Pre Eclampsia

Acta Obstet Gynecol Scand 2002; 81: 25–30 Copyright C Acta Obstet Gynecol Scand 2002

Printed in Denmark ¡ All rights reservedActa Obstetricia et

Gynecologica ScandinavicaISSN 0001-6349

ORIGINAL ARTICLE

Nifedipine or hydralazine as a first-line agent tocontrol hypertension in severe preeclampsiaBIBI SHAHNAZ AALI AND SAMIRA SHAHABI NEJAD

From Kerman Medical University, Niknafs Maternity Center, Kerman, Iran

Acta Obstet Gynecol Scand 2002; 81: 25–30. C Acta Obstet Gynecol Scand 2002

Background. Pre-eclampsia is one of the most serious and common complications of preg-nancy. Nifedipine, a calcium channel blocker, and the vasodilator hydralazine have both beenused as antihypertensive agents in this condition. The aim of this study was to determinewhich of these two agents is the most appropriate antihypertensive in the management ofsevere pre-eclampsia.Methods. One hundred and twenty-six pre-eclamptic patients with a gestational age of morethan 20 weeks were randomized to receive either 8 mg nifedipine sublingually or 5–10 mgintravenous hydralazine. Women with a history of heart failure and women receiving anti-hypertensive treatment during the course of the current pregnancy were excluded. For eachpatient the following data were recorded; the number of drug administrations, the time neededto control blood pressure, mean urinary output, the time interval between effective controland a new hypertensive crisis after each drug administration and relevant adverse effects inmother or fetus.Results. Effective control of blood pressure was achieved in both treatment arms. Data analy-sis indicated significantly fewer drug administrations in the nifedipine arm of the study. Thetime interval before a new hypertensive crisis following initial effective control of blood press-ure was significantly longer in the nifedipine group when compared with hydralazine. Effectivecontrol of blood pressure was achieved more rapidly in multiparous patients receiving nifedi-pine (pΩ0.026). Mean urinary output before and after delivery was greater in the nifedipinearm of the study. There were no significant differences between the two groups in othervariables. In addition, in neither group were there any serious adverse effects in mother orfetus.Conclusion. Nifedipine is safe and more effective than hydralazine in controlling blood press-ure in severe pre-eclampsia. It has the added advantage of being cheaper and more widelyavailable than the latter and is easily administered.

Key words: hydralazine; hypertension; nifedipine; severe pre-eclampsia

Submitted 23 July, 2001Accepted 8 August, 2001

Pre-eclampsia is one of the most serious compli-cations of pregnancy and causes considerable mat-ernal and perinatal mortality and morbidity (1, 2).

Dangerous complications such as placental ab-ruption, renal failure, and disseminated intravascu-lar coagulation can result. Maternal death probably

Abbreviations:BP: blood pressure; mmHg: millimeter mercury; T: t-studenttest; DF: degree of freedom; cc: cm3; p value: probability value;RR: relative risk; CI: confidence interval; h: hour.

C Acta Obstet Gynecol Scand 81 (2002)

occurs due to cerebral hemorrhage secondary to hy-pertension (1, 3, 4). An important part of any man-agement protocol is administration of antihyperten-sive agents where hypertension is severe, i.e.BPØ160/110 (5). The ideal drug for this aim shouldact quickly, without causing fetal or maternal sideeffects. Such a drug is not available currently. Hy-dralazine has been used as the standard treatmentfor this purpose (6), but there are problems associ-ated with its use. It can cause a precipitous reduc-tion in BP and thus adversely affect the mother as

26 B. S. Aali and S. Shahabi Nejad

well as the uteroplacental flow (7). It may also leadto maternal tachycardia and fluid retention andmust be administered parenterally (8).

