Secondary Logo

Journal Logo

The Accuracy of the CNAP®Device Compared with Invasive Radial Artery Measurements for Providing Continuous Noninvasive Arterial Blood Pressure Readings at a Medical Intensive Care Unit

A Method-Comparison Study

Smolle, Karl-Heinz MD*; Schmid, Martin MSc; Prettenthaler, Helga MD*; Weger, Christian MD

doi: 10.1213/ANE.0000000000000965
Technology, Computing, and Simulation: Research Report
Free

BACKGROUND: In cases of intensive care unit (ICU) patients without an arterial line, noninvasive continuous blood pressure monitoring could be very beneficial. The CNAP® monitor (CNSystems Medizintechnik AG) provides noninvasive, beat-to-beat blood pressure (BP) estimates using the volume clamp method to measure finger arterial pressure calibrated to brachial pressure values. The aim of this study was to compare noninvasive BP estimates of the CNAP monitor with invasive blood pressure (IBP) measurements obtained via a radial arterial catheter in unselected medical ICU patients under routine clinical conditions.

METHODS: In 40 adult patients, IBP and noninvasive CNAP blood pressure (CBP) were measured simultaneously for 30 minutes. Bland-Altman analysis accounting for repeated measurements revealed accuracy and precision of CBP toward IBP. Percentage errors were calculated using the summary measures method and tested for interchangeability. Trending analysis was assessed using 4-quadrant plots and polar plots, whereby each reported statistical calculation used the sample size of n = 40 patients.

RESULTS: A total of 7200 measurement pairs of CBP and IBP were analyzed. For mean arterial pressure, accuracy ± precision resulted in 4.6 ± 6.7 mm Hg (limits of agreement −8.7 to 17.8 mm Hg) with a percentage error of 6.77% (95% confidence interval [CI], 6.57%–6.97%). Trending analysis of 3-minute intervals showed a concordance rate of 94.6% (95% CI, 94.0%–95.2%; exclusion zone 10%) and a polar concordance rate of 99.50% (95% CI, 99.48%–99.52%) for changes lying within 10% limits.

CONCLUSIONS: The CNAP device provided feasible estimates of BP in unselected medical ICU patients under routine clinical conditions. Mean arterial pressure met interchangeability criteria for accuracy toward radial arterial pressure, as well as for percentage error, and showed good trending capabilities according to the Critchley predefined criteria.

From the *Department of Internal Medicine and Intensive Care, University Hospital Graz, Graz, Austria; Institute for Medical Engineering, University of Technology Graz, Graz, Austria; and Department of Orthopaedic Surgery, University Hospital Graz, Graz, Austria.

Accepted for publication July 6, 2015.

Funding: None.

Conflict of Interest: See Disclosures at the end of the article.

Address correspondence to Karl-Heinz Smolle, MD, Department of Internal Medicine and Intensive Care, University Hospital Graz, Auenbruggerplatz 15, A-8036 Graz, Austria. Address e-mail to karl-heinz.smolle@klinikum-graz.at.

The accurate assessment and control of arterial pressure is a key issue in the intensive care unit (ICU). Arterial pressure conveys crucial information concerning diagnosis, pathophysiology, prognosis, and treatment and is therefore monitored and used for hemodynamic optimization by >90% of anesthesiologists in the United States and Europe.1 A significant decrease in mean arterial pressure (MAP) in the critically ill patient often signifies inadequate blood flow to vital tissue: autoregulatory mechanisms in the vasculature of the brain and the kidney may fail because of impaired oxygen delivery, and the perfusion of these organs is then a direct function of blood pressure.2 Shoemaker et al.3 showed that critically ill patients who survive have a significantly higher MAP than nonsurvivors.

