Journal of Clinical Monitoring and Computing

https://doi.org/10.1007/s10877-018-0125-4

ORIGINAL RESEARCH

Impact of continuous non-invasive blood pressure monitoring

on hemodynamic fluctuation during general anesthesia: a randomized
controlled study
Takashi Juri1 · Koichi Suehiro1   · Aya Kimura1 · Akira Mukai1 · Katsuaki Tanaka1 · Tokuhiro Yamada1 · Takashi Mori1 ·
Kiyonobu Nishikawa1

Received: 25 September 2017 / Accepted: 2 March 2018

© Springer Science+Business Media B.V., part of Springer Nature 2018

Abstract
This study assessed the ability of a continuous non-invasive blood pressure (BP) monitoring system to reduce intra-anesthetic
hemodynamic fluctuation compared with intermittent BP cuff measurement. Forty patients undergoing total knee arthroplasty
under general anesthesia were enrolled and randomly divided into two groups (Control and CS group). BP management
was performed using the same protocol with BP measured by intermittent BP cuff in the Control and that by continuous
non-invasive BP monitoring in the CS group. We assessed the accuracy and precision of the continuous non-invasive BP
monitoring compared with BP cuff measurement using Bland–Altman, four-quadrant plot, and polar-plot analyses. Addition-
ally, the occurrence of hypotension and hypertention during general anesthesia was compared between the two groups. The
continuous non-invasive BP monitoring showed excellent accuracy of − 1.1 ± 8.1 mmHg during surgery and an acceptable
trending ability with a concordance rate of 95.1% according to the four-quadrant plot analysis and an angular concordance
rate of 86.7% by polar-plot analysis. Hypotension was less common in the CS group during induction of anesthesia (p = 0.002)
and surgery (p = 0.008). Hypertension occurred more frequently in the Control group during emergence from anesthesia
(p = 0.037). The duration of hemodynamic stability (systolic BP 80–110% of baseline) intraoperatively was longer in the CS
group than in the Control group (87.7 vs. 61.9%; p < 0.001). Accuracy and trending ability of the continuous non-invasive
BP monitoring was clinically acceptable, and lead to hemodynamic stability and reduction of intra-anesthetic hypotension
and hypertension intraoperatively.

Keywords  Continuous blood pressure monitoring · Cuff measurement · General anesthesia · Hemodynamic fluctuation ·
Non-invasive

1 Introduction a deep anesthetic level [2]. Also, cardiovascular complica-

tions frequently occur in the perioperative period, and can
With advances in surgical techniques and perioperative care, lead to mortality [7]. Therefore, reducing the incidence of
the number of patients undergoing major non-cardiac sur- perioperative cardiovascular complications remains a con-
gery is increasing and intraoperative management of anes- cern [8]. One anesthetic approach to decrease intraoperative
thesia can impact short and long-term outcomes [1–3]. For and postoperative cardiac morbidity is to maintain a stable
example, postoperative outcome is improved by intraopera- hemodynamic status during surgery [9, 10].
tive goal-directed fluid therapy [4], restrictive blood trans- Hemodynamic fluctuation are common during general
fusion [5], and avoiding intraoperative hypothermia [6] and anesthesia because of the vasodilation and myocardial
suppression caused by anesthetic drugs, surgical stress,
and blood loss [11–16]. Both hypotension and hyperten-
* Koichi Suehiro sion may increase organ oxygen demand and reported to
[email protected] be correlated to postoperative morbidity and mortality [17].
1 Intraoperative blood pressure (BP) measurement is typically
Department of Anesthesiology, Osaka City University
Graduate School of Medicine, 1‑5‑7 Asahimachi, Abenoku, performed using an automated digital sphygmomanometer
Osaka City, Osaka 545‑8586, Japan and is usually taken every 5 min as recommended by the

