Background

The optimal method for blood pressure monitoring in obese surgical patients remains unknown. Arterial catheters can cause potential complications, and noninvasive oscillometry provides only intermittent values. Finger cuff methods allow continuous noninvasive monitoring. The authors tested the hypothesis that the agreement between finger cuff and intraarterial measurements is better than the agreement between oscillometric and intraarterial measurements.

Methods

This prospective study compared intraarterial (reference method), finger cuff, and oscillometric (upper arm, forearm, and lower leg) blood pressure measurements in 90 obese patients having bariatric surgery using Bland–Altman analysis, four-quadrant plot and concordance analysis (to assess the ability of monitoring methods to follow blood pressure changes), and error grid analysis (to describe the clinical relevance of measurement differences).

Results

The difference (mean ± SD) between finger cuff and intraarterial measurements was −1 mmHg (± 11 mmHg) for mean arterial pressure, −7 mmHg (± 14 mmHg) for systolic blood pressure, and 0 mmHg (± 11 mmHg) for diastolic blood pressure. Concordance between changes in finger cuff and intraarterial measurements was 88% (mean arterial pressure), 85% (systolic blood pressure), and 81% (diastolic blood pressure). In error grid analysis comparing finger cuff and intraarterial measurements, the proportions of measurements in risk zones A to E were 77.1%, 21.6%, 0.9%, 0.4%, and 0.0% for mean arterial pressure, respectively, and 89.5%, 9.8%, 0.2%, 0.4%, and 0.2%, respectively, for systolic blood pressure. For mean arterial pressure and diastolic blood pressure, absolute agreement and trending agreement between finger cuff and intraarterial measurements were better than between oscillometric (at each of the three measurement sites) and intraarterial measurements. Forearm performed better than upper arm and lower leg monitoring with regard to absolute agreement and trending agreement with intraarterial monitoring.

Conclusions

The agreement between finger cuff and intraarterial measurements was better than the agreement between oscillometric and intraarterial measurements for mean arterial pressure and diastolic blood pressure in obese patients during surgery. Forearm oscillometry exhibits better measurement performance than upper arm or lower leg oscillometry.

Editor’s Perspective
What We Already Know about This Topic
  • Optimal blood pressure monitoring in obese surgical patients remains unknown because multiple invasive and noninvasive monitoring methods are available with little understanding of agreement between different methods.

What This Article Tells Us That Is New
  • In a study of 90 obese patients having bariatric surgery, the agreement between finger cuff and intraarterial measurements was better than the agreement between oscillometric and intraarterial measurements for mean arterial and diastolic blood pressure, but not systolic blood pressure. Forearm oscillometry demonstrated better measurement performance than upper arm or lower leg oscillometry.

Close blood pressure monitoring is important in obese patients having surgery because those patients are at high risk of cardiac complications.1,2  However, in clinical practice, choosing the optimal method for perioperative blood pressure monitoring in obese patients is a challenge.

Invasive continuous blood pressure monitoring using an arterial catheter (i.e., intraarterial blood pressure monitoring) is the clinical reference method, but is associated with potential complications such as permanent ischemic damage (mean incidence, 0.1%), bleeding (mean incidence, 0.5%), or local infection (mean incidence, 0.7%).3,4  Additionally, arterial catheter placement can be technically challenging in morbidly obese patients. Oscillometry using an inflatable cuff enables noninvasive blood pressure monitoring, but provides blood pressure values only intermittently, and the measurement performance depends on the selection of the appropriate cuff size and cuff positioning.5,6  Ideally, the cuff size needs to be matched to the circumference of the extremity used for oscillometry—usually the upper arm—to obtain reliable measurements. Obesity is frequently associated with an increased circumference and a conical rather than cylindrical shape of the upper arm. These changes can make correct cuff placement difficult. Therefore, it has been proposed to use the forearm instead of the upper arm for oscillometry in obese patients.7 

Considering the limitations of both continuous intraarterial and intermittent noninvasive oscillometric blood pressure monitoring, innovative technologies allowing for continuous noninvasive blood pressure monitoring may be a promising option in obese patients during surgery. One of those continuous noninvasive blood pressure monitoring technologies is the finger cuff–based vascular unloading technology that is also referred to as the volume clamp method.8 

For the finger cuff technology to be considered for routine blood pressure monitoring, its measurement performance needs to be investigated in comparison to the established reference methods: continuous intraarterial blood pressure monitoring using an arterial catheter and intermittent noninvasive blood pressure monitoring using oscillometry.

