Anesth Analg 2006;102:690-693
© 2006 International Anesthesia Research Society
doi: 10.1213/01.ane.0000196512.96019.e4
CARDIOVASCULAR ANESTHESIA
Vasotrac® Arterial Blood Pressure and Direct Arterial Blood Pressure Monitoring During Liver Transplantation
James Y. Findlay, FRCA,
Bhargavi Gali, MD,
Mark T. Keegan, MRCPI,
Christopher M. Burkle, MD, and
David J. Plevak, MD
Department of Anesthesiology, Mayo Clinic College of Medicine, Rochester, Minnesota
Address correspondence and reprint requests to James Y. Findlay, FRCA, Department of Anesthesiology, Mayo Clinic College of Medicine, 200 First Street SW, Rochester MN 55905. Address e-mail to findlay.james{at}mayo.edu.
 |
Abstract
|
|---|
During liver transplantation two arterial catheters are often placed. The Vasotrac® is a noninvasive monitor that provides radial arterial blood pressures by a tonometric method. We investigated whether the Vasotrac® would be an accurate substitute for an arterial catheter by comparing Vasotrac® blood pressures with simultaneous direct radial blood pressures recorded from the contralateral arm in 14 patients undergoing liver transplantation. Correlation between the two methods was calculated and a Bland-Altman analysis performed to assess agreement. A total of 6468 simultaneous measurements were made over a duration of 1.5–7.5 h per case. For mean arterial blood pressure 57% of Vasotrac® measurements were within 10% of direct arterial measurement. Correlation (r) was 0.82. Vasotrac® bias was +5.4 mm Hg and limits of agreement were ±18.6 mm Hg. For systolic arterial blood pressure 65% of Vasotrac® measurements were within 10% of direct arterial measurement. Correlation was 0.78. Vasotrac® bias was +7.6 mm Hg and limits of agreement ±25 mm Hg. For diastolic arterial blood pressure 57% of Vasotrac® measurements were within 10% of direct arterial measurement. Correlation was 0.82. Vasotrac® bias was +3.3 mm Hg and limits of agreement ±15 mm Hg. We conclude that the Vasotrac® is not adequately accurate to substitute for direct arterial blood pressure monitoring in liver transplantation.
|
Introduction
|
|---|
The Vasotrac® (Medwave Inc., Arden Hills, MN) is a noninvasive arterial blood pressure monitor that provides a radial arterial waveform and pressures by a nonocclusive tonometry method (1). This is achieved by cyclical nonocclusive compression and decompression of the radial artery by a transducer. A number of variables are measured, primarily over three beats at the time of maximum energy transfer, and blood pressures are algorithmically calculated. The accuracy of the Vasotrac® has been reported in both adult volunteers and surgical and intensive care unit patients (1,2), as well as in pediatric patients after cardiac surgery (3). In its "continuous" mode the Vasotrac® can provide an arterial blood pressure reading every 12 to 15 beats. During anesthesia for liver transplantation surgery two arterial catheters are often inserted (4). We hypothesized that the Vasotrac® would be a suitable substitute for one radial arterial catheter during liver transplantation.
|
Methods
|
|---|
After informed patient consent and IRB approval adult patients undergoing orthotopic liver transplantation were enrolled. Exclusion criteria were absence of radial pulse or the presence of an anatomic deformity of the distal forearm causing inability to place the Vasotrac® device. Patients had oscillometric arterial blood pressure measurements by cuff made in both arms. If the mean blood pressure differed by more than 5 mm Hg the patient was excluded from the study. Inability to obtain an adequately functioning arterial catheter in one arm led to the patient's removal from the study.
