Search

The amount of transfused blood is related to blood loss calculated for the specific type of surgical procedure, transfusion hematocrit trigger and patient's red blood cell mass on the day before surgery. To optimise the benefit/cost and benefit/risk ratios of blood transfusion, a correct prescription must be done in accordance with the patient's red blood cell mass and surgical blood loss. Indeed, there is a clear need to define the appropriate uses of blood management methods and to seek new methods of improving perioperative blood management. The number of moderately anaemic patients undergoing surgery is currently thought to be 20%. Where transfusion requirements are estimated at two to three blood units, as for instance in the most common types of orthopaedic surgery, preoperative haemoglobin is the key factor governing transfusion needs. In this case, the simplest approach is to prescribe Epoetin Alfa subcutaneous at a dose of 600 IU/kg/week starting three weeks before the surgery. In addition, it is important in all cases to give concomitant iron supplements. Concomitant use of other methods to decrease allogeneic blood requirements is of no value. Obviously, the higher the haematocrit the day prior to surgery, the higher the patient's RBC mass and the greater the patient's permitted blood loss, decreasing the transfusion trigger. In this way, allogeneic blood loss is reduced, but without the need for the patient to attend the blood transfusion center and to undergo laboratory screening and testing of donated blood, and without the risk of inducing preoperative anaemia compared with sequential autologous blood donation. But, to optimise the benefit/cost ratio, we try to define precisely the patient populations likely to benefit from preoperative erythropoietin. Using different examples, management is proposed with algorithms.

Disseminated intravascular coagulation (DIC) is an abnormal extension of a coagulation process normally limited in time and space. DIC results from the release of free thrombin in circulating blood, as thrombin synthesis and inactivation mechanisms non longer function properly. In addition to microthrombosis, acute DIC may lead to haemorrhage related to activation of the coagulation process. According to the disease aetiology, emergency treatment may become a top priority, in case of haemorrhage, shortage of platelets and coagulation factors related to increased use of these blood components requires replacement treatment with infusion of platelets and fresh frozen plasma. As treatment of the procoagulation process, heparin should not be recommanded in acute DIC with haemorrhage. Based on rational theoretical data, current evaluation of antithrombin therapy is in progress.

Objectives: To compare O2 uptake values measured by indirect calorimetry (VO2R) with Fick-derived values (VO2Fick) over a wide range of VO2 in experimental conditions and to analyze the influence of cardiac output (QC) on the difference between VO2R and VO2Fick., Study Design: Animal study., Material: Nineteen 2.5-month-old, 29.2 +/- 2.8-kg, Large White pigs., Methods: A step-by-step decrease in venous return was obtained either by haemorrhage or progressive inferior vena cava and portal veins clamping. Measurements, of 5 to 7 minutes duration included VO2R (Deltatrac), thermodilution QC and arterial and mixed venous blood O2-content. The VO2 values were compared using Bland and Altman's bias analysis. The relationship between QC values and relative error (ER = [VO2R - VO2Fick]/0.5x[VO2R + VO2Fick] was analyzed using a Spearman rank correlation coefficient., Results: The VO2R and VO2Fick arithmetic mean ranged from 108 to 253 mL.min-1 in 130 measurements with broad limits of agreement between both methods. On average, VO2R values were higher than VO2Fick values by +15 +/- 25 mL.min-1. ER significantly decreased with QC (rs = 0.417; P = 0.0001)., Conclusion: VO2R values exceed VO2Fick values. This bias does not occur in the low QC range, presumably owing to artifactual over estimation of thermodilution QC.

Human serum albumin (HSA) is quantitatively the most important non specific transport protein. HSA binds a wide variety of both endogenous and exogenous ligands. Hypoalbuminaemia may lead to a decreased plasma binding capability of some compounds. Biological/pharmacological consequences depend on the ligand and the target tissue. Many experimental studies suggest that hypoalbuminaemia may influence the metabolism and toxicity of endogenous ligands (bilirubin, metallic ions, oxygen radicals) and the pharmacological effect of some drugs (among others: furosemide, phenytoin, warfarin). The relevance of such information for human surgical situations remains unclear. Clinical studies are scarce and inconclusive. There is a lack of pertinent data supporting the necessity of HSA infusions in order to maintain a minimal plasma concentration and a convenient plasma transport. However, experimental data indicate that major hypoalbuminaemia should be considered with caution.

