Review: Antihypertensive drugs improve maternal outcomes in mild chronic and pregnancy-induced hypertension
ACP J Club. 1999 Nov-Dec;131:61. doi:10.7326/ACPJC-1999-131-3-061
Magee LA, Ornstein MP, von Dadelszen P. Management of hypertension in pregnancy. BMJ. 1999 May 15;318:1332-6.
How effective are drug and nondrug therapies for chronic and pregnancy-induced hypertension for mothers and infants?
Studies were identified by searching MEDLINE (1966 to 1997), Hypertension in Pregnancy (to 1997), and the bibliographies and text of relevant studies and review articles.
English- and French-language randomized controlled trials were selected if they studied pregnant women with hypertension, evaluated either drug or non-drug (including bed rest and hospital admission) therapies, and assessed clinical outcomes.
Data were extracted in duplicate on study design, patient characteristics, definition of hypertension (chronic or pregnancy-induced), therapies, and outcomes (severe hypertension, additional anti-hypertensive treatment, pre-eclampsia, eclampsia, maternal mortality, cesarean section, placental abruption, change of drug because of side effects, perinatal mortality, premature delivery, small size for gestational age, admission to special care nurseries, low Apgar score, intra-ventricular hemorrhage, necrotizing enterocolitis, respiratory distress syndrome, neonatal bradycardia, hypotension, hypoglycemia, jaundice, and long-term neurodevelopment).
Although no changes in either maternal or perinatal mortality in mild chronic or pregnancy-induced hypertension were found, improvements in markers of morbidity were detected (Table). The choice of antihypertensive drug did not matter; some outcomes were better with drug rather than nondrug approaches. For women with severe hypertension remote from delivery, aggressive rather than expectant management increased the risk for total neonatal morbidity (odds ratio [OR] 5.2, 95% CI 1.5 to 18), admission to special care nurseries (OR 9.0, CI 2.7 to 30.0), necrotizing enterocolitis (OR 8.7, CI 1.4 to 52.0), and the respiratory distress syndrome (OR 3.3, CI 1.6 to 6.9) and decreased the risk for having a small-for-gestational-age infant (OR 0.31, CI 0.12 to 0.82). Intravenous hydralazine was associated with more maternal hypertension and cesarean sections and lower Apgar scores than other agents and more neonatal bradycardia than labetolol.
Drug treatment of mild chronic or pregnancy-induced hypertension improves maternal outcomes but does not clearly improve perinatal outcomes.
Source of funding: Physicians' Services Incorporated Foundation.
For correspondence: Dr. L.A. Magee, Department of Medicine, Mount Sinai Hospital, Toronto, Ontario M5G 1X5, Canada. FAX 416-586-8434.
Table. Drug treatment of hypertension in pregnancy (summary odds ratios)
|Type of hypertension||No. of studies||Outcomes||Odds ratio (95% CI)|
|Mild chronic||3||Severe hypertension||0.27 (0.14 to 0.53)|
|1||Hospitalization||0.23 (0.07 to 0.70)|
|Pregnancy induced||13||Severe hypertension||0.36 (0.26 to 0.49)|
|4||Hospitalization||0.45 (0.30 to 0.67)|
|12||Proteinuria at delivery||0.67 (0.51 to 0.89)|
|5||Respiratory distress||0.27 (0.13 to 0.54)|
The ambitious meta-analysis by Magee and colleagues tries to cover what the Canadian Hypertension Society Consensus Conference covered in 3 articles (1-3) (1 of which was previously reviewed in ACP Journal Club ). This review is primarily complementary but is improved by the use of Cochrane-style analysis. The methods, however, are not as clear, and it is difficult for the reader to determine which trials are included in the quantitative analysis. It is thus possible that important articles are missing and that the power (i.e., numbers of women in included trials) is inadequate to exclude clinically important differences.
Despite these criticisms, it is apparent that pregnant women with hypertension tolerate antihypertensive medications and generally benefit from them. However, the review fails to provide guidance on when to implement antihypertensive drugs; timing was an important issue in the previous review (3).
The benefits for the fetus or neonate, except when pre-eclampsia presents before 34 weeks of gestation, are equivocal but lean toward the benefit side of the ledger. Because reduction of maternal blood pressure does not explicitly benefit the fetus, the clinician must not assume that if the mother is fine, the baby is fine. Fetal surveillance remains clinically important. Similarly, expectant management entails intensive treatments and observations and is best relegated to experienced clinicians in tertiary care hospitals.
Robert Burrows, MD
Monash UniversityClayton, Victoria, Australia