In contrast, nifedipine, a calcium channel blockeragent, is theoretically an ideal candidate for use insevere pre-eclampsia (9). It promptly and effectivelyreduces elevated BP, whilst selectively increasing re-nal perfusion and diuresis (10). It is widely used tocontrol hypertensive crises unrelated to pregnancy(11, 12). There are no reported serious adverse ef-fects on mother or fetus as long as a stable BP canbe maintained (13, 14). The use of nifedipine in themanagement of preterm labor and hypertensive dis-orders of pregnancy has not been associated withadverse perinatal outcome (15). Barton et al. re-ported beneficial effects on urine output and meanarterial pressure in patients with severe pre-eclamp-sia treated with nifedipine during the immediatepuerperium (6). We therefore performed a ran-domized controlled study to compare the efficacyand safety profile of nifedipine with hydralazine inwomen with severe pre-eclampsia.

Materials and methods

We designed a single blind, randomized clinicaltrial for 126 patients with severe pre-eclampsia andgestational age of more than 20 weeks in NiknafsMaternity Center of Kerman, southeast Iran, fromApril to December 1999.

All patients had BPØ160/110 and met the cri-teria of severe pre-eclampsia according to theAmerican College of Obstetrics & Gynecology (1).Excluding criteria were a history of heart failurediagnosed by a cardiologist and any history oftreatment with an antihypertensive agent duringthe course of the current pregnancy.

After giving informed consent, the women wererandomized into two treatment arms. The block-randomized technique was used and each blockhad four cases. Women were allocated consecutive,numbered, opaque, sealed envelopes indicatingtheir medication. One group of patients receivednifedipine 8 mg (4 drops) sublingually and theother one 5–10 mg hydralazine intravenously. Theinitial dose of hydralazine was 5 mg with furtherdoses of 10 mg at intervals according to the proto-col recommended by the American College of Ob-stetrics & Gynecology (1). Patients in both groupsalso received IV magnesium sulphate (loading dose4 g, maintenance dose 1–2 g/hr, as determined bypatellar reflexes, respiratory rate and urinary out-put), which was stopped 24 hours after delivery.Induction of labor and vaginal or abdominal deliv-ery were performed as appropriate. Further moni-toring included measurements of BP every 5 min-utes in the supine position taken with a standard


sphygmomanometer on the right arm and usingphase 4 of Korotkoff sounds as the diastolic press-ure. The observer who measured BP was blind tothe type of treatment.

Effective control of BP was considered to beachieved by attaining a diastolic value between 90–100 mm Hg and not lower than 90 mm Hg re-corded every 5 minutes after drug administration.Repeated doses of either drug were given after 20minutes if effective BP control was not achieved.Urinary output was monitored hourly using an in-dwelling Foley catheter and continuous fetal heartrate monitoring was performed for all patients toidentify any signs of fetal compromise.

The study protocol was approved by the localethics committee.

Data collected on every patient included: age,parity, gestational age, history of known renal dis-ease, diabetes mellitus and chronic hypertension.The following parameters were considered as theendpoints of the study:

1) the number of drug administrations,2) the time needed to achieve effective BP control,

Table I. Demographic variables in two treatment arms

Nifedipine Hydralazine StatisticalCharacteristic Number (percent) Number (percent) test

Number of patients 65 (51.6) 61 (48.4)Age16–20 12 (18.500) 10 (16.4)21–25 15 (23.1) 17 (27.9)26–30 20 (30.8) 19 (31.1) Chi2*Ω0.0731–35 9 (13.8) 8 (13.2) pΩ0.7936–40 8 (12.3) 6 (9.8).40 1 (1.5) 1 (1.6)

Means (s.d.) 27.1 (6.4) 26.8 (6.1)tΩ0.26pΩ0.79

Gestational age (weeks)28–33 9 (13.8) 4 (6.6)34–37 24 (36.9) 22 (36) Chi2*Ω0.338–42 21 (32.4) 20 (32.8) pΩ0.25Apparently term% 11 (16.9) 15 (24.6)

Means (s.d.) 37 (3.3) 37.7 (8.3)tΩ0.51pΩ0.54

Gravidity‡

1-gravid 32 (49.2) 29 (47.5) Chi2Ω0.04Gravida 2π 33 (50.8) 32 (52.5) pΩ0.85

Underlying diseaseChronic hypertension 10 (15.4) 7 (11.5)

Chi2$Ω1.34Gestational/overt D.M. 3 (4.6) 2 (3.3)

pΩ0.25Renal disease 0 (0) 0 (0)

* Chi square for trend.% In this analysis the apparently term and 38–42 were merged.‡ The patients in the second row were paraΩ.1.$ All underlying diseases were merged to increase the power of test.

