Liver transplantation is a widely accepted treatment for end-stage liver diseases such as cirrhosis. Often, important haemodynamic abnormalities combining increased cardiac output (CO) and systemic vasodilatation accompany these diseases.1 Additionally, the surgical procedure of liver transplantation is usually associated with profound haemodynamic variations.2 The intervention is divided into three different surgical stages: hepatectomy or preanhepatic stage, anhepatic stage and postreperfusion or neohepatic stage. Each of these periods is associated with specific haemodynamic changes related to surgical procedures. During hepatectomy and the anhepatic stage, bleeding, impairment of vena cava flow and peritoneal exudation are frequently associated with a hypovolaemic profile. The postreperfusion stage is usually described as a sepsis-like haemodynamic profile associating a decrease in systemic vascular resistances and an increase in CO.3
Given these complex haemodynamic changes during the same intervention and evidence in favour of some beneficial effects of haemodynamic goal-directed strategies in a surgical context,4,5 monitoring of CO and/or oxygen utilization seems appropriate for haemodynamic management for liver transplantation.4,5 A pulmonary artery flotation catheter could be considered as an ideal tool during liver transplantation. Indeed, it allows continuous pulmonary arterial pressures and CO measurement and concomitant monitoring of mixed venous saturation (SvO2). This last parameter enables integration of data about CO, haemoglobin (Hb) concentration and arterial oxygen saturation (SaO2) into a composite parameter.
However, right heart catheterization has been seriously challenged because of the absence of any significant impact on outcome.6,7 Moreover, this catheter can be associated with severe complications such as disruption of the pulmonary artery and catheter-related infections.8–11 Consequently, less invasive alternatives to the pulmonary artery flotation catheter are advocated. Two studies confirmed recently that less or noninvasive monitoring devices, such as the aortic Doppler [Ultrasonic cardiac output monitor (USCOM); USCOM Ltd, Coffs Harbour, New South Wales, Australia], can reliably measure CO during liver transplantation.12
Central venous saturation (ScvO2) is measured from blood drawn from a central venous catheter. Some evidence suggests that ScvO2 values could be highly correlated to SvO2 ones, with similar trends in variation and clinical significance.13
The objective of the present work was to study the agreement between SvO2 and ScvO2 during hepatectomy and after reperfusion in liver transplantation in cirrhotic patients.
Materials and methods
The study was approved by the Institutional Review Board. Written informed consent was obtained prior to surgery from 30 unpremedicated patients undergoing liver transplantation.
Anaesthesia was performed according to a standardized protocol. A rapid sequence induction was performed using propofol and succinylcholine followed by intubation. Standard monitoring included peripheral pulse oximetry, ECG, noninvasive blood pressure monitoring (before invasive arterial catheter placement) and capnography. Inspiration oxygen fraction, tidal volume and respiratory frequency were adjusted to maintain end-tidal CO2 (ETCO2) between 30 and 35 mmHg (values of ETCO2 are lower than normal because of the increased gradient between paCO2 and ETCO2 during cirrhosis), arterial saturation above 95% and peak airway pressure below 25 cmH2O. In addition, after induction of general anaesthesia, a pulmonary artery flotation catheter and an arterial pressure line were inserted. For right heart catheterization, a 7.5 Fr, three-lumen, 110 cm long pulmonary artery catheter (PAC) with a right atrial proximal port at 30 cm from the tip (Edwards Lifesciences, Irvine, California, USA) was inserted through the internal jugular vein using a percutaneous 8 Fr sheath introducer (Edwards Lifesciences). Placement of the right heart catheter was guided by wedge pressure monitoring until both pulmonary artery pressure and occlusive pressure waveforms were obtained. The postoperative chest radiography and the pattern of the pulmonary artery pressure tracings confirmed the location of the distal port in the pulmonary artery.
Anaesthesia was maintained with desflurane delivered in a constant O2/air gas mixture (inspired oxygen fraction of 50%). Analgesia and muscle relaxation were assured using continuous administration of sufentanil and atracurium, respectively. Muscle relaxation was monitored by relaxometry, whereas the sufentanil dose was adjusted to maintain the cardiac frequency and arterial pressure within 20% of the preoperative values under normovolaemic conditions (defined as CO unchanged after a crystalloid or colloid challenge). Haemodynamic stability was defined by a mean arterial pressure and a cardiac frequency within 20% of the preoperative values. The haemodynamic management through titration of anaesthetics, fluid challenge and catecholamine use (neosynephrine bolus and norepinephrine infusion) was left to the discretion of the anaesthesiologist in charge after careful evaluation of the haemodynamic parameters. Hb level was constantly kept above 9 g dl−1 using autologous or homologous (cell saver device, if not contraindicated) blood transfusion.
