Secondary Logo

Journal Logo

Lactate is Correlated with the Indocyanine Green Elimination Rate in Liver Resection for Cirrhotic Patients

Orii, Ryo MD*,; Sugawara, Yasuhiko MD†,; Hayashida, Masakazu MD*,; Uchida, Kanji MD*,; Yamada, Yoshitsugu MD*,; Takayama, Tadatoshi MD†,; Makuuchi, Masatoshi MD†,; Hanaoka, Kazuo MD*

doi: 10.1097/00000539-200104000-00049
General Articles: Research Report

The role of lactate in liver ischemia-reperfusion injury in cirrhosis has not been clarified. Fifty patients with hepatocellular carcinoma who underwent partial liver resection under Pringle’s maneuver were included in this study. We performed the indocyanine green clearance test before the operation and three times during the surgery to calculate its elimination rate. Blood lactate and base excess were measured at the corresponding times. Systolic and diastolic systemic arterial pressure, heart rate, cardiac index, and esophageal temperature were monitored. Aminotransferase levels were recorded the day before the operation, 1 h after the operation, and on the first and third postoperative days. We calculated the increase or decrease in lactate levels during the preischemic, ischemic, and postischemic phases, and examined the correlation between these results and the changes in indocyanine green elimination rate and some clinical factors. The lactate levels increased before reperfusion and began to decrease after reperfusion. The lactate increase and decrease during the ischemic and postischemic phases correlated with the change in indocyanine green elimination rate (P < 0.0001 and P = 0.02 for the respective phases). The lactate increase during the preischemic phase correlated with the duration of the preischemic phase (P < 0.0001). In cirrhotic patients who undergo liver resection with Pringle’s maneuver and who do not show postoperative liver failure, the blood lactate profile might be a reliable indicator of liver metabolic capacity during surgery.

*Department of Anesthesiology and †Hepatobiliary Pancreatic Surgery Division, Department of Surgery, Graduate School of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan

Supported, in part, by a Grant-in-aid for Scientific Research from the Ministry of Education, Science and Culture of Japan.

November 15, 2000.

Address correspondence and reprint requests to Yasuhiko Sugawara, MD, Hepatobiliary Pancreatic Surgery Division, Department of Surgery, Graduate School of Medicine, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan.

IMPLICATIONS: In cirrhotic patients who underwent liver resection with Pringle’s maneuver, the lactate increase and decrease during the ischemic and postischemic phases correlated with the change in the indocyanine green elimination rate. The blood lactate profile might be a reliable indicator of liver metabolic capacity during surgery.

Lactate is produced at a rate of 1,300 mmol per day in a resting adult under steady-state conditions. Forty to 60% of the lactate is removed by the liver by glucogenesis, and to a lesser extent by oxidation to CO2 and H2O. Other sites of lactate uptake and metabolism include the muscles, kidneys, and heart (1). Hypoperfusion, hypoxia, and severe ischemic damage of the liver convert the liver from a lactate-consuming to a lactate-producing system (2). The conditions that induce this kind of hepatic damage are encountered during the preanhepatic and anhepatic phases of liver transplantation and during the ischemic phase of liver resection.

Liver transplantation is a unique model for studying the relationship between lactate metabolism and liver metabolic capacities. This has led some workers to propose that early postoperative changes in the blood lactate concentration could be used as an early indicator of graft viability and of the quality of the functional recovery of the graft in orthotopic liver transplantation (3–5). de Gasperi et al. (6) concluded that blood lactate profiles measured in the neohepatic phase indicated the speed of recovery of liver metabolic capacities immediately after orthotopic liver transplantation. A recent study (7) revealed that, in living-related liver transplantation, the preanhepatic, anhepatic and neohepatic phases each had a characteristic blood lactate profile. The lactate increase observed before reperfusion reflected depressed hepatic function during anesthesia and surgery. The graft size strongly affected lactate levels during the early neohepatic phase.

However, these suggestions rest basically on comparison of the blood lactate profile with indirect clinical indicators for liver function such as primary nonfunction of the graft and postoperative mortality and morbidity after orthotopic liver transplantation. No study has demonstrated correlation between the blood lactate profile and more direct indicator of liver function. In addition, usefulness of the blood lactate profile has not been demonstrated in liver surgery other than liver transplantation. We noticed that in patients who undergo partial resection of the liver, lactate tends to increase during the preischemic phase, increases markedly during the ischemic phase, and decreases during the postischemic phase. We hypothesized that lactate could reflect liver metabolic capacities during liver resection. To demonstrate the hypothesis, we examined the blood lactate concentration and indocyanine green elimination rate (ICG-K) (8–11), which is a reliable indicator of metabolic liver capacities as well as hepatic blood flow (12,13) in patients who underwent liver resection.

