Antihypertensive drug therapy for mild to moderate hypertension during pregnancy

  • Review
  • Intervention

Authors


Abstract

Background

Mild to moderate hypertension during pregnancy is common. Antihypertensive drugs are often used in the belief that lowering blood pressure will prevent progression to more severe disease, and thereby improve the outcome.

Objectives

To assess the effects of antihypertensive drug treatments for women with mild to moderate hypertension during pregnancy.

Search methods

We searched the Cochrane Pregnancy and Childbirth Group's Trials Register (30 April 2013) and reference lists of retrieved studies.

Selection criteria

All randomised trials evaluating any antihypertensive drug treatment for mild to moderate hypertension during pregnancy defined, whenever possible, as systolic blood pressure 140 to 169 mmHg and diastolic blood pressure 90 to 109 mmHg. Comparisons were of one or more antihypertensive drug(s) with placebo, with no antihypertensive drug, or with another antihypertensive drug, and where treatment was planned to continue for at least seven days.

Data collection and analysis

Two review authors independently extracted data.

Main results

Forty-nine trials (4723 women) were included. Twenty-nine trials compared an antihypertensive drug with placebo/no antihypertensive drug (3350 women). There is a halving in the risk of developing severe hypertension associated with the use of antihypertensive drug(s) (20 trials, 2558 women; risk ratio (RR) 0.49; 95% confidence interval (CI) 0.40 to 0.60; risk difference (RD) -0.10 (-0.13 to -0.07); number needed to treat to harm (NNTH) 10 (8 to 13)) but little evidence of a difference in the risk of pre-eclampsia (23 trials, 2851 women; RR 0.93; 95% CI 0.80 to 1.08). Similarly, there is no clear effect on the risk of the baby dying (27 trials, 3230 women; RR 0.71; 95% CI 0.49 to 1.02), preterm birth (15 trials, 2141 women; RR 0.96; 95% CI 0.85 to 1.10), or small-for-gestational-age babies (20 trials, 2586 women; RR 0.97; 95% CI 0.80 to 1.17). There were no clear differences in any other outcomes.

Twenty-two trials (1723 women) compared one antihypertensive drug with another. Alternative drugs seem better than methyldopa for reducing the risk of severe hypertension (11 trials, 638 women; RR (random-effects) 0.54; 95% CI 0.30 to 0.95; RD -0.11 (-0.20 to -0.02); NNTH 7 (5 to 69)). There is also a reduction in the overall risk of developing proteinuria/pre-eclampsia when beta blockers and calcium channel blockers considered together are compared with methyldopa (11 trials, 997 women; RR 0.73; 95% CI 0.54 to 0.99). However, the effect on both severe hypertension and proteinuria is not seen in the individual drugs. Other outcomes were only reported by a small proportion of studies, and there were no clear differences.

Authors' conclusions

It remains unclear whether antihypertensive drug therapy for mild to moderate hypertension during pregnancy is worthwhile.

Résumé scientifique

Traitement antihypertenseur pour l'hypertension légère à modérée pendant la grossesse

Contexte

Une hypertension légère à modérée pendant la grossesse est fréquente. Les antihypertenseurs sont souvent utilisés, avec l'idée que le fait de baisser la pression sanguine empêchera la progression vers une maladie plus grave, et par conséquent améliorera le résultat.

Objectifs

Évaluer les effets des traitements antihypertenseurs sur les femmes avec une hypertension légère à modérée pendant la grossesse.

Stratégie de recherche documentaire

Nous avons effectué des recherches dans le registre des essais du groupe Cochrane sur la grossesse et l'accouchement (le 30 avril 2013) et les références bibliographiques des études trouvées.

Critères de sélection

Tous les essais randomisés évaluant un traitement antihypertenseur utilisé pour une hypertension légère à modérée pendant la grossesse définie, lorsque possible, comme une pression sanguine systolique de 140 à 169 mmHg et une pression sanguine diastolique de 90 à 109 mmHg. Les comparaisons étaient effectuées entre un ou plusieurs antihypertenseurs et un placebo, aucun autre antihypertenseur, ou un autre antihypertenseur, et lorsque le traitement était sensé durer au moins sept jours.

Recueil et analyse des données

Deux auteurs de la revue ont indépendamment extrait les données.

Résultats principaux

Quarante-neuf essais (4 723 femmes) ont été inclus. Vingt-neuf essais comparaient un antihypertenseur avec un placebo / aucun antihypertenseur (3 350 femmes). Nous avons constaté une réduction de moitié du risque de développer une hypertension sévère associée à l'utilisation de médicament(s) antihypertenseur(s) (20 essais, 2 558 femmes ; risque relatif (RR) 0,49 ; intervalle de confiance (IC) à 95 % 0,40 à 0,60 ; différence de risques (DR) -0,10 (-0,13 à -0,07) ; nombre nécessaire pour nuire (NNN) 10 (8 à 13)), mais peu de preuves d'une différence dans le risque de pré-éclampsie (23 essais, 2 851 femmes ; RR 0,93 ; IC à 95 % 0,80 à 1,08). De même, il n'existe aucun effet évident sur le risque de décès du bébé (27 essais, 3 230 femmes ; RR 0,71 ; IC à 95 % 0,49 à 1,02), d'accouchement prématuré (15 essais, 2 141 femmes ; RR 0,96 ; IC à 95 % 0,85 à 1,10), ou de bébés petits pour l'âge gestationnel (20 essais, 2 586 femmes ; RR 0,97 ; IC à 95 % 0,80 à 1,17). Il n'y avait aucune différence claire dans les autres critères de jugement.

Vingt-deux essais (1 723 femmes) comparaient un antihypertenseur à un autre. Les autres médicaments semblaient plus efficaces que la méthyldopa pour réduire le risque d'hypertension sévère (11 essais, 638 femmes ; RR (effets aléatoires) 0,54 ; IC à 95 % 0,30 à 0,95 ; DR -0,11 (-0,20 à -0,02) ; NNN 7 (5 à 69)). On observe également une réduction du risque global de développer une protéinurie / une pré-éclampsie lorsque les bêta-bloquants et les inhibiteurs des canaux calciques analysés ensemble sont comparés à la méthyldopa (11 essais, 997 femmes ; RR 0,73 ; IC à 95 % 0,54 à 0,99). Cependant, l'effet sur l'hypertension sévère et sur la protéinurie n'est pas observé pour les médicaments individuels. Les autres critères de jugement étaient uniquement mentionnés pour une proportion réduite d'études, et il n'a été constaté aucune différence précise.

Conclusions des auteurs

L'incertitude persiste quant à l'utilité d'un traitement antihypertenseur pour l'hypertension légère à modérée pendant la grossesse.

Plain language summary

Antihypertensive drug therapy for mild to moderate hypertension during pregnancy

There is not enough evidence to show whether antihypertensive drug treatment for mild to moderate hypertension during pregnancy is worthwhile.

During the early weeks of normal pregnancy, blood pressure falls and climbs slowly in later pregnancy to reach pre-pregnancy levels at term. Mild to moderate hypertension (high blood pressure) is common during pregnancy. In some women, it can become more serious, resulting in hospital admission, pre-eclampsia (a complication of pregnancy that includes high blood pressure) and possible premature delivery. Antihypertensive drugs are often used to lower blood pressure in the belief that they will prevent this progression. This review of 49 trials, involving 4723 women, found there was not enough evidence to show the benefit of antihypertensive drugs for mild to moderate hypertension during pregnancy. More research is needed.

Résumé simplifié

Traitement antihypertenseur pour l'hypertension légère à modérée pendant la grossesse

Il n'existe pas suffisamment de données permettant de démontrer si un traitement hypertenseur utilisé pour une hypertension légère à modérée pendant la grossesse est valable.

Au cours des premières semaines d'une grossesse normale, la tension artérielle chute, et remonte ensuite lentement en fin de grossesse pour atteindre à terme les niveaux d'avant la grossesse. L'hypertension (pression artérielle élevée) légère à modérée est fréquente pendant la grossesse. Chez certaines femmes, elle peut devenir plus grave, entraînant une hospitalisation, la pré-éclampsie (une complication de la grossesse qui inclut une pression artérielle élevée) et potentiellement un accouchement prématuré. Les médicaments antihypertenseurs sont souvent utilisés pour faire baisser la pression sanguine dans l'optique d'empêcher cette progression. Cette revue de 49 essais, totalisant 4 723 femmes, a conclu qu'il n'y avait pas suffisamment de données pour démontrer l'avantage des antihypertenseurs en cas d'hypertension légère à modérée pendant la grossesse. Des recherches supplémentaires doivent être effectuées.

Notes de traduction

Traduit par: French Cochrane Centre 22nd June, 2014
Traduction financée par: Financeurs pour le Canada : Instituts de Recherche en Santé du Canada, Ministère de la Santé et des Services Sociaux du Québec, Fonds de recherche du Québec-Santé et Institut National d'Excellence en Santé et en Services Sociaux; pour la France : Ministère en charge de la Santé

Background

Description of the condition

During the early weeks of normal pregnancy blood pressure falls, climbing slowly in later pregnancy to reach pre-pregnancy levels at term (Hytten 1980; Villar 1989). These changes are related to multiple physiological and environmental factors, and complicate the diagnosis of hypertension during pregnancy. There is no consensus about the definition of hypertensive disorders during pregnancy (Chappell 1999), and several classifications have been suggested (ASSHP 1993; Davey 1988; Gifford 1990; North 1999; Roberts 1993). Nevertheless, there is now general agreement about the broad categories. These are: (a) gestational hypertension or pregnancy-induced hypertension, which is hypertension without proteinuria; (b) pre-eclampsia, which is hypertension with proteinuria; (c) chronic hypertension, or essential hypertension, which is pre-existing hypertension; and (d) chronic hypertension with superimposed pre-eclampsia.

Variations in the systems for classification are largely to do with how high blood pressure is defined. The system suggested by the International Society for the Study of Hypertension in Pregnancy defines hypertension as a diastolic blood pressure of 90 mmHg or above on two consecutive occasions at least four hours apart, or a single diastolic blood pressure of 110 mmHg or more (Davey 1988). In the past there was disagreement about which auscultatory sound to use for measuring diastolic blood pressure. However, Korotkoff phase V (disappearance of sounds) is now widely recommended as more reliable than phase IV (muffling) (Brown 2001; Rubin 1996).

The suggestion that a change in blood pressure is more important than any absolute level (Redman 1988) is no longer included in the definition of gestational hypertension due to lack of evidence that it is related to outcome (Brown 2000; Brown 2001; NHBPEP 2000). Pre-eclampsia is defined as high blood pressure (using the criteria above) plus significant proteinuria, usually taken as at least 300 mg/24 hour or 1+ on dipsticks (Davey 1988).

For this review we have accepted broad and pragmatic criteria for identifying women with mild to moderate hypertension during pregnancy. This reflects clinical practice, and is justifiable in the context of randomised trials as within each study the same criteria will have been used for women in both groups.

Hypertension during pregnancy is common. One in 10 women will have high blood pressure at some time before delivery, and pre-eclampsia complicates between 2% to 8% of pregnancies (WHO 1988). Pre-eclampsia is discussed in more detail in the generic protocol of interventions for prevention of pre-eclampsia (Meher 2005).

Description of the intervention

A wide variety of drugs have been advocated for lowering blood pressure in hypertensive pregnant women, and each group has different potential side-effects and adverse events. In this review we evaluate individual agents within the class or family to which they belong, as each class has a similar mechanism of action. Alpha agonists inhibit vasoconstriction via a centrally mediated effect (Ingenito 1970). Methyldopa is the most widely used alpha agonist, and became available in 1963. Clonidine is also an alpha agonist, although it has the disadvantage that sudden withdrawal may cause a hypertensive crisis (Isaac 1980). Beta-adrenoceptor blocking drugs block adrenoceptors in the heart, peripheral blood vessels, airways, pancreas and liver (Frishman 1979). Labetalol has an additional arteriolar vasodilating action that lowers peripheral resistance, but is usually classified with the beta blockers. Calcium channel blockers include isradipine, nifedipine, nicardipine, nimodipine and verapamil. These drugs inhibit influx of calcium ions to vascular smooth muscle resulting in arterial vasodilatation (Robinson 1980). Hydralazine is a vasodilator with a direct relaxing effect on smooth muscle in the blood vessels, predominantly in the arterioles (Stunkard 1954). Ketanserin, is a selective serotonin receptor antagonist with weak adrenergic receptor blocking properties (Frishman 1995). The drug is effective in lowering blood pressure in essential hypertension. It also inhibits platelet aggregation. Glyceryl trinitrate is a nitric oxide donor with vasodilator effect in perivascular smooth-muscle cells (Seligman 1994).

How the intervention might work

The role of antihypertensive therapy for pregnant women with mild to moderate hypertension is unclear. As there is no immediate need to lower blood pressure, the rationale for treatment is that it will prevent or delay progression to more severe disease, thereby benefiting the woman or her baby, or both, and reducing consumption of health service resources. As well as reducing blood pressure, the belief has been that these drugs reduce the risk of preterm delivery and placental abruption and improve fetal growth.

Why it is important to do this review

The aim of this review is to assess the potential benefits and hazards, to the woman and baby, of antihypertensive drugs for the treatment of mild to moderate hypertension during pregnancy. If antihypertensive agents are overall beneficial, a secondary aim will be to assess the comparative effects of alternative agents.

There are other types of interventions for women with mild to moderate hypertension during pregnancy that are not considered in this review. Interventions covered by other reviews include salt restriction (Duley 1999), antiplatelet agents (Duley 2007), abdominal decompression (Hofmeyr 2012) and bed rest, with or without hospitalisation (Meher 2005a). Diuretics are no longer widely used in pregnancy, and are usually reserved for women with renal or cardiac problems (ASSHP 1993; CHSCC 1997). The role of diuretics for women with hypertension during pregnancy is covered by a separate Cochrane review (Churchill 2007), as is prevention and treatment of postpartum hypertension (Magee 2013).

For women with severe hypertension, usually defined as 160 to 170 mmHg or more systolic blood pressure or 110 mmHg or more diastolic blood pressure, there is a risk of direct arterial damage and so antihypertensive drugs are used to lower blood pressure (Gifford 1990; Redman 1993). The question of which drug is best in this situation is considered in another Cochrane review and not discussed further here (Duley 2013).

A separate review assessing the effect of oral beta blockers in mild to moderate hypertension during pregnancy is available (Magee 2003). However, beta blockers are included in this review as part of all the spectrum of antihypertensive drugs.

Objectives

To determine the possible benefits, risks and side-effects of antihypertensive drug treatments for women with mild to moderate hypertension during pregnancy (defined whenever possible as a systolic blood pressure of 140 to 169 mmHg or diastolic blood pressure of 90 to 109 mmHg, or both). Also, to compare the differential effects of alternative drug regimens.

The comparisons are of:

  1. any antihypertensive drug with either no drug or placebo;

  2. one antihypertensive drug compared with another. For this review, the commonly used drugs are regarded as control and compared with all other agents (for example, any antihypertensive versus methyldopa, any antihypertensive versus calcium channel blockers).

Methods

Criteria for considering studies for this review

Types of studies

All randomised trials evaluating any antihypertensive drug treatment for mild to moderate hypertension during pregnancy. Quasi-randomised designs were excluded.

Types of participants

The review includes women with mild to moderate hypertension during pregnancy, defined whenever possible as those with systolic blood pressure 140 to 169 mmHg and diastolic blood pressure 90 to 109 mmHg. Studies in which participants were described as having 'mild to moderate' hypertension but the range of blood pressures was not clearly specified were also included. Women were included regardless of whether they had proteinuria or not, and irrespective of previous antihypertensive treatment or whether the pregnancy was singleton or multiple.

Women who had given birth before trial entry were excluded, as were women with severe hypertension (defined whenever possible as either systolic blood pressure of 170 mmHg or more, or diastolic blood pressure 110 mmHg or more). Studies that included a substantial proportion of women who did not have mild to moderate hypertension were excluded, unless data were available on outcomes for those with mild to moderate hypertension only.

Types of interventions

Any comparison of one or more antihypertensive drug with either placebo or no antihypertensive drug was included, as were comparisons of one antihypertensive drug with another. Studies were excluded if the intention was to treat for less than seven days, as a longer period of treatment would be necessary for any substantive clinical effect. Comparisons of two drugs of the same class were also excluded, although these may be included in future updates if clinically relevant.

Drugs that aimed to reduce the risk of pregnancy-induced hypertension progressing to pre-eclampsia but are not antihypertensive agents were also excluded.

Types of outcome measures

Primary outcomes

Main outcomes were prespecified as follows.

  • Severe hypertension: defined whenever possible as either systolic blood pressure 170 mmHg or more, or diastolic blood pressure 110 mmHg or more. Trials where the definition of severe hypertension was not clear, or where the cut-off was up to 10 mmHg lower were also included and were clearly documented.

  • Proteinuria: defined whenever possible as new proteinuria (1+ or more or 300 mg/24 hour).

  • Any reported baby death: fetal deaths included miscarriage (fetal losses before viability, usually taken as 20 or 24 weeks) and stillbirths (after 24 weeks, or however defined). Perinatal deaths are stillbirths plus deaths in the first week of life. Neonatal deaths are deaths in the first 28 days.

  • Small-for-gestational age: low birthweight for gestational age, below the third, fifth or 10th percentile, using the most severe reported.

  • Preterm birth: all births before 37 completed weeks.

Secondary outcomes
(i) For the woman
  • Severe pre-eclampsia: defined whenever possible as severe hypertension with proteinuria 2+ or more, or 2 g or more/24 hours, with or without other signs of symptoms, or as moderate hypertension with proteinuria 3+ or more. Haemolysis, elevated liver enzymes and low platelets (HELLP) syndrome is a form of severe pre-eclampsia and was therefore included here as well as a separate measure. Trials reporting imminent eclampsia, or where the definition of severe pre-eclampsia was not clear were also included.

  • Eclampsia.

  • HELLP syndrome.

  • Severe maternal morbidity: such as liver or renal failure, disseminated intravascular coagulation and cerebrovascular accident (stroke).

  • Need for another drug to control blood pressure.

  • Miscarriage (fetal losses before viability, usually taken as before 20 or 24 weeks).

  • Elective delivery: combines elective caesarean sections and elective induction of labour at term or before term.

  • Caesarean section.

  • Antenatal hospital admission and length of stay more than seven days: hospital and day care unit were to be reported separately.

  • Placental abruption.

  • Side-effects: any reported side-effects or severe adverse events.

  • Drug stopped because of side-effects.

(ii) For the baby
  • Very low, less than four, five-minute Apgar score.

  • Severe prematurity, all births less than 32 or less than 34 weeks' gestation.

  • Admission to neonatal or intensive care nursery.

  • Respiratory distress syndrome.

  • Other morbidity possibly related to maternal drug therapy, such as hypo or hypertension, hypoglycaemia and bradycardia (with beta blockers).

  • Impaired long-term growth and development in infancy and childhood.

Search methods for identification of studies

Electronic searches

We searched the Cochrane Pregnancy and Childbirth Group's Trials Register by contacting the Trials Search Co-ordinator (30 April 2013). 

The Cochrane Pregnancy and Childbirth Group’s Trials Register is maintained by the Trials Search Co-ordinator and contains trials identified from: 

  1. monthly searches of the Cochrane Central Register of Controlled Trials (CENTRAL);

  2. weekly searches of MEDLINE;

  3. weekly searches of Embase;

  4. handsearches of 30 journals and the proceedings of major conferences;

  5. weekly current awareness alerts for a further 44 journals plus monthly BioMed Central email alerts.

Details of the search strategies for CENTRAL, MEDLINE and Embase, the list of handsearched journals and conference proceedings, and the list of journals reviewed via the current awareness service can be found in the ‘Specialized Register’ section within the editorial information about the Cochrane Pregnancy and Childbirth Group

Trials identified through the searching activities described above are each assigned to a review topic (or topics). The Trials Search Co-ordinator searches the register for each review using the topic list rather than keywords.

For detail of additional searching carried out for the previous version (Abalos 2007), see Appendix 1.

Searching other resources

We searched reference lists of retrieved studies.

We did not apply any language restrictions.

Data collection and analysis

For the methods used when assessing the trials identified in the previous version of this review, see Abalos 2007.

For this update we used the following methods when assessing the reports identified by the updated search.

Selection of studies

Two authors (E Abalos (EA), L Duley (LD)) assessed for inclusion all potential studies identified as a result of the search strategy. Disagreements were resolved through discussion. If agreement could not be reached, a third review author (DW Steyn (DWS)) was consulted.

Data extraction and management

Data were extracted in an ad-hoc form and entered into Review Manager (RevMan 2012) by EA, and checked by LD. Review authors were not blinded to the authors, sources of the articles, or results. Discrepancies were resolved by discussion between the two review authors and, when necessary, the third review author (DWS) was consulted. When information regarding any of the above was unclear, we attempted to contact authors of the original reports to provide further details.

Assessment of risk of bias in included studies

One review author (EA) assessed the risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions ( Higgins 2011). Results were checked and agreed by a second review author (LD). Disagreements were resolved by discussion or by involving the third review author.

(1) Random sequence generation (checking for possible selection bias)

The method used to generate the allocation sequence was described for each included study in sufficient detail to allow an assessment of whether it should produce comparable groups.

We assessed the method as:

  • low risk of bias (any truly random process, e.g. random number table; computer random number generator);

  • high risk of bias (any non-random process, e.g. odd or even date of birth; hospital or clinic record number);

  • unclear risk of bias.   

(2) Allocation concealment (checking for possible selection bias)

The method used to conceal allocation to interventions prior to assignment was described (when information was available) for each included study and whether intervention allocation could have been foreseen in advance of, or during recruitment, or changed after assignment was assessed.

We assessed the methods as:

  • low risk of bias (e.g. telephone or central randomisation; consecutively numbered sealed opaque envelopes);

  • high risk of bias (open random allocation; unsealed or non-opaque envelopes, alternation; date of birth);

  • unclear risk of bias.   

(3.1) Blinding of participants and personnel (checking for possible performance bias)

We considered that studies were at low risk of bias if they were blinded, or if we judged that the lack of blinding would be unlikely to affect results. We assessed blinding separately for different outcomes or classes of outcomes.

We assessed the methods as:

  • low, high or unclear risk of bias for participants;

  • low, high or unclear risk of bias for personnel.

(3.2) Blinding of outcome assessment (checking for possible detection bias)

We evaluated for each included study the methods used, if any, to blind outcome assessors from knowledge of which intervention a participant received. We assessed blinding separately for different outcomes or classes of outcomes.

We assessed methods used to blind outcome assessment as:

  • low, high or unclear risk of bias.

(4) Incomplete outcome data (checking for possible attrition bias due to the amount, nature and handling of incomplete outcome data)

We described for each included study, and for each outcome or class of outcomes, the completeness of data including attrition and exclusions from the analysis. We stated whether attrition and exclusions were reported and the numbers included in the analysis at each stage (compared with the total randomised participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes. Where sufficient information was reported, or could be supplied by the trial authors, we re-included missing data in the analyses.

We assessed methods as:

  • low risk of bias (e.g. no missing outcome data; missing outcome data less tan 10% of randomised women);

  • high risk of bias (e.g. numbers or reasons for missing data imbalanced across groups; ‘as treated’ analysis done with substantial departure of intervention received from that assigned at randomisation, missing outcome data more than 10% of randomised patients);

  • unclear risk of bias.

Trials were excluded if it was not possible to enter data on an intention-to-treat basis or if 20% or more participants were excluded, or both.

(5) Selective reporting (checking for reporting bias

For each included study the possibility of selective outcome reporting bias was sought.

We assessed the methods as:

  • low risk of bias (where it is clear that all of the study’s pre-specified outcomes and all expected outcomes of interest to the review have been reported either in the paper or through personal communication);

  • high risk of bias (where not all the study’s pre-specified outcomes have been reported; one or more reported primary outcomes were not pre-specified; outcomes of interest are reported incompletely and so cannot be used; study fails to include results of a key outcome that would have been expected to have been reported);

  • unclear risk of bias.

