Anaesthesia 2014, 69, 143–165

doi:10.1111/anae.12445

Review Article Prophylactic phenylephrine for caesarean section under spinal anaesthesia: systematic review and meta-analysis M. Heesen,1 S. Kl€ ohr,2 R. Rossaint3 and S. Straube4 1 Professor, 2 Consultant, Department of Anaesthesia, Klinikum am Bruderwald, Bamberg, Germany 3 Professor, Department of Anaesthesia, University Hospital Aachen, Aachen, Germany 4 Occupational physician and researcher, Institute of Occupational, Social and Environmental Medicine, University Medical Center G€ottingen, G€ottingen, Germany

Summary We conducted a systematic review to determine the harm and benefit associated with prophylactic phenylephrine for caesarean section under spinal anaesthesia. We included 21 randomised controlled trials with 1504 women. The relative risk (95% CI) of hypotension with phenylephrine infusion – as defined by authors – before delivery was 0.36 (0.18–0.73) vs placebo, p = 0.004; 0.58 (0.39–0.88) vs an ephedrine infusion, p = 0.009; and 0.73 (0.55–0.96) when added to an ephedrine infusion, p = 0.02. After delivery, the relative risks of hypotension and nausea and vomiting with phenylephrine compared with placebo were 0.37 (0.19–0.71), p = 0.003, and 0.39 (0.17–0.91), p = 0.03, respectively. There was no evidence that hypertension, bradycardia or neonatal endpoints were affected. Phenylephrine reduced the risk for hypotension and nausea and vomiting after spinal doses of bupivacaine generally exceeding 8 mg, but there was no evidence that it reduced other maternal or neonatal morbidities. .................................................................................................................................................................

Correspondence to M. Heesen Email: [email protected] Accepted: 16 August 2013

Phenylephrine is the a-agonist recommended to treat hypotension that affects approximately half of caesarean sections under spinal anaesthesia [1, 2]. Phenylephrine is associated with a lower rate of fetal acidosis compared with ephedrine [3]. Prophylactic administration of a vasopressor has been recommended [4] and in some institutions it is routine, but it is unclear whether any benefit might be outweighed by harm [5, 6]. The aim of this systematic review was to determine the harm and the benefit associated with prophylactic phenylephrine infusion.

Methods We searched PubMed, Embase, CINAHL, LILACS, CENTRAL and ISI WOS from inception to February © 2013 The Association of Anaesthetists of Great Britain and Ireland

26th, 2013 (Appendix). We also screened the reference lists of retrieved studies. We included randomised controlled trials (RCTs) of prophylactic phenylephrine vs placebo or other intervention started before, during or after neuraxial blockade. We included any dose or mode of phenylephrine administration, as well as its combination with another vasopressor. Two authors (MH, SS) independently assessed as adequate, unclear or inadequate the risk of bias in the following domains: selection (random sequence generation, allocation concealment); performance (blinding of participant and personnel); detection (blinding of assessor). We excluded studies categorised as at high risk of selection bias and also studies with more than one of the following categor143

Identification

Anaesthesia 2014, 69, 143–165

Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia

Records identified through database searching: n = 421

Additional records identified through other sources: n = 5

Screening

Records after duplicates removed: n = 423

Records screened: n = 423

Records excluded: n = 396

Records excluded:

Eligibility

Not double-blind: n = 2 Full-text articles assessed for eligibility: n = 27

Double blinding inadequate: n = 1 Vasopressor not given prophylactically: n = 2 No control included n = 1

Included

Number of studies included in qualitative synthesis: n = 21

Number of studies included in quantitative synthesis (meta-analysis): n = 13

Figure 1 Flow diagram of article selection.

ised as inadequate: patient blinding; operator blinding; or assessor blinding. We recorded maternal rates of hypotension, hypertension, bradycardia and nausea or vomiting, using the authors’ definitions for each, which therefore varied between studies. We also recorded uterine circulatory variables, arterial and venous umbilical pH and base excess, and neonatal Apgar scores. We focused on hypotension before delivery as most cases of hypotension occur in the first 15 min after spinal injection [7], whilst haemodynamic changes after delivery do not directly influence fetal outcome. We calculated pooled estimates for comparisons with two or more RCTs and described other comparisons without meta-analysis. Consequently, we only 144

used meta-analysis to compare prophylactic phenylephrine infusion: (1) vs placebo; (2) vs ephedrine infusion; (3) combined with ephedrine vs ephedrine alone. Relative risks (RR) with 95% CI were calculated for dichotomous variables, and weighted mean differences (WMD) with 95% CI for continuous variables. We combined data from multiple phenylephrine infusions in single RCTs; by simple addition for dichotomous variables and by the Cochrane Handbook formula for continuous variables [8]. We calculated means and SDs from medians and ranges using the formula described by Hozo and colleagues [9]. © 2013 The Association of Anaesthetists of Great Britain and Ireland

© 2013 The Association of Anaesthetists of Great Britain and Ireland

Mercier et al. 2001 [20]

Ayorinde et al. 2001 [19]

Hall et al. 1994 [18]

Alahuhta et al. 1992 [4]

Operator blinding: unclear

Allocation concealment: unclear

Operator blinding: adequate (“solutions were prepared independently and coded by a recovery nurse”)

Allocation concealment: adequate (“solutions were prepared independently and coded by a recovery nurse”)

Allocation concealment: adequate (“sealed, opaque envelopes”)

Random sequence generation: adequate (“random table”)

Allocation concealment: adequate (“randomised by a coded, opaque, sealed envelope technique”)

Random sequence generation: adequate (“randomised by a coded, opaque, sealed envelope technique”)

Operator blinding: adequate (“double-blind design”, “The anaesthetist who prepared the infusions and who acted as investigator was aware to which group each mother had been assigned, but he was not involved in the management of the mother’s anaesthetic.”)

Allocation concealment: unclear

Assessor blinding: adequate (“Both the patient and the investigator were blinded to group assignment.”)

Operator blinding: adequate (“Study solutions were prepared by an anesthesiologist or a nurse anesthetist not involved in the patients’ care, according to the group indicated in numbered, sealed, opaque envelopes.”)

Patient blinding: adequate (“double-blind”, “Both the patient and the investigator were blinded to group assignment.”)

Assessor blinding: unclear

Operator blinding: adequate (patients managed by an anaesthetist blind to the identity of the study medication)

Patient blinding: adequate (“double-blind”)

Assessor blinding: inadequate (“The anaesthetist who prepared the infusions and who acted as investigator was aware to which group each mother had been assigned, but he was not involved in the management of the mother’s anaesthetic.”, “The time taken for anaesthesia to reach T4 and maximum height of sensory loss was recorded by the investigator, and the time between insertion of the spinal needle to delivery of the fetus and uterine incision to delivery time of the neonate. The incidence and nature of any complications were noted.”)

Patient blinding: adequate (“double-blind design”)

Random sequence generation: unclear (“allocated randomly”)

Assessor blinding: unclear

Patient blinding: adequate (“double-blind”)

Random sequence generation: adequate (“randomization list”)

Assessor blinding: unclear (Apgar and Early Neonatal Neurobehavioural Scale scoring performed by blinded personnel)

Patient blinding: adequate (“double-blind fashion”)

Random sequence generation: unclear (“randomised”)

Moran et al. 1991 [17]

Blinding

Randomisation

Study

Table 1 Assessment of bias, supported by quotations from the respective articles.

Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia Anaesthesia 2014, 69, 143–165

145

146

Saravanan et al. 2006 [24]

Loughrey et al. 2005 [23]

Ngan Kee et al. 2004 [7]

Cooper et al. 2004 [22]

Random sequence generation: adequate (“randomly allocated by envelope selection”)

Cooper et al. 2002 [21]

Allocation concealment: unclear

Random sequence generation: adequate (“computer-generated random code”)

Allocation concealment: adequate (“randomised … by sealed envelope”, “pre-prepared syringes marked “Study Vasopressor”.”)

