Neurocritical Care

, Volume 12, Issue 1, pp 17–23

Phenylephrine but not Ephedrine Reduces Frontal Lobe Oxygenation Following Anesthesia-Induced Hypotension

  • Peter Nissen
  • Patrice Brassard
  • Thomas B. Jørgensen
  • Niels H. Secher
Original Article

DOI: 10.1007/s12028-009-9313-x

Cite this article as:
Nissen, P., Brassard, P., Jørgensen, T.B. et al. Neurocrit Care (2010) 12: 17. doi:10.1007/s12028-009-9313-x



Vasopressor agents are used to correct anesthesia-induced hypotension. We describe the effect of phenylephrine and ephedrine on frontal lobe oxygenation (ScO2) following anesthesia-induced hypotension.


Following induction of anesthesia by fentanyl (0.15 mg kg−1) and propofol (2.0 mg kg−1), 13 patients received phenylephrine (0.1 mg iv) and 12 patients received ephedrine (10 mg iv) to restore mean arterial pressure (MAP). Heart rate (HR), MAP, stroke volume (SV), cardiac output (CO), and frontal lobe oxygenation (ScO2) were registered.


Induction of anesthesia was followed by a decrease in MAP, HR, SV, and CO concomitant with an elevation in ScO2. After administration of phenylephrine, MAP increased (51 ± 12 to 81 ± 13 mmHg; P < 0.001; mean ± SD). However, a 14% (from 70 ± 8% to 60 ± 7%) reduction in ScO2 (P < 0.05) followed with no change in CO (3.7 ± 1.1 to 3.4 ± 0.9 l min−1). The administration of ephedrine led to a similar increase in MAP (53 ± 9 to 79 ± 8 mmHg; P < 0.001), restored CO (3.2 ± 1.2 to 5.0 ± 1.3 l min−1), and preserved ScO2.


The utilization of phenylephrine to correct hypotension induced by anesthesia has a negative impact on ScO2 while ephedrine maintains frontal lobe oxygenation potentially related to an increase in CO.


Cerebral autoregulation Cardiac output Arterial pressure Near infrared spectroscopy Drug effect 



Cerebral blood flow


Cardiac output


Heart rate


Mean arterial pressure


Near infrared spectroscopy


Frontal lobe cerebral oxygenation


Stroke volume


Induction of anesthesia may lead to a decrement in arterial pressure that could be considered below the lower limit of cerebral autoregulation. Cerebral autoregulation refers to the intrinsic ability of the cerebral vasculature to maintain cerebral blood flow (CBF) relatively stable within a range of mean arterial pressure (MAP) from 60 to 150 mmHg [1]. Outside this range of MAP, CBF is considered to decrease with a low MAP and to rise with a high MAP. Accordingly, when MAP decreases below 60 mmHg following the induction of anesthesia, an elevation in MAP may be indicated to secure CBF and, in turn, frontal lobe oxygenation (ScO2). ScO2 monitored by near infrared spectroscopy (NIRS) detects effective cerebral autoregulation for septic patients [2] and patients undergoing liver transplantation [3]. However, the lower level of cerebral autoregulation is highly variable, i.e., ScO2 may be maintained at a MAP of 42 mmHg in some patients, while for other patients, ScO2 decreases when MAP become lower than 90 mmHg. The preservation of ScO2 during surgery is of importance since a reduction in jugular venous oxygen saturation [4] or ScO2 [5] has been associated with early postoperative neurological dysfunction in patients undergoing cardiopulmonary bypass surgery.

One effective means of restoring MAP after induction of anesthesia is provided by the administration of vasopressor agents, such as phenylephrine and ephedrine. Phenylephrine is a selective α-adrenergic receptor agonist, and the increase in MAP is accounted for by an elevation in total peripheral resistance although it may not affect CBF [6]. Still if the cerebral vasculature has a significant sympathetic innervation [7], it may be that phenylephrine induces cerebral vascular constriction in parallel with its effect on other vascular beds [8]. Ephedrine is a sympathomimetic drug, which stimulates both β- and α-adrenergic receptors directly and indirectly by endogenous release of norepinephrine with an elevation in MAP, cardiac output (CO), and stroke volume (SV).