(6) Other bias (checking for bias due to problems not covered by (1) to (5) above)

We assessed whether each study was free of other problems that could put it at risk of bias:

  • low risk of other bias;

  • high risk of other bias;

  • unclear whether there is risk of other bias.

(7) Overall risk of bias

We made explicit judgements about whether studies were at high risk of bias, according to the criteria given in the Cochrane Handbook (Higgins 2011).

Measures of treatment effect

Dichotomous data

We only collected dichotomous data, presenting the results as summary risk ratio with 95% confidence intervals. 

Continuous data

Continuous data were not collected.

Unit of analysis issues

The unit of analysis was individual randomised women.

Cluster-randomised trials

No cluster-randomised trials were found for this review. If we had identified cluster-randomised trials, they would have been eligible for inclusion.

Cross-over trials

Given the objectives of this review, cross-over trials were not included.

Other unit of analysis issues

No other units of analyses were used in this review.

Dealing with missing data

For all outcomes, we conducted the analyses on an intention-to-treat basis, i.e. we attempted to include all participants randomised to each group in the analyses, and all participants were analysed in the group to which they were allocated, regardless of whether or not they received the allocated intervention. The denominator for each outcome in each trial was the number randomised minus any participants whose outcomes were known to be missing. Trials were excluded if it was not possible to enter data on an intention-to-treat basis, or if 20% or more participants were excluded, or both.

Assessment of heterogeneity

Statistical heterogeneity in each meta-analysis was assessed using the Tau², I² and Chi² statistics. We regarded heterogeneity as substantial if the I² was greater than 30% and either the Tau² was greater than zero, or there was a low P value (less than 0.10) in the Chi² test for heterogeneity. 

Assessment of reporting biases

We investigated reporting biases (such as publication bias) using funnel plots. We assessed funnel plot asymmetry visually, and used formal tests for funnel plot asymmetry. As all outcomes were dichotomous, we used the test proposed by Harbord 2006.

Data synthesis

We carried out statistical analysis using the Review Manager software (RevMan 2012). We used fixed-effect meta-analysis for combining data where it was reasonable to assume that studies were estimating the same underlying treatment effect: i.e. where trials were examining the same intervention, and the trials’ populations and methods were judged sufficiently similar. If there was clinical heterogeneity sufficient to expect that the underlying treatment effects differed between trials, or if substantial statistical heterogeneity was detected, we used random-effects meta-analysis to produce an overall summary, if an average treatment effect across trials was considered clinically meaningful. The random-effects summary was treated as the average of the range of possible treatment effects as an input to discuss the clinical implications of treatment effects differing between trials. Trials were not combined if the average treatment effect was not clinically meaningful.

Subgroup analysis and investigation of heterogeneity

For the comparison of antihypertensive drug/s with placebo or no treatment the following subgroup analyses were prespecified:

  1. by class of drug (such as alpha agonists, beta blockers and calcium channel blockers);

  2. by type of hypertensive disorder at trial entry: mild to moderate hypertension alone; mild to moderate hypertension with proteinuria; chronic hypertension; unspecified;

  3. by gestational age at trial entry: less than about 32 weeks' gestation; about 32 weeks or more gestation; or unclassified/mixed;

  4. by whether a placebo was used: placebo, no placebo.

The subgroup analysis by type of drug is presented for all outcomes. The remaining subgroups are presented for the prespecified primary outcomes only.

We assessed subgroup differences by interaction tests available within RevMan (RevMan 2012). We reported the results of subgroup analyses quoting the Chi² statistic and P value, and the interaction test I² value.

Sensitivity analysis

Sensitivity analyses were not conducted.

Results

Description of studies

Details for each trial can be found in the Characteristics of included studies table and the Characteristics of excluded studies table.

Included studies

Forty-nine trials (4723 women) were included in this review. Of these, 34 (3480 women) were conducted in industrialised countries (Australia, France, Hong Kong, Ireland, Israel, Italy, Sweden, UK and USA), and 15 (1243 women) were performed in middle- or low-income countries (Argentina, Brazil, Caribbean Islands, India, South Africa, Sudan and Venezuela). Three trials were published in the 1960s and 1970s, 22 in the 1980s, 17 in the 1990s, and seven after the year 2000. All included trials are small. The largest study recruited 300 women. Two three-arm trials of methyldopa versus labetalol versus no treatment (India 2012; USA 1990) were included in the comparison of any antihypertensive with placebo/no antihypertensive, and in the comparison of one drug with another. Only five studies had comparison arms containing more than 100 women.

(One report from an updated search in April 2013 has been added to Studies awaiting classification.)

Interventions

The antihypertensive drugs used in these trials include: alpha agonists (methyldopa), beta blockers (acebutolol, atenolol, labetalol, mepindolol, metoprolol, pindolol, oxprenolol and propranolol), calcium channel blockers (isradipine, nicardipine, nifedipine, nimodipine and verapamil), vasodilators (hydralazine and prazozin), ketanserin and glyceryl trinitrate (GTN). All drugs were given orally, except glyceryl trinitrate, which was given transdermally. The dose for several agents varied considerably between studies, in both amount and duration of therapy.

The antihypertensive drug was compared with placebo, or no antihypertensive drug, in 29 trials (3350 women). Of these trials, 16 evaluated beta blockers (1602 women), seven using a placebo for the control group and nine comparing with no treatment. Methyldopa was evaluated in eight trials (986 women), one comparison with placebo, and seven with no antihypertensive treatment. One trial (118 women) compared isradipine with placebo, another trial (199 women) compared verapamil with placebo, and three studies (583 women) compared nifedipine with no drug treatment. Prazozin was compared with placebo in one trial (32 women), and GTN was compared with placebo patches in another (16 women).

Alternative drug regimens were compared in 22 trials (1723 women). Twenty of these studies compared methyldopa with another agents. In 16 trials (1507 women) the comparison was with beta blockers, in two it was nifedipine (49 women), in one (111 women) it was nimodipine and in another ketanserin (20 women). One small trial (36 women) compared nifedipine with glyceryl trinitrate. In one study (100 women), metoprolol was compared with nicardipine.

Gestation at trial entry

Twenty of the 49 included studies recruited women during the second and third trimester of pregnancy, and 19 recruited only during the third trimester. Two studies recruited women during the first trimester (Argentina 1988; USA 1990). Gestational age at trial entry was not reported in eight studies.

Severity and type of hypertension disease at trial entry

Mild to moderate hypertension was defined as a diastolic blood pressure of 90 mmHg or more in 35 studies. In seven trials, the definition was 95 mmHg or more. In two trials, the cut-off was 85 mmHg, In five studies, authors merely stated 'mild to moderate hypertension', or 'pregnancy-induced hypertension', or 'diagnosed hypertension'. Women with proteinuria were excluded from trial entry in 17 studies whilst in six trials all women had proteinuria at recruitment. Twelve trials included women regardless of whether or not they had proteinuria, and the proportion of women with proteinuria ranged from 4% to 47%. In the remaining 14 trials the presence of proteinuria at trial entry was not reported.

Eight studies only recruited women with chronic hypertension. Women with chronic hypertension were excluded from 16 trials (16/49). Nine trials included women regardless of whether or not they had chronic hypertension, although outcomes were often not reported separately. In the remaining 16 trials, chronic hypertension at trial entry was not mentioned.

Methods for measuring blood pressure

Only four trials masked the assessment of blood pressure by using a random zero sphygmomanometer (Australia 1983; UK 1976; UK 1983; UK 1983a). For assessment of diastolic blood pressure, Korotkoff phase IV sound was used in 14 trials and Korotkoff phase V was used for eight studies. Criteria for blood pressure measurement were not mentioned in 27 trials.

Definition of small-for-gestational age

Small-for-gestational age was defined in a variety of ways in the 29 trials reporting this outcome. Four studies used birthweight below the fifth centile and 11 below the 10th centile. Five trials used other definitions, and in the remaining nine trials, small-for-gestational age was not defined.

One outcome specified in our protocol, very low (less than four) five-minute Apgar score, is not included in this review as it was not reported by any trial.

Excluded studies

Eighty studies were excluded from the review. Of these, 36 were conducted in high-income countries (Australia, Belgium, Denmark, Finland, France, Germany, Hong Kong, Israel, Italy, Japan, Spain, Sweden, UK and USA), and 44 in low- and middle-income countries (Argentina, Brazil, China, Cuba, Czech Republic, Dominican Republic, Egypt, Hungary, India, Iran, Kuwait, Mexico, Pakistan, Panama, Philippines, Russia, Singapore, Slovakia, South Africa, Sri Lanka, Uganda and Venezuela). The oldest excluded study was published 1957, one was published in 1978, 54 were published in the 1980s and 1990s, and 24 have been published since the millennium.

The language of publication was English (for 63 papers), Chinese (five), Spanish (five), Portuguese (three), Czech (two), French (one), and Russian (one). Language was not a reason for exclusion. Twenty-two papers (22/80) were published only as congress abstracts. Authors of 11 papers provided additional information about methods and/or clinical outcomes.

The reasons for exclusions were as follows.

  • Methodological problems (22 studies): either they were not randomised trials (11 studies) or they used quasi-random methods for treatment allocation (seven), or more than 20% of women were excluded after randomisation (four).

  • Participants were not eligible for the review (nine studies): either some or all of the women had severe hypertension (six), or some women had normal blood pressure (three), and data were not presented separately for the women with mild to moderate hypertension.

  • Intervention was not eligible for the review (38 studies): either the comparison was of drugs within the same class (eight), or the allocated intervention was for less than seven days (13), or it was not an antihypertensive drug (16) or mixed control group (1).

  • No clinical outcomes reported (11 studies): there are no data on relevant clinical outcomes (eight congress abstracts, of which four authors were contacted but with no responses to date). Two reports were personal communications of planned randomised controlled trials, but no information could be found about their completion.

Risk of bias in included studies

Overall, the quality of the studies included in this review is moderate to poor (Figure 1; Figure 2).

Figure 1.

'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 2.

'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.

Allocation

Methods for generating the random sequence were described in 14 trials (29%). These included: computer generation (France 1994; Italy 1998; USA 1987; USA 1990; USA 1992), random-number tables (UK 1980; UK 1989; Venezuela 1988), series of random numbers (Israel 1992). Although three studies did not describe the method used for allocation, the authors mentioned that it was in blocks of six (Sweden 1995),10 (Caribbean Is.1990), or nine at each centre (France 1987). Inadequate methods were used in two trials, namely "mixed-up opaque envelopes" (India 2012) and "cards shuffled into a random order and numbered in sequence" (Ireland 1991).

Concealment of allocation was adequate for only 17 of the 49 trials (35%), being unclear whether concealment was adequate in the remaining 32 trials. Methods for concealing the allocation included telephone randomisation (Italy 1998; Sweden 1984), blinded treatment packs (Brazil 1988; Brazil 2000a; Caribbean Is.1990; Israel 1992; Italy 1999; Italy 2000; UK 1989), consecutive, sealed identical envelopes (UK 1992) or blinded or sealed envelopes (France 1987; France 1994; India 2012; Ireland 1991; USA 1987; USA 1990; USA 1992). Most trials with unclear concealment of allocation were described as 'randomised' with no details on how this was achieved. Some of these studies were stated to be double blind, but with no information about how this was achieved. Three trials with unclear concealment used random-number tables without mentioning any attempt to conceal the allocation (Sweden 1995; UK 1980; Venezuela 1988).

Blinding

All trials evaluating alternative treatments for mild/moderate hypertension during pregnancy were open label trials, and no attempts were made to blind outcome assessment in order to prevent performance or detection bias. From 29 trials evaluating whether or not hypertension should be treated with antihypertensives, only 12 (41%) compared the active drug with placebo, 10 of them were described by authors as double blind (Brazil 2000a; Caribbean Is.1990; Hong Kong 1990; Israel 1992; Sweden 1984; Sweden 1995; UK 1983; UK 1989; UK 1990; USA 1987a). One placebo-controlled trial was reported as single blind (Australia 2001), and another, even though placebo controlled, did not mention whether it was single or double blinded (South Africa 1991).

Incomplete outcome data

As per protocol, trials were excluded from the review (or a particular outcome was excluded from the analysis) if it was not possible to enter data on an intention-to-treat basis, or lf 20% or more participants were excluded or withdrawn. We also downgraded the score of a trial to a high risk of bias status if more than 10% of randomised women were withdrawn from the analysis. Most trials reported the outcomes for all (or nearly all) women randomised, although only four trials described the women who met the study eligibility criteria, but were not randomised (Sweden 1984; Sweden 1985; UK 1976; UK 1983). Four trials reported losses greater than 10%: Italy 1999 (17%), South Africa 1993 (10%), UK 1990 (12%) and USA 1990 (12%).

Selective reporting

Thirty-three trials (67%) reported on the risk of a woman for developing severe hypertension (one of the primary outcomes of this review). However, all trials evaluated the effect that the antihypertensive under scrutiny may have on blood pressure (mean blood pressure after treatment, % fall in blood pressure, etc.). Proteinuria (either new proteinuria or the aggravation of previously existing) was reported in 46 (46/49) trials. Fetal or neonatal deaths were evaluated in 48 trials, preterm birth in 25 and small-for-gestational-age babies in 28. Most analyses were carried out using data from published reports. Authors were contacted for unpublished information related to the main outcomes; we obtained data from Brazil 1988; Caribbean Is.1990; Italy 1999; Italy 2000; Sweden 1984; UK 1982; UK 1990; USA 1990.

Other potential sources of bias

Baseline characteristics were similar between groups in 23 trials, thus in the absence of other potential sources of bias they were scored as at low risk. In one trial (Australia 1985) co-interventions were unevenly distributed: 48% of oxprenolol group and 35% of methyldopa group received a second or third antihypertensive. In another study (Caribbean Is.1990), treatment was unblinded in 23 women (15%) and other treatment started, 16 for uncontrolled BP (five experimental, 11 control) and seven for poor compliance/side-effects (four experimental, three control).

Informed consent was mentioned in the majority of trials. In one study, informed consent was obtained only from women allocated to the treatment arm (Ireland 1991).

Sample size and power calculations were reported for five trials (Caribbean Is.1990; France 1987; Ireland 1991; Italy 1998; UK 1989).

Effects of interventions

This review includes 49 trials, involving 4723 women.

(1) Any antihypertensive drug versus none

Overall, 29 trials with a total of 3350 women compared an antihypertensive drug with placebo or no antihypertensive drug.

Primary outcomes
Severe hypertension

There is a halving in the risk of developing severe hypertension associated with the use of antihypertensive drug/s (20 trials, 2558 women; risk ratio (RR) 0.49; (95% confidence interval (CI) 0.40 to 0.60); risk difference (RD) -0.10 (-0.13 to -0.07); number needed to treat to harm (NNTH) 10 (8 to 13)), Analysis 1.3. This effect is strikingly consistent regardless of the gestation at trial entry, or whether a placebo was used for the control group. Although the magnitude of this reduction varies when the class of drug (Test for subgroup differences: Chi² = 17.03, df = 6 (P = 0.009), I² = 64.8%, Analysis 1.3) and hypertensive disorder at trial entry (Test for subgroup differences: Chi² = 10.20, df = 3 (P = 0.02), I² = 70.6%, Analysis 2.1) are considered, a beneficial trend is still observed.

Pre-eclampsia

There is no clear evidence of an overall difference in the risk of pre-eclampsia/proteinuria in the 23 trials (2851 women) reporting this outcome (RR 0.93; 95% CI 0.80 to 1.08), Analysis 1.4. These results are consistent regardless of the use of placebo. However, when other subgroups are considered, there is evidence of a difference between subgroups: class of drug (Test for subgroup differences: Chi² = 18.28, df = 8 (P = 0.02), I² = 56.2%), Analysis 1.4; type of hypertensive disorder (Test for subgroup differences: Chi² = 8.27, df = 3 (P = 0.04), I² = 63.7%), Analysis 2.2.

There is a reduction in the risk of developing proteinuria in women treated with beta blockers versus none (eight trials, 883 women; RR 0.73; 95% CI 0.57 to 0.94) Analysis 1.4.1, in those with hypertension alone at trial entry (eight trials, 684 women; RR 0.72; 95% CI 0.57 to 0.92), Analysis 2.2.1, and in two trials of 120 women recruited after 32 weeks' gestation (RR 0.34; 95% CI 0.12 to 0.96), Analysis 3.2.2. However, in trials evaluating calcium channel blockers versus none (four trials, 725 women), the risk of pre-eclampsia appears to increase (RR 1.40; 95% CI 1.06 to 1.86), Analysis 1.4.6.

Fetal or neonatal deaths

Although there is no statistically significant difference in the overall risk of the baby dying, the point estimate is for a 29% reduction associated with the use of antihypertensive drugs (27 trials, 3230 women; RR 0.71; 95% CI 0.49 to 1.02), Analysis 1.16. The only subgroup in which this reduction reaches statistical significance is for methyldopa versus none (three trials, 337 women; RR 0.35; 95% CI 0.13 to 0.94). Assessing by time of death of the baby, the only statistically significant difference was for miscarriage (seven trials, 1058 women; RR 0.39; 95% CI 0.17 to 0.93). This was also seen in 12 trials (1422 babies) with unclassified/mixed hypertensive disorders at trial entry (RR 0.55; 95% CI 0.31 to 0.96). However, no differences were observed between any of the subgroups as indicated by subgroup interaction tests.

Preterm birth (less than 37 weeks)

There is no overall difference in the 15 trials (2141 women) reporting this outcome (RR 0.96; 95% CI 0.85 to 1.10), Analysis 1.18. No differences were found between any of the subgroups considered.

Small-for-gestational age

There is no overall difference in the 20 trials (2586 women) reporting small-for-gestational-age babies (RR 0.97; 95% CI 0.80 to 1.17), Analysis 1.20. This result remains consistent across all the subgroups. However, for the comparison of beta blockers versus none, there is a strong trend towards an increase (nine trials, 904 women; RR 1.38; 95% CI 0.99 to 1.92), Analysis 1.20.1. Three of these beta blocker trials (287 women) are the only studies in the subgroup for birthweight less than fifth centile (RR 3.04; 95 % CI 1.25 to 7.40), Analysis 1.21.2.

Secondary outcomes

Of the remaining outcomes, use of additional antihypertensives was reported in 10 trials (1285 women) (RR 0.42; 95% CI 0.30 to 0.58) Analysis 1.8; changed drugs due to side-effects in 15 trials (1403 women) (RR 2.59; 95% CI 1.33 to 5.04) Analysis 1.9; caesarean section in 20 trials (2624 women) (RR 0.92; 95% CI 0.83 to 1.02) Analysis 1.14; placental abruption in 11 trials (1433 women) (RR 1.41; 95% CI 0.65 to 3.07) Analysis 1.15; and admission to special care nursery was reported in nine trials (1470 babies) (average RR 1.04; 95% CI 0.83 to 1.30; Heterogeneity: Tau² = 0.04; Chi² = 12.46, df = 8 (P = 0.13); I² = 36%), Analysis 1.22. Remaining outcomes were only reported for less than half the women in the comparison.

(2) One hypertensive drug versus another

Overall, 22 trials with a total of 1723 women compared one antihypertensive drug with another.

Primary outcomes
Severe hypertension

Beta blockers and calcium channel blockers together appear to be more effective than methyldopa in avoiding an episode of severe hypertension (11 trials, 638 women; RR (random-effects) 0.54; 95% CI 0.30 to 0.95; RD -0.11 (-0.20 to -0.02); NNTH 7 (5 to 69)), Analysis 5.1. There is no clear difference between any class of antihypertensive drugs: beta blockers versus methyldopa (nine trials, 592 women); RR 0.59; 95% CI 0.33 to 1.05, Analysis 5.1.1; calcium channel blockers versus methyldopa (two trials, 46 women) RR 0.23; 95% CI 0.04 to 1.22, Analysis 5.1.2; beta blockers versus calcium channel blockers (one trial, 100 women) 2.14; 95% CI 0.96 to 4.80, Analysis 6.1.2; and for the comparison of glyceryl trinitrate with nifedipine (one trial, 36 women), it is 1.56; 95% CI 1.07 to 35.67, Analysis 6.1.1.

Pre-eclampsia

There is a reduction in the overall risk of developing proteinuria/pre-eclampsia when either beta blockers or calcium channel blockers are compared with methyldopa (11 trials, 997 women; RR 0.73; 95% CI 0.54 to 0.99) Analysis 5.2. However, this effect is not seen in the subgroup analysis for beta blockers versus methyldopa (10 trials, 903 women; RR 0.75; 95% CI 0.53 to 1.05), Analysis 5.2.1, or for calcium channel blockers versus methyldopa (one trial, 94 women; RR 0.66; 95% CI 0.34 to 1.27), Analysis 5.2.2.

One trial (92 women) compared beta blockers with calcium channel blockers (RR 2.67; 95% CI 0.75 to 9.42), Analysis 6.2.2. One trial compared glyceryl trinitrate with calcium channel blockers, but was too small for any reliable conclusion (one trial, 36 women; RR 1.00; 95% CI 0.10 to 9.96), Analysis 6.2.1.

Total reported fetal deaths or deaths before discharge from hospital

There is no difference in the risk of the baby dying (19 trials, 1339 women; RR 0.73; 95% CI 0.42 to 1.27), Analysis 5.10, when any antihypertensive drug was compared with methyldopa. No differences were found when metoprolol or glyceryl trinitrate were compared with calcium channel blockers (two trials, 136 women; RR 1.00; 95% CI 0.06 to 15.55), Analysis 6.10.

Preterm birth (less than 37 weeks)

Only nine trials comparing any antihypertensive with methyldopa (623 women, RR 0.76; 95% CI 0.55 to 1.05), Analysis 5.11, and the small trial (36 women) of glyceryl trinitrate versus nifedipine (RR 0.63; 95% CI 0.20 to 1.91), Analysis 6.11, reported this outcome.

Small-for-gestational age

Only seven small trials reported this outcome. Six trials (577 women) compared beta blockers with methyldopa (RR 0.85; 95% CI 0.55 to 1.32), Analysis 5.12.1, and one trial (20 women) compared nifedipine versus methyldopa (RR 0.40; 95% CI 0.10 to 1.60), Analysis 5.12.2.

Secondary outcomes

Of the remaining outcomes, use of additional antihypertensives was reported in 12 trials (989 women) comparing any antihypertensive with methyldopa (RR 0.84; 95% CI 0.66 to 1.06), Analysis 5.3, and in one trial (100 women) comparing metoprolol with nicardipine (RR 2.14; 95% CI 0.96 to 4.80), Analysis 6.4. In the comparison of any antihypertensive versus methyldopa changed drugs due to side-effects was reported by four trials (272 women) (RR 2.80; 95% CI 0.12 to 67.91), Analysis 5.8; caesarean section by 10 trials (878 women) (RR 0.93; 95% CI 0.78 to 1.12) Analysis 5.6; placental abruption by one trial (173 women) (RR 2.02; 95% CI 0.19 to 21.90) Analysis 5.9; and admission to special care nursery was reported by four trials (478 babies) (RR 0.92; 95% CI 0.67 to 1.26), Analysis 5.13. Similarly, there were no clear differences in the other comparisons where these outcomes were reported.

Discussion

Summary of main results

Antihypertensive drugs half the risk that a pregnant woman with mild or moderate hypertension will have one or more episodes of severe hypertension. This is unsurprising, as the antihypertensive effects of these agents have been well demonstrated in non-pregnant people. Also unsurprising is that women allocated an antihypertensive were less likely to need another agent, and more likely to experience side-effects than those allocated placebo or no antihypertensive treatment. Between eight to 13 women need to be treated with an antihypertensive drug to prevent an episode of severe hypertension. Whether this reduction in risk would, alone, be worthwhile is likely to depend on whether there are associated reductions in the consequences of severe hypertension, such as admission to hospital and stroke. There are insufficient data for any firm conclusions about these more substantive outcomes. However, if the reduction in severe hypertension was clinically important, you might expect to see an impact in terms of fewer preterm births and fewer caesarean sections. There is no clear evidence of such an effect in this review.