Random sequence generation: adequate (“randomised … by sealed envelope”)

Allocation concealment: adequate (“sealed, sequentially numbered envelopes”, “Two identical syringes were prepared containing either phenylephrine 100 lg/mL or saline”)

Random sequence generation: adequate (“computer-generated randomization codes”)

Allocation concealment: adequate (“randomly allocated by envelope selection”)

Random sequence generation: adequate (“randomly allocated by envelope selection”)

Allocation concealment: adequate (“randomly allocated by envelope selection”, “A third party, not involved with the study, opened an envelope containing the code for the patient group and gave the investigator the relevant unlabelled syringe.”)

Randomisation

Study

Table 1. (Continued) Blinding

Assessor blinding: adequate (“Vasopressor solutions were freshly prepared at room temperature immediately before caesarean section by an anaesthetist who had no other role in the study. An observer blinded to the nature of the vasopressor infusion assessed the efficacy of each solution.”)

Operator blinding: adequate (“Vasopressor solutions were freshly prepared at room temperature immediately before caesarean section by an anaesthetist who had no other role in the study.”)

Patient blinding: adequate (“double-blind”)

Assessor blinding: unclear

Operator blinding: adequate (“The anesthesiologist remained blinded to the study solution throughout.”)

Patient blinding: adequate (“double-blinded fashion”)

Assessor blinding: adequate (“an investigator or anesthesiologist not involved with patient care or assessment … selected one of the syringes to be used for IV infusion and the other for IV boluses.”, “Apgar scores were assessed 1 and 5 min after delivery by the attending pediatrician, who was blinded to the patient’s group.”)

Operator blinding: adequate (“an investigator or anesthesiologist not involved with patient care or assessment … selected one of the syringes to be used for IV infusion and the other for IV boluses. The investigator administering the solutions … and the patient were thus blinded to the contents of the syringes.”)

Patient blinding: adequate (“double-blinded”, “The investigator administering the solutions … and the patient were thus blinded to the contents of the syringes.”)

Assessor blinding: unclear

Operator blinding: adequate (“The patients, anesthetists, and nurses involved with patient care were blinded to the patient grouping.”)

Patient blinding: adequate (“double blind”, “The patients, anesthetists, and nurses involved with patient care were blinded to the patient grouping.”)

Assessor blinding: unclear

Operator blinding: adequate (“The patients, anesthetists, nurses, and midwives involved with patient care were blinded to the patient grouping.”)

Patient blinding: adequate (“double blind”, “The patients, anesthetists, nurses, and midwives involved with patient care were blinded to the patient grouping.”)

Anaesthesia 2014, 69, 143–165 Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia

© 2013 The Association of Anaesthetists of Great Britain and Ireland

© 2013 The Association of Anaesthetists of Great Britain and Ireland

Hennebry et al. 2009 [27]

Ngan Kee et al. 2008 [26]

Langesæter et al. 2008 [12]

Allocation concealment: adequate (“randomised in pairs using sealed opaque envelopes”)

Random sequence generation: adequate (“randomised in pairs using sealed opaque envelopes”)

Allocation concealment: adequate (“opaque, sealed, sequentially numbered envelopes”)

Random sequence generation: adequate (“randomly allocated patients … according to computer generated codes”)

Allocation concealment: adequate (“The senior author, who was not involved in the handling of the drugs or the participants, performed the randomization … Block size and randomization codes were not revealed to the investigators until all measurements and calculations had been entered into the database.”, “opaque, sealed envelopes”, “All data were constructed before breaking the randomization codes to ensure against biased handling of the data.”)

Random sequence generation: adequate (“randomization in blocks of eight to four groups of equal size using a list of random numbers according to the Moses-Oakford algorithm.”)

Operator blinding: adequate (“The patients, anaesthetists, and nurses involved with patient care were blinded to the patient grouping.”)

Allocation concealment: adequate (“computer-generated code that was kept in a numbered envelope”)

Assessor blinding: adequate (“An observer was blinded to the dose of bupivacaine and also to the nature of the vasopressor used. This was achieved by covering up the infusion bag and positioning the cardiovascular monitor out of sight of the observer. The observer assessed the efficacy of the block to light touch using ethyl chloride spray bilaterally”, “In this study the block assessor was blinded to the dose of bupivacaine and also to the nature of the vasopressor used.”)

Operator blinding: inadequate (“Vasopressor solutions were freshly prepared at room temperature immediately before use by the unblinded anaesthetist performing the CSE.”, “Blood pressure was managed by the unblinded anaesthetist according to a strict protocol.”)

Patient blinding: adequate (“double-blind”)

Assessor blinding: unclear

Operator blinding: adequate (“The vasopressor solutions were prepared in identical 50 mL syringes by an investigator not involved in patient care.”)

Patient blinding: adequate (“double-blind”)

Assessor blinding: adequate (“To maintain blinding of both patients and examiner throughout the study, syringes for each patient were prepared in the morning of surgery by a doctor or nurse not involved in the treatment or assessment of the patients.”)

Operator blinding: adequate (“A 50-ml syringe containing 30 ml phenylephrine or placebo, and a 10-ml syringe containing the spinal solution marked with the randomization number and neutral study information were delivered to the primary investigator in the operating theater. Unblinding of the investigators was tested by registering a guess at the treatment combination just after induction of spinal anesthesia and a second guess when the intravenous test drugs were stopped after 20 min.”)

Patient blinding: adequate (“double-blinded study”, “To maintain blinding of both patients and examiner throughout the study, syringes for each patient were prepared in the morning of surgery by a doctor or nurse not involved in the treatment or assessment of the patients.”)

Assessor blinding: unclear

Patient blinding: adequate (“double blind”, “The patients, anaesthetists, and nurses involved with patient care were blinded to the patient grouping.”)

Random sequence generation: adequate (“randomly allocated to group by a computer-generated code”)

Cooper et al. 2007 [25]

Blinding

Randomisation

Study

Table 1. (Continued)

Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia Anaesthesia 2014, 69, 143–165

147

148

Mohta et al. 2010 [30]

das Neves et al. 2010 [13]

Allen et al. 2010 [15]

Ngan Kee et al. 2009 [29]

Random sequence generation: adequate (“random numbers generated … by a computer”)

~es Magalha et al. 2009 [28]

Operator blinding: adequate (“The solution of vasopressor for infusion was prepared by an assistant who was not involved in the study”)

Allocation concealment: adequate (“randomly divided … using a sealed envelope technique”)

Assessor blinding: adequate (“the investigator as well as the patient were thus blinded to the identity of vasopressor used”)

Patient blinding: adequate (“double-blind”, “the investigator as well as the patient were thus blinded to the identity of vasopressor used”)

Assessor blinding: adequate (“The result of the distribution was not revealed, and it was not known by patients or physicians responsible for collection and analysis of the data.”)

Operator blinding: inadequate (“Only the anesthesiologist responsible for the anesthesia was aware of which group the patient belonged to and which medications were being administered.”)

Patient blinding: adequate (“double blind”, “The result of the distribution was not revealed, and it was not known by patients or physicians responsible for collection and analysis of the data.”)

Assessor blinding: unclear

Operator blinding: adequate (“To maintain blinding, the infusions were prepared in identical 50-mL syringes … by a physician not involved in the study.”)

Patient blinding: adequate (“double-blind”)

Assessor blinding: adequate (“To facilitate double-blinding, the drugs were prepared in identical syringes by one of the investigators who was not involved with subsequent patient management or data collection.”)

Operator blinding: adequate (“To facilitate double-blinding, the drugs were prepared in identical syringes by one of the investigators who was not involved with subsequent patient management or data collection.”)

