Spinal anesthesia for cesarean delivery is frequently complicated by hypotension with a reported incidence of up to 80%.1 Spinal anesthesia has been associated with a greater risk of fetal acidosis during cesarean delivery compared with epidural and general anesthesia.2,3
Vasopressors are the mainstay for the prevention and treatment of hypotension associated with spinal anesthesia. Phenylephrine and ephedrine are routinely used to treat spinal anesthesia-induced hypotension. Fetal outcomes assessed by umbilical arterial and venous pH are improved, and maternal nausea and vomiting reduced if maternal systolic blood pressure (SBP) is maintained at its baseline level with phenylephrine compared with ephedrine.1,4,5 However, the optimum dose of phenylephrine for the treatment of hypotension has not been determined. In this double-blind, up-down study, we sought to estimate the 90% effective dose (ED90) of IV phenylephrine for the treatment of hypotension associated with spinal anesthesia for cesarean delivery.
Institutional research ethics board approval was obtained. The trial was registered at www.clinicaltrials.gov (NCT00781157). Adult (age ≥18 yr), ASA physical status I and II, nonlaboring women undergoing an elective cesarean delivery with spinal anesthesia at term with singleton gestation (37–42 wk) were recruited. Exclusion criteria were morbid obesity (body mass index ≥45 kg/m2), height <152 cm, history of severe hypertensive disease of pregnancy (SBP >160 mm Hg, diastolic blood pressure >110 mm Hg, and/or requirement for antihypertensive treatment or presence of significant proteinuria), and significant maternal cardiac disease or diabetes.
After obtaining written informed consent and before entering the operating room, subjects had their arterial blood pressure measured 3 times (every 2 min), supine with left uterine displacement. The mean of the 3 SBP readings was deemed the baseline SBP. Hypotension was defined as SBP <20% of baseline or SBP <90 mm Hg.
All women received spinal anesthetic administered between L3 and L5 in the sitting position using hyperbaric bupivacaine 12 mg, fentanyl 15 μg, and preservative-free morphine 100 μg. Subjects were immediately laid supine with left uterine displacement. Each subject received an IV preload and coload of lactated Ringer's solution (approximately 15–20 mL/kg) with the aim of administering 500 mL before spinal anesthesia and 2 L before delivery. After the spinal anesthetic procedure was completed, arterial blood pressure was measured every minute for 10 min and then every 2.5 min for the duration of the study. If the study medication was administered, arterial blood pressure was measured every minute until SBP returned to within 20% of baseline.
If the SBP decreased >20% of baseline or to an SBP <90 mm Hg, a 5-mL syringe with saline and a predetermined dose of phenylephrine was administered. An anesthesiologist not involved with patient recruitment, anesthetic care, or data collection prepared the coded syringes of phenylephrine. If the SBP returned to within 20% of baseline or ≥90 mm Hg within 1 min, treatment was considered a success. Any resultant hypertension (SBP >20% above baseline) was noted but not considered a failure. If hypotension persisted after 1 min, the anesthesiologist treated the hypotension with a vasopressor of his or her choice. This was recorded as a failure. The study concluded with the response to the blinded phenylephrine bolus or delivery, whichever occurred first.
The initial dose of phenylephrine was 100 μg. This dose was chosen based on our clinical experience and statistical simulation at various doses. Each subsequent dose was based on the response of the preceding subject, as per a biased-coin design up-down sequential method.6 The dosing changes were in increments of 20 μg. The anesthesia provider was blinded to the dose of phenylephrine, as was the subject. If a failure was observed in the previous subject, the dose was stepped up in the next subject. If a success was observed, the next subject was randomized with probability of 0.1 to the next lower dose and with probability of 0.9 to the same dose.* If no hypotension occurred before the birth of the neonate, the subject was withdrawn and the next subject was assigned the same dose.
The ED90 with 95% confidence intervals (CIs) and estimated probabilities were calculated using maximum likelihood estimation (MLE) and Firth logistic regression with penalized MLE using Minitab 15 (Minitab, State College, PA) and LogXact 8.0 (Cytel, Cambridge, MA).7 A sample size of at least 40 patients was selected after testing a variety of scenarios, each with a million simulations of both the responses and the corresponding doses selected by the sequential allocation method described above, and beginning with various starting doses.