All liver transplantations were performed using preservation of the inferior vena cava and piggy-back technique, with a temporary porto-caval anastomosis and without veno-venous bypass. In all patients at the time of reperfusion of the graft, the portal vein was first unclamped before the hepatic artery.
The anaesthesiologist in charge collected three simultaneous blood samples for gas analysis (simultaneous measurement using the cooximetry technique) from the pulmonary artery, the right atrium and the arterial line in each patient. Two of these measurement sequences were performed in the hepatectomy stage (as determined by the clinical conditions), and two other sequences were performed 30 and 40 min after reperfusion after the hepatic artery unclamping.
The data gathered from these samples and invasive haemodynamic monitoring with the PAC allowed determination and calculation of the cardiac index (CI, expressed in l min−1 m−2), SaO2 (determined by blood gas analysis), Hb (determined by blood gas analysis), SvO2, ScvO2, arterial and venous oxygen content [CaO2 and CvO2, calculation based on the formula: Ca(v)O2 = 1.34 × Sa(v)O2 × Hb expressed in mlO2], arterial oxygen delivery (DO2, calculated based on the formula: DO2 = CaO2 × CI × 10 expressed in ml min−1 m−2), venous to arterial oxygen saturation difference [(a − v)dO2, calculated based on the formula: (a − v)dO2 = CaO2 − CvO2 expressed in mlO2), oxygen consumption (VO2, calculated based on the formula: VO2 = VADO2 × CI × 10 expressed in ml min−1 m−2) and oxygen extraction [EO2, calculated based on the formula EO2 = (VADO2 × 100/CaO2) expressed as a percentage].
Agreement between the SvO2 and ScvO2 during the hepatectomy stage, the reperfusion stage and for overall measurements was tested using the Bland–Altman analysis.14 Agreement between the two parameters was considered acceptable when limits of agreement, defined as the mean of ScvO2–SvO2, were below ±5% in accordance with a previous recommendation.15
Comparison of data before and after reperfusion of the graft was carried out using a Student's t-test. For each stage of the intervention, Pearson correlation coefficients were determined between the bias of the measured venous saturation (ΔScv2SvO2, the differences between ScvO2 and SvO2) and the haemodynamic and oxygenation parameters. In order to minimize between-patient variability and assuming a normal distribution of bias (ScvO2 − SvO2) and the other haemodynamic and oxygenation parameters, 30 patients were included in this study.
Statistical packages used for the analysis were Graph Pad (version 5.00 for Windows; Graph Pad Software, San Diego, California, USA) and SPSS (version 15 for Windows; SPSS Inc., Chicago, Illinois, USA).
Demographic and intraoperative data are summarized in Table 1. Norepinephrine was continuously infused in 18 patients during the hepatectomy stage (0.5–1 mg h−1). After reperfusion, all the patients required an in increase in or the introduction of continuous norepinephrine infusion (range from 1 to 2.5 mg h−1).
A total of 60 pairs (central and mixed venous samples) of measurements were performed, 30 during the hepatectomy stage and 30 after graft reperfusion. SvO2 [mean ± SD (minimum–maximum)] was 84.5 ± 7.7% (63.9–98.6%) for the global measurements, 86.13 ± 7.8% (63.9–98.6%) during hepatectomy and 82.9 ± 7.3% (67–97.9%) after reperfusion. Oxygenation and haemodynamic parameters, including CI, VO2, Hb level, SaO2, DO2 and EO2, are summarized in Table 2. SvO2 decreased over time from hepatectomy to reperfusion, whereas CI, VO2, DO2 and EO2 showed significantly higher values after reperfusion. Hb level, ScvO2 and SaO2 did not differ between the two periods. Throughout the two periods, SvO2 was lower than ScvO2; however, the difference was not significant.
Bland–Altman analysis found a bias (limits of agreements) of −1.2% (−9.1 to 6.6%), −0.3% (−4.8 to 4%) and −2.1% (−12 to 7.8%) for the overall measurements and the hepatectomy (Fig. 1a) and after reperfusion (Fig. 1b) measurements, respectively. According to our limits of agreement fixed for the equivalence between ScvO2 and SvO2 (less than 5%), the two measurements were equivalent only during the hepatectomy stage.