Patient blinding: adequate (“double-blind”)

Assessor blinding: adequate (“The result of the allocation was ignored by both the patients and the physicians responsible for collecting and analyzing the study parameters.” “The syringes with the study drugs were prepared by a physician who was not involved in the collection of the data and analysis of the results.”)

Operator blinding: unclear

Patient blinding: adequate (“double blind”, “The result of the allocation was ignored by both the patients and the physicians responsible for collecting and analyzing the study parameters.”)

Blinding

Random sequence generation: adequate (“randomly divided … using a sealed envelope technique”)

Allocation concealment: adequate (“Patients were randomly distributed in three groups, using sequential sealed envelopes”)

Random sequence generation: adequate (“random computer-generated numbers”)

Allocation concealment: adequate (“Each syringe was identified by a study number”, “identical 50-mL syringes”)

Random sequence generation: adequate (“computer-generated randomization”)

Allocation concealment: adequate (“sealed envelopes”)

Random sequence generation: adequate (“computer-generated randomization code”)

Allocation concealment: adequate (“sequential, sealed envelopes”)

Randomisation

Study

Table 1. (Continued)

Anaesthesia 2014, 69, 143–165 Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia

© 2013 The Association of Anaesthetists of Great Britain and Ireland

© 2013 The Association of Anaesthetists of Great Britain and Ireland

Bhardwaj et al. 2013 [14]

Das et al. 2011 [32]

Allocation concealment: adequate (“sealed envelopes”)

Random sequence generation: adequate (“randomly allocated … by drawing of sequentially numbered sealed envelopes that contained a computer generated randomization code”)

Allocation concealment: adequate (“The sealed opaque envelope technique was used for concealment.”)

Random sequence generation: adequate (“Patients were randomly allocated by block randomization method, where one patient had every chance to get allocated in any group by using computer generated random numbers.”)

Allocation concealment: adequate (“Group assignments were sealed within opaque envelopes”)

Assessor blinding: adequate (data recorded by anaesthetist conducting the anaesthesia)

Operator blinding: adequate (unlabelled syringe prepared by investigators different from the investigators conducting the anaesthesia)

Patient blinding: adequate (“double-blind”)

Assessor blinding: adequate (“The anaesthesiologist, who prepared and supplied the drug as per randomization number, was not related to data collection, monitoring or conduct of anaesthesia. … Outcome assessors were unaware of group allocation, patient profile and the study drugs.”)

Operator blinding: adequate (“The anaesthesiologist, who prepared and supplied the drug as per randomization number, was not related to data collection, monitoring or conduct of anaesthesia.”)

Patient blinding: adequate (“double blind”)

Assessor blinding: adequate (“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 anaesthesiologist 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.”)

Operator blinding: adequate (“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 anaesthesiologist 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.”)

Random sequence generation: adequate (“computer generated random number table.”)

Stewart et al. 2010 [31]

Blinding Patient blinding: adequate (“double-blind”, “The patient and second and third anesthesiologists were blinded to group assignment.”)

Randomisation

Study

Table 1. (Continued)

Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia Anaesthesia 2014, 69, 143–165

149

Anaesthesia 2014, 69, 143–165

We used Review Manager (RevMan 5.1; The Nordic Cochrane Centre, Copenhagen, Denmark) for metaanalyses. We used the random-effects model because we expected methodological and clinical heterogeneity. Statistical significance was assumed for p values < 0.05. We measured the percentage of variation between studies due to heterogeneity, excess to that expected by chance, with the I2 statistic, categorising values > 50% as heterogeneous. Guidelines for the reporting of systematic reviews were followed where appropriate [10].

Results We included 21 RCTs with 1504 participants (Fig. 1). Table 1 describes our assessments of bias and Tables 2 and 3 contain detailed descriptions of the included studies. We contacted authors to clarify the period during which hypotension was assessed [11–13], blinding and the observation period [14]. Five studies compared prophylactic phenylephrine vs placebo infusion in 371 parturients (Figs 2 and 3). Phenylephrine reduced the rates of hypotension before delivery, RR 0.36 (0.18–0.73), p = 0.004 and after delivery, RR 0.37 (0.19–0.71), p = 0.003; and the composite ‘nausea or vomiting’, RR 0.39 (0.17–0.91), p = 0.03. There was significant heterogeneity between studies for hypotension before and after delivery. There was no evidence for significant effects of phenylephrine on the rates of maternal bradycardia, hypertension, vomiting or umbilical artery variables (Table 4a). The number of neonates with an Apgar score below 7, at 1 or 5 min, did not differ between groups in the studies by Ngan Kee et al. [11] and Langeseater et al. [12]. In the study by das Neves and colleagues [13], prophylactic phenylephrine, by either infusion or bolus, increased the number of babies with a 1-min Apgar score above 8 compared with treatment of hypotension after it had happened (p = 0.01 for both comparisons). Allen et al. [15] reported no differences in median Apgar scores at 1 and 5 min. Doherty et al. [16] reported no differences in the number of neonates with an Apgar value below 7 or 9, at 1 or 5 min. Seven studies compared prophylactic phenylephrine vs ephedrine infusions in 499 parturients. Phenylephrine reduced the rate of hypotension before delivery, RR 0.58 (0.39–0.88), p = 0.009 (Fig. 4). Hall 150

Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia

et al. [18] studied infusion of phenylephrine vs ephedrine (two doses). The rate of hypotension was lower with ephedrine (12/19) than with phenylephrine (9/10). If the data of Hall et al., which might have included hypotensive episodes after delivery, were included in the meta-analysis of hypotension before delivery, phenylephrine would not show any difference to ephedrine: RR 0.65 (0.39–1.09), p = 0.10. There was no evidence of effect for other maternal or neonatal variables (Table 4b). There was no difference in arterial or venous umbilical pH or base excess. Alahuhta et al. [4] studied fetal outcomes; they found increased arterial pulsatility (uterine, arcuate and fetal renal) after spinal injection when accompanied by an infusion of phenylephrine, but not ephedrine. There were no differences in fetal myocardial contractility, ventricular size, acidosis or Apgar score. Hypotension was more common after an 80-lg intravenous bolus of phenylephrine (28/ 30) vs 10 mg ephedrine (21/30) in a study by Magalh~aes et al. [28]. Umbilical arterial pH and Apgar score at 1 min were lower with ephedrine, whilst the 5 min Apgar scores were no different. Ayorinde et al. [19] studied intramuscular phenylephrine (4 mg, 2 mg) vs ephedrine (45 mg) or placebo. Phenylephrine at either dose reduced the rate of hypotension compared with placebo. Both ephedrine and phenylephrine (4 mg) reduced the rate of rescue ephedrine for hypotension compared with the placebo group. There were no differences in fetal pH or Apgar scores at 1 and 5 min. Three studies compared a prophylactic infusion of ephedrine with or without phenylephrine in 200 parturients (Table 4c). The addition of phenylephrine reduced the rate of hypotension before delivery, RR 0.73 (0.55–0.96), p = 0.02 (Fig. 5). There were no differences in rates of nausea or hypertension before delivery. In one study, the addition of intravenous phenylephrine (40 lg) had no effect on rates of hypotension or rescue ephedrine, or umbilical artery pH, when added to a bolus of intravenous ephedrine (10 mg) [23]. One study compared infusions of phenylephrine vs mephentermine and found no differences in hypotension, umbilical artery pH or Apgar scores, whilst hypertension occurred in 8/30 participants receiving phenylephrine and 0/30 receiving mephentermine [30]. © 2013 The Association of Anaesthetists of Great Britain and Ireland

Inclusion/exclusion criteria

Inclusion: ASA 1 or 2, term singleton pregnancies, elective section, allocated to one of two groups Exclusion: pre-existing or pregnancy-induced hypertension, cardiovascular or cerebrovascular disease, known fetal abnormalities, or contra-indications to spinal anaesthesia

