Large spontaneous portal-systemic shunts have been occasionally described in patients with cirrhosis. This study was undertaken to assess the prevalence of portal-systemic shunts in patients with cirrhosis with recurrent or persistent hepatic encephalopathy (HE) as compared with patients with cirrhosis without HE. Fourteen patients with cirrhosis with recurrent or persistent HE (cases) and 14 patients with cirrhosis without previous or present signs of overt HE matching for age and degree of liver failure (controls) were studied. Each patient underwent neurological assessment and cerebral magnetic resonance (MR) imaging to exclude organic neurological pathological conditions. HE evaluation included psychometric performance (Trail-Making Test A), electroencephalogram (EEG), mental status examination and grading, arterial, venous, and partial pressure of ammonia determination. The presence of portal-systemic shunts was assessed by portal venous phase multidetector-row spiral computed tomography (CT). Large spontaneous portal-systemic shunts were detected in 10 patients with HE and in only 2 patients without HE (71% vs. 14%; chi square ‫؍‬ 9.16; df ‫؍‬ 1.0; P ‫؍‬ .002). The patients with HE presented ascites (P ‫؍‬ .002) and medium/large esophageal varices (P ‫؍‬ .02) less frequently than the control group. In conclusion, our study suggests that large spontaneous shunts may often sustain the chronicity of HE; the presence of large shunts should be sought in patients with cirrhosis with recurrent or persistent HE.

No abstract

Background/Aims: The effects of acute insulin deficiency on the kidney have been investigated in animal models of experimental diabetes; however, the impact of long-term diabetes has not been determined. Methods: We measured renal glycogen contents in streptozotocin (STZ)-diabetic rats 3 weeks (n = 12) or 9 months (n = 12) after the induction of diabetes, and in 2 groups of control rats of similar age (n = 16 and n = 12, respectively), in the fed state and after a 24-hour fast. Results: Diabetic rats had high glucose levels, low insulin but normal glucagon concentrations in portal blood. In the fasting state, kidney glycogen content was very low in both young control and young diabetic rats (54 ± 15 and 189 ± 26 µg/g, respectively, mean ± SD); in contrast, glycogen levels were markedly elevated in rats with long-standing diabetes as compared to old nondiabetic animals (2,628 ± 1,023 ± and 1,968 ± 989 µg/g of diabetic rat, fasting and fed, respectively, p < 0.001 vs. 0 ± 0 and 4 ± 6 µg/g of control rats). On electron microscopy, large glycogen clusters were localized to the renal tubules. Kidney phosphorylase activity was higher, and synthase activity lower in diabetic than control rats (p < 0.05 for both), whereas kidney glycogen was strongly related to plasma glucose levels, suggesting that the enzyme changes were secondary to glycogen accumulation itself. Renal hexosephosphates and fructose-2,6-bisphosphate contents were both increased in long-term diabetic rats (p < 0.05), implying enhanced fluxes through both glycolysis and gluconeogenesis. Conclusion: In chronic, untreated diabetes glycogen accumulates in the renal tubules; prolonged hyperglycemia is the sole driving force for this phenomenon.

Increased plasma atrial natriuretic peptide (ANP) levels and impaired ANP action have been reported in patients with diabetes or insulin resistance. The aim of this study was to assess the interaction between insulin and ANP in type 2 diabetes. In 12 normotensive, normoalbuminuric type 2 diabetics, we infused insulin at a high (6.6 pmol/min/kg) or, on a different day, at a low rate (0.6 pmol/min/kg) during 4 hours of isoglycemia under isovolumic, isoosmolar conditions. The normal response was established in 12 healthy volunteers using an identical protocol. Despite higher baseline ANP levels (17.7 +/- 2.8 vs. 10.8 +/- 1.8 pg/ml, p = 0.04), urinary sodium excretion was similar in diabetics and controls (113 +/- 8.5 vs. 102 +/- 8.8 mEq/24 hours, p = ns). In both groups, hyperinsulinemia caused a decrease in blood volume (0.33 +/- 0.10 l, p < 0.01), diastolic blood pressure (6 %, p < 0.02), and natriuresis. However, plasma ANP decreased in controls (from 12.7 +/- 1.9 to 8.6 +/- 1.4 pg/ml, p = 0.01) but not in type 2 diabetics (15.1 +/- 2.7 vs. 17.2 +/- 3.8 pg/ml, p = ns). We conclude that ANP release is resistant to volume stimulation in type 2 diabetic patients, and natriuresis is resistant to ANP action. This dual disruption of ANP control may play a role in blood pressure regulation in diabetes.