Beta blockers and calcium channel blockers considered together seem to be more effective than methyldopa for preventing severe hypertension, although the comparative effects on other outcomes are unclear.

One of the main objectives in treating women with mild to moderate hypertension with antihypertensive drugs is to prevent or delay progression to pre-eclampsia. Although this review excludes a large reduction in the overall risk of pre-eclampsia associated with antihypertensive therapy in general, a moderate but clinically important effect remains possible. The confidence intervals suggest the true effect is somewhere between a 20% reduction in the risk of pre-eclampsia and a 8% increase. Although with beta blockers, or in the subgroup of women with hypertension alone at trial entry (independently of the antihypertensive drug used) this effect is likely, this should be interpreted with caution as asymmetric funnel plots in these comparisons cannot exclude the possibility of publication bias (Figure 3; Figure 4). Similarly, a moderate but clinically important reduction in the risk of fetal or neonatal death (point estimate 29%, confidence intervals consistent with everything from a 51% reduction in risk to a 2% increase) is possible. Although many studies did not define stillbirths and miscarriages, these data suggest that antihypertensive treatment for mild to moderate hypertension may have a greater potential for reducing early pregnancy loss than later stillbirths or neonatal deaths.

Figure 3.

Funnel plot of comparison: 1 Any antihypertensive drug versus none (subgrouped by class of drug), outcome: 1.4 Proteinuria/pre-eclampsia.

Figure 4.

Funnel plot of comparison: 2 Any antihypertensive drug versus none (subgrouped by type of hypertensive disorder at trial entry), outcome: 2.2 Proteinuria/pre-eclampsia.

For small-for-gestational-age babies, the confidence intervals include everything from a 20% reduction in the risk of a small-for-gestational-age baby to a 17% increase, and the point estimate is for a 3% reduction in risk. It has been argued that lowering maternal blood pressure may cause fetal growth restriction (von Dadelszen 2000). This hypothesis is based on meta-regression, however. So, although the included studies were randomised trials, the analysis is prone to all the biases of observational studies. Also, combining data from all trials there is no overall effect on the relative risk of having a small-for-gestational-age baby (RR 0.97; 95% CI 0.80 to 1.17) for women allocated antihypertensive drugs rather than placebo. Yet amongst the trials of beta blockers an increase appears likely (nine trials, 904 women; RR 1.38; 95% CI 0.99 to 1.92). It therefore remains plausible that the observed association with fetal growth restriction is related to beta blockers, rather than any general effect of reducing blood pressure.

Being small-for-gestational age is an important marker for neurodevelopmental delay. The ideal timing of delivery for such babies is unclear, regardless of whether the woman has raised blood pressure or not (Thornton 2004).

Few children exposed to antihypertensive drugs in utero have been followed up beyond the perinatal period. Two trials (Italy 1998; UK 1983) have reported follow-up of a total of 110 children at age one year, and of 190 children at age 18 months, respectively. Another (UK 1976) reported follow-up at 7½ years of age for children randomised before 28 weeks' gestation. All studies were too small to provide reliable estimates of the benefits and adverse effects for surviving children.

The question of which antihypertensive drug to use is less relevant until it becomes clearer whether attempting to control mild to moderate hypertension during pregnancy is worthwhile. However, beta blockers seem to be better tolerated by women than methyldopa (RR 0.69; 95% CI 0.52 to 0.91), although there is potential for reporting or detection bias as this outcome was only reported by half the trials, all of them open label studies. Beta blockers also seem to be more effective than methyldopa in avoiding an episode of severe hypertension. However, concerns remain about their possible role in the risk of having small-for-gestational-age babies (see above).

Overall completeness and applicability of evidence

We believe that this is a comprehensive review that covers the most up to date evidence from randomised controlled trials conducted around the world. We have not applied language restrictions for selecting studies.

Surprisingly, only four eligible trials (recruiting 511 women) were published since this review was first available in 2001. Twenty-five (out of 49) trials included in this review were published before 1990, more than 20 years ago. Evidence from these trials account for 48% of the total number of women recruited. Maternal and perinatal survival, availability of resources, technologies for diagnosis and treatment of complications, as well as indications for termination of pregnancy for maternal or perinatal conditions could have changed substantially since then. On the other hand, only a quarter of women considered in this review were recruited in low- and middle-income countries, where keeping the woman in a stable condition aiming at prolonging pregnancy may have major impact on neonatal survival.

These considerations, amongst others, may lead to a different response to the uncertainties presented in this review in different settings.

Quality of the evidence

Overall, the quality of the studies included in this review is moderate to poor. Concealment of allocation was adequate for only 17 of the 49 trials (35%). Most trials with unclear concealment of allocation were described as 'randomised' with no further details. Some of these studies were stated to be double blind, but with no information about how this was achieved. Only 12 of the studies evaluating a single agent used a placebo for the control group. All trials comparing alternative antihypertensive drugs were open label trials. None of the trials comparing a single drug against no treatment, or those comparing one agent with another, mentioned blinding in the assessment of outcome. The primary outcome reported by trials usually was mean blood pressure, or a fall in blood pressure, thus sample size to test the hypothesis was in general small for other clinically important outcomes, such as maternal or perinatal morbidity. The largest trial in this review recruited only 300 women, and only five studies had comparison arms of more than 100 women. Five trials (out of 49) reported sample size and power calculations.

Potential biases in the review process

A large number of outcomes are reported in these trials, and for many, data are only available from a small number of studies. There is therefore considerable potential to be misled by reporting bias. For example, in the comparison of any antihypertensive with none, only five of the 29 trials reported respiratory distress syndrome, and all had results favouring antihypertensive treatment. Without further information, it is impossible to know whether the other 24 trials did not collect these data, or whether they did not report it because it did not favour antihypertensive therapy.

We also report data from a large number of subgroups. Although these subgroups were all prespecified, the numbers in many cells are small, and for one in 20 the difference will be statistically significant purely by chance. Results from these subgroups should therefore be interpreted with caution, as there is considerable potential to be misled by random errors.

Agreements and disagreements with other studies or reviews

The review of Magee and Duley on beta blockers for non-severe hypertension also concludes that the effect of beta blockers on perinatal outcome is uncertain and large randomised trials are needed to determine whether antihypertensive therapy in general (rather than beta blocker therapy specifically) results in greater benefit than risk, for treatment of non-severe pregnancy hypertension (Magee 2003).

In the National Institute for Health and Clinical Excellence (NICE) guideline CG107Hypertension in pregnancy: The management of hypertensive disorders during pregnancy, the authors do not recommend antihypertensive drugs in mild pregnancy-induced hypertension or pre-eclampsia. Only when blood pressure is greater than 150/100 mmHg, labetalol, methyldopa or nifedipine are advised as the drugs of choice (NICE 2011).

Authors' conclusions

Implications for practice

It remains unclear whether antihypertensive drug therapy for mild to moderate hypertension during pregnancy is worthwhile. Whether the reduction in the risk of severe hypertension is considered sufficient to warrant treatment is a decision that should be made by women in consultation with their obstetrician. If an antihypertensive is used, there is insufficient evidence to conclude that one antihypertensive is better than another. The choice should therefore depend on the previous experience of the clinician and the woman's preference.

Implications for research

Large simple trials are required in order to provide reliable estimates of the benefits and adverse effects of antihypertensive treatment for mild to moderate hypertension. We need to know the effects for both mother and baby, as well as the costs to the health services, to women and to their families. Outcomes relevant to women should be included in such studies, such as admission to hospital, clinic visits, and disruption to their family and working life. Trials should also assess the level of blood pressure at which antihypertensive treatment becomes worthwhile. Long-term follow-up of children entered into such trials as fetuses is needed in order to assess whether there are any effects on infant and child development.

[Note: The one citation in the awaiting classification section of the review may alter the conclusions of the review once assessed (Canada 2008).]

Acknowledgements

Thanks to Rory Collins and David Henderson Smart for their contribution to earlier versions of this review.

Thanks to Marta Roqué i Figuls from the Iberoamerican Cochrane Centre, Barcelona, (Spain), and to Zulma Ortiz from the Centro de Investigaciones Epidemiologicas, Academia Nacional de Medicina, Buenos Aires, (Argentina) for their help in locating original papers in Spanish and Portuguese.

Special thanks to Greg Davis (Australia 2001), Salvio Freire (Brazil 1988), Soubhi Kahhale and Silvana Darcie (Brazil 2000), Denis J Nascimento (Brazil 2000a), Pierre-Francois Plouin (Caribbean Is.1990), Elba Gomez Sosa (Cuba 1994), Fakhrolmolouk Yassaee (Iran 2000), Sean Blake and Dermot MacDonald (Ireland 1991), Isabella Neri and Fabio Facchinetti (Italy 1999), Claudio Borghi (Italy 2000), Klas Wichman and Orvar Finnström (Sweden 1984), Stellan Högstedt (Sweden 1985), James Walker (UK 1982), Peter C Rubin (UK 1990), DJ Cruickshank (UK 1992), Baha Sibai (USA 1990), and Silvia Sanchez (Venezuela 1985), who kindly provided unpublished data for this review.

The National Institute for Health Research (NIHR) is the largest single funder of the Cochrane Pregnancy and Childbirth Group. The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the NIHR, NHS or the Department of Health.

Data and analyses

Download statistical data

Comparison 1. Any antihypertensive drug versus none (subgrouped by class of drug)
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Maternal death5525Risk Ratio (M-H, Fixed, 95% CI)1.08 [0.24, 4.83]
2 Eclampsia5578Risk Ratio (M-H, Fixed, 95% CI)0.34 [0.01, 8.15]
3 Severe hypertension202558Risk Ratio (M-H, Fixed, 95% CI)0.49 [0.40, 0.60]
3.1 Beta blocker versus none8762Risk Ratio (M-H, Fixed, 95% CI)0.38 [0.26, 0.57]
3.2 Beta blocker + other drug versus none2322Risk Ratio (M-H, Fixed, 95% CI)0.30 [0.12, 0.77]
3.3 Methyldopa versus none2310Risk Ratio (M-H, Fixed, 95% CI)0.32 [0.17, 0.58]
3.4 Methyldopa + other drug versus none158Risk Ratio (M-H, Fixed, 95% CI)0.67 [0.12, 3.70]
3.5 Beta blocker or methyldopa versus none2412Risk Ratio (M-H, Fixed, 95% CI)0.43 [0.25, 0.74]
3.6 Calcium channel blocker versus none4662Risk Ratio (M-H, Fixed, 95% CI)0.81 [0.60, 1.11]
3.7 Alpha blocker versus none132Risk Ratio (M-H, Fixed, 95% CI)0.07 [0.00, 1.09]
4 Proteinuria/pre-eclampsia232851Risk Ratio (M-H, Fixed, 95% CI)0.93 [0.80, 1.08]
4.1 Beta blocker versus none8883Risk Ratio (M-H, Fixed, 95% CI)0.73 [0.57, 0.94]
4.2 Beta blocker + other drug versus none2322Risk Ratio (M-H, Fixed, 95% CI)1.26 [0.63, 2.51]
4.3 Methyldopa versus none2267Risk Ratio (M-H, Fixed, 95% CI)1.22 [0.55, 2.71]
4.4 Methyldopa + other drug versus none2158Risk Ratio (M-H, Fixed, 95% CI)0.74 [0.43, 1.30]
4.5 Beta blocker or methyldopa versus none2412Risk Ratio (M-H, Fixed, 95% CI)0.77 [0.50, 1.19]
4.6 Calcium channel blocker versus none4725Risk Ratio (M-H, Fixed, 95% CI)1.40 [1.06, 1.86]
4.7 Alpha blocker versus none132Risk Ratio (M-H, Fixed, 95% CI)0.33 [0.04, 2.52]
4.8 Glyceryl trinitrate versus none116Risk Ratio (M-H, Fixed, 95% CI)0.43 [0.06, 3.28]
4.9 Regular antihypertensive therapy versus none136Risk Ratio (M-H, Fixed, 95% CI)0.19 [0.02, 1.39]
5 Severe pre-eclampsia3416Risk Ratio (M-H, Fixed, 95% CI)0.54 [0.24, 1.23]
6 HELLP syndrome1197Risk Ratio (M-H, Fixed, 95% CI)2.02 [0.38, 10.78]
6.1 Calcium channel blocker versus none1197Risk Ratio (M-H, Fixed, 95% CI)2.02 [0.38, 10.78]
7 Pulmonary oedema2325Risk Ratio (M-H, Fixed, 95% CI)1.24 [0.30, 5.12]
7.1 Beta blocker versus none1176Risk Ratio (M-H, Fixed, 95% CI)5.23 [0.25, 107.39]
7.2 Beta blocker or methyldopa versus none1149Risk Ratio (M-H, Fixed, 95% CI)0.51 [0.07, 3.48]
8 Additional antihypertensive101285Risk Ratio (M-H, Fixed, 95% CI)0.42 [0.30, 0.58]
8.1 Beta blocker versus none3245Risk Ratio (M-H, Fixed, 95% CI)0.40 [0.23, 0.70]
8.2 Beta blocker + other drug versus none2315Risk Ratio (M-H, Fixed, 95% CI)0.32 [0.13, 0.82]
8.3 Methyldopa + other drug versus none158Risk Ratio (M-H, Fixed, 95% CI)0.31 [0.11, 0.83]
8.4 Beta blocker or methyldopa versus none1263Risk Ratio (M-H, Fixed, 95% CI)0.52 [0.22, 1.20]
8.5 Calcium channel blocker versus none2372Risk Ratio (M-H, Fixed, 95% CI)0.67 [0.35, 1.28]
8.6 Alpha blocker versus none132Risk Ratio (M-H, Fixed, 95% CI)0.07 [0.00, 1.09]
9 Changed/stopped drugs due to maternal side-effects151403Risk Ratio (M-H, Fixed, 95% CI)2.59 [1.33, 5.04]
9.1 Beta blocker versus none9745Risk Ratio (M-H, Fixed, 95% CI)1.90 [0.76, 4.75]
9.2 Beta blocker + other drug versus none2315Risk Ratio (M-H, Fixed, 95% CI)1.30 [0.30, 5.61]
9.3 Methyldopa versus none125Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
9.4 Calcium channel blocker versus none2302Risk Ratio (M-H, Fixed, 95% CI)4.10 [0.46, 36.21]
9.5 Glyceryl trinitrate versus none116Risk Ratio (M-H, Fixed, 95% CI)18.75 [1.25, 281.11]
10 Maternal side-effects11934Risk Ratio (M-H, Fixed, 95% CI)1.53 [1.10, 2.12]
10.1 Beta blocker versus none7554Risk Ratio (M-H, Fixed, 95% CI)2.07 [1.21, 3.54]
10.2 Beta blocker + other drug versus none1154Risk Ratio (M-H, Fixed, 95% CI)1.30 [0.30, 5.61]
10.3 Methyldopa versus none125Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
10.4 Calcium channel blocker versus none1185Risk Ratio (M-H, Fixed, 95% CI)0.96 [0.60, 1.52]
10.5 Glyceryl trinitrate versus none116Risk Ratio (M-H, Fixed, 95% CI)18.75 [1.25, 281.11]
11 Antenatal hospital admission4455Risk Ratio (M-H, Fixed, 95% CI)0.87 [0.73, 1.03]
11.1 Beta blocker versus none152Risk Ratio (M-H, Fixed, 95% CI)0.84 [0.57, 1.24]
11.2 Beta blocker + other drug versus none2254Risk Ratio (M-H, Fixed, 95% CI)0.96 [0.78, 1.17]
11.3 Beta blocker or methyldopa versus none1149Risk Ratio (M-H, Fixed, 95% CI)0.51 [0.26, 0.98]
12 Induction of labour5563Risk Ratio (M-H, Fixed, 95% CI)0.91 [0.77, 1.07]
12.1 Beta blocker versus none3384Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.83, 1.16]
12.2 Beta blocker + other drug versus none1154Risk Ratio (M-H, Fixed, 95% CI)0.49 [0.26, 0.90]
12.3 Methyldopa versus none125Risk Ratio (M-H, Fixed, 95% CI)1.54 [0.46, 5.09]
13 Elective delivery (induction of labour + elective caesarean section)4710Risk Ratio (M-H, Fixed, 95% CI)0.91 [0.83, 1.00]
13.1 Beta blocker versus none2240Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.84, 1.13]
13.2 Beta blocker + other drug versus none3470Risk Ratio (M-H, Fixed, 95% CI)0.87 [0.77, 0.99]
14 Caesarean section202624Risk Ratio (M-H, Fixed, 95% CI)0.92 [0.83, 1.02]
14.1 Beta blocker versus none8850Risk Ratio (M-H, Fixed, 95% CI)1.05 [0.86, 1.30]
14.2 Beta blocker + other drug versus none2322Risk Ratio (M-H, Fixed, 95% CI)0.57 [0.38, 0.84]
14.3 Methyldopa versus none2272Risk Ratio (M-H, Fixed, 95% CI)0.86 [0.57, 1.30]
14.4 Methyldopa + other drug versus none158Risk Ratio (M-H, Fixed, 95% CI)1.8 [0.69, 4.72]
14.5 Beta blocker or methyldopa versus none2412Risk Ratio (M-H, Fixed, 95% CI)0.89 [0.67, 1.19]
14.6 Calcium channel blocker versus none3642Risk Ratio (M-H, Fixed, 95% CI)0.95 [0.83, 1.09]
14.7 Alpha blocker versus none132Risk Ratio (M-H, Fixed, 95% CI)0.63 [0.20, 1.91]
14.8 Regular antihypertensive therapy versus none136Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
15 Placental abruption111433Risk Ratio (M-H, Fixed, 95% CI)1.41 [0.65, 3.07]
15.1 Beta blocker versus none3364Risk Ratio (M-H, Fixed, 95% CI)5.11 [0.25, 104.96]
15.2 Beta blocker + other drug versus none2322Risk Ratio (M-H, Fixed, 95% CI)2.24 [0.34, 14.98]
15.3 Methyldopa versus none170Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
15.4 Beta blocker or methyldopa versus none2412Risk Ratio (M-H, Fixed, 95% CI)0.52 [0.13, 2.03]
15.5 Calcium blocker versus none1197Risk Ratio (M-H, Fixed, 95% CI)1.52 [0.26, 8.87]
15.6 Alpha blocker versus none132Risk Ratio (M-H, Fixed, 95% CI)3.33 [0.34, 32.96]
15.7 Regular antihypertensive therapy versus none136Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
16 Total reported fetal or neonatal death (including miscarriage)273230Risk Ratio (M-H, Fixed, 95% CI)0.71 [0.49, 1.02]
16.1 Beta blocker versus none111045Risk Ratio (M-H, Fixed, 95% CI)1.02 [0.46, 2.29]
16.2 Beta blocker + other drug versus none2322Risk Ratio (M-H, Fixed, 95% CI)1.21 [0.33, 4.43]
16.3 Methyldopa versus none3337Risk Ratio (M-H, Fixed, 95% CI)0.35 [0.13, 0.94]
16.4 Methyldopa + other drug versus none2158Risk Ratio (M-H, Fixed, 95% CI)0.58 [0.23, 1.44]
16.5 Beta blocker or methyldopa versus none2443Risk Ratio (M-H, Fixed, 95% CI)0.58 [0.22, 1.55]
16.6 Calcium channel blocker versus none5857Risk Ratio (M-H, Fixed, 95% CI)0.75 [0.28, 2.05]
16.7 Alpha blocker versus none132Risk Ratio (M-H, Fixed, 95% CI)0.83 [0.25, 2.73]
16.8 Regular antihypertensive therapy versus none136Risk Ratio (M-H, Fixed, 95% CI)2.24 [0.22, 22.51]
17 Fetal or neonatal death (subgrouped by time of death)25 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
17.1 Miscarriage71058Risk Ratio (M-H, Fixed, 95% CI)0.39 [0.17, 0.93]
17.2 Stillbirth182480Risk Ratio (M-H, Fixed, 95% CI)1.14 [0.60, 2.17]
17.3 Perinatal death202382Risk Ratio (M-H, Fixed, 95% CI)0.96 [0.60, 1.54]
17.4 Neonatal death4557Risk Ratio (M-H, Fixed, 95% CI)0.79 [0.14, 4.34]
18 Preterm birth (< 37 weeks)152141Risk Ratio (M-H, Fixed, 95% CI)0.96 [0.85, 1.10]
18.1 Beta blocker versus none4361Risk Ratio (M-H, Fixed, 95% CI)0.90 [0.62, 1.32]
18.2 Beta blocker + other drug versus none2322Risk Ratio (M-H, Fixed, 95% CI)0.96 [0.63, 1.46]
18.3 Methyldopa versus none2272Risk Ratio (M-H, Fixed, 95% CI)1.61 [0.73, 3.58]
18.4 Beta blocker or methyldopa versus none2412Risk Ratio (M-H, Fixed, 95% CI)0.66 [0.42, 1.05]
18.5 Calcium channel blocker versus none4742Risk Ratio (M-H, Fixed, 95% CI)1.03 [0.88, 1.21]
18.6 Alpha blocker versus none132Risk Ratio (M-H, Fixed, 95% CI)0.67 [0.27, 1.66]
19 Preterm birth (subgrouped by gestational age)15 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
19.1 < 37 weeks101569Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.85, 1.13]
19.2 < 36 weeks2304Risk Ratio (M-H, Fixed, 95% CI)0.77 [0.37, 1.59]
19.3 < 34 weeks5792Risk Ratio (M-H, Fixed, 95% CI)1.19 [0.77, 1.83]
19.4 Unspecified4423Risk Ratio (M-H, Fixed, 95% CI)1.26 [0.87, 1.82]
20 Small-for-gestational age202586Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.80, 1.17]
20.1 Beta blocker versus none9904Risk Ratio (M-H, Fixed, 95% CI)1.38 [0.99, 1.92]
20.2 Beta blocker + other drug versus none2322Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.46, 2.02]
20.3 Methyldopa versus none2227Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.26, 3.70]
20.4 Methyl dopa + other drug versus none158Risk Ratio (M-H, Fixed, 95% CI)1.0 [0.28, 3.62]
20.5 Beta blocker or methyldopa versus none2412Risk Ratio (M-H, Fixed, 95% CI)0.66 [0.43, 1.01]
20.6 Calcium channel blocker versus none3640Risk Ratio (M-H, Fixed, 95% CI)0.84 [0.60, 1.16]
20.7 Alpha blocker versus none123Risk Ratio (M-H, Fixed, 95% CI)0.43 [0.05, 3.57]
21 Small-for-gestational age (subgrouped by severity)19 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
21.1 Birthweight < 10th centile101265Risk Ratio (M-H, Fixed, 95% CI)0.99 [0.79, 1.24]
21.2 Birthweight < 5th centile3287Risk Ratio (M-H, Fixed, 95% CI)3.04 [1.25, 7.40]
21.3 Unspecified71090Risk Ratio (M-H, Fixed, 95% CI)0.75 [0.51, 1.10]
22 Admission to special care baby unit91470Risk Ratio (M-H, Random, 95% CI)1.04 [0.83, 1.30]
22.1 Beta blocker versus none3449Risk Ratio (M-H, Random, 95% CI)1.07 [0.82, 1.41]
22.2 Beta blocker + other drug versus none1151Risk Ratio (M-H, Random, 95% CI)0.66 [0.38, 1.14]
22.3 Methyldopa versus none2272Risk Ratio (M-H, Random, 95% CI)1.56 [0.88, 2.78]
22.4 Beta blocker or methyldopa versus none1149Risk Ratio (M-H, Random, 95% CI)0.51 [0.27, 0.95]
22.5 Calcium channel blocker versus none2449Risk Ratio (M-H, Random, 95% CI)1.18 [0.87, 1.62]
23 Respiratory distress syndrome5825Risk Ratio (M-H, Fixed, 95% CI)0.28 [0.12, 0.63]
23.1 Beta blocker versus none3412Risk Ratio (M-H, Fixed, 95% CI)0.28 [0.11, 0.71]
23.2 Beta blocker + other drug versus none1157Risk Ratio (M-H, Fixed, 95% CI)0.33 [0.03, 3.10]
23.3 Calcium channel blocker versus none1256Risk Ratio (M-H, Fixed, 95% CI)0.20 [0.01, 4.06]
24 Neonatal hypoglycaemia5862Risk Ratio (M-H, Fixed, 95% CI)0.77 [0.51, 1.17]
24.1 Beta blocker versus none2261Risk Ratio (M-H, Fixed, 95% CI)0.73 [0.24, 2.24]
24.2 Beta blocker + other drug versus none1157Risk Ratio (M-H, Fixed, 95% CI)0.81 [0.35, 1.84]
24.3 Beta blocker or methyldopa versus none1261Risk Ratio (M-H, Fixed, 95% CI)2.07 [0.23, 18.24]
24.4 Calcium channel blocker versus none1183Risk Ratio (M-H, Fixed, 95% CI)0.69 [0.39, 1.21]
25 Neonatal bradycardia3418Risk Ratio (M-H, Fixed, 95% CI)1.93 [1.05, 3.53]
25.1 Beta blocker versus none2261Risk Ratio (M-H, Fixed, 95% CI)2.61 [1.32, 5.15]
25.2 Beta blocker + other drug versus none1157Risk Ratio (M-H, Fixed, 95% CI)0.25 [0.03, 2.16]
26 Neonatal jaundice3529Risk Ratio (M-H, Fixed, 95% CI)0.78 [0.56, 1.09]
26.1 Beta blocker versus none1144Risk Ratio (M-H, Fixed, 95% CI)0.53 [0.19, 1.47]
26.2 Methyldopa versus none1202Risk Ratio (M-H, Fixed, 95% CI)1.02 [0.65, 1.61]
26.3 Calcium channel blocker versus none1183Risk Ratio (M-H, Fixed, 95% CI)0.62 [0.35, 1.10]
27 Follow-up of the children at 1 year: cerebral palsy1110Risk Ratio (M-H, Fixed, 95% CI)0.33 [0.01, 8.01]
28 Follow-up of the children at 7 1/2 years1 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
28.1 Chronic ill health1190Risk Ratio (M-H, Fixed, 95% CI)0.70 [0.16, 3.06]
28.2 Impaired hearing1190Risk Ratio (M-H, Fixed, 95% CI)1.10 [0.38, 3.14]
28.3 Impaired vision1190Risk Ratio (M-H, Fixed, 95% CI)0.47 [0.20, 1.11]
Analysis 1.1.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 1 Maternal death.

