Widespread use of spinal anesthesia for cesarean delivery has been accompanied by a reduction in maternal mortality.1 However, spinal anesthesia is associated with hypotension, and this is more common and profound in the pregnant population, with an incidence in excess of 80% without prophylactic management.2 The resulting hypotension can cause nausea and vomiting, cardiovascular collapse, and loss of consciousness in the mother, as well as acidosis in the fetus.3 Both spinal anesthesia and maternal physiological changes contribute to the hypotension. A reduction in systemic vascular resistance (SVR) as a consequence of sympathetic blockade, more extensive neuroblockade because of a contracted subarachnoid space, and aortocaval compression have all been implicated as mechanisms of the hypotension.4
Many strategies have been described to prevent and treat hypotension in the obstetric population. There has been growing evidence to support the use of coloading with IV fluids in combination with the use of vasopressors.5,6 The ideal vasopressor would maintain maternal cardiovascular stability and prevent nausea and vomiting, but have little adverse effect on uteroplacental perfusion, therefore with no resulting compromise to the fetus. The ideal vasopressor has been the subject of much controversy and debate, but it is now widely accepted that the vasopressor of choice in the parturient is phenylephrine.7–10 Large doses of vasopressor are often required during cesarean delivery after spinal anesthesia because of increased baroreceptor sensitivity.11 There have been concerns that large doses of phenylephrine, although maintaining maternal systolic blood pressure (SBP), may cause reflex bradycardia and consequently a reduction in cardiac output (CO).a
Although previous work has shown that maternal CO correlates closely with uteroplacental blood flow,12 most of the work done to date examining the cardiovascular effects of phenylephrine has concentrated on its effect on maternal heart rate (HR) and SBP.13 It is a long-held belief that measurements of SBP can be used as a surrogate for maternal CO in predicting uterine blood flow, in the absence of a reliable noninvasive method of assessing maternal CO.14 However, with the advent of newer noninvasive techniques, maternal CO can now be measured reliably and accurately. The suprasternal measurement of aortic blood flow is a noninvasive technique for assessing the CO. It measures the stroke distance, in the aortic arch, providing a linear measure of CO. It has been validated against volumetric measures derived from thermodilution15 and has been used successfully in the pregnant and nonpregnant populations.16,17
This double-blind, randomized, controlled study was designed to investigate the dose-dependent effects of 3 infusion doses of phenylephrine on maternal cardiovascular stability in women undergoing elective cesarean delivery under combined spinal-epidural anesthesia. The primary outcome measure was change in CO. Secondary outcomes included effects on HR, SBP, and measures of fetal well-being.
After ethics committee (Royal Free Hampstead NHS Trust, London, UK) and MHRA (Medicines and Healthcare Regulatory Authority) approvals (EudraCT number 2006-004858-25), the study was conducted over a 7-month period from May until November 2007. Written informed consent was obtained from 75 healthy term parturients undergoing elective cesarean delivery under combined spinal-epidural anesthesia. Exclusion criteria included cardiovascular disease or cardiac medication, pregnancy-related hypertensive disease, height <150 cm or >180 cm, or weight <50 kg or >100 kg.
Randomization was performed using a computer-generated random number table. Parturients were randomly assigned to 1 of 3 groups of 3 infusion regimens of phenylephrine. Group assignments were sealed within opaque envelopes. All phenylephrine infusions were infused at the same rate, but the concentration differed among groups, such that the parturients received 25 μg/min (group 25), 50 μg/min (group 50), or 100 μg/min (group 100).
Three anesthesiologists were involved in the study. One anesthesiologist, unconnected with the clinical care or data collection, prepared the phenylephrine infusion according to the randomization group and instructions provided in the sealed envelope opened just before drug preparation. A second anesthesiologist monitored and recorded the HR and SBP every minute from the initiation of anesthesia (intrathecal injection) until delivery of the fetus and controlled the phenylephrine infusion. A third anesthesiologist performed all of the Doppler measurements of CO, as well as the combined spinal-epidural technique. The patient and second and third anesthesiologists were blinded to group assignment.
Baseline measurements (HR, SBP, and oxygen saturation) were obtained before the initiation of the fluid preload in the supine position with the bed tilted 15 degrees to the left. A handheld suprasternal ultrasound device (SupraQ® cardiac function monitor; Deltex Medical Ltd., Chichester, UK) was used to measure CO, stroke volume (SV), and other Doppler variables. Baseline HR and SBP were taken as an average of 3 readings, and 80% value of the baseline SBP was calculated.