It remains unknown whether the use of vasopressor agents during anesthesia influences CBF and, in turn, ScO2. The utilization of norepinephrine, which is an α-agonist, reduces middle cerebral artery mean flow velocity and cerebral oxygenation characterized by ScO2 and jugular venous oxygen saturation in normotensive healthy subjects [9]. Phenylephrine has also been shown to markedly increase cerebral perfusion pressure and intracranial pressure without an effect on cerebral oxygenation [10]. The present report describes the ScO2 response to the elevation of MAP by routine use of phenylephrine in patients who underwent elective surgery and experienced anesthesia-induced hypotension. Correction of anesthesia-induced hypotension was then shifted to the administration of ephedrine to evaluate whether that drug affects ScO2 when it increases MAP. Accordingly, we describe the experience of utilizing both phenylephrine and ephedrine in regard to ScO2 when anesthesia-induced hypotension is corrected.


Twenty-five female patients (phenylephrine group: 59 ± 10 years old; 74 ± 18 kg; 1.64 ± 0.07 m, and body mass index: 28 ± 6 kg/m2; ephedrine group: 53 ± 19 years old, 71 ± 16 kg, 1.67 ± 0.05 m and 25 ± 5 kg/m2; mean ± SD), who underwent elective mastectomy, thyroidectomy, or parathyroidectomy and experienced hypotension following the induction of anesthesia, were eligible to participate in this prospective clinical study. Inclusion criteria were age >18 years old; ASA score I or II and a reduction of MAP below 60 mmHg following the induction of anesthesia. Exclusion criteria included preoperative treatment for hypertension, previous treatment with bleomycin and raised values of plasma bilirubin [11]. A total number of 53 patients who underwent elective mastectomy, thyroidectomy, and parathyroidectomy provided informed consent to the study as approved by the Ethics Committee of Copenhagen (01-012/02) in accordance with the Declaration of Helsinki.


Following the patient’s arrival in the operating room, MAP, SV, CO, heart rate (HR), and ScO2 were recorded before anesthesia and reported also following placement of an endotracheal tube or a laryngeal mask after induction of anesthesia and with the administration of the vasoactive drugs. After anesthesia, a bolus of phenylephrine (0.1 mg) was administered to 13 patients as their MAP had decreased to a value considered below the lower limit of cerebral autoregulation (60 mmHg). The strategy to correct anesthesia-induced hypotension changed and ephedrine was administered (10 mg) to 12 patients to evaluate its impact on ScO2. All measurements were obtained before the beginning of surgery. The specific periods for which data are presented were 1) baseline, 2) time of administration of the drugs, 3) highest MAP achieved after drug administration, and 4) lowest ScO2 measured within 10 min following the highest MAP.

Induction of Anesthesia

Before the induction of anesthesia, the patient’s legs were elevated by ~10 cm to maintain central blood volume, and an intravenous infusion of lactated Ringer was started [12] and maintained at 3 ml kg−1 h−1 throughout anesthesia. Anesthesia was induced by fentanyl (0.15 mg) and propofol (2.0 mg kg−1) and maintained with a propofol infusion of 25–33 μg kg−1 min−1. For 23 patients undergoing mastectomy, a laryngeal mask was used, while one patient undergoing mastectomy and a patient undergoing thyroidectomy were orally intubated after neuromuscular blockade with cisatracurium (1.5 mg kg−1). Controlled ventilation was established and adjusted to an end-tidal carbon dioxide tension of 3.5–4.5 kPa and the ventilation was not changed during the study period and, therefore, the end-tidal carbon dioxide tension did not change. Also, oxygen (O2) and atmospheric air were mixed to ascertain an inspired O2 fraction of 0.7 [13]. The arterial O2 saturation was monitored by a finger pulse oximeter and kept >97%.

Hemodynamic Monitoring

Hemodynamic monitoring included an ECG for HR, MAP, SV, and CO. MAP was determined by a Finometer device (Finapres Medical Systems, Amsterdam, The Netherlands) with the cuff applied at the third finger of the left hand in patients who underwent thyroidectomy and parathyroidectomy, and from the opposite hand to the operation side in patients who underwent a mastectomy. The Finometer calculates SV from the arterial pressure wave according to the Modelflow method [14]. This method uses a non-linear three-element model of the aortic input impedance and simulates aortic flow waveforms from a peripheral arterial pressure signal. Two of the three model elements, i.e., aortic characteristic impedance and arterial compliance depend on the aorta’s elastic properties and are computed using a built-in database of arctangent aortic pressure–area relationships given age, height, weight, and gender. Integrating the aortic flow waveform per beat provides left ventricular SV, and CO is computed from SV and HR. The third model element, peripheral vascular resistance, is calculated for each heartbeat as the quotient of arterial pressure and the modeled flow. The software used was an online real-time version of Beatscope® (FMS, Amsterdam, The Netherlands). This derived CO has been successfully validated against a thermodilution estimate of CO during a reduction in central blood volume induced by head-up tilt in healthy subjects [15], during cardiac and liver transplantation surgery [16, 17] as well as in intensive care medicine [18].