Multiple-organ failure (MOF), defined as the failure of initially uninvolved organs, is the final step of definitive and massive liver necrosis. Emergency liver transplantation (ELT) has radically modified the outcome of acute liver failure and early primary graft failure, but the results of ELT in cases of MOF are unknown. From May 1988 to June 1993, 243 patients underwent a liver transplantation (LT). Thirty-seven patients (15.2%) who had an acute liver necrosis complicated by a MOF underwent an ELT. Twenty-one patients were children. An emergency retransplantation was performed in 16 patients. Three or 4 organ-system failures (OSF) were present in 13 patients. Before ELT, the mean Acute Physiology and Chronic Health Evaluation (APACHE) II score was 26.3 +/- 5.1. Six-month and 1-year survival rates were 37.8% and 25.9%, respectively, after ELT complicated by MOF, and 78% and 73.5%, respectively, in other cases of LT. Twenty-six patients had surgical complications (70%), whereas thirty-one patients had medical complications (84%). Twenty-two patients died during the postoperative period (60%). Before ELT, infection (P < .05), cardiovascular failure (P < .03), and more than two OSF (P < .05) were more frequent in patients who died after intervention. The APACHE II score (P < .05) and the length of stay in the intensive care unit before ELT (P < .05) were lower among survivors. In the context of liver allograft shortage, our results suggest that an ELT should not be performed in patients with cardiac failure, more than two OSF, or an APACHE II score higher than 30.

The relative contribution of 14 preoperative risk factors to a high intraoperative blood loss was studied in 95 consecutive first pediatric orthotopic liver transplantations (OLT). Patients were distributed in two groups according to red blood cell (RBC) requirements. Wide interindividual RBC requirements were observed (median, 79 mL/kg; range, 4-586). The upper quartile of the population was defined as the high blood loss group and required 123 mL/kg or more (median, 161). On univariate analysis, the high blood loss group had a significantly higher proportion of patients with portal vein hypoplasia, intraabdominal malformations, signs of severe liver failure (encephalopathy, ascites, prolonged prothrombin time), and requiring inpatient support. Age, previous abdominal surgery, and platelet count had no prognostic value. All variables used in the univariate analysis were included in a stepwise logistic regression analysis. Only presence of portal vein hypoplasia, inpatient support, and use of a reduced-size liver graft were independently associated with a high blood loss. Adjusted odds ratios were 40.4 (95% confidence interval; 5.9-278), 5.4 (1.6-17.9), and 3.8 (0.9-15.2), respectively, highlighting the importance of portal vein hypoplasia as a risk factor for high blood loss.

Perioperative intensive care has several objectives in the context of liver transplantation. Prior to the operation, it is essential for the management of fulminant hepatic failure in order to limit the consequences of cerebral oedema. During the operation, intensive care consists of monitoring and correction of the occasionally severe circulatory and metabolic disorders induced by vascular exclusion of the liver and revascularization of the cold, ischaemic, orthotopic transplant. The possibility of major haemorrhage requiring massive transfusion with stored blood aggravates these disorders. Complex clotting disorders also play an important role; their mechanism is imperfectly understood and they are difficult to control. The severity of the problems encountered after the operation depends on graft function.

Seven partial liver resections of at least six hepatic segments (five extended left hepatectomies, also called left trisegmentectomies, and two subtotal hepatectomies) were performed for tumor (of malignant nature in 6 cases). Six of seven patients were admitted for liver transplantation and one for liver resection. These procedures were performed under total vascular exclusion of the liver in six cases (with a veno-venous bypass in two cases) and with the Pringle maneuver in one case. Postoperative liver function was satisfactory in six cases whereas one patient died of hepatic failure on the tenth postoperative day. There was one bile duct stricture at eight months, after extended left hepatectomy. Four patients survived more than one year, three with hepatic tumor recurrence and one with distant metastases. Two patients were still alive, one after fourteen months with tumor recurrence in the liver, the other after seventeen months. In selected cases, partial hepatic resection of more than six segments can be performed for very large tumors. In cases of malignant tumors, a short period of survival with good quality of life can be offered to these patients.

