of Science Core Collection, and Cochrane CENTRAL databases (Table S1). The search strings included natural language and database-specific controlled vocabulary covering the concepts HDP and diuretic medications. The searches were limited to the English language as well as the years 1980 to present. Additionally, clinicaltrials.gov was searched for relevant trials. On May 11, 2021, 3718 references were downloaded into the EndNote reference management software. The references were deduplicated twice: first using the Amsterdam Efficient Deduplication method and then the Bramer et al. method, resulting in a total of 2994 citations.^15,16 These citations were uploaded to the DistillerSR software (Distiller, Evidence Partners, Ottawa, Canada) for screening. To supplement these electronic searches, reference lists of pertinent articles were reviewed to identify any additional potential eligible studies. Ongoing surveillance was conducted using PubMed article alerts to identify any potential additional studies published through December 1, 2024. The search was rerun on June 29, 2022, and after deduplication of the search, 246 new citations were uploaded to Distiller.¹⁷

Study Selection

In Distiller, two of four cardiology or obstetric physician investigators (SK, MC, AK, and AH) independently reviewed each title, abstract, and full text article when needed using prespecified inclusion and exclusion criteria (Table S2). Disagreements were resolved by discussion. RCTs and observational cohort studies (prospective or retrospective) were eligible for inclusion. Case control studies, cross-sectional studies, case reports, conference presentations or posters, and narrative review articles were excluded. Eligible interventions included intravenous or oral diuretics or diuretics in conjunction with another antihypertensive pharmacotherapy, while eligible comparators included placebo or non-diuretic antihypertensive pharmacotherapy. The primary outcome was “persistent HTN” defined as systolic BP ≥ 140 mmHg or diastolic BP ≥ 90 mmHg before delivery hospitalization discharge (day 0) or up to 10 days postpartum. As guidelines recommend checking an ambulatory BP within 10 days of delivery in individuals with HDP, we performed our analysis by grouping persistent HTN from 0–10 days postpartum.¹⁸ Additional outcomes included systolic and diastolic BP, persistent HTN at 30 days or 6 weeks postpartum, need for additional antihypertensive medication, length of hospitalization, hospital readmission, emergency department visits, heart failure, b-type natriuretic peptide levels, and lactation outcomes.

Data Extraction and Quality Assessment

For each included study, one of four reviewers (SK, MC, AK, or KS) abstracted relevant study characteristics into a structured form. A second reviewer (SK, MC, AK, or KS) verified all data for accuracy. Two reviewers (SK, MC, AK, or KS) independently assessed methodological quality. The Cochrane collaboration’s risk-of-bias tool for randomized trials version 2–0 was used for risk of bias assessment of the included RCTs.¹⁹ This tool allowed reviewers to assess bias in 5 categories including bias arising from the randomization process, bias due to deviations from intended interventions, bias due to missing outcome data, bias in measurement of the outcome, and bias in selection of the reported result. Through a series of yes/no questions, a risk of bias score (low, some concerns, high) was assigned per domain for each study. A low risk of bias meant that the study was judged to be a low risk of bias in all domains. A “some concern” risk of bias meant that the study was judged to have risk of concerns in some domains but no high risk of bias in any domains. Finally, a high risk of bias meant that the study was judged to have at least one high risk domain or multiple domains with some concern of bias overall lowering the confidence of the results. In turn, these domain-level judgements provided the basis for an overall risk-of-bias judgment for the specific trial result being assessed through an algorithm built in the tool. The Newcastle-Ottawa scale score was used for risk of bias assessment of the one included observational cohort study. The Newcastle-Ottawa Scale score is a score combining three different domains, selection (maximum 4 stars), comparability (maximum 1 star), and outcome (maximum 3 stars) for a total maximum score of 9. Studies with 7–9 point reflect high-quality.²⁰ At the end of this thorough review, two reviewers (SK and KS) compared risk of bias scores per trial and any disagreements in data extraction or study quality assessments were resolved by consensus or with a third reviewer.