Inclusion: healthy, term singleton pregnancy elective section, allocated to one of four groups Exclusion: pre-existing or gestational hypertension, pre-eclampsia, cardiovascular or cerebrovascular disease, a height less than 160 cm or greater than 180 cm, prepregnancy body mass index greater than 32 kg.m 2, or contra-indications to spinal anaesthesia

Inclusion: ASA 1, term single fetus, elective section, allocated to one of three groups Exclusion: pre-existing or gestational hypertension, pre-eclampsia, cardiovascular or cerebrovascular disease, a height less than 160 cm or greater than 180 cm, prepregnancy body mass index greater than 32 kg.m 2, or contra-indications to spinal anaesthesia

Study

Ngan Kee et al. 2004 [7]

Langesaeter et al. 2008 [12]

das Neves et al. 2010 [13]

50-lg bolus of phenylephrine (n = 40)

9 lg.kg 1.h 1 phenylephrine (n = 40) or 50-lg phenylephrine bolus (n = 40) or no prophylaxis (n = 40) 10 ml.kg 1 lactated Ringer’s solution, co-load

27 G, L 3–4 10 mg bupivacaine hyperbaric with 100 lg morphine Injection speed of 1 ml every 15 s

30-lg bolus of phenylephrine or 5 mg of ephedrine when hypotension was combined with bradycardia (heart rate < 55 min 1) (n = 40)

15 lg.kg 1.h 1 phenylepyrine (n = 40) or placebo (n = 40)

750 ml NaCl 0.9%, co-load

27 G, L 2–3 7 or 10 mg isobaric bupivacaine with 4 lg sufentanil

Rescue 100-lg bolus of phenylephrine (n = 24)

1

Intervention 6 mg.h phenylephrine (n = 26) or placebo (n = 24)

5 ml.min Ringer’s lactate, co-load

1

Intravenous preload

25 G, L 2–3 or L 3–4 10 mg bupivacaine hyperbaric with 15 lg fentanyl

Spinal anaesthesia

Table 2 Prophylactic phenylephrine infusion vs no prophylaxis.

© 2013 The Association of Anaesthetists of Great Britain and Ireland

2-min intervals noninvasive

Continuous, LiDCO technique

1-min intervals; noninvasive

Blood pressure monitoring

ND

SBP < 90 mmHg

ND

ND

SBP < 80% of baseline

SBP and/or DBP < 80% of baseline

Hypertension

Hypotension

Definitions

Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia Anaesthesia 2014, 69, 143–165

151

152

Inclusion: ASA 1 or 2, singleton, 36 weeks’ gestation, elective section, allocated to one of two groups Exclusion: ND

Inclusion: ASA 1 or 2, ≥ 18 years, weight 50–100 kg, height 150–180 cm, elective section, allocated to one of two groups Exclusion: patient refusal, inability to communicate in English, allergy or hypersensitivity to phenylephrine, hypertension, cardiovascular or cerebrovascular disease, fetal abnormalities, diabetes (excluding gestational diabetes), or contraindications to spinal anaesthesia

Allen et al. 2010 [15]

Doherty et al. 2012 [16]

7.2 mg.h 1 phenylephrine (n = 30) or placebo (n = 30)

10 ml.kg 1 lactated Ringer’s solution, co-load (max. 1 l)

13.5 mg hyperbaric bupivacaine with 10 lg fentanyl and 100 lg morphine

Intervention Phenylephrine 0 lg.min 1 (n = 20); 25 lg.min 1 (n = 20); 50 lg.min 1 (n = 20); 75 lg.min 1 (n = 19); 100 lg.min 1 (n = 22)

2 l Ringer’s solution, co-load

Intravenous preload

25 G, L 3–4 or L 4–5 12 mg hyperbaric bupivacaine with 15 lg fentanyl and 150 lg morphine

Spinal anaesthesia 1-min intervals for first 10 min, then 2.5 min until end of study; noninvasive 1-min intervals; noninvasive

120 lg phenylephrine if SBP ≤ baseline 5 mg of ephedrine if SBP < 80% of baseline 10 mg of ephedrine if SBP remained < 80% of baseline after 1 min (n = 30)

Rescue

Blood pressure monitoring

100-lg bolus of phenylephrine (n = 20)

DBP, diastolic blood pressure; LiDCO, lithium dilution cardiac output; ND, not defined; SBP, systolic blood pressure.

Inclusion/exclusion criteria

Study

Table 2. (Continued)

Hypertension SBP > 120% of baseline

SBP > 120% of baseline

Hypotension SBP < 80% of baseline

SBP < 80% of baseline

Definitions

Anaesthesia 2014, 69, 143–165 Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia

© 2013 The Association of Anaesthetists of Great Britain and Ireland

Inclusion: 38– 42 weeks’ gestation, healthy singleton pregnancy, elective section because of fetal breech presentation or cephalopelvic disproportion Exclusion: not detailed

Inclusion: healthy parturients, singleton fetus, no evidence of placental pathology, > 37 weeks’ gestation, elective section Exclusion: not detailed

Inclusion: ASA 1 or 2, elective section Exclusion: known hypertensive patients, resting arterial pressure greater than 160/ 90 mmHg,

Inclusion: ASA 1 or 2, ≥ 18 years, ≤ 90 kg, ≥ 152 cm, term singleton pregnancy, elective section

Alahuta et al.1992 [4]

Hall et al. 1994 [18]

Ayorinde et al. 2001 [19]

Mercier et al. 2001 [20]

Study

Inclusion/exclusion criteria

© 2013 The Association of Anaesthetists of Great Britain and Ireland

CSE: spinal 11 mg hyperbaric bupivacaine 0.5%, fentanyl 20 lg, sitting position

not detailed

500 ml Ringer‘s lactate

Spinal: 0.5% hyperbaric bupivacaine; 13.75 mg if maternal height < 160 cm, 15 mg otherwise, sitting position

Spinal: 11 mg hyperbaric bupivacaine 0.5%, 2.5 lg sufentanil, 0.1 mg morphine, sitting position

Ephedrine 6-mg bolus then 60 mg.h 1 (n = 10) or 120 mg.h 1 (n = 9), or phenylephrine 20-lg bolus then 0.6 mg.h 1 (n = 10)

20 ml.kg 1 Hartmann‘s solution

CSE: 11.5–13 mg bupivacaine 0.5% with 8.0% dextrose; right lateral position

Ephedrine 120 mg.h 1 alone (n = 20) or with phenylephrine 0.6 mg.h 1 (n = 19)

Phenylephrine 2 or 4 mg i.m. or ephedrine 45 mg i.m. (n = 27 for each)

Ephedrine 5-mg bolus (n = 9) followed by infusion of 50 mg.h 1, or phenylephrine 100-lg bolus (n = 8) followed by infusion of 1 mg.h 1

20 ml.kg 1 balanced electrolyte solution over 20 min

Spinal anaesthesia

Intervention

Intravenous preload

Table 3 Prophylactic phenylephrine vs any other prophylactic vasopressor.

