Yasumi Katsuta

Intrapulmonary vascular dilatation and nitric oxide in hypoxemic rats with chronic bile duct ligation.

BACKGROUND/AIMS Nitric oxide (NO) has been suggested as the major cause of pulmonary vascular dilatation and hypoxemia in hepatopulmonary syndrome (HPS). The aim of this study was to assess the effect of NO on arterial oxygenation in rats with common bile duct ligation (CBDL rats), a model of HPS. METHODS Arterial blood gases were measured in 44 CBDL rats and 44 Sham rats under unrestrained conditions. Intrapulmonary shunting was assessed with (141)Ce-labeled microspheres (15-mum diameter) and serum nitrate/nitrite levels were measured by HPLC. The effect of NOS inhibition on A-aDO(2) was studied using L-NAME. RESULTS A decrease of PaO(2) below 82.7 mmHg (the mean value-2sigma in Sham rats) was seen in 43% of CBDL rats. Intrapulmonary shunting was greater in CBDL rats than in Sham rats (P<0.001). A correlation between the extent of shunting and A-aDO(2) was found in all animals studied (r=0.89, P<0.001, n=16). Serum levels of nitrate/nitrite increased significantly across the lungs, and the increase was significantly correlated with A-aDO(2) in the total population of animals studied. Administration of L-NAME to CBDL rats achieved a significant improvement of A-aDO(2). CONCLUSIONS These results suggest that pulmonary vascular dilatation due to NO leads to hypoxemia in CBDL rats.

Pulmonary blood transit time and impaired arterial oxygenation in patients with chronic liver disease.

BackgroundContrast-enhanced echocardiography (CEE) using agitated saline can detect intrapulmonary vasodilatation (IPVD) in patients with hepatopulmonary syndrome (HPS). We estimated the pulmonary transit time of erythrocytes (PTT) by CEE, using microbubbles, and studied its relationship to arterial oxygenation in chronic liver disease.MethodsSixteen patients with chronic liver disease and seven healthy subjects were studied. PTT was defined as the time between opacification of the right atrium and left atrium on CEE, using human serum albumin-air microbubble complexes with a mean diameter of 4 µm (Albunex). IPVD was detected by CEE with agitated saline. Arterial blood gases were analyzed with patients in the supine position, and while they were seated. Cardiac output (CO) was determined by Doppler echocardiography.ResultsThe mean PTT value for all of the patients was 4.0 ± 1.4 s. One of the 3 patients who showed IPVD was normoxemic. Mild orthodeoxia was observed in the patients with abnormal alveolar-arterial oxygen difference (A-aDO2) values (>15 mmHg), but not in those with normal A-aDO2 values, or in the healthy subjects. PTT was correlated with PaO2 (r = 0.52; P < 0.05; n = 16) and A-aDO2 (r = −0.54; P < 0.05; n = 16) in the seated position. CO was significantly correlated with PTT (r = −0.62; P < 0.05; n = 15), but not with PaO2 and A-aDO2, in both positions.ConclusionsPTT may be a useful parameter for evaluating arterial oxygenation in patients with chronic liver disease with early HPS.

Long-term haemodynamic effects of a 4-week regimen of nipradilol, a new β-blocker with nitrovasodilating properties, in patients with portal hypertension due to cirrhosis: A comparative study with propranolol