Statistical Analyses

Bias corrected standardized mean differences (Hedges’ g) were used to present the difference between outcomes. A random effects model was used to estimate a pooled relative risk for persistent postpartum HTN given that a small number of studies with relatively small numbers of participants were anticipated, likely to result in some unexplained heterogeneity. A Forest plot was used to illustrate the individual and pooled relative risk, estimate the confidence interval for the pooled relative risk, and report an overall p-value. Secondary continuous outcomes were assessed using a random effects model to compute a pooled mean difference. The prospective cohort study was not included in quantitative meta-analysis given its non-randomized design. All statistical tests were performed in R 4.4.1 (R version 4.4.1 (2024–06-14 ucrt)) using the “meta” and “metafor” packages.^21–23 A pooled treatment effect model (theta) p-value <0.05 was considered significant.

RESULTS

Characteristics of Included Studies

The searches yielded 3975 citations as summarized in the PRISMA flow diagram (Figure 1). After removing duplicates, 3242 unique studies were screened and 97 were sought for retrieval. Out of these 97 studies, 86 were excluded due to various ineligibilities including study design, population, timing of intervention, comparison, outcomes, interventions, and additional duplicate records.

Figure 1. Literature search flow and study selection diagram.

The main characteristics of included studies are summarized in Table 1. A total of 1624 subjects were included from 10 RCTs and one prospective observational cohort study. Five studies were conducted in the United States,^24–28 2 studies in Iran,^29,30 1 study in Bangladesh,³¹ 1 study in India,³² 1 study in Brazil ,³³ and 1 study in the United Kingdom.³⁴ Patients in the intervention group received diuretics for a period of 5 to 10 days. Most RCTs assigned their intervention group to furosemide 20–40 mg daily except for Viteri et al.²⁵ who assigned intervention group patients to torsemide 20 mg daily and Bartal et al. who assigned intervention group patients to a combination of hydrochlorothiazide 15 mg and lisinopril 20 mg daily. Matthews et al.³⁴, Viteri et al.²⁵, Lopes Perdigao et al.²⁶, and Cursino et al.³³ assigned their control groups to placebo. Four studies assigned their control groups to alternative antihypertensive medications. The control groups in Veena et al.²⁴ and Bozorgan et al.³⁰ were assigned to nifedipine 10 mg every 8 hours, the control group in Bartal et al.²⁷ was assigned to nifedipine 30 mg daily, and the control group in Siamansoori et al.²⁹ was assigned to methyldopa 250 mg three times daily. The control groups in Ascarelli et al.²⁴ and Dabaghi et al.³¹ were assigned to no medication. The prospective cohort study enrolled patients in a remote postpartum blood pressure monitoring program as standard of care for which use of oral furosemide for 5 days postpartum was implemented. Of the 258 included patients, 203 received the furosemide who were then compared to the 55 patients who did not receive furosemide.²⁸ The trials with the largest number of subjects had 384²⁶ and 264²⁴ subjects, respectively. The study with the least number of participants had 19 subjects.³⁴

Table 1.