We report our results with orthotopic liver transplantation in children with fulminant liver failure. Thirty-five children with fulminant liver failure were evaluated for liver transplantation. The main causes of liver failure were viral hepatitis (54.2%), drug-induced liver injury (14.2%) and Wilson's disease (11.4%). Children were considered as candidates for liver transplantation only if hepatic encephalopathy was associated with a decrease in the level of factor V to below 25%. Seven children (20%) did not meet this criterion and recovered spontaneously. Six children (17.1%) had contraindications for liver transplantation and died. In three of these six children, contraindications included irreversible brain damage at the time of admission. Twenty-two children (62.8%) met the criteria for liver transplantation and were placed on the emergency transplant list. Three of them died awaiting grafts. Nineteen children underwent liver transplantation; 13 of them (68.4%) are alive without sequelae, after 6 mo to 4 yr of follow-up, at this writing. Four of the children who died after surgery had severe encephalopathy on admission that did not improve after liver transplantation. In conclusion, emergency liver transplantation appears to be an effective treatment for children with fulminant liver failure. Nevertheless, irreversible brain damage developed in 10 patients, and they died before or after surgery. We postulate that many of these deaths could have been avoided if children had been transferred to a liver transplantation facility and had undergone transplantation earlier. We emphasize that children with acute liver failure should be transferred to a center that performs liver transplantation before the development of hepatic encephalopathy.

Cell-mediated immunity is impaired during cholestasis, and there is evidence that bile acids play a role in this immune defect. Ursodeoxycholic acid (UDCA), which corrects the immunological abnormalities observed in primary biliary cirrhosis, could counter the detrimental effects of the endogenous bile acids. Accordingly, we assessed the respective effects of cholestasis, chenodeoxycholic acid (CDCA), and UDCA, using mixed lymphocyte culture as a model of allogeneic immune response. CDCA induced a dose-dependent inhibition of the proliferative response (0-150 mumol/L). Mononuclear cells obtained from bile duct-ligated mice had a normal immunostimulatory effect, whereas responder cells obtained from such animals showed a profoundly impaired proliferative response, suggesting that responder T cels are the main target of the cholestasis-induced immune defect. Supplementation of cultures with exogenous interleukins partially compensated for the inhibitory effect of 25 mumol/L CDCA, but not for that of 50 mumol/L CDCA, suggesting that impaired secretion of interleukins is not the only factor involved in the effect of bile acids. In contrast to CDCA, UDCA had no inhibitory effect on the allogenic immune response at concentrations of up to 50 mumol/L.

This study retrospectively reviewed the results of 81 orthotopic liver transplantations in 72 pediatric patients with either a full-size graft (n = 41) or a reduced-size graft (n = 40) during a 4-year period. Two types of liver reduction were used to obtain a left lobe graft (n = 30) or a left lateral segment graft (n = 10). The choice of the reduction technique was based on two criteria: the donor/recipient body weight ratio and the transverse size of the recipient's abdomen. The patients who underwent transplantation with a reduced-size graft were younger and smaller than those who received a full-size graft. The mortality rate of the candidates on the pediatric transplantation waiting list was 2.7% during the same period. The 3-year survival rate after elective transplantation was 85% and 75% in the full-size graft group and the reduced-size graft group, respectively (difference not significant). The use of reduced-size grafts was associated with higher amounts of red cell and fresh-frozen plasma transfusions after graft revascularization. The incidence of postoperative bleeding and bile leakage from the transected surface was low, whereas hepatic artery and biliary complications were less frequent in the reduced-size graft group than in the full-size graft group. At the time of follow-up, liver biochemical test results and quality of life were not different between the two groups. Despite the constraining technique and the increase in intraoperative blood loss, orthotopic liver transplantation with a reduced-size graft allows us to overcome the shortage of pediatric donors and improve the overall survival rate of children with end-stage liver disease.

