Ain-Shams Journal of Anaesthesiology

: 2014  |  Volume : 7  |  Issue : 3  |  Page : 263--268

Impact of sevoflurane versus isoflurane on coagulation profile in living donor liver transplant recipients: a prospective randomized trial

Waleed M.R. Elsarraf1, Tarek Salah2, Ahmad M Sultan2, Mohamed M El Shobari2, Mohamed Abdel Wahab2,  
1 Department of Anesthesia and Intensive Care, Gastroenterology Surgical Center, Egypt
2 Department of Surgery, Mansoura Faculty of Medicine, Mansoura University, Egypt

Correspondence Address:
Waleed M.R. Elsarraf
Department of Anesthesia and Intensive Care, Gastroenterology Surgical Center, Mansoura University, Mansoura, Egypte


Background End-stage liver disease is characterized by severe coagulopathy, and bleeding is common during liver transplantation (Ltx). Previous studies reported a depressant effect of some inhalational anesthetics on platelet function in normal patients. None of them investigated this effect in Ltx recipients with end-stage liver disease. In this study, we evaluated the effects of sevoflurane versus isoflurane on coagulation function, blood loss, and transfusion requirements in living donor liver transplantation recipients. Patients and methods A total of 32 patients of either sex, with MELD score between 12 and 18 scheduled for living donor liver transplantation, were randomly allocated into either the sevoflurane group (N = 18) or the isoflurane group (N = 14), based on the inhalational anesthetics used. Standard hemodynamic monitoring was applied. All operations were performed by the same anesthesia and surgery teams. All patients were administered propofol/fentanyl and rocuronium bromide for induction, followed by continuous infusion of fentanyl for analgesia and rocuronium bromide for muscle relaxation. Samples for INR, prothrombin time, bleeding time, Factor V, Factor VII, serum fibrinogen, complete blood picture, and aPTT were collected before the induction of anesthesia, end hepatectomy, 60 min after reperfusion, and 6, 12, 24 h in the ICU. Amounts of blood loss and blood components transfusion were also recorded at the end of operation and first postoperative day. Results Both the groups had comparable demographics, coagulation profiles, and MELD scores. Bleeding time was significantly longer in the sevoflurane group (P = 0.04), starting at end hepatectomy and persisted till the sixth hour postoperative reading, compared with the isoflurane group. INR, aPTT, PT, Factor V and Factor VII, and Hg were comparable in both groups. Platelet count was also not significantly different between both groups. Blood loss and transfusion requirements were comparable in both groups. No outcome difference was observed between groups. Conclusion In Ltx recipients, sevoflurane induced prolongation of the bleeding time compared with the isoflurane group. Yet, no clinically significant impact was encountered regarding blood loss and transfusion requirements either during intraoperative or the early postoperative period.

How to cite this article:
Elsarraf WM, Salah T, Sultan AM, El Shobari MM, Wahab MA. Impact of sevoflurane versus isoflurane on coagulation profile in living donor liver transplant recipients: a prospective randomized trial.Ain-Shams J Anaesthesiol 2014;7:263-268

How to cite this URL:
Elsarraf WM, Salah T, Sultan AM, El Shobari MM, Wahab MA. Impact of sevoflurane versus isoflurane on coagulation profile in living donor liver transplant recipients: a prospective randomized trial. Ain-Shams J Anaesthesiol [serial online] 2014 [cited 2022 Jan 20 ];7:263-268
Available from:

Full Text


Liver transplantation is the most effective management for end-stage liver disease (ESLD) [1]. Intraoperative bleeding during living donor liver transplantation is a common problem, and unfortunately plasma and red blood cell transfusion are linked adversely to 1-year survival rates [2].

ESLD is usually complicated by coagulation defects, which are known to be multifactorial. Decreased plasma levels and synthesis of coagulation factors, thrombocytopenia and thromboasthenia, are important factors [3].

Some anesthetics have antiaggregatory effect on platelets, and by using them we may increase surgical bleeding and hemorrhagic complications [4].

In a previous study, sevoflurane had no effect or increased the expression of P-selectin on platelets in-vitro samples; P-selectin was found to be increased in patients with stroke and acute artery syndromes, treated with angioplasty or thrombolysis [5].