Non-invasive, 1-min intervals for 10 min, then every 2 min until delivery

Non-invasive

Non-invasive and invasive

Automated non-invasive, 2-min intervals from induction to delivery

Blood pressure monitoring Hypertension ND

SBP > 120% of baseline

ND

SBP > 120% of baseline

Hypotension SBP < 90% baseline

SBP < 80% baseline

SBP ≤ 75% of baseline

SBP < 80% of baseline or < 100 mmHg

Definitions

Before delivery

ND

ND

Before delivery

Observation period

Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia Anaesthesia 2014, 69, 143–165

153

154

Cooper et al. 2004 [22]

Cooper et al. 2002 [21]

Study

Spinal: 12.5 mg hyperbaric bupivacaine 0.5%, 20 lg fentanyl Or CSE: 10 mg spinal levobupivacaine 0.5%, 20 lg fentanyl, sitting position or left lateral position Or CSE: 12.5 mg spinal levobupivacaine 0.5%, 10 lg fentanyl, left lateral position CSE: 10 mg spinal levobupivacaine 0.5%, 20 lg fentanyl, sitting then left lateral

Inclusion: ASA 1 or 2, elective section, singleton pregnancy, no known fetal abnormality, and no history of pre-eclampsia or diabetes mellitus

Spinal anaesthesia

Inclusion: ASA 1 or 2, elective section, singleton pregnancy, with no known fetal abnormality, and no history of pre-eclampsia or diabetes mellitus Exclusion: not detailed

Exclusion: pregnancy-induced hypertension, cardiac disease, diabetes, or fetal complications, being in labour

Inclusion/exclusion criteria

Table 3. (Continued)

10 ml.kg 1 Hartmann‘s solution

10 ml.kg 1 Hartmann‘s solution, then at 4 ml.min 1

Intravenous preload

Phenylephrine 0.25–4 mg.h 1 (n = 30), or ephedrine 7.5–120 mg.h 1 (n = 30)

Phenylephrine 0.13–4 mg.h 1 (n = 48), or ephedrine 4– 120 mg.h 1 (n = 50), or phenylephrine 0.067–4 mg.h 1 and ephedrine 2–60 mg.h 1 (n = 49)

Rate reduced if SBP increased to 105–110% of baseline, infusion stopped if SBP > 120% of baseline

Intervention

Automated oscillometric 1-min intervals

Automated oscillometric 1-min intervals

Blood pressure monitoring

SBP < 80% baseline

SBP < 75% baseline

Hypotension

Definitions

SBP > 120% of baseline

SBP > 125% of baseline

Hypertension

Up to 20 min after spinal

Before delivery

Observation period

Anaesthesia 2014, 69, 143–165 Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia

© 2013 The Association of Anaesthetists of Great Britain and Ireland

Inclusion: ASA 1 or 2, elective section, term, uncomplicated, singleton pregnancies, < 100 kg Exclusion: cardiac, pulmonary or renal diseases, or systemic diseases that could influence haemodynamic responses, including preeclampsia, hypertension, and diabetes; taking or had a history of taking any medications that could influence haemodynamic responses, including magnesium sulphate, terbutaline, or ß-blockers

Inclusion: ASA 1 or 2, elective section, > 18 years, 50– 120 kg, 150– 180 cm, normal singleton pregnancy, > 36 weeks’ gestation

Saravanan et al. 2006 [24]

Exclusion: not detailed

Loughrey et al. 2005 [23]

Study

Inclusion/exclusion criteria

Table 3. (Continued)

Saline 0.9% 500 ml infusion commenced after iv access was established

CSE: 13 mg hyperbaric bupivacaine 0.5%, 400 lg spinal diamorphine

1

10 ml.kg lactated Ringer‘s solution within 15 min

Spinal: 12 mg hyperbaric bupivacaine 0.75%, 10 lg fentanyl, 200 lg morphine; sitting position

Spinal anaesthesia

Intravenous preload

© 2013 The Association of Anaesthetists of Great Britain and Ireland

Up-down titration; ephedrine increased by 5 mg when preceding dose ineffective, decreased by 5 mg when preceding dose effective

Ephedrine 10 mg (n = 20) with or without phenylephrine 40 lg (n = 20)

Intervention

ND

ND

SBP < 75% of baseline or < 100 mmHg Non-invasive

Hypertension

SBP ≤ 80% of baseline or < 100 mmHg

Hypotension

Definitions

Automated non-invasive, 1-min intervals until delivery

Blood pressure monitoring

Before delivery, or 30 min after spinal, whichever was earlier

ND

Observation period

Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia Anaesthesia 2014, 69, 143–165

155

156

Inclusion: ASA 1 or 2, elective section, singleton pregnancy > 36 weeks’ gestation, no fetal abnormality, no history of preeclampsia or diabetes mellitus Exclusion: not detailed

Inclusion: ASA 1 or 2, elective section, term singleton pregnancies Exclusion: hypertension (SBP > 140 mmHg or diastolic BP > 90 mmHg), cardiovascular or cerebrovascular

Ngan Kee et al. 2008 [26]

Exclusion: pregnancyinduced hypertension, history of diabetes, cardiovascular and cerebrovascular problems, fetal abnormalities and contraindications to spinal anaesthesia

Cooper et al. 2007 [25]

Study

Inclusion/exclusion criteria

Table 3. (Continued)

Spinal: 10 mg hyperbaric bupivacaine 0.5%, 15 lg fentanyl, right lateral

Spinal: 14 mg hyperbaric bupivacaine 0.5%, 400 lg diamorphine

Spinal anaesthesia

Phenylephrine 0.25–4 mg.h 1, or ephedrine 12–180 mg.h 1 (n = 27 for both)

Phenylephrine 6 mg.h 1 (n = 24), or phenylephrine 4.5 mg.h 1 with ephedrine 120 mg.h 1 (n = 24), or phenylephrine 3 mg.h 1 with ephedrine 240 mg.h 1 No prehydration, warmed lactated Ringers‘s solution (maximum 2 l) with spinal

(n = 40), or phenylephrine increased by 50 lg when preceding dose ineffective, decreased by 50 lg when preceding dose effective (n = 40)

Intervention

10 ml.kg 1 Hartmann‘s solution after spinal injection

Intravenous preload

Non-invasive, 1-min intervals

Automated oscillometric 3-min intervals

Blood pressure monitoring

SBP < 80% of baseline

SBP < 80% of baseline

Hypotension

Definitions

SBP > 120% of baseline

SBP > 120% of baseline

Hypertension

Before delivery

ND

Observation period

Anaesthesia 2014, 69, 143–165 Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia

© 2013 The Association of Anaesthetists of Great Britain and Ireland

© 2013 The Association of Anaesthetists of Great Britain and Ireland

Hennebry et al. 2009 [27]

Ngan Kee et al. 2009 [29]

Study

Inclusion: ASA 1 or 2, elective section, term singleton pregnancies Exclusion: hypertension (SBP > 140 mmHg or diastolic BP > 90 mmHg), cardiovascular or cerebrovascular disease, known fetal abnormality, contra-indications to spinal anaesthesia, signs of onset of labour Inclusion: ASA 1 or 2, elective section, 50– 120 kg, 150– 180 cm, normal

disease, known fetal abnormality, contraindications to spinal anaesthesia, signs of onset of labour

Inclusion/exclusion criteria

Table 3. (Continued)

CSE up-down titration: hyperbaric bupivacaine increased by

Spinal: 10 mg hyperbaric bupivacaine 0.5%, 15 lg fentanyl, right lateral position

Spinal anaesthesia

Hartmann‘s solution 500 ml commenced shortly after

No prehydration, warmed lactated Ringers‘s solution (maximum 2 l) with spinal

Intravenous preload

Phenylephrine 1 mg.h 1 (n = 31), or ephedrine

Phenylephrine 6 mg.h 1 (n = 52), or ephedrine 240 mg.h 1 (n = 52): infusion rate constant for 2 min and continued unless SBP > 120% of baseline

(n = 25), or phenylephrine 1.5 mg.h 1 with ephedrine 360 mg.h 1 (n = 24), or ephedrine 480 mg.h 1 (n = 25): infusion rate constant for 2 min and then continued unless SBP > 120% of baseline

Intervention

SBP > 120% of baseline

ND

SBP < 80% of baseline Non-invasive, 1-min intervals until delivery,

Hypertension

SBP < 80% of baseline

Hypotension

Definitions

Non-invasive, 1-min intervals

Blood pressure monitoring

Before delivery

Before delivery

Observation period

Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia Anaesthesia 2014, 69, 143–165

157

158

~es Magalha et al. 2009 [28]