Immediately after the measurement of the baseline hemodynamic variables a fluid preload of 500 mL Hartmann solution was infused over a 5-minute period through a wide-bore peripheral IV cannula with the aid of a simple pressurized infusion system. No further fluid was administered until after delivery of the fetus. On completion of the fluid preload, SBP, HR, CO, and other Doppler variables were measured, followed by initiation of anesthesia. HR and SBP were then recorded and acted upon every minute from the start of spinal injection until delivery of the fetus. CO, SV, and other Doppler measurements were performed at 5-minute intervals for a period of 20 minutes before the start of surgery. The incision was delayed for purposes of the study until the 20-minute measurement interval was completed.
Combined spinal-epidural anesthesia was initiated in the sitting position at the L3-4 or L4-5 interspace with 0.5% hyperbaric bupivacaine 11 mg and fentanyl 15 μg. Block height to cold and loss of touch were recorded at 5-minute intervals using ethyl chloride spray. If a block height of T5 to touch sensation was not achieved by 20 minutes, 5-mL boluses of levobupivacaine 0.5% were given via the epidural catheter to extend neuroblockade.
The phenylephrine infusion was administered at a rate of 120 mL/h from the time of the intrathecal injection until delivery of the fetus if SBP was at or below the baseline SBP. The infusion was stopped if the SBP went above the baseline. Hypotension, defined as an SBP <80% of baseline SBP for 2 consecutive readings, despite the phenylephrine infusion, was treated with a bolus of phenylephrine 100 μg. If no improvement was seen after a further 2 consecutive readings, a bolus of ephedrine 6 mg was administered. Bradycardia (HR <50 bpm) for 2 consecutive readings was treated by stopping the phenylephrine infusion if the SBP was at or above the baseline, but if the SBP was below the baseline, the phenylephrine infusion was continued and a bolus of glycopyrrolate 200 μg was administered. The presence of nausea and vomiting (none, mild, moderate, or severe) was assessed at 5-minute intervals until 20 minutes after spinal injection.
Obstetric data collected included time interval from the intrathecal injection to the start of surgery, uterine incision to delivery time, Apgar scores at 1 and 5 minutes, and umbilical arterial and venous blood gases obtained from a double-clamped segment of umbilical cord.
CO was measured using a suprasternal ultrasound device, the SupraQ cardiac function monitor (Deltex Medical Ltd.). All measurements were taken from the aortic arch and performed by a single operator trained over a 3-month period in suprasternal ultrasound techniques, so as to achieve reproducibility and reliability. All CO readings used for statistical analysis represented the mean of 3 readings taken in rapid succession at each measurement time. Apart from CO and SV, we also measured stroke distance, minute distance, corrected flow time, and peak velocity.
Data are presented as mean and frequency. Group effects were analyzed using 1-way analysis of variance for numerical data, linear trend for dose, and expanded Fisher exact tests for categorical data. Within- and among-group comparisons of hemodynamic variables from baseline to predefined time points were performed using repeated-measures analysis of variance, analysis of covariance (using baseline hemodynamic variables as covariates), and Tukey-Kramer multiple comparison tests. True group effects were tested at 4 time points, between 5 and 20 minutes after the spinal injection.
An a priori sample size analysis showed that a minimum of 23 patients in each of 3 groups would give 80% power to detect a 20% difference in CO (assuming a coefficient of variation of 20%), at an overall 2-sided P < 0.05 (threshold P < 0.017 with the Bonferroni adjustment for multiple comparisons). This sample size would also provide at least 90% power to detect differences of 20% in SBP and 25% in HR. Additionally, because the aim of the study was to provide similar SBP control in the groups, this was examined by an equivalence analysis approach, which was defined as significant (P < 0.05 after adjustment for multiple comparisons) if the 90% confidence interval estimates of the ratios were contained within the conventional 0.80 to 1.25 margin. Analyses were performed using the following software: Excel 2000 (Microsoft Corp., Redmond, WA), Number Cruncher Statistical Systems 2004 (NCSS Inc., Kaysville, UT), and Prism 5.0 (GraphPad Inc., San Diego, CA).
Data analysis was performed on 75 patients (Fig. 1). Details of maternal characteristics are summarized in Table 1. The 3 groups were similar with respect to maternal age, weight, height, and body mass index.