Conventionally, blood pressure monitoring in the ICU is done either intermittently using an oscillometric pressure device (noninvasive blood pressure [NIBP]) or continuously via an invasive arterial blood pressure (invasive blood pressure [IBP]). Because the oscillometric device provides only discontinuous pressure readings, the Association for the Advancement of Medical Instrumentation (AAMI) standard suggests measurements every 3 to 5 minutes4—hemodynamic events may be missed or noticed with delay. Furthermore, it has been shown that prolonged use of an NIBP device may cause nerve5 and skin damage.6,7 In addition to providing the clinician with easy access for blood sampling and blood gas analysis, arterial cannulation is generally considered a safe procedure.8 Nevertheless, it has also shown a few, but potentially serious, complications, including temporary and permanent artery occlusion, pseudoaneurysm, local infection, and hematoma formation.9 Wax et al.10 recommend concomitant NIBP and IBP monitoring because monitoring of only IBP was associated with an increased use of intraoperative blood transfusion, vasopressor or inotrope infusion, and antihypertensive medication administration—a tendency that may be associated with adverse outcomes. They observed that NIBP tends to show lower values than IBP at high pressures and higher values at low pressures. This was recently confirmed by Lehman et al.11 in a large retrospective study especially for systolic pressure, whereas MAP was found to be independent of measurement modality.

ICU patients who are now monitored by NIBP might benefit from noninvasive continuous blood pressure monitoring. A newly developed device for NIBP measurement (CNAP®; CNSystems Medizintechnik AG, Graz, Austria) provides beat-to-beat blood pressure readings and has already been evaluated during general anesthesia,12–17 vascular surgery,18 interventional endoscopy,19 spinal anesthesia,20–22 analgesic sedation,23 rapid ventricular pacing,24 in critically ill patients,25 in pediatric patients,26 and in the cardiac surgical ICU.27 With 2 exceptions,15,21 the CNAP device has been found to provide reliable real-time estimates of arterial pressure with clinically acceptable agreement compared with invasive arterial pressure, particularly for MAP. The accuracy of the device has, to our knowledge, not been evaluated in a general medical ICU. Therefore, the primary aim of this study was to assess the accuracy and trending ability of CNAP compared with simultaneous IBP measurements of a radial catheter in an unselected group of critically ill patients treated in a medical ICU under routine clinical conditions.

Back to Top | Article Outline

METHODS

Study Design and Patients

Table 1

Table 1

This method-comparison study using the Conformité Européene (CE)-marked CNAP device was appraised and approved by our local ethics commission for human subjects (IRB00002556, University Hospital Graz, Austria, Chairperson: Dr. Peter Rehak, EK.Nr. 20-139 ex 08/09) on July 16, 2009. Patients were included between October 2009 and September 2010 if they were 18 years or older with a body weight between 40 and 180 kg and a body mass index <35 with a clinical indication for invasive radial blood pressure monitoring. Patients with a known history of neurological or neuromuscular seizure, subjects with pronounced disturbance of peripheral blood circulation (peripheral artery occlusive disease stage II and above, Raynaud syndrome, arteriovenous shunt, etc.), subjects with vascular implants at the sites of NIBP measurement (fingers and upper arms), and subjects with excessive movements, as well as with obvious edema on the upper extremities, especially the fingers, were not included in the study. The relatives of all patients were informed about the study when the patient was included, gave written informed consent, and could refuse the patient’s participation at any time. All patients were studied on our medical ICU, were sedated, and were under vasopressor therapy (for details, see Table 1).

Back to Top | Article Outline

Measurement Procedure and Data Extraction

IBP was monitored via a radial artery catheter (20G, Arterial Cannula; BD Critical Care Systems Ltd., Singapore). Damping coefficient and natural frequency of the hydrostatic transducer system were regularly tested using the fast flush test.28

NIBP (CBP) was measured using a CNAP Monitor 500 (CNSystems Medizintechnik AG). The CNAP system consists of a double-finger cuff, a pressure transducer mounted on the forearm, and an upper-arm blood pressure cuff for calibration. The principle of CNAP, the “volume clamp method” (or “vascular unloading technique”), was originally developed by Peňáz29 in the early 1970s and further improved by Fortin et al.30 The finger cuff consists of 2 semirigid cylinders encompassing 2 neighboring fingers between digits II and IV. Within these semirigid cylinders, inflatable cuffs apply pressure to the fingers such that the blood volume flowing through the finger arteries is held constant. The finger cuff is used for continuous NIBP monitoring, one finger at a time, and switches automatically between fingers every 5 to 60 minutes (set to 30 minutes for this study as recommended by the manufacturer). An upper-arm blood pressure cuff derives measurements of oscillometric blood pressure and serves for calibration of the CNAP device every 15 to 30 minutes (set to 15 minutes for this study as recommended by the manufacturer) using a scaling operation.