Sixty-nine subjects were screened to participate in this study (Fig. 1). Three subjects were excluded before consent. Sixty-six subjects consented; however, 1 subject was withdrawn before the spinal anesthetic because she received atropine to treat a vasovagal response to local infiltration in preparation for the anesthetic procedure. Of the 65 subjects who completed the trial, 20 (31%) did not experience hypotension and were withdrawn from the study.
Forty-five enrolled subjects (69%) experienced spinal anesthesia-induced hypotension and received phenylephrine per study protocol. There was no difference in age, weight, height, crystalloid volume, or sensory level between women who became hypotensive and received phenylephrine compared with those who did not (Table 1).
Those subjects who developed hypotension received doses of phenylephrine ranging between 80 and 180 μg with failures noted in all dosages except for 180 μg (Fig. 2). Of the 45 subjects who experienced hypotension and received a bolus of phenylephrine, the mean reduction in SBP was 25% ± 7% from the baseline SBP. No subjects experienced hypertension (SBP >20% above baseline) after receiving the allocated phenylephrine dose. The time (mean ± sd) from spinal anesthesia administration to hypotension was 5.8 ± 3.6 min. Determined with the MLE method with raw data, the ED90 of phenylephrine was 147 μg (95% CI, 98–222 μg) (Fig. 3). Using Firth regression, the probability of a successful response at 150 μg is 90.5% (95% CI, 66.0%–99.0%) (Fig. 4).
Two subjects who received phenylephrine boluses for hypotension required treatment of bradycardia. The first subject was a treatment failure (phenylephrine 140 μg) and received atropine for a heart rate of 55 bpm and hypotension. The subject's arterial blood pressure returned within 20% of baseline after receiving atropine with no subsequent hypertension. The second subject was a treatment success (phenylephrine 160 μg) and received glycopyrrolate for a heart rate of 49 bpm without hypotension. There was no reactive hypertension after the administration of glycopyrrolate.
In this study, we found that the ED90 of phenylephrine bolus required to reverse hypotension induced by spinal anesthesia in cesarean delivery is 147 μg. This dose is 50% more than our suspected results and what is typically reported in clinical use. Although the 95% CIs span over 100 μg, it is worth noting that our current clinical dose is at the bottom end of this CI. As evidenced by lack of maternal hypertension and significant bradycardia, our study agrees with previous study results suggesting that doses of phenylephrine in this range are unlikely to have any negative impact on neonatal or maternal outcomes.8 Recently, Tanaka et al.9 estimated the 95% effective dose (ED95) of a phenylephrine bolus to prevent spinal anesthesia-induced hypotension and nausea to be 135 and 159 μg, respectively. The results of Tanaka et al.9 and those of our trial suggest that the doses of phenylephrine required to prevent and treat hypotension are higher than previously suspected, suggesting that further evaluation of phenylephrine in this dose range is warranted.
Hypotension during spinal anesthesia is common, and nonpharmacologic measures to combat hypotension are frequently not effective.10–12 Vasopressors such as ephedrine and phenylephrine are frequently required whether or not aggressive fluid management is part of prophylaxis of spinal anesthesia-induced hypotension. Ephedrine is no more effective than phenylephrine and may result in a lower umbilical cord pH compared with phenylephrine.4,5,13,14 However, the increased fetal oxygen consumption and excretion of carbon dioxide observed by Ngan Kee et al.4 as the dose of ephedrine increased may reflect a direct fetal effect of ephedrine increasing fetal metabolism. Phenylephrine, perhaps because of preferential maternal splanchnic vasoconstriction over uterine artery vasoconstriction, may increase maternal cardiac output and improve uteroplacental circulation.15
Earlier up-down methodology studies sought the ED50 of various drugs, e.g., the dose of inhaled anesthetics that by definition is effective in 50% of subjects. Anesthesiologists were left to extrapolate the ED50 to more clinically useful doses, i.e., ED90 or ED95 that would be effective in 90%–95% of patients. The biased-coin design allows the researchers to set the quantile effect dose of interest, i.e., 90%, 95%, or even 50% if they choose. Our preliminary simulations, designed to guide our project, led us to the conclusion that the ED90 would give us reasonably precise results. Although striving for the ED95 may be more clinically relevant, we believed this would sacrifice precision of our estimate. Despite this, we do not want to overstate our precision. Table 2 shows results from our data using isotonic regression estimation and the pooled-adjacent-violators algorithm approach. This demonstrates breaches of monotonicity (i.e., as the dose increases, the drug effect increases) that can occur with such data. These approaches can be used to provide better precision and narrower CIs for point estimates. However, given the unpredictability of the degree of hypotension associated with the initiation of spinal anesthesia for cesarean delivery (i.e., wide variance in change of SBP from baseline), such natural variability in response is not surprising. Therefore, we prefer to use the more conservative approach and emphasize the inherent uncertainty by reporting with wider precision.