Frontal Lobe Oxygenation

The ScO2 was monitored by NIRS (INVOS Cerebral Oximeter, Somanetics, Troy, MI) with optodes attached to the forehead above the frontal sinuses. Changes in ScO2 parallel those in internal jugular venous O2 saturation and middle cerebral artery mean flow velocity [19] and NIRS detects cerebral hypoperfusion during surgery [20]. The NIRS determined ScO2 is based on the absorption of light in the spectra for oxygenated and deoxygenated hemoglobin and reports ScO2 as a percentage of light absorption by oxygenated to total hemoglobin. An emitter generates light at 733 and 808 nm, and the reflection is registered by two optodes placed at a distance of 3 and 4 cm from the emitter. This placement of the optodes allows for the subtraction of reflections derived from superficial tissues of the scalp and the skull from ScO2 [21]. With increasing distance between the emitter and the optodes, light penetrates deeper into the tissues and with evaluation of absorption at two distances (spatial resolution), absorption in deep tissue, i.e., in the frontal lobe is appreciated. Thus, values reported for ScO2 account predominantly for hemoglobin oxygenation in the frontal lobe cortex.

Statistical Analysis

Power calculations revealed that to detect an expected reduction in NIRS-determined ScO2 of 10 ± 10% (mean reduction ± SD), a sample size of n = 10 would be sufficient (alpha level 0.05 and statistical power >80%). A one-way analysis of variance on repeated measures evaluated variables over time for data normally distributed. All pairwise multiple comparison procedures were performed by the Holm-Sidak method. A Friedman repeated measures analysis of variance on ranks evaluated parameters over time for data not normally distributed. All pairwise multiple comparison procedures on ranks were performed by the Dunn’s method. A Student’s paired t test was used to evaluate changes between conditions. Results are presented as mean ± SD for data normally distributed and median (range) for data not normally distributed and a value of P < 0.05 was considered statistically significant. Data were analyzed by the software package Sigmastat (SPSS, Chicago, IL).


In accordance with the inclusion criteria for the study, the induction of anesthesia reduced MAP from 90 ± 19 to 51 ± 12 mmHg in the phenylephrine group and from 91 ± 18 to 53 ± 9 mmHg in the ephedrine group. The ScO2 increased from 68 (57–82)% to 70 (57–85)% in the phenylephrine group and from 70 (63–79)% to 75 (67–84)% in the ephedrine group, reaching statistical significance only in the ephedrine group (Table 1). CO declined in both groups (phenylephrine group: from 5.3(3.6–7.4) to 3.7(1.5–4.9) l min−1; P < 0.005; ephedrine group: from 5.2(1.6–8.9) to 3.2(2.3–6.2) l min−1; P < 0.001), when compared against their own baseline. Induction of anesthesia significantly lowered HR and SV, but had no influence on total peripheral resistance in the phenylephrine group. In the ephedrine group, induction of anesthesia lowered HR while neither SV nor total peripheral resistance was affected (Table 1).
Table 1

Hemodynamic variables and frontal lobe oxygenation at baseline and with phenylephrine and ephedrine administration following hypotension induced by anesthesia



Administration of drug

Highest MAP

Lowest CO


 MAP (mmHg)

91 ± 16

51 ± 12*

81 ± 13*

62 ± 13*‡††

 Cardiac output (l min−1)

5.3 (3.6–7.4)

3.7 (1.5–4.9)

3.2 (1.9–4.8)


 Heart rate (bpm)

70 ± 10

53 ± 8*

47 ± 11*§

48 ± 7*§

 Stroke volume (ml)

75 (58–120)

56 (40–120)

64 (46–131)

70 (53–119)

 Total peripheral resistance (mmHg min−1)

18 (13–27)

16 (7–28)

24 (14–49)†§


 Frontal lobe oxygenation (%)

68 (57–82)

70 (57–85)

70 (52–82)

60(52–76) **§


 MAP (mmHg)

91 ± 18

53 ± 9*

79 ± 8‡†

69 ± 12*§**

 Cardiac output (l min−1)

5.2 (1.6–8.9)

