Timing and aetiology of bacterial infections in a liver intensive care unit

We undertook a prospective study of 887 consecutive adult patients admitted over an 11 year period to a liver intensive care unit. One or more bacterial infections occurred in 335 (37.8%) patients. Gram-positive cocci predominated. In relation to the date of admission these infections occurred in a statistically significant sequence. Streptococci infections were earliest (median time to infection two days), followed by Staphylococcus aureus (three days), coagulase-negative staphylococci (six days) and enterococci (eight days). Escherichia coli infections occurred earlier than those due to klebsiella-enterobacter (two vs seven days; P = 0.0001) and, overall, Enterobacteriaceae earlier than non-fermentative Gram-negatives (four vs. eight days; P = 0.0081). This study contributes to the management of high-dependency patients by confirming statistically the timing and sequence of infecting bacteria in patients with acute liver failure.

Severe hepatic dysfunction in pregnancy

Acute liver disease in pregnancy may have fatal consequences. Pre-eclampsia, HELLP syndrome and acute fatty liver of pregnancy form a spectrum of disease that range from mild symptoms to severe life-threatening multi-organ dysfunction. Early recognition of signs and prognostic indicators may enable prompt referral to specialist centres providing the multidisciplinary support required to reduce maternal and perinatal morbidity and mortality. We review the common causes of acute hepatic failure associated with pregnancy, and current management practices.

Liver transplantation in adults coinfected with HIV

OBJECTIVE

To report our experience of prospectively identifying and transplanting livers into HIV-positive patients.

DESIGN

Liver transplantation in HIV-positive patients remains controversial. The finding of HIV is usually considered a contraindication to any form of transplantation. Previously reported cases are few and refer to patients who tested HIV positive after they had their liver transplantations or who seroconverted in the posttransplantation period. This is, to our knowledge, the only report of patients who were known to be HIV positive at the time of decision for listing for transplantation.

METHODS

The medical records of five HIV-positive patients who received liver transplants in King's College Hospital, London, during a 5-year period (January 1995-December 1999) were reviewed. All five were known to be HIV positive at the time of listing for liver replacement. Three of them had end-stage liver disease due to hepatitis C (two of them had underlying Hemophilia A) while the other two had acute liver failure, one due to hepatitis B infection and one due to nonA-nonB-nonC hepatitis. In all but one patient the HIV infection had been asymptomatic.

RESULTS

All patients survived the immediate posttransplantation period, but the three patients with hepatitis C died of complications of recurrent hepatitis C between 6 and 25 months posttransplantation. The other two patients are currently alive 4 and 34 months posttransplantation with good graft function and without complications from their HIV infection.

CONCLUSION

The early outcome of liver transplantation in HIV seropositive patients can be good, and patients should not be excluded from transplantation if their liver disease determines their prognosis. More effective antiviral therapy for hepatitis C given posttransplantation, and for hepatitis B reinfection, should improve the longer-term outcome of HIV patients with end-stage liver disease due to hepatitis.

Acute liver failure as the initial manifestation of acute leukaemia

BACKGROUND/AIMS

Haematological malignancies seldom cause clinically significant liver disease. Acute liver failure as the initial manifestation of acute leukaemia is very rare and carries a very poor prognosis.

METHODS/RESULTS

Three cases of acute liver failure secondary to acute leukaemia are described. Each case presented initially as acute liver failure of uncertain cause. Specific treatment for the leukaemia was instituted; however, all three patients died as a consequence of the liver failure. We describe the clinical course and relevant investigations of these patients and discuss possible mechanisms of acute liver failure in this setting.

CONCLUSION

Acute leukaemia presenting as acute liver failure has a very poor prognosis. Although a rare cause of acute liver failure, it should be considered in any patient presenting with acute liver failure with prodromal symptoms and a raised peripheral white cell count, lactate dehydrogenase and uric acid.