Knowledge of the effect of inhalational anesthetics on coagulation and blood component requirement is very important, especially in ESLD and during and after liver transplantation operations.

We hypothesize that the use of inhalational anesthestics, such as sevoflurane and isoflurane, during long operations and in risky groups, such as, ESLD patients, may neither affect the coagulation profile nor the need for blood transfusion.

The effects of both isoflurane and sevoflurane on coagulation in Ltx patients with terminal cirrhosis is yet to be prospectively investigated, which is the objective of this randomized prospective trial.

 Patients and methods

After approval of the local ethical committee, and securing written informed patient's consent, a total of 32 patients aged 30-60 years, ASA II and MELD score between 16-18 scheduled for living donor liver transplantation in Mansoura university liver transplantation program for ESLD, were included in this study. Exclusion criteria were patients with hepatocellular carcinoma, severe portopulmonary hypertension, veno-occlusive diseases, history of recent intake of drugs affecting coagulation, previous major upper abdominal surgery, and hepatocellular carcinoma. Patients were randomly allocated, using closed envelops, into either the sevoflurane group (N = 18) or the isoflurane group (N = 14), according to inhalational anesthetics used for the maintenance of anesthesia. Patients were premedicated with 3 mg midazolzm (Midazolam Hameln Pharmaceuticals, Germany) and 40 mg of pantoprazole intravenously in the recovery area. The induction of anesthesia was carried out in the operating room using 2 mcg/kg fentanyl (Fentanyl; Janssen-Cilag) and propofol 1 mg/kg (Propofol 1% Fresenius). Rocuronium bromide 0.6 mg (Esmeron; Organon) was used to facilitate endotracheal intubation. Anesthesia was maintained with sevoflurane or isoflurane 1-2 MAC in 0.4 air : oxygen mixture. Analgesia was achieved with continuous infusion of fentanyl 1 mcg/kg/min, whereas muscle relaxation was carried out by continuous infusion of rocuronium 300 mcg/kg/min and mechanical ventilation was adjusted to maintain an end-tidal CO 2 around 33 ± 2 mmHg. Patients in both the groups were monitored intraoperatively using Datex S5 (Datex Ohmeda Monitor) for continuous follow-up and monitoring of ECG, heart rate, oxygen saturation, body temperature (using the esophageal probe), and invasive arterial blood pressure. Pulmonary artery catheter was introduced in all patients after the induction of anesthesia and verification of its position was done using fluoroscopy and waveform readings. Continuous cardiac output monitoring, pulmonary artery pressure, and mixed venous oxygen saturation were recorded.

Sixteen gauge intravenous catheters were inserted in the left and right external jugular veins and the basilic vein under complete aseptic conditions. Blood salvaging system (XTRA; Sorin) was used in all patients to minimize the homologues blood product transfusion, using a fixed standard priming protocol in all cases. Intraoperative blood transfusion followed the ASA guidelines (Massicotte 2006), whereas fluid transfusion was guided using stroke volume, where boluses of 250 ml of albumin 4% or hydroxyethyl starch 130.0.4 (Voluven) were administered, if the stroke volume decreased by 20% of the basal value. Venous blood samples for the monitoring of Hg concentration, INR, prothrombin concentration and time, bleeding time, factor V, factor VII, and aPTT and arterial samples for arterial blood gases were collected 10 min before the induction of anesthesia (basal), at end hepatectomy, 60 min postreperfusion, and 6, 12, and 24 h postoperatively (6PH), (12PH), and (24PH) in the ICU, as all patients were transferred to the ICU at the end of the procedure. Planned early extubation was adopted in all cases (about 2-4 h) in the ICU.

As several components are variably affected by the impaired liver functions in the liver transplant recipients with terminal cirrhosis, no single coagulation test could express the difference between the two inhalational agents clearly; therefore, an overall parameter such as blood transfusion was adopted as a primary outcome objective in our trial, particularly with a fixed surgical team. Secondary objectives in our trial included detailed assessment of the individual coagulation parameters as well as the early postoperative thrombotic complications.

