- Email: [email protected]
Use of an Esophageal Echo-Doppler Device During Liver Transplantation: Preliminary Report
Use of an Esophageal Echo-Doppler Device During Liver Transplantation: Preliminary Report V. Perilli, A.W. Avolio, T. Sacco, C. Modesti, R. Gaspari, R. Caserta, S. Agnes, and L. Sollazzi ABSTRACT Determination of cardiac output (CO) is crucial for perioperative monitoring of orthotopic liver transplant (OLT) recipients. A pulmonary artery catheter (PAC) has always been considered the “gold standard” of hemodynamic monitoring. The aim of this study was to evaluate the suitability of a transesophageal echo-Doppler device (ED) as a minimally invasive device to measure CO in OLT. ED was compared with the standard PAC technique taking into account the disease severity of OLT recipients as defined by the model for end-stage liver disease (MELD) score. We enrolled 42 cirrhotic patients scheduled for OLT 3 thermodilution CO measurements were taken by a PAC and the most recent ED measurement (COED) was also recorded. Paired measurements of CO were performed at standard times, unless there were additional clinical needs. Recipients were stratified into 3 groups according to MELD score: MELD score ⱕ15 (14 patients); MELD score between 16 and 28 (17 patients); and MELD score ⱖ29 (11 patients). We performed 495 paired measurements of CO. Mean bias was 0.34 ⫾ 0.9 L/min and limits of agreement were ⫺1.46 and 2.14 L/min. In patients with MELD score ⬍15, the bias was 0.12 ⫾ 0.55. The ED results were not interchangeable with PAC, because of the large limits of agreement. However, in cirrhotic patients with MELD scores ⬍15, the precision of the new method was similar to that of PAC; therefore, in this subset of patients, it may represent a reliable alternative to PAC.
P
ULMONARY ARTERY CATHETER (PAC) has always been considered the “gold standard” of hemodynamic monitoring techniques, even though the recent introduction of new minimally invasive instruments into anesthetic practice may cause a reevaluation of this opinion.1,2 Minimally invasive methods to measure cardiac output (CO) in critically ill surgical patients include transesophageal echocardiography, arterial waveform analysis, thoracic impedance, the modified Fick technique, and esophageal Doppler (ED).3– 8 Two studies comparing ED with PAC throughout orthotopic liver transplantation (OLT) showed conflicting results, while another study compared these techniques only during hepatic vascular exclusion.9 –11 The aims of the present study were: (1) to compare CO measurements by ED (COED) with those obtained by bolus thermodilution (COPAC) in the OLT setting, and (2) to investigate whether the degree of liver function decompensation, as measured by the model for end-stage liver disease (MELD), affected the correlation between the 2 techniques. 0041-1345/09/$–see front matter doi:10.1016/j.transproceed.2008.09.054 198
MATERIALS AND METHODS After institutional approval and written informed consent, we included 42 adult patients scheduled for OLT, excluding acute liver failure patients. Venovenous bypass (VVBP) was always used with monitoring during surgery performed according to our standards; in addition to a PAC (Opticath, Abbott, Ill, United States), we utilized an esophageal echo-Doppler device (Hemosonic-Arrow, Reading, Penn, United States) to provide continuous, minimally invasive CO measurements. The ED probe was gently inserted orally after tracheal intubation according to the standard technique.8 All Doppler measurements were performed by the same clinicians (V.P., L.S., T.S., and C.M.) who were experienced in the echo-Doppler technique and were blinded to the thermodilution CO values.