Study

Inclusion: ASA 1 or 2, term pregnancy of a single fetus, and indication for section, no information whether elective or urgent Exclusion: refusal to participate in the study, < 18 years, pre-existing or pregnancy-induced hypertension, presence of cardiovascular or cerebrovascular diseases, fetal abnormalities, history of allergy to study drugs,

singleton pregnancy > 37 weeks’ gestation Exclusion: pregnancyinduced hypertension, history of diabetes mellitus, cardiovascular or cerebrovascular problems, fetal abnormalities, and contraindications to spinal anaesthesia

Inclusion/exclusion criteria

Table 3. (Continued)

Spinal: 10 mg hyperbaric bupivacaine 0.5%, 3 lg sufentanil, left lateral decubitus

1 mg if preceding dose ineffective, decreased by 1 mg if preceding dose effective, all with 25 lg fentanyl, sitting then left lateral

Spinal anaesthesia

Immediately after spinal injection 2 l Ringer‘s lactate

iv access was established

Intravenous preload 1

Ephedrine 10-mg bolus (n = 30), or phenylephrine 80-lg bolus (n = 30)

90 mg.h (n = 31)

Intervention

Non-invasive

then 5-min intervals

Blood pressure monitoring

SBP < 80% of baseline

Hypotension

Definitions

SBP > 120% of baseline

Hypertension

ND

Observation period

Anaesthesia 2014, 69, 143–165 Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia

© 2013 The Association of Anaesthetists of Great Britain and Ireland

Inclusion: ASA 1 or 2, elective section, term, uncomplicated, singleton pregnancy Exclusion: pregnancy-induced hypertension, cardiovascular disease, cerebrovascular disease, placental or fetal abnormalities, absolute or relative contra-indications to spinal anaesthesia and with systolic blood pressure < 100 mmHg

Inclusion: ASA 1 or 2, elective section, 50– 90 kg, 145–165 cm, uncomplicated singleton pregnancy > 36 weeks’ gestation Exclusion: fetal malpresentation, pregnancy-induced hypertension, hypertension, cardiac disease, renal disease, fetal anomaly, diabetes mellitus or chronic medication

Das et al. 2011 [32]

contra-indications to spinal

Mohta et al. 2010 [30]

Study

Inclusion/exclusion criteria

Table 3. (Continued)

Phenylephrine increased to 3 mg.h 1 (n = 30), or mephentermine to 36 mg.h 1 (n = 30), if SBP > baseline, rate reduced by aliquots of 10% maximum

Phenylephrine 4 mg.h 1 (n = 31), or ephedrine 120 mg.h 1 (n = 29), or phenylephrine 2 mg.h 1 and ephedrine 60 mg.h 1 (n = 33) Rate reduced if SBP > 110% of baseline, or stopped if SBP > 120% of baseline

10 ml.kg 1 Hartmann‘s solution over 10– 15 min, then 4 ml. kg 1.h 1

15 ml.kg 1 lactated Ringer‘s solution over 15 min

Spinal: 12.5 mg hyperbaric bupivacaine over 10–15 s, left lateral position

Intervention

Spinal: 11 mg hyperbaric bupivacaine 0.5%, 10 mg hyperbaric bupivacaine 0.5% if < 150 cm

Spinal anaesthesia

Intravenous preload

© 2013 The Association of Anaesthetists of Great Britain and Ireland

Non-invasive, 1-min intervals for 10 min, then every 2 min

not detailed

Blood pressure monitoring

SBP > 120% of baseline

ND

SBP < 80% of baseline

Hypertension

SBP ≤ 80% of baseline or ≤ 100 mmHg

Hypotension

Definitions

ND

Before cord clamp

Before delivery

Observation period

Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia Anaesthesia 2014, 69, 143–165

159

Anaesthesia 2014, 69, 143–165

Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia

SBP > 120% of baseline Spinal: 12.5 mg hyperbaric bupivacaine 0.5%, left lateral position

1

None; isotonic saline at 7 ml.kg 1.h

Spinal anaesthesia

Bhardwaj et al. 2013 [14]

Study

Inclusion: ASA 1 or 2, elective section, singleton pregnancy without known fetal abnormalities, preeclampsia, or cerebrovascular disease Exclusion: not detailed

Intravenous preload Inclusion/exclusion criteria

Table 3. (Continued) 160

CSE, combined spinal-epidural analgesia; iv, intravenous; ND, not defined; SBP, systolic blood pressure.

Non-invasive, 1-min intervals

SBP < 80% of baseline

Discussion

Ephedrine 5 mg then 150 mg.h 1 (n = 26), or phenylephrine 30 lg then 0.9 mg.h 1 (n = 32), or metaraminol 0.5 mg then 15 mg.h 1 (n = 27)

Hypotension Intervention

Blood pressure monitoring

Definitions

Hypertension

Observation period

Ephedrine, metaraminol and phenylephrine were given as a bolus followed by infusion after spinal injection in the report by Bhardwaj et al. [14]. Hypertension occurred more frequently in the metaraminol group (10/27) than in the ephedrine (6/26) or phenylephrine (3/32) groups. There were no differences in other maternal or fetal outcomes.

We found that prophylactic phenylephrine infusion reduced the rate of maternal hypotension after spinal anaesthesia for caesarean section, both before and after birth. We would have preferred to choose as our primary outcome an event of direct significance to mother or neonate, but we were limited by the outcomes most commonly reported by the included studies. The incidence of hypotension depends on the definition used, increasing with higher thresholds, without one knowing whether or by how much important morbidity would be associated with each threshold [2]. Most of the studies in our review used a value of 80% of the baseline as the threshold, with five studies [4, 12, 19, 21, 24] using a different value. We therefore think that this may not be an important source of between-study heterogeneity. A probable source of heterogeneity, which limits our review, was the different rates of phenylephrine infusion, which varied between 25 and 120 lg.min 1 [15, 16]. Because of between-study heterogeneity, the conclusions of our review have to be tentative. We used the random-effects model for meta-analysis because of this inter-study heterogeneity. An alternative approach would have been not to pool results and only perform qualitative information synthesis. Ilies et al. [35] recently detected more hypotensive episodes with a continuous non-invasive blood pressure monitor compared with interval measurements, but it remains to be shown whether this could be used to reduce the rate of hypotension, perhaps by means of automated closed-loop systems [36, 37]. In contrast to prophylactic infusions, these systems only administer a vasopressor when the blood pressure falls. Phenylephrine also reduced the rate of postoperative nausea and vomiting compared with placebo. © 2013 The Association of Anaesthetists of Great Britain and Ireland

Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia

Anaesthesia 2014, 69, 143–165

Figure 2 Forest plot for the rate of hypotension during caesarean section before fetal delivery: prophylactic phenylephrine (PE) infusion vs placebo. M-H, Mantel-Haenszel; Random, random-effects model.

Figure 3 Forest plot for the rate of hypotension during caesarean section after fetal delivery: prophylactic phenylephrine (PE) infusion vs placebo. M-H, Mantel-Haenszel; Random, random-effects model.

