Secondary Logo

Journal Logo

Obstetric anaesthesia

Comparison of double intravenous vasopressor automated system using nexfin versus manual vasopressor bolus administration for maintenance of haemodynamic stability during spinal anaesthesia for caesarean delivery

A randomised double-blind controlled trial

Sng, Ban Leong; Du, Wei; Lee, Man Xin; Ithnin, Farida; Mathur, Deepak; Leong, Wan Ling; Sultana, Rehena; Han, Nian-Lin R.; Sia, Alex Tiong Heng

Author Information
European Journal of Anaesthesiology: May 2018 - Volume 35 - Issue 5 - p 390-397
doi: 10.1097/EJA.0000000000000779
  • Free



Maternal hypotension is a common adverse effect after spinal anaesthesia during caesarean section, occurring in up to 90% of cases if not prevented.1 This occurs due to sympathetic blockade from spinal anaesthesia leading to decreased vascular resistance and increased venous pooling, and it is exacerbated by supine hypotension syndrome and increased maternal sensitivity to local anaesthetics during pregnancy.2,3 The consequences of maternal hypotension include maternal side effects (nausea, vomiting, tissue hypoperfusion) and foetal adverse effects (foetal acidosis, foetal hypoxia).4

Various strategies have been attempted to prevent and manage maternal hypotension, such as the use of intravenous fluids and vasopressors. Phenylephrine has been widely accepted as the first-line therapy owing to its selective α1-receptor agonist property that causes vasoconstriction and increases systemic vascular resistance.5 Ephedrine is a mixed α and β agonist commonly used in maternal hypotension with concomitant bradycardia.2 However, none of these strategies have been proven to eliminate hypotension6 and the optimal timing and dose of vasopressors is still debatable. There are rapid fluctuations in blood pressure (BP) during spinal anaesthesia for caesarean section and standard. Oscillometric BP monitoring has the limitation of being an intermittent measurement that may not be sensitive enough to guide vasopressor administration. Furthermore, the large inter-individual variability in the severity of BP changes and their response to treatment make it difficult to determine the optimal intervention, often resulting in under- or overtreatment.

We have previously described the safety and efficacy of our updated close-loop double vasopressor automated system in achieving better haemodynamic stability during spinal anaesthesia for caesarean section.7–9 A closed-loop double-vasopressor automated system that administered phenylephrine or ephedrine according to continuous noninvasive arterial pressure (CNAP) monitoring was first developed in 2010 and our preliminary results were published in 2012.7 The system was designed to overcome the possibility of reactive bradycardia which can occur with the use of phenyl-ephrine alone. Subsequently, an enhanced control algorithm was evolved with a reduced time interval for the administration of phenylephrine boluses.8 Our randomised trial showed that the augmented algorithm achieved better systolic arterial pressure control, and a reduced incidence of nausea compared with women who received manual vasopressor boluses (MVBs), without a difference in the incidence of reactive hypertension or total vasopressor dose.8 A further development integrates beat-to-beat SBP data from a continuous monitor, Nexfin (BMEYE B.V, Amsterdam, the Netherlands), and automatically administers vasopressors based on a novel algorithm.9 The Nexfin device measures finger arterial pressure directly from a finger cuff using a volume-clamp method, and reconstructs brachial arterial pressure using wave filtering with pressure level correction.10 Unlike the CNAP system,11,12 this method eliminates the need for recalibration and provides both CNAP and cardiac output (CO) monitoring. The accuracy and precision of the BP and CO measurement by Nexfin is comparable with invasive intra-arterial BP monitoring and transcardiopulmonary thermodilution respectively.12,13 Our previous study showed that the automated vasopressor delivery system was able to respond efficiently to rapid changes in BP and tailor vasopressor administration based on the haemodynamic profile.7 We used phenylephrine as the first-line vasopressor for hypotension, and ephedrine is used to treat hypotension with concomitant bradycardia.5 A bolus regimen for vasopressor delivery was chosen as a comparator, as it was shown to maintain maternal BP closer to baseline and required a lower total dose of phenylephrine to maintain BP compared with an infusion regimen.13

The goal of this study is to individualise therapy and provide a tighter control of maternal BP by reducing the incidence of maternal hypotension (defined as SBP <80% of baseline) and reactive hypertension (defined as SBP >120% of baseline). In this double-blind, randomised controlled trial (RCT), we compared the performance and reliability of the double intravenous vasopressor automated (DIVA) system against MVB administration. Maternal and neonatal outcomes after spinal anaesthesia were also evaluated.