From Department of Anaesthesia and Intensive Care (V.P., T.S., C.M., R.G., R.C., L.S.), Department of Surgery – Transplantation Service (A.W.A., S.A.), Catholic University of Rome, Italy. Address reprint requests to Alfonso Wolfango Avolio, MD, Catholic University of Rome, L. go Gemelli 8, Rome 00168, Italy. E-mail: [email protected] © 2009 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710 Transplantation Proceedings, 41, 198 –200 (2009)
ESOPHAGEAL ECHO-DOPPLER DEVICE DURING OLT
199
Three separate thermodilution measurements of CO were randomly performed within the respiratory cycle by injection of 10 mL of 5% dextrose water at room temperature. Paired measurements of CO were performed at standard times unless there were additional clinical needs: (1) after tracheal intubation; (2) after laparotomy; (3) during hepatectomy; (4) just before VVBP; (5) at the beginning of VVBP; (6) during VVBP; (7) at the end of portal bypass; (8) after venous reperfusion; (9) after arterial reperfusion; and (10) at the end of surgery. Recipients were stratified into 3 groups according to their biochemical MELD scores: MELD score ⱕ15 (14 patients); MELD score between 16 and 28 (17 patients); and MELD score ⱖ29 (11 patients). Bland-Altman analysis was performed by plotting the differences between the means of COPAC and COED versus the average of the means.12,13 In addition, the mean percentage error (PE) was calculated (2 SD/mean), as well as the relative error by the difference between each 2-paired measurement, related to their average value12,13:
These statistical parameters were calculated both for the entire dataset as well as for each MELD group. Results are presented as mean values ⫾ SD. P ⬍ .05 was considered statistically significant.
RESULTS
We performed 495 paired CO measurements; however, signal detection was not optimal in 55 determinations (11.1%), which were excluded for the comparisons. ED was inserted uneventfully in all patients with no episode of gastrointestinal bleeding. The mean difference between the paired values of the entire dataset, representing the mean bias of ED measurements with respect to PAC, was 0.34 L/min. The limits of agreement (2 SD) were ⫺1.46 and 2.14 L/min); the PE was 24.1% and the mean relative percentage error was 3.85% ⫾ 11%. Moreover, we observed a positive correlation between COPAC and the relative error (r ⫽ .41, P ⬍ .01; Fig 1). As 40
30
RELATIVE ERROR
20
10
0
-10
-20
4
6
8
10
12
14
16
18
COED COPAC PE Relative error Bias
MELD ⱕ15
MELD ⬎15–⬍29
MELD ⱖ29
6.75 ⫾ 2.04 6.75 ⫾ 1.87 12.8 1.66 ⫾ 8.1 0.12 ⫾ 0.55
8.5 ⫾ 3.5* 8.85 ⫾ 3.5* 22.9 4.67 ⫾ 11.3* 0.38 ⫾ 1.10*
7.44 ⫾ 1.77 7.67 ⫾ 1.56 19.6 4.41 ⫾ 11.05* 0.33 ⫾ 0.84*
COED, CO from esophageal echo-Doppler; COPAC, CO from pulmonary artery catheter; PE, mean percentage of error; relative error, relative error calculated as follows: (COPAC⫺ COED) * 100/0.5* (COPAC ⫹ COED). *P ⬍ .05 compared with MELD ⱕ15 group.
for the 3 patient groups, the low MELD group showed a significantly lower mean bias and relative error than the other 2 groups (P ⬍ .05; Table 1). DISCUSSION
(COPAC ⫺ COED) * 100 ⁄ 0.5* (COPAC ⫹ COED)
-30 2
Table 1. CO Measurements and Statistical Parameters in the 3 MELD Groups
20
22
CO PAC
Fig 1. Regression analysis between relative error and cardiac output measured by PAC (CO PAC; r ⫽ .41, P ⬍ .01).