The CNAP finger cuff was placed contralaterally to IBP to allow the medical staff to easily access IBP for possible treatment. Although the manufacturer recommends placing the CNAP upper-arm cuff on the same arm as the CNAP finger cuff, the upper-arm cuff was applied to the same arm as IBP to eliminate possible pressure differences of the arms. IBP and CNAP transducers were placed approximately at the level of the heart. The CNAP monitor was connected to the patient monitor and zero-leveled as recommended by the manufacturer. CNAP and IBP waveforms were synchronously displayed on the bedside patient monitor (Infinity Delta; Dräger, Lübeck, Germany) and recorded by a personal computer using data acquisition software (Dräger Data Grabber). This software allows a matched export of CBP and IBP systolic, diastolic, and mean blood pressure values, providing synchronous values every second. Although phases during standard ICU interventions (e.g., fluid administration and drug therapy) were included in the analysis, phases of intense patient care by medical staff likely to disturb IBP and/or CBP signals (e.g., IBP flushing, washing or moving the patient, blood extraction for blood gas analysis) were noted manually on a patient sheet for later exclusion from statistical analysis.

Back to Top | Article Outline

Statistical Analysis

Phases of obvious artifact (e.g., because of IBP flushing or CBP recalibration) were manually removed after visual inspection. Furthermore, phases of intense patient manipulation by medical staff were excluded from statistical analysis based on the notes on the patient sheet (see aforementioned details). From the remaining data, data of a complete 30-minute period were extracted and analyzed as follows: 10-second averages were computed, resulting in 180 consecutive systolic, diastolic, and mean pressures for IBP and CBP, respectively.

From these data, systolic, diastolic, and MAP values of CBP were plotted against the respective IBP values as scatter plots. Mean and SD of the differences between CBP and IBP were defined as accuracy and precision, respectively, and were compared with the criterion formulated by the AAMI4: according to the AAMI SP10, a minimum of 15 subjects and 10 readings per subject should be reported for the proof of accuracy for NIBP devices, whereby average differences (mean ± SD) should be below 5 ± 8 mm Hg. On the basis of these requirements, we considered analyzing 40 subjects with 180 readings per subject (i.e., a total of 7200 data points) as sufficient without further a priori power analysis.

Bland-Altman analysis accounting for multiple observations per individual was performed to derive accuracy, precision, and limits of agreement (i.e., the region within which approximately 95% of all data points should lie) according to the literature.31 To better illustrate which amount of data belong to the same subject, all data points from one patient bear the same color. To investigate whether the differences between CBP and IBP were related to MAP, linear regression analysis was performed.

Percentage errors (PEs), as well as results of the trending analysis, were calculated according to the summary measures method, that is, single values were calculated for each patient (n = 40) and then those 40 values were summarized. For the calculation of concordance rate and polar concordance rate, the percentage values were close to 100%, and therefore, the arcsine transformation was used for standard error of the mean (SEM) and calculation of confidence interval (CI). CIs with a confidence level of 95% are provided as mean ± 2.02 × SEM (2-sided t distribution with 39 degrees of freedom).

PE was calculated as 1.96 × (SD of the differences between CBP and IBP)/(mean of measurements) for each patient (n = 40) before averaging and then calculating SEM. The PE values were then compared with previously published cutoff values for interchangeability with IBP: the PE of systolic, diastolic, and mean arterial BP should be <14.7%, 17.5%, and 18.7%, respectively.12