Nifedipine or hydralazine to control severe preeclampsia 27

3) the time interval between effective BP controland a new hypertensive crisis,

4) Mean of urinary output before and after de-livery,

5) any adverse effect from the drugs on motherand fetus.

Data were analyzed using the EPI6 and SPSSsoftware packages. First the variables were de-scribed; then the differences between the two treat-ment arms were assessed. Pearson chi-square test,chi-square for trend and t-student test were appliedto test possible differences between the two treat-ment arms. Probability values less than 0.05 wereconsidered significant.

Results

A total of 126 patients enrolled in the study. Sixty-five patients received nifedipine and 61 hydralazine.There were no significant differences between thetwo groups regarding maternal age, gestationalage, parity, overt or gestational diabetes, renal dis-ease and chronic hypertension (p.0.05) (Table I).The mean number of drug administrations areshown in Table II. Nine point three percent of thenifedipine patients required more than 3 doses incomparison with 19.7% in the hydralazine groupto achieve effective BP control. Fewer doses ofnifedipine were required to achieve effective BPcontrol when compared with hydralazine. The dif-ference was statistically significant (pΩ0.03). Pa-tients receiving nifedipine achieved effective BPcontrol more quickly than those receiving hydrala-zine (mean 9.6 minutes, s.d.Ω3.4 for nifedipine andmean 10.4 minutes, s.d.Ω3.8 for hydralazine), al-though they did not differ significantly as shownin Table II (pΩ0.24). Failure to achieve effectiveBP control within 20 minutes occurred in 6 and15 patients treated with nifedipine and hydralazine,respectively. However, comparison of the timeneeded for effective control of BP in multiparouspatients revealed significant differences (t-test, p Ω0.026) (Table IV).

The time interval before a new hypertensive cri-

Table III. Short-term adverse effect according to treatment group

Treatment Nifedipine Hydralazine Chi square (DF) PV RR* (95% CI)

No adverse effect 54 51 0.01 (1) 1.03Adverse effect 11 10 0.94 0.47–2.26Maternal tachycardia 4 (6.1) 8 (13.1)

Headache 7 (10.8) 2 (3.3)Hypotension 0 (0) 0 (0)

* Risk ratio of any kind of adverse effect in Nifedipine versus hydralazine group.


Table II. Efficacy parameters of nifedipine and hydralazine in controlling of BPin severe pre-eclamptic patients

Nifedipine Hydralazine StatisticalNumber (percent) Number (percent) test

Number of drug administrations1 15 (23) 9 (14.8)2 24 (36.9) 16 (26.2)3 20 (30.8) 24 (39.3) Chi2*Ω4.54 5 (7.7) 7 (11.5) pΩ0.025 0 (0) 5 (8.2)6 1 (1.6) 0 (0)

Mean (s.d.) 2.3 (1.01) 2.7 (1.11)tΩ2.1

pΩ0.03

Needed time to achieve effective BP control (minutes)$

5 35 (23.5) 34 (20.5)10 88 (59.1) 76 (45.8) Chi2*Ω6.2515 16 (10.7) 36 (21.7) pΩ0.01320 4 (2.7) 5 (3)F. 6 (4) 15 (9)

Mean (s.d.) 9.6 (3.4) 10.4 (3.8)t‡Ω1.18pΩ0.24

Time interval for a new hypertensive crisis following effective BP control(hours)1 7 (10.8) 9 (14.8)2 13 (20) 25 (41)3 7 (10.8) 4 (6.6)4 2 (3.1) 3 (4.9) Chi2*Ω4.415 3 (4.6) 0 (0) pΩ0.0366 4 (6.1) 1 (1.6)7 3 (4.6) 0 (0)No crises 26 (40) 19 (31.1)

Mean (s.d.) 3.1 (1.9) 2.1 (0.99)tΩ3.5

pΩ0.005

Mean of urinary output (cc)Before delivery 117 78After delivery 157 103.5

* Chi square for trend.$ The figures in the second and third columns stand for the number of doses.. Failure to achieve effective control of BP within 20 minutes.‡ t test for those who were controlled.

