Thromb Haemost

Patterns of hepatitis G viraemia and liver disease in haemophiliacs previously exposed to non-virus inactivated coagulation factor concentrates.

Year 1998
Hanley JP. Jarvis LM. Hayes PC. Lee AJ. Simmonds P. Ludlam CA.
Department of Haematology, Royal Infirmary of Edinburgh, Scotland, UK.
Hepatitis G virus (HGV), a novel flavivirus, has been implicated as a cause of posttransfusion hepatitis. We have performed a longitudinal study in a cohort of haemophiliacs (n = 68) who previously received non-virus inactivated coagulation factor concentrates to assess both patterns of HGV viraemia and any associated liver disease. Hepatitis C virus (HCV) RNA was present in 58/68 and co-infection with human immunodeficiency virus (HIV) was present in 15/68. HGV RNA was detected in 17/68 (25%) samples from the mid-1980s. There was no association between either HIV infection (p = 0.74) or co-infection with a particular HCV genotype (p = 0.62). However, there was a relationship between HGV viraemia and the severity of haemophilia (p = 0.0004) with HGV RNA detected in 5/19, 9/16 and 3/32 patients with mild, moderate and severe haemophilia respectively. A longitudinal study was performed in 15/17 haemophiliacs with HGV viraemia using stored serum samples from the 1980s and 1990s. HGV viraemia persisted in 8/15 and cleared in 7/15 over a variable period of time. A Weibull model was constructed to estimate the duration of HGV viraemia in the study group. The 75th and 90th percentiles for the duration of HGV were estimated to be 8.7 years (95%, confidence interval 4.8-15.7) and 23.6 years (95% confidence interval 11.8-47.1) respectively. Laparoscopic liver inspection/biopsy was performed in 25/68. There was no association between severity of liver disease and HGV viraemia (p = 0.43). This study demonstrates considerable variation in patterns of HGV viraemia in haemophiliacs. We found little evidence to implicate HGV as a major cause of chronic liver disease in haemophiliacs.

Increased thromboxane metabolites excretion in liver cirrhosis.

Year 1998
Davi G. Ferro D. Basili S. Iuliano L. Camastra C. Giammarresi C. Santarone S. Rocca B. Landolfi R. Ciabattoni G. Cordova C. Violi F.
Department of Medicine, University of Chieti, Italy. gdavi@unich.it
An augmented systemic production of thromboxane (TX) A2, as assessed by urinary excretion of the thromboxane metabolites, has been described in severe liver cirrhosis. However, the significance of this finding remains unclear since in liver cirrhosis a number of phenomena i.e. altered hepatic TXA2 metabolism, increased intrasplenic platelet destruction, may affect TXA2 entry into systemic circulation as well as its metabolism. In order to further clarify this, we measured both major enzymatic metabolites of TXB2 in the urine of 44 patients affected by liver cirrhosis, subdivided in three classes on the basis of Child-Pugh criteria. Urinary 11-dehydro-TXB2 and 2,3-dinor-TXB2 were assayed with previously validated RIA techniques. The urinary excretion rate of 11-dehydro-TXB2 was significantly (p = 0.0001) increased in the cirrhotic patients (673.5 pg/mg cr, median) in comparison with the controls (275 pg/mg cr, median) but no significant difference could be demonstrated among the excretion rates of the three patient subgroups. The excretion rate of 2,3 dinor-TXB2 was also significantly (p = 0.0001) increased in the patients (824 pg/mg cr, median) in comparison with controls (175 pg/mg cr, median), with a significant (p < 0.05) increase from class A (381 pg/mg cr) to class C (1337 pg/mg cr). The sum of the two enzymatic metabolites was significantly (p = 0.0001 ) increased in the cirrhotic patients in comparison to controls, with a progressive increase from class A (1003 pg/mg cr, median) to class C (2240 pg/mg cr, median). The urinary excretion of 2,3 dinor-TXB2 was significantly (p = 0.008) related to plasma prothrombin fragment 1+2 (F1+2). This study provides further evidence of increased thromboxane biosynthesis in liver cirrhosis. Moreover, we demonstrate intraliver shift of thromboxane metabolic disposition, due to progressive liver decompensation, because only the fraction undergoing beta-oxidation to 2,3-dinor-TXB2 was progressively increased with the degree of liver failure. We, also, find a significant correlation between urinary excretion of 2,3-dinor-TXB2 and plasma F1+2, suggesting that clotting activation could partly account for in vivo platelet activation.