We therefore compared continuous noninvasive finger cuff with continuous intraarterial blood pressure measurements using a radial artery catheter in a prospective method comparison study in obese patients having bariatric surgery. We also compared intermittent noninvasive oscillometric blood pressure measurements at the upper arm, forearm, and lower leg with intraarterial blood pressure measurements.

Study Design and Patients

This prospective method comparison study was reviewed and approved by the Tufts Health Sciences Institutional Review Board (Boston, Massachusetts; No. 9743). We obtained written informed consent from all patients. Adult patients with a body mass index greater than or equal to 40 kg/m2 scheduled for elective laparoscopic bariatric surgery (gastric bypass, sleeve gastrectomy, and gastric banding) and an American Society of Anesthesiologists Physical Status classification of less than IV were eligible for study inclusion. Exclusion criteria were the presence of peripheral upper or lower extremity edema, vascular or anatomical abnormalities, history of ipsilateral axillary or inguinal lymph node dissection, carpal tunnel syndrome, negative modified Allen test and absence of a palpable ipsilateral ulnar pulse, and atrial fibrillation. Data were collected between September 2011 and February 2013.

Anesthesia Management

All patients received a similar general anesthetic regimen at the discretion of the anesthesia team. A representative combination of medications to facilitate induction of general anesthesia and endotracheal intubation consisted of intravenous propofol (1 to 2 mg/kg adjusted body weight), fentanyl (1 µg/kg adjusted body weight), lidocaine 100 mg, and rocuronium bromide (0.8 mg/kg adjusted body weight), or succinylcholine (1.5 mg/kg adjusted body weight). Either sevoflurane or desflurane in combination with fentanyl and hydromorphone was administered for maintenance of general anesthesia. During surgical preparation of the abdomen and at skin closure, the patients were positioned horizontally, and during surgery, a reverse Trendelenburg position was maintained. Both arms were extended and secured on padded arm boards.

Blood Pressure Measurements

We recorded continuous intraarterial blood pressure using a radial artery catheter (reference method), as well as continuous noninvasive finger cuff blood pressure and intermittent noninvasive oscillometric upper arm, forearm, and lower leg blood pressure (test methods).

To measure continuous intraarterial blood pressure, we inserted a 20-gauge arterial catheter in the radial artery after induction of general anesthesia and connected it to a disposable pressure transducer. The pressure transducer was leveled to the level of the right atrium throughout the study. The transducer system was examined for its damping properties with a fast flush test, and, if necessary, actions were taken to correct abnormal damping.

To record continuous noninvasive finger cuff blood pressure, we used the finger cuff system ccNexfin (BMEYE B.V., The Netherlands)—now ClearSight (Edwards Lifesciences, USA). The ccNexfin/ClearSight system uses an inflatable finger cuff with an infrared plethysmograph to measure the blood volume in the finger arteries, keep the blood volume constant throughout the cardiac cycle based on an automated feedback system that inflates and deflates the finger cuff, and indirectly reconstruct the blood pressure waveform based on the required cuff pressure. The finger cuff was fitted and placed on the middle phalanx of the third or fourth finger according to the manufacturer’s specification for proper fit. The heart reference system that compensates for hydrostatic pressure differences between the level of the heart and the finger cuff was positioned at the level of the right atrium during the entire study period.

Intermittent noninvasive oscillometric blood pressure was measured using a standard large cuff at the upper arm, a standard adult cuff at the forearm, and a standard large cuff at the lower leg (all cuffs Criticon; GE Healthcare, USA). To reflect daily clinical practice, cuff fit was accepted when blood pressure measurements could be obtained. A cuff that would not allow for circumferential closure or that would spontaneously come undone during an inflation cycle despite the most feasible limb alignment was considered a cuff failure for that limb site, and no aids such as tape were used to rescue potential measurements. Oscillometric lower leg blood pressure measurements obtained during steep reverse Trendelenburg positioning were not included in the analysis because hydrostatic pressure differences would have influenced the difference between lower leg and intraarterial blood pressure measurements.

Intraarterial, finger cuff, upper arm, and forearm blood pressure measurements were taken at the ipsilateral arm of the patient so as to avoid any effect of blood pressure differences between both upper extremities on study measurements. For lower leg blood pressure measurements, the cuff was applied to the right leg. Intraarterial and oscillometric blood pressure measurements were displayed on the patient monitor (Philips Intellivue MP 90 anesthesia monitor; Philips Healthcare, USA). Finger cuff blood pressure measurements were displayed on and extracted from the proprietary monitor of the ccNexfin system.