Our results may have been influenced by our decision to use the 1-min mark to judge the effectiveness of phenylephrine. The peak action of centrally administered phenylephrine is 30–45 s,16 and therefore we believe that the 1-min interval for peripherally administered phenylephrine was an appropriate interval. Additionally, in this clinical scenario, if hypotension has not resolved within 1 min, most obstetrical anesthesiologists would typically opt for alternate or additional therapy. Furthermore, we note in our study that those subjects who failed treatment with the allocated phenylephrine dose and received additional therapy did not have any subsequent hypertension. This suggests that they likely required a higher dose than the allocated study treatment dose and truly would have failed treatment no matter how long we waited. This study concluded with delivery of the neonate, and therefore the results are applicable to the hypotension associated with the evolution of the spinal blockade in pregnant women with high thoracic neuroblockade. The study also was restricted to the treatment of the first episode of hypotension. Therefore, these results may not be generalizable to recurrent episodes of hypotension that may occur at a later stage of the anesthetic when the block has stabilized, or to hypotension associated with oxytocin administration or blood loss.
In conclusion, the estimated ED90 of phenylephrine for the treatment of hypotension after spinal anesthesia for cesarean delivery is approximately 150 μg. Recognizing that the 95% CI of the ED90 is between 98 and 220 μg, our traditional starting dose of 100 μg may be reasonable; however, efficacy may be improved by starting with a higher dose. Based on our results, further evaluations of phenylephrine in this dose range are warranted.
1. 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
2. Roberts SW, Leveno KJ, Sidawi JE, Lucas MJ, Kelly MA. Fetal acidemia associated with regional anesthesia for elective cesarean delivery. Obstet Gynecol 1995;85:79–83
3. Mueller MD, Bruhwiler H, Schupfer GK, Luscher KP. Higher rate of fetal acidemia after regional anesthesia for elective cesarean delivery. Obstet Gynecol 1997;90:131–4
4. Ngan Kee WD, Lee A, Khaw KS, Ng FF, Karmakar MK, Gin T. A randomized 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. Anesth Analg 2008;107:1295–302
5. 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
6. Pace NL, Stylianou MP. Advances in and limitations of up-and-down methodology: a precis of clinical use, study design, and dose estimation in anesthesia research. Anesthesiology 2007; 107:144–52
7. Firth D. Bias reduction of maximum likelihood estimates. Biometrika 1993;80:27–38
8. 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
9. Tanaka M, Balki M, Parkes RK, Carvalho JC. ED95 of phenylephrine to prevent spinal-induced hypotension and/or nausea at elective cesarean delivery. Int J Obstet Anesth 2009;18:125–30
10. Rocke DA, Rout CC. Volume preloading, spinal hypotension and caesarean section. Br J Anaesth 1995;75:257–9
11. Rout CC, Rocke DA. Prevention of hypotension following spinal anesthesia for cesarean section. Int Anesthesiol Clin 1994;32:117–35
12. Rout CC, Rocke DA, Levin J, Gouws E, Reddy D. A reevaluation of the role of crystalloid preload in the prevention of hypotension associated with spinal anesthesia for elective cesarean section. Anesthesiology 1993;79:262–9
13. 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
14. Reynolds F, Seed PT. Anaesthesia for caesarean section and neonatal acid-base status: a meta-analysis. Anaesthesia 2005;60: 636–53
15. Vallejo MC, Ramanathan S. Should alpha-agonists be used as first line management of spinal hypotension? Int J Obstet Anesth 2003;12:243–5
16. Goertz AW, Seeling W, Heinrich H, Lindner KH, Rockemann MG, Georgieff M. Effect of phenylephrine bolus administration on left ventricular function during high thoracic and lumbar epidural anesthesia combined with general anesthesia. Anesth Analg 1993;76:541–5
*The probability P = (1 − γ)/γ. For γ = 0.9, then P = 0.1/0.9, thus if a positive response is observed, then the dose for the next patient is assigned at probability 1/9 to the next lower dose and probability 8/9 to the same dose. These values were rounded to the nearest tenth, i.e., 0.1 and 0.9.