Keiichiro Yoneyama

Risk factors for infections in cirrhotic patients with and without hepatocellular carcinoma

Background: Patients with liver cirrhosis (LC) frequently have complications with bacterial infections, and these infections increase the mortality rate. However, a detailed analysis of infections associated with LC patients has not yet been performed. Methods: We analyzed 325 patients with LC with and without hepatocellular carcinoma (HCC) who were hospitalized between 1997 and 1999. Results: Infections developed in 70 (21.5%) patients and 48 (68.6%) of these developed infections during hospitalization. The mortality rate of 28.6% (20/70) in patients with infectious complications was higher than that of 12.5% (32/255) in patients without infectious complications. Forty (57.1%) of the 70 patients had infections of unknown causes; 11 (15.7%) had sepsis; 6 (8.6%) had intravenous hyperalimentation (IVH) infection; 3 (4.3%) each had spontaneous bacterial peritonitis (SBP), liver abscess, and cholecystitis; and 4 (5.7%) had other infections. Bacterial cultures of blood were prepared from 73 of the 325 patients (22.5%), and were positive in 22 of the 73 patients (30.1%). Of these 22 culture-positive patients, 11 had sepsis, 6 had IVH infection, 2 had liver abscess, 1 had cholecystitis, 1 had pneumonia, and 1 had decubitus ulcer. Gram-positive bacterial strains were detected most frequently, in 16 of the 24 strains isolated. Univariate analysis revealed significant differences between the groups with and without infectious complications with regard to hepatitis B virus infection, Child-Pugh classification, ascites, esophageal varices, survival rate, total-bilirubin (T-Bil), albumin (Alb), lactate dehydrogenase (LDH), total cholesterol (T-chol), and prothrombin time (PT). On multivariate analysis, the Alb level was selected as a significant independent factor contributing to the development of infections. Conclusions: Patients with advanced cirrhosis with low Alb levels should be carefully treated, and the administration of broadspectrum antibiotics covering gram-positive bacteria needs to be considered in the treatment of infections.

Prognostic Index of Cirrhotic Patients with Hepatic Encephalopathy with and Without Hepatocellular Carcinoma

This study was carried out in Japanese patients to clarify the state of liver cirrhosis complicated by hepatic encephalopathy with and without hepatocellular carcinoma and its prognosis. The subjects were 100 patients with liver cirrhosis complicated by hepatic encephalopathy. Clinical data were investigated, and prognostic factors were extracted using Coxs proportional hazard model. The cumulative survival rate after the first episode of hepatic encephalopathy was 59.1% after 1 year, 48.3% after 2 years, and 22.2% after 5 years. The prognostic index (PI) was calculated using the following formula consisting of these six factors. PI = 0.806 × Child–Pugh classification + 1.149 × presence or absence of HCC + 0.024 × BUN + 0.036 × LDH + 0.093 × WBC + 0.381 × PIVKA-II. The PI value was suggested to be useful for the prognosis of liver cirrhosis after the first episode of hepatic encephalopathy.

Efficacy and safety of prednisolone in patients with autoimmune hepatitis.

A retrospective cohort study involving 29 Japanese patients with autoimmune hepatitis (AIH) was performed to clarify factors that predict the efficacy of prednisolone and the occurrence of various serious adverse effects. Independent predictors were identified by logistic analysis and with use of the Cox proportional hazard model. Responses to prednisolone were noted in 28 patients, who were classified into the complete remission group (52%) or the relapse group (48%). Multivariate analysis identified alanine aminotransferase, alkaline phosphatase, and immoglobulin G levels as independent predictors of relapse. The adverse effects most frequently observed were diabetes mellitus (37.9%), psychiatric/ neurologic symptoms (34.5%), and circulatory symptoms (34.5%). Predictive factors included lactate dehydrogenase, albumin, and fasting blood glucose levels for diabetes mellitus, alkaline phosphatase and C-reactive protein for psychiatric/neurologic symptoms, and autoimmune hepatitis score and lactate dehydrogenase for circulatory symptoms. Selection of an optimal treatment method for individual patients may be possible after the risks of relapse and adverse effects have been estimated.

