Increasing Maternal Blood Pressure with Ephedrine Increases Uterine Artery Blood Flow Velocity during Uterine Contraction

Ten healthy women with uncomplicated pregnancies, at term, and in the active phase of labor were studied after giving informed consent. Our Institutional Review Board approved the protocol. Subjects with preexisting medical conditions or known fetal antepartum disease, such as cardiac or vascular disorders, maternal gestational or chronic arterial hypertension, preeclampsia, HELLP syndrome (hemolysis, elevated liver enzymes, and low platelet count), abnormal resistance index of uterine or umbilical arteries, body temperature greater than 38.5°C, or suspicion of chorioamnionitis, were excluded from the study. In addition, all women with fetuses suspected to be small for gestational age or with meconium appearance in the amniotic fluid were excluded. Adequate analgesia (epidural analgesia or intravenous meperidine) was provided before the study. At the time of inclusion, all women had a normal heart rate (< 110 beats/min) and stable arterial pressure (90 < systolic pressure < 130 mmHg, and diastolic pressure < 90 mmHg) with less than 10% variation during three consecutive measurements 10 min apart. All fetuses had reactive tracings with baseline heart rate between 120 and 150 beats/min.

When oxytocin was required, the infusion rate was kept constant throughout the protocol. Intrauterine pressure measurement using intrauterine catheter was indicated for obstetric management in two subjects.

Fetal heart rate and uterine contractions were monitored continuously during labor with an external transducer (Hewlett Packard 8040, Bobligen, Germany). To assess fetal well-being during the protocol, Doppler flow waveforms were acquired in the umbilical and middle cerebral arteries of the fetus to compute resistance indices. 12,13Maternal heart rate and systolic, diastolic, and mean arterial pressure (MAP) were measured every 5 min using an automated sphygmomanometer (Dinamap^®; Critikon Ltd., Tampa, FL). After administration of bolus intravenous ephedrine, these measurements were acquired every 3 min. Blood flow velocity waveforms from the maternal uterine artery were obtained using a 3.75-MHz sector transducer with color flow mapping and pulsed-wave Doppler (Toshiba SSA-340A, Tokyo, Japan) by a trained operator. To ensure accurate and reproducible velocity measurements, the angle between the vessel axis and the ultrasound beam was minimized and kept similar for all recordings. Both peak systolic (V^sys) and end-diastolic (V^dia) flow velocities were measured. Reproducibility of velocity measurements was assessed in a group of pregnant women at term but in the absence of labor. The mean intraobserver difference for velocity measurement was 3.5% for systole and 8% for diastole.

The resistance index (RI) was calculated according to the formula of Pourcelot 14:

Uteroplacental perfusion pressure during diastole (UPP^dia) was calculated as:

where DAP is diastolic arterial pressure, and IUP is intrauterine pressure (intrauterine perfusion pressure is the most accurate way to estimate myometrial activity). 15However, monitoring of IUP was accepted by our Institutional Review Board only when indicated for an obstetric reason.

All parameters were acquired successively at (1) baseline, (2) peak effect of ephedrine, and (3) after recovery, when MAP returned to control values. At each time, Doppler parameters were obtained both between uterine contractions and at the peak amplitude of uterine contractions. After baseline measurements, 6-mg boluses of ephedrine (ephedrine chlorohydrate; Aguettant, Lyon, France) were infused every 2 min until MAP increased by 20% above the baseline value, not exceeding a maximum total dose of 30 mg. If MAP did not return to baseline, ultrasound measurements were recorded after a maximum of 100 min, corresponding to a complete washout of ephedrine. 16 

Data were considered suitable for analysis when the Doppler envelope was well-defined and consistent over several consecutive cardiac cycles. Blood flow velocities were obtained as the averaged value of three consecutive cardiac cycles.

Data are presented as mean ± standard error of the mean (hemodynamic, demographic, and labor data) or median and quartiles (25th–75th percentile) (Doppler data). We compared the parameter values during ephedrine administration with baseline values using paired t  test for hemodynamic, demographic, and labor data, and Wilcoxon test for Doppler data because assumptions of normal distribution or equal variance were generally not fulfilled for this type of measurement. The same procedures were applied when comparing values obtained during the recovery period with baseline values. All tests were two sided. A P  value of less than 0.05 was considered significant.