13
Vol.:(0123456789)
 Journal of Clinical Monitoring and Computing

American Society of Anesthesiologists [18]. Anesthesi- non-invasive BP cuff (IntelliVue MP70; Philips Electron-
ologists make decisions about hemodynamic management ics Japan Corp., Tokyo, Japan) measurement were attached.
including administration of vasopressors and fluid loading In the CS group, the ClearSight device was attached to an
based on the values obtained using a non-invasive BP cuff. index finger. Patients rested in a supine position and kept
Detection of hypertension and hypotension would be more calm in the operating theatre for 5 min. Non-invasive BP
difficult when BP was measured intermittently compared cuff measurement was then started in both groups. Systolic
with an arterial line. However, intra-arterial catheterization BP ­(SBPcuff), diastolic BP ­(DBPcuff), and mean BP ­(MBPcuff)
can cause complications, including hematoma, nerve injury, were measured on three occasions separated by 1-min inter-
and pseudo-aneurysm [18]. The ClearSight system (Edwards vals, and baseline value was calculated as the average of
Lifesciences, Irvine, CA, USA) is a monitoring system that three values. In the CS group, BP measurement by the con-
allows for continuous non-invasive BP measurement. This tinuous non-invasive BP monitoring was started simultane-
system was developed using Wesseling’s Finapres method. ously, and systolic BP (­ SBPCS), diastolic BP ­(DBPCS), and
The Finapres system utilizes the “Physiocal” (physiologic mean BP ­(MBPCS) were recorded as the baseline values.
calibration) algorithm for vascular unloading. The Physiocal An intravenous line was inserted after the baseline BP
algorithm induces stepwise and intermittent BP changes, and measurements were completed. Administration of fluid was
simultaneously repeats the analysis of the photoplethysmo- initiated at 10 ml/min using Ringer’s solution containing
gram waveforms. When alternate pressure settings produce bicarbonate (Bicanate®, Otsuka Pharmaceutical Co. Ltd.,
larger photoplethysmogram amplitudes, the feedback is used Tokyo, Japan). Five minutes after initiation of fluid adminis-
to achieve an unloaded state of the finger artery [18]. We tration, induction of anesthesia was performed using propo-
hypothesized that this continuous non-invasive BP monitor- fol 1.5 mg/kg, remifentanil 0.3 µg/kg/min, and rocuronium
ing could allow for reduced hemodynamic fluctuation during 0.9 mg/kg. Bag-mask ventilation was then performed at
general anesthesia when compared with standard intermit- around 8 ml/kg tidal volume with 100% oxygen and 4% des-
tent BP cuff measurement. flurane, and the patients were intubated after 5 min. Anesthe-
This study aimed to investigate: (1) the accuracy and sia was maintained using desflurane and remifentanil. The
precision of BP measured with the continuous non-invasive attending anesthesiologists were responsible for intraopera-
BP monitoring compared with intermittent BP cuff meas- tive administration of fluids. The depth of anesthesia was
urement during general anesthesia including induction and controlled to maintain bispectral index values (v4.0, Aspect
emergence and (2) the efficacy of the continuous non-inva- Medical System Inc, Natick, MA, USA) between 45 and 60.
sive BP monitoring allowing for reduction of hypotension Before closing the incision, the surgeon injected a steroid-
and hypertension during general anesthesia when compared containing local anesthetic periarticularly into all the soft
with BP cuff measurement was evaluated. tissues surrounding the knee. Fentanyl 2–4 µg/kg and aceta-
minophen 15 mg/kg were administered at the end of surgery
for analgesia. After the operation, sugammadex (2–4 mg/
2 Methods kg) was administered, and then the patients were extubated.

2.1 Anesthetic management
2.2 Blood pressure management
The study protocol was approved by the institutional review
board. Written informed consent was obtained from all In both groups, BP measurement by non-invasive cuff was
patients. Patients undergoing total knee arthroplasty for performed at 1-min intervals during anesthetic induction and
knee osteoarthritis under general anesthesia were enrolled. emergence from anesthesia, and with 5-min intervals dur-
The exclusion criteria were arrhythmia (atrial fibrillation, ing surgery. The investigators treating the CS group were
frequent ventricular or supraventricular premature beats); blinded to the BP values from the cuff. The BP in the CS
decreased cardiac function (preoperative ejection frac- group was controlled using the values measured by the sys-
tion < 50%); and valvular heart disease [significant (at least tem, whereas that in the Control group was managed using
moderate) regurgitation or stenosis of the aortic, mitral, or the non-invasive BP cuff values. In both groups, BP man-
tricuspid valve]. Patients were randomly divided into a Con- agement was performed using the same protocol (Fig. 1).
trol group and a CS group preoperatively. During induction of and emergence from anesthesia, BP
No premedication or intravenous infusions were adminis- was controlled to maintain S ­ BPcuff in the Control group,
tered preoperatively, and all patients were able to take clear and ­SBPCS in the CS group within 90–110% of the baseline
liquids until 3 h before surgery. On arrival in the operating value. During surgery, the remifentanil infusion rate was
room, each patient was laid on the operating table and moni- controlled to maintain SBP within 80–110% of the baseline
toring devices, for pulse oximetry, electrocardiography, and value. Vasopressors or vasodilators were administered in the