To study the long-term effects of pharmacological combination therapy, a comparison was made of the haemodynamic changes in patients with cirrhosis and portal hypertension following a 4-week treatment of propranolol or nipradilol, a new nonselective beta-blocker with nitrovasodilating effect. Nipradilol (12 mg/dag, n = 12) significantly diminished wedged hepatic venous pressure (WHVP, 25 +/- 16%), the hepatic venous pressure gradient (HVPG, 20 +/- 12%), and estimated hepatic blood flow (EHBF, 18 +/- 16%). Propranolol (30 mg/day, n = 11) also caused a significant reduction in WHVP (22 +/- 21%) and HVPG (24 +/- 21%), but not in EHBF. The percentage of portal pressure reduction and the frequency of nonresponders did not differ between the nipradilol and propranolol groups. Both agents reduced heart rate by approx. 20%. Nipradilol, however, did not cause a significant reduction in cardiac index (CI) versus a 14% reduction by propranolol. Pulmonary capillary wedge pressure and central venous pressure, an index of preload, were decreased slightly in the nipradilol group. When nonresponders were excluded, there was a significant correlation of the percentage of reduction between WHVP and CI or systemic vascular resistance, in the nipradilol group. These results indicate that nipradilol may have potent hypotensive effects on portal hypertension, similar but not superior to propranolol. Nipradilol, at the dosage used in the present study, did not appear to exert a nitrovasodilating effect to enhance the portal pressure reduction induced by beta-blocking action.

Hepatic chymase level in chronic hepatitis: co-localization of chymase with fibrosis

Chymase, secreted by mast cells, is associated with angiotensin II production and fibrosis of myocardium and renal interstitium. Assuming that chymase also is involved in liver fibrosis, we previously established an enzyme-linked immunosorbent assay for chymase in human liver tissue. In the present study, we explored the localization of mast cells in the liver using antibodies against human chymase and also studied relationships between hepatic chymase level and histologic findings in 49 patients with chronic hepatitis. By the international classification, fibrosis was staged from F0 to F4 and activity was graded from A1 to A3. Cells immunoreactive for chymase were seen throughout portal areas and intralobular sinusoidal walls, largely colocalizing with inflammation and fibrosis. Hepatic chymase levels in F3 and F4 cases were greater than in F1 and F2 (F3+F4, 30.8+/-41.2 ng/mg vs. F1+F2, 5.7+/-6.6 ng/mg; P<0.01). Chymase in A3 casese (39.4+/-50.8 ng/mg) was more abundant than in A1 (3.7+/-4.3 ng/mg) or A2 (12.8+/-19.4 ng/mg); P<0.05 for each. Our findings suggest that hepatic chymase level is implicated in fibrosis and activity in human liver disease.

Plasma volume contraction in portal hypertension

The effect of blood volume contraction induced by a 4-week regimen of spironolactone (100 mg/day) or furosemide (40 mg/day) on the hepatic venous pressure gradient (HVPG), an indicator of portal hypertension, was evaluated in patients with cirrhosis and no ascites. In the spironolactone group (n = 15), HVPG decreased significantly from 16.5 +/- 0.9 mmHg (mean +/- S.E.M.) to 12.9 +/- 1.0 mmHg (P < 0.005) and total blood volume contracted significantly from 4338 +/- 231 ml to 4006 +/- 203 ml (n = 14, P < 0.01). Although the HVPG changes did not correlate significantly with changes in measured total blood volume or in simultaneously determined systemic haemodynamics, a significant inverse correlation (r = -0.74, P < 0.01, n = 12) was found between the HVPG change and posttreatment plasma aldosterone levels, attesting to the effectiveness of spironolactone therapy in lowering HVPG. In the furosemide group (n = 10), neither HVPG (13.7 +/- 0.3 mmHg vs. 13.6 +/- 0.9 mmHg) nor total blood volume (4961 +/- 153 ml vs. 4964 +/- 162 ml) declined significantly. These results show that long-term administration of spironolactone to patients with cirrhosis and no ascites produced a significant reduction in HVPG that may have been due to gradual, sustained volume contraction. Thus, spironolactone may prove to be an effective treatment for portal hypertension in cirrhosis without ascites.