Parturients were 30 ± 2 yr old. Five parturients were primiparae, and five were multiparae. The gestational age was 40 ± 1 weeks. Eight parturients underwent epidural analgesia, one received intravenous analgesia with meperidine, and one declined analgesia. Maternal MAP was unaltered by epidural analgesia (−1 ± 6%). The protocol was started at least 1 (3 ± 1.5) h after epidural or intravenous analgesia. Maternal MAP did not vary (+3.6 ± 3.5%) during the 30-min period immediately before data acquisition, confirming hemodynamic stability. No change in analgesic regimen was required during the protocol. The verbal pain scores at the three periods were 0.2 ± 0.2, 0.4 ± 0.2, and 2.2 ± 1.2, respectively (not significant). Cervical dilation increased from 4 ± 1 cm at baseline to 5 ± 1 cm after ephedrine injection (P = 0.03) and to 7 ± 1 cm at recovery (P = 0.01). The number of uterine contractions per 10 min remained unchanged (4.5 ± 0.4, 4.6 ± 0.4, and 4.7 ± 0.5, respectively). Eight women had vaginal delivery, five required instrumental extraction, and two underwent cesarean delivery because of failure to progress or cephalopelvic disproportion. Data could not be acquired at recovery in two parturients because of lack of time before delivery. No change was noticed on fetal umbilical and mean cerebral artery resistance indices during the protocol (table 1). Fetal heart rate tended to increase after ephedrine and remained increased at recovery, when compared with baseline (table 1). The 5-min Apgar index at birth was 9.8 ± 0.1, and immediate neonatal outcome was normal.

The dose of intravenous ephedrine required to achieve a 20% increase in MAP was 18 ± 2 mg. Recovery data were acquired 62 ± 7 min after ephedrine injection. Table 2summarizes maternal hemodynamic data. Maternal heart rate and systolic, mean, and diastolic arterial pressure increased after administration of intravenous ephedrine when compared with baseline values. The mean increase in MAP was 17 ± 4%. Values at recovery were not significantly different from those obtained at baseline. At baseline, V^sysand V^diawere both reduced during the peak amplitude of uterine contraction, with a major effect on V^diaand a marked increase in RI with respect to values obtained between uterine contraction (figs. 1 and 2). After ephedrine administration, V^sysand V^diaincreased and RI decreased, both during and between uterine contractions (figs. 1 and 2). However, the effect of ephedrine on uterine artery flow velocity and RI was more pronounced during uterine contraction. After administration of intravenous ephedrine, V^diaduring contraction was restored to 74% of the precontraction value. After ephedrine washout, V^diaand RI values during uterine contraction returned to baseline values (fig. 1).

Intrauterine pressure was measured in two parturients who underwent intrauterine catheter placement for obstetric management. Doppler parameters were acquired at a matched level of intrauterine pressure during uterine contraction (40 mmHg), before and after administration of bolus intravenous ephedrine. Diastolic arterial pressure increased from 68 to 94 mmHg and from 50 to 67 mmHg in each patient, respectively, after ephedrine administration. Concomitantly, uterine perfusion pressure in diastole during contractions increased from 28 to 54 mmHg and from 10 to 27 mmHg, respectively. Simultaneously, V^diaincreased from 0.12 to 0.40 m/s and from 0.25 to 0.41 m/s, while resistance indices decreased from 0.75 to 0.72 and from 0.83 to 0.71, respectively.

In the current study, we observed that intravenous ephedrine almost completely reversed the reduction in uterine blood flow velocities and the increase in resistance indices due to uterine contraction. Fetal hemodynamic parameters and progression of labor were not affected after ephedrine administration.