13
Journal of Clinical Monitoring and Computing

a power analysis demonstrated that a sample size of 20 sub-

jects in each group was sufficient to detect a difference of 18%
between the Control and CS groups with a power of 0.80 and
an alpha of 0.05.
We also compared the accuracy and precision of the con-
tinuous non-invasive BP monitoring for BP measurement
with that of BP cuff measurement. In the CS group patients
(n = 20), Bland–Altman analysis was performed to compare
­MBPcuff and M ­ BPCS. The results of this analysis are demon-
strated as the bias and the 95% limits of agreement (LOA)
[19]. We adjusted the repeated measurements for each patient
as described by Myles and Cui [20]. The BP measured by
the continuous non-invasive BP monitoring was defined as
acceptable if accuracy and precision were no greater than 5
and 8 mmHg, respectively, when compared with BP meas-
ured by the cuff, based on the standards for evaluation of BP
monitoring recommended by the Association for the Advance-
ment of Medical Instrumentation [21]. The changes in MBP
between two consecutive measurements, as measured using
a non-invasive BP cuff ­(MBPcuff) and the continuous non-
invasive BP monitoring (­MBPCS), were also compared to
assess the trending ability of the continuous non-invasive BP
monitoring. Four-quadrant plot and polar-plot analyses were
used to evaluate tracking ability [22]. The concordance rate
(CR) is defined as the percentage of data points lying in the
lower left or upper right quadrant, and is evaluated as excellent
when more than 92% [23, 24]. In the polar-plot analysis, the
agreement between the two methods was assessed by the fol-
lowing variables: the mean angular bias (i.e., the average angle
between a data point and the polar axis); the angular CR (i.e.,
Fig. 1  Protocol for BP management during a induction of and emer-
the percentage of data points within a ± 30° radial zone); and
gence from anesthesia and b intraoperatively. BP blood pressure, SBP the radial LOA (i.e., the radial zone containing 95% of all the
systolic blood pressure data points). Using this analysis, the tracking ability is defined
as “excellent” when the mean angular bias is within ± 5°, the
angular CR is more than 95%, and the radial LOA are within
event of a sudden decrease or increase in BP that could not ± 30° [22]. The data within a central zone of radial limits of
be controlled by remifentanil. 10% were excluded from the analysis, because a slight change
in BP may not reflect trending ability.
2.3 Statistical analysis Patient characteristics were compared using the
Mann–Whitney U test, the Student’s t-test, and the Chi
For analysis of BP, we used the data measured by the cuff to square test. For all analyses, a p-value < 0.05 was consid-
standardize the BP outcome in both groups. We set the primary ered to be statistically significant. The statistical analysis
study endpoint to be hemodynamically stable time (HSTime), was performed using StatFlex version 6.0 (Artech Co., Ltd,
i.e., the ratio of the intraoperative duration of hemodynamic Osaka, Japan), MedCalc version 14 (MedCalc Software
stability (SBP in the range of 80–110% of the baseline value) bvba, Ostend, Belgium), and SigmaPlot 13.0 (Systat Soft-
to the total operating time and the secondary study endpoints ware Inc., San Jose, CA, USA).
to be the incidence of hypotension (SBP < 90 mmHg or 80%
of baseline SBP) and of hypertension (SBP > 160 mmHg or
110% of baseline SBP) during induction of and emergence 3 Results
from anesthesia. In a preliminary study of 5 cases, HSTime
was 72 ± 18% without the continuous non-invasive BP moni- Figure 2 shows the CONSORT flow diagram. Forty-five
toring. We hypothesized that the HSTime when using the patients were assessed for inclusion in the study, and five
continuous non-invasive BP monitoring would be 90%, and patients were excluded because of not meeting inclusion

13
 Journal of Clinical Monitoring and Computing

criteria (Control group, n = 20; CS group, n = 20). There We investigated the accuracy and precision of BP
were no exclusions after randomization. As shown in measured by the continuous non-invasive BP monitoring
Table 1, there were no significant preoperative differences system. The results of Bland–Altman analysis are shown
between the two groups. in Fig. 3 and Table 2. The bias was − 3.9 ± 10.3 mmHg

Fig. 2  CONSORT flow diagram

for the present study

Table 1  Preoperative data in Variables Control group (n = 20) ClearSight group p value

both groups (n = 20)

