Prophylactic Phenylephrine Infusions to Reduce Severe Spinal Anesthesia Hypotension During Cesarean Delivery in a Resource-Constrained Environment : Anesthesia & Analgesia

Secondary Logo

Journal Logo

Obstetric Anesthesiology

Prophylactic Phenylephrine Infusions to Reduce Severe Spinal Anesthesia Hypotension During Cesarean Delivery in a Resource-Constrained Environment

Bishop, David G. MBChB, FCA*; Cairns, Carel MBChB, FCA*; Grobbelaar, Mariette MBChB, FCA*; Rodseth, Reitze N. MBChB, FCA, MMed, Cert Crit Care, MSc, PhD*†

Author Information
Anesthesia & Analgesia 125(3):p 904-906, September 2017. | DOI: 10.1213/ANE.0000000000001905
  • Free

Prophylactic phenylephrine infusions are increasingly considered to be standard of care during obstetric spinal anesthesia (SA).1,2 However, few developing countries incorporate prophylactic phenylephrine infusions into their management guidelines.3 Yet, it is in precisely these contexts, faced by the majority of the world’s population, where spinal hypotension has its greatest adverse impact. A recent South African report found that over half of anesthetic deaths were attributed to spinal hypotension.4 The risk of death attributable to anesthesia in sub-Saharan Africa is 1.5 per 1000 women undergoing an obstetric procedure,5 whereas in the United States, the risk of death is 3.8 to 6.5 per million anesthetics.6 This pragmatic study, conducted in a resource-constrained environment with predominantly inexperienced staff, aimed to replicate findings regarding phenylephrine infusions reported in clinical trials conducted under optimal research conditions. We compared 2 management strategies: the first protocol employed a vasopressor bolus strategy to treat hypotension, and the second combined a fixed-rate, low-dose prophylactic phenylephrine infusion with additional vasopressor boluses as required. We hypothesized that prophylactic phenylephrine infusions would lower the incidence of severe hypotension without increasing complications.


Study Design

We undertook a single-center, prospective, alternating intervention study of patients undergoing cesarean delivery (CD) under SA. Ethics approval was given by the University of KwaZulu-Natal Ethics Committee (BFC013/15). Because the intervention for this study was at an institutional level and made use of management protocols with equivalent therapies and equivalent risk profiles, the local ethics and provincial ethics committees waived the need for consent. Patients were informed of the study and were given the opportunity to withdraw their data if they so desired.

Patients and Setting

Edendale Hospital is a regional referral center that caters to both routine and complex obstetric cases. The delivery rate is 700 births per month, with a CD rate of 30%. Both obstetric and anesthesia departments are staffed largely with inexperienced junior doctors. Medical officers with less than 1 year of anesthesia experience commonly provide obstetric anesthesia.

Recruitment and Protocols

Recruitment occurred from June 29, 2015, to November 25, 2015, in patients scheduled for CD under SA between 7:30 am and 4:00 pm on normal working days. Alternating spinal hypotension protocols were used. The first protocol was used exclusively for a 2-week period, followed by the second for the subsequent 2 weeks. Choice of first protocol was decided by the flip of a coin.

Management and Data Collection

Before administering SA, baseline heart rate (HR) and noninvasive blood pressure were recorded by the attending anesthetist. Automated noninvasive blood pressure measurements were taken at 1-minute intervals. Patients received metoclopramide 10 mg intravenously and oral sodium citrate 30 mL preoperatively. Spinal anesthesia was standardized: all patients received 9 mg of 0.5% hyperbaric bupivacaine with 10 μg fentanyl, administered in the sitting position and then immediately put into the supine position with left lateral tilt. A 1-L coload of Ringer’s lactate was administered via an 18-G intravenous line before delivery. After delivery, 3 international units (IU) of oxytocin were administered as a slow injection and 7 IU of oxytocin were given by infusion over the remainder of the operation.