Analysis 1.2.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 2 Eclampsia.

Analysis 1.3.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 3 Severe hypertension.

Analysis 1.4.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 4 Proteinuria/pre-eclampsia.

Analysis 1.5.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 5 Severe pre-eclampsia.

Analysis 1.6.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 6 HELLP syndrome.

Analysis 1.7.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 7 Pulmonary oedema.

Analysis 1.8.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 8 Additional antihypertensive.

Analysis 1.9.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 9 Changed/stopped drugs due to maternal side-effects.

Analysis 1.10.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 10 Maternal side-effects.

Analysis 1.11.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 11 Antenatal hospital admission.

Analysis 1.12.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 12 Induction of labour.

Analysis 1.13.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 13 Elective delivery (induction of labour + elective caesarean section).

Analysis 1.14.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 14 Caesarean section.

Analysis 1.15.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 15 Placental abruption.

Analysis 1.16.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 16 Total reported fetal or neonatal death (including miscarriage).

Analysis 1.17.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 17 Fetal or neonatal death (subgrouped by time of death).

Analysis 1.18.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 18 Preterm birth (< 37 weeks).

Analysis 1.19.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 19 Preterm birth (subgrouped by gestational age).

Analysis 1.20.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 20 Small-for-gestational age.

Analysis 1.21.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 21 Small-for-gestational age (subgrouped by severity).

Analysis 1.22.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 22 Admission to special care baby unit.

Analysis 1.23.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 23 Respiratory distress syndrome.

Analysis 1.24.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 24 Neonatal hypoglycaemia.

Analysis 1.25.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 25 Neonatal bradycardia.

Analysis 1.26.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 26 Neonatal jaundice.

Analysis 1.27.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 27 Follow-up of the children at 1 year: cerebral palsy.

Analysis 1.28.

Comparison 1 Any antihypertensive drug versus none (subgrouped by class of drug), Outcome 28 Follow-up of the children at 7 1/2 years.

Comparison 2. Any antihypertensive drug versus none (subgrouped by type of hypertensive disorder at trial entry)
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Severe hypertension202558Risk Ratio (M-H, Fixed, 95% CI)0.49 [0.40, 0.60]
1.1 Hypertension alone4493Risk Ratio (M-H, Fixed, 95% CI)0.27 [0.15, 0.47]
1.2 Hypertension + proteinuria2256Risk Ratio (M-H, Fixed, 95% CI)0.26 [0.13, 0.54]
1.3 Chronic hypertension4538Risk Ratio (M-H, Fixed, 95% CI)0.57 [0.34, 0.98]
1.4 Unclassified/mixed101271Risk Ratio (M-H, Fixed, 95% CI)0.60 [0.47, 0.77]
2 Proteinuria/pre-eclampsia232851Risk Ratio (M-H, Fixed, 95% CI)0.93 [0.80, 1.08]
2.1 Hypertension alone8684Risk Ratio (M-H, Fixed, 95% CI)0.72 [0.57, 0.92]
2.2 Hypertension + proteinuria2383Risk Ratio (M-H, Fixed, 95% CI)1.65 [0.92, 2.97]
2.3 Chronic hypertension4605Risk Ratio (M-H, Fixed, 95% CI)0.92 [0.66, 1.28]
2.4 Unclassified/mixed91179Risk Ratio (M-H, Fixed, 95% CI)1.04 [0.80, 1.36]
3 Total reported fetal or neonatal death (including miscarriage)273230Risk Ratio (M-H, Fixed, 95% CI)0.71 [0.49, 1.02]
3.1 Hypertension alone7668Risk Ratio (M-H, Fixed, 95% CI)0.83 [0.42, 1.65]
3.2 Hypertension + proteinuria3475Risk Ratio (M-H, Fixed, 95% CI)0.89 [0.30, 2.59]
3.3 Chronic hypertension5665Risk Ratio (M-H, Fixed, 95% CI)0.94 [0.40, 2.21]
3.4 Unclassified/mixed121422Risk Ratio (M-H, Fixed, 95% CI)0.55 [0.31, 0.96]
4 Preterm birth (< 37 weeks)152141Risk Ratio (M-H, Fixed, 95% CI)0.96 [0.85, 1.10]
4.1 Hypertension alone5607Risk Ratio (M-H, Fixed, 95% CI)0.75 [0.56, 1.00]
4.2 Hypertension + proteinuria2267Risk Ratio (M-H, Fixed, 95% CI)1.24 [0.92, 1.67]
4.3 Chronic hypertension2447Risk Ratio (M-H, Fixed, 95% CI)1.15 [0.72, 1.86]
4.4 Unclassified/mixed6820Risk Ratio (M-H, Fixed, 95% CI)0.96 [0.80, 1.15]
5 Small-for-gestational age202586Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.80, 1.17]
5.1 Hypertension alone6623Risk Ratio (M-H, Fixed, 95% CI)0.84 [0.57, 1.24]
5.2 Hypertension + proteinuria2391Risk Ratio (M-H, Fixed, 95% CI)1.54 [0.93, 2.54]
5.3 Chronic hypertension5628Risk Ratio (M-H, Fixed, 95% CI)0.80 [0.53, 1.23]
5.4 Unclassified/mixed7944Risk Ratio (M-H, Fixed, 95% CI)0.96 [0.71, 1.30]
Analysis 2.1.

Comparison 2 Any antihypertensive drug versus none (subgrouped by type of hypertensive disorder at trial entry), Outcome 1 Severe hypertension.

Analysis 2.2.

Comparison 2 Any antihypertensive drug versus none (subgrouped by type of hypertensive disorder at trial entry), Outcome 2 Proteinuria/pre-eclampsia.

Analysis 2.3.

Comparison 2 Any antihypertensive drug versus none (subgrouped by type of hypertensive disorder at trial entry), Outcome 3 Total reported fetal or neonatal death (including miscarriage).

Analysis 2.4.

Comparison 2 Any antihypertensive drug versus none (subgrouped by type of hypertensive disorder at trial entry), Outcome 4 Preterm birth (< 37 weeks).

Analysis 2.5.

Comparison 2 Any antihypertensive drug versus none (subgrouped by type of hypertensive disorder at trial entry), Outcome 5 Small-for-gestational age.

Comparison 3. Any antihypertensive drug versus none (subgrouped by gestation at trial entry)
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Severe hypertension202558Risk Ratio (M-H, Fixed, 95% CI)0.49 [0.40, 0.60]
1.1 Entry < 32 weeks71071Risk Ratio (M-H, Fixed, 95% CI)0.63 [0.47, 0.83]
1.2 Entry > 32 weeks1120Risk Ratio (M-H, Fixed, 95% CI)0.29 [0.06, 1.32]
1.3 Unclassified/mixed121367Risk Ratio (M-H, Fixed, 95% CI)0.40 [0.30, 0.53]
2 Proteinuria/pre-eclampsia232851Risk Ratio (M-H, Fixed, 95% CI)0.93 [0.80, 1.08]
2.1 Entry < 32 weeks81147Risk Ratio (M-H, Fixed, 95% CI)1.05 [0.81, 1.36]
2.2 Entry > 32 weeks2120Risk Ratio (M-H, Fixed, 95% CI)0.34 [0.12, 0.96]
2.3 Unclassified/mixed131584Risk Ratio (M-H, Fixed, 95% CI)0.91 [0.75, 1.10]
3 Total reported fetal or neonatal death (including miscarriage)273230Risk Ratio (M-H, Fixed, 95% CI)0.71 [0.49, 1.02]
3.1 Entry < 32 weeks101276Risk Ratio (M-H, Fixed, 95% CI)0.66 [0.39, 1.14]
3.2 Entry > 32 weeks1120Risk Ratio (M-H, Fixed, 95% CI)0.5 [0.05, 5.37]
3.3 Unclassified/mixed161834Risk Ratio (M-H, Fixed, 95% CI)0.77 [0.47, 1.27]
4 Preterm birth (< 37 weeks)152141Risk Ratio (M-H, Fixed, 95% CI)0.96 [0.85, 1.10]
4.1 Entry < 32 weeks6993Risk Ratio (M-H, Fixed, 95% CI)0.99 [0.82, 1.20]
4.2 Entry > 32 weeks00Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
4.3 Unclassified/mixed91148Risk Ratio (M-H, Fixed, 95% CI)0.95 [0.79, 1.13]
5 Small-for-gestational age202586Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.80, 1.17]
5.1 Entry < 32 weeks101185Risk Ratio (M-H, Fixed, 95% CI)0.84 [0.63, 1.11]
5.2 Entry > 32 weeks1117Risk Ratio (M-H, Fixed, 95% CI)1.11 [0.46, 2.67]
5.3 Unclassified/mixed91284Risk Ratio (M-H, Fixed, 95% CI)1.09 [0.83, 1.43]
Analysis 3.1.

Comparison 3 Any antihypertensive drug versus none (subgrouped by gestation at trial entry), Outcome 1 Severe hypertension.

Analysis 3.2.

Comparison 3 Any antihypertensive drug versus none (subgrouped by gestation at trial entry), Outcome 2 Proteinuria/pre-eclampsia.

Analysis 3.3.

Comparison 3 Any antihypertensive drug versus none (subgrouped by gestation at trial entry), Outcome 3 Total reported fetal or neonatal death (including miscarriage).

Analysis 3.4.

Comparison 3 Any antihypertensive drug versus none (subgrouped by gestation at trial entry), Outcome 4 Preterm birth (< 37 weeks).

Analysis 3.5.

Comparison 3 Any antihypertensive drug versus none (subgrouped by gestation at trial entry), Outcome 5 Small-for-gestational age.

Comparison 4. Any antihypertensive drug versus none (subgrouped by use of placebo)
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Severe hypertension202558Risk Ratio (M-H, Fixed, 95% CI)0.49 [0.40, 0.60]
1.1 Placebo10937Risk Ratio (M-H, Fixed, 95% CI)0.50 [0.36, 0.69]
1.2 No placebo101621Risk Ratio (M-H, Fixed, 95% CI)0.48 [0.38, 0.62]
2 Proteinuria/pre-eclampsia232851Risk Ratio (M-H, Fixed, 95% CI)0.93 [0.80, 1.08]
2.1 Placebo10869Risk Ratio (M-H, Fixed, 95% CI)0.88 [0.71, 1.09]
2.2 No placebo131982Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.78, 1.20]
3 Total reported fetal or neonatal death (including miscarriage)273230Risk Ratio (M-H, Fixed, 95% CI)0.71 [0.49, 1.02]
3.1 Placebo10911Risk Ratio (M-H, Fixed, 95% CI)0.79 [0.37, 1.69]
3.2 No placebo172319Risk Ratio (M-H, Fixed, 95% CI)0.69 [0.46, 1.04]
4 Preterm birth (< 37 weeks)152141Risk Ratio (M-H, Fixed, 95% CI)0.96 [0.85, 1.10]
4.1 Placebo5566Risk Ratio (M-H, Fixed, 95% CI)0.85 [0.62, 1.17]
4.2 No placebo101575Risk Ratio (M-H, Fixed, 95% CI)1.00 [0.86, 1.15]
5 Small-for-gestational age202586Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.80, 1.17]
5.1 Placebo8714Risk Ratio (M-H, Fixed, 95% CI)0.94 [0.64, 1.38]
5.2 No placebo121872Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.78, 1.22]
Analysis 4.1.

Comparison 4 Any antihypertensive drug versus none (subgrouped by use of placebo), Outcome 1 Severe hypertension.

Analysis 4.2.

Comparison 4 Any antihypertensive drug versus none (subgrouped by use of placebo), Outcome 2 Proteinuria/pre-eclampsia.

Analysis 4.3.

Comparison 4 Any antihypertensive drug versus none (subgrouped by use of placebo), Outcome 3 Total reported fetal or neonatal death (including miscarriage).

Analysis 4.4.

Comparison 4 Any antihypertensive drug versus none (subgrouped by use of placebo), Outcome 4 Preterm birth (< 37 weeks).

Analysis 4.5.

Comparison 4 Any antihypertensive drug versus none (subgrouped by use of placebo), Outcome 5 Small-for-gestational age.

Comparison 5. Any antihypertensive versus methyldopa (subgrouped by class of drug)
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Severe hypertension11638Risk Ratio (M-H, Random, 95% CI)0.54 [0.30, 0.95]
1.1 Beta blocker versus methyldopa9592Risk Ratio (M-H, Random, 95% CI)0.59 [0.33, 1.05]
1.2 Calcium channel blockers versus methyldopa246Risk Ratio (M-H, Random, 95% CI)0.23 [0.04, 1.22]
2 Proteinuria/pre-eclampsia11997Risk Ratio (M-H, Fixed, 95% CI)0.73 [0.54, 0.99]
2.1 Beta blocker versus methyldopa10903Risk Ratio (M-H, Fixed, 95% CI)0.75 [0.53, 1.05]
2.2 Calcium channel blockers versus methyldopa194Risk Ratio (M-H, Fixed, 95% CI)0.66 [0.34, 1.27]
3 Additional antihypertensive12989Risk Ratio (M-H, Fixed, 95% CI)0.84 [0.66, 1.06]
3.1 Beta blocker versus methyldopa10853Risk Ratio (M-H, Fixed, 95% CI)0.88 [0.69, 1.13]
3.2 Calcium channel blocker versus methyldopa2136Risk Ratio (M-H, Fixed, 95% CI)0.59 [0.29, 1.22]
4 Antenatal hospital admission2275Risk Ratio (M-H, Fixed, 95% CI)0.77 [0.58, 1.03]
4.1 Beta blocker versus methyldopa2275Risk Ratio (M-H, Fixed, 95% CI)0.77 [0.58, 1.03]
5 Elective delivery (induction of labour + elective caesarean section)5443Risk Ratio (M-H, Fixed, 95% CI)1.01 [0.86, 1.18]
5.1 Beta blocker versus methyldopa4333Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.84, 1.15]
5.2 Calcium channel blocker versus methyldopa1110Risk Ratio (M-H, Fixed, 95% CI)1.56 [0.59, 4.07]
6 Caesarean section10878Risk Ratio (M-H, Fixed, 95% CI)0.93 [0.78, 1.12]
6.1 Beta blocker veresus methyldopa9852Risk Ratio (M-H, Fixed, 95% CI)0.94 [0.78, 1.13]
6.2 Calcium channel blocker versus methyldopa126Risk Ratio (M-H, Fixed, 95% CI)0.67 [0.13, 3.35]
7 Maternal side-effects6412Risk Ratio (M-H, Fixed, 95% CI)0.69 [0.52, 0.91]
7.1 Beta blocker versus methyldopa5302Risk Ratio (M-H, Fixed, 95% CI)0.69 [0.52, 0.91]
7.2 Calcium channel blockers versus methyldopa1110Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
8 Changed/stopped drugs due to maternal side-effects4272Risk Ratio (M-H, Fixed, 95% CI)2.80 [0.12, 67.91]
8.1 Beta blocker versus methyldopa4272Risk Ratio (M-H, Fixed, 95% CI)2.80 [0.12, 67.91]
9 Placental abruption1173Risk Ratio (M-H, Fixed, 95% CI)2.02 [0.19, 21.90]
9.1 Beta blocker versus methyldopa1173Risk Ratio (M-H, Fixed, 95% CI)2.02 [0.19, 21.90]
10 Total reported fetal or neonatal death (including miscarriage)191339Risk Ratio (M-H, Fixed, 95% CI)0.73 [0.42, 1.27]
10.1 Beta blocker versus methyldopa151160Risk Ratio (M-H, Fixed, 95% CI)0.73 [0.40, 1.33]
10.2 Calcium channel blocker versus methyldopa3159Risk Ratio (M-H, Fixed, 95% CI)0.31 [0.04, 2.65]
10.3 Ketanserin versus methyldopa120Risk Ratio (M-H, Fixed, 95% CI)3.0 [0.14, 65.90]
11 Preterm birth (< 37 weeks)9623Risk Ratio (M-H, Fixed, 95% CI)0.76 [0.55, 1.05]
11.1 Beta blocker versus methyldopa7574Risk Ratio (M-H, Fixed, 95% CI)0.81 [0.57, 1.15]
11.2 Calcium channel blocker versus methyldopa249Risk Ratio (M-H, Fixed, 95% CI)0.53 [0.24, 1.17]
12 Small-for-gestational age7597Risk Ratio (M-H, Fixed, 95% CI)0.80 [0.53, 1.21]
12.1 Beta blocker versus methyldopa6577Risk Ratio (M-H, Fixed, 95% CI)0.85 [0.55, 1.32]
12.2 Calcium channel blocker versus methyldopa120Risk Ratio (M-H, Fixed, 95% CI)0.4 [0.10, 1.60]
13 Admission to special care baby unit4478Risk Ratio (M-H, Fixed, 95% CI)0.92 [0.67, 1.26]
13.1 Beta blocker versus methyldopa3453Risk Ratio (M-H, Fixed, 95% CI)0.91 [0.66, 1.25]
13.2 Calcium channel blocker versus methyldopa125Risk Ratio (M-H, Fixed, 95% CI)1.27 [0.32, 5.12]
14 Neonatal hypoglycaemia4321Risk Ratio (M-H, Fixed, 95% CI)1.05 [0.50, 2.18]
14.1 Beta blocker versus methyldopa4321Risk Ratio (M-H, Fixed, 95% CI)1.05 [0.50, 2.18]
15 Neonatal bradycardia128Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
15.1 Beta blocker versus methyldopa128Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
16 Neonatal jaundice128Risk Ratio (M-H, Fixed, 95% CI)1.2 [0.47, 3.03]
16.1 Beta blocker versus methyldopa128Risk Ratio (M-H, Fixed, 95% CI)1.2 [0.47, 3.03]
Analysis 5.1.

Comparison 5 Any antihypertensive versus methyldopa (subgrouped by class of drug), Outcome 1 Severe hypertension.

Analysis 5.2.

Comparison 5 Any antihypertensive versus methyldopa (subgrouped by class of drug), Outcome 2 Proteinuria/pre-eclampsia.

Analysis 5.3.

Comparison 5 Any antihypertensive versus methyldopa (subgrouped by class of drug), Outcome 3 Additional antihypertensive.

Analysis 5.4.

Comparison 5 Any antihypertensive versus methyldopa (subgrouped by class of drug), Outcome 4 Antenatal hospital admission.

Analysis 5.5.

Comparison 5 Any antihypertensive versus methyldopa (subgrouped by class of drug), Outcome 5 Elective delivery (induction of labour + elective caesarean section).

Analysis 5.6.

Comparison 5 Any antihypertensive versus methyldopa (subgrouped by class of drug), Outcome 6 Caesarean section.

Analysis 5.7.

Comparison 5 Any antihypertensive versus methyldopa (subgrouped by class of drug), Outcome 7 Maternal side-effects.

Analysis 5.8.

Comparison 5 Any antihypertensive versus methyldopa (subgrouped by class of drug), Outcome 8 Changed/stopped drugs due to maternal side-effects.

Analysis 5.9.

Comparison 5 Any antihypertensive versus methyldopa (subgrouped by class of drug), Outcome 9 Placental abruption.

Analysis 5.10.

Comparison 5 Any antihypertensive versus methyldopa (subgrouped by class of drug), Outcome 10 Total reported fetal or neonatal death (including miscarriage).

Analysis 5.11.

Comparison 5 Any antihypertensive versus methyldopa (subgrouped by class of drug), Outcome 11 Preterm birth (< 37 weeks).

Analysis 5.12.

Comparison 5 Any antihypertensive versus methyldopa (subgrouped by class of drug), Outcome 12 Small-for-gestational age.

Analysis 5.13.

Comparison 5 Any antihypertensive versus methyldopa (subgrouped by class of drug), Outcome 13 Admission to special care baby unit.

Analysis 5.14.

Comparison 5 Any antihypertensive versus methyldopa (subgrouped by class of drug), Outcome 14 Neonatal hypoglycaemia.

Analysis 5.15.

Comparison 5 Any antihypertensive versus methyldopa (subgrouped by class of drug), Outcome 15 Neonatal bradycardia.

Analysis 5.16.

Comparison 5 Any antihypertensive versus methyldopa (subgrouped by class of drug), Outcome 16 Neonatal jaundice.