Gender (M/F) 3/17 5/15

Age (year) 75.5 ± 5.2 71.5 ± 10.1 0.120
Height (cm) 152 ± 7.4 153 ± 8.3 0.689
Weight (kg) 61.2 ± 7.6 59.8 ± 9.4 0.608
Operation side (L/R) 13/7 12/8
NYHA classification 1.00
 I/II 20 20
 III/IV 0 0
Preoperative complication
 Hypertension, n (%) 13 (65%) 12 (60%)
 Diabetes mellitus, n (%) 9 (45%) 6 (30%)
 Old myocardial infarction 2 (10%) 2 (10%)
 Asthma, n (%) 2 (10%) 2 (10%)
 Chronic kidney disease, n (%) 0 (0%) 2 (10%)
Baseline hemodynamic data
 Systolic BP (mmHg) 149 ± 20.1 140 ± 15.5 0.119
 Diastolic BP (mmHg) 73.9 ± 15.6 79.3 ± 10.2 0.204
 Mean BP (mmHg) 94.6 ± 15.6 95.2 ± 10.4 0.887
 Heart rate (bpm) 71.8 ± 9.8 75.8 ± 11.8 0.250

p < 0.05 statistically significant

Data are expressed as mean ± SD
NYHA New York Heart Association, BP blood pressure

13
Journal of Clinical Monitoring and Computing

during the procedure. The continuous non-invasive BP induction of and emergence from anesthesia (angular CR:
monitoring showed excellent accuracy and precision of 54.5 and 72.7%, respectively).
− 1.1 ± 8.1 mmHg intraoperatively but poor accuracy and Table 3 shows the intraoperative data for both groups. The
precision during induction of and emergence from anesthe- patients in the CS group had a larger urine output; despite for
sia. The trending ability of the continuous non-invasive BP the volumes of fluid administered were not different between
monitoring was also evaluated using four-quadrant and polar the two study groups. The intraoperative dose of phenyle-
plot-analyses, shown in Fig. 4 and Table 2. In the four-quad- phrine was greater in the CS group, whereas ephedrine was
rant analysis, the continuous non-invasive BP monitoring more likely to be used in the Control group (p < 0.05). The
showed excellent trending ability throughout the procedure incidence of hypotension was lower in the CS group during
(CR, 87.9% during induction of anesthesia, 97.7% intraoper- induction of anesthesia and during surgery (p = 0.002 and
atively, 90.9% during emergence from anesthesia, and 95.1% p = 0.008, respectively). Hypertension during emergence
for total operating time). In the polar-plot analysis, the angu- from anesthesia was more common in the Control group
lar bias and CR was − 9.3 ± 21.4° and 79.2%, respectively, (p = 0.037). Further, the hemodynamic stability time was
during the procedure. The continuous non-invasive BP mon- longer in the CS group than in the Control group (87.7 vs.
itoring demonstrated an acceptable tracking ability during 61.9%, p < 0.001) (Fig. 5).
surgery of − 10.7 ± 17.1° and 86.7% in angular bias and CR,
respectively. However, the trending ability worsened during
4 Discussion

We evaluate the ability of the continuous non-invasive BP

monitoring allowing for reduction of hemodynamic fluc-
tuation during general anesthesia. First, we assessed the
accuracy and precision of BP measured by the continuous
non-invasive BP monitoring with that of intermittent BP
cuff measurement during general anesthesia. The bias shown
by the Bland–Altman analysis was − 3.9 ± 10.3 mmHg dur-
ing the procedure. Many validation studies have been per-
formed for the continuous non-invasive BP monitoring of
BP measurement [21, 25–28]. Fisher et al. [25] evaluated
the accuracy of this system in patients undergoing cardiac
surgery. There was a correlation between BP measured by
the continuous non-invasive BP monitoring and invasive
arterial pressure measurement (p < 0.001), with a bias of
Fig. 3  Bland–Altman analysis of agreement of BP measurements 4.6 ± 6.5 mmHg. In critically ill patients, the correlation
obtained by an BP intermittent cuff and the continuous non-invasive between MBP measured by the continuous non-invasive BP
BP monitoring. M­ BPcuff, mean blood pressure measured by intermit-
tent cuff, ­MBPCS, mean blood pressure measured by the continuous monitoring and that recorded by invasive arterial pressure
non-invasive BP monitoring. BP blood pressure measurement was acceptable ­(r2 = 0.50, bias 6 ± 12 mmHg),

Table 2  Statistical data from Anesthetic induction During operation Emergence Total

Bland–Altman analysis and from anesthesia
polar plots
Bland–Altman analysis
 Mean bias (mmHg) − 8.7 ± 14.4 − 1.1 ± 8.1 − 9.9 ± 9.4 − 3.9 ± 10.3
 95% LA (mmHg) 28.3 16.2 18.5 20.5
Four quadrant plot
 CR (%) 87.9 97.7 90.9 95.1
Polar plot
 Mean angular bias (degree) 6.3 ± 31.6 − 10.7 ± 17.1 − 5.9 ± 25.3 − 9.3 ± 21.4
 Radial LA (degree) 61.9 33.5 49.6 41.9
 Angular CR (%) 54.5 86.7 72.7 79.2