We recorded mean arterial pressure, systolic blood pressure, and diastolic blood pressure at six different time points during the study period. Each measurement cycle commenced with the simultaneous recording of intraarterial and finger cuff blood pressure measurements followed by forearm, upper arm, and finally, lower leg blood pressure measurements. The six time points were chosen as follows: within 15 min before abdominal insufflation in the horizontal position; 3, 15, 30, and 45 min after the beginning of insufflation in the 30° reverse Trendelenburg position; and 3 min after desufflation in the horizontal position. Visual inspection of the intraarterial blood pressure waveform just before each point of observation allowed for correction of any observed obvious blood pressure waveform artifacts or abnormalities.

Statistical Analysis

Patient characteristics are described as mean ± SD or absolute numbers with percentages, blood pressure values as mean ± SD and range. To illustrate the distribution of finger cuff, upper arm, forearm, and lower leg blood pressure measurements and their relation to intraarterial blood pressure measurements, we calculated the Pearson correlation coefficient.

To assess the agreement between finger cuff and intraarterial blood pressure measurements, as well as between upper arm, forearm, and lower leg blood pressure measurements and intraarterial blood pressure measurements, we performed Bland–Altman analysis accounting for repeated measurements,9,10  and calculated the mean of the differences (bias), SD, and 95% limits of agreement (mean of the differences ± 1.96 × SD). The relationships between means and differences are shown in the Bland–Altman plots and were examined by linear regression analyses.

We computed four-quadrant plots with an exclusion zone of 5 mmHg and calculated the concordance rate to describe the trending ability of finger cuff, upper arm, forearm, and lower leg blood pressure measurements in comparison with intraarterial blood pressure measurements.11 

We performed continuous error grid analysis to classify measurement differences between blood pressure monitoring methods according to their clinical importance.12,13  An error grid is a scatter plot with reference blood pressure measurements on the x-axis and corresponding blood pressure measurements from the investigated method on the y-axis, and it contains five risk zones.12,13  Each blood pressure measurement pair is classified into one of five different risk zones. Those risk zones describe the risk for patients resulting from therapeutic interventions that may be triggered by measurement differences between the investigated method and the reference method and are classified as follows: risk zone A: no risk; risk zone B: low risk; risk zone C: moderate risk; risk zone D: significant risk; and risk zone E: dangerous risk.12,13  The risk zone boundaries in the error grid result from a survey among 25 experts who were first asked to define five blood pressure ranges that would, according to the experts’ experience and judgment, need to result in the following therapeutic reactions: emergency treatment for low blood pressure necessary; treatment of low blood pressure appropriate; no intervention needed; treatment of high blood pressure appropriate; and emergency treatment for high blood pressure necessary.12,13  The experts then attributed risk levels to combinations of “blood pressure measurement with investigated method is in range i” and “reference blood pressure measurement is in range j,” considering patient risk that would result from measurement differences.12,13  The experts’ aggregated and weighted answers regarding the five risk levels resulted in the error grid risk zones.12,13  Details on how smoothing polygons were fitted to the boundaries between the zones including coordinates of the resulting polygons and the aggregated risk scores are provided in our original paper.12,13  To better illustrate the different risk zones in the error grid, colors are used ranging from green (no risk) to red (dangerous risk).12,13  Patient risk resulting from measurement differences can thus be visually assessed from the color-coded continuous error grid and quantitatively analyzed by calculating absolute and relative numbers of measurement pairs within the risk zones.12,13 

For statistical analysis, we used Excel (Version 1912; Microsoft, USA), R (Version 3.6.2.; The R Foundation for Statistical Computing, Vienna, Austria), and, for error grid analysis, MATLAB (The MathWorks Inc., USA).

No statistical power calculation was conducted before the study. The sample size was based on our previous experience with method comparison studies.