Comparison 6. Any antihypertensive versus calcium channel blocker (subgrouped by class of drug)
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Severe hypertension2136Risk Ratio (M-H, Fixed, 95% CI)2.09 [0.96, 4.57]
1.1 Glyceryl trinitrate versus calcium channel blockers136Risk Ratio (M-H, Fixed, 95% CI)1.56 [0.07, 35.67]
1.2 Beta blockers versus calcium channel blockers1100Risk Ratio (M-H, Fixed, 95% CI)2.14 [0.96, 4.80]
2 Proteinuria/pre-eclampsia2128Risk Ratio (M-H, Fixed, 95% CI)2.15 [0.73, 6.38]
2.1 Glyceryl trinitrate versus calcium channel blockers136Risk Ratio (M-H, Fixed, 95% CI)1.0 [0.10, 9.96]
2.2 Beta blockers versus calcium channel blockers192Risk Ratio (M-H, Fixed, 95% CI)2.67 [0.75, 9.42]
3 HELLP syndrome1100Risk Ratio (M-H, Fixed, 95% CI)1.5 [0.26, 8.60]
3.1 Beta blockers versus calcium channel blockers1100Risk Ratio (M-H, Fixed, 95% CI)1.5 [0.26, 8.60]
4 Additional antihypertensive1100Risk Ratio (M-H, Fixed, 95% CI)2.14 [0.96, 4.80]
4.1 Beta blockers versus calcium channel blockers1100Risk Ratio (M-H, Fixed, 95% CI)2.14 [0.96, 4.80]
5 Changed/stopped drug due to side-effects2136Risk Ratio (M-H, Fixed, 95% CI)2.6 [0.13, 50.25]
5.1 Glyceryl trinitrate versus calcium channel blockers136Risk Ratio (M-H, Fixed, 95% CI)2.6 [0.13, 50.25]
5.2 Beta blockers versus calcium channel blockers1100Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
6 Maternal side-effects1100Risk Ratio (M-H, Fixed, 95% CI)1.2 [0.39, 3.68]
6.1 Beta blockers versus calcium channel blockers1100Risk Ratio (M-H, Fixed, 95% CI)1.2 [0.39, 3.68]
7 Elective delivery (induction of labour + elective caesarean section)1100Risk Ratio (M-H, Fixed, 95% CI)0.89 [0.69, 1.15]
7.1 Beta blockers versus calcium channel blockers1100Risk Ratio (M-H, Fixed, 95% CI)0.89 [0.69, 1.15]
8 Caesarean section1100Risk Ratio (M-H, Fixed, 95% CI)1.57 [0.91, 2.71]
8.1 Beta blockers versus calcium channel blockers1100Risk Ratio (M-H, Fixed, 95% CI)1.57 [0.91, 2.71]
9 Placental abruption1100Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
9.1 Beta blockers versus calcium channel blockers1100Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
10 Total reported fetal or neonatal death (including miscarriage)2136Risk Ratio (M-H, Fixed, 95% CI)1.0 [0.06, 15.55]
10.1 Glyceryl trinitrate versus calcium channel blockers136Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
10.2 Beta blockers versus calcium channel blockers1100Risk Ratio (M-H, Fixed, 95% CI)1.0 [0.06, 15.55]
11 Preterm birth (< 37 weeks)136Risk Ratio (M-H, Fixed, 95% CI)0.63 [0.20, 1.91]
11.1 Glyceryl trinitrate versus calcium channel blockers136Risk Ratio (M-H, Fixed, 95% CI)0.63 [0.20, 1.91]
12 Small-for-gestational age136Risk Ratio (M-H, Fixed, 95% CI)1.0 [0.10, 9.96]
12.1 Glyceryl trinitrate versus calcium channel blockers136Risk Ratio (M-H, Fixed, 95% CI)1.0 [0.10, 9.96]
13 Admission to special care baby unit199Risk Ratio (M-H, Fixed, 95% CI)1.47 [0.44, 4.89]
13.1 Beta blockers versus calcium channel blockers199Risk Ratio (M-H, Fixed, 95% CI)1.47 [0.44, 4.89]
Analysis 6.1.

Comparison 6 Any antihypertensive versus calcium channel blocker (subgrouped by class of drug), Outcome 1 Severe hypertension.

Analysis 6.2.

Comparison 6 Any antihypertensive versus calcium channel blocker (subgrouped by class of drug), Outcome 2 Proteinuria/pre-eclampsia.

Analysis 6.3.

Comparison 6 Any antihypertensive versus calcium channel blocker (subgrouped by class of drug), Outcome 3 HELLP syndrome.

Analysis 6.4.

Comparison 6 Any antihypertensive versus calcium channel blocker (subgrouped by class of drug), Outcome 4 Additional antihypertensive.

Analysis 6.5.

Comparison 6 Any antihypertensive versus calcium channel blocker (subgrouped by class of drug), Outcome 5 Changed/stopped drug due to side-effects.

Analysis 6.6.

Comparison 6 Any antihypertensive versus calcium channel blocker (subgrouped by class of drug), Outcome 6 Maternal side-effects.

Analysis 6.7.

Comparison 6 Any antihypertensive versus calcium channel blocker (subgrouped by class of drug), Outcome 7 Elective delivery (induction of labour + elective caesarean section).

Analysis 6.8.

Comparison 6 Any antihypertensive versus calcium channel blocker (subgrouped by class of drug), Outcome 8 Caesarean section.

Analysis 6.9.

Comparison 6 Any antihypertensive versus calcium channel blocker (subgrouped by class of drug), Outcome 9 Placental abruption.

Analysis 6.10.

Comparison 6 Any antihypertensive versus calcium channel blocker (subgrouped by class of drug), Outcome 10 Total reported fetal or neonatal death (including miscarriage).

Analysis 6.11.

Comparison 6 Any antihypertensive versus calcium channel blocker (subgrouped by class of drug), Outcome 11 Preterm birth (< 37 weeks).

Analysis 6.12.

Comparison 6 Any antihypertensive versus calcium channel blocker (subgrouped by class of drug), Outcome 12 Small-for-gestational age.

Analysis 6.13.

Comparison 6 Any antihypertensive versus calcium channel blocker (subgrouped by class of drug), Outcome 13 Admission to special care baby unit.

Appendices

Appendix 1. Additional author searching

Additional searching carried out for the previous version (Abalos 2007),

The Cochrane Central Register of Controlled Trials (The Cochrane Library 2005, Issue 3), MEDLINE (1966 to November 2005), LILACS (1984 to November 2005) and EMBASE (1974 to November 2005) using the terms hypertens*, pre-eclamp*, preeclamp*, pre eclamp* and pregnan*.

What's new

DateEventDescription
14 June 2013New citation required but conclusions have not changedThree new trials were added for this update: one of beta blockers versus methyldopa; one of calcium channel blockers versus methyldopa; and a three-arm trial of labetalol versus methyldopa versus no treatment.
31 May 2013New search has been performed

Review authors updated. April 2013: search updated.

Nineteen trials added to excluded studies (not all are new studies as, for some, new information has become available leading them to be reclassified as excluded).

Four figures added: 'Risk of bias' graph, 'Risk of bias' summary, and two funnel plots.

History

Protocol first published: Issue 3, 2000
Review first published: Issue 2, 2001

DateEventDescription
6 August 2012AmendedSearch updated. Nineteen new reports added to Studies awaiting classification.
8 May 2008AmendedConverted to new review format.
14 November 2006New search has been performed

March 2006: Search updated. Six new trials were added to included studies. Twenty-seven new trials added to excluded studies (not all are new studies as, for some, new information has become available leading them to be reclassified as excluded).

Changes in the outcome tree (calcium channel blockers versus beta blockers are no longer referred to the beta blockers review (Magee 2000) as this comparison is now part of the group 'any antihypertensive versus calcium channel blockers').  Changes in the text to reflect new data.   An entire section describing the general characteristics of the excluded trials has been added in the text.

Acknowledgements to authors for unpublished data.

Contributions of authors

For this update, all changes on the text of the review were drafted by E Abalos with input by L Duley. E Abalos and L Duley performed the methodological assessment of all new studies, and extracted the data when appropriate. E Abalos, L Duley and W Steyn have commented on and agreed the final version.

E Abalos is the guarantor of the review.

Declarations of interest

E Abalos updated this systematic review in order to compile the new evidence for the update of the WHO recommendations for Prevention and treatment of pre-eclampsia and eclampsia. L Duley has been awarded an NIHR research grant for a programme of work on care at very preterm birth.

Sources of support

Internal sources

  • Centre for Perinatal Health Services Research, University of Sydney, Australia.

  • Centro Rosarino de Estudios Perinatales. Rosario, Argentina.

  • Nottingham Clinical Trials Unit, University of Nottingham, UK.

External sources

  • Human Reproduction Programme. World Health Organization. Geneva, Switzerland.

Differences between protocol and review

Methods updated.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Argentina 1985

MethodsAllocation concealment: not stated. Authors said '...randomly divided into two groups...'.
Participants60 women with SBP >/= 160 mmHg and/or DBP >/= 100 mmHg x 2, 24 hr apart, with or without proteinuria at trial entry.
Excluded: > 1 drug to control BP, or contraindication for beta blockers.
InterventionsExp: atenolol 50-250 mg/day.
Control: methyldopa 750-2000 mg/day.
OutcomesWomen: BP (mean).
Babies: gestational age, birthweight, Apgar score, stillbirth, neonatal deaths.
NotesMain paper in Spanish. Methods for measuring BP not mentioned.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...randomly divided into two groups..." no further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear risk.

Argentina 1987

MethodsAllocation concealment: not stated. Authors said 'open randomised study'.
Participants20 women with SBP > 159 mmHg and/or DBP > 99 mmHg x 2, 24 hr apart, +/- proteinuria.
Excluded: > 1 drug to control BP, or hypertensive emergency.
InterventionsExp: ketanserin 20-80 mg/day.
Control: methyldopa 500-2000 mg/day.
OutcomesWomen: none reported.
Babies: stillbirth, neonatal death, birthweight (mean), gestation at delivery (mean).
NotesInterim report of study ongoing in 1987. Methods for measuring BP not mentioned.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...open randomised study..." no further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

Argentina 1988

MethodsAllocation concealment: not stated. Authors said 'randomised' 'divided into 2 equal groups'.
Participants36 women > 14 weeks' gestation with BP >/= 140/90 mmHg and </= 170/110 mmHg.
InterventionsExp: mepindolol, increasing weekly doses, from 5-10 mg/day.
Control: methyldopa, increasing weekly doses from 500-2000 mg/day.
OutcomesWomen: additional antihypertensive, caesarean section, side-effects, maternal complications.
Babies: stillbirth, SGA (undefined).
NotesMethods for measuring BP not mentioned. Available only as an abstract.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...randomised study..." no further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study abstract.
Other biasUnclear risk.

Australia 1983

MethodsAllocation concealment: not stated. Authors said 'randomly allocated'.
Participants28 women in antenatal clinics with mild-moderate PIH (BP >/= 140/90 mmHg x 2 at least 24 hr apart).
Excluded: impaired renal function.
InterventionsExp: propranolol 30-160 mg/day.
Control: methyldopa 500-1000 mg/day.
OutcomesWomen: severe hypertension, proteinuria (undefined), additional antihypertensive, changed drugs due to side-effects, caesarean section.
Babies: perinatal death, preterm delivery, jaundice, bradycardia, hypoglycaemia, birthweight (mean).
NotesLondon School of Hygiene sphygmomanometer (random zero) used. No mention of which Korotkoff sound used for DBP.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...randomly allocated..." no further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear risk.

Australia 1985

MethodsAllocation concealment: not stated. Authors said 'allocated by series of random numbers'.
Participants183 women with singleton pregnancy and mild hypertension (DBP >/= 90 mmHg x 2, 24 hr apart, or DBP >/= 95 mmHg x 2, 12 hr apart, or DBP >/= 100 mmHg x 2, 8 hr apart).
Interventions

Exp: oxprenolol 40-320 mg x 2/day.
Control: methyldopa 250 mg x 2/day-1000 mg x 3/day.

If blood pressure not controlled, hydralazine in both groups.

OutcomesWomen: severe hypertension, proteinuria ('heavy and increasing requiring delivery'), additional antihypertensive, induction of labour, caesarean section,
Babies: stillbirth, neonatal death, admission to SCBU, days in SCBU, RDS, birthweight. (mean), Apgar (mean).
NotesKorotkoff phase IV used for DBP.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described: "...were allocated by a series of random numbers to...", no further details.
Allocation concealment (selection bias)Unclear riskNot described.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasHigh riskCointerventions: 48% of oxprenolol group and 35% of methyldopa group received a second or third antihypertensive.

Australia 2001

MethodsAllocation concealment: central telephone randomisation Although authors stated it was a placebo-controlled trial, data provided by authors suggest that they may have used a patch for the control, but not a matching placebo.
Participants16 women with gestational hypertension, defined as "de novo" hypertension after 20 weeks' gestation of > 140 and/or 90 mmHg on 2 readings, 6 hr apart; or a rise in systolic pressure of > 25 mmHg or a diastolic of 15 mmHg from a BP pre-pregnancy or in the first trimester.
InterventionsExp: transdermal glyceryl trinitrate patches 10 mg.
Control: patch for the control, but not a matching placebo.
OutcomesWomen: pre-eclampsia, side-effects.
Babies: not reported.
NotesTrial planned to recruit 220 women and stopped early due to side-effects (headache) in the treatment group. Additional data provided by authors.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described.
Blinding of participants and personnel (performance bias)
All outcomes
High riskDescribed as "...single blind, placebo controlled trial...".
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study letter to editor.
Other biasUnclear risk.

Brazil 1985

MethodsAllocation concealment: not stated. Authors said '...patients were randomly divided into two groups...'.
Participants100 women with chronic hypertension diagnosed before 20th week, BP >/= 140/90 mmHg x 2, 5 min apart. With no proteinuria and no contraindication to beta blockers.
InterventionsExp: pindolol 10-30 mg/day.
Control: no treatment.
OutcomesWomen: MAP, severe pre-eclampsia, side-effects.
Babies: abortions, fetal deaths, neonatal deaths, gestational age, birthweight, IUGR, Apgar score, congenital malformations, hypoglycaemia.
NotesMethods for measuring blood pressure not mentioned. Main paper in Portuguese.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: : "...patients were randomly divided into two groups...", no further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

Brazil 1988

MethodsAllocation concealment: consecutive numbered treatment boxes.
Participants40 pregnant women with chronic hypertension with DBP =/> 95 mmHg, without proteinuria.
InterventionsExp: pindolol 10-30 mg/day.
Control: methyldopa 500-2000 mg/day.
OutcomesWomen: BP, need for additional antihypertensives, severe HT, superimposed pre-eclampsia.
Babies: birthweight, Apgar score, fetal and neonatal death, preterm birth, SGA (undefined).
NotesMain paper in Portuguese. Methods for measuring BP not mentioned. Additional data provided by authors.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot stated.
Allocation concealment (selection bias)Low riskConsecutive numbered treatment boxes (personal communication).
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Low riskPublished and unpublished data provided by author.
Other biasUnclear risk.

Brazil 2000a

MethodsAllocation concealment: trial drug supplied by pharmacy in packs with serial numbers.
Withdrawals: 15 women (7.5%) excluded from the analysis (5 delivered in other hospitals, 9 dropped the study or failed to comply with treatment, 1 due to side-effects).
Participants199 singleton pregnant women with mild/moderate chronic hypertension (DBP > 90 mmHg and =/< 110 mmHg before 20 weeks' gestation, or with history of chronic hypertension), before 25 weeks' gestation and giving informed consent. Excluded: renal, cardiac or hepatic disease, IUGR diagnosed before trial entry, alcohol/drug abuse.
InterventionsExp: verapamil 240 mg x 3/day.
Control: placebo.
OutcomesWomen: BP, heart rate, severe hypertension, superimposed pre-eclampsia, side-effects, mode of delivery.
Babies: birthweight, gestational age, SGA, Apgar score, jaundice, hypoglycaemia, mortality.
NotesKorotkoff phase IV used for DBP. Main report in Portuguese, presented as a Doctoral Thesis. Additional data provided by authors.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Low riskConsecutive numbered treatment packs containing the study drug or placebo.
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble blind, only pharmacist providing treatment packs aware of the codes.
Blinding of outcome assessment (detection bias)
All outcomes
Low riskDouble blind, only pharmacist providing treatment packs aware of the codes.
Incomplete outcome data (attrition bias)
All outcomes
Low risk15 withdrawals (7.5%): 5 women delivered in other hospitals, 10 withdrew consent (1 due to side-effects).
Selective reporting (reporting bias)Unclear riskAssessment from published study thesis.
Other biasLow riskGroups appear comparable at baseline.

Caribbean Is.1990

MethodsAllocation concealment: women given number corresponding to sealed envelope and treatment batch. Envelope contained unblinding, kept by investigator and only opened when necessary. Envelopes collected at end of study. 2 centres.
Withdrawals: 1 woman, from placebo group.
Participants155 women with singleton pregnancy at 20-36 weeks' gestation, DBP < 85 mmHg x 2 before 20 weeks and > 84 mmHg after 20 weeks.
Excluded: type I diabetes, congestive heart failure, cardiac block, asthma, pre-pregnancy hypertension, antihypertensive treatment during current pregnancy.
Interventions

Exp: oxprenolol 160-320 mg x 2/day. Hydralazine 50-100 mg added if necessary to keep DBP < 86 mmHg.

Control: placebo, identical appearance.

OutcomesWomen: death, mean BP, severe hypertension, proteinuria (> 1+ or 0.25 g/L), additional antihypertensive, eclampsia, changed drugs due to side-effects, elective delivery, caesarean section, hospital admission, days in hospital, placental abruption.
Babies: perinatal death, preterm delivery (< 37 weeks), birthweight (mean), SGA (undefined, excludes stillbirth), 5 min Apgar < 7, admission to SCBU, RDS.
NotesKorotkoff phase IV used for DBP. For 23 women (15%), treatment unblinded and other treatment started. 16 for uncontrolled BP (5 exp, 11 control) and 7 for poor compliance/side-effects (4 exp, 3 control). Additional data provided by authors.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk"Randomisation was in stratified blocks of ten..."
Allocation concealment (selection bias)Low risk"...sealed envelopes and individual batch of drugs containing either the maximum amount of active drug that could be used in 20 weeks of treatment or placebos of identical appearance."
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble blind, placebo controlled.
Blinding of outcome assessment (detection bias)
All outcomes
Low riskYes.
Incomplete outcome data (attrition bias)
All outcomes
Low risk1 women (0.6%) lost to follow-up.
Selective reporting (reporting bias)Low riskPublished and unpublished data provided by author.
Other biasHigh riskFor 23 women (15%), treatment unblinded and other treatment started. 16 for uncontrolled BP (5 exp, 11 control) and 7 for poor compliance/side-effects (4 exp, 3 control).

France 1987

MethodsAllocation concealment: 'blinded envelopes'. Stratified in blocks of 10 at each clinic. Multicentre, 12 hospitals.
Withdrawals: 12 women (6%). 5 labetalol (3 lost to follow-up and 2 given methyldopa) and 7 methyldopa (all lost to follow-up).
Participants188 women with singleton pregnancy at 12-34 weeks' gestation, booked < 20 weeks and DBP >/= 90 mmHg.
Excluded: previous antihypertensive treatment this pregnancy, diabetes, depression, contraindication to beta blockers.
InterventionsExp: labetalol 200-600 mg x 2/day.
Control: methyldopa 250-750 mg x 2/day.
OutcomesWomen: proteinuria (> 2+ or 0.5 g/L), admission to hospital, caesarean section, elective delivery, additional antihypertensive, side-effects, changed drugs due to side-effects.
Babies: stillbirth, neonatal death, admission to SCBU, SGA (< 5th centile, excludes stillbirths), preterm delivery (< 37 weeks), 5 min Apgar < 8.
NotesKorotkoff phase IV used for DBP.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk"...stratified blocks of ten at each clinic..."
Allocation concealment (selection bias)Low riskBlinded envelopes.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low risk12 women (6.4%) excluded from analysis: 10 lost to follow-up (3 from labetalol and 7 from methyldopa group), and 2 "protocol violation".
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

France 1988

MethodsAllocation concealment: not stated. Authors said 'random order'.
3-arm study.
Participants63 women at 7-36 weeks' gestation with DBP > 90 mmHg x 2, 8 days apart).
InterventionsExp: (1) acebutolol 400-1200 mg; (2) labetalol 400-1200 mg.
Control: methyldopa 500-1500 mg.
OutcomesWomen: PE, caesarean section.
Babies: perinatal death, preterm delivery, birthweight (mean), Apgar, admission to SCBU, hypoglycaemia.
NotesNo mention about Korotkoff sound considered for DBP. Main paper in French.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"Apres tirage au sort, effectué par groupe de 9 patientes..."
Allocation concealment (selection bias)Unclear riskNot described.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

France 1994

MethodsAllocation concealment: sealed envelopes drawn by physician. Ordered using list of computer-generated random numbers.
Participants100 women with singleton pregnancy at > 20 weeks' gestation and mild-moderate hypertension (BP >/= 140/90 mmHg x 2). No other antihypertensive medication at trial entry.
InterventionsExp: nicardipine 20 mg x 3/day.
Control: metoprolol (slow release) 200 mg/day.
OutcomesWomen: severe hypertension, proteinuria (undefined), HELLP syndrome, additional antihypertensive, changed drug due to side-effects, induction of labour, caesarean section.
Babies: perinatal death, umbilical Doppler, admission to SCBU.
NotesKorotkoff phase IV used for DBP.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskList of computer-generated random numbers.
Allocation concealment (selection bias)Low riskSealed envelopes.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

Hong Kong 1990

MethodsAllocation concealment: not stated.
Authors said: 'randomised double-blind'.
Participants41 healthy nulliparous women admitted for PE (BP >/= 140/90 mmHg x 2 within 24 hours).
InterventionsExp: labetalol 200 mg x 3/day.
Control: placebo (character not stated).
OutcomesWomen: mean BP, severe hypertension, additional antihypertensive.
Babies: birthweight (mean), SGA (< 10th centile), gestation at delivery (mean).
NotesTrial reported as in progress in 1990. Missing data for some babies. No description of how BP measured. Available only as an abstract.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: authors only said: 'randomised double-blind controlled study'.
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble blind, placebo controlled.
Blinding of outcome assessment (detection bias)
All outcomes
Low riskYes.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study abstract.
Other biasUnclear risk.

India 1992

MethodsAllocation concealment: not stated. Authors said 'randomly allocated'.
Participants30 primigravid women at 24-37 weeks' gestation with mild-moderate PIH (BP>/ = 140/90 mmHg x 2, 6 hr apart).
Excluded: UTI, heart disease or other cause of hypertension.
InterventionsExp: metoprolol 50-150 mg x 2/day.
Control: methyldopa 250 mg x 3/day, increased to 2000 mg/day.
OutcomesWomen: severe hypertension.
Babies: perinatal death, preterm delivery, gestation at delivery, birthweight, Apgar at 1 and 5 min (mean).
NotesMethod for measuring BP not mentioned.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...randomly allocated...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear risk.

India 2002

MethodsMethods not stated, authors only say "...randomised by number..."
Participants

118 primigravid women with hypertension (BP > 140/90 x 2, 60 hours apart) during pregnancy with and without proteinuria at > 20 weeks' gestation.

Excluded: 2nd or more pregnancies, history of chronic HT, use of anti HT drugs.

Interventions

Exp: Nimodipine 30 mg every 6 hours.

Control: Methyldopa 250 mg every 6 hours.

OutcomesMAP, additional antihypertensives, serum creatinine and liver enzymes, induction of labour, serious maternal side-effects, perinatal deaths.
NotesMethod for measuring BP not mentioned.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...randomised by number...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low risk7 dropouts (6%), not specified in which group.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

India 2012

MethodsMixed sealed envelopes containing the assigned intervention. 3-arm trial.
Participants

149 women with PIH (140-159/90-109 mmHg x 2, 6 hours apart, without proteinuria) at 20-38 weeks' gestation.

Excluded: chronic HT, previous anti HT medication, secondary HT, UTI, diabetes, known medical/psychiatric disorders, multiple pregnancy, placenta previa, Rh isoimmunisation, congenital abnormality.

Interventions

Exp 1: Labetalol 100 mg x 2/day (up to 2500 mg).

Exp 2: Methyldopa 250 mg x 2/day (up to 2000 mg).

Control: no treatment.