During normal pregnancy, trophoblast invasion converts the spiral arteries into wide flaccid low resistance vessels, resulting in high-flow uteroplacental circulation. 17,18End-diastolic flow velocity measured in the uterine artery outside the myometrium is commonly used as an index of uteroplacental blood flow. 9,18During pregnancy, uteroplacental flow velocity waveform changes progressively toward a pattern of high end-diastolic velocity, reflecting the increasing blood flow associated with a low resistance index. During labor, the placenta is mainly perfused between uterine contractions. During contractions, blood flow decreases and the resistance index increases dramatically because surrounding myometrial pressure compresses vessels. 8–10,17 

In the current study, we observed an 85% reduction in uterine end-diastolic velocity during uterine contraction versus  precontraction values, associated with a 58% increase in the resistance index. Ephedrine increased blood pressure and restored end-diastolic velocity to 74% of the precontraction value. Whether the increase in uterine artery flow velocity corresponds to an in increase in uteroplacental blood flow is open to debate. Indeed, if vessel diameter is reduced, an increased velocity can coexist with similar or reduced flow. In the current study, the resistance index was reduced, and this does not support the hypothesis of a reduction in uterine artery diameter. In addition, during pregnancy, the main uterine artery, where Doppler measurements were obtained, has been shown to be poorly constricted by ephedrine 5and is not squeezed during uterine contractions because of its extramyometrial location. Therefore, variations in flow velocity are likely to reflect true variations in blood flow. This restoration of flow could be secondary to the increase in pressure during diastole (uteroplacental driving pressure). This is supported by the data obtained in the two patients in whom intrauterine pressure was measured. The increase in uterine perfusion pressure was accompanied by an increase in end-diastolic flow velocity in the uterine artery. The effect of intravenous ephedrine on diastolic velocity and resistance index during uterine contractions may reflect the reopening of uteroplacental vessels compressed by the myometrium. Conversely, when the myometrium is relaxed, the effect of intravenous ephedrine is minor because uterine vessels are already patent.

Intravenous ephedrine improves systemic perfusion because it increases both mean arterial pressure (increased systemic arterial resistance) and cardiac output (increased heart rate and better venous return due to a decrease in venous capacitance). Interestingly, ephedrine has minor direct effects on the uterine artery during pregnancy compared with other α-agonist agents 5and does not constrict smaller uteroplacental spiral arteries because they are devoid of a muscular layer. 19Angiotensin II has also been shown to restore maternal blood pressure with limited increase in maternal heart rate and limited effects on uteroplacental circulation when administered before spinal anesthesia for cesarean delivery. 20Angiotensin II might yield similar increases in uterine blood flow velocities to ephedrine during uterine contractions, but this agent has not yet gained wide acceptance for the care of women in labor.

Similar findings might be obtained in patients with increased uteroplacental resistances (i.e. , preeclampsia, diabetes), although the increase in uterine blood flow would probably be less marked. However, during preeclampsia, trophoblastic invasion is not associated with the normal loss of the uterine arteriolar muscular layer, 21and potent vasoconstriction could occur with the use of high doses of ephedrine.

During normal pregnancy, provided that less than 30 mg ephedrine is administered, placental transfer of ephedrine is minimal and does not affect fetal hemodynamic parameters and neonatal outcome, as confirmed by the current findings. 4,22This is not true with atropine, which has been shown to diminish fetal oxygenation when used to correct fetal bradycardia. 23 

This was not a blinded study; however, each parturient was used as her own control, and the return to baseline values after ephedrine was assessed to rule out a time factor. Almost every woman underwent epidural analgesia, which may alter uteroplacental circulation for at least two reasons: (1) pain may affect uterine vascular resistance, and (2) the presence of sympathectomy may also alter uteroplacental hemodynamics. Therefore, the relevance of these findings cannot be generalized to women without epidural analgesia. This study was performed in women with a normal obstetric course and without any sign of fetal asphyxia. Further studies are required to assess whether ephedrine administration is therapeutically valuable in normotensive women in labor presenting with signs of fetal asphyxia.

In conclusion, intravenous bolus ephedrine increased baseline uteroplacental blood flow velocity and, most importantly, prevented its dramatic reduction during uterine contraction while resistance index decreased. These results suggest that the increase in uterine perfusion pressure during labor could in part restore uterine blood flow to the placenta during uterine contraction.

The authors thank Alexandre Mebazaa, M.D., Ph.D., Department of Anesthesiology and Intensive Care, Centre Hospitalier Universitaire Lariboisière, Paris, France, for his helpful comments.