We enrolled 108 patients in this study. Eighteen patients dropped out for the following reasons: technical difficulties with the insertion of the arterial catheter (n = 11); technical failures with blood pressure monitoring equipment during the procedure (n = 3); change of the surgical procedure (n = 1); patient’s withdrawal from participation (n = 1); discovery of missed exclusion criterion (n = 1); and case canceled (n = 1). Hence, we included 90 patients in the final analysis with a mean ± SD body mass index of 48 ± 7 kg/m2. Patient characteristics are summarized in table 1. Of paired blood pressure measurements available for analysis, 538 were finger cuff, 443 were upper arm, 535 were forearm, and 111 were lower leg (Supplemental Digital Content 1, http://links.lww.com/ALN/C516). Cuff failure occurred in 8 upper arm, 1 forearm, and 11 lower leg blood pressure measurements. In some cases, there was no cuff large enough to fit the upper arm (6 patients) or the lower leg (24 patients), whereas this problem did not occur on the forearm. Details on missing values are provided in Supplemental Digital Content 2 (http://links.lww.com/ALN/C517).

Table 1.

Patient Characteristics

Patient Characteristics
Patient Characteristics

The distribution of finger cuff, upper arm, forearm, and lower leg blood pressure measurements and their relation to intraarterial blood pressure measurements are shown in Supplemental Digital Content 3 (http://links.lww.com/ALN/C518).

Bland–Altman analysis revealed a mean of the differences (± SD, 95% limits of agreement) between finger cuff and intraarterial blood pressure measurements of −1 mmHg (± 11 mmHg, −23 to 21 mmHg) for mean arterial pressure, of −7 mmHg (± 14 mmHg, −35 to 20 mmHg) for systolic blood pressure, and of 0 mmHg (± 11 mmHg, −22 to 22 mmHg) for diastolic blood pressure (fig. 1A; table 2). Bland–Altman analyses for the comparison of upper arm, forearm, and lower leg blood pressure measurements with intraarterial blood pressure measurements are shown in figure 1, B through D, and table 2. For mean arterial pressure and diastolic blood pressure, the mean of the differences and SD between oscillometric (at each of the three measurement sites) and intraarterial blood pressure measurements was higher than between finger cuff and intraarterial blood pressure measurements.

Table 2.

Results of Noninvasive (Finger Cuff and Oscillometry) versus Intraarterial (Radial Artery Catheter) Blood Pressure Measurements

Results of Noninvasive (Finger Cuff and Oscillometry) versus Intraarterial (Radial Artery Catheter) Blood Pressure Measurements
Results of Noninvasive (Finger Cuff and Oscillometry) versus Intraarterial (Radial Artery Catheter) Blood Pressure Measurements
Fig. 1.

Bland–Altman plots showing the agreement between radial artery catheter–derived intraarterial blood pressure measurements and blood pressure measurements obtained using the finger cuff technology (A), upper arm oscillometry (B), forearm oscillometry (C), and lower leg oscillometry (D) for systolic blood pressure (red squares), diastolic blood pressure (blue triangles), and mean arterial pressure (black circles). The continuous horizontal lines represent the mean of the differences between the measurement methods, and the dashed horizontal lines represent the upper and lower 95% limits of agreement. The relationship between mean values and differences is shown by linear regression analyses (red line for systolic blood pressure, blue line for diastolic blood pressure, and black line for mean arterial pressure).

Fig. 1.

Bland–Altman plots showing the agreement between radial artery catheter–derived intraarterial blood pressure measurements and blood pressure measurements obtained using the finger cuff technology (A), upper arm oscillometry (B), forearm oscillometry (C), and lower leg oscillometry (D) for systolic blood pressure (red squares), diastolic blood pressure (blue triangles), and mean arterial pressure (black circles). The continuous horizontal lines represent the mean of the differences between the measurement methods, and the dashed horizontal lines represent the upper and lower 95% limits of agreement. The relationship between mean values and differences is shown by linear regression analyses (red line for systolic blood pressure, blue line for diastolic blood pressure, and black line for mean arterial pressure).

Four-quadrant plot analysis revealed a concordance rate between changes in finger cuff and in intraarterial blood pressure measurements of 88% for mean arterial pressure, 85% for systolic blood pressure, and 81% for diastolic blood pressure (fig. 2; table 2). Four-quadrant plot analyses for upper arm and forearm blood pressure measurements in comparison with intraarterial blood pressure measurements are shown in table 2 and Supplemental Digital Content 4 (http://links.lww.com/ALN/C519). For mean arterial pressure, the concordance rate between changes in oscillometric (at each of the three measurement sites) and intraarterial blood pressure measurements was lower than between changes in finger cuff and intraarterial blood pressure measurements.

Fig. 2.