OutcomesBP, laboratory parameters, gestational age at delivery, birthweight, 5 min Apgar score, maternal adverse events, maternal death, major morbidity, severe HT, proteinuria, severe pre-eclampsia, antenatal hospital admission, caesarean section, perinatal death, SGA, preterm babies, Admission to NICU.
NotesKorotkoff phase V for DBP.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)High risk"Allocation cards labelled L (labetalol), M (methyldopa) or N (no antihypertensive) were prepared by a staff clerk in equal numbers of 50 each and put into opaque envelopes that were subsequently sealed, mixed up and numbered 1-150."
Allocation concealment (selection bias)Low riskConsecutively numbered opaque sealed envelopes.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low risk1 woman withdrew consent in methyldopa group.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

India 2013

Methods"...patients were divided into two groups randomly..."
Participants180 pregnant women at > 20 weeks' gestation with BP > 140/90 mmHg x 2, 6 hours apart and proteinuria 1+ on dipstick x 2, 4 hours apart.
Interventions

Exp: Labetalol 100 mg t.i.d. (doubled every 48 hours until BP control).

Control: Methyldopa 250 mg t.i.d. (doubled every 48 hours until BP control).

OutcomesFall in BP after 7 days treatment, average dose of drug used, induction of labour, maternal side-effects.
NotesMethod for measuring BP not mentioned.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...patients were divided into two groups randomly...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

Ireland 1991

MethodsAllocation concealment: cards with 'test' or 'control' sealed in envelopes, shuffled and then numbered in sequence. Consecutive envelopes opened.
Participants36 women < 38 weeks' gestation with BP >/= 140/90 mmHg on 2 separate days, without proteinuria.
Excluded: if lived too far from the hospital to attend for frequent examinations.
InterventionsExp: choice between atenolol 50-100 mg/day and methyldopa 750-2250 mg/day. If monotherapy inadequate, 2 drugs combined. Bendrofluazide 2.5-5.0 mg added as a third agent when necessary.
Control: no antihypertensive.
OutcomesWomen: MAP, proteinuria.
Babies: perinatal death, Apgar, gestation age at delivery, birthweight, birthweight < 50th centile.
NotesKorotkoff phase V used for DBP. Additional data provided by authors.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)High risk"...shuffled into random order..."
Allocation concealment (selection bias)Low riskSealed, identical numbered envelopes.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

Israel 1986

MethodsAllocation concealment: not stated. Authors said 'randomly allocated'.
Participants32 women with singleton pregnancy at 27-33 weeks' gestation with PIH (DBP >/= 95 mmHg x 2 at least 6 hr apart).
Excluded: history of chronic renal disease or essential hypertension.
InterventionsExp: pindolol 15 mg/day.
Control: methyldopa up to 2000 mg/day (no other details).
OutcomesWomen: severe hypertension, new proteinuria (> 2+ or 0.5 g/L), eclampsia, side-effects, additional antihypertensive, changed drugs due to side-effects.
Babies: neonatal death, birthweight (mean), abnormal antenatal fetal heart rate, gestation at delivery (mean).
NotesMethods for BP measurement not mentioned.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...randomly allocated...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

Israel 1986a

MethodsAllocation concealment: not stated. Authors said 'randomly allocated'.
2 women with side-effects on hydralazine crossed over to pindolol + hydralazine, and reported in this group. Data only included if available as intention-to-treat.
Participants44 women at < 37 weeks with BP >/= 150/90 mmHg x 2 at least 24 hr apart.
Excluded: insulin-dependent diabetes, obstructive lung disease, contraindication to pindolol or hydralazine.
InterventionsExp: hydralazine 50-100 mg/day + pindolol 10-25 mg/day (in 2 daily doses).
Control: hydralazine 50-100 mg/day (in 2 daily doses).
OutcomesWomen: severe hypertension, proteinuria (> 1 g in 24 hr), side-effects, changed drug due to side-effects, caesarean section.
Babies: preterm delivery, SGA (< 250 on Usher's curve), hypoglycaemia, hypothermia, low Apgar score.
NotesNo mention of which Korotkoff sound used.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...randomly allocated...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline. 2 women with side-effects on hydralazine crossed over to pindolol + hydralazine, and reported in this group. Data only included if available as intention-to-treat.

Israel 1992

MethodsAllocation concealment: trial drug supplied by pharmacy in packs with serial numbers, in blocks of 6.
Participants60 women < 35 weeks' gestation with DBP 85-99 mmHg x 2, 12 hours apart, and no treatment for hypertension during this pregnancy.
Excluded: multiple pregnancy, contraindication to beta blockers or insulin-dependent diabetes.
Interventions

Exp: pindolol 5 mg x 2/day. If DBP still >/= 85 mmHg on day 3, increased to 5 mg x 3/day, if no response next day, increased to 10 mg x 2/day.
Control: identical placebo.

If DBP 100-109 mmHg x2 or > 110 mmHg x1, hydralazine added for pindolol group. In placebo group, pindolol given first, followed by hydralazine if DBP > 100 mmHg.

OutcomesWomen: additional antihypertensive, days in hospital, proteinuria > 2+ or > 0.5 g/L, treatment stopped due to side-effects, caesarean section.
Babies: perinatal death, gestation at delivery (mean), birthweight, 5 min Apgar > 7, SGA (< 10th centile), hypoglycaemia, jaundice.
NotesKorotkoff IV used for DBP.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk"...in a random fashion, in blocks of six, using serial numbers..."
Allocation concealment (selection bias)Low riskPindolol and placebo tablets supplied by pharmacy in numbered packets.
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble blind : "...placebo of identical appearance...".
Blinding of outcome assessment (detection bias)
All outcomes
Low riskCodes were broken only in the presence of severe side-effects.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

Israel 1995

MethodsAllocation concealment: not stated. Authors said 'randomly allocated'.
3-arm trial.
Participants51 women with BP 140-160/95-110 mmHg.
Excluded: proteinuria > 2+, contraindication to beta blockers, or any other disease.
InterventionsExp: (1) hydralazine 60-200 mg/day + propranolol 40-120 mg/day; (2) hydralazine 60-200 mg/day + pindolol 5-15 mg/day.
Control: hydralazine 60-200 mg/day.
OutcomesWomen: eclampsia, severe maternal morbidity, side-effects, caesarean section.
Babies: perinatal death, preterm delivery, SGA (< 10th centile), birthweight (mean).
NotesKorotkoff phase V used for DBP.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...randomly allocated...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear risk.

Italy 1997

MethodsAllocation concealment: not stated. Authors said: 'randomly allocated'.
Participants100 primigravid women at 26-36 weeks' gestation with SBP 140-160 mmHg, and DBP 90-110 mmHg in first 24 hr after admission and proteinuria < 300 mg/24 hr.
Excluded: if other medical maternal or fetal pathology (IUGR or altered biophysical profile).
InterventionsExp: nifedipine 40-120 mg/day orally and bed rest.
Control: bed rest alone.
OutcomesWomen: severe hypertension, proteinuria, days in hospital before delivery.
Babies: stillbirth, neonatal death, gestation at delivery (mean), birthweight, placental weight, SGA (undefined).
NotesMethods for measuring blood pressure not stated. Article in Italian.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...randomly allocated...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear risk.

Italy 1998

MethodsAllocation concealment: central telephone randomisation, stratified by centre and type of hypertension (chronic, gestational or unclassified). Multicentre, 33 hospitals.
Withdrawals: 22 women (8%), 13 exp and 9 control lost to follow-up.
Follow-up of children at 18 months: 190/252 (77%) responded to postal survey.
Participants283 women at 12-34 weeks' gestation, with mild-moderate hypertension (DBP 90-110 mmHg x 2, 4 hours apart).
Excluded: chronic diseases (such as diabetes or renal disease), fetal malformations, previous antihypertensive treatment or contraindications to nifedipine.
InterventionsExp: slow-release nifedipine 20-80 mg x 2/day orally.
Control: no antihypertensive.
OutcomesWomen: severe hypertension, proteinuria, caesarean section, admission to intensive care.
Babies: perinatal death, birthweight, SGA (< 10th centile), preterm delivery (< 34 and < 37 weeks), admission to SCBU, hyperglycaemia, jaundice, RDS, other serious neonatal problems.
NotesClassification of hypertensive disorders using Davey and MacGillivray system. Methods for measuring blood pressure not mentioned.
Data from follow-up excluded as > 20% lost.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated random numbers balanced by centre and stratified by type of hypertension.
Allocation concealment (selection bias)Low riskAllocation by telephone.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low risk22 women (7.8%) lost to follow-up.13 in nifedipine group and 9 in control group.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

Italy 1999

MethodsAllocation concealment: consecutive-numbered, opaque, sealed envelopes. 3-arm study. 6 women (17%) left the study due to side-effects (2 women) or mother's or baby's worsening conditions (4).
Participants36 women with singleton pregnancy, gestation > 24 weeks and PIH or PE (BP 140/90 mmHg or more, PE if proteinuria > 300 mg/24 hr).
Excluded: fetal abnormalities or chromosomal disorders, renal or hepatic disease, chronic hypertension.
InterventionsExp (1): transdermal glyceryl trinitrate 10 mg continuously 24 hr/day. Exp (2): transdermal glyceryl trinitrate 10 mg intermittently for 16 hr/day.
Control: Nifedipine 40 mg/day orally.
OutcomesWomen: caesarean section, BP (mean), stopped drug due to side-effects, severe hypertension, proteinuria/pre-eclampsia.
Babies: birthweight, fetal/neonatal deaths, preterm birth, IUGR, gestation at birth (mean).
NotesKorotkoff phase IV used for DBP. In the analysis the 2 GTN arms have been combined. Additional data provided by authors.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Low riskConsecutive opaque sealed envelopes (personal communication).
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
High risk6 withdrawals (16.6%). 3 in group 1, 1 in group 2 and 2 in group 3.
Selective reporting (reporting bias)Low riskAdditional data provided by authors.
Other biasUnclear risk.

Italy 2000

MethodsAllocation concealment: consecutive-numbered treatment boxes.
Participants20 women with pre-eclampsia (no further details).
InterventionsExp: nifedipine GITS 30-60 mg/day.
Control: methyldopa 500-1000 mg/day.
OutcomesWomen: BP, PE, Doppler abnormalities, need for drug adjustment, severe hypertension.
Babies: fetal and neonatal death, preterm birth, SGA (undefined), Apgar score.
NotesPublished as an abstract only. Method for measuring BP not stated. Additional data provided by authors.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Low riskConsecutive numbered treatment boxes provided by pharmacy (personal communication).
Blinding of participants and personnel (performance bias)
All outcomes
High riskSingle blind (only for participants).
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Low riskAdditional data provided by authors.
Other biasUnclear risk.

South Africa 1991

MethodsAllocation concealment: cards labelled R and Q picked blindly from a box, these identified drug container.
Participants32 women at 12-30 weeks' gestation with a singleton pregnancy and BP >/= 140/90 mmHg x 2 at least 6 hr apart, no proteinuria, no antihypertensive therapy and no other drug treatment.
InterventionsExp: prazosin 1-5 mg x 3/day.
Control: identical placebo.
OutcomesWomen: severe hypertension, proteinuria, duration of treatment, placental abruption, caesarean section.
Babies: perinatal death, gestation at delivery (mean), birthweight, SGA (< 10th centile) preterm delivery (< 37 weeks).
NotesMethod for measuring blood pressure not mentioned. The trial stopped early.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...in a randomised way, participants either received...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskAlthough identical placebo tablets were used, there is no information about whether it was a double-blind or a single-blind trial.
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNot described.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear risk.

South Africa 1993

MethodsAllocation concealment: not stated. Authors said: 'randomised open study'.
Withdrawals: 3 women (10%) lost to follow-up, but outcome for babies reported.
Participants29 women at 29-36 weeks' gestation with mild-moderate hypertension (DBP 90-110 mmHg).
Interventions

Exp: nifedipine started at 30 mg/day.
Control: methyldopa started at 750 mg/day.

Stated that 'dose adjustments were made, when necessary, every second day until control of BP was obtained'.

OutcomesWomen: additional antihypertensive, caesarean section, induction of labour, side-effects.
Babies: stillbirth, preterm delivery, gestation at delivery (mean), admission to SCBU.
NotesMethod for measuring BP not mentioned.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...randomised, prospective open study...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
High risk3 women (10.3%) did not complete the trial. 2 from one group and 1 from the other, not clear from the report in which group they were.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear risk.

Sudan 2002

MethodsAllocation concealment: not stated. Authors said: '...patients were randomly allocated...'.
Participants70 primigravid women with pre-eclampsia (BP =/> 90/109 mmHg x 2, 6 hr apart plus 2+ proteinuria in dipsticks) at 28-36 weeks' gestation. Singleton pregnancy.
Interventions

Exp: methyldopa 750-4000 mg/day
Control: no drug treatment.

All women in both groups were admitted to hospital for bed rest.

OutcomesWomen: BP, abruptio, imminent eclampsia, eclampsia, preterm delivery, caesarean section, maternal death.
Babies: birthweight, IUGR, admission to SCBU (reported as 'referral of baby'), perinatal deaths, Apgar score.
NotesKorotkoff IV sound used for DBP.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...patients were randomly allocated...".
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

Sweden 1984

MethodsAllocation concealment: telephone randomisation, no further details.
Participants52 women in antenatal clinic at < 37 weeks' gestation with singleton pregnancy, BP >/= 140/90 mmHg or an increase of at least 30 mmHg SBP or 15 mmHg DBP x 2 within 24 hr.
Excluded: imminent eclampsia, serious fetal distress, severe hypertension (> 170/110 mm Hg), Rh disease, diabetes, contraindication to beta blockers, 'social or psychological handicaps'.
InterventionsExp: metoprolol 100-200 mg x 2/day.
Control: identical placebo x 2/day.
OutcomesWomen: proteinuria (>/= 2+), severe hypertension, changed drugs due to side-effects, hospital admission, placental abruption, caesarean section.
Babies: perinatal death, gestation at delivery (mean) Apgar (mean).
NotesKorotkoff phase V used for DBP. Additional data provided by authors.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Low riskTelephone randomisation (personal communication).
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble blind. Placebos of identical appearance.
Blinding of outcome assessment (detection bias)
All outcomes
Low riskYes.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Low riskUnpublished data provided by author.
Other biasUnclear risk.

Sweden 1985

MethodsAllocation concealment: 'envelope randomisation'. No further information.
Withdrawals: 7 women (4%) dropped out (4 exp, 3 control). Multicentre, not stated how many hospitals.
Participants168 women in antenatal ward with singleton pregnancy at < 37 weeks, DBP >/= 90 mmHg x 2, no proteinuria.
Excluded: diabetes, asthma, heart disease, psychiatric or psychological disorders.
InterventionsExp: metoprolol 50-200 mg/day + hydralazine 50-300 mg/day.
Control: no antihypertensive.
OutcomesWomen: severe hypertension, proteinuria (> 1+ or 0.25 g/L), changed drugs due to side-effects, placental abruption, caesarean section.
Babies: stillbirth, neonatal death, preterm delivery (< 37 and < 34 weeks), SGA (undefined), bradycardia, hypoglycaemia, Apgar < 7 at 1 and 5 min, RDS.
NotesKorotkoff phase V used for DBP. Additional data provided by authors.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described "...randomly allocated...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

Sweden 1995

MethodsAllocation concealment: authors said: 'randomised by numbers to treatment with capsules'. Randomisation in blocks of 6. Information about allocation kept in sealed envelopes, opened if severe complications or side-effects. 5 centres in Sweden, 1 in Denmark.
Withdrawals: 7 women (6%), 1 dropout, 6 not re-evaluated after 3 days (4 exp, 2 control).
Participants118 women at 26-37 weeks, with singleton pregnancy and DBP 95-110 mmHg.
Excluded: if delivery expected within a week, history of alcohol or drug abuse, or other medication known to be toxic.
InterventionsExp: isradipine (slow release) 5 mg x 2/day.
Control: placebo x 2/day.
OutcomesWomen: eclampsia, severe hypertension (DBP >/= 110 mmHg), proteinuria >/=2+, need for additional antihypertensive, MAP, caesarean section, induction of labour, side-effects.
Babies: perinatal death, gestation at delivery (mean), admission to SCBU, birthweight (mean), placental weight.
NotesKorotkoff phase IV used for DBP. Description of BP measurements technique, and of criteria used to define hypertension and proteinuria.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk"...by numbers in blocks of 6..."
Allocation concealment (selection bias)Unclear riskNot described: "...parallel double blind study...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble blind.
Blinding of outcome assessment (detection bias)
All outcomes
Low riskYes.
Incomplete outcome data (attrition bias)
All outcomes
Low risk7 withdrawals (6%).
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

UK 1968

MethodsAllocation concealment: not stated. Authors said 'allocated at random'.
Participants100 pregnant women with DBP >/= 90 mmHg or more x 2, 48 hr apart.
InterventionsExp: methyldopa 250-1,000 mg x 2/day + bendrofluazide 5-10 mg/day.
Control: no treatment.
OutcomesWomen: mean BP, proteinuria, residual hypertension, length of gestation.
Babies: birthweight (mean), perinatal death.
NotesMethods for measuring BP not mentioned. According with BP at entry, women were divided in 2 groups: 'moderate' for those with DBP = or > 90 mmHg at entry (n = 42), and 'severe' for those with DBP = or > 100 mmHg (n = 58). For the main outcomes results are presented together.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...allocated at random...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear risk.

UK 1976

MethodsAllocation concealment: not stated. Authors said 'randomly allocated'.
Withdrawals: 5 women (2%) withdrawn from exp group.
Follow-up of 202 live born children. At 4 years, 34 (17%) lost to follow-up. At 7, years 7 (3%) lost to follow-up.
Participants247 women with BP >/= 140/90 mmHg if < 28 weeks' gestation, or >/= 150/95 mmHg if > 28 weeks' gestation x 2, 24 hr apart.
Excluded: diabetes, multiple pregnancy, Rh immunisation. Women > 36 weeks' gestation excluded during first year of the trial, thereafter excluded if > 32 weeks' gestation.
Interventions

Exp: methyldopa 750-4000 mg/day.
Control: no antihypertensive.

Hydralazine if severe hypertension.

OutcomesWomen: severe hypertension, proteinuria, caesarean section, elective delivery, side-effects, changed drug due to side-effects.
Babies: perinatal death, birthweight (mean), gestation at delivery (mean), SGA (< 2 SD below mean), babies nursed in an incubator, neurodevelopment at 4 and 7 years.
NotesKorotkoff phase IV used for DBP. Random zero sphygmomanometer.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "Treatment was allocated randomly...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low risk5 withdrawals (2%).
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

UK 1980

MethodsAllocation concealment: randomly allocated using random-number table.
Participants26 women < 38 weeks' gestation with PIH and no contraindication to beta blockers.
InterventionsExp: labetalol 400-800 mg/day.
Control: methyldopa 750-1500 mg/day.
OutcomesWomen: proteinuria, severe hypertension, caesarean section, induction of labour, side-effects.
Babies: stillbirth, birthweight (mean), gestation at delivery (mean), 1 min Apgar, admission to SCBU, jaundice.
NotesKorotkoff phase IV used for DBP.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandom number table.
Allocation concealment (selection bias)Unclear riskNot described: "...randomly allocated...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

UK 1982

MethodsAllocation concealment: envelope randomisation, no further information.
Participants126 women with either chronic hypertension or PIH, and DBP > 95 mmHg if < 20 weeks or 95-109 mmHg if > 20 weeks.
Interventions

Exp: labetalol 100 mg x 2/day, increased to maximum of 1200 mg/day.
Control: no antihypertensive.

If BP not controlled, hydralazine 25 mg x 3/day, increased to maximum of 200 mg/day.

OutcomesWomen: severe hypertension, proteinuria (undefined), caesarean section, placental abruption.
Babies: perinatal death, SGA (< 10th centile).
NotesMethods for measuring BP not mentioned. Additional data provided by authors.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskEnvelope randomisation (personal communication). No details on whether they were opaque, sealed and consecutively numbered.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Low riskAdditional data provided by author.
Other biasUnclear risk.

UK 1983

MethodsAllocation concealment: authors said 'allocated in double-blind and randomised manner'.
Withdrawals: some data missing for 35 women (29%). Data for each outcome only included if < 20% excluded.
Follow-up: 110 children (92%) seen at 1 year.
Participants120 women with PIH in third trimester admitted for bed rest, SBP 140-170 mmHg and DBP 90-110 mmHg x 2, 24 hr apart.
Excluded: women with contraindication to beta blockers.
InterventionsExp: atenolol 100-200 mg/day.
Control: placebo.
OutcomesWomen: proteinuria (> 0.5 g/24 hr), severe hypertension, additional antihypertensive, changed treatment due to side-effects, side-effects, admission to hospital prior to delivery, caesarean section.
Babies: perinatal death, SGA (< 10th centile), bradycardia, hypoglycaemia, jaundice, RDS. At 1 year: cerebral palsy, IQ < 1 SD below mean, weight.
NotesKorotkoff phase IV used for DBP. Random zero sphygmomanometer used for measuring blood pressure.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...allocated in double-blind, randomised manner...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble blind, placebo controlled.
Blinding of outcome assessment (detection bias)
All outcomes
Low riskYes.
Incomplete outcome data (attrition bias)
All outcomes
Low riskWithdrawals: some data missing for 35 women (29%). Data for each outcome only included if < 20% excluded. Follow-up: 110 children (92%) seen at 1 year.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear risk.

UK 1983a

MethodsAllocation concealment: not stated. Authors said 'allocated at random'. Stratified by gestational age.
Participants100 women with singleton pregnancy and DBP >/= 95 mmHg x 2 at least 24 hr apart, or > 105 mmHg x 1.
Excluded: asthma, heart failure, or heart block, diabetes, renal disease, or taking other hypertensive medication.
Interventions

Exp: oxprenolol 80-320 mg x 2/day.
Control: methyldopa 250-1000 mg x 3/day.

If BP not controlled, hydralazine added to both groups.

OutcomesWomen: severe hypertension, proteinuria (> trace on dipstick), induction of labour, caesarean section, additional antihypertensive, hospital admission.
Babies: perinatal death, birthweight (mean), 5 min Apgar < 7, antenatal fetal heart rate.
NotesKorotkoff phase IV used for DBP. Random zero sphygmomanometer.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described "...allocated at random...". Stratified by gestational age.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low risk4 women (4%) excluded, 2 lost to follow-up, 2 abortions.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear risk.

UK 1984

MethodsAllocation concealment: not stated. Authors said 'randomised trial'.
Participants60 women at 18-36 weeks' gestation with undefined hypertension.
InterventionsExp: atenolol 100 mg/day.
Control: methyldopa 250 mg x 3/day.
OutcomesWomen: proteinuria (undefined).
Babies: stillbirth, birthweight, SGA (< 10th centile) bradycardia, hypoglycaemia.
NotesKorotkoff phase V used for DBP.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...randomised trial...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear risk.

UK 1989

MethodsAllocation concealment: drug and placebo sent by manufacturer to hospital pharmacy with list of random numbers. Then dispensed by pharmacists. 5 centres.
Withdrawals: 8 (5%), 6 exp, 2 control. 2 women withdrew, 1 treated with ward stock labetalol, 1 developed rash, and 4 did not fulfil entry criteria.
Participants152 women from antenatal wards at 20-38 weeks' gestation with SBP 140-160 mmHg and DBP 90-105 mmHg x 2, 24 hr apart, and no proteinuria.
Excluded: history of hypertension, renal, metabolic, cardiovascular, respiratory or collagen disease.
InterventionsExp: labetalol 100-200 mg x 3/day.
Control: identical placebo.
OutcomesWomen: mean BP, severe hypertension, proteinuria (undefined), induction of labour, caesarean section, days in hospital (mean), side-effects.
Babies: perinatal death, preterm delivery (< 37 weeks), SGA (< 5th centile), admission to SCBU, RDS.
NotesKorotkoff phase IV used for DBP. Conventional sphygmomanometers used to measure blood pressure.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskList of random numbers.
Allocation concealment (selection bias)Low riskDrug and placebo sent by manufacturer to hospital pharmacy with list of random numbers. Then dispensed by pharmacists.
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble blind, placebo controlled.
Blinding of outcome assessment (detection bias)
All outcomes
Low riskYes.
Incomplete outcome data (attrition bias)
All outcomes
Low risk8 women excluded after randomisation (5.3%).
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear risk.