For MAP, the ability of CBP to track changes in IBP was assessed using 4-quadrant plots and polar plots.32 Trend analysis was based on relative changes of BP computed as the percentage of change between 10-second epochs 1-minute apart, resulting in 30 trend evaluations per patient and a total of 1200 data points. The same was done for 3-minute intervals, resulting in 10 trend evaluations per patient and a total of 400 data points. In the 4-quadrant plot analysis, the level of concordance was calculated for each patient (n = 40) as the percentage of data points with the same direction of change (i.e., having the same sign) after applying exclusion zones of 5% and 10% to remove statistical noise, as recommended in the literature.32 The polar plot concordance rate was calculated for each patient (n = 40) as the percentage of data points lying within the 5% and 10% of change lines. Furthermore, polar data were converted to positive directional changes to provide half-circle polar plots and derive angular bias and radial limits of agreement.33

Statistical analysis was performed using Microsoft Office Excel 2007 (Microsoft Corporation, Redmond, WA) and SPSS for Windows (SPSS 17.0, Chicago, IL). Plotting of data was performed using Microsoft Office Excel 2007 and Matlab R2009a (The MathWorks, Natick, MA). Continuous variables are presented as mean, SD, median, and range, whereas categorical variables are presented as number and percentage.

Back to Top | Article Outline

RESULTS

Patients and Patients’ Characteristics

A total of 47 patients were included into the study. From these, 4 patients had to be excluded from statistical analysis because of technical problems (CNAP could not obtain a blood pressure signal, patients exhibited clinical signs of skin hypoperfusion of the finger). Three patients had to be excluded from statistical analysis because of operational problems: sufficient data were not available for analysis after removal of artifacts and phases of intense patient manipulation. Thus, data of 40 patients were available for statistical analysis (for patient characteristics, diagnoses, comorbidities, and clinical characteristics on the day of study inclusion see Table 1).

Back to Top | Article Outline

Data Analysis

Data of 40 patients were included in the data analysis, providing 7200 ten-second epochs. The ranges of IBP were 74 to 185 mm Hg for systolic pressure, 34 to 105 mm Hg for diastolic pressure, and 50 to 133 mm Hg for MAP; the corresponding ranges of CBP were 66 to 184 mm Hg for systolic pressure, 41 to 108 mm Hg for diastolic pressure, and 52 to 128 mm Hg for MAP. Figure 1 shows scatter plots for systolic, diastolic, and MAP values, including lines of identity. The scatter plots show that CBP measured with the CNAP device reflects IBP over a wide range of pressures. Table 2 lists the averages of IBP pressures, CBP pressures, and CBP–IBP differences, their limits of agreement, PEs, and correlation coefficients, as well as within-subject variability and within-subject precision.

Table 2

Table 2

Figure 1

Figure 1

Figure 2 shows Bland-Altman plots plus overall limits of agreement for systolic pressure, diastolic pressure, and MAP. Regression analysis of systolic pressure, diastolic pressure, and MAP revealed no significant relation between CBP–IBP differences and their average (CBP + IBP)/2 values, indicating that the differences between CBP and IBP are similar over the complete range of pressure values.

Figure 2

Figure 2

As can be seen in Figure 2, most patients remained hemodynamically stable during intensive care. This is further quantified in Table 2 by the small within-subject variability. In most patients, CBP tended to show a small bias toward IBP, but this subject-specific bias seemed to remain near-constant throughout measurement. Moreover, CBP–IBP differences did not depend on BP magnitude, which indicates that CBP is comparable with IBP over the complete range of blood pressures. Within-subject precision is best for MAPs, followed by diastolic and then systolic pressures.

Table 3

Table 3

Figure 3

Figure 3

Figure 4

Figure 4

Figure 5

Figure 5

The results of trending analysis for MAP are shown in Figures 3 to 5. Figure 3 shows the 4-quadrant trending plots of the data based on 1- and 3-minute intervals where the exclusion zones of 5% and 10% are indicated by blue squares of different shades (see figure caption for details). Figure 4 shows the polar plots of the trending data based on 1- and 3-minute intervals, whereas Figure 5 shows the half-circle polar plots of the same polar data converted to positive directional changes (see figure captions for more details). All trending analysis results are summarized in Table 3.