Four-quadrant plot with an exclusion zone of 5 mmHg showing the trending agreement between temporal changes of radial artery catheter-derived intraarterial blood pressure measurements and finger cuff blood pressure measurements obtained using the finger cuff technology for systolic blood pressure (red squares), diastolic blood pressure (blue triangles), and mean arterial pressure (black circles).

Fig. 2.

Four-quadrant plot with an exclusion zone of 5 mmHg showing the trending agreement between temporal changes of radial artery catheter-derived intraarterial blood pressure measurements and finger cuff blood pressure measurements obtained using the finger cuff technology for systolic blood pressure (red squares), diastolic blood pressure (blue triangles), and mean arterial pressure (black circles).

Error grid analysis for the comparison between finger cuff and intraarterial blood pressure measurements revealed that the proportions of measurement pairs in risk zones A to E were 77.1%, 21.6%, 0.9%, 0.4%, and 0.0%, respectively, for mean arterial pressure and 89.5%, 9.8%, 0.2%, 0.4%, and 0.2%, respectively, for systolic blood pressure (fig. 3, A and B; table 3). Error grids analyses for upper arm, forearm, and lower leg blood pressure measurements in comparison with intraarterial blood pressure measurements are provided in table 3 and Supplemental Digital Content 5 (http://links.lww.com/ALN/C520).

Table 3.

Results of Error Grid Analysis

Results of Error Grid Analysis
Results of Error Grid Analysis
Fig. 3.

Error grid analysis comparing radial artery catheter–derived intraarterial blood pressure measurements and finger cuff blood pressure measurements obtained using the finger cuff technology for mean arterial pressure (A) and systolic blood pressure (B). The background colors correspond to the continuous risk level for each pair of measurement. The continuous risk level ranges from 0 to 100% as shown below.

Fig. 3.

Error grid analysis comparing radial artery catheter–derived intraarterial blood pressure measurements and finger cuff blood pressure measurements obtained using the finger cuff technology for mean arterial pressure (A) and systolic blood pressure (B). The background colors correspond to the continuous risk level for each pair of measurement. The continuous risk level ranges from 0 to 100% as shown below.

In this method comparison study, we compared intraarterial with finger cuff blood pressure measurements and with upper arm, forearm, and lower leg blood pressure measurements in obese patients during bariatric surgery across a wide range of blood pressure values. The absolute agreement and the trending agreement of finger cuff blood pressure measurements compared to intraarterial blood pressure measurements were only moderate. However, error grid analysis showed that about 99% of finger cuff blood pressure measurements lay in risk zones A (“no risk”) and B (“low risk”). For mean arterial pressure and diastolic blood pressure, the absolute agreement and the trending agreement between finger cuff and intraarterial blood pressure measurements were better than between oscillometric (at each of the three measurement sites) and intraarterial blood pressure measurements. Forearm blood pressure monitoring performed better than upper arm and lower leg blood pressure monitoring with regard to absolute and trending agreement with intraarterial blood pressure monitoring.

Several previous studies compared ccNexfin/ClearSight blood pressure measurements with intraarterial reference blood pressure measurements using an arterial catheter in mixed patient populations not explicitly focusing on obese patients.14  Between these studies, there is substantial variability regarding the blood pressure measurement performance of the finger cuff technology, with several studies demonstrating interchangeability between blood pressure values obtained by either method, but others showing poor agreement.14 

However, data on the measurement performance of finger cuff technologies in severely obese patients are scarce. A study in 35 severely obese patients having bariatric surgery showed that ClearSight blood pressure measurements showed good agreement with intraarterial blood pressure measurements for mean arterial pressure and diastolic blood pressure.15  In addition, trending capabilities of finger cuff blood pressure measurements were good, and more than 99% of those measurements were in no- or low-risk zones in error grid analysis.15  In a very similar patient population of 29 severely obese patients having bariatric surgery, blood pressure values measured using another commercially available finger cuff system (CNAP System; CNSystems Medizintechnik GmbH, Austria) showed moderate accuracy and precision but good trending capabilities when compared with intraarterial blood pressure measurements.16 