sis following initial stabilization was significantlylonger in the nifedipine group compared with thehydralazine arm of the study (mean 3.1 hours, s.d.1.9 for nifedipine versus mean 2.1 hours, s.d. 0.99for hydralazine). No hypertensive crisis occurred


Table IV. Efficacy parameters of nifiedinipe and hydralyzine according to treatment group in in primigravid and multiparous patients

Nifedipine Number Hydralazine NumberParity Endpoint (percent) (percent) Statistical test

Total doses for effective BP control1 10 (31.25) 5 (17.24) Chi2*Ω3.722 9 (28.13) 6 (20.69) pΩ0.05

Primigravida 3 11 (34.38) 12 (41.38).Ω4 2 (6.25) 6 (20.68)

Mean (s.d.) 2.2 (1.09) 2.86 (1.3) tΩ2.15 pΩ0.03

1 5 (15.15) 4 (12.5) Chi2*Ω1.442 15 (45.45) 10 (31.25) pΩ0.23

Multipara 3 9 (27.27) 12 (37.5).Ω4 4 (12.12) 6 (18.3)

Mean (s.d.) 2.5 (1.1) 2.8 (1.2) tΩ1.04 pΩ0.3

Time needed for effective BP control (minutes)5 19 (27.54) 20 (25) Chi2*Ω1.96 pΩ0.15

10 37 (53.62) 34 (42.5)15 8 (11.59) 15 (18.75)

Primigravida 20 3 (4.35) 2 (2.5)F. 2 (2.9) 9 (11.25)

Mean (s.d.) 9.6 (3.8) 9.9 (3.9) t‡Ω0.47 pΩ0.63

5 16 (20) 14 (16.28) Chi2*Ω4.6210 51 (63.75) 42 (48.84) pΩ0.0315 8 (10) 21 (24.42)

Multipara 20 1 (1.25) 3 (3.49)F. 4 (5) 6 (6.98)

Mean (s.d.) 9.6 (3.03) 10.8 (3.8) t‡Ω2.24 pΩ0.026

* Chi square for trend.‡ t test for those who were controlled.. Failure to achieve effective BP control within 20 minutes.

following initial stabilization in 26 versus 19 pa-tients in the nifedipine and hydralazine patients,respectively (Table II). The most common durationof BP control was 2 hours. Mean urinary outputwas greater in the nifedipine group in comparisonwith hydralazine (50% before and 34% after deliv-ery). Adverse effects occurred in 16.9% and 16.4%(RR 1.03, 95% CI 0.47–2.26) of the nifedipine andhydralazine groups, respectively. Tachycardia wasobserved more frequently with hydralazine (eightversus four) but headache was more common inthe nifedipine patients (seven versus two). No sud-den fall in blood pressure was observed (Table III).

Apgar scores of newborns were not significantlydifferent between the two groups. Thirteen neo-nates in the nifedipine group and 12 in the hydrala-zine arm had Apgar scores of 4–6 and all of themhad a gestational age lower than 30 weeks. No car-diotocographic abnormalities were recorded.

Discussion

Nifedipine has recently been used for the reductionof blood pressure in severe pre-eclampsia. A cal-cium channel blocker, it causes peripheral arterialvasodilatation (15). Extensive reviews on its


pharmacokinetics and pharmacodynamics indi-cate that nifedipine is associated with a 25% reduc-tion in systolic BP, diastolic BP and mean arterialBP in 98% of cases (16, 17). The study of Fenakelet al. indicated greater efficacy of nifedipine whencompared to hydralazine in achieving effective con-trol of BP in severe pre-eclampsia (9). Kwawu-kume & Ghosh in their study on 114 severe pre-eclamptic patients also found that nifedipine wasmore effective in controlling BP than hydralazine(18). In the present study both nifedipine and hy-dralazine could effectively control BP. In contrast,the study of Fenakel et al. demonstrated effectiveBP control in 98% versus 68% of patients in thenifedipine and hydralazine groups, respectively (9).The reason for this discrepancy may be in differentdoses and modes of drug administration. Herenifedipine was administered in a sublingual mode,while in the study of Fenakel et al. nifedipine wasgiven orally. In addition, hydralazine was switchedfrom the intravenous to oral route after the initialstabilization of patients and administration ofdrugs was performed over a longer period (7–15days). Although both nifedipine and hydralazinewere found to be quite effective in our study, fewerdoses of nifedipine were required to achieve BP