Data are expressed as mean ± SD

LA limit of agreement, CR concordance rate

13
 Journal of Clinical Monitoring and Computing

Fig. 4  Polar-plot analysis to
assess the trending ability of
changes of mean blood pressure
measured by the continuous
non-invasive BP monitoring
compared with intermittent cuff
for all measurements. The dis-
tance from the center of the plot
shows the mean percent changes
in mean blood pressure, and the
angle with the horizontal axis
means disagreement. The data
inside the limits of agreement
(i.e. ± 30° shown by dotted
lines) represents good trending
ability. ΔMBP mean percentage
changes in mean blood pressure

but worsened in patients with edema and those in whom first year after surgery (relative risk 1.036/min; p = 0.0125).
noradrenaline was administered (­r2 = 0.40 and ­r2 = 0.28, In another study [17], intraoperative hypotension increases
respectively). Chen et  al. [27] compared SBP recorded the risk of 30-day postoperative mortality. Even short dura-
by the continuous non-invasive BP monitoring with that tions of an intraoperative MBP less than 55 mmHg can cause
measured by an intermittent BP cuff in surgical patients, acute kidney and myocardial injury [31]. In vascular sur-
and report a bias of − 1.0 ± 12.6 mmHg. They also assessed gery, intraoperative hypotension defined as a 40% decrease
the trending ability for BP measurement and reported an from the preoperative MBP with a duration for > 30 min is
angular CR of 81.3% by polar-plot analysis. In the meta- associated with postoperative myocardial injury [32]. When
analysis by Kim et al. [21], the accuracy was good with using the surgical Apgar score, the lowest MBP was useful
a bias of − 2 ± 8, 5 ± 6.5, and 4 ± 7 mmHg for SBP, DBP, for predicting an outcome after surgery [33]. A continuous
and MBP, respectively. Furthermore, intraoperative cardiac recording allows the anesthesiologist to intervene soon so
output (CO) measurement using the continuous non-invasive that the patient is stable in regard to blood pressure, which
BP monitoring had a strong correlation to CO measured by could lead to a decrease in postoperative morbidity and mor-
esophageal Doppler even after phenylephrine administra- tality as well as improved long-term outcomes. As indicated
tion [29]. Therefore, the impact of acute changes in vascular in Fig. 5, the risk of intra-anesthetic hypotension was lower
resistance may not be so large in this system, which may lead in the CS group than in the Control group, particularly dur-
to acceptable trending ability of this system for BP measure- ing induction of anesthesia (90 vs. 45%, p = 0.002). It is
ment. Except in critically ill patients, the accuracy and trend- important to be able to deal with hypotension quickly during
ing ability of the continuous non-invasive BP monitoring induction of anesthesia by using the continuous blood pres-
was clinically acceptable, which can lead to intraoperative sure monitoring, because anesthesiologists may be distracted
hemodynamic stability and reduction of the incidence of during mask-ventilation and subsequent tracheal intubation.
intra-anesthetic hypotension and hypertension. Moreover, patients in the CS group were exposed to less
Perioperative morbidity and mortality as well as the hypertension during emergence from anesthesia. Although
long-term outcomes after surgery may be influenced by there were no significant differences in infusion and blood
anesthetic management, especially management of arterial loss between the two groups, the amount of urine output
BP. Intraoperative hypotension is a risk factor for periopera- was larger in the CS group due to maintenance of the renal
tive morbidity and mortality [17, 30–33]. Monk et al. [30] blood flow.
reported that every minute of hypotension in the operating As shown by Meidert et al. [34], continuous intraopera-
room increased the risk of postoperative mortality in the tive BP measurement by the ClearSight system decreased

13
Journal of Clinical Monitoring and Computing

Table 3  Intraoperative data in Variables Control group (n = 20) ClearSight group p value

both groups (n = 20)