Baseline cardiovascular variables are detailed in Table 2. To control for differences in baseline CO values, any differences among groups after spinal anesthesia were compared using analysis of covariance. CO decreased significantly (P < 0.001) with time within each group (Fig. 2). Compared with baseline values, the reductions in CO at 20 minutes were 0.3, 0.4, and 1.1 L/min in groups 25, 50, and 100, respectively. There were significant between-group differences (P = 0.03) in CO that were concentration dependent (linear trend; P = 0.007). The maximum percentage reductions in CO from baseline values were 7.8%, 15.2%, and 22% in groups 25, 50, and 100, respectively. The decrease from baseline CO was highest in group 25 and lowest in group 100 at all time points.
There was a significant between-group difference in HR (P = 0.02), and HR decreased significantly with time in all groups (linear trend; P < 0.001) (Fig. 3). Compared with baseline values, the reductions in HR at 20 minutes were 8, 12, and 22 bpm in groups 25, 50, and 100, respectively. There were significant concentration-dependent reductions in HR (linear trend; P < 0.007). At all time points from the start of the phenylephrine infusion, the HR was the most rapid in group 25 and slowest in group 100. No significant differences were found in the number of glycopyrrolate boluses (Table 3).
SV remained stable with time within each group (Table 4). Other variables obtained from the Doppler readings, corrected flow time, mean acceleration, peak velocity, did not demonstrate any significant differences among the groups (data not shown).
There were small but significant differences in SBP among groups (P = 0.04) (Fig. 4), which were concentration dependent (linear trend; P = 0.01). Our results suggest that the highest concentration group had a more “stable” SBP compared with the lower concentrations, which tended to drift down with time. Although statistically significant, the difference among groups at any time point was <15 mm Hg and therefore less than the clinically significant minimum difference of 20% as demanded by the protocol.
At 20 minutes, the SBP was 110, 113, and 125 mm Hg in groups 25, 50, and 100, respectively. Overall, SBP was 6% higher (P = 0.049) in group 100 compared with group 25 (Fig. 4). These small differences were further examined by an equivalence analysis, which confirmed that there were no significant clinical deviations of SBP among the groups as the ratio 1.06 (95% confidence interval at 0.99–1.13) was wholly contained within the 0.80 to 1.25 interval (P < 0.05). The number of minutes SBP was recorded as above baseline was significantly higher in group 100 (linear trend; P = 0.01) (Table 3), and the number of minutes SBP was recorded below baseline was significantly lower in group 100 (linear trend; P = 0.02.)
There were significant concentration-dependent effects on the duration of the infusion, infusion dose, and total dose received (linear trend; P < 0.05). The lower the concentration of phenylephrine, the longer the infusion time, but with lower total dose (Table 3). There was no difference in the median number of interventions (i.e., stopping/starting the infusion) while receiving the phenylephrine infusion among groups. The median number of phenylephrine and ephedrine boluses required per patient was not significantly different among the groups. Fewer boluses of phenylephrine were administered as the phenylephrine infusion concentration increased (trend analysis for proportions, P = 0.01). There were no differences among groups in the number of patients requiring ephedrine boluses. Fewer patients received phenylephrine or ephedrine boluses (some patients received both) as the phenylephrine infusion concentration increased (trend analysis for proportions, P < 0.02).
All patients received a pretest 500-mL crystalloid preload after the baseline readings. Preload was associated with increases in SV and CO (P < 0.001) and a small reduction in HR (P = 0.02).
Before the start of surgery, all patients achieved a bilateral sensory anesthesia level to touch ranging from T5 to T2. Six patients needed supplementation with 0.5% wt/vol levobupivacaine (3 patients in group 25, 2 in group 50, and 1 in group 100). Six patients in group 25 (2 mild, 3 moderate, 1 severe), 1 patient in group 50 (mild), and no patients in group 100 had symptoms of nausea and vomiting (P = 0.01).
Obstetric data, including interval from spinal anesthesia to skin incision and delivery, as well as interval from skin incision to delivery, are shown in Table 5. There were no significant differences among groups. There were also no significant differences in Apgar scores or neonatal cord gas values (Table 5).