Our data showed an underestimation of CO using ED compared with PAC with only a small mean systemic bias (0.34 L/min), a fairly large limit of agreement (⫺1.46 L and 2.14 L/min), and a percentage of error of 24.1%. These results were consistent with previous studies revealing a satisfactory accuracy of the measurements; the relatively large scatter of differences between the 2 methods involved the combination of errors of each measurement technique, reflecting the lack of precision of each.4,9,10,13 Since the average values of triplicate, randomly injected series of thermodilution measurements show errors of 10%, according to Critchley’s statistical criteria the error of the new method was calculated as follows: 兹共24.12 ⫺ 102兲 ⫽ 21.9%.12,13,14 Under these conditions, the 2 methods of measurement may not be defined as interchangeable, since the new method (ED) was significantly less precise than the old one (PAC; 21.9% vs 10%).13 However, when we evaluated these statistical parameters in each MELD group, we observed a significant reduction in the error among the low severity patients (Table 1). In this group, the precision of the new method was quite similar to that of the old one (12.8% vs 10%), and the mean bias was close to zero (0.12 L/min), which may indicate that the new technique is clinically suitable for cirrhotic patients with MELD scores ⬍15. The different behavior of more critically cirrhotic patients has been observed also with other noninvasive measurement techniques.15 It may have different explanations: First, a weak positive correlation between COPAC and relative error; therefore, more severely cirrhotic patients, who usually have higher CO values, may exhibit a lower precision of echo-Doppler CO measurements. Second, a redistribution of blood flow among body compartments may take place in more severely cirrhotic recipients due to profound alterations of their arterial tone, and vasoactive stimulation.11,16 These factors could influence CO measurements making the echo-Doppler device more precise in low-risk patients. The present study suffers from several limitations: (1) the relative absence of patients with low CO; (2) the use of a
200
standard bolus thermodilution technique for comparison, since more recent models of PAC have been proposed; and (3) the impossibility to evaluate reproducibility in our series as only one pair of values was recorded for each patient at a given surgical time. In conclusion, ED cannot replace PAC for the anesthetic management of severely cirrhotic patients undergoing OLT because of its larger error (21.9% vs 10%).14 However, in patients with biochemical MELD scores ⬍15, including many subjects with hepatocellular carcinoma, this device has the potential to yield reliable values of CO and replace PAC. In these circumstances, one must consider the advantage of a minimally invasive monitoring technique.
REFERENCES 1. Diaz S, Perez-Pena J, Sanz J, et al: Haemodynamic monitoring and liver function evaluation by pulsion cold system Z-201 (PCS) during orthotopic liver transplantation. Clin Transplant 17:47, 2003 2. Burternshaw AJ, Isaac L: The role of trans-oesophageal echocardiography for perioperative cardiovascular monitoring during orthotopic liver transplantation. Liver Transpl 12:1577, 2007 3. Berton C, Cholley B: Equipment review: new techniques for cardiac output, oesophageal Doppler, Fick principle using carbon dioxide, pulse contour analysis. Crit Care 6:216, 2002 4. Botero M, Lobato EB: Advances in noninvasive cardiac output monitoring: an update. J Cardiothorac Vasc Anesth 15:631, 2001 5. Cholley B, Payen D: Noninvasive techniques for measurements of cardiac output. Curr Opin Crit Care 11:424, 2005
PERILLI, AVOLIO, SACCO ET AL 6. Laupland KB, Bands CJ: Utility of esophageal Doppler as a minimally invasive hemodynamic monitor: a review. Can J Anesth 49:393, 2002 7. Singer M: Esophageal Doppler monitoring. In Pinsky MR, Payen D (eds): Functional Hemodynamic Monitoring. New York: Springer; 2005, p 193 8. Cariou A, Monchi M, Joly LM, et al: Noninvasive cardiac output monitoring by aortic blood flow determination: evaluation of the Sometec Dynemo 3000. Crit Care Med 26:2066, 1998 9. Colbert S, O’Hanlon DM, Duranteau J, et al: Cardiac output during liver transplantation. Can J Anesthesiol 45:133, 1988 10. Odenstedt H, Anemen A, Svensson Y, et al: Descendent aortic blood flow and cadiac output. A clinical and experimental study of continuous oesophageal echo-Doppler flowmetry. Acta Anaesthesiol Scand 45:180, 2001 11. Boucaud C, Bouffard Y, Dumortier J, et al: Transesophageal echo Doppler