Materials and methods

This study was approved by the SingHealth Centralised Institutional Review Board (CIRB reference: 2013/322/D), and registered in (NCT02277730). We recruited women with American Society of Anesthesiologists physical status 1 and 2, with singleton full-term pregnancies presenting for elective caesarean section under spinal anaesthesia. Inclusion criteria were age 21 to 45 years, weight 40 to 90 kg and height 145 to 170 cm. The exclusion criteria were women with contraindications to spinal anaesthesia, allergy to drugs used in the study, obstetric complications (preeclampsia, placenta previa) or uncontrolled medical conditions (hypertension, diabetes mellitus, cardiovascular disease).

The patients were screened at a preanaesthetic clinic and written information was provided. The day before surgery (or the same day) informed consent was obtained. On the day of caesarean section, patients were randomly allocated to two groups using a computer-generated code that was stored in serial sealed opaque envelopes prepared by a biostatistician not involved in treatment allocation. All patients received sodium citrate preoperatively. The patient rested supine with left uterine displacement for 15 min in a quiet room. Baseline BP was measured using an oscillometric device on the right arm and calculated by taking the average values of three consecutive measurements at 1-min intervals. Intravenous access was established on the forearm using an 18-gauge cannula. The Nexfin finger cuff was placed on the right second or middle finger, and a pulse oximeter probe on a finger of the left hand. Electrocardiographic monitoring was applied. Spinal anaesthesia was performed in the sitting position using a 27-G pencil-point needle (BD Medical, Franklin Lakes, New Jersey, USA). After ensuring free flow of cerebrospinal fluid, 11 mg hyperbaric bupivacaine 0.5% with 15 μg fentanyl and 100 μg morphine was injected over 15 s with the orifice of the spinal needle facing cephalad. The patient was then positioned supine with a left lateral table tilt of about 15° and intravenous cohydration with 1000 ml Hartmann's solution using a pressure bag was started.

In the DIVA group, phenylephrine (100 μg ml−1) and ephedrine (8 mg ml−1) were prepared in separate 50 ml syringes connected via fine-bore extension tubes to the intravenous cannula by three-way stopcocks. SBP and heart rate (HR) measured by the Nexfin were uploaded continuously to a computer via a National Instruments data acquisition card; this integrated the data every 10 s (LabVIEW running on Windows XP operating system; Microsoft Corporation, Redmond, Washington, USA). When SBP fell below baseline, commands were sent to activate either of two syringe driver pumps (B. Braun, Melsungen, Germany). When SBP fell between 90 and 100% of baseline, a 25 μg phenylephrine bolus was administered, or a 2 mg ephedrine bolus was administered if HR was less than 60 beats min−1 concomitantly. If SBP was less than 90% of baseline, a 50 μg phenylephrine bolus was administered instead, or a 4 mg ephedrine bolus was administered if the HR was less than 60 beats min−1 concomitantly (Fig. 1). It takes more than 10 s for each of the two vasopressors to take effect14 so we programmed a 10-s lockout period after the bolus was administered. The system was monitored by the investigator, who would manually administer atropine, additional phenylephrine or ephedrine if SBP remained less than 70% baseline for more than 3 min.

Fig. 1
Fig. 1:
Schematic diagram of the algorithm used in the DIVA group.

In the MVB group, vasopressor delivery was via two separate syringes of phenylephrine and ephedrine. SBP measurements at 1-min intervals using Nexfin were provided to the blinded investigator, who manually administered phenylephrine or ephedrine boluses at a maximum of one bolus every minute according to a set algorithm (Fig. 2). If a patient was assigned to the DIVA group, active vasopressors were delivered via the DIVA system and 0.9% saline placebo were delivered as MVB and vice versa. The unblinded investigators prepared all medications.