The significantly larger dose of phenylephrine received by group 100 had a marked effect on maternal HR and CO. Maternal HR was decreased steadily with time in all groups, and at all times HR was slower in group 100, and this dose-dependent effect was significantly different 20 minutes after spinal anesthesia. The decrease in HR had a profound effect on maternal CO. SV remained stable, therefore we can attribute the changes that occurred in CO solely to a decrease in HR. Initiation of a phenylephrine infusion is usually associated with an increase in SVR, a consequent reduction in SV, and a baroreceptor-mediated bradycardia.18 We did not commence Doppler recordings until 5 minutes after induction of spinal anesthesia, which may account for the fact that we did not see an initial decrease in SV values.
Previous work has also demonstrated a reduction in CO associated with phenylephrine infusions.18,19 Work done by Langesaeter et al.19 led to questioning the routine clinical practice of maintaining SBP at baseline with such high concentrations of phenylephrine, at the expense of a negative effect on maternal CO. Using the minimally invasive lithium dilution technique (LiDCO™ Plus; LiDCO Ltd., Lake Villa, IL), the investigators demonstrated that even a low-dose phenylephrine infusion (0.25 μg/kg/min [equivalent to 16 μg/min in a woman weighing 65 kg ]) resulted in significantly lower HR and CO compared with a group receiving a placebo infusion.
Although the LiDCO Plus technique allows continuous measurement of CO, it can only be described as a minimally invasive technique because it requires arterial line placement. The technique has been recently criticized as a tool for assessing CO during periods of hemodynamic instability20; errors >33% were noted during cardiac surgery. In our study, we used a suprasternal Doppler technique that measures flow across the aortic arch to estimate serial changes in maternal CO. The advantage of this technique over the LiDCO Plus technique is that it is truly noninvasive, making it ideal in both the elective and emergency setting. It is a well-validated technique15 and has been used in the pregnant and nonpregnant populations.16,17 However, its main drawbacks are that it is an intermittent technique, it does not allow for continuous monitoring, and it does require some training in its use to achieve reproducibility. It gives a calculation of CO taken from across the descending part of the aortic arch, distal to the common carotid and subclavian arteries; therefore, it typically underestimates CO by approximately 10%. However, in our study, it was not actual values of CO that we were interested in, but the trend in each patient that was important.
Dyer et al.21 studied hemodynamic changes associated with spinal anesthesia for cesarean delivery in severe preeclampsia using the LiDCO Plus and pulse wave form analysis. They found that the administration of small boluses of phenylephrine in response to a 20% decrease in mean arterial blood pressure restored SVR to levels close to baseline. This was associated with a trend toward a reduction in CO. The investigators suggested that further work was required to establish whether a mixed acting vasopressor with α and β agonist activity could have advantages for the mother with severe preeclampsia and her fetus.21 In a more recent study, the same group compared the hemodynamic effects of bolus ephedrine and phenylephrine administered in response to a 20% decrease in mean arterial blood pressure. Bolus phenylephrine was found to reduce maternal CO, and CO changes correlated strongly with HR changes. It was concluded that a low bolus dose of phenylephrine is the most appropriate intervention in most cases to restore SVR and CO, and that HR is the best surrogate marker for CO.18 Their findings support our view, based on the present study, that continuous infusions of phenylephrine sufficient to cause sinus bradycardia should be avoided.
This prospective, double-blind, randomized, controlled trial demonstrated that all 3 infusion regimens of phenylephrine maintained the maternal SBP equally. However, to achieve this control, those who received 100 μg/min, despite receiving a significantly shorter mean infusion time, received a significantly higher total dose of phenylephrine (twice that received by patients who received 50 μg/min and nearly 3 times that received by parturients who received 25 μg/min). Although there was a small and statistically significant 6% increase of SBP in the high-dose compared with the low-dose phenylephrine group, control in the 3 groups was deemed equivalent, as defined for usual clinical purposes. There was, however, a greater degree of variability in SBP measurements in the low-dose group, and this variability was reduced with higher doses of phenylephrine. Although the higher SBP observed in the high-dose group may have contributed to the lower incidence of nausea and vomiting, it is possible that the greater variability in SBP control in the low-dose group was the more important factor. This would support work done by Ngan Kee et al.,22 which showed an improvement in the incidence of nausea and vomiting in a group of parturients whose SBP was controlled to 100% of baseline with a phenylephrine infusion, compared with groups controlled to 80% and 90% of baseline.