Measurement of twitch transdiaphragmatic, esophageal, and endotracheal tube pressure with bilateral anterolateral magnetic phrenic nerve stimulation in patients in the intensive care unit

OBJECTIVE

In the critically ill, respiratory muscle strength usually has been assessed by measuring maximum inspiratory pressure. The maneuver is volitional, and results can be unreliable. The nonvolitional technique of bilateral anterolateral magnetic stimulation of the phrenic nerves, producing twitch transdiaphragmatic pressure, has been successful in normal subjects and ambulatory patients. In this study we used the technique in the intensive care unit and explored the measurement of twitch endotracheal tube pressure as a less invasive technique to assess diaphragmatic contractility.

DESIGN

Clinical study to quantify diaphragm strength in the intensive care unit.

SETTING

Patients from three London teaching hospital intensive care units and high-dependency units.

PATIENTS

Forty-one intensive care patients were recruited. Of these, 33 (20 men, 13 women) were studied.

INTERVENTIONS

Esophageal and gastric balloon catheters were passed through the anaesthetized nose, and an endotracheal tube occlusion device was placed in the ventilation circuit, next to the endotracheal tube. Two 43-mm magnetic coils were placed anteriorly on the patient's neck, and the phrenic nerves were stimulated magnetically.

MEASUREMENTS AND MAIN RESULTS

On phrenic nerve stimulation, twitch gastric pressure, twitch esophageal pressure, twitch transdiaphragmatic pressure, and twitch endotracheal tube pressure were measured. Forty-one consecutive patients consented to take part in the study, and twitch pressure data were obtained in 33 of these. Mean transdiaphragmatic pressure was 10.7 cm H2O, mean twitch esophageal pressure was 6.7 cm H2O, and mean twitch endotracheal tube pressure was 6.7 cm H2O. The mean difference between twitch esophageal pressure and twitch endotracheal tube pressure was 0.02 cm H2O. Correlation of the means of twitch endotracheal tube pressure to twitch esophageal pressure was 0.93, and that for twitch endotracheal tube pressure to transdiaphragmatic pressure was 0.78.

CONCLUSIONS

Transdiaphragmatic pressure can be measured in the critically ill to give a nonvolitional assessment of diaphragm contractility, but not all patients can be studied. At present, the relationship of twitch endotracheal tube pressure to transdiaphragmatic pressure is too variable to reliably represent a less invasive measure of diaphragm strength.

Administering granulocyte colony-stimulating factor to acute liver failure patients corrects neutrophil defects

OBJECTIVES

Neutrophil function is defective in acute liver failure (ALF) and the in vitro ability of granulocyte colony-stimulating factor (G-CSF) to reverse these defects has been reported. The effects of administering G-CSF to ALF patients are presented in this study.

DESIGN

This was a prospective, phase I/II, open label, study.

SETTING

The liver intensive therapy unit at King's College Hospital, London.

PARTICIPANTS

Sequential patients admitted with acute liver failure due to acetaminophen overdose.

INTERVENTIONS

G-CSF was given to four groups (each n = 6) of ALF patients as a daily infusion at 25, 50, 100 or 150 microg/m2. A control group of eight patients did not receive G-CSF.

MAIN OUTCOME MEASURES

Neutrophil phagocytosis and killing of Staphylococcus aureus and superoxide release before G-CSF administration and at 24 and 96 h thereafter.

RESULTS

Neutrophils from patients receiving 50, 100 or 150 microg/m2 G-CSF, but not from control patients or those receiving 25 microg/m2, showed significantly increased phagocytosis and killing at 96 h. Doses of 50 or 150 microg/m2 G-CSF resulted in increased superoxide production at 96 h. No patients discontinued treatment as a consequence of side effects related to G-CSF administration.

CONCLUSIONS

G-CSF administration is a safe and effective means of reversing the neutrophil defects of ALF, and may have a role in the prevention and treatment of infection in these patients. A dose of 50 microg/m2/day is as effective as higher doses and was associated with fewer side effects.