Fig. 2
Fig. 2:
Schematic diagram of the algorithm used in the MVB group.

The sensory block level was measured 5 min after spinal anaesthesia by assessing the loss of sensation to cold. Motor block at 5 min was measured using the modified Bromage scale. The time of spinal injection, surgical incision and delivery were recorded. We recorded continuous SBP, HR and the number of vasopressor boluses in a data file. We also recorded nausea or vomiting, the total amount of intravenous fluids given up to delivery and any technical problems during surgery. The attending midwife or neonatologist assessed the Apgar scores at 1 and 5 min after delivery and neonatal birth weight, umbilical cord blood pH and lactate concentration were measured.

Statistical analysis

Our primary outcome was the incidence of maternal hypotension. A sample size of 236 (118 in each group) was required to show that the difference in incidence of hypotension (any beat to beat systolic pressure reading <80% of baseline) of 23% (MVB group 58.9% and DIVA group 36%) in this double-blinded, two-arm, RCT based on the following assumptions: 1 : 1 allocation ratio, α = 0.05, power = 90%, 10% drop-out rate.15

The secondary outcomes were reactive hypertension, total vasopressor requirement, maternal outcomes (nausea, vomiting, bradycardia) and neonatal outcomes (Apgar scores at 1 and 5 min, cord arterial and venous pH, cord arterial and venous lactate concentration).

All categorical data were represented as frequency (proportion). Continuous data were expressed as mean (± standard deviation [SD]) or median (interquartile range (IQR) [range]), as appropriate. Categorical data were compared using Fisher's exact test and continuous data were compared using a two-sample independent t-test or Mann–Whitney U test, whichever was applicable. A P value less than 0.05 was considered as statistically significant. We assessed the performance of our system using measures previously employed on similar closed-loop systems,16,17 adjusted with a pooled-data approach that gives additional consideration to the wide variations in the number of measurements taken for each patient.

Percentage performance error

The performance error was defined as the percentage difference between each measured value of systolic pressure from the baseline. The performance error for the ith patient at the jth second was calculated as follows:

Median absolute performance error

The median absolute performance error (MDAPE) is a measure of inaccuracy and indicates the absolute magnitudes of the differences between measured and baseline BPs. For each patient, it was defined as the median of the absolute values of performance error (|PE|) and was calculated as follows:

where Ni is the number of values of |PE| for the ith patient and M is the number of patients in the study.

Median performance error

The median performance error (MDPE) is a measure of bias and indicates whether the differences between measured BPs were systematically above or below baseline values. For each patient, it was defined as the median of performance error and was calculated as follows:


Wobble measures how much performance error fluctuates around the MDPE with time, for each patient (i.e. intra-individual performance error variability). It was calculated as follows:


Divergence is defined as the slope obtained from linear regression of each patient's |PE| against time, and describes the trend of changes in |PE| with time. It gives an indication of whether the accuracy of our system improves (negative divergence) or decreases (positive divergence) with time. Divergence (per minute) was calculated as follows:

where tij is the time of ith individual measurement, in min.

All calculations were performed using SAS version 9.3 software (SAS Institute, Cary, North Carolina, USA).


A total of 236 patients were recruited from July 2013 to June 2016 for the study. One patient was excluded in the DIVA group due to failed spinal with conversion to general anaesthesia. In the MVB group, there were two failed spinals with conversion to general anaesthesia and three with faulty finger cuffs. Overall, 230 patients completed the study. The CONSORT diagram is shown in Fig. 3.

Fig. 3
Fig. 3:
CONSORT diagram for study recruitment.