There is strong evidence to suggest that CO, specifically the maximum change in CO, correlates more closely with uteroplacental blood flow than upper arm blood pressure measurement.23 Uterine blood flow increases throughout pregnancy, reaching 800 mL/min at term (10%–15% of the maternal CO). Uterine blood flow, and therefore oxygen delivery to the fetus, is dependent on the maternal CO. The fetus has no storage capacity for oxygen and is vulnerable if placental oxygen delivery should fail. The well-being of the fetus requires that maternal CO, uterine blood flow, and maternal PaO2 are maintained at or above normal values for pregnancy.24
We did not see any adverse effects on the fetus in our study as indicated by Apgar scores and umbilical arterial and venous gases. However, this degree of decrease in CO may have a detrimental effect on the fetus in the emergency situation in which fetal acidosis might already be present. Under these circumstances, we should be doing all we can to improve oxygen delivery to the fetus by maintaining maternal CO and blood pressure as near to normal as possible. Our study has demonstrated that a decrease in the maternal HR, with phenylephrine use, is associated with a similar decrease in maternal CO. Therefore, if high doses of phenylephrine are required to maintain the maternal SBP, and a decrease in HR is observed, the anesthesiologist should be alerted that the maternal CO is compromised and initiate prompt and aggressive treatment of the slow HR, either by stopping the phenylephrine infusion if SBP is satisfactory or by using a chronotropic drug.
Theoretically, treatment of decreasing HR should minimize the risk of a sustained decrease in maternal CO and optimize oxygen delivery to the fetus. However, no studies examining the effects of phenylephrine on maternal CO and/or umbilical cord gases have evaluated the effects on uteroplacental blood flow or oxygen delivery to the fetus. We suggest that further work in this area should incorporate Doppler measurements of umbilical blood flow. An additional limitation of our study design is that we did not include a control group. It was thought at the time that we could not have a group that received an infusion containing no phenylephrine because it is well recognized that a sustained decrease in maternal SBP is associated with nausea and vomiting in the mother and acidosis in the fetus. However, we could have included a bolus-only group and compared this with the other 3 infusion groups.
In summary, good SBP control using phenylephrine during elective cesarean delivery can mask significant underlying maternal hemodynamic effects that are not obvious to the anesthesiologist using routine monitoring. The dose-dependent decrease in HR associated with a continuous infusion of phenylephrine during spinal anesthesia for cesarean delivery is associated with a similar reduction in maternal CO. Infusion rates of phenylephrine insufficient to cause a sinus bradycardia should be used to maintain maternal CO and therefore oxygen delivery to the fetus.
Supported by loan of the SupraQ, suprasternal Doppler machine from Deltex Medical, Chichester, West Sussex, UK. AS was supported by an unrestricted research grant from the Obstetric Anaesthetists' Association, UK. SM, RH, and TJ were supported by an unrestricted research grant from Smiths Medical, USA. RF was supported by the University College London Hospitals/University College London Comprehensive Biomedical Research Centre, which receives a proportion of funding from the United Kingdom Department of Health's National Institute of Health Research Biomedical Research Center's funding scheme.
a Ashpole K, Fernando R, Tamilselvan P, Columb M. Maternal cardiac output changes with phenylephrine and ephedrine infusions after spinal anaesthesia for caesarean section [abstract]. Int J Obstet Anesth 2005;14:S5.