After the induction of general anesthesia a 20-gauge radial artery catheter was inserted into one radial artery; the side chosen was at the discretion of the attending anesthesiologist. The catheter was connected to a commercially available disposable transducer (Edwards Lifesciences, Irvine, CA) via a 4-foot noncompliant tubing. Before each study, the transducer to be used was calibrated against reference pressures of 0, 50, 100, and 150 mm Hg. After a successful self-calibration check the Vasotrac® unit was placed over the contralateral radial artery in the manner described by the manufacturer, using the appropriate sized guide. All investigators had instruction in device placement and operation by representatives of the manufacturer before commencement of the study.
The output from the arterial transducer was routed via the monitoring system (M1046A; Hewlett-Packard, Andover, MA) to a laptop computer where the arterial waveform was continually monitored. The Vasotrac® output was also routed to this computer. Computer software synchronized the two inputs and collected paired blood pressure data from the Vasotrac® measured beat and the corresponding arterial waveform. Data were stored on the laptop for subsequent analysis. Data were collected with the Vasotrac® in continuous mode for the duration of the case. Episodes in which either Vasotrac® recordings or arterial catheter recordings were lost were noted by the investigator. All data points were stored on the laptop computer for subsequent analysis.
At 1-h intervals the Vasotrac® site was inspected for signs of inadequate perfusion or venous congestion distal to the device as well as for evidence of excessive pressure beyond the expected skin conformational changes. If any of these three changes were noted, the study was terminated at that point and the Vasotrac® device removed.
All paired readings were analyzed. Systolic, mean, and diastolic blood pressures were analyzed separately. For each, the percentage of paired measurements for which the Vasotrac® pressure was within 5% and 10% of the direct arterial blood pressure were identified, as were as the percentage of measurements within 5 and 10 mm Hg of the direct measurement. Pearson product moment correlation (r) was calculated and a Bland-Altman analysis (5) performed to identify bias and limits of agreement. Continuous variables are given as mean ± sd.
At the conclusion of the data analysis the results were presented to the liver transplant anesthesiologists at our institution. They were asked if, given the results, they would consider the Vasotrac® an acceptable replacement for one arterial catheter during liver transplantation.
|
Results
|
|---|
Fourteen patients undergoing liver transplantation were enrolled; four were female. The patients were aged 48 ± 16 yr (range 34–67 yr), and average body mass index was 30 ± 8 kg/m2 (range, 23–55 kg/m2). Baseline hemodynamics revealed a mean cardiac output of 7.5 ± 2.1 L/min and systemic vascular resistance (SVR) of 648 ± 156 dynes · s/cm5. Duration of measurements ranged from 1.5 to 7.5 h, with a total of 6468 paired measurements recorded and analyzed.
Vasotrac® recordings became unobtainable during 2 cases (14%). In one patient the signal was lost during a period of low cardiac output during the anhepatic phase with subsequent signal return postrecirculation. In the other the signal was lost after 2.5 h of surgery for no identifiable reason.
The study was terminated before the end of surgery in two cases; one because of significant pressure markings without skin damage noted after 4 h, the other consequent to the finding of a cool finger distal to the device after 4 h. Neither patient suffered long-term adverse sequelae.
The correspondence of the Vasotrac® blood pressures to direct arterial blood pressures for systolic, diastolic, and mean pressures in terms of percentage difference, correlation and Bland–Altman analyses for all patients combined are given in Table 1. For mean arterial blood pressure, the correlation was 0.82. Vasotrac® bias was +5.4 mm Hg, limits of agreement were ±18.6 mm Hg. For systolic arterial blood pressure 37% of Vasotrac® measurements were within 5 mm Hg of direct arterial measurement, 60% were within 10 mm Hg. Correlation was 0.78. Vasotrac® bias was +7.6 mm Hg and limits of agreement were ±25 mm Hg. For diastolic arterial blood pressure 53% of Vasotrac® measurements were within 5 mm Hg of direct arterial measurement, 78% were within 10 mm Hg. Correlation was 0.82. Vasotrac® bias was +3.3 mm Hg and limits of agreement were ±15 mm Hg. The corresponding Bland-Altman plots are shown in Figures 1–3.