Patient baseline characteristics and haemodynamic parameters were similar between DIVA and MVB groups (Table 1). The intraoperative haemodynamic outcomes are summarized in Table 2. In the DIVA group, 83.1% of SBP readings fell within ± 20% variation from baseline values (95 400 out of 114 817), whereas 75% of those from the MVB group fell within ± 20% from baseline (98 403 out of 131 084). There was similar incidence of reactive hypertension in the two groups: in the DIVA group, 94 out of 117 (80.3%) patients had reactive hypertension, compared with 99 out of 113 (87.6%) patients in the MVB group (Table 2).

Table 1
Table 1:
Patient characteristics and baseline haemodynamic values in DIVA and MVB groups
Table 2
Table 2:
Haemodynamic outcomes measured during caesarean section under spinal anaesthesia in DIVA and MVB groups

Hypotension was present in 46 (39.3%) patients in the DIVA group and this was significantly less than in the MVB group with 65 (57.5%) (P = 0.008). These results included any beat-to-beat SBP less than 80% of baseline value, regardless of the duration of hypotension and hypertension. Hence, there is high sensitivity for detection and overestimation of hypotension and hypertension compared with the intermittent oscillometric technique used in most clinical settings. There was no significant difference in the highest SBP, lowest SBP, highest and lowest HR between the two groups.

In terms of system performance errors, the wobble was lower in the DIVA group than the MVB group (6.7 versus 8.2%), whereas all other performance measures were comparable between the two groups (Table 2). No additional atropine, phenylephrine or ephedrine was given by the investigators. In all cases of hypertension, the BP decreased spontaneously towards the baseline without additional interventions within 3 min.

The DIVA group had less nausea than the MVB group (11 versus 16), although the difference did not reach statistical significance. Other maternal outcomes such as vomiting and bradycardia were comparable across the groups (Table 3). Similarly, neonatal outcomes (Apgar score, umbilical cord pH and lactate levels) were comparable between the DIVA and MVB groups (Table 4). In the DIVA group, one neonate had an Apgar score of 7 at 1 min and one neonate in the MVB group with a score of 4. All neonates had Apgar scores of 8 or 9 at 5 min.

Table 3
Table 3:
Maternal outcomes in DIVA and MVB groups
Table 4
Table 4:
Neonatal outcomes in DIVA and MVB groups


We found that the updated DIVA system using the Nexfin with a two-step algorithm achieved significantly less number of hypotensive readings, lower incidence of maternal hypotension and lower wobble than the MVB group. Other maternal and neonatal outcomes were comparable between the two groups.

With our previously described automated CNAP system,8,9 we found that there was better control of BP compared with a MVB control group (hypotension in 34.9 versus 58.9% in the control group). However, as the study design administered vasopressors only if BP fell less than 90% of baseline, it might have resulted in a higher clinical incidence of hypotension than a regime of giving vasopressors whenever the SBP fell below baseline.18 Moreover, one technical limitation of the CNAP system was that it required recalibration at 15-min intervals, which might have resulted in missed readings.

To detect hypotension more accurately and to initiate earlier treatment, we refined our algorithm to allow a smaller dose of vasopressors to be administered when SBP is between 90 and 100% of baseline, with the aim to reduce further BP drop, whilst reducing the risk of causing reactive hypertension. We have demonstrated that our improved double-vasopressor system (DIVA) with the built-in noninvasive continuous haemodynamic monitoring device Nexfin was able to maintain maternal SBP close to the baseline.7 Similar to our original automated vasopressor delivery system, the new system required minimal intervention from the attending anaesthetist and resulted in satisfactory clinical maternal and neonatal outcomes.

The incidence of hypotension in this study (39.3%) was lower than that found in our pilot single-arm study (47.4%) evaluating the effectiveness of this updated vasopressor system7 but higher than that observed in the RCT of the original system using CNAP (34.9%).8 However, neither our pilot single-arm study using Nexfin nor previous RCT using CNAP showed a statistically significant difference in the incidence of hypotension using Fisher's exact test when compared with our present study (pilot single-arm study versus present study: odds ratio (OR) 0.72 (95% confidence interval, 95% CI [0.36 to 1.44]); P = 0.3969. RCT using CNAP versus present study: OR 1.13 (95% CI [0.68 to 1.87]); P = 0.6997). Despite using different devices, there was no significant difference in the incidence of hypotension.