Statistical analysis

A prior power analysis, using a-priori computation of one-tailed t-test using linear bivariate regression for the difference between the two groups intercepted, based on pilot data at our center, indicating that 18 patients in each group would be sufficient, based on effect size of 30% in bleeding and a type-I error of 0.05 and a power of 80%.

Statistical package for social sciences, Ver 17, was used to analyze the results. Data were tested for normality using Kolmogorov-Smirnov test. Two-tailed independent sample Student's t-test assuming equal variance was used to compare the parametric values in both groups, whereas Mann-Whitney U-test was conducted to compare the nonparametric values. Data were expressed as mean ± SD or percentage, as appropriate. A value of P less than 0.05 was considered to represent statistical significance. Nominal parameters were assessed by χ2 -test.


A total of 43 patients were examined for enrollment in this trial. Of them, 11 did not meet the inclusion criteria, whereas 32 were enrolled in the isoflurane and sevoflurane groups, respectively (n = 14 and 18). All enrolled patients completed the study and their data were analyzed [Figure 1]. Both groups were matched with regard to their demographic and operative data ([Table 1] and [Table 2]). [Table 3] demonstrates the coagulation parameters in both groups where no significant difference could be detected between the study groups [Table 3]. Neither blood loss nor the blood products transfusion exhibited any statistically or clinically significant differences between the studied groups [Figure 2]. Bleeding time exhibited a statistically significant higher value in the sevoflurane group compared with the Isoflurane group starting from end hepatectomy until 6 h postoperative [Figure 3].{Figure 1}{Figure 2}{Figure 3}{Table 1}{Table 2}{Table 3}


The only significant difference between the study groups in this trial was the bleeding time that was significantly prolonged in the sevoflurane group compared with the isoflurane group, starting at end hepatectomy and continuing up to 6 h postoperatively without being reflected on either blood loss or blood transfusion.

Coagulation process is normally modulated by the liver, which produces the coagulation factors (except Von Willebrand factor) and the coagulation inhibitors, and then the reticuloendothelial system clears the circulation from the activated coagulation factors and fibrin degradation products. ESLD is characterized by a failure of hemostasis owing to several factors: inability to synthesize clotting factors, abnormal fibrinolysis, thrombocytopenia, and platelet dysfunction [6].

Bleeding time is an easily conducted, frequently used test for primary hemostasis and a reliable indicator of platelet function [2]. Thrombocytopenia may partly explain the abnormal bleeding time as it occurs in 12-25% of cirrhotic patients with normal platelet counts [7-9]. In ESLD, 49-64% of patients suffer from mild-to-moderate thrombocytopenia, with platelet count varying between 30 000 and 40 000, although spontaneous bleeding is uncommon [10,11]. Thrombocytopenia may be caused by platelet sequestration in enlarged spleen, decreased platelet synthesis, and platelet destruction [6].

In our study, there was no significant difference in the platelet count in the sevoflurane and isoflurane groups (53 ± 14 and 49 ± 10 platelets/mcl), respectively; hence, we could exclude thrombocytopenia as a cause of the statistically different prolonged bleeding time in the sevoflurane group.

Impaired platelet aggregation occurs with ESLD and is reflected by a prolonged bleeding time in 40% of patients with cirrhosis [7-9,12]. It may be caused by circulating platelet inhibitors, plasmin degradation of platelet receptors, dysfibrinogenemia and excess nitric oxide production, which is a powerful vasodilator, and platelet adhesion and aggregation inhibitor produced in ESLD by vascular endothelium [8,9,13-16].

Dordoni et al. [17] studied the effect of thiopental, propofol, and sevoflurane on platelet function both in vitro and in vivo during thyroid operation. They showed that combination of thiopental, fentanyl, and sevoflurane decreased collagen-induced aggregation at the end of anesthesia induction; meanwhile, ADP-induced aggregation and generation of thromboxane were not affected. On the contrary, the combination of propofol, fentanyl, and sevoflurane did not affect platelet function. They concluded that thiopental had an inhibitory effect on platelet function in vitro in a dose-dependent manner, whereas propofol and fentanyl did not [17].

In the present study, anesthesia was induced using propofol, fentanyl with sevoflurane, or isoflurane for maintenance. On the basis of Dordoni study, excluding an effect of propofol and fentanyl leaves us with effects induced only by the inhalational agent used in this trial.