View larger version (26K):
[in this window]
[in a new window]
|
Figure 1. Bland-Altman plot for mean blood pressure showing bias (solid line) and limits of agreement (broken lines).
|
|
View larger version (27K):
[in this window]
[in a new window]
|
Figure 2. Bland-Altman plot for systolic blood pressure showing bias (solid line) and limits of agreement (broken lines).
|
|
View larger version (26K):
[in this window]
[in a new window]
|
Figure 3. Bland-Altman plot for diastolic blood pressure showing bias (solid line) and limits of agreement (broken lines).
|
|
Individual patient data were also examined. Individual patient correlations ranged between 0.73–0.98 for mean arterial blood pressure, 0.78–0.99 for systolic arterial blood pressure, and 0.69–0.95 for diastolic arterial blood pressure for 13 of 14 patients. For the remaining patient all correlations were <0.3. Individual patient biases for mean arterial blood pressure ranged between –7 and 15 mm Hg. There were no significant correlations between individual patient mean arterial blood pressure biases or limits of agreement and either body mass index or initial SVR. In addition the percentage of direct arterial blood pressures <50 mm Hg was calculated. This ranged from 0% to 6.5% (overall 2.6% of all measurements). The proportion of mean direct arterial blood pressures <50 mm Hg was not significantly correlated with individual patient mean blood pressure bias or limits of agreement.
The determination of the liver transplant anesthesiologists was that the Vasotrac® would not be an acceptable arterial blood pressure monitoring substitute for one arterial catheter during liver transplantation.
|
Discussion
|
|---|
We found that, during liver transplantation, the Vasotrac®-measured blood pressures had limits of agreement that we considered too wide to allow the substitution of Vasotrac® blood pressure measurements for direct arterial measurements during this surgery. This determination was the opinion of the members of our liver transplantation group, the clinicians who would be relying on the measurements provided to guide intraoperative management. More objective criteria for the comparison of blood pressure measurement devices have been published, but these have protocol requirements not achievable in an intraoperative study (6,7). However, using only the numeric requirements of the European Society of Hypertension criteria that combine the commonly used British Hypertension Society and Association for the Advancement of Medical Instrumentation criteria the Vasotrac® would be "not recommended" based on the results of the current study, with systolic performance being outside the Association for the Advancement of Medical Instrumentation limits (mean difference >5 mm Hg or standard deviation >8 mm Hg) and graded "D" using British Hypertension Society criteria (based on cumulative percentages of readings within 5, 10, and 15 mm Hg of the standard) (8).
One patient in our study did demonstrate notably poorer correlation between the two methods than did the remainder of patients. We retained the results from this patient in the overall analysis, as the Vasotrac® blood pressures were presented as actual measurements (i.e., without an error reported). Bias and limits of agreement analyses were reperformed with the exclusion of all data from this one patient; the results were only negligibly different from those of the analysis of the complete dataset (recalculated bias 5.9 ± 19 mm Hg (limits of agreement) for mean arterial blood pressure, 7.7 ± 24 mm Hg for systolic arterial blood pressure, and 3.2 ± 14 mm Hg for diastolic arterial blood pressure).
Our results are in contrast to previously published studies. In a study using healthy volunteers Belani et al. (2) reported the mean error for mean arterial blood pressure as –0.38 mm Hg with 95% confidence intervals of –6.5 to +5.7 mm Hg. In a subsequent multicenter study involving 17,468 paired measurements in 80 surgical and critically ill patients, the same author reported a bias of –0.2 mm Hg and precision 2.1 mm Hg (a limit of agreement of ±4.2 mm Hg) (1). More recently Cua et al. (3) compared 4102 paired measurements on 16 pediatric patients after cardiac surgery and reported a bias of –0.3 mm Hg with limits of agreement –11.8 to 11.2 mm Hg. The differences seen in our results may relate to the cardiovascular pathophysiology of our patients. Patients undergoing liver transplantation often have a high cardiac output, low peripheral resistance hemodynamic state; it is possible that alterations in arterial wall compliance associated with the vasodilated state may influence the accuracy of the Vasotrac® monitor. We were unable to demonstrate a correlation between initial SVR and Vasotrac® accuracy in our patients; however, the majority had the anticipated low SVR. Hemodynamic data are not presented in the previously reported studies, making comparisons impossible. A study involving subjects with a range of hemodynamic states would be helpful in elucidating this point. The possibility that the Vasotrac® may be less accurate at low arterial blood pressures has been previously noted because of the small number of such pressures observed in previous studies (1). Our proportion of direct arterial mean blood pressures less than 50 mm Hg was not large overall, and there was no significant correlation between the proportion of such blood pressures and individual patient bias, suggesting that a larger proportion of low arterial blood pressures does not explain the difference between our results and those previously reported.