The total dose of phenylephrine used by the two systems was not significantly different, although the DIVA group had slightly lower consumption. Despite this, we still observed a lower incidence of hypotension with the DIVA system. Therefore, the timely administration of vasopressor delivery using a continuous noninvasive BP monitor could be important in reducing maternal hypotension.

We chose a bolus regimen instead of infusion for vasopressor delivery in the control group, as the bolus regimen is known to maintain maternal BP closer to baseline during the initial minutes after spinal anaesthesia and required a lower total dose of phenylephrine to maintain BP than the infusion regime.13,19 das Neves et al.20 reported that prophylactic phenylephrine infusion had a lower incidence of hypotension (18 versus 33%), nausea (10 versus 15%) and vomiting (0 versus 8%) than a prophylactic bolus regimen. However, this study had limitations in design and statistical analysis that did not adequately address the testing between prophylactic infusion and prophylactic bolus groups. Therefore, the conclusion could be overstated. In a recent randomised trial, Ngan Kee et al.19 showed that the precision of BP control was greater in the bolus group when compared with the infusion, whereas phenylephrine consumption was smaller in the bolus group. There were no differences in hypotension, CO, nausea, vomiting or neonatal outcome between groups. Further studies are required to investigate the optimal dosing regimen for maintaining maternal BP.

We assessed the performance of our system using the same methods for evaluating similar closed-loop systems,16,17 with the modification of using a pooled-data method that assigns a weight to each BP measurement according to the total number of measurements obtained for each patient. The rationale behind this was to avoid over-representation of patients with fewer BP readings and under-representation of patients with more BP readings when taking the average of all BP measurements, owing to variability in SBP between patients and the very large number of BP readings obtained from each patient. This method allowed us to assign more weight to BP readings from patients in which our system performance was known to have produced a greater accuracy (i.e. in those with a greater number of BP measurements), and vice versa.16 The positive MDPE means that our system maintained SBP above baseline values by a median of 11.0%, with the wobble indicating that the SBP fluctuated around MDPE by 6.7%. The MDAPE variable shows that our system achieved SBP values that differed from baseline by a median of 6.1%. Compared with our original system, the MDAPE (6.1 versus 8.5%) and wobble (6.7 versus 6.3%) were all comparable, indicating that the DIVA system was relatively robust in maintaining tight BP control.

The usefulness of noninvasive haemodynamic parameters in reducing the risk of hypotension could be limited by the accuracy and precision of the monitoring device. Movement of patients and stress, anxiety or cold temperature leading to peripheral vasoconstriction might affect the accuracy of fingertip pulse contour analysis used by the Nexfin. However, the accuracy of BP measurement using the Nexfin has been validated against intra-arterial BP monitoring.11

In conclusion, when compared with the MVB group, we have shown that the DIVA system with continuous noninvasive monitoring is able to achieve a lower incidence of maternal hypotension, less hypotensive readings and less wobble, without higher vasopressor consumption. There were good maternal and foetal outcomes. Future research is needed to determine the optimal method and timing of vasopressor delivery in the management of obstetric patients undergoing spinal anaesthesia.

Acknowledgements relating to this article

Assistance with the study: the authors would like to thank the clinical research coordinators (Agnes Teo, Liu Juan, Yvonne Yong) and the staff of the major operating theatres at KK Women's and Children's Hospital, Singapore, for their support.

Financial support and sponsorship: we would like to acknowledge the SingHealth Foundation Clinical Trial Grant (Ref: SHF/CTG047/2012) and Centre Grant (NMRC/CG/006/2013) for the research funding.

Conflicts of interest: none.

Presentation: Oral presentation at the Australian and New Zealand College of Anaesthetists Conference in 2017, Brisbane, Australia.