Operation time (min) 131 ± 33.0 119 ± 37.3 0.304

Anesthesia time (min) 179 ± 37.2 163 ± 37.6 0.170
Tourniquet time (min) 110 ± 36.5 103 ± 29.6 0.471
Infusion (mL) 974 ± 294 830 ± 401 0.203
Blood loss (mL) 40.8 ± 51.6 42.8 ± 77.2 0.924
Urine output (mL) 269 ± 192 668 ± 564 0.005*
Total amount of vasoactive drugs
 Phenylephrine (mg) 0.74 ± 0.58 1.51 ± 0.52 < 0.001*
 Ephedrine (mg) 22.3 ± 9.7 12.9 ± 8.0 0.002
 Nicardipine (mg) 0.48 ± 0.93 0.82 ± 1.54 0.403
Maximum blood pressure
 Systolic BP (mmHg) 174 ± 18.0 159 ± 21.8 0.023*
 Diastolic BP (mmHg) 78.3 ± 12.3 84.0 ± 17.1 0.233
 Mean BP (mmHg) 102 ± 10.9 106 ± 28.8 0.526
Minimum blood pressure
 Systolic BP (mmHg) 91.2 ± 18.5 100 ± 12.7 0.075
 Diastolic BP (mmHg) 48.8 ± 10.5 53.3 ± 8.4 0.147
 Mean BP (mmHg) 58.8 ± 12.1 64.2 ± 8.5 0.114
Incidence of hypotension, cases (%)
 Anesthetic induction 18 (90%) 9 (45%) 0.002*
 During operation 19 (95%) 12 (60%) 0.008*
 Emergence from anesthesia 8 (40%) 6 (30%) 0.507
Incidence of hypertension, cases (%)
 Anesthetic induction 8 (40%) 9 (45%) 0.749
 During operation 8 (40%) 9 (45%) 0.749
 Emergence from anesthesia 6 (30%) 1 (5%) 0.037*
Hemodynamic stability (%) 61.9 ± 22.9 87.7 ± 10.9 < 0.001*

Data are expressed as mean ± SD

BP blood pressure
*p < 0.05 statistically significant

Fig. 5  Duration of hemody-
namic stability and incidence
of hypotension in the control
group and the CS group