3.2 (2.3–6.2)*

5.0 (3.4–7.1)

4.1 (3.2–6.7)

 Heart rate (bpm)

71 (58–87)

60 (41–75)*

64 (41–81)

63 (44–81)

 Stroke volume (ml)

75 (23–114)

62 (32–107)

88 (54–107)§

74 (51–107)§

 Total peripheral resistance (mmHg min−1)

18 (10–33)

18 (8–21)

16 (10–21)

16 (9–23)

 Frontal lobe oxygenation (%)

70 (63–79)

75 (67–84)*

73 (70–81)

72 (67–81)§

Data are median ± SD for data normally distributed and median(range) for data not normally distributed MAP: mean arterial pressure

* P < 0.001 vs. baseline; † P < 0.05 vs. baseline; ‡ P < 0.001 vs. administration of drug; § P < 0.05 vs. administration of drug; ** P < 0.05 vs. highest MAP; †† P < 0.001 vs. highest MAP

Effects of Phenylephrine Administration

Phenylephrine was effective at increasing MAP to above what is considered the lower limit of cerebral autoregulation (51 ± 12 to 81 ± 13 mmHg; P < 0.001) with the highest MAP following its administration after 41 (30–83) s (Fig. 1). However, a 14% reduction [70(50–82) to 60(52–76)%, P < 0.05] in ScO2 followed the administration of phenylephrine, while there was no change in CO, a decrease in HR (53 ± 8 to 47 ± 11 bpm) and an increase in total peripheral resistance (Table 1).
Fig. 1

Heart rate, mean arterial pressure, cardiac output, and frontal lobe oxygenation responses to phenylephrine or ephedrine administration following hypotension induced by anesthesia. * P < 0.001 vs. baseline, † P < 0.05 vs. baseline, ‡ P < 0.001 vs. administration of drug, § P < 0.05 vs. administration of drug, ** P < 0.05 vs. highest MAP, †† P < 0.001 vs. highest MAP

Effects of Ephedrine Administration

The administration of ephedrine led to an increase in MAP from 53 ± 9 to 79 ± 8 mmHg (P < 0.001) and the time to achieve the highest MAP was 83 (21–134) s (Fig. 1; Table 1). The administration of ephedrine restored CO, but had no influence on HR or total peripheral resistance and ScO2 was preserved.


The results suggest that the use of phenylephrine to increase MAP in patients with hypotension induced by induction of anesthesia has a negative impact on frontal lobe and presumably cerebral oxygenation as total peripheral resistance increases. After having been convinced that phenylephrine had a negative impact on ScO2, we evaluated the effect of ephedrine following anesthesia-induced hypotension on ScO2. In contrast to phenylephrine, the increase in MAP in response to the administration of ephedrine had no negative impact on ScO2, potentially related to an increase in CO.

Anesthesia impacts arterial pressure by a depressive effect on sympathetic tone and alters baroreflex control of arterial pressure [22] although, in the present study, hypotension was also induced by a lowering in CO supporting previous findings [3]. Even if changes in MAP are well tolerated in healthy patients, it may not always be the case in patients with cardiovascular risk factors. One priority in anesthesia can be maintenance of MAP to ensure adequate perfusion pressure to vital organs. In regards to cerebral perfusion, if MAP is reduced below what is considered to be the lower limit of cerebral autoregulation, CBF would be expected to vary passively with a further decrease in MAP, exposing the patient to possible cerebral hypoperfusion and eventually a reduction in ScO2. However, this study corroborates the observation that the induction of anesthesia is not followed by a reduction in ScO2 in patients experiencing hypotension [23]. In contrast, there is an immediate reduction in CBF and ScO2 if MAP drops below ~80 mmHg in response to a reduction in the central blood volume, e.g., during head-up tilt [24] or in response to hemorrhage during surgery [25]. In light of the preserved ScO2 following the induction of anesthesia, it may be questioned why MAP should be increased. We acknowledge that in some clinical populations including patients with a history of cerebrovascular disease, renal dysfunction, peripheral claudications, hypertension, or hypovolaemia, the maintenance of hypotension during anesthesia is contraindicated. However, we suggest that the increase in MAP is not a prerequisite to secure ScO2, and in some patients hypotensive anesthesia appears to be without an effect on cerebral oxygenation.