The effect of halothane on platelet function was first studied by Ueda [18]. He discovered that halothane inhibited ADP-induced platelet aggregation using aggregometry in canine blood [18].

In agreement with our results, Bozdogan et al. [19] reported that sevoflurane has an inhibitory effect on platelet function that was continued for 1 h after surgery, but was significantly decreased from levels found at 15 min after intubation.

Whereas several studies have shown the inhibitory effect of sevoflurane on platelet aggregation [4,17,20-25], others negated any platelet aggregation inhibitory effects of sevoflurane [21,26,27].

Yokubol et al. [20] showed that sevoflurane exerts its inhibitory effect on platelet function through decreasing cyclo-oxygenase activity and decreasing thromboxane A2 activity, leading to prolonged bleeding time in patients undergoing minor elective surgery. In agreement with our study, they noted that, despite prolonged bleeding time, blood loss and blood products transfusion were not affected [20].

Dogan et al. [21] showed that the inhibitory effect of sevoflurane on platelet function continued up to 1 h postoperatively after minor surgeries.

Contradicting our results, Hirakata et al. [24] showed that sevoflurane (0.13-0.91 mmol/l) and halothane (0.49-1.25 mmol/l) had inhibitory effects on platelet aggregation induced by ADP and epinephrine, whereas isoflurane (0.28-0.84 mmol/l) did not have that effect. They explained these effects of sevoflurane through the inhibition of cyclo-oxygenase activity and thromboxane A2 formation [24].

Huang et al. [5] also found that P-selectin surface expression decreased in both ADP-stimulated and ADP-unstimulated platelets in whole blood and in platelet-rich plasma in patients anesthetized with sevoflurane, indicating the inhibitory effect of sevoflurane on platelet aggregation. The P-selectin plays an important role in formation of platelet-leucocytes conjugates, a role that may be induced also by PSGL-1 (1ry legend for P-selectin), bridging with fibrinogen or thrombospondin [5].

In contrast, Horn et al. [22,28] demonstrated the increased levels or no effect on P-selectin in-vitro samples stimulated by ADP and thrombin receptors agonists peptide 6 during sevoflurane anesthesia. He also showed that sevoflurane anesthesia may increase binding of platelets to lymphocytes, neutrophils, and monocytes. This may be explained by the variations in circumstances in-vitro media, as in clinical situations platelet aggregation is thought to be induced by combined activation of multiple pathways, involving strong agonists and weak agonists separately. Therefore, the effect of sevoflurane on platelet aggregation in-vitro condition is not the same as in-vivo situations [28]. In addition, Lichtenfeld et al. studied the effect of volatile anesthetics on platelet aggregation, using aggregometry, and reported that platelet function is affected by anesthetics as well as by surgical stress [29].

The use of thromboelstography or platelet aggregometry would have given more insight into our study; yet the absence of such technology in our institute during implementation of the study is a major study limitation.


Sevoflurane did prolong bleeding time; yet, this effect was not reflected on blood loss, blood products transfusion, or patient outcome in this risky patient group.