13
 Journal of Clinical Monitoring and Computing

the hypotensive events during orthopedic surgery com- References

pared with intermittent BP monitoring. Other continuous
non-invasive monitoring could induce hemodynamic sta- 1. Etzioni DA, Liu JH, Maggard MA, Ko CY. The aging popu-
bility in various clinical situations [35, 36]. Benes et al. lation and its impact on the surgery workforce. Ann Surg.
2003;238:170–7.
[35] reported that continuous BP monitoring by the CNAP 2. Sessler DI, Sigl JC, Kelley SD, Chamoun NG, Manberg PJ,
monitor (CNSystems Medizintechnik AG, Graz, Austria) Saager L, Kurz A, Greenwald S. Hospital stay and mortality are
allowed for intraoperative BP stability in surgical patients increased in patients having a “triple low” of low blood pressure,
in upright position. The CNAP system can also detect low bispectral index, and low minimum alveolar concentration
of volatile anesthesia. Anesthesiology 2012;116:1195–203.
intraoperative hypotension more precisely than intermit- 3. Hallqvist L, Martensson J, Granath F, Sahlen A, Bell M. Intra-
tent BP measurement after spinal anesthesia for cesarean operative hypotension is associated with myocardial damage in
delivery [36]. As with the continuous non-invasive BP noncardiac surgery: an observational study. Eur J Anaesthesiol.
monitoring, invasive arterial pressure measures BP con- 2016;33:450–6.
4. Phan TD, Ismail H, Heriot AG, Ho KM. Improving periop-
tinuously, and may allow for better BP management. erative outcomes: fluid optimization with the esophageal
Our study has some limitations. Firstly, we measured Doppler monitor, a metaanalysis and review. J Am Coll Surg.
BP using an intermittent BP cuff in the Control group. 2008;207:935–41.
Secondly, we did not assess the postoperative course. 5. Koch CG, Li L, Duncan AI, Mihaljevic T, Cosgrove DM, Loop
FD, Starr NJ, Blackstone EH. Morbidity and mortality risk
Thirdly, although the ClearSight System can provide vari- associated with red blood cell and blood-component transfu-
able hemodynamic parameters such as CO and stroke vol- sion in isolated coronary artery bypass grafting. Crit Care Med.
ume variation, we focused on blood pressure in the current 2006;34:1608–16.
study. Perioperative goal-directed therapy (GDT) using 6. Yamasaki H, Tanaka K, Funai Y, Suehiro K, Ikenaga K, Mori
T, Osugi H, Nishikawa K. The impact of intraoperative hypo-
minimally invasive CO monitoring has been reported to thermia on early postoperative adverse events after radical
decrease postoperative morbidity [37]. Therefore, optimi- esophagectomy for cancer: a retrospective cohort study. J Car-
zation of CO and stroke volume variation using the Clear- diothorac Vasc Anesth. 2014;28:943–7.
Sight system can lead to better postoperative outcomes. 7. Sharifpour M, Moore LE, Shanks AM, Didier TJ, Kheterpal S,
Mashour GA. Incidence, predictors, and outcomes of periopera-
tive stroke in noncarotid major vascular surgery. Anesth Analg.
2013;116:424–34.
8. Tsuchiya M, Takahashi R, Furukawa A, Suehiro K, Mizutani
5 Conclusions K, Nishikawa K. Transversus abdominis plane block in com-
bination with general anesthesia provides better intraoperative
hemodynamic control and quicker recovery than general anes-
We demonstrated that hemodynamic management using thesia alone in high-risk abdominal surgery patients. Minerva
the continuous non-invasive BP monitoring can induce Anestesiol. 2012;78:1241–7.
intra-anesthetic hemodynamic stability. The accuracy 9. Charlson ME, MacKenzie CR, Gold JP, Ales KL, Topkins M,
Fairclough GP Jr, Shires GT. The preoperative and intraop-
and trending ability of the continuous non-invasive BP erative hemodynamic predictors of postoperative myocardial
monitoring for BP measurement was clinically accept- infarction or ischemia in patients undergoing noncardiac sur-
able, and led to intraoperative hemodynamic stability and gery. Ann Surg. 1989;210:637–48.
a reduced incidence of intra-anesthetic hypotension and 10. Charlson ME, MacKenzie CR, Gold JP, Ales KL, Topkins M,
Shires GT. Preoperative characteristics predicting intraoperative
hypertension. hypotension and hypertension among hypertensives and diabet-
ics undergoing noncardiac surgery. Ann Surg. 1990;212:66–81.
Funding  Only departmental funds were used for this study. 11. Bijker JB, van Klei WA, Kappen TH, van Wolfswinkel L, Moons
KG, Kalkman CJ. Incidence of intraoperative hypotension as a
Compliance with ethical standards  function of the chosen definition: literature definitions applied
to a retrospective cohort using automated data collection. Anes-
thesiology 2007;107:213–20.
Conflict of interest  None of the authors has any potential conflict of 12. Zhang J, Critchley LA. Inferior vena cava ultrasonography
interest, including no commercial relationships such as consultation or before general anesthesia can predict hypotension after induc-
equity interests that could have direct or potential influence or impart tion. Anesthesiology 2016;124:580–9.
bias on the work. 13. Lima EQ, Zanetta DM, Castro I, Massarollo PC, Mies S,
Machado MM, Yu L. Risk factors for development of acute renal
Research involving human participants and/or animals  All procedures failure after liver transplantation. Ren Fail. 2003;25:553–60.
performed in studies involving human participants were in accordance 14. Latson TW, Ashmore TH, Reinhart DJ, Klein KW, Giesecke
with the ethical standards of the institutional and/or national research AH. Autonomic reflex dysfunction in patients presenting for
committee and with the 1964 Helsinki Declaration and its later amend- elective surgery is associated with hypotension after anesthesia
ments or comparable ethical standards. induction. Anesthesiology 1994;80:326–37.
15. Lonjaret L, Lairez O, Minville V, Geeraerts T. Optimal perio-
Informed consent  Informed consent was obtained from all individual perative management of arterial blood pressure. Integr Blood
participants included in the study. Press Control. 2014;7:49–59.