The two vasoactive agents used were effective in correcting the hypotension induced by anesthesia. However, the use of phenylephrine led to a 14% reduction in ScO2 following the maximal vasoconstrictive effect of the drug (Fig. 1), while it was administered to provide for an increment, or at least the maintenance of MAP and thereby ScO2. A 50% reduction in middle cerebral artery mean flow velocity and a 10–15% reduction in ScO2 are associated with pre-syncopal symptoms [24, 25, 26, 27] and gravity-induced loss of consciousness is reported at an ~15% decrease in ScO2 [28]. Accordingly, the 14% reduction in ScO2 following the infusion of phenylephrine is not trivial and the ability of NIRS to detect clinical meaningful cerebral hypoperfusion has been shown in several studies [3, 29, 30].

A reduction in CO induced by the administration of phenylephrine has been documented in both animals and humans [31, 32]. Also, a reduced ability to increase CO during exercise induces peripheral vasoconstriction not only in working skeletal muscles [33] but also in the brain [34]. In addition, significant linear relationships exist between CO and middle cerebral artery mean flow velocity both at rest and during exercise [35]. The ability to increase cerebral perfusion during exercise becomes limited in patients in whom CO does not increase such as in patients with atrial fibrillation and heart failure [36]. In contrast, CBF and ScO2 increase during exercise in healthy subjects [37] and the ability of CO to increase in response to an elevation in MAP by pharmacological means following hypotension induced by anesthesia could be of importance in the maintenance of CBF and ScO2. Besides phenylephrine per se may provoke vasoconstriction in the cerebral vasculature. Although phenylephrine increases cerebral vascular tone in humans, it can increase [38] or has no influence [39] on cerebral vascular resistance. The use of phenylephrine increases CBF in cardiopulmonary bypass patients [40], healthy subjects [38], and anesthetized patients [41]. However, in other situations, CBF seems unaffected by phenylephrine [6]. The present findings, along with evidence from lower body negative pressure, head-up tilt, and hemorrhage suggest that the lower limit of cerebral autoregulation depends on the central blood volume and/or the ability to increase CO, through the modulation of sympathetic nervous activity.

This study was not randomized nor was it designed to compare the effect of phenylephrine and ephedrine on ScO2 when administered to correct hypotension induced by anesthesia. Still a double-blind, randomized controlled trial would be unlikely to provide a different result since there was no exception to the negative effect of phenylephrine on ScO2. There is a small probability that the absence of a reduction in ScO2 in the ephedrine group can be ascribed to too few patients included in the study. Yet, we consider that any reduction in ScO2 during surgery should be avoided; although we acknowledge that the significance of a transient reduction in ScO2 on postoperative neurological dysfunction is unlikely to be significant considering that a vasovagal syncope is not regularly associated with such sequelae. Another limitation of this study is that we did not measure CBF but rather oxygenation of the brain. Under circumstances ranging from hypovolemic shock [24] to maximal exercise [42] and in septic patients [2], there is a parallel variation in ScO2 and middle cerebral artery mean flow velocity in basal cerebral arteries and it may be considered that an important role for CBF is to provide adequate oxygenation for the brain. It may further be considered that the individual variation in the NIRS derived ScO2 was large. We accept that the initial value is somewhat arbitrary since we did not know which frontal lobe vessels were evaluated. However, we consider changes in ScO2 important. Further, the use of NIRS for ScO2 could be problematic with interventions having the potential to influence venous pressure, the relative amount of blood in the cerebral arteries, capillaries and veins, as well as the contribution of scalp versus cerebral blood. Indeed, changes in SsO2 may eventually be accounted for by, at least partially, the modulation of the variables. Also, cerebral venous blood pressure is of interest since phenylephrine might induce venoconstriction and, in turn, a decrease in CBF velocity and ScO2 [43]. Finally, the use of methodology based upon pulse wave analysis, to calculate CO could be problematic during interventions which influence systemic vascular impedance, but the Modelflow has been validated under such circumstances [18].


The results suggest that the use of phenylephrine to correct hypotension induced by anesthesia has a negative impact on ScO2, while it does not affect CO despite an effective elevation in MAP. If there is a possibility to choose between phenylephrine and ephedrine to correct hypotension induced by anesthesia, ephedrine is the drug that preserves ScO2.

Copyright information

© Springer Science+Business Media, LLC 2009

Authors and Affiliations

  • Peter Nissen
    • 1
  • Patrice Brassard
    • 1
  • Thomas B. Jørgensen
    • 1
  • Niels H. Secher
    • 1
  1. 1.Department of AnesthesiaRigshospitalet 2041, University of CopenhagenCopenhagen ØDenmark

Personalised recommendations