Aims: The aim of this study was to evaluate whether prophylactic magnesium sulfate given in women with mild preeclampsia or gestational hypertension brings any significant change in umbilical artery and fetal middle cerebral artery pulsatility index.
Material and Methods: Forty-eight women with gestational age greater than 34 weeks with mild preeclampsia or gestational hypertension were randomly assigned to receive either magnesium sulfate or placebo. Pre- and post-magnesium sulfate pulsatility index of umbilical and fetal middle cerebral artery were compared by Mann–Whitney U-test between the groups. Within-group comparison was conducted using Wilcoxon's signed rank test.
Results: There was a significant reduction in the post-magnesium sulfate umbilical artery pulsatility index in the intervention group (median 0.88 [0.82−1.03]) when compared to the placebo group (median 1.00 [0.89–1.10]). Post-magnesium sulfate fetal middle cerebral artery pulsatility index in the intervention group (median 1.78 [1.63–1.98]) did not show a significant change compared to the placebo group (median 1.65 [1.42–1.91]). Within-group comparison showed change in both variables after treatment in the intervention group.
Conclusion: Magnesium sulfate produces a significant decrease in fetal umbilical artery and middle cerebral artery pulsatility index.
Magnesium sulfate reduces the risk of eclampsia in antepartum and postpartum women, irrespective of the severity of preeclampsia.1
Despite many years of use, the effects of magnesium sulfate on the unborn fetus are inconclusive. It protects against neonatal cerebral hemorrhage, decreases neuronal damage following brain injury and improves survival in extremely low-birthweight infants.2–6 In a recently published double-blind placebo controlled trial, magnesium sulfate reduced the incidence of cerebral palsy among newborns born before 31 weeks of gestation.7
On the contrary, there have been reports which suggest that administration of magnesium sulfate to the mother in the antenatal period leads to decreased fetal breathing and heart rate variability.8,9One study has also reported that at a neonatal serum magnesium level of 2–3 meq/L, neonatal respiratory and cardiovascular depression occur.10 This serum level is very similar to the magnesium level attained in the umbilical artery during prophylactic magnesium sulfate therapy in preeclampsia.11 An alarming observation of increased pediatric mortality among preterm neonates following antenatal magnesium sulfate therapy has also been reported.12 Some studies have found that magnesium sulfate does not have a major impact on disease progression among women with mild preeclampsia.13,14These findings have raised questions about the routine use of magnesium sulfate prophylaxis in mild preeclampsia. In the background of these conflicting evidences, it is essential to critically analyze the effect of prophylactic magnesium sulfate on the unborn fetus. We designed a randomized control trial to evaluate the effects of magnesium sulfate on umbilical artery waveform in cases of mild preeclampsia because abnormal umbilical artery waveform is a strong and independent predictor of fetal outcome.15
Doppler ultrasound provides us with a unique opportunity to observe the changes in uteroplacental and fetal circulation. Pulsatility index measured by Doppler ultrasound is the most reliable marker of the degree of resistance in a particular vessel. Increase in the pulsatility index of the umbilical artery is commonly associated with adverse perinatal outcome.16 The current study was designed to determine whether prophylactic magnesium sulfate given in women with mild preeclampsia or gestational hypertension brings any significant change in umbilical artery and/or fetal middle cerebral artery pulsatility index. Umbilical artery pulsatility index was considered as the primary and middle cerebral artery pulsatility index was considered the secondary outcome variable.
Material and Methods
The study was conducted in a postgraduate medical institute of West Bengal, India, serving chiefly a rural population of low socioeconomic strata from January to June 2009. Patients were selected from those attending the antenatal clinic of the institute. All pregnant women with fetal gestational age greater than 34 weeks (confirmed by last menstrual period and first trimester ultrasonography) and mild preeclampsia or gestational hypertension were evaluated for inclusion in the study. For the present study, mild preeclampsia or gestational hypertension was defined as systolic blood pressure ≥140 and <160 mmHg and/or diastolic blood pressure ≥90 and <110 mmHg, with or without proteinuria. Women with multiple pregnancy, diagnosed congenital fetal anomaly, abnormal fetal heart rate on auscultation, gross intrauterine growth restriction, amniotic fluid index – <5 and history of medical disorders, such as essential hypertension, chronic renal disorder, seizure disorder, overt diabetes and hypersensitivity to magnesium were excluded from the study. The selected patients after routine examination in the antenatal clinic were admitted for further evaluation. Subsequently they underwent ultrasound examination on the same day. During the period of admission they were allowed to conduct their daily normal activities and consume a normal diet. No antihypertensive medications were prescribed unless blood pressure exceeded the defined range of mild gestational hypertension. Blood pressure was monitored every 4 h while the patient was under admission.