Main characteristics of included studies

Year, author	Country	Study design	Included HDP	Important exclusion criteria	Total (n)	Intervention (n)	Comparison (n)	Primary outcome	Secondary outcomes
Matthews et al34 (1997)	UK	RCT	Preeclampsia	Diabetes mellitus, renal or hepatic impairment, contraindication to diuretic	19	Furosemide 40 mg orally every 24 hours for 7 days (10)	Placebo orally every 24 hours for 7 days (8)	Mean and maximum BP	Postpartum antihypertensive usage Length of hospital stay
Ascarelli et al24 (2005)	USA	RCT	Chronic HTN, gestational HTN, preeclampsia	Less than 20 weeks of gestation, hypokalemia (K < 3.0 mEq/L), already taking diuretics or potassium supplementation, hemodynamic instability	264	Furosemide 20 mg orally every 24 hours for 5 days (132)	No medication (132)	Systolic and diastolic BP	Length of hospital stay Postpartum antihypertensive usage
Veena et al32 (2017)	India	RCT	Severe preeclampsia	Hemodynamic instability, hypokalemia (K< 3mEq/L), already on potassium supplementation and diuretics, expulsion of the fetus at <20 weeks	100	Furosemide 20 mg orally every 24 hours with nifedipine 10 mg orally every 8 hours until not needed (50)	Nifedipine 10 mg every 8 hours until not needed (50)	Reduction in mean, systolic, and diastolic BP	Length of hospital stay Postpartum antihypertensive usage Antihypertensive requirement at discharge
Viteri et al25 (2018)	USA	RCT	Preeclampsia with or without severe features, superimposed preeclampsia with or without severe features	Oliguria, heart failure, hypokalemia (serum potassium below 3 mEq/L), diuretic use within the past 24 hours, hypersensitivity to torsemide or sulfonylureas, and pulmonary edema	118	Torsemide 20 mg orally every 24 hours for 5 days (59)	Placebo orally every 24 hours for 5 days (59)	BP >150/100 by day 5 or hospital discharge	Persistent HTN at discharge 7–10 days and 6 weeks after delivery Postpartum antihypertensive usage, Length of hospital stay Hospital readmission for HTN Adverse events to diuretics
Dabaghi et al31 (2019)	Bangladesh	RCT	Severe preeclampsia HELLP syndrome	Gestational age under 20 weeks, hypokalemia (k < 3.0 mEq/L) on admission, history of diuretics or potassium supplements use, any hemodynamic instability surrounding the events of delivery	90	Furosemide 20 mg and potassium supplementation every 24 hours for 5 days (45)	No medication (45)	Mean systolic BP	Postpartum antihypertensive usage Length of hospital stay Eclampsia
Siamansoori et al29 (2020)	Iran	RCT	Preeclampsia	Underlying cardiac disease, complicated pregnancies including gestational diabetes mellitus, renal disease, hypersensitivity to furosemide or methyldopa, major adverse drug complications	100	Furosemide 20 mg twice daily (50)	Methyldopa 250 mg three times daily (50)	Systolic and diastolic BP at 6 hours, 48 hours and one week	Diuresis at 24 and 48 hours post intervention
Lopes Perdigao et al26 (2021)	USA	RCT	All forms of HDP: gestational HTN, preeclampsia with or without severe features, preeclampsia superimposed on chronic HTN with or without severe features	Underlying cardiac disease, rheumatologic disease, advanced diabetes, elevated creatinine (>1.2 mg/dL), significant hypokalemia (K<3 mEq/L), allergy to furosemide, or those who received diuretics before randomization	384	Furosemide 20 mg orally every 24 hours for 5 days (192)	Placebo orally every 24 hours for 5 days (192)	Persistent HTN 7 days postpartum (defined as at least 2 consecutive BP readings over 48 hours of systolic BP ≥140 mm Hg and diastolic BP ≥90 mm Hg) and the number of days required to achieve resolution of HTN (defined as at least 2 consecutive BP readings over 48 hours of systolic BP <140 mm Hg and diastolic BP <90 mm Hg)	Percentage of postpartum BP in the severe range (systolic BP ≥160 mm Hg or diastolic BP ≥110 mm Hg) Postpartum antihypertensive usage Hospital readmissions or ER visits for HTN Length of hospital stay HTN complications
Bozorgan, et al30 (2022)	Iran	RCT	Preeclampsia	Chronic HTN, BP less than 150/100 mmHg, administration of diuretics, renal disease, diabetes, hemodynamic instability, potassium level lower than 3 mEq/L, contraindication to furosemide, hematocrit more than 37%	116	Nifedipine 10 mg every 8 hours and furosemide 20 mg daily for 5 days (58)	Nifedipine 10 mg every 8 hours for 5 days (58)	Systolic and diastolic BP at day 2–5	Postpartum antihypertensive usage
Bartal, et al27 (2023)	USA	RCT	Chronic HTN or any form of HDP	Urine output <30 cc/h, creatinine > 1.4 mg/dL, potassium > 5 mEq/L, end-stage renal disease, hypersensitivity to ACEIs or sulfa drugs, idiopathic or hereditary angioedema, or pulmonary edema	58	HCTZ with lisinopril 12.5 mg/20 mg daily up to maximal dose of 25/20 mg daily (30) up to the first postpartum visit at 7–10 days after delivery	Nifedipine 30 mg daily up to maximum daily dose of 120 mg up to the first postpartum visit at 7–10 days after delivery	Systolic BP≥140 mmHg and/or diastolic BP ≥90 mmHg at 7–10 days after delivery or readmission for BP control	Severe maternal morbidity up to 6 weeks after delivery, severe BP (systolic BP ≥160 mmHg and/or diastolic BP ≥110 mmHg), postpartum hospital length of stay, systolic and diastolic BP at postpartum visit, medication compliance, adverse events
Cursino, et al33 (2025)	Brazil	RCT	Preeclampsia	Chronic HTN, blood pressure <140/90 mmHg, diuretic use, renal impairment, diabetes, sickle cell disease, rheumatologic disease, hemodynamic instability, potassium less than 3 mEq/L, contraindication for furosemide use	118	Furosemide 40 mg orally every 24 hours for 5 days (58)	Placebo orally every 24 hours for 5 days (60)	Mean blood pressure from 24 hours after delivery to first 15 days of delivery	Postpartum antihypertensive usage Time to blood pressure control Daily urine output Length of hospital stay Frequency of adverse events Frequency of maternal complications
Chandrasekaran, et al (2025)28	USA	Prospective cohort study	Any form of HDP	Chronic HTN without superimposed preeclampsia, creatinine > 1.2 mg/dL, cardiac disease, sickle cell disease, serum potassium < 3mEq/L, furosemide or sulfa allergy	258	Furosemide 20 mg orally every 24 hours for 5 days (203)	No medication (55)	Hospital readmission up to six weeks postpartum	BP values occurring over the first six weeks postpartum ER evaluations and readmissions