The goal of this study was to identify high-risk groups for cytomegalovirus infection after liver transplantation. Sixty-one patients were evaluated. Twenty-five patients (41 percent) had infection. Among the 16 patients who were seronegative for the virus before transplantation, 11 received a liver graft and blood products from seronegative donors and none of them developed infection. All seronegative recipients of a liver from seropositive donors (5/5) developed primary infection. Among the 45 patients seropositive before transplantation, 20 developed a cytomegalovirus infection, whatever the donor serologic status. The incidence of symptomatic reactivation or reinfection was high (14/20), and, for 12/14 of them, associated with early acute rejection. Two high-risk groups of patients, eligible for cytomegalovirus prophylaxis, were identified: seronegative recipients of seropositive donors and seropositive recipients with early acute rejection.

Nowadays the UW solution is usually used in liver transplantation for graft preservation. The importance of each of its components has not been fully resolved. The omission of some of the components does not result in poor hepatic function. The aim of this controlled study was to assess the early graft function (ALAT, factor V and total bilirubin at days 1, 2, 3 and 6) after use of the UW solution and SLF Cochin. SLF Cochin is a modified UW solution without hydroxyethyl starch, adenosine, penicillin, dexamethazone and allopurinol. No significant difference was observed in the early graft function between the two solutions for a cold ischemia time below 15 hours. Considering the qualities of the SLF Cochin and its three fold lower cost, it is concluded that the use of the SLF Cochin solution is justified for liver transplantation in which foreseeable cold ischemia time is less than 15 hours.

Of 220 patients undergoing liver transplantation between March 1982 and April 1991, eighteen (8.1 percent) already had a surgical portasystemic shunt. Four patients had a distal splenorenal shunt, six a side-to-side portacaval shunt, three an end-to-side portacaval shunt, and five, a mesocaval shunt. The splanchnic venous system was assessed by Doppler ultrasound examination and angiography before liver transplantation. Perioperative mortality rate was higher but not significantly different from that observed in the population of patients without previous portasystemic shunt (33.3 percent versus 21.7 percent). Liver transplantation was particularly difficult in the six patients with a previous side-to-side portacaval shunt, in which the perioperative mortality rate was high (66.6 percent). Liver transplantation is feasible in patients with a previous portasystemic shunt but carries a higher risk. It is easier when no previous dissection of the hepatic pedicle has been done. Suppression of the shunt after revascularization of the graft must be performed in order to provide optimal venous flow to the hepatic graft.

Hepatic artery thrombosis and early acute rejection are severe complications of orthotopic liver transplantation (OLT). Their rapid detection is most desirable. The purpose of this study was to assess the usefulness of monitoring hepatic artery (HABF) and portal vein (PVBF) blood flows during the first week after OLT. At the end of operation, microprobes were sutured to the vessels, and their connecting tubes were externalized and connected to a pulsed Doppler flowmeter operating at 8 MHz. In 10 patients (ages ranged from 2 to 54 years) of 106, measurements of HABF and PVBF were done during alternative clamping of both vessels and before and after abdominal closure, every 12 hours during 7 days, and at day 7, before and after a 150 gm carbohydrate meal. At day 7 the probes were pulled out by gentle traction without complication, and all patients were allowed to go home. Reciprocal increase of flow during selective clamping was only observed for HABF (+45.8% +/- 47.6%; p less than 0.01). Abdominal closure decreased both HABF and PVBF by 13.8%, p less than 0.01, and 26%, p less than 0.05, respectively. In seven cases no significant variation of HABF and PVBF was observed during 7 days. In two patients with histologically confirmed early acute rejection, a marked decrease of diastolic HABF, without modification in PVBF, was the first manifestation and was rapidly corrected by boluses of steroids. In one patient disappearance of systolic and diastolic HABF led us to diagnose an arterial obliteration caused by a plicature, which was successfully surgically treated in the emergency department. In all patients, after oral ingestion of the carbohydrate meal, and only after this type of diet, a significant and deep decrease (-87%, p less than 0.001) of HABF was observed between 7 and 120 minutes without any change in PVBF. Such an effect was not observed in control patients. We conclude that this Doppler flowmetric technique with implantable microprobes is useful for rapid diagnosis of and strategy in treating early complications and is a new tool for pathophysiologic study of OLT consequences.