Rapid changes in hemodynamic variables are not uncommon during liver transplantation secondary to surgical events such as clamping of major vessels or sudden blood loss; it may be that the rapid changes influence the accuracy of the Vasotrac®. Vasotrac® recordings were lost in one case during a period of low cardiac output, suggesting that this may influence the device.
Other features of liver transplantation must also be considered. Frequently, large volumes of fluids are administered, both of crystalloid and colloid solutions, in addition to red blood cell transfusion. Peripheral edema that may accompany such therapy could conceivably interfere with the monitoring ability of the Vasotrac® device, as it compresses the radial artery, or even cause displacement of an initially appropriately positioned device over time.
In two patients the Vasotrac® was removed before the end of the case secondary to concerns regarding local pressure from the device. Although we observed no adverse consequences in those cases, we are unable to be certain that if left in position there would have been morbidity associated with the findings. Until this issue can be clarified we would recommend that if the Vasotrac® is used, the site and the hand distal be available for inspection during the case.
In conclusion, we found that the Vasotrac® blood pressure monitor is not an accurate substitute for direct arterial blood pressure monitoring during surgery for liver transplantation. Hemodynamic changes are very common during liver transplantation surgery and until further work is done in this field, clinicians desiring more than one arterial blood pressure monitor should continue to place a second intraarterial catheter. Although this study considered only patients undergoing liver transplantation, our findings may also be of relevance to patients undergoing other procedures with similar hemodynamic profiles.
The authors thank Steve Dunn and the staff at Medwave for technical and analytic assistance during this study.
|
Footnotes
|
|---|
Supported, in part, by the Mayo Clinic College of Medicine.
Accepted for publication October 17, 2005.
|
References
|
|---|
- Belani KG, Ozaki M, Hynson J, et al. A new non-invasive method to measure blood pressure. Anesthesiology 1999;91:686–92.[ISI][Medline]
- Belani KG, Buckley JJ, Poliac MO. Accuracy of radial artery blood pressure determination with the Vasotrac. Can J Anesth 1999;46:488–96.[Abstract/Free Full Text]
- Cua C, Thomas K, Zurakowski D, Laussen P. A comparison of the Vasotrac with invasive arterial blood pressure monitoring in children after pediatric cardiac surgery. Anesth Analg 2005;100:1289–94.[Abstract/Free Full Text]
- Schumann R. Intraoperative resource utilization in anesthesia for liver transplantation in the United States. Anesth Analg 2003;97:21–8.[Abstract/Free Full Text]
- Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307–10.[ISI][Medline]
- Association for the Advancement of Medical Instrumentation. American national standard: electronic or automated sphygmomanometers. ANSI/AAMI SP 10-1992 Arlington, VA: AAMI, 1993.
- O’Brien F, Petrie J, Littler W, et al. The British Hypertension Society protocol for the evaluation of blood pressure measuring devices. J Hypertension 1993;11:S43–63.[ISI]
- O’Brien E, Waeber B, Parati G, et al. Blood pressure measuring devices: recommendations of the European Society of Hypertension. Br Med J 2001;322:531–6.[Free Full Text]
Top
Abstract
Introduction