Fetal growth parameters, liquor volume (amniotic fluid index) and Doppler evaluation of placental and fetal vessels were performed with the GE Logiq 400 Pro Ultrasound System (GE Healthcare, Munich, Germany) using a 5-MHz sector transducer with spatial peak temporal average intensities below 50 mW/cm2 and a high-pass filter at 50–100 Hz. Measurements were obtained from the umbilical artery at the midsection of umbilical cord and the distal (straight) portion of middle cerebral artery. Umbilical artery pulsatility index and fetal middle cerebral artery pulsatility index were measured during a period of fetal apnea. After this baseline evaluation, the selected patients were randomly allocated into two different groups by sealed envelopes containing computer-generated random numbers. One group was designated as the placebo group while the other was the intervention group. Both the patient and the examining radiologist were blinded regarding group allocation. Treatment packs containing identical ampoules of either normal saline or magnesium sulfate in equal number and volume were prepared. The intervention group received prophylactic magnesium sulfate given at doses of 4 g (8 mL) i.v. diluted in 12 mL normal saline given slowly over 10–15 min and 10 g (20 mL) i.m. (5 g [10 mL]) in each buttock. This dose was followed by 5 g (10 mL) i.m. magnesium sulfate every 4 h for the next 24 h. The placebo group received normal saline in the same amount. The patient's vital parameters (pulse, blood pressure, urine output and neurological evaluation) were recorded. Occurrences of any possible side-effects of drug administration, such as flushing, nausea, respiratory depression, hypotension and syncope were recorded. Patient's complaints regarding the injection site, such as pain, burning, phlebitis and any local abscess formation were also noted After completion of treatment Doppler parameters were repeated for each patient. Post-magnesium sulfate Doppler evaluation was performed in all cases within 4 h of administration of the last i.m. dose of magnesium sulfate. The examining radiologist was masked regarding each patient's clinical profile, any medications given and any side-effects. Patients were thoroughly counseled and informed consent was given by all patients prior to trial entry. The study was granted permission by the ethical committee of the institute.
General demographic parameters of the two groups were compared using Mann–Whitney U-test. Pre- and post-magnesium sulfate pulsatility index of umbilical and fetal middle cerebral artery between the groups were compared by Mann–Whitney U-test. Within-group comparison (both placebo and intervention) of these two variables was conducted using Wilcoxon's matched pairs signed rank test. Statistical significance was set at P < 0.05.
It was estimated that 24 subjects in each group were required to detect a difference of 0.1 in the mean umbilical artery pulsatility index between groups with 80% power and 5% probability of type I error. This calculation assumed a standard deviation of 0.12 for the primary outcome variable.
Figure 1 is a flow chart of the study design. Five hundred and twelve women were assessed for eligibility during the study period. Eighty-eight fulfilled the criteria for trial entry. Twenty-two did not give consent for participation in the study after counseling. Sixty-six were randomized into two groups (thirty-three in each group) and started on the decided protocol of treatment. Five women in the intervention arm and six in the placebo arm delivered within 24 h of trial entry and could not complete treatment protocol. In three women in the intervention arm and two in the placebo arm post-treatment repeat Doppler parameter could not be arranged. In one woman in each group treatment was stopped early because they could not tolerate the intramuscular dose because of severe pain. Forty-eight women with 24 women in each arm completed the study. ‘On treatment’ analysis was done for the present study (i.e. trial participants who fulfilled the protocol in terms of eligibility, intervention and outcome assessments were only analyzed).
Women in both groups were comparable in terms of general demographic parameters (Table 1). Table 2 shows the median pre-treatment and post-treatment umbilical artery and fetal middle cerebral artery pulsatility index in the placebo and intervention groups. Pre-treatment values of both the variables were comparable. Statistical analysis showed a significant reduction in post-magnesium sulfate umbilical artery pulsatility index in the intervention group (median value with interquartile range = 0.88 [0.82–1.03]) compared to the placebo group (median value with interquartile range = 1.00 [0.89–1.1]). Post-magnesium sulfate fetal middle cerebral artery pulsatility index in the intervention group (median value with interquartile range = 1.78 [1.63–1.98]) did not show a significant change compared to the placebo group (median value with interquartile range = 1.65 [1.42–1.91]).
Table 1. General demographic parameters in the placebo and intervention groups
(n = 24)
(n = 24)
BMI, body mass index; MABP, mean arterial blood pressure. Values expressed in mean and standard deviation or median and range as applicable.