1Juttner B, Brock J, Weissig A, et al. Dependence of platelet function on underlying liver disease in orthotopic liver transplantation. Thromb Res 2009; 124:433-438.
2 Gordon PC, Batty DJ. Haemostatic problems in liver surgery: a review. South Afr J Anaesth Analg 2009; 15:16-18.
3 Lisman T, Leebeek FW De Groot PG. Haemostatic abnormalities in patients with liver disease. J Hepatol 2002; 37:280-287.
4 Hirakata H, Nakamura K, Sai S, et al. Platelet aggregation is impaired during anaesthesia with sevoflurane but not with isoflurane. Can J Anaesth 1997; 44:1157-1161.
5 Huang GS, Li CY, Hsu PC, et al. Sevoflurane anesthesia attenuates adenosine diphosphate-induced P-selectin expression and platelet-leukocyte conjugate formation. Anesth Analg 2004; 99:1121-1126.
6 Kujovich JL. Hemostatic defects in end stage liver disease. Crit Care Clin 2005; 21:563-587.
7 Blake JC, Sprengers D, Grech P, et al. Bleeding time in patients with hepatic cirrhosis. BMJ 1990; 301:12-15.
8 Violi F, Leo R, Vezza E, et al. Bleeding time in patients with hepatic: relation with degree of liver failure and clotting abnormalities. J Hepatol 1994; 20:531-536.
9 Hsu WC, Lee FY, Lee SD, et al. Prolonged bleeding time in cirrhotic patients: relationship to peripheral vasodilation and severity of cirrhosis. J Hepatol 1994; 9:437-441.
10Bashour FN, Teran JC, Mullen KD. Prevalence of peripheral blood cytopenias (hypersplenism) in patients with nonalcoholic chronic liver disease. Am J Gastroenterol 2000; 95:2936-2939.
11Jabbour N, Zyjko A, Orons P, et al. Does trasjugular intrahepatic portosystemic shunt (TIPS) resolve thrombocytopenia associated with cirrhosis? Dig Dis Sci 1998; 43:2459-2462.
12Basili S, Ferro D, Leo R, et al. Bleeding time does not predict gastrointestinal bleeding in patients with cirrhosis. J Hepatol 1996; 24:574-580.
13Ballard HS, Marcus AJ. Platelet aggregation in portal cirrhosis. Arch Intern Med 1976; 136:316-319.
14Thomas DP, Ream VJ, Stuart RK. Platelet aggregation in patients with 6.
15Albornoz L, Bandi JC, Otaso JC, et al. Prolonged bleeding time in experimental cirrhosis: role of nitric oxide. J Hepatol 1999; 30:456-460.
16Coller BS. Platelets and thrombolytic therapy. N Engl J Med 1990; 322:33-42.
17Dordoni PL, Frassanito L, Bruno MF, et al. In vivo effects of different anesthetics on platelet function. Br J Hematol 2004; 125:79-82.
18Ueda I. The effect of volatile general anesthetics on adenosine diphosphate-induced platelet aggregation. Anesthesiology 1971; 34:405-408.
19Bozdogan N, Madenoglu H, Dogru K, et al. Effects of isoflurane, sevoflurane and desflurane on platelet function: a prospective, randomized, single blind, in vivo study. 2005; 66:375-384.
20Yokubol B, Hirakata H, Nakamura K, et al. Anesthesia with sevoflurane, but not with isoflurane, prolongs bleeding time in humans. J Anesth 1999; 13:193-196.
21Dogan IV, Ovali E, Eti Z, et al. The in vitro effects of isoflurane, sevoflurane, and propofol on platelet aggregation. Anesth Analg 1999; 88:432-436.
22Horn NA, de Rossi L, Robitzsch T, et al. Sevoflurane inhibits unstimulated and agonist-induced platelet antigen expression and platelet function in whole blood in vitro. Anesthesiology 2001; 95:1220-1225.
23Frolich D, Rothe G, Schmitz G, Hansen E. Volatile anaethetics induce changes in the expression of P-selectin and glycoprotein Ib on the surface of platelets in vitro. Eur J Anaesthesiol 1998; 15:641-648.
24Hirakata H, Ushikubi F, Toda H, et al. Sevoflurane inhibits human platelet aggregation and thromboxane A2 formation, possibly by suppression of cyclo oxygenase activity. Anesthesiology 1996; 85:1447-1553.
25Nozuchi S, Mizobe T, Aoki H, et al. Sevoflurane does not inhibit human platelet aggregation induced by thrombin. Anesthesiology 2000; 92:164-170.
26Mielke L, Kling M, Entholzner E, et al. The effect of general anesthesia with desflurane, sevoflurane or isoflurane on thrombocyte function. Anesth Analg 1997; 84:S1-S599.
27Mammen EF, Comp PC, Gossellin R, et al. PFA-100 system: a new method for assessment of platelet dysfunction. Semin Thromb Hemost 1998; 24:195-202.
28Horn NA, de Rossi L, Robitzsch T, et al. The effects of sevoflurane and desflurane in vitro on platelet-leukocyte adhesion in whole blood. Anesthesia 2003; 58:312-319.
29Lichtenfeld KM, Schiffer CA, Helrich M. Platelet aggregation during and after general anesthesia and surgery. Anesth Analg 1979; 58:293-296.