Hypotension was managed either according to the bolus or the phenylephrine infusion protocol. The infusion protocol differed from the bolus protocol only by the addition of a fixed-rate 25 μg/min (50 μg/mL at 30 mL/h) prophylactic phenylephrine infusion commenced immediately after spinal insertion. Boluses were given in both protocols if mean arterial pressure (MAP) was less than 80% of baseline MAP or systolic blood pressure (SBP) was <90 mm Hg. If the HR was >70 beats per minute, phenylephrine (50–100 μg) was administered, and if the HR was less than 70 beats per minute, ephedrine was administered (5–10 mg). If the MAP increased to greater than 20% above baseline, the infusion was discontinued. Infusions were otherwise continued until after administration of oxytocin, and then weaned rapidly, aiming to maintain MAP within 20% of its initial value and SBP >90 mm Hg.

Hemodynamic data were recorded with Nihon-Kohden Lifescope monitors (BSM 3562; Nihon-Kohden, Tokyo, Japan). Intraoperative vomiting was assessed by direct observation. Blood loss was estimated by the anesthetist based on visual inspection of the abdominal swabs and suction containers, and discussion with the obstetric surgeon. Data were entered into a Microsoft Excel spreadsheet by a research assistant and checked by a study investigator. Data were then exported to a statistical program (StataCorp. 2013. Stata: Release 13. Statistical Software; StataCorp LP, College Station, TX) for further analysis.


The primary study outcome was severe hypotension defined as either MAP < 70% below baseline or SBP < 80 mm Hg. Hypertension was defined as MAP > 20% above baseline. We also recorded adverse events: mortality within 24 hours of SA, intensive care unit admission within 24 hours of SA, conversion to general anesthesia (GA) due to loss of consciousness, intensive care unit admission due to on-table loss of consciousness, cardiac arrest requiring cardiopulmonary resuscitation, and emergent administration of atropine or adrenaline.


Baseline demographics were compared using the Student t test and the χ2 test. The Shapiro-Wilk test was used for testing whether data were normally distributed. The primary outcome was the incidence of severe hypotension: this was tested using a χ2 test. For all analysis, an α value of <.05 defined statistical significance. A priori we deemed a 30% reduction in the primary outcome to be clinically important. Although initial sample size calculations were incorrect, the study was still adequately powered. With the achieved 253 patients per group, we had 93% power at the 0.5 significance level to detect a relative reduction of 30% in the proportion with severe hypotension, assuming a 50% incidence in the worst group. The STROBE guideline for the reporting of observational studies was followed.7


A total of 523 patients were included in the study. Seventeen patients were excluded for the following reasons: 1 duplicate entry; 2 patients erroneously included who had a GA as the primary anesthetic; 1 received the incorrect bupivacaine dose; and 13 required conversion to GA during the study period. For the patients converted to GA, 11 of these had failed SA, 1 suffered postpartum hemorrhage (bolus group), and 1 had an iatrogenic bowel injury (bolus group). The final analysis included 506 patients: 253 in the bolus group and 253 in the phenylephrine infusion group. The patient characteristics of the 2 groups are outlined in the Table.

Patient Characteristics

For the primary outcome of severe hypotension (MAP < 70% of baseline or SBP <80 mm Hg), the estimated incidence in the phenylephrine group was 47.4% (n = 120/253) vs 62.1% (n = 157/253) in the bolus group. This constituted a 23.6% reduction in severe hypotension in the phenylephrine when compared with the bolus group (P = .001, estimated relative risk 0.84, 95% confidence interval, 0.69–1.02).

The estimated incidence of hypertension was 13% (n = 66/506) and was not different between 15% (n = 39/253) in the phenylephrine group and 11% (n = 27/253) in the bolus group (P = .11). Eleven patients became bradycardic (HR < 50) during the study period: 7 in the phenylephrine group and 4 in the bolus group. No patient required atropine administration.

There were no other adverse outcomes.


This study found that fixed-rate, low-dose prophylactic phenylephrine infusions reduced the incidence of severe hypotension in real-world, resource-limited conditions. Although prophylactic phenylephrine infusions are increasingly becoming standard of care, there has been a reluctance to apply these recommendations to daily obstetric anesthesia practice, even in similar clinical contexts to the research environment.8 Furthermore, because research in this field has been conducted predominantly in elective, healthy patients, both international9 and local guidelines10 reflect caution in generalizing applicability of the findings. This pragmatic study applied the protocols to a diverse heterogeneous population group and included any patient deemed fit for SA, including both elective and urgent CD.