Analysis of Background Factors Influencing Long-Term Prognosis of Patients with Chronic Hepatitis C Treated with Interferon

Objective: Interferon (IFN) therapy has been used as antiviral therapy for chronic hepatitis C (CH-C); however, complete response to the therapy is observed in only about 30% of patients in Japan. Background factors involved in the responsiveness to IFN therapy, and progression to liver cirrhosis (LC) and hepatocellular carcinoma (HCC) after IFN therapy have not yet been sufficiently investigated. Methods: One hundred twenty-one patients with CH-C who received IFN therapy at Showa University Hospital between 1984 and 1999 were analyzed. Results: At 6 months after the termination of IFN therapy, 53 patients achieved a complete response, 11 patients incomplete response, and 57 patients no response. During a mean follow-up of 52.7 months, 12 patients progressed to LC, and 10 patients developed HCC. Multivariate analysis showed that significant independent factors involved in progression to LC were platelet count and the efficacy of IFN therapy. The significant independent factor involved in the development of HCC was platelet count. The factor involved in the therapeutic effect at 6 months after the termination of IFN administration was the serum hepatitis C virus (HCV) RNA levels before IFN therapy. Conclusion: Patients with high HCV RNA levels and low platelet counts should be considered to be at high risk of progressing to LC and developing of HCC and should be carefully followed after IFN therapy using ultrasonography, CT scan and MRI.

Prognostic index of liver cirrhosis with ascites with and without hepatocellular carcinoma

Background: In Japan, the incidence of liver cirrhosis caused by hepatitis viruses is higher, and cirrhosis is more likely to be complicated by hepatocellular carcinoma, than in Western countries. The aim of this study was to predict the outcome in liver cirrhosis with ascites with and without hepatocellular carcinoma. Methods: The subjects were 146 patients with liver cirrhosis complicated by ascites. Forty-six factors were evaluated concerning clinical laboratory parameters and extracted prognostic factors using the Cox proportional hazards model. Results: The mean duration of the follow-up period was 634.9 days, during which 89 (61%) of the patients died, 27 (18.5%) survived, and 30 (20.6%) were lost to follow-up. The cumulative survival rate after the onset of ascites was 59.7% after 1 year, 44.5% after 2 years, and 29.5% after 5 years. Multivariate analysis indicated 9 factors, i.e. age, total bilirubin (T-Bil), alkaline phosphatase (ALP), blood urea nitrogen (BUN), α-fetoprotein (AFP), mean arterial pressure (MAP), gastrointestinal bleeding, infection, and portal vein tumor thrombosis (PVTT), as independent prognostic factors. The prognostic index (PI) was calculated by the following formula using these 9 factors. PI = 0.045 × age + 0.180 × T-Bil + 0.088 × ALP + 0.020 × BUN + 0.467 × AFP + (−0.022 × MAP) + 0.662 × gastrointestinal bleeding + 0.521 × infections + 0.882 × PVTT. Conclusion: Prediction of the outcome using PI based on the 9 factors provides additional information for the determination of the therapeutic approach in cirrhotic patients with ascites with and without hepatocellular carcinoma.

Prognostic Index for Survival in Patients After Treatment for Primary Hepatocellular Carcinoma

We retrospectively evaluated clinical factors affecting long-term survival after treatment for hepatocellular carcinoma (HCC) to predict the reliability of the prognosis of patients with HCC. We analyzed 157 patients who received treatment for HCC. The prognostic index (PI) was evaluated using the Cox regression model. The overall cumulative survival rates were 79.7% at 1 year, 51.1% at 3 years, and 24.9% at 5 years. The PI was calculated by the following formula, consisting of five factors: PI=0.637×number of tumor lesions+0.103×tumor diameter+1.003×ascites+0.915×portal vein tumor thrombosis – 0.006×cholinesterase+2.0. It was found that liver function and progression of the tumor are the most important factors for prognosis after treatment for HCC. The PI, based on five factors, will in future be an appropriate index to predict the prognosis of patients with HCC.