Back to Top | Article Outline

DISCUSSION

This study evaluates the CNAP device in unselected critically ill patients in a medical ICU. The main finding of this study is that the CNAP device could provide blood pressure estimates comparable with IBP measured in 91.5% (43 of 47 patients) of unselected ICU patients under routine clinical conditions. For MAP, accuracy ± precision met the AAMI criteria (4.6 ± 6.7 mm Hg, limits of agreement −8.7 to 17.8 mm Hg), as well as the interchangeability criterion of PE (6.77%; 95% CI, 6.57%–6.97%). The trending analysis of MAP over 1- and 3-minute intervals (levels of concordance [exclusion zone of 10%] >90%, polar concordance rates >95%, angular bias <5°, radial limits of agreement <±30°; see Table 3 for details) all suggest good trending capabilities of CNAP according to the criteria defined by Critchley et al.33 Also notable are the generally narrow CIs (see Tables 2 and 3).

The complex population of critically ill patients found at a medical ICU has been a big challenge for other NIBP monitors, so far showing mainly unfavorable results. In the study by Stover et al.,34 10 critically ill patients were monitored by using Nexfin® (BMEYE, Amsterdam, The Netherlands), which is also based on finger cuff measurements: the bias of MAP toward invasive measurements was −2 ± 8 mm Hg (no values for systolic or diastolic pressures were given), and the authors concluded that the noninvasive technique cannot properly replace an invasive system. Nowak et al.35 investigated 40 critically ill patients in their emergency department comparing the Nexfin blood pressure measurements with intermittent oscillometric NIBP measurements. They found a bias of 0.87 ± 23.34 mm Hg for systolic pressures, −1.24 ± 15.25 mm Hg for diastolic pressures, and −2.05 ± 15.89 mm Hg for MAPs, thus missing AAMI requirements in all 3 cases. A recent study by Hohn et al.36 on 25 critically ill patients evaluating the Nexfin toward invasive femoral and radial measurements showed a bias of 6 ± 12 mm Hg for MAPs and −9 ± 25 mm Hg for systolic pressures (values for diastolic pressures not given), thus missing the AAMI requirements in both cases. In the study by Ameloot et al.,37 Nexfin was validated toward invasive femoral measurements in 45 critically ill adult patients at five 2-hour intervals: they found a bias of 8.3 ± 13.8 mm Hg, −9.4 ± 6.9 mm Hg, and −1.8 ± 5.1 mm Hg for systolic, diastolic, and MAP, respectively. Trending analysis over the intervals revealed a level of concordance of 85.5% (exclusion zone of 15%) and a polar concordance rate of 96.7% within the 10% lines.

Our results conform to the basic AAMI requirements for MAPs (4.6 ± 6.7 mm Hg). For diastolic pressures, the AAMI requirements are missed in terms of bias (7.0 mm Hg) but not in terms of SD (±5.7 mm Hg), indicating that a systematic offset exists between CNAP and IBP for diastolic pressures, which might be because of the calibration toward brachial pressure. For systolic pressures, the AAMI requirements are met in terms of bias (−3.2 mm Hg) but not in terms of SD (±10.1 mm Hg). As for other NIBP monitors, estimates of systolic pressure seem the hardest to achieve with sufficient precision. However, the results by Lehman et al.11 suggest that MAP (rather than systolic or diastolic pressure) should be the preferred metric in the ICU to guide therapy.

In a recent meta-analysis, Kim et al.38 reviewed the presentation of accuracy and precision in evaluation reports of continuous noninvasive BP monitoring devices. They found that there exists a large heterogeneity between studies and their way of evaluating interchangeability between measurement methods. Moreover, they point out that the AAMI guidelines, which were initially developed for oscillometric NIBP evaluation but are also used for the evaluation of continuous BP measurement devices, can be and often are misinterpreted. We share the view of Kim et al.38 that the community “should adopt more specific standards for conducting and reporting method-comparison studies.” Most importantly, trending analysis should be reported along with Bland-Altman analysis.