vs thermodilution cardiac output measurement during hepatic vascular exclusion in liver transplantation. Eur J Anaesthesiol 25:485, 2008 12. Bland JM, Altmann GD: Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1:307, 1986 13. Critchley LAH, Critchley JAJH: A meta-analysis of studies using bias and precision statistics to compare cardiac output measurement techniques. J Clin Monit Comput 15:85, 1999 14. Stetz CW, Miller RG, Kelly GE: Reliability of thermodilution method in the determination of cardiac output in clinical practice. Am Rev Respir Dis 125:1001, 1982 15. Biais M, Nouette-Gaulain K, Cottenceau V, et al: Cardiac output measurement in patients undergoing liver transplantation: pulmonary artery catheter versus uncalibrated arterial pressure waveform analysis. Anesth Analg 106:1480, 2008 16. Kamal GD, Symreng T, Starr J: Inconsistent esophageal Doppler cardiac output during acute blood loss. Anesthesiology 72:95, 1990
P
ULMONARY ARTERY CATHETER (PAC) has always been considered the “gold standard” of hemodynamic monitoring techniques, even though the recent introduction of new minimally invasive instruments into anesthetic practice may cause a reevaluation of this opinion.1,2 Minimally invasive methods to measure cardiac output (CO) in critically ill surgical patients include transesophageal echocardiography, arterial waveform analysis, thoracic impedance, the modified Fick technique, and esophageal Doppler (ED).3– 8 Two studies comparing ED with PAC throughout orthotopic liver transplantation (OLT) showed conflicting results, while another study compared these techniques only during hepatic vascular exclusion.9 –11 The aims of the present study were: (1) to compare CO measurements by ED (COED) with those obtained by bolus thermodilution (COPAC) in the OLT setting, and (2) to investigate whether the degree of liver function decompensation, as measured by the model for end-stage liver disease (MELD), affected the correlation between the 2 techniques. 0041-1345/09/$–see front matter doi:10.1016/j.transproceed.2008.09.054 198
MATERIALS AND METHODS After institutional approval and written informed consent, we included 42 adult patients scheduled for OLT, excluding acute liver failure patients. Venovenous bypass (VVBP) was always used with monitoring during surgery performed according to our standards; in addition to a PAC (Opticath, Abbott, Ill, United States), we utilized an esophageal echo-Doppler device (Hemosonic-Arrow, Reading, Penn, United States) to provide continuous, minimally invasive CO measurements. The ED probe was gently inserted orally after tracheal intubation according to the standard technique.8 All Doppler measurements were performed by the same clinicians (V.P., L.S., T.S., and C.M.) who were experienced in the echo-Doppler technique and were blinded to the thermodilution CO values.
From Department of Anaesthesia and Intensive Care (V.P., T.S., C.M., R.G., R.C., L.S.), Department of Surgery – Transplantation Service (A.W.A., S.A.), Catholic University of Rome, Italy. Address reprint requests to Alfonso Wolfango Avolio, MD, Catholic University of Rome, L. go Gemelli 8, Rome 00168, Italy. E-mail: [email protected] © 2009 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710 Transplantation Proceedings, 41, 198 –200 (2009)
ESOPHAGEAL ECHO-DOPPLER DEVICE DURING OLT
199
Three separate thermodilution measurements of CO were randomly performed within the respiratory cycle by injection of 10 mL of 5% dextrose water at room temperature. Paired measurements of CO were performed at standard times unless there were additional clinical needs: (1) after tracheal intubation; (2) after laparotomy; (3) during hepatectomy; (4) just before VVBP; (5) at the beginning of VVBP; (6) during VVBP; (7) at the end of portal bypass; (8) after venous reperfusion; (9) after arterial reperfusion; and (10) at the end of surgery. Recipients were stratified into 3 groups according to their biochemical MELD scores: MELD score ⱕ15 (14 patients); MELD score between 16 and 28 (17 patients); and MELD score ⱖ29 (11 patients). Bland-Altman analysis was performed by plotting the differences between the means of COPAC and COED versus the average of the means.12,13 In addition, the mean percentage error (PE) was calculated (2 SD/mean), as well as the relative error by the difference between each 2-paired measurement, related to their average value12,13:
These statistical parameters were calculated both for the entire dataset as well as for each MELD group. Results are presented as mean values ⫾ SD. P ⬍ .05 was considered statistically significant.