However, we pooled results from only two studies [11, 16], so future studies could easily change this finding. We did not pool results from Allen et al. [15] who scored the severity of intra-operative nausea and vomiting: in that study the group receiving the highest dose of phenylephrine (100 lg.min 1) did not report nausea. A major limitation of our study is the paucity of available data. Studies, participants and events were few, certainly less than the two hundred or more events suggested to make results reliable [33]. Large, single randomised controlled trials have occasionally disagreed with meta-analyses of small studies that preceded them [34]. We chose not to pool data from studies more prone to bias, as we were concerned that there was a greater chance of generating misleading results. The doses of bupivacaine used in the studies that we pooled were 10 mg [11, 13], 12 mg [15] and 14 mg [25], which are more likely to induce hypotension, nausea and vomiting, compared with doses less than 9 mg [38]. In a previous review, we found that a © 2013 The Association of Anaesthetists of Great Britain and Ireland

drop in blood pressure below 80% of baseline was induced in half of the parturients who had had a low-dose bupivacaine spinal anaesthetic [2]. Half of the prophylactic vasopressor infusions would thus be given to patients whose blood pressure would not fall this low, but who would be exposed to possible adverse effects. Prophylactic phenylephrine might harm by inducing hypertension [5], with half the participants in one study being categorised as hypertensive [7]. We did not calculate any differences in rates of hypertension, but as mentioned above, this was based on few events. Takotsubo cardiomyopathy has been observed after hypertension (200/130 mmHg) and tachycardia (170 min 1) was induced in a parturient by phenylephrine (two doses of 50 lg) and atropine (0.5 mg) [39]. Atropine is often co-administered with phenylephrine to avoid bradycardia. Ventricular arrhythmia has also been associated with phenylephrine infusion [40]. Ephedrine has long been favoured as the vasopressor of first choice for the treatment of spinal-induced 161

Anaesthesia 2014, 69, 143–165

Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia

Table 4 Additional maternal and neonatal outcomes after prophylactic phenylephrine infusion: (a) prophylactic phenylephrine infusion vs no prophylaxis; (b) prophylactic phenylephrine infusion vs prophylactic ephedrine infusion; and (c) prophylactic infusion of a combination of phenylephrine and ephedrine vs prophylactic ephedrine infusion. Number of patients Outcomes

Phenylephrine

Control

RR (95% CI)

111 152 151 111 121 56

50 91 90 50 60 54

0.37 1.38 2.00 1.89 0.30 0.39

70 141

70 89

WMD (95% CI)

References

(a) Maternal Post-delivery hypotension Bradycardia Pre-delivery hypertension Post-delivery hypertension Vomiting Nausea or vomiting Fetal Uterine artery pH Uterine artery base excess

(0.19–0.71) (0.71–2.66) (0.36–11.17) (0.15–3.45) (0.06–1.55) (0.17–0.91)

[15, 16] [15, 16] [9, 15, 16] [15, 16] [9, 15] [7, 16] 0.01 ( 0.01 to 0.02) 0.25 ( 0.31 to 0.81)

[12, 16] [12, 15, 16]

WMD (95% CI)

References

Number of patients Outcomes

Phenylephrine

Ephedrine

RR (95% CI)

99 99 30

100 100 30

0.97 (0.74–1.27) 5.3 (0.9–30.2) 0.54 (0.35–0.85)

55 55 15 15 69 69

53 53 13 13 70 70

(b) Maternal Pre-delivery hypertension Pre-delivery bradycardia Pre-delivery tachycardia Fetal Umbilical artery pH Umbilical artery BE Umbilical vein pH Umbilical vein BE Apgar ≥ 7 at 1 min Apgar ≥ 7 at 5 min

[22, 26, 29] [22, 26, 29] [22] 0.05 1.85 0.04 2.20

(0.02–0.08) (0.56–3.15) ( 0.03 to 0.11) (0.29–4.11)

0.99 (0.94–1.03) 1.00 (0.96–1.04)

[24, 27] [24, 27] [27] [27] [26, 29] [26, 29]

Number of patients Maternal outcomes

Combination

Ephedrine

RR (95% CI)

References

101 52 33 82

99 49 29 79

0.73 0.68 1.32 0.73

[20, 21, 32] [20, 32] [32] [21, 32]

(c) Pre-delivery hypotension Pre-delivery hypertension Pre-delivery bradycardia Nausea

(0.55–0.96) (0.43–1.07) (0.24–7.35) (0.29–1.86)

RR, risk ratio; WMD, weighted mean differences.

hypotension during caesarean section. Compared with placebo, prophylactic ephedrine reduced the risk of hypotension to 0.73 (0.63–0.86) in one meta-analysis [41]. We found that phenylephrine reduced the risk of hypotension compared with ephedrine to 0.58 (0.39–0.88), although this result was sensitive to the inclusion of other data. One might have expected a greater risk reduction when the combination of phenylephrine and ephedrine was compared with ephedrine alone, but we calculated a less impressive relative risk of 0.73 (0.55–0.96), the interpretation of 162

which is hampered by different doses in different studies. In summary, we found limited evidence that prophylactic phenylephrine infusion reduced the risk of hypotension induced by spinal bupivacaine, before and after caesarean delivery, as well as the risk of nausea and vomiting. It remains to be determined whether the same effects will be seen with lower doses of bupivacaine. We found no differences for other maternal haemodynamic variables or clinically relevant fetal outcomes. In view of the paucity of data, this may well be due to © 2013 The Association of Anaesthetists of Great Britain and Ireland

Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia

Anaesthesia 2014, 69, 143–165

Figure 4 Forest plot for the rate of hypotension during caesarean section before fetal delivery: prophylactic phenylephrine (PE) infusion vs ephedrine (E) infusion. M-H, Mantel-Haenszel; Random, random-effects model.

Figure 5 Forest plot for the rate of hypotension during caesarean section before fetal delivery: prophylactic ephedrine (E) infusion with or without phenylephrine (PE). inf, infusion; M-H, Mantel-Haenszel; Random, randomeffects model. the absence of relevant evidence, rather than any evidence of the absence of an effect. More evidence is therefore needed to inform clinical practice. A large double-blind randomised controlled trial with sufficient power would help, with an emphasis on important maternal and neonatal outcomes, rather than blood pressure.

References 1. Cooper DW. Caesarean delivery vasopressor management. Current Opinion in Anesthesiology 2012; 25: 300–8. €hr S, Roth R, Hofmann T, Rossaint R, Heesen M. Definitions 2. Klo of hypotension after spinal anaesthesia for caesarean section: literature search and application to parturients. Acta Anaesthesiologica Scandinavica 2010; 54: 909–21. €hr S, Rossaint R, Heesen M. 3. Veeser M, Hofmann T, Roth R, Klo Vasopressors for the management of hypotension after spinal anesthesia for elective caesarean section. Systematic review and cumulative meta-analysis. Acta Anaesthesiologica Scandinavica 2012; 56: 810–6. n AI. 4. Alahuhta S, R€ as€anen J, Jouppila P, Jouppila R, Hollme Ephedrine and phenylephrine for avoiding maternal hypotension due to spinal anaesthesia for caesarean section. Effects on uteroplacental and fetal haemodynamics. International Journal of Obstetric Anesthesia 1992; 1: 129–34. © 2013 The Association of Anaesthetists of Great Britain and Ireland

5. Beilin Y. The treatment should not be worse than the disease. Anesthesiology 2006; 104: 1348–9. 6. Cooper D, Schofield L, Hynd R, et al. Prospective evaluation of systolic arterial pressure control with a phenylephrine infusion regimen during spinal anaesthesia for caesarean section. International Journal of Obstetric Anesthesia 2012; 2: 245–52. 7. 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. 8. Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011], Chapter 7.7. http://handbook.cochrane.org/ (accessed 10/ 09/2013). 9. Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and variance from the median, range, and the size of a sample. BMC Medical Research Methodology 2005; 5: 13. 10. Moher D, Liberati A, Tetzlaff J, Altman DG. The PRISMA Group (2009). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA Statement. British Medical Journal 2009; 339: b2535. 11. Ngan Kee WD, Khaw KS, Ng FF, Lee BB. Prophylactic phenylephrine infusion for preventing hypotension during spinal anesthesia for cesarean delivery. Anesthesia and Analgesia 2004; 98: 815–21. 12. Langesaeter E, Rosseland LA, Stubhaug A. Continuous invasive blood pressure and cardiac output monitoring during cesarean delivery: a randomised, double-blind comparison of low-dose versus high-dose spinal anesthesia with intravenous phenyl163