Validation of confocal laser scanning microscopy for detecting intracellular calcium heterogeneity in liver slices

To investigate changes in the intracellular Ca(2+) ([Ca(2+)]i) in liver lobules under aerobic and hypoxic conditions, we measured [Ca(2+)]i in liver slices using a confocal laser scanning microscope (CLSM). The liver lobule is divided into 3 equal parts between the central vein and portal area, Zones 1, 2, and 3 from the portal side. [Ca(2+)]i in each zone of cultured rat liver lobules was measured by CLSM and a fluorescent Ca(2+) indicator (Rhod 2 AM). After the culture solution was changed to an Na(+)-free solution under aerobic conditions, the percentage of cells showing an increase in [Ca(2+)]i was 66.0+/-9.7% in Zone 1, 70.0+/-10.5% in Zone 2, and 94.0+/-9.7% in Zone 3. The percentage was significantly higher in Zone 3 than in Zones 1 and 2 (P< .01). Under hypoxic conditions, the percentage of cells showing an increase in [Ca(2+)]i was 6.0+/-9.7% in Zone 1, 8.0+/-10.3% in Zone 2, and 10.0+/-10.5% in Zone 3. There were no differences among the 3 zones. In all zones, the percentage was higher under aerobic conditions than under hypoxic conditions (P< .01). These results indicated that the increase in [Ca(2+)]i in liver lobules was heterogeneous. Measurement of [Ca(2+)]i in liver slices by CLSM was considered useful for studying heterogeneity between liver lobules, as well as between liver cells.

Clinical predictive factors for preterm birth in women with threatened preterm labour or preterm premature ruptured membranes

Background: Independent predictive factors of preterm delivery were evaluated using clinical data at hospitalisation by multivariate analysis.

Factors involved in resistance to early treatment of acute cholangitis patients.

BACKGROUND/AIMS Severity-based treatment is not homogenously effective for acute cholangitis patients and some are resistant to early treatment. We performed a retrospective cohort study involving acute cholangitis patients and analyzed factors strongly associated with resistance to early treatment. METHODOLOGY The subjects were 94 patients admitted to the Department of Gastroenterology, Showa University Hospital and diagnosed with acute cholangitis. The endpoint was set as the presence or absence of resistance to early treatment. Background and blood test results of the patients immediately after admission were surveyed and significant factors independently contributing to resistance to early treatment were extracted from the surveyed factors employing a logistic regression model. RESULTS The mean age of the patients was 73.2 ± 11.6 years and 58 were male (61.7%). Jaundice, fever and abdominal pain were observed in 46 (48.9%), 66 (70.2%) and 85 patients (90.4%), respectively. Twenty-eight patients (29.8%) were resistant to early treatment. On multivariate analysis, 3 factors (fever, serum amylase level and systolic blood pressure (below 100 mm Hg)) were extracted as significant factors independently contributing to resistance to early treatment (p<0.05). CONCLUSIONS If such resistance can be predicted before treatment, appropriate treatment may be selected to shorten the persistence of symptoms, improving the patients QOL.

Factors predicting the appearance of neutropenia in patients with advanced pancreatic cancer undergoing gemcitabine therapy.

BACKGROUND/AIM Factors predicting the appearance of neutropenia were evaluated in patients with advanced pancreatic cancer undergoing gemcitabine hydrochloride (GEM) therapy. METHODOLOGY The subjects were 92 patients who were diagnosed with unresectable advanced pancreatic cancer and underwent GEM therapy. Mono- and multivariate analyses were performed concerning each evaluated factor. The toxicity index (TI) was also prepared by combining the extracted predictive factors. RESULTS Severe neutropenia occurred in 26 patients (28.2%). As a result of multivariate analysis, the white blood cell count (WBC), CA19-9 and liver metastasis were extracted as factors independently and significantly contributing to the appearance of severe neutropenia (p<0.05). The TI was prepared by combining these 3 factors and their regression coefficients: TI = 4.777-0.605xWBC (x103/microL)-0.511xlog (CA19-9)-1.285xliver metastasis. CONCLUSIONS The WBC, CA19-9 and liver metastasis before treatment were shown to be related to the appearance of severe neutropenia after GEM therapy.