1. Mercier FJ, Augé M, Hoffmann C, et al. Maternal hypotension during spinal anaesthesia for caesarean delivery. Minerva Anestesiol 2013; 79:62–73.
2. Dyer RA, Reed AR. Spinal anesthesia during elective cesarean delivery: closer to a solution. Anesth Analg 2010; 111:1093–1095.
3. Ngan Kee WD, Khaw KS. Vasopressors in obstetrics: what should we be using? Curr Opin Anaesthesiol 2006; 19:238–243.
4. Maayan-Metzger A, Schushan-Eisen I, Todris L, et al. Maternal hypotension during elective cesarean section and short-term neonatal outcome. Am J Obstet Gynecol 2010; 202:56.e1–56.e5.
5. Veeser M, Hoffman T, Roth R, et al. Vasopressors for the management of hypotension after spinal anaesthesia for elective caesarean section. Systematic review and cumulative meta-analysis. Acta Anaesthesiol Scand 2012; 56:810–816.
6. Cyna AM, Andrew M, Emmett RS, et al. Techniques for preventing hypotension during spinal anaesthesia for caesarean section. Cochrane Database Syst Rev 2006; CD002251.
7. Sia ATH, Tan HS, Sng BL. Closed-loop double-vasopressor automated system to treat hypotension during spinal anaesthesia for caesarean section: a preliminary study. Anaesthesia 2012; 67:1348–1355.
8. Sng BL, Tan HS, Sia ATH. Closed-loop double-vasopressor automated system vs manual bolus vasopressor to treat hypotension during spinal anaesthesia for caesarean section: a randomised controlled trial. Anaesthesia 2014; 69:37–45.
9. Sng BL, Wang H, Assam PN, Sia AT. Assessment of an updated double-vasopressor automated system using Nexfin for the maintenance of haemodynamic stability to improve peri-operative outcome during spinal anaesthesia for caesarean section. Anaesthesia 2015; 70:691–698.
10. Imholz BP, Wieling W, van Montfrans GA, Wesseling KH. Fifteen years experience with finger arterial pressure monitoring: Assessment of the technology. Cardiovasc Res 1998; 38:605–616.
11. Martina JR, Westerhof BE, van Goudoever J, et al. Noninvasive continuous arterial blood pressure monitoring with Nexfin®. Anesthesiology 2012; 116:1092–1103.
12. Broch O, Renner J, Gruenewald M, et al. A comparison of the Nexfin(R) and transcardiopulmonary thermodilution to estimate cardiac output during coronary artery surgery. Anaesthesia 2012; 67:377–383.
13. Doherty A, Ohashi Y, Downey K, Carvalho JC. Phenylephrine infusion versus bolus regimens during cesarean delivery under spinal anesthesia: a double-blind randomized clinical trial to assess hemodynamic changes. Anesth Analg 2012; 115:1343–1413.
14. Dyer RA, Reed AR, van Dyk D, et al. Hemodynamic effects of ephedrine, phenylephrine, and the coadministration of phenylephrine with oxytocin during spinal anesthesia for elective cesarean delivery. Anesthesiology 2009; 111:753–765.
15. 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. Br J Anaesth 2013; 110:59–65.
16. Varvel JR, Donoho DL, Shafer SL. Measuring the predictive performance of computer-controlled infusion pumps. J Pharmacokinet Biopharm 1992; 20:63–94.
17. Ngan Kee WD, Tam YH, Khaw KS, et al. Closed-loop feedback computer-controlled infusion of phenylephrine for maintaining blood pressure during spinal anaesthesia for caesarean section: a preliminary descriptive study. Anaesthesia 2007; 62:1251–1256.
18. 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; 103:744–750.
19. Ngan Kee WD, Tam YH, Khaw KS, et al. Closed-loop feedback computer-controlled phenylephrine for maintenance of blood pressure during spinal anesthesia for cesarean delivery: a randomized trial comparing automated boluses versus infusion. Anesth Analg 2017; 125:117–123.
20. das Neves JF, Monteiro GA, de Almeida JR, et al. Phenylephrine for blood pressure control in elective cesarean section: therapeutic versus prophylactic doses. Rev Bras Anestesiol 2010; 60:391–398.
© 2018 European Society of Anaesthesiology