BP: blood pressure; HDP: hypertensive disorders of pregnancy; HTN: hypertension; MAP: mean arterial pressure; ER: emergency room; HELLP: hemolysis, elevated liver enzyme, low platelet count

Primary outcomes varied between studies and included mean systolic and diastolic BPs, reduction in mean BPs, and persistent HTN at various time points up to 10 days post-hospital discharge. Observation periods varied among studies and lasted up to 6 weeks post-delivery and post-hospitalization.

Primary Outcome

The primary outcome, persistent HTN at 0–10 days post-delivery, was available in 9 studies.^24–32,34 Persistent HTN was significantly less frequent in the intervention group in two of the more recent studies,^26,30 including the largest study in our analysis.²⁶ However, in our random effects model there was no significant difference in persistent HTN among individuals who received postpartum diuretics compared with controls (odds ratio [OR]: 0.69; 95% confidence interval [CI]: 0.45; 1.08), shown in Figure 2. Study heterogeneity was moderate for the primary outcome (I^2 of 41%, tau ^2 0.14, p=0.12).

Figure 2. Persistent hypertension at 0–10 days postpartum among individuals with hypertensive disorders of pregnancy treated with diuretic versus control.

OR: odds ratio; CI: confidence interval

In subgroup analyses by comparison group, persistent hypertension was significantly lower for studies that compared diuretics to an alternative antihypertensive medication as the control (OR: 0.47; 95% CI: 0.27; 0.80) (Figure 3), but was not significantly different among studies that compared diuretics to placebo (Figure S1) or no medication (Figure S2). A sensitivity analysis restricted to the five low risk of bias studies demonstrated postpartum diuretics were not significantly associated with persistent HTN (OR: 0.71; 95% CI: 0.35; 1.43; 0.33).^{25–27,33,34}

Figure 3. Persistent hypertension at 0–10 days postpartum among individuals with hypertensive disorders of pregnancy treated with diuretic versus alternative antihypertensive medication.

OR: odds ratio; CI: confidence interval

The one prospective cohort study reported no difference in rates of persistent hypertension between those who did versus did not receive furosemide (32.8% vs 36.2%, p=0.66.

Secondary Outcomes

Systolic and diastolic blood pressure in the postpartum period

The systolic and diastolic BP measurements were available for 6 studies^29–31 BP measurements were monitored for up to 10 days postpartum. Postpartum diuretic use was associated with lower systolic BP (square mean difference [SMD]: −0.44; 95% CI: −0.66; −0.219) (Figure 4) without a difference in diastolic BP (SMD: −0.15; 95% CI: −0.47; 0.16) (Figure S3) compared with controls. The lower systolic BP effect was largest in the Bozorgan et al. study. The cohort study demonstrated no difference in systolic or diastolic blood pressure.

Figure 4. Mean difference in systolic blood pressure among individuals with hypertensive disorders of pregnancy with postpartum diuretic treatment compared with control.