Back to Top | Article Outline

Patients and Methods

The study protocol was approved by the ethics committee of our local authority and informed consent was obtained from each patient.

Back to Top | Article Outline


From December 1997 to October 1999, 172 patients underwent liver resection at the Hepatobiliary Pancreatic Division, Department of Surgery, Tokyo University Hospital. Fifty of these patients (37 males and 13 females aged 35 to 80 yr) were enrolled in this study. Each patient was classified in ASA physical status II. To lessen the heterogeneity of the patient population and to exclude patients with severe cirrhosis, the inclusion criteria were defined as follows:

  • 1. Subjects were cirrhotic patients with hepatocellular carcinoma.
  • 2. The tumors were limited to one subsegment of the liver.
  • 3. The preoperative ICG-K was over 0.06 min−1.
  • 4. Either subsegmentectomy (14) or limited resection of the tumor was scheduled.
  • 5. Pringle’s maneuver alone was performed to induce liver ischemia.

In our institute, Pringle’s maneuver is usually applied during liver resection to lessen blood loss. Briefly, a 15-min period of vascular clamping at the liver hilus was repeated at intervals of 5 min until liver resection was completed (15). Limited resection indicated nonanatomical resection of the tumor(s) with a sufficient surrounding nontumorous liver tissue margin. Subsegmentectomy indicated complete anatomical resection of the corresponding subsegment fed by the portal branches. Our method of identifying the subsegment in concern has been described in detail elsewhere (16). Patients were excluded if intraoperative ultrasound (16) revealed that the tumor extended beyond one segment or a procedure other than Pringle’s maneuver was applied to induce ischemia.

The preoperative laboratory data were as follows: total bilirubin 0.5–1.3 mg/dL (mean ± se, 0.9 ± 0.1), albumin 3.0–4.2 mg/dL (3.6 ± 0.1), prothrombin time 10.8–13.0 s (11.8 ± 0.1), aspartate aminotransferase (AST) 17–77 IU/L (50.6 ± 3.1), and alanine aminotransferase (ALT) 18–115 IU/L (48.8 ± 3.7).

Anesthesia was induced with midazolam (0.05–0.1 mg/kg), thiopental sodium (2–4 mg/kg), and fentanyl (2 μg/kg), and maintained with 0.8%-1.5% end-tidal isoflurane, 50% nitrous oxide in oxygen, and fentanyl (1–2 μg/kg/h). Muscle relaxation was maintained with pancuronium (0.08 mg · kg−1 · h−1). Mechanical ventilation was set at a respiratory frequency of 10–15/min and an inspiratory/expiratory duration ratio of 1:2. The respiratory tidal volume was reduced to approximately 60% just before starting liver resection to reduce the thoracic and right atrial pressure and, consequently, back-bleeding from the hepatic veins and their tributaries (17). The concentration of inspired oxygen was set to give a Pao2 of more than 120 mm Hg. The Paco2 was maintained at 35–45 mm Hg. The following variables were monitored rou- tinely: invasive arterial blood pressure, heart rate, and electrocardiogram (5 leads and 2 channels: II and V5), pulse oximetry, end-tidal carbon dioxide tension, and esophageal temperature.

A crystalloid solution (sodium, 90 mmol/L; chloride 70 mmol/L; lactate 20 mmol/L) was given at a rate of 2–3 mL/kg/h. Fresh frozen plasma was transfused when necessary to maintain stable hemodynamics and to supply clotting factors. The blood glucose level was measured once per hour, and when it was over 16.6 mmol/L, regular insulin was administered IV.

The following variables were examined and recorded by one of the authors (M.H.) and the anesthesiologist in charge of anesthesia. The ICG (Diagnogreen®, Daiichi Pharmaceutical Co., Tokyo, Japan) test was performed before surgery as a baseline (T0). During surgery, it was performed three times: 1) just before the induction of liver ischemia by Pringle’s maneuver (T1); 2) just after reperfusion of the liver, i.e., the completion of liver resection (T2); and 3) 60 min after reperfusion (T3). The optical sensor of an ICG clearance meter (DDG-2001™, Nihon Kohden Industry Co. Ltd., Tokyo, Japan) was applied to the right second finger. ICG-K is originally calculated as the slope of the regression line in the plasma ICG concentration (μg/mL) versus time (min) plot multiplied by −1. ICG clearance meter was determined by measuring light transmission through the finger at two wavelengths, 810 nm and 940 nm (8,18). During surgery, the cardiac index (CI) was also measured using the dye dilution method by calculating the area under the very initial phase of the ICG concentration curve with a computer.