Our trending analysis results investigated 2 different time intervals and 2 exclusion zone sizes for the 4-quadrant plots: as can be seen in Table 3, the concordance rate values are similar regardless of time interval length or exclusion zone size. Although exclusion zones were introduced to remove statistical noise of small changes,32 they naturally remove a large amount of data points when hemodynamically stable patients are investigated, and thus, changes of BP over 1-minute or even 3-minute intervals are relatively small. Therefore, our results should not be overinterpreted. Likewise, although the results for angular bias are low for both studied time intervals with radial limits of agreement within the recommended values of ±30°,33 these results should not be overrated. However, the examination of polar concordance rate in the full polar plot might be more informative: investigating the distance of the data points from the horizontal polar axis (which corresponds to the line of identity in the 4-quadrant plots), the percentage of data points lying within specified limits of change can be calculated without applying an exclusion zone, and thus use the complete set of data points. Here, the results of 98.25% and 99.25% for 5% limits of change, as well as those of 99.50% and 99.83% for 10% limits of change (all with narrow CIs), can be considered as showing good trending capabilities according to the Critchley criteria.33

Our study has several limitations: First, we report results based on critically ill patients in an intensive care setting at a single center under routine clinical conditions, which may not easily apply to other patient groups. Second, the majority of patients were hemodynamically stable, and thus, blood pressure changes requiring immediate treatment were rarely observed. Third, no subgroup of patients with the same medical problem was large enough to allow for separate analysis. Thus, further studies should be performed to investigate the performance of the device on less hemodynamically stable patients and to study more pronounced blood pressure changes (e.g., because of pain, sympathetic stimulation, or cardiovascular instability). Other studies should also focus on the agreement of CNAP to invasive monitoring during specific drug or fluid administrations. Naturally, studying such interventions could provide trending data which, to the most part, can be expected to lie outside exclusion zones and thus make results more reliable and informative.

In conclusion, the CNAP monitor measured mean arterial blood pressure noninvasively and continuously, meeting basic AAMI requirements, as well as interchangeability criteria for PE, and showing good trending capability in critically ill patients at our medical ICU. We agree with Stover et al. who state that noninvasive finger technologies, being easy to use and install, might offer a fast initial hemodynamic overview and provide important information on the patient’s blood pressure trend (i.e., stable or deteriorating).34 The CNAP monitor might thus be useful in bridging the time until invasive monitoring is finally installed, as well as in cases where invasive monitoring is not indicated.

Back to Top | Article Outline

DISCLOSURES

Name: Karl-Heinz Smolle, MD.

Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.

Attestation: Karl-Heinz Smolle attests to the integrity of the original data and the analysis reported in this manuscript and is the archival author. He approved the final manuscript.

Conflicts of Interest: None.

Name: Martin Schmid, MSc.

Contribution: This author helped collect and analyze the data and write the manuscript.

Attestation: Martin Schmid attests to the integrity of the original data and the analysis reported in this manuscript. He approved the final manuscript.

Conflicts of Interest: Martin Schmid received a grant from CNSystems Medizintechnik AG for this work.

Name: Helga Prettenthaler, MD.

Contribution: This author helped design and conduct the study and write the manuscript.

Attestation: Helga Prettenthaler approved the final manuscript.

Conflicts of Interest: None.

Name: Christian Weger, MD.

Contribution: This author helped design and conduct the study.

Attestation: Christian Weger approved the final manuscript.

Conflicts of Interest: None.

This manuscript was handled by: Maxime Cannesson, MD, PhD.

Back to Top | Article Outline

ACKNOWLEDGMENTS

The CNAP monitor was kindly provided by CNSystems Medizintechnik AG (Graz, Austria) on a loan basis. The authors thank Dräger Medical Systems, Vienna, for providing data collection software.