SD: standard deviation; SMD: standardized mean difference; CI: confidence interval

Antihypertensive use postpartum

Additional antihypertensive medication use during an observation period up to 10 days was reported in 10 studies (Figure 5).^{24–28,30–34} Although Veena et al. and Bozorgan et al. reported significantly lower use of additional antihypertensive drugs in the intervention group compared with the control group, there was no significant difference in the remaining studies.^30,32 This effect was largest in the Bozorgan et al. study (OR: 0.23; 95% CI: 0.080.67).³⁰ Pooling data from all 9 studies, there was no significant difference in postpartum antihypertensive use in patients who received diuretics compared with controls (OR: 0.768; 95% CI: 0.46–1.03; p=0.10). Study heterogeneity was moderate with I^2 of 39%.

Figure 5. Use of additional antihypertensive medication among individuals with hypertensive disorders of pregnancy treated with diuretic compared with control.

OR: odds ratio; CI: confidence interval

Length of hospital stay

Although 7 studies commented on length of hospital stay, there was significant heterogeneity in these reports which precluded the ability to calculate a pooled effect size for this secondary outcome. In brief, five studies reported postpartum hospital length of stay^25–28,32 while two others reported total hospital length of stay.^31,34 No RCTs reported a significant difference in these outcomes between treatment groups whereas the prospective cohort study reported a significantly shorter postpartum length of stay among those who did versus did not receive furosemide.²⁸

Postpartum maternal complications

There was limited reporting of our additional secondary outcome variables of interest including hospital readmission, emergency department visits, heart failure, b-type natriuretic peptide levels, and lactation outcomes. Four studies reported outcomes related to emergency department visits or hospital readmissions and three studies reported lactation outcomes without any significant differences between treatment groups for these secondary outcomes. ^25–27,34 However significant variations in outcome definitions and the inconsistent inclusion of a CI or standard error in these studies precluded the calculation of pooled estimates.

Risk of Bias Assessment Results

Detailed study quality assessments of the included RCTs and prospective cohort study are provided in Figure S4 and Table S3, respectively. Of the 10 included RCTs, five studies received a low risk of bias.^{25–27,32–34} One study received some concerns of risk of bias related to reporting of missing outcome data.³² Lastly, Four studies received an overall high risk of bias for significant concerns with multiple domains including the randomization process, intervention assignment, missing outcome data, and outcome measurement.^24,30–32 The prospective cohort study scored 5, reflecting a satisfactory study with some areas of concern, particularly in comparability and adequacy of follow-up given blood pressure measurements were missing for 64% of participants at 6 weeks.

Sample Size Calculation

Assuming a two-sided test with equal group sizes, α = 0.05 and β = 0.20, we estimated the number of participants required to detect the pooled effect from our random-effects model for persistent hypertension (OR 0.69). A total of 1,571 participants would be needed after adjusting for diversity (taû2) and 2,252 participants after adjusting for inconsistency (I^2) to achieve 80% power, which exceeds the 1,624 participants included in this meta-analysis, of which 1,316 had assessment of the primary outcome. Although no consensus exists for a minimal clinically important difference in this setting, relative risk reductions of 20–30% are common and considered meaningful in obstetric and cardiovascular trials such as SPRINT (20% reduction in cardiovascular events)³⁵ and Magpie (15–50% reductions in perinatal and maternal outcomes).³⁶ Our observed effect fell within this range.

DISCUSSION

In this meta-analysis, we assessed the role of postpartum diuretics after HDP on improving postpartum HTN and BP. We found that postpartum diuretic use was not associated with reduced persistent hypertension but was associated with reduced systolic BP compared with controls without a statistically significant difference diastolic BP or use of additional antihypertensive medications. In a sub-analysis of the primary outcome by the type of comparator, there were lower odds of persistent hypertension with the use of diuretic compared with alternative antihypertensive medications. This finding may support the importance of diuretics as first line therapy for blood pressure control in the postpartum period. Ultimately, the available data to assess the impact of postpartum diuretics on individuals with HDP remains limited. The number of available studies was relatively small, with some studies including only a small number of patients and some with significant concerns about risk of bias. Despite these limitations, the rates of persistent postpartum HTN and of use of antihypertensive medications were numerically lower with OR of 0.69 and 0.68, respectively; however, these findings did not reach statistical significance.