Our study has several limitations. Initial power calculations were incorrectly calculated, but fortunately, the study was sufficiently powered for the primary outcome. We also used a composite hypotension definition of either an absolute SBP < 80 mm Hg or a reduction of MAP to < 70% of the baseline value. It is possible that this may overestimate the incidence of severe hypotension, because preoperative anxiety may elevate maternal blood pressure. This may have been compounded by the single blood pressure reading taken to establish the baseline, which was necessary to prevent delays in emergency surgery. This is likely to have affected both groups to the same degree. We also recorded the incidence of severe hypotension as a binary outcome and were thus unable to differentiate between isolated readings of severe hypotension and protracted periods of hypotension.


This pragmatic study, conducted in a resource-limited setting with inexperienced staff, has shown that a prophylactic low-dose phenylephrine infusion is able to significantly reduce the incidence of severe hypotension in patients undergoing both elective and urgent CD under SA. In the light of these findings, and the strong existing evidence base, we propose that consideration be given to the routine use of prophylactic phenylephrine infusions in low-resource environments.


The authors would like to thank Robert A. Dyer, MBChB, FCA, PhD, Professor and Second Chair, Department of Anaesthesia and Perioperative Medicine, University of Cape Town, Cape Town, South Africa, for constructive input into the preparation of the final manuscript.


Name: David G. Bishop, MBChB, FCA.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Name: Carel Cairns, MBChB, FCA.

Contribution: This author helped design the study, conduct the study, and approved the manuscript.

Name: Mariette Grobbelaar, MBChB, FCA.

Contribution: This author helped design the study, conduct the study, and approved the manuscript.

Name: Reitze N. Rodseth, MBChB, FCA, MMed, Cert Crit Care, MSc, PhD.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

This manuscript was handled by: Jill M. Mhyre, MD.


1. Allen TK, George RB, White WD, Muir HA, Habib AS. A double-blind, placebo-controlled trial of four fixed rate infusion regimens of phenylephrine for hemodynamic support during spinal anesthesia for cesarean delivery. Anesth Analg. 2010;111:1221–1229.
2. Heesen M, Kölhr S, Rossaint R, Straube S. Prophylactic phenylephrine for caesarean section under spinal anaesthesia: systematic review and meta-analysis. Anaesthesia. 2014;69:143–165.
3. Nag DS, Samaddar DP, Chatterjee A, Kumar H, Dembla A. Vasopressors in obstetric anesthesia: a current perspective. World J Clin Cases. 2015;3:58–64.
4. NCCEMD. Saving mothers 2011–2013: sixth report on the confidential enquiries into maternal deaths in South Africa. Compiled by the National Committee for Confidential Enquiry into Maternal Deaths. 2014.
5. Sobhy S, Zamora J, Dharmarajah K, et al. Anaesthesia-related maternal mortality in low-income and middle-income countries: a systematic review and meta-analysis. Lancet Glob Health. 2016;4:e320–e327.
6. Hawkins JL, Chang J, Palmer SK, Gibbs CP, Callaghan WM. Anesthesia-related maternal mortality in the United States: 1979-2002. Obstet Gynecol. 2011;117:69–74.
7. Vandenbroucke JP, von Elm E, Altman DG, et al. Strengthening the reporting of observational studies in epidemiology (STROBE): explanation and elaboration. Ann Intern Med. 2007;147:W163W94.
8. Smiley RM. Burden of proof. Anesthesiology. 2009;111:470–472.
9. NICE Caesarean Section. National Institute for Health and Care Excellence. 2011. Available at: Accessed June 13, 2016.
10. Dyer RA, Rout CC, Kruger AM, van der Vyver M, Lamacraft G, James MF. Prevention and treatment of cardiovascular instability during spinal anaesthesia for caesarean section. S Afr Med J. 2004;94:367–372.
Copyright © 2017 International Anesthesia Research Society