In 10 patients undergoing therapy for a mild exacerbation of their chronic obstructive pulmonary disease (COPD), a quantitative two-dimensional echocardiographic (2DE) study was performed together with hemodynamics to assess left ventricular (LV) function. From the 2DE examination, which was made up of parasternal, subcostal, and apical views, measurements of LV short axis end-diastolic and end-systolic areas (A) at the high papillary muscle level and long axis end-diastolic and end-systolic length (L) permitted us to calculate LV end-systolic and end-diastolic volumes (V) using the formula V = 5/6 AL. Compared with the same measurements obtained in a group of 12 normal volunteers, patients with COPD exhibited a markedly reduced LV cavity (LVES, 28.9 +/- 14.6 ml/m2 versus 51.5 +/- 11.0 ml/m2; LVEDV, 67.7 +/- 24.6 ml/m2 versus 103.2 +/- 19.9 ml/m2). An increased thickness of both left ventricular free wall and interventricular septum was also evidenced in patients with COPD. Left ventricular systolic function, assessed using both peak systolic blood pressure/end-systolic volume ratio and calculated left ventricular ejection fraction, was found to be clearly enhanced in patients with COPD. The influence of right ventricular enlargement on left ventricular diastolic function was also investigated in patients with COPD using progressive volume loading and 2DE right ventricular measurements. After a given threshold of volume loading, reduction in stroke index, opposite variations in right and left ventricular size and septal flattening, suggested the occurrence of ventricular interaction.

Circulatory failure occurs in about 10% of patients with pulmonary embolism, resulting from a massive obstruction of the pulmonary arterial bed. Hemodynamic and respiratory features are well established; they involve precapillary pulmonary hypertension, low cardiac output state, elevated filling pressure for the right ventricle, and venous admixture. More recently, two-dimensional echocardiography permitted the visualization of pulmonary artery and right heart enlargement, reduced right ventricular ejection fraction, and tricuspid regurgitation. Evaluated by this latter means, left ventricular systolic function appeared unchanged, but diastolic function might be reduced by septal bulging.

The effect of ascites on splanchnic and systemic hemodynamics was retrospectively studied in 256 patients with alcoholic cirrhosis. The patients were divided into 3 classes: no ascites, moderate ascites, and large ascites. They were also classed into 3 groups according to a modified Pugh classification (ascites was not taken into account). In patients without ascites, cardiac output was positively related to the severity of liver disease. This result was not found in patients with large ascites. Similarly, in patients with no or moderate ascites, the degree of portal hypertension estimated by the hepatic venous pressure gradient was associated with liver failure. In patients with large ascites, hepatic venous pressure gradient was higher than in patients without ascites but this relation was observed only in patients without liver failure. These results show that ascites reduces the relation between cardiac output and liver failure and increases the degree of portal hypertension but not cardiac output.

Massive obstruction of the right branches of the pulmonary artery by choriocarcinomatous cells was found at embolectomy in a 36 year old woman who had had a hydatidiform mole 10 years previously. Several febrile lung infections resistant to antibiotic therapy over an 11 months period, had led to a reduction in the volume and vascularisation of the right lung. The diagnosis of recurrent pulmonary embolism was only suspected when a further embolic episode occurred after curettage for a spontaneous abortion at 9 weeks of pregnancy. The diagnostic value of increased HCG was masked by the pregnancy. The patient died a few days after surgery despite administration of chemotherapy postoperatively. The authors discuss the problems in the diagnosis of recurrent pulmonary embolism in the absence of a suggestive cause and the problem of detecting the tumoural nature of these emboli.

Acute right ventricular failure is commonly observed during respiratory intensive care, particularly in patients suffering from massive pulmonary embolism, chronic obstructive pulmonary disease or adult respiratory distress syndrome. The haemodynamic effects of a continuous dobutamine infusion at the rate of 9.4 +/- 3.7 micrograms/kg/min were assessed in a group of 15 patients with acute and isolated right ventricular failure, as evidenced by haemodynamic and two dimensional echocardiographic measurements. This inotropic agent induced at 37% increase in mean cardiac index and a 25% increase in mean stroke index, with only a 10% increase in mean heart rate. Moreover, measurement of the right ventricular ejection fraction by a thermal dilution technique performed in 10 patients demonstrated that dobutamine consistently and significantly increased right ventricular systolic function and also significantly reduced right ventricular end-diastolic tension. In conclusion, it appeared that dobutamine was able to improve circulatory status in patients with acute right heart failure or various origins.

