Paolo Buonamico’s research while affiliated with University of Bari Aldo Moro and other places

Background: Liver involvement is a common manifestation of hereditary haemorrhagic telangiectasia (HHT). Although a number of studies have been carried out in adult patients, no study has ever been focused on investigating HHT-related hepatic involvement in paediatric patients. The present study aimed for the first time to systematically estimate the prevalence of HHT-associated liver involvement and to characterize HHT-associated hepatic angiodynamic features in paediatric age. Patients and methods: The study was designed as a cross-sectional survey in an HHT paediatric cohort, subclassified as HHT1 and HHT2 according to the mutated gene. The evaluation of the angiodynamic profile was performed by duplex ultrasound examination. Investigation by multi-slice computed tomography (MSCT) or magnetic resonance angiography (MRA) was performed in patients >12 years. Results: MSCT/MRA examination disclosed silent hepatic involvement in 7/20 (35.0 %) children, and nodular regenerative hyperplasia in two cases. Diameter of common hepatic artery was significantly larger in HHT2 (0.45 ± 0.15 cm) compared to HHT1 (0.33 ± 0.09, p < 0.01) and control children (0.32 ± 0.08, p < 0.05). None of the patients had clinical manifestations of liver involvement. Angiodynamic profiles were different between paediatric and adult HHT patients. Conclusions: Liver involvement can be detected in paediatric HHT patients, albeit with a lower frequency compared to adults. Paediatric HHT2 children show a higher frequency of liver involvement and a trend to hepatic artery dilation when compared to HHT1 children.

Hepatic arterio-venous malformations (HAVMs) have been found in 74% of hereditary hemorrhagic telangiectasia (HHT) patients with multislice CT (MSCT). This single-blind study aimed to compare the diagnostic accuracy of echo-color-Doppler with MSCT and identify the most sensitive ultrasound criteria indicating hepatic shunts. One hundred and fifty-three HHT patients were systematically screened for HAVMs by biological tests, abdominal MSCT and echo-color-Doppler. Twenty-five normal subjects and 15 cirrhotic patients were also included as control groups. Both intrahepatic ("color spots" and hypervascularization) and extrahepatic parameters (diameter, flow velocity and tortuosity of hepatic artery and diameter and flow velocity of portal/hepatic vein) were utilized. "Color-spots" are defined as subcapsular vascular spots with a high-velocity arterial blood flow and low resistivity index and can identify extremely small HAVMs. CT was positive in 128/153 (84%) patients and Doppler color spots were found in 131/153 (86%) patients. The sensitivity, specificity and diagnostic accuracy of "color spots" compared to MSCT were 95.3%, 68.0% and 91.8%, respectively. The "color-spot" showed a greater correlation to CT (V(index)=0.655; p<0.0001) than extrahepatic criteria (V=0.317). In 20/29 (69%) subjects, echo-color-Doppler, confirmed by CT, identified the third criterion for definite HHT diagnosis. Intrahepatic criteria was superior to extrahepatic criteria for identification of HAVMs. A new Doppler parameter ("color-spots") with an optimal accuracy for detecting HAVMs is proposed for easy periodic screening of HHT patients.

Hereditary hemorrhagic telangiectasia (HHT), also known as Rendu-Osler-Weber disease, is an autosomal-dominant vascular disease characterized by mucocutaneous or visceral angiodysplastic lesions (telangiectases and arteriovenous malformations) that may be widely distributed throughout the cardiovascular system. The recognition of mucocutaneous telangiectases, the occurrence of spontaneous and recurrent episodes of epistaxis, the presence of visceral involvement, and a family history of this disease are the clinical criteria that allow diagnosis. In comparison with skin, lungs, gastrointestinal tract, and brain involvement, hepatic involvement defined by clinical criteria alone has long been considered uncommon. Our experience with a large group of HHT patients, even those asymptomatic for liver involvement, demonstrates that it is more frequent than reported and is characterized by the presence of intrahepatic shunts, disseminated intraparenchymal telangiectases, and other vascular lesions. Congestive cardiac failure, portal hypertension, portosystemic encephalopathy, cholangitis, and atypical cirrhosis have been reported as possible serious complications related to this condition. Thus, a correct diagnosis is important, and diagnostic imaging has a fundamental role in detecting alterations involving the liver. The possibilities to perform a multiphasic study and to provide high-quality multiplanar and angiographic reconstructions, gives multidetector row helical computed tomography the ability to detect and characterize the complex anatomopathologic alterations typical of this disease.