UK 1990

MethodsAllocation concealment: not stated. Authors said 'randomised' but no other information.
Withdrawals: 4 (12%), 1 exp (changed her mind), 3 control (2 severe hypertension, 1 breathlessness).
Participants33 women 12-24 weeks' gestation with SBP 140-170 mmHg and DBP 90-110 mmHg x 2, 24 hr apart.
Excluded: if 'usual' contraindications to beta blockers.
InterventionsExp: atenolol 50-200 mg/day.
Control: placebo (character not stated).
OutcomesWomen: mean BP, severe hypertension, stopped drug due to side-effects.
Babies: stillbirth, birthweight, SGA (< 5th centile), placental weight, gestation at delivery (mean).
NotesKorotkoff phase V used for DBP. The trial was stopped early when the principal investigator left Glasgow. Additional data provided by authors.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...randomised, placebo-controlled, double blind study...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble-blind, placebo-controlled trial.
Blinding of outcome assessment (detection bias)
All outcomes
Low riskYes.
Incomplete outcome data (attrition bias)
All outcomes
High risk4 withdrawals (12.1%).
Selective reporting (reporting bias)Low riskAdditional data provided by authors.
Other biasUnclear risk.

UK 1992

MethodsAllocation concealment: numbered, sealed opaque envelopes. Stratified by parity.
Participants114 women with singleton pregnancy at 24-39 weeks' gestation with DBP > 90 mmHg for > 24 hr and no proteinuria.
Excluded: psychoneurosis, cardiac abnormality, diabetes, asthma, contraindication to beta blockers, antenatal antihypertensive treatment.
InterventionsExp: labetalol 100 mg x 2/day, increased up to 400 mg x 3/day.
Control: no antihypertensive.
OutcomesWomen: proteinuria (> 1+ or 0.25 g/L), duration of stay in hospital (mean), side-effects, changed drug due to side-effects, elective delivery, caesarean section.
Babies: perinatal death, gestation at delivery (mean), preterm delivery (< 37 weeks), SGA (<5th centile), admission to SCBU, length of stay in hospital (mean).
NotesKorotkoff phase IV used for DBP. Additional data provided by authors.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Low riskNumbered, sealed opaque envelopes. Stratified by parity.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Low riskAdditional data provided by authors.
Other biasLow riskGroups appear comparable at baseline.

USA 1979

MethodsAllocation concealment: not stated. Authors said 'allocated randomly to treatment or no treatment'.
Participants58 women with hypertension before pregnancy or BP >/= 140/90 mmHg x 2 more than 24 hr apart before 20 weeks' gestation. Excluded: DBP > 100 mmHg, nulliparous, other major medical or obstetric problem.
InterventionsExp: methyldopa 750-2000 mg/day, hydrochlorothiazide 50 mg/day, hydralazine 75-250 mg/day.
Control: no antihypertensive.
OutcomesWomen: severe hypertension, proteinuria (> 1+ or > 300 mg/L in 24 hr), caesarean section.
Babies: perinatal death, gestation at delivery, birthweight < 2500 g, fetal distress, SGA (undefined).
NotesNo information about how BP measured. In exp group, 11 women had methyldopa + hydrochlorothiazide, 10 hydralazine + hydrochlorothiazide, 8 had all 3 drugs.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...allocated randomly to treatment or no treatment...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

USA 1987

MethodsAllocation concealment: physician drew a sealed envelope containing assignment.
Withdrawals: 14 women (7%), 8 exp and 6 control refused hospitalisation, but data reported for perinatal death.
Participants200 primigravid women in hospital at 26-35 weeks' gestation with SBP 140-160 mmHg and DBP 90-110 mmHg, proteinuria > 0.3 g/L and uric acid > 4.6 mg/dL.
Excluded: associated medical and obstetrical complications, other antihypertensive medication.
InterventionsExp: hospitalisation + labetalol 300 mg/day, increased every few days to max 2400 mg/day.
Control: hospitalisation alone.
OutcomesWomen: severe hypertension, increased proteinuria, eclampsia, placental abruption, caesarean section, renal function, days gained during management.
Babies: perinatal death, gestation at delivery (mean), birthweight (mean), placental weight, admission to SCBU, SGA (< 10th centile).
NotesNo mention of how BP measured.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated list of random numbers.
Allocation concealment (selection bias)Low riskSealed envelopes.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low risk14 withdrawals (7%). 8 in treatment group and 6 in control group refused hospitalisation, but data reported for perinatal death.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear risk.

USA 1987a

MethodsAllocation concealment: not stated. Authors said 'randomly allocated', no further information.
Participants25 women at < 34 weeks' gestation, singleton pregnancy with BP 140/90 mmHg x 2 at least 6 hr apart and no proteinuria. Presumed chronic hypertension.
InterventionsExp: methyldopa 750 mg x 3/day to 2000 mg x 4/day.
Control: placebo, in the same way.
If severe pre-eclampsia, hydralazine or MgSO4 added.
OutcomesWomen: MAP, new proteinuria (2+ or greater on urine dipsticks), PE (defined as a sudden rise of 30 mmHg SBP or 15 mmHg DBP and weight gain > 2 lbs/week, or proteinuria > 2+), elective delivery, side-effects.
Babies: perinatal death, gestation at delivery (mean), birthweight (mean and < 50th centile).
NotesNo information about how BP measured.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot described.
Allocation concealment (selection bias)Unclear riskNot described: "...randomly allocated...". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble blind, placebo controlled.
Blinding of outcome assessment (detection bias)
All outcomes
Low riskYes.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear risk.

USA 1990

MethodsAllocation concealment: envelope randomisation, using computer-generated random numbers. 3-arm study.
Withdrawals: 37 women (12%). 27 exp (21 excluded due to poor compliance, 3 twin, 1 abortion and 2 lost to follow-up) and 10 control (8 due to poor compliance, 1 twin and 1 spontaneous abortion).
Participants300 women in antenatal ward with chronic mild-moderate hypertension at 6-13 weeks' gestation. All had chronic hypertension before pregnancy and no associated medical complications.
InterventionsExp: (1) methyldopa 750-4000 mg/day (no other details). (2) labetalol 300-2400 mg/day (no other details).
Control: no antihypertensive.
OutcomesWomen: PE (defined as hypertension, proteinuria, and hyperuricaemia), additional antihypertensive, days in hospital, placental abruption, congestive heart failure, serum creatinine, uric acid.
Babies: perinatal death, gestation at delivery, birthweight < 2.5 kg, preterm delivery (< 37 weeks), SGA (undefined), admission to SCBU, hypoglycaemia, 5 min Apgar < 7.
NotesKorotkoff phase IV used for DBP. 36% of women were taking an antihypertensive at the time of trial entry. Additional data provided by authors.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated list of random numbers.
Allocation concealment (selection bias)Low riskSealed envelopes (personal communication).
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
High riskWithdrawals: 37 women (12%). 27 exp (21 excluded due to poor compliance, 3 twin, 1 abortion and 2 lost to follow-up) and 10 control (8 due to poor compliance, 1 twin and 1 spontaneous abortion).
Selective reporting (reporting bias)Low riskAdditional data provided by authors.
Other biasLow riskGroups appear comparable at baseline.

USA 1992

MethodsAllocation concealment: physician drew sealed envelope containing assignment. Computer-generated random numbers.
Withdrawals: 3 women (1.5%) lost to follow-up (2 exp, 1 control).
Participants200 primigravid women at 26-36 weeks' gestation with SBP 140-160 mmHg and/or DBP 90-110 mmHg 24 hr after hospitalisation, proteinuria > 300 mg/24 hr, and/or uric acid > 6 mg/dL.
Excluded: associated medical or obstetric complications, or fetal compromise (suspected abnormal fetal growth by US, abnormal fetal testing).
InterventionsExp: nifedipine 40-120 mg/day.
Control: bed rest alone.
OutcomesWomen: MAP, severe proteinuria (> 5 g/24 hr), antenatal hospital stay (mean), days gained during management, caesarean section, placental abruption, HELLP syndrome.
Babies: stillbirth, neonatal death, birthweight, preterm delivery (< 37 weeks), SGA (< 10th centile), admission to SCBU, days in SCBU (mean).
NotesMethod of measuring blood pressure not mentioned.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated list of random numbers.
Allocation concealment (selection bias)Low riskSealed envelopes.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low risk3 women lost to follow-up (1.5%). 2 in treatment group and 1 in control group.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasLow riskGroups appear comparable at baseline.

Venezuela 1988

  1. a

    BP: blood pressure
    DBP: diastolic blood pressure
    exp: experimental
    GITS: gastrointestinal therapeutic system
    GTN: glyceryl trinitrate
    HELLP: syndrome of haemolysis, elevated liver enzymes and low platelets
    hr: hour(s)
    HT: hypertension
    IUGR: intrauterine growth restriction
    IV: intravenous
    MAP: mean arterial pressure
    MgSO4: magnesium sulphate
    min: minutes
    NICU: neonatal intensive care unit
    PE: pre-eclampsia
    PIH: pregnancy-induced hypertension
    RDS: respiratory distress syndrome
    SBP: systolic blood pressure
    SCBU: special care baby unit
    SD: standard deviation
    SGA: small-for-gestational age
    t.i.d.: dosing three times daily
    US: ultrasound
    UTI: urinary tract infection

MethodsAllocation concealment: not stated. Treatment assigned using random-number tables.
Participants31 women > 14 weeks' gestation with either chronic hypertension or mild-moderate PIH (BP 140-169/90-109 mmHg x 2 after 5 min rest).
Excluded: contraindication to beta blockers, Rh or haemorrhagic disorders.
InterventionsExp: mepindolol 5 mg/day, increased weekly to 10 mg/day.
Control: methyldopa 250 mg x 2/day increased weekly to 250 mg x 4/day.
OutcomesWomen: severe hypertension, caesarean section, induction of labour.
Babies: perinatal death, gestation at delivery, birthweight, Apgar score.
NotesMain paper in Spanish.
Method of measuring blood pressure not mentioned.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandom-number table.
Allocation concealment (selection bias)Unclear riskNot described: [Estudio prospectivo randomizado] "randomised, prospective trial". No further details.
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label trial.
Blinding of outcome assessment (detection bias)
All outcomes
High riskNo.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo.
Selective reporting (reporting bias)Unclear riskAssessment from published study report.
Other biasUnclear risk.

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
  1. a

    ADP: adenosine diphosphate
    b.i.d.: twice a day
    BP: blood pressure
    DBP: diastolic blood pressure
    HT: hypertension
    IUGR: intrauterine growth retardation
    IV: intravenous
    LBW: low birthweight
    min: minutes
    MAP: mean arterial pressure
    NICU: neonatal intensive care unit
    NO-donor: Nitric oxide donor
    NST: non-stress test
    PIH: pregnancy-induced hypertension
    PI: pulsatility index
    PE: pre-eclampsia
    RCT: randomised controlled trial
    RDS: respiratory distress syndrome
    RI: resistance index
    SBP: systolic blood pressure
    SCBU: special care baby unit
    SD: standard deviation
    S/D ratio: ratio between peak systolic to end-diastolic flow velocity
    SGA: small-for-gestational age
    t.i.d.: dosing three times daily
    vs: versus

Argentina 1990

Most women with severe chronic hypertension. Women with mild-moderate HT not analysed separately.

Methods: open, prospective, randomised, comparative 3-arm trial.

Participants: 90 women with severe chronic hypertension.

Intervention: Atenolol (50-200 mg daily), vs methyldopa (500-2000 mg daily) vs ketanserin (80-120 mg daily).

Outcomes: BP, gestational age at delivery, birthweight, 1-min Apgar score, fetal and neonatal mortality.

Argentina 1994Not clearly randomised. Available as abstract only.
Methods: 'divided into two groups'. No further information.
Participants: 187 women with chronic hypertension (n = 66) or gestational hypertension (n = 121).
Interventions: atenolol 40-100 mg/day versus methyldopa 250-2000 mg/day.
Outcomes: superimposed pre-eclampsia, maternal BP, birthweight.
Australia 1985aComparison of 2 alpha agonists.
Methods: 'prospective, double blinded'. Women entered in a numerical sequence. No numbers missed or used a second time.
Participants: 100 women with BP > 130/85 mmHg or a rise of 30/15 mmHg from previous values.
Intervention: clonidine 150-1200 mcg/day versus methyldopa 250-2000 mg/day. If additional treatment needed, hydralazine.
Outcomes: severe hypertension, need for additional drug, stopped treatment due to side-effects, stillbirth, neonatal death, preterm delivery, birthweight (mean), SGA, 5-min Apgar.
Australia 1991Entry criteria was DBP greater than 1 SD above the reported mean for gestational age. Mean BP of recruited women was 129/84 mmHg at entry to the trial (122-136/79-89 mmHg) for the placebo group, and 126/82 mmHg (118-134/79-85) for the treatment group.
Participants: 52 nulliparous with singleton pregnancies between 28 and 34 weeks of gestation, without proteinuria.
Intervention: clonidine from 200 to 800 mcg a day plus hydralazine from 50 to 200 mg a day, and placebo.
Outcomes: severe hypertension, imminent eclampsia, eclampsia, severe proteinuria, antepartum haemorrhage, HELLP syndrome, fetal distress, fetal death, IUGR.
Belgium 1988Comparison of 2 beta blockers. Available as abstract only.
Methods: 'randomised', no further information.
Participants: 23 women with BP at least 140/90 mmHg x 2 and no proteinuria.
Intervention: atenolol 100 mg a day vs pindolol 15 mg a day.
Outcomes: umbilical PI, maternal BP, birthweight, Apgar score.
Brazil 1997Single-dose intervention. No clinical outcomes studied (effect of nifedipine in placental blood flow). Article in Portuguese. Only abstract translated into English.
Methods: double blind, placebo-controlled, RCT.
Participants: 65 women with pre-eclampsia.
Intervention: nifedipine 20 mg orally (only dose) vs placebo.
Outcomes: placental blood flow prior and 30 minutes after the intervention.
Brazil 2000Quasi-random design. Main paper in Portuguese.
Methods: alternated allocation (data extracted from original thesis). 11 women (10.5%) excluded after trial entry.
Participants: 105 women with singleton pregnancies diagnosed with pre-eclampsia, chronic hypertension, and pre-eclampsia superimposed to chronic hypertension.
Intervention: isradipine (slow release), 5 mg every 12 hr vs atenolol 50 mg every 12 hr.
Outcomes: BP, maternal heart rate, proteinuria, maternal side-effects, mode of delivery, gestational age, birthweight, SGA babies, Apgar score.
Brazil 2000b40 women (24%) excluded after randomisation. Reasons for exclusion were: missed appointment for Doppler (70%), non-compliance (20%), side-effects (7.5%), preterm delivery (2.5%). Data were not presented by treatment arm. Main paper in Portuguese.
Methods: randomised, double-blind, placebo-controlled trial.
Participants: 123 pregnant women with chronic hypertension.
Intervention: verapamil 240 mg/day vs placebo during 30 days.
Outcomes: Doppler PI, RI and S/D ratio, incidence of pre-eclampsia, birthweight, gestational age at delivery, SGA.
China 1991Herbal medicine vs magnesium sulphate. No clinical data available. Article in Chinese. Only abstract translated into English.
Methods: not reported. Authors said: '...randomly designed to...'.
Participants: 75 women with 'hypertension syndrome of pregnancy'.
Intervention: Magnesium sulphate 20-25 g/day vs ligustrazine 120-160 mg/day.
Outcomes: MAP, proteinuria, haematocrit, side-effects, positive rate of NST, Apgar score.
China 1993Only dose intervention. Sublingual nifedipine previous to caesarean section. Article in Chinese. Only abstract translated into English.
Methods: not reported. Indexed as publication type: RCT.
Participants: 33 women with pre-eclampsia undergoing emergent caesarean section.
Intervention: sublingual nifedipine, 16 mg (only dose). Control group not reported in abstract.
Outcomes: MAP, systolic and diastolic BP, maternal heart rate, postoperative haematocrit, side-effects.
China 1998Single-dose intervention. No clinical outcomes studied (effect of nimodipine in retinal blood flow). Article in Chinese. Only abstract translated into English.
Methods: not stated. Indexed as publication type: RCT.
Participants: 28 women with PIH.
Intervention: nimodipine 30 mg orally (only dose) vs IV magnesium sulphate.
Outcomes: retinal PI.
China 1999Herbal medicine + nifedipine vs nifedipine. No clinical outcomes studied. Article in Chinese. Only abstract translated into English.
Methods: not stated. Indexed as publication type: RCT.
Participants: 95 women with PIH.
Intervention: prepared rhubarb + nifedipine vs nifedipine.
Outcomes: serum lipids, and other blood tests.
China 2000Less than 7 days treatment. Treatment was given only during labour. Article in Chinese. Only abstract translated into English.
Methods: "64 cases of PIH were randomly divided into...".
Participants: 64 women with PIH.
Interventions: Nifedipine orally given every 6 hrs during labour vs no treatment.
Outcomes: postpartum haemorrhage.
Cuba 1994Quasi-random design. Article in Spanish.
Methods: alternate allocation (data provided by author).
Participants: 90 pregnant women with chronic hypertension.
Intervention: methyldopa (1-2 g/day) or hydralazine (100-200 mg/day) vs no treatment.
Outcomes: BP, superimposed pre-eclampsia, abruption, preterm delivery, LBW, Apgar score, RDS, hypoxia, fetal death.
Czech Republic 1993Comparison of 2 beta blockers. Article in Czech. Only abstract translated into English.
Methods: 'divided at random'. No further information.
Participants: 40 women with DBP 95-105 mmHg.
Intervention: atenolol 50-100 mg/day versus bisoprolol 5-10 mg/day.
Outcomes: BP, maternal heart rate, side-effects.
Denmark 1991Intervention is not an antihypertensive: magnesium vs placebo.
Methods: "...patients were allocated in a double-blind and randomised manner, based in a computer-generated list of numbers...".
Participants: 61 women with PIH. Chronic HT excluded. Withdrawals: 3 women (2 from intervention group, 1 from control group) excluded after randomisation.
Intervention: 48-hr of either IV magnesium or placebo infusion followed by daily oral magnesium or placebo tablets.
Outcomes: MAP, caesarean section, induction of labour, side-effects, gestational age, birthweight, Apgar score, admission to SCBU and days of stay.
Denmark 2000Intervention is not an antihypertensive: magnesium vs methyldopa.
Methods: RCT. Allocation concealment by numbered sealed opaque envelopes.
Participants: 33 women with PIH. Chronic HT excluded.
Intervention: magnesium, 48-hr IV infusion followed by daily oral magnesium vs methyldopa 250 mg x 4/day.
Outcomes: BP, gestational age, birthweight, admission to SCBU and length of stay, serum magnesium.
Dominican Rep 1992Not a RCT. Article in Spanish. Only abstract translated into English.
Methods: not stated. Authors only says "...divided into 2 groups...". Women known as given the drugs under study were also included.
Participants: 50 pregnant women with chronic HT + superimposed pre-eclampsia.
Intervention: slow-release nifedipine 20 mg every 8 hr vs methyldopa 500 mg every 12 hr.
Outcomes: BP, Apgar score.
Dominican Rep 1992aNot a RCT. Article in Spanish. Only abstract translated into English.
Methods: not stated. Authors only says "...divided into 2 groups...".
Participants: 30 pregnant women with severe pre-eclampsia.
Intervention: methyldopa 250-500 mg every 5 hr vs hydralazine 20-50 mg every 8 hr.
Outcomes: BP, maternal side-effects.
Egypt 1988No relevant clinical outcomes reported. Available as abstract only.
Methods: 'patients were randomly allocated to three treatment groups'. No further information.
Participants: 50 primigravidae with pre-eclamptic toxaemia and 20 multigravidae with essential hypertension in their late pregnancy.
Interventions: 3-arm trial: bromocriptine 5 mg, methyldopa 1 gr, and placebo, in different combinations. No further information.
Outcomes: serum prolactin and serum placental lactogen, BP. 1-year follow-up reported.
Egypt 1993One-week intervention. Outcomes measured at 30 min, 3 and 7 days.
Methods: 'randomly allocated' .
Participants: 30 women with PE in the third trimester. 25 women had mild PE with DBP 100-109 mmHg and 5 had severe PE with DBP >/= 110 mmHg.
Intervention: nifedipine 20 mg every 8 hr for 7 days or placebo in the same time and duration.
Outcomes: BP and fetal heart rate measured at 30 min, 3 and 7 days. Renal function tests and Doppler scans of umbilical cord.
Egypt 1997The intervention is not an antihypertensive. Naltrexone vs placebo. Available as an abstract only.
Methods: "...were randomly allocated to either naltrexone (...) or placebo".
Participants: 20 women with PIH at 30-36 weeks' gestation.
Intervention: naltrexone (opioid receptor antagonist), 50 mg every 12 hr vs placebo.
Outcomes: BP, proteinuria, oedema, prolactin levels, gestational age, status of the baby at birth.
Egypt 2009The intervention is not an antihypertensive. Ozone therapy (rectal insufflations).
Methods: "...were randomly assigned into two groups in equal numbers as follow...".
Participants: 30 hypertensive pregnant women at 24 weeks' gestation.
Intervention: ozone therapy by rectal Insufflations 3 sessions/week for 7 weeks + methyldopa vs methyldopa.
Outcomes: BP, resistance and pulsatility index, dose of methyldopa required.
Finland 1988Comparison of 2 beta blockers.
Methods: 'according to randomisation table'. No further information.
Participants: 51 women with BP > 149/94 mmHg x 2 in sitting position after 2 days bed rest in hospital.
Intervention: atenolol 50-100 mg/day versus pindolol 10-20 mg/day. If needed, hydralazine 150 mg/day added.
Outcomes: stillbirths, side-effects, need for additional drug, caesarean section, gestation at delivery (mean), birthweight (mean), 5 min Apgar.
Finland 1988aNo relevant clinical outcomes reported, report of ongoing study. Available as abstract only.
Methods: 'randomised pilot trial'. No further information.
Participants: 25 women with PIH.
Interventions: nifedipine 30-60 mg a day versus no treatment.
Outcomes: mean DBP, birthweight (mean).
Finland 1995Comparison of 2 beta blockers. Less than 7 days treatment, single-dose study. Women with mild hypertension not reported separately from severe hypertension.
Methods: 'randomly chosen'. No further information.
Participants: 24 women with a singleton pregnancy at 28-40 weeks, and either mild or severe pre-eclampsia (BP > 160/110 mmHg plus proteinuria > 5 g/24 hr, or BP 140/90-160/110 mmHg plus proteinuria < 5 g/24 hr).
Intervention: atenolol 0.15 mg/kg IV vs pindolol 0.006 mg/kg IV in 100 mL of Ringer's solution. Infusion time 15-20 min.
Outcomes: utero and umbilicoplacental vascular impedance, fetal haemodynamics and cardiac function.
Finland 1999Main outcomes were assessed only at 5-7 days of inclusion. 29% of women were excluded from the analysis.
Methods: randomised, double-blind, double-dummy study.
Participants: 24 women with singleton pregnancies between 29 and 39 weeks with BP > 140/90 mmHg x2, 6 hr apart, and proteinuria > 0.3 g in 24 hr urine collection.
Intervention: isradipine 2.5 mg twice daily or placebo vs metoprolol 50 mg twice daily or placebo (double-dummy study).
Outcomes: insulin sensitivity, uric acid, degree of proteinuria, lipids and lipoproteins, BP, umbilical artery RI, birthweight, placental weight, caesarean section, Apgar scores.
France 1988aNo clinical outcomes reported. Outcomes assessed at 4 weeks after trial entry. Available as abstract only.
Methods: 'randomised' no further information. 3-arm study.
Participants: 29 women with isolated hypertension after 'a mean period of 18 weeks of pregnancy'.
Intervention: pindolol vs atenolol versus methyldopa.
Outcomes: BP, maternal heart rate, serum sodium, potasium, uric acid, creatinine, plasma renin activity and aldosterone.
France 1990No clinical data reported. Available as congress abstracts only (1 in English, 3 in French).
Methods: 'randomised protocol'.
Participants: 21 women with moderate hypertension (SBP 140-180 mmHg and DBP 90-120 mmHg).
Intervention: oral atenolol (n = 12) vs nifedipine (n = 9) (no doses reported).
Outcomes: BP, Doppler measures, birthweight and length, Apgar score, admission to SCBU.
Germany 2012

Not mild-moderate HT. Prevention study.