13
Journal of Clinical Monitoring and Computing

16. Lin FQ, Li C, Zhang LJ, Fu SK, Chen GQ, Yang XH, Zhu CY, pressure monitoring on intraoperative hemodynamic management.
Li Q. Effect of rapid plasma volume expansion during anes- J Clin Monit Comput. 2012;26:133–40.
thesia induction on haemodynamics and oxygen balance in 28. Joosten A, Desebbe O, Suehiro K, Murphy LS, Essiet M, Alexan-
patients undergoing gastrointestinal surgery. Int J Med Sci. der B, Fischer MO, Barvais L, Van Obbergh L, Maucort-Boulch
2013;10:355–61. D, Cannesson M, Paul M. Accuracy and precision of non-inva-
17. Monk TG, Bronsert MR, Henderson WG, Mangione MP, Sum- sive cardiac output monitoring devices in perioperative medi-
Ping ST, Bentt DR, Nguyen JD, Richman JS, Meguid RA, Ham- cine: a systematic review and meta-analysisdagger. Br J Anaesth.
mermeister KE. Association between intraoperative hypotension 2017;118:298–310.
and hypertension and 30-day postoperative mortality in noncar- 29. Chen G, Meng L, Alexander B, Tran NP, Kain ZN, Cannesson M.
diac surgery. Anesthesiology 2015;123:307–19. Comparison of noninvasive cardiac output measurements using
18. Bartels K, Esper SA, Thiele RH. Blood pressure monitor- the Nexfin monitoring device and the esophageal Doppler. J Clin
ing for the anesthesiologist: a practical review. Anesth Analg. Anesth. 2012;24:275–83.
2016;122:1866–79. 30. Monk TG, Saini V, Weldon BC, Sigl JC. Anesthetic management
19. Bland JM, Altman DG. Statistical methods for assessing agree- and one-year mortality after noncardiac surgery. Anesth Analg.
ment between two methods of clinical measurement. Lancet 2005;100:4–10.
1986;1:307–10. 31. Walsh M, Devereaux PJ, Garg AX, Kurz A, Turan A, Rodseth
20. Myles PS, Cui J. Using the Bland–Altman method to measure RN, Cywinski J, Thabane L, Sessler DI. Relationship between
agreement with repeated measures. Br J Anaesth. 2007;99:309–11. intraoperative mean arterial pressure and clinical outcomes after
21. Kim SH, Lilot M, Sidhu KS, Rinehart J, Yu Z, Canales C, Can- noncardiac surgery: toward an empirical definition of hypotension.
nesson M. Accuracy and precision of continuous noninvasive Anesthesiology 2013;119:507–15.
arterial pressure monitoring compared with invasive arterial 32. van Waes JA, van Klei WA, Wijeysundera DN, van Wolfswinkel
pressure: a systematic review and meta-analysis. Anesthesiology L, Lindsay TF, Beattie WS. Association between intraoperative
2014;120:1080–97. hypotension and myocardial injury after vascular surgery. Anes-
22. Critchley LA, Yang XX, Lee A. Assessment of trending ability of thesiology 2016;124:35–44.
cardiac output monitors by polar plot methodology. J Cardiothorac 33. Gawande AA, Kwaan MR, Regenbogen SE, Lipsitz SA, Zinner
Vasc Anesth. 2011;25:536–46. MJ. An Apgar score for surgery. J Am Coll Surg. 2007;204:201–8.
23. Suehiro K, Tanaka K, Yamada T, Matsuura T, Funao T, Mori 34. Meidert AS, Nold JS, Hornung R, Paulus AC, Zwißler B, Czerner
T, Nishikawa K. The utility of intra-operative three-dimensional S. The impact of continuous non-invasive arterial blood pressure
transoesophageal echocardiography for dynamic measurement of monitoring on blood pressure stability during general anaesthesia
stroke volume. Anaesthesia 2015;70:150–9. in orthopaedic patients: a randomised trial. Eur J Anaesthesiol.
24. Suehiro K, Tanaka K, Yamada T, Matsuura T, Mori T, Funao T, 2017;34:716–22.
Nishikawa K. The ability of the Vigileo-FloTrac system to meas- 35. Benes J, Simanova A, Tovarnicka T, Sevcikova S, Kletecka J,
ure cardiac output and track cardiac output changes during one- Zatloukal J, Pradl R, Chytra I, Kasal E. Continuous non-invasive
lung ventilation. J Clin Monit Comput. 2015;29:333–9. monitoring improves blood pressure stability in upright position:
25. Fischer MO, Avram R, Carjaliu I, Massetti M, Gerard JL, Hanouz randomized controlled trial. J Clin Monit Comput. 2015;29:11–7.
JL, Fellahi JL. Non-invasive continuous arterial pressure and 36. Ilies C, Kiskalt H, Siedenhans D, Meybohm P, Steinfath M, Bein
cardiac index monitoring with Nexfin after cardiac surgery. Br J B, Hanss R. Detection of hypotension during Caesarean section
Anaesth. 2012;109:514–21. with continuous non-invasive arterial pressure device or intermit-
26. Hohn A, Defosse JM, Becker S, Steffen C, Wappler F, Sakka SG. tent oscillometric arterial pressure measurement. Br J Anaesth.
Non-invasive continuous arterial pressure monitoring with Nexfin 2012;109:413–9.
does not sufficiently replace invasive measurements in critically 37. Michard F, Giglio MT, Brienza N. Perioperative goal-directed
ill patients. Br J Anaesth. 2013;111:178–84. therapy with uncalibrated pulse contour methods: impact on
27. Chen G, Chung E, Meng L, Alexander B, Vu T, Rinehart J, fluid management and postoperative outcome. Br J Anaesth.
Cannesson M. Impact of non invasive and beat-to-beat arterial 2017;119:22–30.

13