This double-blind study was designed to evaluate the haemodynamic effect of two drugs, propranolol and octreotide, and their combination in patients with cirrhosis. Fifteen patients with cirrhosis were randomly assigned to two groups receiving either octreotide subcutaneously at 100 microg ('octreotide' group, n = 9) or propranolol orally at 40 mg followed by a subcutaneous dose of octreotide (100 microg) after 1 h ('propranolol + octreotide' group, n = 6); then, after 30 min, a standard meal was administered to both groups. The hepatic vein pressure gradient by hepatic vein catheterization, portal and superior mesenteric artery blood flow velocity, superior mesenteric artery pulsatility index by the echo-Doppler duplex system were recorded at baseline, 1 h after propranolol in the 'propranolol + octreotide' group, and in both groups 30 min after octreotide and 30 min after meal. At fast, propranolol was more active in decreasing portal pressure (from 16 +/- 2.2 to 12.7 +/- 3.8 mmHg, -20%, P < 0.05) as compared to octreotide (from 18.6 +/- 4.8 to 16.6 +/- 4.3 mmHg, -11%, P < 0.05). Conversely, octreotide was more active on the mean blood flow velocity of superior mesenteric artery (from 22.8 +/- 5 to 19 +/- 4.5 cm/ s, -17%; P< 0.05). Octreotide administration in patients receiving beta-blockers showed, also, a trend to increase the mesenteric vascular resistances (pulsatility index from 3.14 +/- 0.69 to 3.68 +/- 1.29, +17%, not significant (NS)) which had not been affected by previous treatment with propranolol. After the meal, a reduction of the expected hyperaemic response occurred in both groups. The combined acute haemodynamic effect of this association suggests the possible combination of these two drugs in critical situations, such as variceal bleeding in patients receiving beta-blockers. The simultaneous use of echo-Doppler and hepatic vein catheterization permitted us a more complete analysis of the acute haemodynamic events.

To assess the long-term effect of the addition of long-acting nitrates to beta-blockers on liver blood flow and liver metabolic activity in patients with cirrhosis and portal hypertension. Eleven patients with cirrhosis and portal hypertension were investigated by using hepatic vein catheterization and indocyanine green (ICG) constant infusion on baseline conditions, after 1 month of treatment with nadolol, after 3 months of treatment with nadolol plus isosorbide mononitrate, and (in seven cases) after 1 year of combined treatment. The hepatic venous pressure gradient decreased significantly after nadolol, and more so after addition of isosorbide mononitrate. Hepatic blood flow, and ICG intrinsic hepatic clearance did not change significantly, although few cases showed an increase or decrease in either parameter. A significant correlation was found between changes in ICG intrinsic hepatic clearance and in hepatic venous pressure gradient (r = 0.62, P = 0.04). Liver blood flow and liver metabolic activity are not consistently affected by addition of isosorbide mononitrate to nadolol. Substantial decreases in portal pressure may be associated with a decrease in ICG intrinsic hepatic clearance.

Background/aims: Doppler arterial resistance indices are used to evaluate alterations in arterial hemodynamics in the liver, spleen, and kidney. The purpose of this study was to determine the interobserver and interequipment variability of hepatic, splenic, and renal arterial Doppler resistance indices, and the influence of a cooperative training program of the operators on the reproducibility of the results. Methods: In the first part of the study, hepatic (PI-L, RI-L), splenic (PI-S, RI-S), and renal (PI-K, RI-K) pulsatility and resistive indices were measured by echo-color-Doppler in eight control subjects and ten patients with cirrhosis by three operators using three different machines. In the second part of the study, measurements were taken by the three operators in nine controls and nine patients with cirrhosis, after cooperative training, with a single machine. Results: Significant interobserver variability was present for all parameters except RI-L. Significant interequipment variability was present for all parameters except PI-S and RI-S. Only 0-3% of variance was equipment- or operator-related, while 58-72% was patient-related. Hepatic and renal coefficients of variation were similar in patients with cirrhosis and controls, while splenic coefficients of variation were higher in patients with cirrhosis than in controls. After training, differences among operators disappeared for all variables except RI-K, and the operator-related component of variance nearly disappeared for all parameters. Conclusions: Hepatic, splenic, and renal arterial resistance indices show small but significant interobserver and interequipment variability. Interobserver variability can be decreased to non-significant levels by a common training program. Thus, these indices can be widely applied to the study of arterial circulation in these organs.

This study aimed to evaluate the protein and lipid redox status in plasma erythrocytes and erythrocyte ghosts of alcoholics and of patients with non-alcoholic liver disease; we also investigated the relation to glutathione levels and the role of acetaldehyde and xanthine oxidase activity in plasma. Carbonyl and sulfhydryl proteins, glutathione and malondialdehyde levels and the activity of the circulating xanthine oxidase were determined in: active and abstinent alcoholics, patients with chronic viral hepatitis and healthy controls. Active alcoholics showed a decrease of sulfhydryl protein and glutathione concentrations in plasma, erythrocytes and ghosts compared to the other groups. Also, an increase of the carbonyl protein and malondialdehyde levels and of the activity of circulating xanthine oxidase (9.2 +/- 1.8 nmol.min.ml, p < 0.001) were observed. Significant correlations between carbonyl protein and malondialdehyde concentrations in plasma (r = 0.775, p < 0.001), as well as between daily alcohol intake and carbonyl protein content in plasma (r = 0.879, p < 0.001) and erythrocytes (r = 0.605, p < 0.01) were observed. However, carbonyl protein levels did not correlate with the degree of liver injury. Incubation of plasma with acetaldehyde, but not with ethanol, significantly increased the carbonyl protein formation. Administration of N-Ethylmaleimide, a thiol depletor, or glutathione significantly increased or delayed, respectively, the carbonyl protein formation. Proteins are oxidatively modified in plasma and erythrocytes of active alcoholics, whereas no such alterations are detectable in patients with non-alcoholic liver disease. Protein oxidation in alcoholics does not seem to result directly from ethanol; circulating xanthine oxidase, delivered from injured cells, may play a contributory role and glutathione appears to be directly involved in the protection of plasma proteins against acetaldehyde toxicity.