A gastric tonometer (Tonocap™ regional capnometry monitor, Tonometrics, Helsinki, Finland) was used in all patients. The gastric mucosal Pco2 was measured, and the Pco2 gap (tonometer Pco2 − arterial Pco2) was calculated.

Systolic systemic arterial pressure (SAP), diastolic systemic arterial pressure (DAP), heart rate (HR), and esophageal temperature (ET) were also recorded at T0, T1, T2, and T3. The lactate concentration and base excess of arterial blood were measured with an analyzer (ABL 625; Radiometer, Copenhagen, Denmark) at the same times.

The observation period was divided into three phases; the preischemic phase (from T0 to T1), the ischemic phase (from T1 to T2), and the postischemic phase (from T2 to T3). The lactate increase in the preischemic phase (ΔLacpre) was calculated by the difference between blood lactate levels at T1 and T0:MATH

The changes in lactate levels in the ischemic phase (ΔLacisc) and postischemic phase (ΔLacpost) were calculated as follows:MATHMATH

The changes in the ICG-K level (ΔICG-K) and Pco2 (ΔPco2) were calculated in a similar manner.

Blood AST and ALT levels were measured before surgery (D-1), one hour after surgery (D0) and on the first and third postoperative days (D1 and D3, respectively).

Back to Top | Article Outline

Statistical Analysis

The data are presented as range (mean) or mean ± sem. The variables measured during and after anesthesia were compared with the corresponding preoperative levels using an analysis of variance with Bonferroni’s correction. The differences at P < 0.02 (CI) or P < 0.008 (except for CI) were considered to be significant in analysis of variance.

The relationship between ΔICG-K and ΔLac in each phase was examined by a simple regression analysis. The relationships between demographic or surgical characteristics and ΔLac were examined by a multiple regression analysis in each phase and were considered to be significant at P < 0.05.

Back to Top | Article Outline


Surgical Results

Overall, the duration of surgery ranged from 125–673 (384) min. The durations of the preischemic and ischemic phases were 40–420 (166) min and 28–115 (71) min, respectively. Blood loss was 115–1081 (512) mL or 2 -22 (10) mL/kg. The weight of the resected liver was 36–450 (114) g. The total volume of fresh frozen plasma was 0–1200 (610) mL or 0–26 (14) mL/kg. None of the patients received packed red cells or fresh whole blood transfusion. Complications after liver resection included bile leakage from a dissection plane of the liver in four patients and pleural effusion in five. No postoperative deaths were observed during the first 30 days after the operation.

Back to Top | Article Outline

Changes in Hemodynamics

SAP and DAP did not change significantly during the observation period. (Fig. 1). HR did not change significantly from T0 through T2, but did change significantly at T3 compared to the T0 levels (P = 0.002). ET and CI remained stable throughout the observation period.

Figure 1

Figure 1

Back to Top | Article Outline

Changes in ICG-K

ICG-K did not change significantly at T1 compared to the baseline level at T0 (Fig. 2). However, just after reperfusion, ICG-K decreased significantly at T2 versus baseline in the control group (P = 0.006). The ICG-K level at T3 did not differ from the baseline level.

Figure 2

Figure 2

Back to Top | Article Outline

Changes in Lactate and Base Excess

The lactate level increased during the preischemic and ischemic phases and tended to decrease during the postischemic phase (Fig. 2). The lactate levels at T1, T2, and T3 were significantly increased from those at baseline (P = 0.004, P < 0.0001, and P < 0.0001 in each comparison).

Base excess continued to decline until T2 (Fig. 2). Base excess at T2 and T3 was significantly decreased from that at baseline (P < 0.0001 and P = 0.002, respectively).

Back to Top | Article Outline

Changes in the Pco2 Gap

The Pco2 gap increased during the preischemic and ischemic phases and tended to decrease during the postischemic phase (Fig. 2). The lactate level at T2 was significantly higher than that at baseline (P = 0.0009).

Back to Top | Article Outline

Changes in AST and ALT

AST levels were maximal (219 ± 20 IU/L) on the first postoperative day (Fig. 2). The changes in the average ALT level were similar to those of AST.

Back to Top | Article Outline

Relationships between Lactate and Clinical Factors

The relationship between ΔICG-K and ΔLac in each phase is summarized in Table 1 and Figure 3. ΔICG-Kisc and ΔICG-Kpost correlated well with ΔLacisc and ΔLacpost (P < 0.0001 and P = 0.02 for the respective phases). However, there was no significant correlation between ΔICG-Kpre and ΔLacpre.