Nifedipine or hydralazine to control severe preeclampsia 29

control. Nifedipine was also shown to have alonger effect. The mean time interval for a newhypertensive crisis was significantly longer in thenifedipine group than hydralazine. Although themean values of the time needed for effective BPcontrol were comparable in the two groups (pΩ0.24), significantly less time was required toachieve effective BP control in mutiparous patientsreceiving nifedipine (pΩ0.026). Failure to achieveeffective BP control within 20 minutes occurredtwice as frequently with hydralazine as with nifedi-pine (9% versus 4%). Faster action of nifedpine re-flects its pharmacokinetics and pharmacodyn-amics. Some studies revealed that its onset of ac-tion is 3 minutes and reaches a maximum within60 minutes (16, 17). Walters and Redman reportedits onset of action as prompt and not longer than20 minutes in most cases. The antihypertensive ef-fects last at least 4 hours after giving 10 mg nifedi-pine orally (19).

For intravenous hydralazine the effect beginswithin 10–15 minutes, the maximum effect isreached within 60 minutes and lasts for 4–6 hours(17). Mean urinary output of the nifedipine pa-tients was greater than the hydralazine group. Thisfinding correlates with the positive renal effects de-scribed by Barton et al. (6), Fenakel et al. (9) andHall et al. (20). Evidence shows that nifedipine in-creases urine production by a natriuretic effect atthe level of the proximal tubule (10). None of ourpatients who received concomitant nifedipine andmagnesium sulfate developed hypotension. Thisfinding supports the results of other reports (6, 9).Barton concluded that the absence of this side ef-fect in their study was due to the oral mode ofadministration of nifedipine rather than sublingualroute (6) but we did not find such complication viasublingual route. The most frequent adverse effectswere headache and tachycardia in the nifedipineand hydralazine groups, respectively, and none ofthem was serious enough to discontinue treatment.Fenakel et al. also reported reflex tachycardia as acommon side effect of hydralazine whilst flushingand headache occurred more frequently in the nife-dipine group. They reported 11 cases of fetal dis-tress with hydralazine and only one similar casein the nifedipine group and related this finding toreduction of uteroplacental blood flow. We did notfind any abnormal cardiotocography in our studyand 25 Apgar scores of 4–7 seemed to be relatedto prematurity rather than drug adverse effect.Various investigators have shown that nifedipinedoes not compromise uteroplacental perfusion (21,22).

Our study demonstrates that compared with hy-dralazine, nifedipine is effective for longer and atfewer doses. It quickly achieves good BP control


in multiparous patients and the diuresis noted mayimpact on improvement in condition of these pa-tients. It is safe for mother and fetus. In additionit can be conveniently and easily administered viathe sublingual route. These properties make nifedi-pine a more appropriate treatment for hyperten-sion in severe pre-eclampsia when compared withhydralazine.

Acknowledgments

This study was funded by research grants from the KermanMedical University. Hereby the authors acknowledge the spon-sors for their cooperation. Invaluable advice provided by Dr.Catherine Holland and Dr. Ali Akbar Haghdoost is appreci-ated.

References

1. Cunningham FG, MacDonald PC, Gant NF, Leveno KJ,Gilstrap LC, Hankins GDV, Clark SL. William’s Obstetrics.20th ed. London: Prentice Hall International, 1997: 693–735.

2. Remuzzi G, Ruggeneti P. Prevention and treatment of preg-nancy associated hypertension, what we learned in the last10 years? Am J Kidney Dis 1991; 18/3: 285–305.

3. Nicoloso E, d’Ercole C, Cassel N, Azoulay P, Cravello L,Boubli L et al. Serious forms of arterial pregnancy relatedhypertension. Rev Fr Gynecol Obstet 1994; 89/10: 476–88.

