Abstract
Volatile anesthetics can elevate cytosolic free Ca2+ by releasing calcium from internal calcium stores and uptaking calcium from extracellular medium. Sevoflurane is an inhaled anesthetic used worldwide. A clear understanding of the exact mechanism of hepatic injury induced by sevoflurane remains elusive. A 69-year-old man with preexisting mild renal dysfunction, having undergone sevoflurane general anesthesia twice in 2 days, developed moderate jaundice. Liver enzymes strongly increased and remained elevated until death, which occurred on the 6th day after the first surgical intervention. The microscopic liver examination revealed an extensive and confluent hepatic necrosis, characterised by a large amount of calcium deposition in hepatic cell cytoplasm. These data were confirmed by confocal laser scanning microscopy and a 3-D visualization of calcium depositions was evident in hepatocytes cytoplasm. Our findings are suggestive with the previous experimental reports that consider elevation of cytoplasmic calcium may be the basis of sevoflurane – induced hepatotoxicity.
Introduction
Volatile anesthetics-induced hepatic dysfunction remains a significant diagnostic problem; it is difficult to be certain what caused the liver dysfunction in these subjects. Generally, the diagnosis of inhalation anesthetics-related hepatic injury is one of exclusion (Lee and Senior, 2005) after traumatic causes, other hepatotoxic drugs, abnormal physiological states (i.e., hypoxia) and infectious causes have been excluded. The effects of sevoflurane anesthesia on hepatic function and hepatocellular damage due to sevoflurane have been extensively investigated in animals (Topal et al., 2003). A case of liver dysfunction, but without fulminant hepatic failure, has been reported after sevoflurane anesthesia (Watanabe et al., 1993). To the best of our knowledge, no fatal cases due to hepatic failure following sevoflurane anesthesia have been reported. We report the case of a fatal hepatic failure in a 69-year-old man, with preexisting mild renal dysfunction, having undergone sevoflurane general anesthesia twice in 2 days.
Case Report
A 69-year-old man affected by chronic renal failure due to diabetic nephropathy, underwent surgery to repair the right succlavian artery, torn during central venous catheter insertion. The routine preoperative check was unremarkable except for a mild anemia (RBC 2.93, Hb 9.08, Hct 26.5); serologic tests for Epstein-Barr Virus, Cytomegalovirus and hepatitis viruses were negative. At admission arterial blood pressure was normal (120/80), AST, ALT, alkaline phosphatase, and LDH were normal. Bilirubin was within the normal range. BUN 60 mg/dL; creatinine 1.8 mg/dL, potassium 5.5 mEq/L, calcium 7.8 mg/dL, glucose 158 mg/dL. He had no previous history of general anesthesia or any known allergies. Preoperative arterial blood pressure was 140/70, with a heart rate of 130 b/min. The patient received an intravenous injection of thiopental (3 mg/kg) and succinylcholine (1 mg/kg) to facilitate tracheal intubation.
Anaesthesia was maintained with sevoflurane in oxygen (total flow 5 L/m). Surgical vascular repair was rightly performed and the anaesthetic concentration was adjusted by the anaesthesiologist to maintain mean arterial blood pressure within 80% –100% of the baseline value. End-tidal carbon dioxide (CO2) tension was maintained between 30–35 mmHg by controlled ventilation. No adjunct anaesthetics or vasoactive drugs were used. The endotracheal tube was removed after surgery in the operating room; when extubated, the patient was conscious and no neurological defect were observed; emogas analysis was nearly normal. Arterial blood pressure remained quite normal (125/60) postoperatively during the initial demonstration of increased AST and ALT.
From the 20th postoperative hour, he developed moderate deeping jaundice; from the 30th postsurgery hour, hepatic enzymes were strongly increased (AST 2196 U/L, ALT 641 U/L). Hypoproteinemia, decreased PT, ATIII and platelets concentration (PT 45.9%, ATIII 48.97%, PLT 81 × 103/μL) and increased fibrinogen degradation products (D-Dimeri 567.6) were detected. A surgical chest exploration was performed 40 hours after the first operation, by means of sevoflurane anesthesia; no bleeding was found. Also in the course of the second general anaesthesia hypotension or other evidence of ischemia were not demonstrated. Patient’s clinical condition rapidly got worse and he progressively developed hepatic dysfunction with severe coagulopathy, renal, respiratory, and cardiocirculatory failure (anuria, pO2 56.4 mmHg, arterial blood pressure 52/30 mmHg); a coma state (GCS 3) was declared. BUN 188 mg/dL; creatinine 4.5 mg/dL, potassium 5.2 mEq/L, calcium 8 mg/dL, glucose 218 mg/dL. Computerized tomography scans revealed a severe and diffuse cerebral edema. Hepatic enzymes persisted strongly increased; AST and ALT levels peaked at 2 days after the second sevoflurane exposure (AST 6169 U/L; ALT 1690 U/L), LDH peaked at 3 days (LDH 6397) and they remained elevated until death, which occurred on the 8th day after the first surgical intervention (Figure 1).