Table 1

Table 1

Figure 3

Figure 3

Simple regression analysis gave the following equations:MATHMATH

None of the clinical factors correlated with ΔLacisc or ΔLacpost. However, the duration of the preischemic phase correlated well with ΔLacpre (P < 0.0001, Table 2).

Table 2

Table 2

Back to Top | Article Outline


In our study, blood lactate tended to change inversely to ICG-K throughout surgery. During the preischemic, ischemic, and postischemic phases, lactate increased, markedly increased, and tended to decrease, respectively. On the contrary, ICG-K tended to decrease, decreased significantly, and increased significantly in respective phase. The change in blood lactate levels during the ischemic and postischemic phases correlated well with the concomitant ICG-K change. These findings strongly suggested that blood lactate levels could reflect liver dysfunction after the induction of ischemia in liver resection for cirrhotic patients.

During the preischemic phase, however, the change in blood lactate levels did not correlate significantly with the concomitant ICG-K change. Lactate increased significantly whereas ICG-K tended to decrease but insignificantly. During this phase, hepatic blood flow should have been reduced by some hemodynamic instability. For example, it can be reduced as a result of dissection of the liver vasculature or compression of the liver parenchyma by surgical retractors (6,11). Anesthetics also might affect hepatic function unfavorably. Although isoflurane does not depress hepatic function as measured by the ICG test in patients with normal hepatic function (19), a dose-dependent decrease in hepatic blood flow induced by isoflurane might depress liver function significantly in cirrhotic patients and thus enhance lactate accumulation (20–22). Presumably because of this reduction in hepatic blood flow, lactate uptake by the liver was impaired, which in turn caused time-dependent increase in the lactate level even before the liver was subjected to ischemia. However, it seems difficult to explain an insignificant correlation between the change in blood lactate levels and the concomitant ICG-K change during this phase.

During the ischemic phase, lactate increased markedly and conversely ICG-K decreased significantly. Both of these changes might indicate a significant reduction in liver function in this phase. In this study, hepatic venous flow was not measured. This might make it unclear whether decreased lactate uptake because of the abolishment of hepatic blood flow during vascular occlusion or increased lactate production resulting from tissue ischemia primarily contributed to the lactate increase during this phase (8,9). The lactate increase was closely correlated with the ICG-K change, which indicated reduced hepatic blood flow. In contrast, there was no significant correlation between the lactate increase and the Pco2 gap, which is a quite sensitive indicator of tissue oxygenation in splanchnic organs (23,24). These results suggest that the lactate increase was mainly caused by reduced lactate uptake by the deteriorated liver. The role of tissue ischemia seemed to be limited.

The lactate level immediately after ischemia tended to decrease and this change correlated well with ΔICG-Kpost. These changes indicated that hepatic blood flow as well as the capacity of the liver to uptake and metabolize lactate were recovering after ischemia. These results also indicated that several ten-minute periods of warm ischemia induced by Pringle’s maneuver did not cause irreversible damage to the liver with mildly cirrhotic change.

The effect of Pringle’s maneuver on the hemodynamic state remains to be clarified. In our study, systemic arterial pressure and the cardiac index remained stable throughout the observation period. In contrast, although heart rate did not change significantly until the completion of liver resection, it did change significantly after reperfusion. Delva et al. (25) showed that Pringle’s maneuver caused the following significant hemodynamic changes: mean arterial pressure increases by 40% and the cardiac index decreases by 10%.

Base excess continued to decline until the completion of liver resection, and then increased after reperfusion in our study. The pattern of the changes in base excess was in contrast to that of lactate. Fresh-frozen plasma is converted to bicarbonate, which will neutralize the H+ that accompanies lactate acidosis. Therefore, the level of base excess might be affected by lactate acidosis or the infusion of fresh-frozen plasma (26). However, the exact relation may be difficult to explain.

In conclusion, in liver resection in cirrhotic patients, lactate continued to increase before reperfusion and began to decrease after reperfusion. The change in lactate levels correlated well with that of ICG-K after the induction of ischemia.