A recent systematic review and meta-analysis on the use of loop diuretics in the context of HDP similarly found no reduction in persistent HTN at 1 or 6 weeks postpartum and no reduction in the need for additional antihypertensive medications at discharge.⁶ Our meta-analysis expands on these findings with the inclusion of four additional studies. We assessed 10 trials and 1 prospective cohort study comparing diuretics to controls. Most of the included trials were individually limited by smaller sample size and short-term follow-up. With our current sample size of 1,624, of which 1,316 participants had assessment of the primary outcome of persistent hypertension, we are underpowered by 936 participants for our primary outcome according to our sample size calculation. Ongoing RCTs will hopefully contribute to narrowing knowledge gaps about the benefits of postpartum diuretics among individuals with HDP. The largest ongoing randomized placebo-controlled trial of postpartum diuretics in individuals with HDP will include an estimated 612 participants and will compare hydrochlorothiazide 50 mg daily to placebo for 14 days after delivery (NCT03298802). Primary outcomes include need for additional antihypertensive drug and rate of readmission or triage visits 1–6 weeks postpartum. The results of this trial would expand the number of subjects to more definitively assess the utility of postpartum diuretics in HDP.

Our finding of reduced systolic BP is novel, although the effect size is small and may reflect limited clinical significance. Improved BP control in the immediate short-term postpartum period may confer durable clinical benefit. Pharmaceutical intervention in the immediate postpartum period has been shown to result in improved BP control sustained up to 3–4 years postpartum^37,38 and be associated with more favorable left atrial and left ventricular remodeling, suggesting that prompt postnatal BP control may reverse the adverse remodeling known to result from HDP.³⁹ Nonetheless, the lack of a diastolic BP benefit is relevant in this patient population. The association between HDP and later-life cardiovascular disease is partially mediated through the development of chronic HTN, with up to 37% of individuals with a HDP developing chronic HTN at 2–7 years after delivery.^40–42 In a study of BP trajectories in the first year postpartum following a HDP, the majority of those with persistent HTN had isolated diastolic HTN, which may suggest that targeting an improvement in postpartum diastolic BP is of particular importance following an HDP.⁴³

Efficacy of postpartum diuretics may vary by severity of HDP. Lopes Perdigao et al. found that reduction in persistent HTN at 7 days postpartum was more significant when results were stratified to only individuals with non-severe HDP. Interestingly, there was no difference in persistent HTN in the severe HDP group (systolic BP ≥160 mm Hg or diastolic BP ≥110 mm Hg).²⁶ In contrast, Ascarelli et al. found that reduction in persistent HTN 2 days postpartum was more significant in patients with severe preeclampsia only.²⁴ There was heterogeneity among the studies in their inclusion of only severe versus both severe and non-severe HDP, and future studies should give particular attention to or perform subgroup analyses for these subgroups of HDP severity to define which group may particularly benefit from postpartum diuretics. There was also variation in the spectrum of HDP subtypes which were included in the studies. Importantly, some studies included chronic HTN which may be less diuretic responsive with regards to BP recovery and thus attenuate the measured benefit by the total studied population of individuals with HDP.

There may also be a role for postpartum diuretics for prevention of HDP. Risk factors for de novo postpartum hypertension including advanced maternal age >35 years, non-Hispanic Black race, obesity, and tobacco use may identify additional clinical subgroups with particular benefit for postpartum diuretic use.^8,44 The recently published Lasix for the Prevention of De Novo Postpartum Hypertension (LAPP) randomized placebo-controlled trial extends our understanding of the preventive role of postpartum diuretics by comparing primary prevention furosemide to placebo among 82 individuals at high risk for de novo postpartum HTN. The LAPP study found no difference in mean arterial pressure 24 hours prior to discharge or in the development of de novo postpartum HTN between the two groups.⁴⁵ As 42% of individuals with postpartum hypertension have preceding antepartum normotension, the identification of individuals at high risk to develop postpartum HTN and strategies to mitigate associated volume overload are needed.⁴⁶

Limitations

The RCTs available for inclusion in our meta-analysis resulted in a modest sample size. Four of the studies were felt to have high risk of bias. Restricting the primary outcome analysis to only the low risk of bias trials yielded a similar, nonsignificant estimate with the same effect direction as the primary analysis, supporting the robustness of our overall findings while underscoring residual imprecision.