To evaluate the influence of CI, VO2, DO2 and EO2 on the differences between SvO2 and ScvO2, statistical correlation analysis between ΔScvO2SvO2 and these variables was performed (Table 3). During hepatectomy, ΔScvO2SvO2 was negatively correlated with VO2 and DO2. After reperfusion, the ΔScvO2SvO2 was negatively correlated with CI but positively correlated with VO2, DO2 and EO2.
The present work demonstrates a good level of agreement between SvO2 and ScvO2 during the first stage of liver transplantation (hepatectomy) but not after graft reperfusion in patients with cirrhosis.
Although we observed a correlation between SvO2 and ScvO2 for both stages of the intervention, the limits of agreement defined a priori, after Bland–Altman analysis, were met only during hepatectomy. This finding is in accordance with previous studies that focused on ScvO2. In both animal models and clinical studies,16,17 without sepsis or systemic inflammatory response, SvO2 and ScvO2 were rather concordant. However, this observation seems no longer valid in patients with a hyperdynamic, haemodynamic or sepsis-like profile.15,18
The main hypothesis explaining this lack of concordance between SvO2 and ScvO2 in this situation could be the quality of mixing of hepatomesenteric blood drained by the inferior vena cava and blood drained by the superior vena cava. Under physiological conditions, SvO2 exceeds ScvO2 by an average of 2 or 3% because of the lower oxygen extraction from the hepato-mesenteric territory compared with cephalic territories such as the brain.19 This fact has been clearly demonstrated by Gutierrez et al.,20 who found a decreasing gradient in venous saturation and lactate between the superior vena cava and the pulmonary artery in stable nonseptic patients. After reperfusion, splanchnic VO2 increases due to increased oxygen consumption by the liver graft.21 Reduction in SvO2 without significant changes of ScvO2 observed in this study probably results from a predominant effect of graft reperfusion on the splanchnic system without any effect on ScvO2 drawn in the superior vena cava. In order to test whether this hypothesis explains the lack of equivalence between the two measures after graft reperfusion, we performed correlations between the ΔScvO2SvO2 and the simultaneously assessed haemodynamic and oxygenation parameters. These analyses showed a significant negative correlation between ΔScvO2SvO2 and both CI and DO2 during the hepatectomy stage. However, during the neoliver stage, analysis showed negative correlations between ΔScvO2SvO2 and CI and a positive significant correlation between ΔScvO2SvO2 and VO2 and EO2 (Table 3). This emphasized the effect of VO2 and CO on the difference between ScvO2 and SvO2.
Another factor that might impair the equivalence between SvO2 and ScvO2 levels is oxygen saturation of blood coming from the coronary sinus. Under physiological conditions, venous blood coming from the coronary sinus is highly desaturated.22 However, during septic states, the increase in myocardial oxygen demand is compensated by the increase in coronary blood flow.23 Given the similarity in haemodynamic profiles observed during septic shock and after graft reperfusion, we can hypothesize that this phenomenon might negatively influence the level of agreement between ScvO2 and SvO2 during the reperfusion stage. As ScvO2 samples were taken from the right atrium in which venous blood coming from the coronary sinus is supposed to be mixed with central venous blood, we assume that this mechanism was unlikely to have affected the present study. However, to our knowledge, no data are available on the effect of neoliver reperfusion on cardiac circulation and metabolism to refute this hypothesis.
The present study has some limitations. Our study found a low VO2 of 47.3 ± 31.6 ml min−1 m−2 before graft reperfusion in comparison with other studies that demonstrated VO2 up to 90 ml min−1 m−2 during the same stage.24,25 This discrepancy with regard to the low VO2 may be explained by the nonstandardization of sample timing acquirement during the hepatectomy stage in our study. Samples were taken at the discretion of the anaesthesiologist. This may have favoured samples being taken during haemodynamically unstable periods such as bleeding, liver manipulations or positioning. The subsequent reduction in tissue flow may have affected the VO2 and contribute to its low values in comparison with the previously published results. Another explanation might result from our practice of using norepinephrine in more than 50% of the patients before reperfusion and in all patients after reperfusion. However, data from a previous study26 have shown no significant effect of norepinephrine use on oxygenation variables in comparison with dopamine. Future studies may better elucidate the role of timing of measurements and norepinephrine use in the comparison between SvO2 and ScvO2.