A complete postmortem examination was performed 24 hours after death. At autopsy the body weighed 80 kg length was 175 cm. The brain and the lungs were oedematous; the left pleural cavity contained 150 ml yellow-colored fluid, and the right one 200 ml. The peritoneal cavity contained 900 ml of yellow-colored fluid. The liver was reduced in size and weight, the capsule was wrinkled, and the organ was softened with multiple necrotic areas on section. The kidneys were small and showed a finely granular external surface (Figure 2). In cross-section, a gross diminution of cortical thickness with loss of demarcation between cortex and medulla was evident. Samples of organs were taken for histological examination and sections were stained by Hematoxylin & Eosin.
On liver specimens, periodic acid Schiff, reticulin stain, Perls and Von Kossa stainings were also performed. An immunohistochemical study to rule out an immune-mediated hepatitis was conducted with a panel of antibodies directed against the common lymphocyte antigen (CD 45), T lymphocytes (CD3, CD4, CD8), B lymphocytes (CD20), and macrophages (CD68 MAC387 and CD 68 KP 1). Additional tests were carried out in order to identify possible pathogenic agents through immunohistochemical studies with a panel of antibodies directed against Type 1 and Type 2 Herpes Virus antigens, Cytomegalovirus, Hepatitis B (HBsAg and HBcAg), Hepatitis C and D. In addition, the TUNEL assay was performed on liver samples; the positive reaction was visualised by a substrate kit for peroxidase, according to standard methods, the sections were counterstained with Methyl Green.
Microscopic examination revealed extensive and confluent bridging perivenular necrosis; cells were swollen, and vacuoles, pycnosis and cell fusion appeared; the presence of dark fine particles in cytoplasm was observed. In the perivenular zone we observed drop out of liver cells, a fine light brown PAS + intracytoplasmic pigment accumulation in hypertrophied Kupfer cells, and mononuclear cell infiltration. A surviving rim of hepatocytes around portal areas was evident (Figure 3). The Perl’s staining was negative. The Von Kossa staining showed an intensive reaction for calcium precipitation as a large amount of fine particles + in cytoplasm. Few apoptotic cells appeared evident in all the hepatic fields (Figure 4).
The immunohistochemical tests for the pathogenic agents were negative for all of the antigens tested. The histological examination of other organs was unremarkable except for massive cerebral edema, alveolar edema and hemorrhage, myocardial contraction band necrosis. In the kidney thickening of the glomerular basement membrane and expansion of the mesangium due to accumulation of extracellular matrix was observed; solid space of the tuft were increased and PAS positive material and acellular accumulations was observed within these areas. Calcium deposits were found only in the tubular lumen and they were seen as an intra-tubular brown material, hardly remembering granular casts (Figure 4).
The liver samples were examined under a confocal laser scanning microscope and a three-dimensional reconstruction was performed. A CLSM illuminates and detects the scattered or fluorescent light from the same volume within the specimen. CLSM allows the precise spatial and temporal analysis of intracellular Ca2+ activity at the subcellular level in addition to measurement of its concentration. This is due to the fact that the emitted fluorescence detected by the CLSM and multiphoton microscopes are limited to a specific focal plane. The CLSM produces images by moving a scanning point across the specimen and collecting the emitted fluorescence through a pinhole that is located at the confocal point of the scanned focus. By excluding the out-of-focus fluorescence from the fluorophore, CLSM offers high vertical and horizontal spatial resolution (<1 μm).
CLSM is able to produce thin and unblurred optical sections, offering the prospect of three-dimensional spatial reconstruction of the parameters of interest. Intracellular calcium detection was performed using a True Confocal Scanner microscope, Leica TCS SPE, Cambridge, UK equipped with an Leica DM5000 microscope. We used a 506 nm beam of an argon-ion laser and the fluorescence was acquired at wavelengths >531 nm in the line scan mode of the confocal system at rate of 5 ms per scan. A sequence of these sections collected along the z-axis can be used to render and quantitatively analyze the structure three-dimensionally. Calcium deposits were seen in the cytoplasm of the hepatocytes in the area of necrosis as needle—like deposits (Figure 5).