To elucidate the relationship between oxygen transport and uptake in cirrhosis, we studied the effects of three vasoactive drugs that change O2 transport. Systemic hemodynamics, blood gases and lactate concentration were measured in patients with alcoholic cirrhosis before and after intravenous dobutamine, propranolol and nitroglycerin. Nine patients received successively dobutamine and then propranolol. Ten patients received nitroglycerin. Three other patients without cirrhosis (controls) received dobutamine. In patients with cirrhosis, dobutamine infusion was accompanied by a significant increase in cardiac output (+21%), systemic O2 transport (+21%) and O2 uptake (+12%), whereas O2 extraction ratio and arterial lactate concentration did not change significantly. Propranolol administration was followed by a significant decrease in cardiac output (-24%) and systemic O2 transport (-25%) and a significant increase in O2 extraction ratio (+19%), whereas O2 uptake and arterial lactate concentration did not change. Nitroglycerin infusion was accompanied by a significant decrease in cardiac output (-21%), systemic O2 transport (-26%) and O2 uptake (-10%), whereas O2 extraction ratio (+18%) and arterial lactate concentration (+31%) significantly increased. In control patients, dobutamine infusion was accompanied by an increase in cardiac output and in systemic O2 transport and by a decrease in O2 extraction ratio, whereas O2 uptake was not modified. These results suggest that O2 uptake may be abnormally dependent on O2 transport in patients with cirrhosis.

We investigated the effect of liver denervation on cardiovascular homeostasis. Three days after surgical denervation of the liver, hemodynamic studies (radioactive microsphere method) were conducted in conscious rats. The efficacity of the liver denervation procedure was confirmed by a significant decrease in norepinephrine content in various lobes of the liver. Liver denervation did not affect either systemic or splanchnic resting hemodynamics. However, hemorrhage (2 ml per 100 gm body weight) induced a decrease in cardiac index which was significantly more marked in rats with liver denervation (-62 +/- 3%) than in sham-operated rats (-47 +/- 5%; p less than 0.05). This more severe response to hemorrhage may be due to a lack of portal territory vasoconstriction since the fraction of cardiac output reaching portal territory did not decrease during hemorrhage in rats with liver denervation (16 +/- 1% to 14 +/- 1%) but significantly decreased in sham-operated rats (from 15 +/- 1% to 11 +/- 1%, p less than 0.05), this value being significantly lower, after hemorrhage, in sham-operated rats than in denervated rats. The results confirm the lack of tonic neural influence on hepatic circulation during physiological experimental conditions and indicate that hepatic nerve function has a significant contribution to the overall cardiovascular homeostasis.