In vitro pentamer formation as a biomarker of tacrolimus-related immunosuppressive activity after liver transplantation

Therapeutic drug monitoring of tacrolimus (FK) is widely performed to assist adjustments of drug dosage but may be an inadequate surrogate of the immunosuppression induced. The aim of this investigation was to develop an alternative method for measuring FK-related immunosuppressive activity in blood samples from liver transplant recipients. A pentamer formation assay (PFA) was devised based on the attachment of the 12 kDa FK-binding protein (FKBP12) to microtitre plates in the presence of calcineurin, calmodulin, Ca++ and FK. Pentamer formation could be detected at FK concentrations > or = 0.2 microg/l by optimising assay conditions, particularly by including Ca++ (0.5 mM) only during the formation of the pentameric complex. Three methods (blood lysis, proteolytic digestion and use of commercial solutions used in a microparticle enzyme immunoassay (MEIA) technique) were incompatible with PFA measurements after extracting immunosuppressive FK-related material from patients' blood samples. However, therapeutic amounts of FK-related material could be quantified by the PFA assay after extraction of blood samples with methanol. There was a moderate correlation (r = 0.689) of FK equivalents assayed by PFA with results using MEIA in 56 blood samples from 14 liver graft recipients, but no obvious relationship of results to variables reflecting their clinical status.

The systemic inflammatory response syndrome in acute liver failure

The systemic inflammatory response syndrome (SIRS) in acute liver failure (ALF), in which infection is common, has not been studied. In this study, SIRS components were recorded on admission and during episodes of infection, in 887 ALF patients admitted to a single center during an 11-year period. Overall, 504 (56.8%) patients manifested a SIRS during their illness, with a maximum of 1, 2, and 3 concurrent SIRS components in 166, 238, and 100 patients, respectively. In 353 (39.8%) patients who did not become infected, a SIRS on admission was associated with a more critical illness, subsequent worsening of encephalopathy, and death. Infected patients more often developed a SIRS and one of greater magnitude. The magnitude of the SIRS in 273 patients with bacterial infection correlated with mortality, being 16.7%, 28.4%, 41.2%, and 64.7% in patients with 0, 1, 2, and 3 maximum concurrent SIRS components, respectively. Similar correlations with mortality were seen for SIRS associated with fungal infection, bacteremia, and bacterial chest infection. Fifty-nine percent of patients with severe sepsis died, as did 98% of those with septic shock. A significant association was found between progressive encephalopathy and infection. Infected patients with progressive encephalopathy manifested more SIRS components than other infected patients. For patients with a SIRS, the proportions of infected and noninfected patients manifesting worsening encephalopathy were similar. In ALF, the SIRS, whether or not precipitated by infection, appears to be implicated in the progression of encephalopathy, reducing the chances of transplantation and conferring a poorer prognosis.

Acute liver failure

Worldwide, viral infection is responsible for the majority of cases of acute liver failure, and the presence of co-existing chronic viral hepatitis may increase its severity. The newly described hepatotrophic viruses, hepatitis G virus and transfusion-transmitted virus, are unlikely to be major aetiological agents. In the USA and western Europe drug-induced hepatotoxicity is the most common cause, and most frequently results from acetaminophen. Hepatotoxicity caused by Ecstasy is increasingly important, particularly in young adults. Hepatic encephalopathy and cerebral oedema remain important and life-threatening complications, and their pathogenesis is not completely understood. The effects of the cerebral metabolism of the high levels of ammonia that circulate in hepatic failure appear to be important. Induced hypothermia is a promising modality of treatment for refractory cerebral oedema, but the only form of treatment known to improve survival is emergency liver transplantation. Living donor and auxiliary liver transplantation are likely to improve survival rates further and reduce the number of patients requiring long-term post-transplant immunosuppression.