Back to Top | Article Outline

REFERENCES

1. Cannesson M, Pestel G, Ricks C, Hoeft A, Perel A. Hemodynamic monitoring and management in patients undergoing high risk surgery: a survey among North American and European anesthesiologists. Crit Care. 2011;15:R197
2. Andrews FJ, Nolan JP. Critical care in the emergency department: monitoring the critically ill patient. Emerg Med J. 2006;23:561–4
3. Shoemaker WC, Wo CC, Chan L, Ramicone E, Kamel ES, Velmahos GC, Belzberg H. Outcome prediction of emergency patients by noninvasive hemodynamic monitoring. Chest. 2001;120:528–37
4. Association for the Advancement of Medical Instrumentation: American National Standard ANSI/AAMISP:10.2002. Manual, Electronic, or Automated Sphygmomanometers. 2003 Arlington, VA AAMI:30–57
5. Lin CC, Jawan B, de Villa MV, Chen FC, Liu PP. Blood pressure cuff compression injury of the radial nerve. J Clin Anesth. 2001;13:306–8
6. Bause GS, Weintraub AC, Tanner GE. Skin avulsion during oscillometry. J Clin Monit. 1986;2:262–3
7. Pedley CF, Bloomfield RL, Colflesh MJ, Rodriguez-Porcel M, Porcel MR, Novikov SV. Blood pressure monitor-induced petechiae and ecchymoses. Am J Hypertens. 1994;7:1031–2
8. Brzezinski M, Luisetti T, London MJ. Radial artery cannulation: a comprehensive review of recent anatomic and physiologic investigations. Anesth Analg. 2009;109:1763–81
9. 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
10. 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
11. Lehman LW, Saeed M, Talmor D, Mark R, Malhotra A. Methods of blood pressure measurement in the ICU. Crit Care Med. 2013;41:34–40
12. Ilies C, Bauer M, Berg P, Rosenberg J, Hedderich J, Bein B, Hinz J, Hanss R. Investigation of the agreement of a continuous non-invasive arterial pressure device in comparison with invasive radial artery measurement. Br J Anaesth. 2012;108:202–10
13. Jeleazcov C, Krajinovic L, Münster T, Birkholz T, Fried R, Schüttler J, Fechner J. Precision and accuracy of a new device (CNAPTM) for continuous non-invasive arterial pressure monitoring: assessment during general anaesthesia. Br J Anaesth. 2010;105:264–72
14. Hahn R, Rinösl H, Neuner M, Kettner SC. Clinical validation of a continuous non-invasive haemodynamic monitor (CNAP&trade; 500) during general anaesthesia. Br J Anaesth. 2012;108:581–5
15. Gayat E, Mongardon N, Tuil O, Sievert K, Chazot T, Liu N, Fischler M. CNAP(®) does not reliably detect minimal or maximal arterial blood pressures during induction of anaesthesia and tracheal intubation. Acta Anaesthesiol Scand. 2013;57:468–73
16. Biais M, Stecken L, Ottolenghi L, Roullet S, Quinart A, Masson F, Sztark F. The ability of pulse pressure variations obtained with CNAP™ device to predict fluid responsiveness in the operating room. Anesth Analg. 2011;113:523–8
17. Dewhirst E, Corridore M, Klamar J, Beebe A, Rice J, Barry N, Tobias JD. Accuracy of the CNAP monitor, a noninvasive continuous blood pressure device, in providing beat-to-beat blood pressure readings in the prone position. J Clin Anesth. 2013;25:309–13
18. Biais M, Vidil L, Roullet S, Masson F, Quinart A, Revel P, Sztark F. Continuous non-invasive arterial pressure measurement: evaluation of CNAP device during vascular surgery. Ann Fr Anesth Reanim. 2010;29:530–5
19. Siebig S, Rockmann F, Sabel K, Zuber-Jerger I, Dierkes C, Brünnler T, Wrede CE. Continuous non-invasive arterial pressure technique improves patient monitoring during interventional endoscopy. Int J Med Sci. 2009;6:37–42
20. Ilies C, Kiskalt H, Siedenhans D, Meybohm P, Steinfath M, Bein B, Hanss R. Detection of hypotension during Caesarean section with continuous non-invasive arterial pressure device or intermittent oscillometric arterial pressure measurement. Br J Anaesth. 2012;109:413–9