RESULTS
We performed 495 paired CO measurements; however, signal detection was not optimal in 55 determinations (11.1%), which were excluded for the comparisons. ED was inserted uneventfully in all patients with no episode of gastrointestinal bleeding. The mean difference between the paired values of the entire dataset, representing the mean bias of ED measurements with respect to PAC, was 0.34 L/min. The limits of agreement (2 SD) were ⫺1.46 and 2.14 L/min); the PE was 24.1% and the mean relative percentage error was 3.85% ⫾ 11%. Moreover, we observed a positive correlation between COPAC and the relative error (r ⫽ .41, P ⬍ .01; Fig 1). As 40
30
RELATIVE ERROR
20
10
0
-10
-20
4
6
8
10
12
14
16
18
COED COPAC PE Relative error Bias
MELD ⱕ15
MELD ⬎15–⬍29
MELD ⱖ29
6.75 ⫾ 2.04 6.75 ⫾ 1.87 12.8 1.66 ⫾ 8.1 0.12 ⫾ 0.55
8.5 ⫾ 3.5* 8.85 ⫾ 3.5* 22.9 4.67 ⫾ 11.3* 0.38 ⫾ 1.10*
7.44 ⫾ 1.77 7.67 ⫾ 1.56 19.6 4.41 ⫾ 11.05* 0.33 ⫾ 0.84*
COED, CO from esophageal echo-Doppler; COPAC, CO from pulmonary artery catheter; PE, mean percentage of error; relative error, relative error calculated as follows: (COPAC⫺ COED) * 100/0.5* (COPAC ⫹ COED). *P ⬍ .05 compared with MELD ⱕ15 group.
for the 3 patient groups, the low MELD group showed a significantly lower mean bias and relative error than the other 2 groups (P ⬍ .05; Table 1). DISCUSSION
(COPAC ⫺ COED) * 100 ⁄ 0.5* (COPAC ⫹ COED)
-30 2
Table 1. CO Measurements and Statistical Parameters in the 3 MELD Groups
20
22
CO PAC
Fig 1. Regression analysis between relative error and cardiac output measured by PAC (CO PAC; r ⫽ .41, P ⬍ .01).
Our data showed an underestimation of CO using ED compared with PAC with only a small mean systemic bias (0.34 L/min), a fairly large limit of agreement (⫺1.46 L and 2.14 L/min), and a percentage of error of 24.1%. These results were consistent with previous studies revealing a satisfactory accuracy of the measurements; the relatively large scatter of differences between the 2 methods involved the combination of errors of each measurement technique, reflecting the lack of precision of each.4,9,10,13 Since the average values of triplicate, randomly injected series of thermodilution measurements show errors of 10%, according to Critchley’s statistical criteria the error of the new method was calculated as follows: 兹共24.12 ⫺ 102兲 ⫽ 21.9%.12,13,14 Under these conditions, the 2 methods of measurement may not be defined as interchangeable, since the new method (ED) was significantly less precise than the old one (PAC; 21.9% vs 10%).13 However, when we evaluated these statistical parameters in each MELD group, we observed a significant reduction in the error among the low severity patients (Table 1). In this group, the precision of the new method was quite similar to that of the old one (12.8% vs 10%), and the mean bias was close to zero (0.12 L/min), which may indicate that the new technique is clinically suitable for cirrhotic patients with MELD scores ⬍15. The different behavior of more critically cirrhotic patients has been observed also with other noninvasive measurement techniques.15 It may have different explanations: First, a weak positive correlation between COPAC and relative error; therefore, more severely cirrhotic patients, who usually have higher CO values, may exhibit a lower precision of echo-Doppler CO measurements. Second, a redistribution of blood flow among body compartments may take place in more severely cirrhotic recipients due to profound alterations of their arterial tone, and vasoactive stimulation.11,16 These factors could influence CO measurements making the echo-Doppler device more precise in low-risk patients. The present study suffers from several limitations: (1) the relative absence of patients with low CO; (2) the use of a
200
standard bolus thermodilution technique for comparison, since more recent models of PAC have been proposed; and (3) the impossibility to evaluate reproducibility in our series as only one pair of values was recorded for each patient at a given surgical time. In conclusion, ED cannot replace PAC for the anesthetic management of severely cirrhotic patients undergoing OLT because of its larger error (21.9% vs 10%).14 However, in patients with biochemical MELD scores ⬍15, including many subjects with hepatocellular carcinoma, this device has the potential to yield reliable values of CO and replace PAC. In these circumstances, one must consider the advantage of a minimally invasive monitoring technique.