Regional Differences in Peripheral Circulation between Upper and Lower Extremity in Patients with Cirrhosis

In 42 patients with compensated cirrhosis and 31 control subjects, blood flow (BF) and vascular resistance (VR) were measured at the forearm and calf, using a pneumoplethysmograph. In some of the subjects deep-body temperature (DBT) was also measured by the zero heat flow method. In cirrhosis, BF and DBT were significantly higher and VR was significantly lower in the forearm than in the calf. Corresponding differences were not observed in control subjects. When these indices of the forearm were compared between cirrhosis and controls, BF and DBT were significantly higher and VR was significantly lower in cirrhosis than in controls. In cirrhotics in whom the gradient between forearm BF and calf BF was 1 ml.dl-1.min-1 or more (forearm greater than calf), the vascular response of the forearm to cold stimulation was reduced, whereas in the remaining patients and in controls the forearm BF and VR responded significantly. These results suggest that there is a regional difference in peripheral circulation in cirrhotics, partly with participation of impaired sympathetic nervous activity, which may account for the selective distribution observed in the clinical manifestations of vascular spider, palmar erythema, and warm hand, inclined toward the upper extremities or the upper part of the body.

Arterial hypoxemia and intrapulmonary vasodilatation in rat models of portal hypertension.

BackgroundRats with chronic bile duct ligation (CBDL) and portal vein ligation (PVL) are used as models of portal hypertension. CBDL rats show hypoxemia with intrapulmonary vasodilatation (IPVD), and are recognized as a model of hepatopulmonary syndrome (HPS), while PVL rats are normoxemic. We investigated the differences in arterial oxygenation between these models, and the key factors leading to HPS.MethodsForty-eight Sprague-Dawley rats were prepared as CBDL or PVL models, or as Sham rats. Arterial oxygenation, hemodynamics (reference sample method), and IPVD were simultaneously evaluated in conscious and unrestrained animals, using 141Ce- or 113Sn-labeled microspheres (15 µm in diameter), respectively. Endothelin-1 (ET-1) and nitrate/nitrite (end products of nitric oxide; NOx) production by the lung tissue (increment across the lungs) was also determined.ResultsThe extent of IPVD was similar in both models, but hypoxemia was only observed in CBDL rats. The ET-1 level and the increment in NOx were significantly increased in CBDL rats, and the increment was directly correlated with impairment of oxygenation. Blood flow through the bronchial arteries (anatomical shunting) was increased in CBDL rats, reaching more than three times the level in PVL rats or Sham rats.ConclusionsThese results support the hypothesis that NO derived from the lung tissues contributes to hypoxemia, and IPVD appears to be a prerequisite for impaired oxygenation. The considerable increase of anatomical shunting may potentially contribute to impaired oxygenation in CBDL rats.

Establishment of an assay method for human mast cell chymase

Chymase secreted by mast cells found in fibroblast-containing interstitial connective tissue has been implicated in collagen fiber formation and extracellular matrix production. We established a method for determination of human chymase activity, and applied this technique to measurements in serum and liver tissue. The mean chymase concentration in liver biopsy specimens from 26 patients with chronic hepatitis was 5.23+/-5.98 ng/mg (ranges 0.32-21.4). The serum chymase concentration was below the limit of detection, in both chronic hepatitis patients and healthy individuals. No significant relationship was seen between chymase activity in liver tissue and severity of liver fibrosis, but further investigation in larger numbers of patients is warranted.

Severe hepatitis during ketoconazole therapy

SummaryKetoconazole is an imidazole derivative recently developed as an antifungal agent. There have been only a few established cases of hepatotoxicity in the literature. This report describes a case of presumed ketoconazole hepatotoxicity characterized by the development of severe hepatitis with marked centrilobular necrosis. The lack of hypersensitivity reactions in clinical findings and centrilobular location of necrosis suggested a host idiosyncrasy with metabolic abnormality as the effect of ketoconazole in the present case.

Multiple coronary arteriovenous fistulas associated with polymyositis.

We report a case of a 60-year-old female presenting a histologically proven polymyositis with multiple coronary arteriocameral fistulas draining into the left ventricle. The exact pathogenesis of coronary artery fistulas and their relation to polymyositis are not clear.