In contrast to these two previous studies comparing exclusively finger cuff and intraarterial blood pressure measurements, we also compared oscillometric with intraarterial blood pressure measurements in the current study. The agreement of upper arm, forearm, and lower leg blood pressure measurements with intraarterial blood pressure measurements was poor. This finding is in line with a study including 30 noncardiac surgery patients with a body mass index greater than 30 kg/m2 showing that neither upper arm nor forearm oscillometry was clinically acceptable in comparison with intraarterial blood pressure monitoring.17  The authors even used two different ways of wrapping the upper arm cuff: one following the conical shape of the upper arm found in many obese patients, the other keeping cuff edges parallel.17  A study comparing oscillometric blood pressure measurements with intraarterial blood pressure measurements during 24,225 noncardiac surgeries demonstrated that oscillometry overestimates blood pressure during periods of intraoperative hypotension.18  Hence, oscillometry may be associated with an increased risk of postoperative complications because it underestimates the severity of intraoperative hypotension. However, in our study, error grid analysis comparing upper arm, forearm, and lower leg blood pressure measurements with intraarterial blood pressure measurements revealed that more than 90% of measurement pairs were in no- or low-risk zones for mean arterial pressure.

As a new and important finding in our study, we found that the absolute agreement and the trending agreement of forearm with intraarterial blood pressure measurements were better than the agreement of upper arm with intraarterial blood pressure measurements. Our findings in patients during surgery agree with results from obese postsurgical patients, leading to the suggestion of using the forearm for oscillometry in this population.7  While these findings further justify consideration of the forearm for oscillometric blood pressure measurements as an appropriate alternative to the upper arm as a standard site, we believe further prospective studies comparing forearm and upper arm oscillometry in obese surgical patients are needed before a general recommendation of which site oscillometry should be performed in obese patients can be made.

For mean arterial pressure and diastolic blood pressure, the mean of the differences and its SD between finger cuff and intraarterial blood pressure measurements was lower, and the concordance was higher, than between oscillometric (at each of the three measurement sites) and intraarterial blood pressure measurements in the current study. Thus, for mean arterial pressure and diastolic blood pressure, the finger cuff technology was superior to oscillometry with regard to the absolute agreement and the trending agreement with intraarterial blood pressure monitoring. Additionally, the finger cuff technology—in contrast to oscillometry—enables continuous blood pressure monitoring. Therefore, the finger cuff technology may be a reasonable alternative to intermittent noninvasive oscillometric blood pressure monitoring in obese patients undergoing invasive procedures that require anesthesia or sedation. A study in patients having elective general surgery under general anesthesia showed that the agreement between finger cuff and intraarterial mean arterial pressure measurements was noninferior to the agreement of upper arm cuff oscillometry and intraarterial mean arterial pressure measurements.19  Considering that the finger cuff technology provides continuous beat-to-beat blood pressure monitoring, it may be a promising alternative blood pressure monitoring method in patients in whom oscillometry is currently being used as standard blood pressure monitoring. In a nonobese general surgical population, continuous noninvasive blood pressure monitoring using a finger cuff technology reduced the amount of intraoperative hypotension compared to intermittent oscillometric blood pressure monitoring.20,21  Studies that link such findings to postoperative outcomes for patients with severe obesity are yet to be conducted. Continuous blood pressure data from finger cuff monitoring may also be used to predict hypotension using machine learning.22,23 

Considering cost is important when choosing methods for perioperative blood pressure monitoring. Expenses for blood pressure monitoring equipment, maintenance, and appropriately trained personnel vary substantially across different healthcare settings, between different countries, and nationally, even between different healthcare systems. Insertion of arterial catheters requires trained medical personnel and is more time-consuming than applying noninvasive blood pressure monitoring using finger cuffs or oscillometry. Material costs for finger cuff blood pressure monitoring currently usually exceed costs for arterial catheter sets or oscillometry. Our research question did not focus on a cost-effectiveness analysis for the different blood pressure monitoring methods. Future studies investigating which method should be used for perioperative blood pressure monitoring should also compare personnel and material costs of the different methods in light of their impact on postoperative complications.

Patients in our study reflect a typical bariatric surgical population, rather than obese patients across an extended age range, with possibly advanced cardiovascular comorbid conditions presenting for a broader range of general surgical procedures. Such a relatively selected study population may limit the generalizability of our findings. We decided to perform all upper extremity blood pressure measurements on the same arm to avoid confounding by interarm blood pressure differences. The existence and potential impact of a partial blood flow obstruction to the hand from the arterial catheter on finger cuff blood pressure measurements are unknown and remain undetermined in our study.

In conclusion, the agreement between finger cuff and intraarterial measurements was better than the agreement between oscillometric and intraarterial measurements for mean arterial pressure and diastolic blood pressure in obese patients during surgery. In these patients, forearm oscillometry exhibits better measurement performance than upper arm or lower leg oscillometry.