Parity, median (range)
Gestational age (weeks), median (range)
No. women with proteinuria
Expected fetal weight, median (range)
Amniotic fluid index, median (range)
Table 2. Changes observed in primary and secondary outcome variables (n = 24) in the placebo and intervention groups
Within-group analysis showed that in the placebo group the pre-treatment umbilical artery pulsatility index values (median with interquartile range = 1.03 [0.92–1.14]) and post-treatment values (median with interquartile range = 1.00 [0.89–1.1]) were not significantly different. Middle cerebral artery pulsatility index also did not show any significant change before and after treatment (median with interquartile range = 1.76 [1.58–1.90]) and 1.65 [1.42–1.91]), respectively. In both the variables a decrease was noted in the intervention group. Post-magnesium sulfate umbilical artery pulsatility index (median and interquartile range = 0.88 [0.82–1.03]) and middle cerebral artery pulsatility index (median and interquartile range = 1.78 [1.63–1.98]) dropped significantly in comparison to pre-magnesium sulfate umbilical artery pulsatility index (median and interquartile range = 1.02 [0.93–1.16]) and middle cerebral artery pulsatility index (median and interquartile range = 1.88 [1.70–1.98]), respectively. Figure 2a,b shows the paired pre- and post-magnesium sulfate umbilical and middle cerebral artery pulsatility index, respectively, of the placebo group. Figure 2c,d shows the paired, pre- and post-magnesium sulfate umbilical and middle cerebral artery pulsatility index, respectively, of the intervention group.
Fourteen patients in the intervention arm had flushing and nausea during administration of loading dose compared to six in the placebo arm. Twelve of these 14 patients in the intervention arm had hypotension and three had syncope during administration of the loading dose. None of the patients in the control arm had hypotension or syncope attacks. The rate of occurrence of injection site problems was similar between the groups with four patients in the intervention arm and three patients in the control arm complaining of pain and phlebitis at the site of administration.
The results of the present study suggests that prophylactic magnesium sulfate administered to mothers antenatally with mild preeclampsia leads to a reduction in pulsatility index of the umbilical artery and improves placental blood flow. Both umbilical and middle cerebral artery pulsatility indices were reduced when within-group comparisons were made. This decrease was secondary to increases in both diastolic and time averaged velocity (TAMX) in umbilical artery blood flow.
In women with hypertensive disorder of pregnancy, placental perfusion is compromised. Decrease in placental perfusion is due to an imbalance in prostacyclin and thromboxane A2 ratio, altered production of nitric oxide and thickening of the vessel wall due to atheromatous deposition, which leads to decreased luminal flow.17,18 There is also a reduction in the number of terminal villous arteries where placental circulatory readjustments occur.19 The current study shows that in women with mild preeclampsia or gestational hypertension, the pulsatility indices of the umbilical and fetal middle cerebral arteries remains within the normal range for ≥34 weeks gestation (a total of five cases, taking both the control and intervention group together show the umbilical artery pulsatility index to be above the 95th percentile for that gestational age).20–22This differs from results of a previous study which showed that the incidence of abnormal umbilical artery pulsatility index in both severe and mild preeclampsia is higher than in normotensive pregnant women.23 The lack of significant change might be due to the extensive vascular bed of placenta and changes in intraplacental villous space volume which dampens any circulatory change in umbilical vessels.19
The mechanism of action of magnesium sulfate in preeclampsia and eclampsia is not clear. Possible mechanisms are in the placental tissue; first, magnesium cleaves plasminogen activator inhibitor type 2 and therefore may reduce placental thrombosis. Second, preeclampsia is a disease of endothelial cell dysfunction. Magnesium has an in vitro protective effect on some endothelial cells. Last, preeclampsia is associated with an increase in inflammatory response demonstrated by leukocyte activation. Magnesium may have some anti-inflammatory properties because it reduces leukocyte activation.24
Most of the previous studies demonstrating the effect of various pharmacological agents used in hypertensive disorders of pregnancy on umbilical arterial blood flow have been in vitro. There have been few in vivo studies that have specifically studied the effect of magnesium sulfate on umbilical artery blood flow.23 These studies (both in vitro and in vivo) were not well designed (none were randomized controlled trials), neither did they select the proper marker of umbilical artery blood flow resistance (systolic/diastolic ratio was taken as primary variable). Most of the studies have observed the change in umbilical artery blood flow immediately after the loading dose, while none have studied the changes after completion of the full dose of magnesium sulfate.23 The present study is the first randomized placebo controlled trial comparing umbilical artery and fetal middle cerebral artery pulsatility index following full dose of prophylactic magnesium sulfate in preeclampsia or gestational hypertension.