21. McCarthy T, Telec N, Dennis A, Griffiths J, Buettner A. Ability of non-invasive intermittent blood pressure monitoring and a continuous non-invasive arterial pressure monitor (CNAP™) to provide new readings in each 1-min interval during elective caesarean section under spinal anaesthesia. Anaesthesia. 2012;67:274–9
22. Sia AT, Tan HS, Sng BL. Closed-loop double-vasopressor automated system to treat hypotension during spinal anaesthesia for caesarean section: a preliminary study. Anaesthesia. 2012;67:1348–55
23. Schramm C, Baat L, Plaschke K. Continuous noninvasive arterial pressure: assessment in older and high-risk patients under analgesic sedation. Blood Press Monit. 2011;16:270–6
24. Schramm C, Huber A, Plaschke K. The accuracy and responsiveness of continuous noninvasive arterial pressure during rapid ventricular pacing for transcatheter aortic valve replacement. Anesth Analg. 2013;117:76–82
25. Monnet X, Dres M, Ferré A, Le Teuff G, Jozwiak M, Bleibtreu A, Le Deley MC, Chemla D, Richard C, Teboul JL. Prediction of fluid responsiveness by a continuous non-invasive assessment of arterial pressure in critically ill patients: comparison with four other dynamic indices. Br J Anaesth. 2012;109:330–8
26. Kako H, Corridore M, Rice J, Tobias JD. Accuracy of the CNAP™ monitor, a noninvasive continuous blood pressure device, in providing beat-to-beat blood pressure readings in pediatric patients weighing 20-40 kilograms. Paediatr Anaesth. 2013;23:989–93
27. Jagadeesh AM, Singh NG, Mahankali S. A comparison of a continuous noninvasive arterial pressure (CNAP™) monitor with an invasive arterial blood pressure monitor in the cardiac surgical ICU. Ann Card Anaesth. 2012;15:180–4
28. Gardner RM. Direct blood pressure measurement–dynamic response requirements. Anesthesiology. 1981;54:227–36
29. Peňáz JAlbert R, Vogt WS, Helberg W. Photoelectric measurement of blood pressure, volume and flow in the finger. In: Digest of the 10th International Conference on Medical and Biological Engineering. 1973:104
30. Fortin J, Marte W, Grüllenberger R, Hacker A, Habenbacher W, Heller A, Wagner Ch, Wach P, Skrabal F. Continuous non-invasive blood pressure monitoring using concentrically interlocking control loops. Comput Biol Med. 2006;36:941–57
31. Bland JM, Altman DG. Agreement between methods of measurement with multiple observations per individual. J Biopharm Stat. 2007;17:571–82
32. Critchley LA, Lee A, Ho AM. A critical review of the ability of continuous cardiac output monitors to measure trends in cardiac output. Anesth Analg. 2010;111:1180–92
33. Critchley LA, Yang XX, Lee A. Assessment of trending ability of cardiac output monitors by polar plot methodology. J Cardiothorac Vasc Anesth. 2011;25:536–46
34. Stover JF, Stocker R, Lenherr R, Neff TA, Cottini SR, Zoller B, Béchir M. Noninvasive cardiac output and blood pressure monitoring cannot replace an invasive monitoring system in critically ill patients. BMC Anesthesiol. 2009;9:6
35. Nowak RM, Sen A, Garcia AJ, Wilkie H, Yang JJ, Nowak MR, Moyer ML. Noninvasive continuous or intermittent blood pressure and heart rate patient monitoring in the ED. Am J Emerg Med. 2011;29:782–9
36. Hohn A, Defosse JM, Becker S, Steffen C, Wappler F, Sakka SG. Non-invasive continuous arterial pressure monitoring with Nexfin does not sufficiently replace invasive measurements in critically ill patients. Br J Anaesth. 2013;111:178–84
37. Ameloot K, Van De Vijver K, Van Regenmortel N, De Laet I, Schoonheydt K, Dits H, Broch O, Bein B, Malbrain ML. Validation study of Nexfin® continuous non-invasive blood pressure monitoring in critically ill adult patients. Minerva Anestesiol. 2014;80:1294–301
38. Kim SH, Lilot M, Sidhu KS, Rinehart J, Yu Z, Canales C, Cannesson M. Accuracy and precision of continuous noninvasive arterial pressure monitoring compared with invasive arterial pressure: a systematic review and meta-analysis. Anesthesiology. 2014;120:1080–97
© 2015 International Anesthesia Research Society