Acknowledgments

The authors wish to acknowledge Omar A. Alyamani, M.B.B.S. (Assistant Professor and Consultant Physician, Department of Anesthesia and Critical Care, Faculty of Medicine, King Abdulaziz University, Jeddah, Makkah, Kingdom of Saudi Arabia), who helped with early protocol development and literature search, and Ingrid Moreno-Duarte, M.D. (Pediatric Anesthesiology Fellow, Department of Anesthesiology, University of Texas Southwestern Children’s Medical Center, Dallas, Texas), who performed preliminary data analyses used for meeting abstract compilation.

Research Support

BMEYE B.V. (Amsterdam, The Netherlands)—now Edwards Lifesciences (Irvine, California)—provided the technical equipment for the study.

Competing Interests

Dr. Schumann receives royalties as author and reviewer for obesity and sleep medicine related chapters in Up-To-Date (Wolters Kluwer; Waltham, Massachusetts). Dr. Wesselink is an employee of Edwards Lifesciences (Irvine, California). Dr. Saugel has received honoraria for consulting, honoraria for giving lectures, and refunds of travel expenses from Edwards Lifesciences; honoraria for consulting, institutional restricted research grants, honoraria for giving lectures, and refunds of travel expenses from Pulsion Medical Systems SE (Feldkirchen, Germany); institutional restricted research grants, honoraria for giving lectures, and refunds of travel expenses from CNSystems Medizintechnik GmbH (Graz, Austria); institutional restricted research grants from Retia Medical LLC (Valhalla, New York); honoraria for giving lectures from Philips Medizin Systeme Böblingen GmbH (Böblingen, Germany); and honoraria for consulting, institutional restricted research grants, and refunds of travel expenses from Tensys Medical Inc. (San Diego, California). The other authors declare no competing interesta.