1. Palmer SK. Anaesthesia care for obstetric patients in the United States. In: Reynolds F, ed. Regional Analgesia in Obstetrics: A Millennium Update. London: Springer-Verlag, 2000:3–10
2. Saravanan S, Kocarev M, Wilson RC, Watkins E, Columb MO, Lyons G. Equivalent dose of ephedrine and phenylephrine in the prevention of post-spinal hypotension in caesarean section. Br J Anaesth 2006;96:95–9
3. Macarthur A, Riley ET. Obstetric anesthesia controversies: vasopressor choice for postspinal hypotension during cesarean delivery. Int Anesthesiol Clin 2007;45:115–32
4. Ngan Kee WD, Khaw KS. Vasopressors in obstetrics: what should we be using? Curr Opin Anaesthesiol 2006;19:238–43
5. Ngan Kee WD, Khaw KS, Ng FF. Prevention of hypotension during spinal anesthesia for cesarean delivery: an effective technique using combination phenylephrine infusion and crystalloid cohydration. Anesthesiology 2005;103:744–50
6. Dyer RA, Farina Z, Joubert IA, Du Toit P, Meyer M, Torr G, Wells K, James MF. Crystalloid preload versus rapid crystalloid administration after induction of spinal anaesthesia (coload) for elective caesarean section. Anaesth Intensive Care 2004;32:351–7
7. Cooper DW, Carpenter M, Mowbray P, Desira WR, Ryall DM, Kokri MS. Fetal and maternal effects of phenylephrine and ephedrine during spinal anesthesia for cesarean delivery. Anesthesiology 2002;97:1582–90
8. Lee A, Ngan Kee WD, Gin T. A quantitative, systematic review of randomized controlled trials of ephedrine versus phenylephrine for the management of hypotension during spinal anesthesia for cesarean delivery. Anesth Analg 2002;94:920–6
9. Lee A, Ngan Kee WD, Gin T. A dose-response meta-analysis of prophylactic intravenous ephedrine for the prevention of hypotension during spinal anesthesia for elective cesarean delivery. Anesth Analg 2004;98:483–90
10. Smiley RM. Burden of proof. Anesthesiology 2009;111:470–2
11. Leduc L, Wasserstrum N, Spillman T, Cotton DB. Baroreflex function in normal pregnancy. Am J Obstet Gynecol 1991;165:886–90
12. Valensise H, Novelli GP, Vasapollo B, Borzi M, Arduini D, Galante A, Romanini C. Maternal cardiac systolic and diastolic function: relationship with uteroplacental resistances—a Doppler and echocardiographic longitudinal study. Ultrasound Obstet Gynecol 2000;15:487–97
13. Cooper DW. Effect of vasopressors on systolic, mean and diastolic arterial pressure during spinal anaesthesia in pregnancy. Int J Obstet Anesth 2008;17:90–2
14. Reynolds F, Seed PT. Anaesthesia for caesarean section and neonatal acid-base status: a meta-analysis. Anaesthesia 2005;60: 636–53
15. Easterling TR, Watts DH, Schmucker BC, Benedetti TJ. Measurement of cardiac output during pregnancy: validation of Doppler technique and clinical observations in preeclampsia. Obstet Gynecol 1987;69:845–50
16. Robson SC, Dunlop W, Moore M, Hunter S. Combined Doppler and echocardiographic measurement of cardiac output: theory and application in pregnancy. Br J Obstet Gynaecol 1987;94:1014–27
17. Mowat DH, Haites NE, Rawles JM. Aortic blood velocity measurement in healthy adults using a simple ultrasound technique. Cardiovasc Res 1983;17:75–80
18. Dyer RA, Reed AR, van Dyk D, Arcache MJ, Hodges O, Lombard CJ, Greenwood J, James MF. Hemodynamic effects of ephedrine, phenylephrine, and the coadministration of phenylephrine with oxytocin during spinal anesthesia for elective cesarean delivery. Anesthesiology 2009;111:753–65
19. Langesaeter E, Rosseland LA, Stubhaug A. Continuous invasive blood pressure and cardiac output monitoring during cesarean delivery: a randomized, double-blind comparison of low-dose versus high-dose spinal anesthesia with intravenous phenylephrine or placebo infusion. Anesthesiology 2008;109:856–63
20. Pauca AL. Pressure wave analysis is useful to understand the pathophysiology of preeclampsia, but perhaps not the rapid changes during cesarean delivery. Anesthesiology 2008;108: 773–4
21. Dyer RA, Piercy JL, Reed AR, Lombard CJ, Schoeman LK, James MF. Hemodynamic changes associated with spinal anesthesia for cesarean delivery in severe preeclampsia. Anesthesiology 2008;108:802–11
22. Ngan Kee WD, Khaw KS, Ng FF. Comparison of phenylephrine infusion regimens for maintaining maternal blood pressure during spinal anaesthesia for caesarean section. Br J Anaesth 2004;92:469–74
23. Robson SC, Boys RJ, Rodeck C, Morgan B. Maternal and fetal haemodynamic effects of spinal and extradural anaesthesia for elective caesarean section. Br J Anaesth 1992;68:54–9
© 2010 International Anesthesia Research Society
24. Ngan Kee WD. Uteroplacental blood flow. In: Chestnut DH, Polley LS, Tsen LC, Wong CA, eds. Obstetric Anesthesia: Principles and Practice. 4th ed. Philadelphia: Mosby Elsevier, 2009:37–53