Methods: randomised, placebo-controlled, double blind study.

Participants: 111 pregnant women with abnormal placental perfusion at 19 - 24 weeks' gestation.

Intervention: NO-donor penterythriltetranitrat (n = 54) vs placebo (n = 57).

Outcomes: utero-placental perfusion, preterm birth, IUGR, pre-eclampsia, fetal deaths.

Hong Kong 1993No clinical data available. Abstract report.
Methods: allocated in 'randomised double manner'. No further information. 4 women (6.2%) excluded after randomisation.
Participants: 65 primigravid women with a singleton pregnancy at > 20 weeks' gestation and BP 140-165/90-105 mmHg x 2, 6 hr apart but no proteinuria.
Interventions: labetalol (dose not reported) vs placebo (vitamin C).
Outcomes: BP, need for additional antihypertensives, induction of labour, proteinuria, gestational age, mode of delivery, birthweight, Apgar score.
Hungary 199928% of women excluded after randomisation (7 because of treatment duration not exceeding 10 days and 2 dropped out).
Methods: allocation 'according to randomisation list'. No further information.
Participants: 32 healthy primigravidae with BP at least 140/90 mmHg x 2 at least 6 hr apart.
Interventions: calcium dobesilate 2 g a day vs placebo.
Outcomes: new proteinuria, caesarean section, placental abruption, preterm delivery.
India 1999Intervention is an antiplatelet agent. Available as abstract only.
Methods: randomised, placebo-controlled trial.
Participants: 163 women with PIH of 20-32 weeks' gestation.
Intervention: aspirin 60 mg a day vs placebo from 22 until 38 weeks of gestation.
Outcomes: prevention of PIH grade B (BP 160/110 mmHg x 2, 4 hr apart), proteinuria 2+ or more, perinatal mortality, maternal mortality, eclampsia, SGA (< 10th centile).
India 2011

Not possible to extract data. Available as abstract only.

Methods: prospective randomised comparative study.

Participants: 200 women with singleton pregnancy, 28-40 weeks' gestation with pregnancy-induced hypertension.

Intervention: oral labetalol versus oral methyldopa.

Outcomes: BP, control of proteinuria, mode of delivery, neonatal Apgar score, NICU admission, neonatal complications, perinatal deaths.

India 2012a

Includes women with severe hypertension. IV labetalol given to this subgroup, not analysed separately.

Methods: prospective, randomised controlled parallel group study.

Participants: 90 women with singleton, vertex pregnancies at 20-40 weeks' gestation with BP >/= 140/90, with and without proteinuria.

Intervention: oral or IV Labetalol versus oral methyldopa.

Outcomes: serum urea and creatinine, significant proteinuria, severe maternal complications, induction of labour, caesarean section, lactation, gestational age at delivery, Apgar score, admission to NICU, RDS, bradycardia, jaundice, hypoglycaemia.

India 2012bIntervention is not an antihypertensive: magnesium sulphate vs placebo. No clinical outcomes studied. 27% women (18/66) excluded after randomisation.
Methods: women were randomly allocated by sealed envelopes containing computer-generated random numbers. Double-blind, placebo-controlled.
Participants: 66 women with mild pre-eclampsia or PIH after 34 weeks' gestation.
Intervention: 24-hr of either IV magnesium sulphate or placebo infusion.
Outcomes: umbilical artery and fetal middle cerebral artery pulsatility index.
Iran 2000Quasi-random design (data from personal communication). Available as abstract only.
Methods: 'patients were sequentially assigned to one of two randomised groups'. Alternate allocation (data obtained from personal communication).
Participants: 37 pregnant women over 26 weeks' gestation with blood pressure over 140/90 mmHg (after 24-48 hr resting) + proteinuria or generalised oedema.
Intervention: nifedipine 10 mg t.i.d. vs hydralazine 10 mg t.i.d.
Outcomes: BP, termination of pregnancy, side-effects.
Iran 2005Intervention is not an antihypertensive: IV magnesium sulphate vs oral magnesium chloride. No clinical outcomes studied.
Methods: randomised controlled open clinical trial.
Participants: 68 women with mild pre-eclampsia.
Intervention: IV magnesium sulphate (2 g/h) or oral magnesium chloride (4 g/2h).
Outcomes: serum Mg levels at 3, 6 and 12 hr after administration.
Israel 1988Comparison of 2 beta blockers. Published as abstract only.
Methods: 'allocated in blind and randomised manner'. No further information.
Participants: 30 women with SBP 140-170 mmHg and DBP 90-110 mmHg x 2, 6 hr apart.
Intervention: Atenolol 100 mg plus 2 placebo tablets vs pindolol 5 mg x 3/day.
Outcomes: gestation at delivery (mean).
Israel 1992aComparison of 2 beta blockers.
Methods: 'randomly allocated to double blind treatment'. No further information.
Participants: 20 women with mild PE, BP >/= 140/90 mmHg.
Interventions: propranolol 40 mg x 3/day vs pindolol 5 mg x 3/day, for 7 days.
Outcomes: BP, umbilical artery Doppler.
Israel 1999Single-dose intervention.
Methods: double-blind, placebo-controlled RCT.
Participants: 23 women with PIH.
Intervention: sublingual tablet of Isosorbide dinitrate (5 mg) or placebo (single dose).
Outcomes: maternal BP and heart rate, umbilical artery Doppler.
Italy 1986Not an RCT (matched controls). Available as abstract only.
Methods: 'randomised protocol', no further information, for group A (nifedipine or atenolol), control group (B) was matched by age and parity with group A. Results in group A were not presented separately.
Participants: 10 women with mild-moderate hypertension in the third trimester (group A).
Interventions: atenolol 100 mg a day or slow-release nifedipine 20 mg x 2/day (group A) vs diuretics or bed rest (group B).
Outcomes: BP, gestational age, birthweight, Apgar score, serum bilirubin, preterm delivery, RDS, side-effects.
Italy 1990Quasi-randomised design. 2 trials with same methods reported in 1 paper (1) 44 women (2) 50 women.
Methods: allocation by 'order of attendance at clinic or department'.
Participants: women with BP =/> 140/90 mmHg x 2 over 8 hr, normal BP before pregnancy.
Intervention: (1) slow-release verapamil 360-480 mcg/day vs pindolol 15-20 mg/day. (2) slow-release verapamil 360-480 mcg/day vs atenolol 100-150 mg/day.
Outcomes: caesarean section, baby death, Apgar (mean), gestation at delivery (mean).
Italy 1990aIntervention is an antiplatelet agent. No clinical outcomes reported. Available as abstract only.
Methods: '...using a random selection...'. No further information.
Participants: 20 women with PIH before 36 weeks' gestation.
Intervention: picotamide (no dose reported) vs no treatment.
Outcomes: platelet aggregation, ADP-threshold values, collagen concentration thresholds.
Italy 2000aWomen had chronic hypertension or history of hypertension or IUGR (results were not presented separately).
Methods: "...patients were randomly allocated to two treatments...".
Participants: 68 women with either chronic hypertension or with previous history of PE or IUGR.
Intervention: glyceryl trinitrate transdermal patch (5 mg/24 hr) for 14-16 hr/day from 16 to 38 weeks' gestation vs observation.
Outcomes: hypertensive syndrome, preterm delivery, abruptio, birthweight, IUGR, Apgar score, admission to SCBU, RDS, neonatal death, umbilical and cerebral artery PI.
Italy 2001Not clearly randomised. No clinical data reported. Available as congress abstract only.
Methods: not stated.
Participants: 24 women with PIH.
Intervention: isosorbide dinitrate sublingual every 6 hr (n = 12) vs nifedipine 20 mg daily (n = 12).
Outcomes: apoptosis in placental tissues.
Italy 2004

Intervention is not an antihypertensive: acupuncture + methyldopa vs. methyldopa. Protocol for a future RCT. Available as abstract only.

Methods: planned RCT

Participants: 60 women with chronic hypertension at 16-20 weeks' gestation.

Intervention: acupuncture + methyldopa vs. methyldopa.

Outcomes:need for antihypertensives.

Italy 2006Intervention is not an antihypertensive. Single-dose treatment.
Methods: double-blind, randomised, cross-over design.
Participants: 15 pregnant women at 30-34 weeks' gestation with mild/moderate PIH.
Intervention: L-Arginine 20 g /500 mL vs placebo infusion.
Outcomes: systolic and diastolic BP, fetal heart rate and fetal movements.
Italy 2008Intervention is not an antihypertensive. Available only as an abstract.
Methods: double-blind, randomised, placebo-controlled trial.
Participants: 20 pregnant women with pre-eclampsia.
Intervention: L-Arginine 1.66g 3 times a day vs placebo.
Outcomes: Serum L-arginine levels, adverse maternal and fetal outcomes.
Italy 2010Intervention is not an antihypertensive.
Methods: double-blind, randomised, placebo-controlled trial.
Participants: 80 pregnant women with mid chronic hypertension.
Intervention: L-Arginine vs placebo.
Outcomes: BP, need for additional antihypertensives, superimposed pre-eclampsia, adverse maternal and fetal outcomes.
Japan 1997Single-dose intervention. No clinical outcomes studied.
Methods: "...randomly allocated into two groups using sealed envelopes...".
Participants: 18 pregnant women with SBP = or > 140 mmHg and DBP = or > 90 mmHg, with or without proteinuria and oedema.
Intervention: isosorbide dinitrate patches (40 mg, only dose) and bed rest vs bed rest alone.
Outcomes: systolic and diastolic BP, uterine and umbilical Doppler velocimetry.
Kuwait 1995Not clearly randomised.
Methods: 'randomly allocated in sequence'. No further information.
Participants: 120 primigravid women > 26 weeks' gestation, with SBP 120-140 mmHg and DBP 95-105 mmHg persisting for 3 days.
Intervention: labetalol 100-300 mg x 3/day vs methyldopa 250-750 mg x 3/day.
Outcomes: maternal MAP, proteinuria (undefined), placental abruption, caesarean section, elective delivery, side-effects, 1 min Apgar score < 5, days on SCBU, birthweight (mean).
Mexico 2008

Intervention is not an antihypertensive. Available only as an abstract.

Methods: randomised, placebo-controlled trial.

Participants: 100 pregnant women with pre-eclampsia.

Intervention: L-Arginine vs placebo.

Outcomes: birthweight, SGA babies.

Pakistan 1994Intervention is an antiplatelet agent.
Methods: 'randomly divided into two groups'. No further information.
Participants: 200 women, 1 group with previous history of PIH (100 women) and other with mild essential hypertension or those developing BP 140/90 mmHg x 2 at least 15 days apart (100 women).
Intervention: aspirin 75 mg b.i.d. vs routine antihypertensive treatment with beta blockers or calcium channel blockers when DBP exceeded 100 mmHg.
Outcomes: development of PE. No other relevant outcomes reported.
Panama 2012

Severe hypertension. Ongoing trial.

Methods: randomised, open label trial.

Participants: 284 (estimated) women with at > 24 weeks' gestation with severe HT (SBP > 160 mmHg/DBP > 110 mmHg).

Intervention: 5 mg IV hydralazine every 15 min until BP controlled vs. IV labetalol ant increasing doses until BP controlled.

Outcomes: BP control.

Philippines 20003 days treatment. No relevant clinical outcomes studied. Available as abstract only.
Methods: randomised, double-blind, placebo-controlled trial.
Participants: 16 pre-eclamptics (no further details).
Intervention: nitrol patch 5 mg for 16 hr for 3 consecutive days vs the same regimen using a gauze only.
Outcomes: uterine and umbilical Doppler velocimetry.
Russia 1993Possibly not a RCT. Full text awaiting translation from Russian. Abstract only in English.
Participants: 92 women with slight and medium-severe hypertension at 24-39 weeks' gestation.
Interventions: venodilators, prazosin and cordafen are all mentioned. Not clear how the groups were constructed.
Singapore 1996More than 20% of women excluded, 6 women (22%) excluded because delivered in the week after trial entry.
Methods: 'by opening a sealed envelope'.
Participants: 27 women with singleton pregnancies, DBP 90 mmHg or above and proteinuria.
Interventions: isradipine (slow release) 5 mg a day vs methyldopa 750 mg a day.
Outcomes: MAP, side-effects, caesarean section, perinatal mortality, birthweight, admission to SCBU, Apgar score, maternal and fetal haemodynamics (by Doppler).
Singapore 1998No relevant clinical outcomes studied.
Methods: 'randomised', no further information.
Participants: 30 women with PE, DBP >/= 90 mmHg and proteinuria >/= 300 mg/24 hr.
Interventions: methyldopa 250-500 mg x 3/day vs isradipine 5-10 mg once/day.
Outcomes: haemostatic parameters only (thrombelastography, fibrinogen, antithrombin III, thrombin-antithrombin-complex, beta-thromboglobulin, plasminogen activators, plasminogen activators inhibitors, and plasminogen).
Slovakia 2002

Not a RCT.

Methods: cohort study.

Participants: women with hypertension during pregnancy.

Interventions: acebutolol vs other antihypertensives.

Outcomes: maternal and perinatal adverse events.

South Africa 1988Quasi-random design. Less than 7 days treatment, single-dose study. No clinical outcomes reported.
Methods: quasi-random design, using last digit of the hospital number.
Participants: 18 women in the last trimester of pregnancy with hypertension +/- proteinuria.
Interventions: nifedipine 5 mg vs placebo (single dose).
Outcomes: measures of uteroplacental blood flow.
South Africa 1990Included women with severe hypertension (DBP 100-120 mmHg).
Methods: 'randomly allocated', no further information.
Participants: 60 women at 28-36 weeks' gestation with mean 24 hr DBP 100-120 mmHg +/- proteinuria.
Intervention: indoramin 50 mg twice daily vs methyldopa 1 g twice daily vs placebo 1 tablet daily.
Outcomes: MAP, need for additional antihypertensive.
South Africa 1991aQuasi-random design. Single-dose intervention.
Methods: allocation 'by virtue of the last digit of their folder number'.
Participants: 19 women at > 28 weeks' gestation, singleton pregnancy and hypertension (defined as mean DBP >/= 90 mmHg).
Intervention: sublingual nifedipine 5 mg vs placebo (single dose).
Outcomes: DBP (mean), maternal and fetal heart rate, gestational age, side-effects.
South Africa 1997Most women did not have hypertension. Eligibility criteria DBP >/= 80 mmHg, before 20 weeks' gestation. Of 138 recruited women, less than half had DBP >/= 90 mmHg. Results for this group were not presented separately.
Methods: sequentially-numbered sealed boxes containing drug or placebo.
Participants: 138 women between 12-20 weeks' gestation with DBP 80-109 mmHg, without antihypertensive therapy.
Intervention: ketanserin 40-80 mg a day vs placebo.
Outcomes: severe HT, proteinuria, placental abruption, other drugs needed, perinatal deaths, SGA (< 10th centile), birthweight.
Spain 1988No clinical outcomes reported. Number of women in each group not reported. Available as abstract only.
Methods: 'double-blind, placebo-controlled trial', no further information.
Participants: 31 women with mild hypertension (BP 140-160/90 110 mmHg) despite bed rest in hospital.
Intervention: labetalol 200-600 mg a day vs placebo.
Outcomes: severe HT, need for additional antihypertensives, MAP, caesarean section, perinatal deaths, fetal distress.
Sri Lanka 1994Quasi-random design.
Methods: 'patients were alternately allocated'.
Participants: 126 women with PIH.
Interventions: nifedipine 30-90 mg/day vs methyldopa 750-2000 mg/day.
Outcomes: severe hypertension, gestation at delivery (mean), birthweight (mean).
Sweden 1992Comparison of 2 beta blockers.
Methods: 'randomly allocated' using 'double-blind dummy technique'. No further information.
Participants: 32 women admitted to hospital with PIH in the third trimester (BP >/= 140/90 mmHg x 2 at least 4 hr apart) and normotensive in the first trimester.
Intervention: atenolol 50 mg x 2/day vs pindolol 5 mg x 2/day, for at least 1 week.
Outcomes: side-effects, caesarean section, maternal haemodynamics, fetal haemodynamics, admission to SCBU, birthweight (mean), 5 min Apgar score.
Sweden 1993It is not clear from papers whether reported data represent only a subgroup of women.
Methods: not stated. Authors said 'allocated at random'.
Participants: 20 women at 26-37 weeks' gestation with 'persistent' DBP >/= 100 mmHg and proteinuria.
Intervention: labetalol 300-1000 mg/day orally (if necessary, IV 25 mg bolus followed by 25-65 mg/hr infusion), vs hydralazine 75-400 mg/day orally (if necessary, 1.5-6.0 mg/hr infusion).
Outcomes: severe hypertension, additional antihypertensive, caesarean section, neonatal death, birthweight (mean), gestation at delivery (mean), SGA (2 SD below mean), bradycardia, hypotension, hypoglycaemia, 5 min Apgar < 7, RDS, cord pH (< 7.20).
Uganda 1992

Status unknown.

Personal communication of a planned RCT of aspirin and methyldopa in moderate hypertension during pregnancy. No further data.

UK 1978Included women with severe hypertension.
Methods: 'randomly allocated'. No further information.
Participants: 74 women with singleton pregnancy with DBP >or = to 170/100 mm Hg x 2 at up to 36 weeks' gestation.
Intervention: labetalol 100 mg (max 1200 mg daily) vs methyldopa 250 mg (up to 4000 mg daily).
Outcomes: severe hypertension, proteinuria ('greater than trace'), additional antihypertensive therapy, changed drugs due to maternal side-effects, caesarean section, perinatal mortality, SGA infants (< 10th centile), intubated, umbilical cord pH.
UK 1991Less than 7 days treatment, single-dose study.
Methods: sequentially-numbered, sealed envelopes.
Participants: 30 women with singleton pregnancy and hypertension, defined as BP >/= 140/90 mmHg.
Intervention: 10 mg hydralazine IV vs or 100 mg labetalol IV, as single dose.
Outcomes: MAP, maternal and fetal heart rate, side-effects, umbilical artery PI.
USA 1957Not randomised. Although a group of women received placebo, results are presented together with a group of matched controls. Included women with severe hypertension.
Methods: not stated.
Participants: 106 pregnant women with chronic hypertension and 28 women with severe pre-eclampsia. In addition 671 women with chronic hypertension were included as controls.
Intervention: oral reserpine 0.25 to 3 mg/day (n = 80) vs placebo (n = 26). 28 women received IV reserpine.
Outcomes: status at birth, birthweight.
USA 1981Study included 63 women, but only 21 randomised. Outcomes not reported separately for randomised women.
Methods: 'randomly and blindly assigned'. No further information.
Participants: 21 women with BP 140/90 mmHg or above in a seated position or at rest, x 2, 6 or more hr apart.
Intervention: hydralazine 25 mg x 3/day vs methyldopa 250 mg x 3/day vs placebo x 3/day.
Outcomes: MAP, caesarean section, induction of labour, birthweight.
USA 1990a

Status unknown.

Personal communication of a planned RCT of oral hypotensive agents in early onset pre-eclampsia. No further data.

USA 1991Not a randomised trial. No clinical outcomes reported.
Methods: placebo group were matched as controls.
Participants: 16 women at 17-22 weeks' gestation.
Intervention: 10 mg sublingual nifedipine vs placebo.
Outcomes: S/D ratio of the uterine artery, maternal BP, maternal heart rate.
USA 2005

Intervention is not an antihypertensive. Ongoing trial.

Methods: randomised controlled trial. Sequentially numbered opaque envelopes.

Participants: 50 pregnant women with chronic hypertension.

Intervention: target blood pressure of 120-130/80-85 mmHg vs target blood pressure 140-150/90-100 mmHg.

Outcomes: BP, superimposed pre-eclampsia, worsening hypertension, HELLP syndrome, gestational age, birthweight, serious perinatal complications.

Venezuela 1985Not randomised. Included women with severe hypertension. Article in Spanish.
Methods: alternated allocation (personal communication).
Participants: 32 pregnant women at > 25 weeks' gestation with severe pre-eclampsia (defined as BP 160/110 or 140/90) and symptoms as headache, epigastric pain, blurred vision or hyperreflexia.
Intervention: labetalol 200-800 mg/day vs methyldopa 750-2000 mg/day.
Outcomes: maternal MAP, maternal pulse rate, gestational age at delivery, birthweight, 1 min Apgar, fetal and neonatal death.
Venezuela 1997Not a RCT. Matched controls. Article in Spanish.
Methods: controls were women treated with methyldopa in the same study period, with the same characteristics than the study group.
Participants: 20 women with PIH.
Intervention: labetalol 200 to 300 mg orally given every 12 hr vs methyldopa from 500 to 1500 mg/day
Outcomes: BP, severe hypertension, gestational age, induction of labour, caesarean section, birthweight.
Venezuela 2001No relevant clinical outcomes reported. Less than 7 days of treatment. Available as abstract only.
Methods: Authors said '...were randomly assigned to...'. No further information.
Participants: 30 pre-eclamptic. No further information.
Intervention: transdermal nitroglycerin (7 mg for 12 hr for 2 consecutive days) vs placebo.
Outcomes: umbilical S/D ratio, PI and RI by Doppler ultrasound.
Venezuela 2007

Mixed control group. Available only as an abstract.

Method: "...were randomly assigned to two study groups...".

Participants: 30 women with pre-eclampsia-eclampsia and gestational hypertension.

Intervention: Carvedilol vs traditional antihypertensive drugs (methyldopa and/or nifedipine).

Outcomes: BP control, additional antihypertensives, maternal and fetal complications, Apgar score, fetal heart rate.

Characteristics of studies awaiting assessment [ordered by study ID]

Canada 2008

MethodsRandomised, multicentre, blinded, placebo-controlled trial.
Participants

Treatment arm: all women who have been diagnosed with pre-eclampsia, are being followed clinically and who provide informed consent. For a diagnosis of pre-eclampsia a patient must meet all 3 criteria:

1. Systolic blood pressure greater than 140 mmHg or an increase of 30 mmHg from the participant’s baseline (with that increase present at 2 measurements taken 6 hours apart).

2. Diastolic blood pressure greater than 90 mmHg or an increase of 15 mmHg from the participant’s baseline (with that increase present at 2 measurements taken 6 hours apart).

3. Proteinuria greater than 0.3 g in 24 hour urine or 2+ on dipstick.

Interventions

Treatment arm: experimental intervention: daily dose of low-dose Isosorbide-5-mononitrate (ISMN) (30 mg) following diagnosis of pre-eclampsia after 24 weeks' gestation till delivery.

Control intervention: matching placebo containing lactose. Patients randomly assigned to either receive low dose ISMN (30 mg) as stated above or placebo.

OutcomesPrimary outcome: randomisation-to-delivery interval between ISMN/placebo groups, measured at delivery. Secondary outcomes: serial change in biochemical markers in treatment/no treatment groups, measured at routine obstetrical visits until delivery (generally every 2 weeks), incidence of any side-effects (major or minor), measured at routine obstetrical visits until delivery (generally every 2 weeks), neonatal outcomes (composite of neonatal morbidity), measured at delivery.
NotesOngoing trial. Author contacted.

Ancillary