1.
Cullen
A
,
Ferguson
A
:
Perioperative management of the severely obese patient: A selective pathophysiological review.
Can J Anaesth
.
2012
;
59
:
974
96
2.
Adams
JP
,
Murphy
PG
:
Obesity in anaesthesia and intensive care.
Br J Anaesth
.
2000
;
85
:
91
108
3.
Scheer
B
,
Perel
A
,
Pfeiffer
UJ
:
Clinical review: Complications and risk factors of peripheral arterial catheters used for haemodynamic monitoring in anaesthesia and intensive care medicine.
Crit Care
.
2002
;
6
:
199
204
4.
Nuttall
G
,
Burckhardt
J
,
Hadley
A
,
Kane
S
,
Kor
D
,
Marienau
MS
,
Schroeder
DR
,
Handlogten
K
,
Wilson
G
,
Oliver
WC
:
Surgical and patient risk factors for severe arterial line complications in adults.
Anesthesiology
.
2016
;
124
:
590
7
5.
Bur
A
,
Hirschl
MM
,
Herkner
H
,
Oschatz
E
,
Kofler
J
,
Woisetschläger
C
,
Laggner
AN
:
Accuracy of oscillometric blood pressure measurement according to the relation between cuff size and upper-arm circumference in critically ill patients.
Crit Care Med
.
2000
;
28
:
371
6
6.
Loenneke
JP
,
Loprinzi
PD
,
Abe
T
,
Thiebaud
RS
,
Allen
KM
,
Grant Mouser
J
,
Bemben
MG
:
Arm circumference influences blood pressure even when applying the correct cuff size: Is a further correction needed?
Int J Cardiol
.
2016
;
202
:
743
4
7.
Leblanc
,
Auclair
A
,
Leclerc
J
,
Bussières
J
,
Agharazii
M
,
Hould
FS
,
Marceau
S
,
Brassard
P
,
Godbout
C
,
Grenier
A
,
Cloutier
L
,
Poirier
P
:
Blood pressure measurement in severely obese patients: Validation of the forearm approach in different arm positions.
Am J Hypertens
.
2019
;
32
:
175
85
8.
Saugel
B
,
Dueck
R
,
Wagner
JY
:
Measurement of blood pressure.
Best Pract Res Clin Anaesthesiol
.
2014
;
28
:
309
22
9.
Bland
JM
,
Altman
DG
:
Measuring agreement in method comparison studies.
Stat Methods Med Res
.
1999
;
8
:
135
60
10.
Bland
JM
,
Altman
DG
:
Agreement between methods of measurement with multiple observations per individual.
J Biopharm Stat
.
2007
;
17
:
571
82
11.
Saugel
B
,
Grothe
O
,
Wagner
JY
:
Tracking changes in cardiac output: Statistical considerations on the 4-quadrant plot and the polar plot methodology.
Anesth Analg
.
2015
;
121
:
514
24
12.
Saugel
B
,
Grothe
O
,
Nicklas
JY
:
Error grid analysis for arterial pressure method comparison studies.
Anesth Analg
.
2018
;
126
:
1177
85
13.
Grothe
O
,
Kaplan
A
,
Kouz
K
,
Saugel
B
:
Computer program for error grid analysis in arterial blood pressure method comparison studies
.
Anesth Analg
.
2020
;
130
:
e71
e74
14.
Saugel
B
,
Hoppe
P
,
Nicklas
JY
,
Kouz
K
,
Körner
A
,
Hempel
JC
,
Vos
JJ
,
Schön
G
,
Scheeren
TWL
:
Continuous noninvasive pulse wave analysis using finger cuff technologies for arterial blood pressure and cardiac output monitoring in perioperative and intensive care medicine: A systematic review and meta-analysis.
Br J Anaesth
.
2020
;
125
:
25
37
15.
Rogge
DE
,
Nicklas
JY
,
Schön
G
,
Grothe
O
,
Haas
SA
,
Reuter
DA
,
Saugel
B
:
Continuous noninvasive arterial pressure monitoring in obese patients during bariatric surgery: An evaluation of the vascular unloading technique (ClearSight system).
Anesth Analg
.
2019
;
128
:
477
83
16.
Rogge
DE
,
Nicklas
JY
,
Haas
SA
,
Reuter
DA
,
Saugel
B
:
Continuous noninvasive arterial pressure monitoring using the vascular unloading technique (CNAP System) in obese patients during laparoscopic bariatric operations.
Anesth Analg
.
2018
;
126
:
454
63
17.
Anast
N
,
Olejniczak
M
,
Ingrande
J
,
Brock-Utne
J
:
The impact of blood pressure cuff location on the accuracy of noninvasive blood pressure measurements in obese patients: An observational study.
Can J Anaesth
.
2016
;
63
:
298
306
18.
Wax
DB
,
Lin
HM
,
Leibowitz
AB
:
Invasive and concomitant noninvasive intraoperative blood pressure monitoring: Observed differences in measurements and associated therapeutic interventions.
Anesthesiology
.
2011
;
115
:
973
8
19.
Vos
JJ
,
Poterman
M
,
Mooyaart
EA
,
Weening
M
,
Struys
MM
,
Scheeren
TW
,
Kalmar
AF
:
Comparison of continuous non-invasive finger arterial pressure monitoring with conventional intermittent automated arm arterial pressure measurement in patients under general anaesthesia.
Br J Anaesth
.
2014
;
113
:
67
74
20.
Maheshwari
K
,
Khanna
S
,
Bajracharya
GR
,
Makarova
N
,
Riter
Q
,
Raza
S
,
Cywinski
JB
,
Argalious
M
,
Kurz
A
,
Sessler
DI
:
A randomized trial of continuous noninvasive blood pressure monitoring during noncardiac surgery.
Anesth Analg
.
2018
;
127
:
424
31
21.
Meidert
AS
,
Nold
JS
,
Hornung
R
,
Paulus
AC
,
Zwißler
B
,
Czerner
S
:
The impact of continuous non-invasive arterial blood pressure monitoring on blood pressure stability during general anaesthesia in orthopaedic patients: A randomised trial.
Eur J Anaesthesiol
.
2017
;
34
:
716
22
22.
Hatib
F
,
Jian
Z
,
Buddi
S
,
Lee
C
,
Settels
J
,
Sibert
K
,
Rinehart
J
,
Cannesson
M
:
Machine-learning algorithm to predict hypotension based on high-fidelity arterial pressure waveform analysis.
Anesthesiology
.
2018
;
129
:
663
74
23.
Maheshwari
K
,
Buddi
S
,
Jian
Z
,
Settels
J
,
Shimada
T
,
Cohen
B
,
Sessler
DI
,
Hatib
F
:
Performance of the Hypotension Prediction Index with non-invasive arterial pressure waveforms in non-cardiac surgical patients
.
J Clin Monit Comput
.
2020
;
doi: 10.1007/s10877-020-00463-5 [Epub ahead of print]