Studies evaluating the effect of other pharmacological agents on umbilical artery blood flow suggest that nifedipine and nicardipine chiefly relax the umbilical vein, with no significant effect on the umbilical artery.25 Hydralazine and labetalol also do not have a significant effect on umbilical artery pulsatility index.26 In fact, adrenergic receptor blockers are unlikely to affect placental vessels as they are devoid of adrenergic nerves and vasa vasorum. Umbilical resistance has been shown to be primarily controlled by autacoids substances (amines, eicosanoids and peptides), hormones and oxygen tension.25 Magnesium sulfate given in in vitro preparations relaxed chiefly the umbilical artery previously treated with thromboxane A2 or its analogues. This is because magnesium ion is known to interfere with membrane calcium channels and it also increases endogenous prostacyclin production, reduces endothelin-1 levels and inhibits platelet aggregation.6,25 Short-term exposure to magnesium sulfate does not show any significant change in Doppler parameters of umbilical vessels in patients treated for preterm labor or preeclampsia.23 This was explained to be due to lesser sensitivity (only 10–15%) of umbilical vessels, secondary to pathological changes in the vessel wall, making them less elastic.18,25 Results of the present study differ from previous studies and suggest that full dose of prophylactic magnesium sulfate does improve both umbilical and fetal middle cerebral artery blood flow. A demonstrable change in umbilical artery pulsatility index value does not occur unless there is at least 60% loss or gain of placental blood flow.19 This suggests that following magnesium sulfate the increase in placental blood flow is significant clinically.
Between-group comparison of middle cerebral artery pulsatility index did not show a significant change, although within-group comparison showed a significant reduction in this variable. The data suggest that in the middle cerebral artery there was a reduction in peak systolic blood flow. Diastolic blood flow and time-averaged velocity remains unchanged. Twickler et al. in their study on effect of magnesium sulfate on fetal cerebral blood flow found no change in middle cerebral artery blood flow; this they attributed to the complex feedback systems of vascular smooth muscle and endothelial cells. This study was performed among women at risk of preterm delivery with a gestational age range of 24–31 weeks, which was less than the gestational age of the present study population which had a range of 35–37 weeks.
Swartz et al. in their study of healthy women with a fetus ≥27 weeks of gestation found a decrease in peak systolic velocity of the fetal middle cerebral artery following fetal heart rate acceleration among which according to them is the result of a protective mechanism which prevent cerebral vascular over-distension or, the result of a decreased fetal myocardial recovery time associated with increased fetal heart rate, resulting in decreased peak systolic velocities in middle cerebral artery.27 In the present study our observation was among women with mild preeclampsia (mean arterial blood pressure = 113.7 mmHg), in response to the loading intravenous dose of magnesium sulfate there was transient bradycardia in the fetal heart rate, which subsequently settled and by the time of the next Doppler evaluation fetal heart rate returned to baseline level. This alteration in fetal heart rate most likely did not affect blood flow at the placental level or at fetal brain. The decrease in pulsatility index of the middle cerebral artery in our study was probably a result of increased perfusion at the placental level.
Flushing, nausea, hypotension and syncope attacks, which are known complications of magnesium sulfate therapy were more common in the intervention arm.2 Most of these problems occurred following the loading dose and effects were short-lasting. The rate of injection site problems was similar between the groups.
In the present study we did not include the fetal outcome parameters following delivery, so the ultimate clinical efficacy of this intervention on the fetus cannot be commented upon. Umbilical artery pulsatility index values are not always reflective of fetal outcome and fetal hemodynamic compromise may occur in the presence of normal umbilical artery pulsatility index values. This reflects a different mechanism of placental insufficiency (e.g. senescent trophoblastic function).28 A small reduction in placental blood flow which does not produce a significant change in umbilical artery pulsatility index may adversely affect a relatively well-grown fetus as well.25 The timing of post-treatment Doppler evaluation was not the same in all cases throughout the study; therefore, the changes in post-treatment Doppler values might have been less uniform.
From the present study it can be concluded that the currently recommended dose of prophylactic magnesium sulfate produces a significant increase in fetal umbilical artery and middle cerebral artery blood flow. Further studies need to be conducted to evaluate the effect of magnesium sulfate on other systems of the unborn fetus, especially the neurological system. Although beneficial to the placental and cerebral blood flow, the true overall impact of antenatally administered magnesium sulfate on the fetal outcome is yet to be settled.
The authors are thankful to the Principal of Bankura Sammilani Medical College, who gave the necessary permission to conduct the study.
None of the authors have any financial relationships with companies that may have a financial interest in the information contained in the manuscript. None of the authors have any potential conflict of interest.