Anti-hepatitis G E2 antibody detection and its relation to serum HGV-RNA in patients with clotting disorders: high prevalence of HGV infection and spontaneous remission.

Year 1998
Sheng L. Soumillion A. Peerlinck K. Verslype C. Schelstraete R. Gyselinck F. Emonds MP. Hess G. Vermylen J. Desmyter J. Yap SH.
Department of Medicine, Rega Institute and University Hospitals, Leuven, Belgium.
In a previous study, we have determined the prevalence of serum HGV-RNA in patients with congenital clotting disorders. Twenty-six (15%) of 175 patients investigated were serum HGV-RNA positive. In addition, HGV-RNA was detectable in peripheral blood mononuclear cells (PBMC) in ten percent of the cases, three of these patients were serum HGV-RNA negative. In the present study, we have determined the prevalence of anti-HGV-E2 antibodies in the same patient population. Anti-HGV-E2 as determined by ELISA was detected in 45 patients (25.7%). Forty of these patients were serum HGV-RNA negative. Ninety-two percent of the 26 HGV viremic patients and all but one patient (44 patients) with detectable anti-HGV-E2 had coinfection with the hepatitis C virus (HCV). Of these coinfected patients, 62.5% of HGV viremic patients and 53% of anti-HGV-E2 positive patients showed elevated serum ALT levels. Anti-HGV-E2 seroconversion is thus not associated with HCV infection. Two patients who were solely infected with HGV had normal serum ALT levels. In a retrospective longitudinal study, we have observed in 15 patients that serum HGV-RNA persisted during one to 19 years of follow-up, while anti-HGV-E2 was repeatedly negative. Five additional patients who were anti-HGV-E2 positive with concomitant detectable HGV-RNA (4 patients in serum and 1 patient in PBMC) became HGV-RNA negative during follow-up, ranging from 1 to 8 years after the first detection of anti-HGV-E2 antibodies. Two patients had lost anti-HGV-E2 antibodies 3 to 6 years after the seroconversion without the re-appearance of serum HGV-RNA. From these findings, it is clear that the prevalence rate of HGV infection in patients with clotting disorders as determined by PCR assay for HGV-RNA and anti-HGV-E2 by ELISA is actually higher than the prevalence of HGV viremia. Although HGV viremia may persist for longer than 19 years, most of the patients infected with HGV may clear the viremia spontaneously. The clearance of viremia is usually associated with seroconversion to anti-HGV-E2. In addition, anti-HGV-E2 may be lost during years of follow-up without the reappearance of the HGV-RNA. Although HGV infection does not seem to influence the fate of HCV infection and does not induce increased levels of serum ALT, the clinical significance of long-term infection remains to be established.

Platelet count and survival in patients with colorectal cancer--a preliminary study.

Year 1998
Monreal M. Fernandez-Llamazares J. Pinol M. Julian JF. Broggi M. Escola D. Abad A.
Department of Surgery, Hospital Universitari Germans Trias i Pujol, Badalona, Spain.
In a previous report we found an inverse correlation between pre-operative platelet count (PlC) levels and the risk of post-operative pulmonary embolism in patients undergoing hip surgery. In the present study, we prospectively evaluated the prognostic significance of pre-operative PlC levels on survival in 180 consecutive patients undergoing surgery for colorectal cancer. Other major clinicopathological parameters studied were age, gender, Dukes' stage, duration of surgery, pre-operative haemoglobin levels and transfusion requirements. There were no significant differences in mean pre-operative PlC levels according to tumor stage. Thirty-three patients (18%) died during follow-up (3-23 months, median: 13 months). Univariate analysis (Kaplan-Meier estimates) showed that advanced tumor stage (p < 0.001), duration of surgery (p < 0.05) and a high pre-operative PlC level (p < 0.001) were significantly associated to a poor survival. The multivariate Cox analysis revealed that tumor stage (RR:5.734; 95%C.I.: 2.644-12.44), a high pre-operative PlC level (RR: 2.467; 95%C.I.: 1.117-5.452), and to a lesser extent the patients' age remained independent prognostic variables for mortality. The findings of this preliminary study may be of interest from the point of view of pathogenesis, but also clinically, since they might be used in the decision as to which patients or groups of patients should receive more aggressive therapeutic intervention.