Relation between liver pathology and prognosis in patients with portal hypertension

Esophageal variceal hemorrhage presenting as sudden death in outpatients

CONTEXT

Some autopsy studies have dealt with histologic features of esophageal varices after different therapeutic procedures. However, to the best of our knowledge, no reports have been published describing outpatient characteristics that are associated with fatal esophageal variceal hemorrhage in a medicolegal autopsy population.

OBJECTIVES

To (1) assess the incidence of sudden deaths from esophageal variceal hemorrhage in an unselected medicolegal autopsy population and (2) determine demographics of outpatients dying from esophageal variceal hemorrhage with special reference to blood alcohol concentrations at the time of death.

DESIGN

We performed a retrospective study of all autopsy cases of sudden death from esophageal variceal hemorrhage from a total of 6038 medicolegal autopsies performed over a 5-year period (1997-2001). We analyzed individual cases to determine gender, age, location and histology of bleeding esophageal varices, pathogenic mechanism for esophageal varices, concomitant underlying diseases contributing to fatal outcome, body mass index, circumstances at the death scene, and blood alcohol levels at the time of death. We reviewed the results of toxicologic analyses of alcohol concentrations in samples of femoral venous blood and urine obtained at autopsy; concentrations had been determined by gas chromatography with mass spectroscopy and enzymatic assays.

RESULTS

We identified 45 cases of fatal esophageal variceal hemorrhage that occurred out of hospital and presented as sudden death; the corresponding 5-year incidence in this autopsy population was 0.75%. All of the deceased were white; the male-female ratio was 1.6:1, and the mean age was 50.6 years. Ruptured esophageal varices were located in the lower third of the esophagus in 44 cases. Cirrhosis of the liver was present in all cases (alcoholic cirrhosis of the liver in 42 cases), and a hepatocellular carcinoma was present in 3 cases. Alcohol-induced pancreatic tissue alterations were frequently found. The results of toxicologic analysis were positive for alcohol in femoral venous blood and urine in 30 cases. Blood alcohol levels at the time of death were less than 100 mg/dL (21.7 mmol/L) in 15 cases, between 100 and 200 mg/dL (21.7 and 43.4 mmol/L) in 8 cases, and greater than 200 mg/dL (43.4 mmol/L) in the remaining 7 cases.

CONCLUSIONS

Apart from abnormalities in coagulation due to poor liver function in long-term alcohol users, acute alcohol intake may represent an important factor influencing mortality in individuals with esophageal variceal hemorrhage. Acute alcohol intake has transient effects on blood clotting time caused by ethanol and its main metabolites. In the present study, bloodstains at the death scene and unusual body positions of the deceased that aroused suspicion of a violent death were leading reasons for conducting a medicolegal autopsy. Apart from aspects of forensic pathology, the demographics of our study population are also noteworthy from the viewpoint of social medicine. The data we present stress the importance of fatal esophageal variceal hemorrhage as a relevant cause of sudden death occurring outside the hospital in socially isolated, alcohol-addicted individuals.

Optimizing emergency care of upper gastrointestinal bleeding in cirrhotic patients

The type of emergency treatment administered to patients with suspected variceal bleeding is important, as the episode is associated with a high mortality rate. Moreover, rebleeding is common during the first few days after the initial haemorrhage. Several techniques are available to control variceal haemorrhage including pharmacotherapy (vasopressin, terlipressin, somatostatin and octreotide), balloon tamponade, endoscopic techniques, transjugular intrahepatic portosystemic shunt and shunt surgery. The majority of these require specialized equipment and/or experienced personnel, which are not always available in every hospital. In such situations, pharmacotherapy represents the most practical method of establishing haemodynamic control prior to the administration of definitive treatment. Pharmacotherapy can be initiated immediately upon admission to stabilize the patient prior to diagnostic endoscopy, which subsequently improves the efficacy and ease of administration of further endoscopic intervention. The optimal pharmacological agent should be both effective and safe. A drug with no side effects will not complicate the management of critical patients and can be administered over an extended period to reduce the incidence of rebleeding and improve prognosis. Meta-analysis of clinical studies has revealed that of the vasoactive drugs available somatostatin is effective with significantly fewer side effects and currently appears to represent the best choice for treatment. The available evidence suggests that the early administration of pharmacotherapy, as part of a specific treatment regimen, offers significant benefit to patients with variceal bleeding and its administration optimizes emergency care.

The role of surgery in the treatment of portal hypertension

The role of surgery in the treatment of portal hypertension continues to evolve. Pharmacologic and endoscopic therapies are the primary treatment modalities for the prophylaxis and treatment of variceal bleeding and ascites. Failure of these therapies is the indication for invasive intervention such as TIPS, surgical shunt, or devascularization. Distal splenoreal shunting provides selective variceal decompression with less encephalopathy and accelerated hepatic failure than portal decompression. Liver transplantation remains the treatment of choice for patients with poor hepatic function.