REFERENCES 1. Diaz S, Perez-Pena J, Sanz J, et al: Haemodynamic monitoring and liver function evaluation by pulsion cold system Z-201 (PCS) during orthotopic liver transplantation. Clin Transplant 17:47, 2003 2. Burternshaw AJ, Isaac L: The role of trans-oesophageal echocardiography for perioperative cardiovascular monitoring during orthotopic liver transplantation. Liver Transpl 12:1577, 2007 3. Berton C, Cholley B: Equipment review: new techniques for cardiac output, oesophageal Doppler, Fick principle using carbon dioxide, pulse contour analysis. Crit Care 6:216, 2002 4. Botero M, Lobato EB: Advances in noninvasive cardiac output monitoring: an update. J Cardiothorac Vasc Anesth 15:631, 2001 5. Cholley B, Payen D: Noninvasive techniques for measurements of cardiac output. Curr Opin Crit Care 11:424, 2005
PERILLI, AVOLIO, SACCO ET AL 6. Laupland KB, Bands CJ: Utility of esophageal Doppler as a minimally invasive hemodynamic monitor: a review. Can J Anesth 49:393, 2002 7. Singer M: Esophageal Doppler monitoring. In Pinsky MR, Payen D (eds): Functional Hemodynamic Monitoring. New York: Springer; 2005, p 193 8. Cariou A, Monchi M, Joly LM, et al: Noninvasive cardiac output monitoring by aortic blood flow determination: evaluation of the Sometec Dynemo 3000. Crit Care Med 26:2066, 1998 9. Colbert S, O’Hanlon DM, Duranteau J, et al: Cardiac output during liver transplantation. Can J Anesthesiol 45:133, 1988 10. Odenstedt H, Anemen A, Svensson Y, et al: Descendent aortic blood flow and cadiac output. A clinical and experimental study of continuous oesophageal echo-Doppler flowmetry. Acta Anaesthesiol Scand 45:180, 2001 11. Boucaud C, Bouffard Y, Dumortier J, et al: Transesophageal echo Doppler vs thermodilution cardiac output measurement during hepatic vascular exclusion in liver transplantation. Eur J Anaesthesiol 25:485, 2008 12. Bland JM, Altmann GD: Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1:307, 1986 13. Critchley LAH, Critchley JAJH: A meta-analysis of studies using bias and precision statistics to compare cardiac output measurement techniques. J Clin Monit Comput 15:85, 1999 14. Stetz CW, Miller RG, Kelly GE: Reliability of thermodilution method in the determination of cardiac output in clinical practice. Am Rev Respir Dis 125:1001, 1982 15. Biais M, Nouette-Gaulain K, Cottenceau V, et al: Cardiac output measurement in patients undergoing liver transplantation: pulmonary artery catheter versus uncalibrated arterial pressure waveform analysis. Anesth Analg 106:1480, 2008 16. Kamal GD, Symreng T, Starr J: Inconsistent esophageal Doppler cardiac output during acute blood loss. Anesthesiology 72:95, 1990