4. Barron WM, Lindheimer MD. Medical disorders duringpregnancy, 2nd ed. St. Louis: Mosby, 1995: 1–30.

5. Belfort MA, Anthony J, Buccimazza A, Davey DA. Hemo-dynamic changes associated with intravenous infusion ofthe calcium antagonist verapamil in the treatment of severegestational proteinuric hypertension. Obstet Gynecol 1990;75: 970–4.

6. Barton JR, Hiett AK, Conover WB. The use of nifedipineduring the postpartum period in patients with severe preec-lampsia. Am J Obstet Gynecol 1990; 162/3: 788–92.

7. Vink GJ, Moodley J, Philpott RH. Effects of hydralazineon the fetus in the treatment of maternal hypertension. Ob-stet Gynecol 1980; 55: 19–22.

8. Naden RP, Redman CWG. Antihypertensive drugs in preg-nancy. Symposium on Medical Disorders During Preg-nancy. Clin Perinatol 1985; 12: 521–38.

9. Fenakel T, Fenakel G, Appelman Z, Lurie M, Katz Z, Sho-ham Z. Nifedipine in the treatment of severe preeclampsia.Obstet Gynecol 1991; 77/3: 331–7.

10. Krusell LR, Christensen CK, Lederballe PO. Acute natriu-retic effect of nifedipine in hypertensive patients and nor-motensive controls – a proximal effect? Eur J Clin Pharma-col 1987; 32: 121–6.

11. Erbel R, Meyer J, Brand G, Effert S. Emergency treatmentof hypertensive crisis with sublingual nifedipine. PostgradMed J Suppl 1983; 59: 134–6.

12. Ellrodt AG, Ault MJ, Murata GH. Efficacy and safety ofsublingual nifedipine in hypertensive emergencies. Am JMed 1985; 79: 19–25.

13. Childress CH, Katz VL. Nifedipine and its indications inobstetrics and gynecology. Obstet Gynecol 1994; 83: 616–19.

14. Seabe SJ, Moodley J, Becker P. Nifedipine in acute hyper-tension emergencies in pregnancy. Afr Med J 1989; 76: 248–50.

15. Ferlinz J. Diagnosis and treatment: drugs five years later:nifedipine in myocardial ischemia, systemic hypertension,


and other cardiovascular disorders. Ann Intern Med 1986;105: 714–29.

16. Barton JR, Ptevost RR, Wilson DA, Whybrew WD, SibaiBM. Nifedipine pharmacokinetics and pharmacodynamicsduring the immediate postpartum period in patients withpreeclampsia. Am J Obstet Gynecol 1991; 165/4: 951–4.

17. Walker JJ, Gant NF. Hypertension in pregnancy. London:Chapman & Hall Medical, 1997: 258–318.

18. Kwawukume EY, Ghosh TS. Oral nifedipine therapy in themanagement of severe preeclampsia. Int J Gynecol Obstet(Ireland) 1995; 49/3: 265–9.

19. Walters BNJ, Redman CWG. Treatment of severe preg-nancy associated hypertension with the calcium antagonistnifedipine. Br J Obstet Gynaecol 1984; 91: 330–6.

20. Hall DR, Odendaal HJ, Steyn DW, Smith M. Nifedipine orprazosin as a second agent to control early severe hyperten-sion in pregnancy: a randomised controlled trial. Br J Ob-stet Gynaecol 2000; 107: 759–65.


21. Lindow SW, Davies N, Davey DA, Smith JA. The effect ofsublingual nifedipine on uteroplacental flow in hypertensivepregnancy. Br J Obstet Gynaecol 1988; 95: 1279–81.

22. Puzey MS, Ackovic KL, Lindow SW, Gonin R. The effectof nifedipine on fetal umbilical artery Doppler waveformsin pregnancies complicated by hypertension. S Afr Med J1991; 79: 192–4.

Address for correspondence:

B. S. Aali, M.D.PO Box 76135-783KermanIrane-mail: shahnaz_aali/excite.com

Copyright of Acta Obstetricia et Gynecologica Scandinavica is the property of Wiley-Blackwell and its content

may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express

written permission. However, users may print, download, or email articles for individual use.

Documents

Pre Eclampsia