Back to Top | Article Outline


1. Cohen RD. Roles of liver and kidney in acid-base regulations and its disorders. Br J Anaesth 1991; 67: 154–64.
2. Moreau R, Lee SS, Soupison T, Roche-Sicot C. Abnormal tissue oxygenation in patients with cirrhotic and liver failure. J Hepatol 1988; 7: 98–105.
3. Fath JJ, Ascher NL, Konstantinided FN, et al. Metabolism during hepatic transplantation: indicators of allograft function. Surgery 1984; 96: 664–73.
4. Begliomini B, DeWolf A, Freeman J, Kang YG. Intraoperative lactate levels can predict graft function after liver transplantation [abstract]. Anesthesiology 1989; 71: A180.
5. Svensson KL, Persson H, Henrikson A, et al. Whole body gas exchange, amino acid and lactate clearance as indicators of initial and early allograft viability in liver transplantation. Surgery 1989; 105: 472–80.
6. de Gasperi A, Mazza E, Corti A, et al. Lactate blood levels in the perioperative period of orthotropic liver transplantation. Int J Clin Lab Res 1997; 27: 123–8.
7. Orii R, Sugawara Y, Hayashida M, et al. Peri-operative blood lactate levels in recipients of living-related liver transplantation. Transplantation 2000; 69: 2124–7.
8. Nakayama M, Kanaya N, Fujita S, Namiki A. Effects of ephedrine on indocyanine green clearance during spinal anesthesia: evaluation by the finger piece method. Anesth Analg 1993; 77: 947–9.
9. Koller J, Wieser C, Furtwangler W, et al. Orthotopic liver transplantation and perioperative lactate metabolism. Transplant Proc 1991; 23: 1989–90.
10. Kawasaki S, Sugiyama Y, Iga T, et al. Pharmacokinetic study on the hepatic uptake of indocyanine green in cirrhotic patients. Am J Gastroenterol 1985; 80: 801–6.
11. Villeneuve JP, Huot R, Marleau D, Huet PM. The estimation of hepatic blood flow with indocyanine green: comparison between the continuous infusion and single injection methods. Am J Gastroenterol 1982; 77: 233–7.
12. Andersen MN, Kuchiba K. Measurement of acute changes in liver function and blood flow. Arch Surg 1970; 100: 541–5.
13. Murray JM, Trinick TR. Hepatic function and indocyanine green clearance during and after prolonged anaesthesia with propofol. Br J Anaesth 1992; 69: 643–4.
14. Miyagawa S, Makuuchi M, Kawasaki S, Kakazu T. Criteria for safe hepatic resection. Am J Surg 1995; 169: 589–94.
15. Pringle JH. Notes on the arrest of hepatic hemorrhage due to trauma. Ann Surg 1908; 48: 541–9.
16. Makuuchi M, Hasegawa H, Yamazaki S. Ultrasonically guided subsegmentectomy. Surg Gynecol Obstet 1985; 161: 346–50.
17. Jones RM, Moulton CE, Hardy KJ. Central venous pressure and its effect on blood loss during liver resection. Br J Surg 1998; 85: 1058–60.
18. Nielsen NC. Spectrophotometric determination of ICG in plasma especially with a view to an improved correction for blank density. Scand J Clin Lab Invest 1963; 15: 613–21.
19. Kanaya N, Nakayama M, Fujita S, Namiki A. Comparison of the effects of sevoflurane, isoflurane and halothane on indocyanine green clearance. Br J Anaesth 1995; 74: 164–7.
20. Frink EJ Jr, Morgan SE, Coetzee A, et al. The effects of sevoflurane, halothane, enflurane, and isoflurane on hepatic blood flow and oxygenation in chronically instrumented greyhound dogs. Anesthesiology 1992; 76: 85–90.
21. Owen OE, Reichle FA, Mozzoli MA, et al. Hepatic, gut, and renal substrate flux rates in patients with hepatic cirrhosis. J Clin Invest 1981; 68: 240–52.
22. Woll PJ, Record CO. Lactate elimination in man: effects of lactate concentration and hepatic dysfunction. Eur J Clin Invest 1979; 9: 397–404.
23. Levy B, Bollaert PE, Lucchelli JP, et al. Dobutamine improves the adequacy of gastric mucosal perfusion in epinephrine-treated septic shock. Crit Care Med 1997; 25: 1649–54.
24. Gomersall CD, Joynt GM, Ho KM, et al. Gastric tonometry and prediction of outcome in the critically ill: arterial to intramucosal pH gradient and carbon dioxide gradient. Anaesthesia 1997; 52: 619–23.
25. Delva E, Camus Y, Paugam C, et al. Hemodynamic effects of portal triad clamping in humans. Anesth Analg 1987; 66: 864–8.
26. Shangraw RE, Winter R, Hromco J, et al. Amelioration of lactic acidosis with dichloroacetate during liver transplantation in humans. Anesthesiology 1994; 81: 1127–38.
© 2001 International Anesthesia Research Society