Anaesthesia 2014, 69, 143–165

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

164

ephrine or placebo infusion. Anesthesiology 2008; 109: 856– 63. das Neves JF, Monteiro GA, de Almeida JR, Sant’Anna RS, Bonin HB, Macedo CF. Phenylephrine for blood pressure control in elective cesarean section: therapeutic versus prophylactic doses. Revista Brasileira de Anestesiologica 2010; 60: 391–8. Bhardwaj N, Jain K, Arora S, Bharti N. A comparison of three vasopressors for tight control of maternal blood pressure during cesarean section under spinal anesthesia: effect on maternal and fetal outcome. Journal of Anaesthesiology Clinical Pharmacology 2013; 29: 26–31. Allen TK, George RB, White WD, Muir HA, Habib AS. A doubleblind, placebo-controlled trial of four fixed rate infusion regimens of phenylephrine for hemodynamic support during spinal anesthesia for cesarean delivery. Anesthesia and Analgesia 2010; 111: 1221–9. Doherty A, Ohashi Y, Downey K, Carvalho JC. Phenylephrine infusion versus bolus regimens during cesarean delivery under spinal anesthesia: a double-blind randomised clinical trial to assess hemodynamic changes. Anesthesia and Analgesia 2012; 115: 1343–50. Moran DH, Perillo M, LaPorta RF, Bader AM, Datta S. Phenylephrine in the prevention of hypotension following spinal anesthesia for cesarean delivery. Journal of Clinical Anesthesia 1991; 3: 301–5. Hall PA, Bennett A, Wilkes MP, Lewis M. Spinal anaesthesia for caesarean section: comparison of infusions of phenylephrine and ephedrine. British Journal of Anaesthesia 1994; 73: 471–4. Ayorinde BT, Buczkowski P, Brown J, Shah J, Buggy DJ. Evaluation of pre-emptive intramuscular phenylephrine and ephedrine for reduction of spinal anaesthesia-induced hypotension during caesarean section. British Journal of Anaesthesia 2001; 86: 372–6. Mercier FJ, Riley ET, Frederickson WL, Roger-Christoph S, Benhamou D, Cohen SE. Phenylephrine added to prophylactic ephedrine infusion during spinal anesthesia for elective cesarean section. Anesthesiology 2001; 95: 668–74. 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. Cooper DW, Jeyaraj L, Hynd R, et al. Evidence that intravenous vasopressors can affect rostral spread of spinal anesthesia in pregnancy. Anesthesiology 2004; 101: 28–33. Loughrey JP, Yao N, Datta S, Segal S, Pian-Smith M, Tsen LC. Hemodynamic effects of spinal anesthesia and simultaneous intravenous bolus of combined phenylephrine and ephedrine versus ephedrine for cesarean delivery. International Journal of Obstetric Anesthesia 2005; 14: 43–7. 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. British Journal of Anaesthesia 2006; 96: 95–9. Cooper DW, Gibb SC, Meek T, et al. Effect of intravenous vasopressor on spread of spinal anaesthesia and fetal acid-base equilibrium. British Journal of Anaesthesia 2007; 98: 649–56. Ngan Kee WD, Lee A, Khaw KS, Ng FF, Karmakar MK, Gin T. A randomised double-blinded comparison of phenylephrine and ephedrine infusion combinations to maintain blood pressure during spinal anesthesia for cesarean delivery: the effects on fetal acid-base status and hemodynamic control. Anesthesia and Analgesia 2008; 107: 1295–302.

Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia 27. Hennebry MC, Stocks GM, Belavadi P, et al. Effect of i.v. phenylephrine or ephedrine on the ED50 of intrathecal bupivacaine with fentanyl for caesarean section. British Journal of Anaesthesia 2009; 102: 806–11. ^ia CS, de Arau jo Ladeira LC, Nascimento 28. Magalh~aes E, Gove BG, Kluthcouski SM. Ephedrine versus phenylephrine: prevention of hypotension during spinal block for cesarean section and effects on the fetus. Revista Brasileira de Anestesiologica 2009; 59: 11–20. 29. Ngan Kee WD, Khaw KS, Tan PE, Ng FF, Karmakar MK. Placental transfer and fetal metabolic effects of phenylephrine and ephedrine during spinal anesthesia for cesarean delivery. Anesthesiology 2009; 111: 506–12. 30. Mohta M, Janani SS, Sethi AK, Agarwal D, Tyagi A. Comparison of phenylephrine hydrochloride and mephentermine sulphate for prevention of post spinal hypotension. Anaesthesia 2010; 65: 1200–5. 31. Stewart A, Fernando R, McDonald S, Hignett R, Jones T, Columb M. The dose-dependent effects of phenylephrine for elective cesarean delivery under spinal anesthesia. Anesthesia and Analgesia 2010; 111: 1230–7. 32. Das S, Mukhopadhyay S, Mandal M, Mandal S, Basu SR. A comparative study of infusions of phenylephrine, ephedrine and phenylephrine plus ephedrine on maternal haemodynamics in elective caesarean section. Indian Journal of Anaesthesia 2011; 55: 578–83. r MR, Collins SL, McQuay HJ. 33. Moore RA, Gavaghan D, Trame Size is everything–large amounts of information are needed to overcome random effects in estimating direction and magnitude of treatment effects. Pain 1998; 78: 209–16. 34. LeLorier J, Gregoire G, Benhaddad A, Lapierre J, Derderian F. Discrepancies between meta-analyses and subsequent large randomised, controlled trials. New England Journal of Medicine 1997; 337: 536–42. 35. Ilies C, Kiskalt H, Siedenhans D, et al. Detection of hypotension during caesarean section with continuous non-invasive arterial pressure device or intermittent oscillometric arterial pressure measurement. British Journal of Anaesthesia 2012; 109: 413–9. 36. Ngan Kee WD, Khaw KS, Ng FF, Tam YH. Randomized comparison of closed-loop feedback computer-controlled with manual-controlled infusion of phenylephrine for maintaining arterial pressure during spinal anaesthesia for caesarean delivery. British Journal of Anaesthesia 2013; 110: 59–65. 37. Sia AT, Tan HS, Sng BL. Closed-loop double-vasopressor automated system to treat hypotension during spinal anaesthesia for caesarean section: a preliminary study. Anaesthesia 2012; 67: 1348–55. 38. Arzola C, Wieczorek PM. Efficacy of low-dose bupivacaine in spinal anaesthesia for caesarean delivery: systematic review and meta-analysis. British Journal of Anaesthesia 2011; 107: 308–18. 39. Zdanowicz JA, Utz AC, Bernasconi I, Geier S, Corti R, Beinder E. ‘‘Broken heart’’ after cesarean delivery. Case report and review of literature. Archives of Gynecology and Obstetrics 2011; 283: 687–94. 40. Lai FM, Jenkins JG. Ventricular bigeminy during phenylephrine infusion used to maintain normotension during caesarean section under spinal anaesthesia. International Journal of Obstetric Anesthesia 2007; 16: 288–90. 41. Lee A, Ngan Kee WD, Gin T. Prophylactic ephedrine prevents hypotension during spinal anesthesia for Cesarean delivery but does not improve neonatal outcome: a quantitative systematic review. Canadian Journal of Anesthesia 2002; 49: 588–99. © 2013 The Association of Anaesthetists of Great Britain and Ireland

Heesen et al. | Prophylactic phenylephrine for spinal anaesthesia

Anaesthesia 2014, 69, 143–165

Appendix Terms used to search electronic bibliographies ((cesarean section OR c section OR surgical delivery OR operative delivery) AND (vasoconstrictor agents OR vasopressor OR phenylephrine) AND (spinal anesthesia OR anesthesia, spinal OR spinal analgesia OR subarachnoid anesthesia OR intrathecal anesthesia OR intrathecal analgesia OR subarachnoid analgesia OR epidural anesthesia OR anesthesia, epidural OR epidural analgesia OR peridural analgesia OR peridural anesthesia OR combined spinalepidural anesthesia)).

© 2013 The Association of Anaesthetists of Great Britain and Ireland

165