Discussion
In the present case, on the basis of clinical data, autoptical and histopathological findings, a cardiocirculatory failure due to fulminant hepatic failure was established as the cause of death and we suggest that liver failure may be related to sevoflurane exposure. The patient had no history of viral hepatitis; no hepatotoxic medications were administered except for inhaled anesthetic agents. The absence of transient hypoxia and hypotension excluding hepatotoxicity resulting from hypoxia. Since there are not, to date, reports of similar deaths, our report provides useful information on the present experience with this volatile anesthetic.
Sevoflurane is an inhaled anesthetic that is used world wide. A clear understanding of the exact mechanism of hepatic injury induced by sevoflurane remains elusive. Genetic factors, anesthetic toxic metabolites and idiosyncratic response have been hypothesized as possible causes (Cullen, 2005; Watkins, 2005). In particular, an immune reaction may be on the base of those rare, fulminant form of severe necrosis, which may be fatal (Puig et al., 2002).
Several studies have addressed on liver and renal function after repeated sevoflurane anaesthesia. Recent studies have reported hepatic damage after sevoflurane anesthesia in man (Watanabe et al., 1993; Reich et al., 2004). Eger et al. (1997) reported mild hepatotoxicity in human volunteers anaesthetized with sevoflurane. A reported case of severe sevoflurane hepatotoxicity in a child with mild renal impairment suggests that the decrease in renal function impairs the excretion of metabolites produced by sevoflurane with resultant hepatic dysfunction (Jang and Kim, 2005). However, many studies demonstrated that sevoflurane is not associated with decreased renal function in adults and children even in the presence of mild-to-moderate renal insufficiency (McGrath et al., 1998).
Tanikawa et al. (1994) concluded that repeated sevoflurane anaesthesia did not exert any adverse effect on hepatic and renal function in a pediatric patient who received sevoflurane anaesthesia 5 times within 40 days. Nishiyama et al. founded no additional risk of increasing serum concentration of liver enzymes or increasing urinary excretion of protein and glucose compared with the first exposure to the same anaesthetic in 30 patients after repeated exposure within 30 to 180 days to sevoflurane (Nishiyama and Hanaoka, 1998). The same authors stated that the second exposure to sevoflurane in ten patients (twice in 30–90 days) did not enhance the changes in hepatic and renal function induced by the first administration. More recently, Takenami et al. suggest that the second exposure to sevoflurane within 7 days interval does not alter the sevoflurane metabolism (Takenami et al., 2002). Experimental studies on dogs confirmed that repeated low-flow sevoflurane anesthesia did not affect hepatic and renal function significantly.
Alterations in intracellular calcium homeostasis have been implicated in hepatic injury induced by several mechanisms such as chemical intoxications and ischemia (Cullen, 2005). It is recognized that volatile anesthetics modulate the homeostasis of intracellular calcium; animal studies have provided evidence supporting a role of altered calcium fluxes in the mechanism of anesthetics induced injury (Yu et al., 2005). The effects of volatile anesthetics on metabolic alterations induced by the calcium ionophore A23187 were studied using isolated liver perfusion in fasted rats; the Authors demonstrated that all the studied anesthetics (halothane, isoflurane and sevoflurane) stimulated a dose—dependent release of radiolabeled calcium from internal calcium stores in isolated rat hepatocytes (Araki et al., 1997). Recently, it has been proposed that hepatic damage occurs due to the disruption of cellular calcium homeostasis mechanisms which appears to be less disturbed by sevoflurane rather than halothane. These anesthetics can elevate cytosolic-free Ca2+ by releasing calcium from internal calcium stores and uptaking calcium from extracellular medium. Cell injury due to loss of Ca2+ homeostasis correlates with blebbing of plasma membranes involving cytoskeletal proteins, Ca2+ ions, and Ca2+ dependant proteases (Yu et al., 2005).
Our findings are consistent with the previous experimental reports that consider elevation of cytoplasmic calcium may be the basis of sevoflurane-induced hepatotoxicity (Araki et al., 1997). In our fatal case, the man had a mild chronic renal failure, but no evidence of hepatic dysfunction was present until the administration of sevoflurane. He had no previous history of general anesthesia, viral hepatitis or any known allergies. No other hepatotoxic drugs were administered; operative hepatic perfusion problem, hypotension neither evidence of ischemia were demonstrated. Exitus was preceded by symptoms consistent with a fulminating hepatic failure. The microscopic liver examination led us to evidence an extensive and confluent hepatic necrosis, characterised by a large amount of calcium deposition in hepatic cell cytoplasm. These data were confirmed by confocal laser microscopy and a 3-D visualization of calcium deposition was evident in hepatocytes cytoplasm.