There was heterogeneity in the control groups, with only four of the studies comparing diuretics to placebo and the remaining studies comparing diuretics to alternative antihypertensive medication or no medication. Unexpectedly, there was a significantly lower odds of persistent hypertension in the studies comparing diuretic to alternative antihypertensive medication which was not observed in the studies comparing diuretic to placebo. This seemingly paradoxical finding may suggest that diuretics would be beneficial to use as first line therapy for treatment of postpartum hypertension but requires further study.

Further, due to the lack of standardized guidance for postpartum BP monitoring, there was notable variation in the timing of study outcome assessments. Some studies reported this data at postpartum day 0–5, others at days 5–10, and others at 30 days. Consequently, study outcome heterogeneity was moderate per our random effect size analysis specifically regarding persistent postpartum HTN. Our analysis may have been affected by this heterogeneity in BP recording postpartum. One study required blood pressure to be controlled (SBP < 140 mmHg, DBP < 110 mmHg) prior to discharge.³³ This may reflect an emerging trend towards stricter BP control in study protocols based on recent clinical trials on postpartum blood pressure management. The CHAP trial reported 18% lower odds of their primary composite outcome of preeclampsia with severe features, preterm birth, placental abruption, and fetal or neonatal death among pregnant patients with chronic hypertension randomized to a BP treatment goal of <140/90 mmHg compared to no treatment unless severe hypertension (BP>160/105 mmHg).⁴⁷ The POP-HT study reported durable lower blood pressure in the first 9 months postpartum in patients randomized to physician guided titration of antihypertensive medications compared with usual care.⁴⁸ The American College of Obstetrics and Gynecology and the Society of Maternal Fetal Medicine subsequently recommended initiation and titration of pharmacologic treatment for chronic hypertension in pregnancy once BP reaches 140/90 mmHg, which may encourage stricter BP control protocols in future trials assessing the impact of postpartum diuretics in HDP.⁴⁹ Accordingly, comparison of the results of future studies with lower BP goals with the results of prior studies that were in an era of more liberal BP goals will require caution.

Conclusions

The data on the benefit of postpartum oral diuretics for the prevention of postpartum HTN among women with HDP remains inconclusive, larger studies are needed to clarify this question and identify specific HDP subgroups who would derive particular benefit Diuretic administration reduces systolic BP and may reduce the risk of HTN up to 10 days postpartum for some clinical subgroups. Use of postpartum diuretics guided by thoughtful assessment of clinical volume status remains reasonable.

Perspectives

This meta-analysis highlights the potential, yet still uncertain, role of postpartum diuretics in BP management among individuals with HDPs. While diuretic use was associated with a significant, but clinically small, reduction in systolic BP, the lack of a significant impact on persistent hypertension underscores the need for more definitive evidence. Given the heterogeneity in study designs, patient populations, and control groups, future research should prioritize large randomized controlled trials with stratified analyses by HDP subtypes.

Prompt postpartum BP control may have durable implications beyond the immediate postpartum period, potentially influencing long-term cardiovascular health through prevention of adverse cardiac remodeling and progression to chronic hypertension. Importantly, the signal towards benefit in specific HDP subtypes suggests that diuretics may hold promise as a first-line postpartum antihypertensive strategy in select populations. Identifying patient characteristics, such as HDP subtype or severity, which predict diuretic responsiveness may refine postpartum BP management and ultimately reduce downstream cardiovascular risk.

NOVELTY AND RELEVANCE.

What is New?

This systematic review and meta-analysis is updated with four recent studies and represents the current state of knowledge on the role of postpartum diuretic use for blood pressure (BP) management among individuals with hypertensive disorders of pregnancy (HDP). Although we were underpowered for our primary outcome of persistent hypertension, we demonstrated a significant, albeit clinically small, improvement in systolic BP.

What is Relevant?

Postpartum diuretics reduce systolic BP among individuals with HDP. Larger studies are needed to definitively evaluate the benefit of postpartum diuretics on persistent hypertension, particularly in certain HDP subgroups.

Clinical/Pathophysiological Implications

This study underscores that postpartum diuretics lower systolic BP and may represent an important tool for a tailored approach to postpartum BP management. Postpartum diuretics may play an important role in the continuum of care for individuals with HDP, bridging pregnancy-related care with long-term cardiovascular disease prevention through improved BP control.

ABBREVIATIONS

HDP

hypertensive disorder of pregnancy

HTN

hypertension

BP

blood pressure

RCT

randomized controlled trials