Sinusoidal Caliber in Alcoholic and Nonalcoholic Liver Disease: Diagnostic and Pathogenic Implications

Open Access
Publisher Website
Google Scholar
Abstract
Portal hypertension in alcoholic liver disease has been attributed to an increased resistance to blood flow either of sinusoidal or of postsinusoidal origin. The former should be accompanied by sinusoidal compression while the latter is expected to result in an increased or a normal sinusoidal diameter. Patients with alcoholic liver disease showed a marked reduction (p < 0.001) in relative sinusoidal area (995 ± 135 μm2; n = 19) when compared to nonalcoholic patients with normal liver histology (5, 100 ± 389 μm2; n = 6), or to patients with nonalcoholic liver disease (6, 242 ± 467 μm2; n = 19). Hepatocyte surface area was significantly increased in patients with alcoholic liver disease when compared to hepatocytes from normal biopsies (563 ± 32 μm2 vs. 301 ± 26 μm2; p < 0.001). Patients with nonalcoholic liver disease had hepatocyte surface areas within the normal range (327 ± 14 μm2). There was a significant inverse correlation between hepatocyte size and sinusoidal area (r = -0.63; p < 10-6; n = 44), indicating that larger hepatocytes were associated with sinusoidal compression. In the alcoholic patients, portal pressure correlated inversely (r = -0.77; p < 0.01) with sinusoidal area only after the sinusoidal area was markedly reduced to areas below 20% of normal. Such a threshold was not reached in patients with nonalcoholic liver disease, in whom no correlation between sinusoidal area and portal pressure was observed. Rats fed chronically with a diet containing 35% of calories as ethanol, in which liver enlargements of 36 to 42% were observed relative to controls fed an isocaloric carbohydrate diet, had a significant reduction in both extracellular space and blood space per unit liver weight. Data presented support the hypothesis that hepatocyte expansion and compression of the sinusoidal space appear to be important determinants in the development of portal hypertension in alcoholic liver disease. In addition, the striking difference in the observable sinusoids in alcoholic and nonalcoholic liver disease should provide an added criterion in the histological differentiation of the two conditions.