Recent improvements in the results of orthotopic liver transplantation (OLT) have made this a well-accepted treatment for patients with severe hepatic failure. Current problems encountered following OLT are discussed. Immediate complications comprise surgical bleeding, primary graft non-function, and graft failure due to hepatic artery occlusion. Secondary complications are frequent. Surgical ones include biliary and vascular (hepatic artery thrombosis most often) problems, as well as intra-abdominal abscesses associated with gastrointestinal perforation, biliary leak, graft ischaemia or an infected haematoma. 40% of patients having undergone OLT will be reoperated on, 2/3 of them within 3 months. Non-surgical complications are mostly pulmonary. The risk of pneumonitis is increased by prolonged mechanical ventilation; it is always potentially disastrous in the immunosuppressed, transplanted patient. Hypertension is also often seen in the early postoperative period; it requires prompt treatment. Early renal impairment after OLT is common, and of better prognosis than late onset renal failure, which is generally associated with shock, graft failure, sepsis or use of nephrotoxic agents. Seizures, usually only one, occur in about 10% of patients; recovery is complete. Encephalopathy with intracranial oedema related to fulminant hepatitis has a worse prognosis, but survival figures are quite encouraging. Three type of rejection are described after OLT: 1) severe accelerated rejection (very rare), 2) acute rejection encountered in about 70% of patients over the first 3 months, and 3) late rejection, which can lead to the vanishing bile duct syndrome (VBDS). Diagnosis of rejection is made by liver biopsy. Prophylactic immunosuppression includes cyclosporin, methylprednisolone and azathioprine. Cyclosporin toxicity and drug interactions are reviewed. Treatment of acute rejection episodes comprises an initial bolus of high doses of corticoid drugs; if there is no response, antilymphocyte globulin or monoclonal antibodies may have to be used. Infection is the main cause of death following OLT. Early infections, mostly intra-abdominal and pulmonary, are bacterial or fungal. Vital (especially CMV) and other opportunistic infections occur generally after the second week. Retransplantation, carried out in 10 to 25% of patients, may be urgent in case of primary graft failure, or hepatic artery thrombosis associated with graft failure, or hepatic artery thrombosis associated with graft failure. Other indications are early graft rejection with severe hepatic dysfunction, chronic rejection with severe VBDS, and recurrence of the initial disease.

The effects of clonidine on portal pressure and splanchnic blood flow were studied in conscious rats with sinusoidal portal hypertension due to cirrhosis induced by bile duct ligation. In cirrhotic and sham-operated rats, clonidine (20 micrograms per kg body weight, intravenously) significantly reduced portal, pressure from 19.0 +/- 0.6 to 14.5 +/- 1.0 mmHg and from 9.8 +/- 0.9 to 7.3 +/- 0.5 mmHg, respectively. No significant change in systemic hemodynamics was observed. In cirrhotic rats, clonidine reduced portal pressure, probably by producing a significant increase in portal tributary vascular resistance leading to a 25% decrease in portal tributary blood flow (radioactive microsphere method). In sham-operated rats, clonidine reduced portal pressure presumably by decreasing hepatic portal vascular resistance, since no significant change in portal tributary blood flow was observed. In both groups, clonidine administration significantly decreased plasma noradrenaline concentration. Placebo administration produced neither significant hemodynamic nor significant plasma noradrenaline concentration change. These findings indicate that the sympathetic regulation of the splanchnic circulation is impaired in cirrhotic rats.

In animals, there may exist a hyperemic response in the portal circulation during intravenous administration of hypertonic glucose, but a hemodynamic response of this kind has never been described in man. This study was designed to evaluate if hyperglycemia itself could induce systemic or splanchnic hemodynamic changes in patients with cirrhosis. Sixteen patients with cirrhosis were studied before and during i.v. infusions of hypertonic (900 mOsmoles per liter) glucose (n = 8), mannitol (n = 4) or saline (n = 4) at 2 ml per min. In the group receiving glucose, there were significant increases in hepatic venous pressure gradient (+12%), azygos blood flow (+27%) and pulmonary capillary pressure (+32%), while calf blood flow decreased (-26%). No changes occurred in the mannitol or saline groups. Changes in plasma osmolality, plasma volume, splanchnic oxygen extraction and vasoactive hormones, including vasoactive intestinal polypeptide and glucagon, did not appear to be involved in the mechanism of these vasoactive phenomena. It is suggested that the possible deleterious effects of increase in portal pressure and azygos blood flow should be taken into consideration when administering hypertonic glucose to patients with portal hypertension.

A retrospective study was carried out a group of 50 adult patients who had presented with acute respiratory distress syndrome and had all been treated by mechanically controlled ventilation with positive end-expiratory pressure (PEEP). The predictive value of the so-called "severity index" was tested. This index, which is used from the second day of treatment onward to predict with strong probability whether the lung lesions will subside or develop, is obtained from the PaO2 after corrections taking into account the FiO2 and the PEEP level. The study also supplied information on the outcome of lung lesions which may result in one of four possible courses of about equal frequency: the patient may either die or be cured rapidly after less than a week of assisted ventilation, or he may die at a later date or be cured slowly after more than a week of assisted ventilation.