ScvO2 samplings were taken from the atrial port of the pulmonary artery flotation catheter. It seems that the level of agreement between ScvO2 and SvO2 improved closer to the pulmonary artery.27,28 This could explain a higher concordance between SvO2 and ScvO2 during hepatectomy than if the samples were obtained in the superior vena cava. Consequently, the use of ScvO2 measured from the superior vena cava might decrease the agreement between ScvO2 and SvO2. Nevertheless, our data do not allow any definite conclusions to be drawn on a nonconcordance between ScvO2 and SvO2 in liver transplantation.
Finally, another concern is the timing of measurements. Hepatectomy and postreperfusion were selected because of the specific haemodynamic modifications associated with these particular stages of liver transplantation. The hepatectomy in liver transplantation is usually the most haemorrhagic phase, and SvO2 values would be expected to decrease. On the contrary, after reperfusion, the haemodynamic profile is probably better characterized by a hyperdynamic profile with a high CO state and a profound reduced systemic arterial resistance. By design, SvO2 and ScvO2 were both determined 30 min after hepatic artery reperfusion to allow haemodynamic stabilization of the patient after the immediate reperfusion of the graft that can be associated with significant hypotension in almost 25% of cirrhotic patients during liver transplantation. However, our study does not include either the anhepatic or the postoperative stages of the liver transplantation, making the extrapolation of the present results to the entire perioperative period difficult.
SvO2 is a robust parameter that integrates information about CO, Hb concentration, arterial saturation and oxygen consumption. It allows early detection of most, if not all, critical situations in the perioperative period, including haemorrhage, hypovolaemia or arterial desaturation. However, this parameter requires the insertion of a right heart catheter. The use of ScvO2 instead of SvO2 could have been an elegant solution to simplify haemodynamic monitoring when central lines are needed or already present and to avoid right heart catheterization.
Goal-directed therapy using stroke volume and CO optimization is now commonly used during open abdominal surgery, with a proven postoperative benefit of preventing complications.5 Recent studies also recommend ScvO2 monitoring as a tool for early goal-directed therapy. Bracht et al.29 and Pearse et al.30,31 have demonstrated an independent association between the occurrence of postoperative complications and low perioperative ScvO2 less than 64%. In addition, Rivers et al.32 in 2001 investigated the optimization of patients in severe sepsis using an ScvO2 target of 70%, central venous pressure, urine output and mean arterial pressure and demonstrated a reduction in in-hospital mortality and the duration of hospitalization. However, despite evidence in favour of the use of perioperative ScvO2 monitoring, target values for this parameter would have to be evaluated during liver transplantation and especially in cirrhosis.
In patients with cirrhosis, ScvO2 measured in the right atrium and SvO2 show an acceptable level of agreement during the hepatectomy stage of liver transplantation but not after liver graft reperfusion. This effect probably resulted from the neoliver-related increase in VO2 that was not detected by ScvO2. Accordingly, at present, ScvO2 cannot be recommended as a substitute for SvO2 monitoring in the context of liver transplantation.
1 Moller S, Henriksen JH. Cardiovascular complications of cirrhosis. Gut 2008; 57:268–278.
2 Paugam-Burtz C, Kavafyan J, Merckx P, et al
. Postreperfusion syndrome during liver transplantation
for cirrhosis: outcome and predictors. Liver Transpl 2009; 15:522–529.
3 Mandell MS, Katz JJ, Wachs M, et al
. Circulatory pathophysiology and options in hemodynamic management during adult liver transplantation
. Liver Transpl Surg 1997; 3:379–387.
4 Donati A, Loggi S, Preiser JC, et al
. Goal-directed intraoperative therapy reduces morbidity and length of hospital stay in high-risk surgical patients. Chest 2007; 132:1817–1824.
5 Abbas SM, Hill AG. Systematic review of the literature for the use of oesophageal Doppler monitor for fluid replacement in major abdominal surgery. Anaesthesia 2008; 63:44–51.
6 Sandham JD, Hull RD, Brant RF, et al
. A randomized, controlled trial of the use of pulmonary-artery catheters in high-risk surgical patients. N Engl J Med 2003; 348:5–14.
7 Shah MR, Hasselblad V, Stevenson LW, et al
. Impact of the pulmonary artery catheter in critically ill patients: meta-analysis of randomized clinical trials. JAMA 2005; 294:1664–1670.
8 Sekkal S, Cornu E, Christides C, et al
. Swan-Ganz catheter induced pulmonary artery perforation during cardiac surgery concerning two cases. J Cardiovasc Surg (Torino) 1996; 37:313–317.
9 Stone JG, Faltas AN, Khambatta HJ, et al
. Temporary unilateral pulmonary artery occlusion: a method for controlling Swan-Ganz catheter-induced hemoptysis. Ann Thorac Surg 1984; 37:508–510.
10 Thomas R, Siproudhis L, Laurent JF, et al
. Massive hemoptysis from iatrogenic balloon catheter rupture of pulmonary artery: successful early management by balloon tamponade. Crit Care Med 1987; 15:272–273.
11 Rubin SA, Puckett RP. Pulmonary artery: bronchial fistula – a new complication of Swan-Ganz catheterization. Chest 1979; 75:515–516.
12 Wong LS, Yong BH, Young KK, et al
. Comparison of the USCOM ultrasound cardiac output monitor with pulmonary artery catheter thermodilution in patients undergoing liver transplantation
. Liver Transpl 2008; 14:1038–1043.
13 Reinhart K, Kuhn HJ, Hartog C, Bredle DL. Continuous central venous and pulmonary artery oxygen saturation monitoring in the critically ill. Intensive Care Med 2004; 30:1572–1578.
14 Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1:307–310.
15 Chawla LS, Zia H, Gutierrez G, et al
. Lack of equivalence between central and mixed venous oxygen saturation. Chest 2004; 126:1891–1896.
16 Turnaoglu S, Tugrul M, Camci E, et al
. Clinical applicability of the substitution of mixed venous oxygen saturation with central venous oxygen saturation. J Cardiothorac Vasc Anesth 2001; 15:574–579.
17 Sander M, Spies CD, Foer A, et al
. Agreement of central venous saturation
and mixed venous saturation
in cardiac surgery patients. Intensive Care Med 2007; 33:1719–1725.
18 Varpula M, Karlsson S, Ruokonen E, Pettila V. Mixed venous oxygen saturation cannot be estimated by central venous oxygen saturation in septic shock. Intensive Care Med 2006; 32:1336–1343.
19 Bloos F, Reinhart K. Venous oximetry. Intensive Care Med 2005; 31:911–913.
20 Gutierrez G, Wulf-Gutierrez ME, Reines HD. Monitoring oxygen transport and tissue oxygenation. Curr Opin Anaesthesiol 2004; 17:107–117.
21 Tietge UJ, Bahr MJ, Manns MP, Boker KH. Decreased splanchnic oxygen uptake and increased systemic oxygen uptake in cirrhosis are normalized after liver transplantation
. Liver Transpl 2001; 7:1015–1022.
22 Schremmer B, Dhainaut JF. Regulation of myocardial oxygen delivery. Intensive Care Med 1990; 16(Suppl 2):S157–S163.
23 Dhainaut JF, Huyghebaert MF, Monsallier JF, et al
. Coronary hemodynamics and myocardial metabolism of lactate, free fatty acids, glucose, and ketones in patients with septic shock. Circulation 1987; 75:533–541.
24 Parker BM, Dishart MK, Pinsky MR, Kang Y. Is there occult tissue ischemia in chronic end-stage liver disease? Liver Transpl Surg 1999; 5:211–218.
25 Kostopanagiotou G, Smyrniotis V, Theodoraki K, et al
. Oxygen availability during orthotopic liver transplantation
. Liver Transpl 2003; 9:1216–1221.
26 Zhang LP, Li M, Yang L. Effects of different vasopressors on hemodynamics in patients undergoing orthotopic liver transplantation
. Chin Med J (Engl) 2005; 118:1952–1958.
27 Edwards JD, Mayall RM. Importance of the sampling site for measurement of mixed venous oxygen saturation in shock. Crit Care Med 1998; 26:1356–1360.
28 Gutierrez G, Venbrux A, Ignacio E, et al
. The concentration of oxygen, lactate and glucose in the central veins, right heart, and pulmonary artery: a study in patients with pulmonary hypertension. Crit Care 2007; 11:R44.
29 Bracht H, Hanggi M, Jeker B, et al
. Incidence of low central venous oxygen saturation during unplanned admissions in a multidisciplinary intensive care unit: an observational study. Crit Care 2007; 11:R2.
30 Pearse R, Dawson D, Fawcett J, et al
. Changes in central venous saturation
after major surgery, and association with outcome. Crit Care 2005; 9:R694–R699.
31 Pearse R, Dawson D, Fawcett J, et al
. Early goal-directed therapy after major surgery reduces complications and duration of hospital stay. A randomised, controlled trial [ISRCTN38797445]. Crit Care 2005; 9:R687–R693.
32 Rivers E, Nguyen B, Havstad S, et al
. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001; 345:1368–1377.