Blunt abdominal trauma in children: how predictive is ALT for liver injury?

BACKGROUND: The liver is the second most commonly injured intra-abdominal organ in children. CT scanning is currently regarded as the "gold standard" in screening for intra-abdominal injury following blunt trauma. However, the risks associated with performing CT in children are not insignificant and, in addition, CT is not always readily available. This study investigates the utility of alanine aminotransferase (ALT) in screening for liver injury in paediatric trauma. METHODS: Two groups of patients were compiled from a prospectively identified trauma registry-one with liver injuries and another with intra-abdominal injuries other than to the liver. Inclusion in the study required that an initial ALT level had been obtained after injury. Where CT had been performed, a paediatric radiologist blind to the ALT results graded the severity of the liver injuries. The study groups were compared and a receiver operating characteristic (ROC) curve generated to derive the optimum ALT threshold to identify liver injury. RESULTS: 51 patients with liver injury and 65 with other intra-abdominal injuries were identified. An ALT level of > or =104 IU/l gave a sensitivity of 96% and a specificity of 80%. When liver injuries were stratified to identify only clinically significant liver injuries (grades III, IV and V), this ALT threshold identified 100% of patients with 70% specificity. CONCLUSIONS: In this sample, ALT appears to be a useful predictor for the presence or absence of liver injury. In haemodynamically stable children with clinical suspicion of isolated liver injury, identification of a normal ALT level (<104 IU/l) may reduce the need for unnecessary transportation for CT scanning with subsequent radiation exposure.

Hepatitis C virus (HCV) infection in bone marrow transplant patients after transfusions from anti-HCV-positive blood donors.

In March 1992, 12 bone marrow transplant patients at the Fred Hutchinson Cancer Research Center received blood components from donors who were anti-HCV-nonreactive by first generation ELISA but whose serum later tested anti-HCV-reactive to a second generation ELISA. All these blood components were further tested for anti-HCV using a second-generation RIBA and for HCV RNA by polymerase chain reaction. Recipient sera were tested for HCV RNA prior to and following blood component infusion. Blood components from four donors were positive for HCV RNA. All recipients of HCV RNA-positive blood components became viremic on the first day tested post-infusion. In addition, two recipients of HCV RNA-negative blood components tested HCV RNA-positive both pre- and post-infusion. Viremia persisted up to the time of death or day 100 in five of the six patients who were HCV RNA-positive post-transplant. No HCV RNA-positive recipient developed symptomatic acute hepatitis, and only two had aminotransferase elevations consistent with chronic hepatitis. We conclude that HCV RNA-positivity in blood components accurately predicts transmission of virus. Infection with HCV did not adversely affect short-term patient outcome following bone marrow transplantation.

Marrow transplantation from hepatitis C virus seropositive donors: transmission rate and clinical course.

Bone marrow transplant recipients are at risk for acquiring hepatitis C infection from the donated marrow. Twelve patients who were hepatitis C virus (HCV) RNA-negative pretransplant received marrow from anti-HCV seropositive donors. HCV RNA was present in the sera of seven of these donors. After transplant, serial serum specimens were obtained from all marrow recipients for determination of HCV RNA and aminotransferase levels. All seven recipients of marrow from HCV RNA-positive donors were HCV RNA-positive after marrow infusion; none cleared virus from the serum. All five recipients of marrow from anti-HCV seropositive, HCV RNA-negative donors remained free of HCV RNA in serum up to day 100. Abnormal serum aminotransferases were common in both HCV RNA-negative and HCV RNA-positive marrow recipients. One HCV-infected recipient developed marked elevation in aminotransferases after immunosuppressive drugs were stopped. We conclude that the presence of HCV RNA in the serum of marrow donors is an accurate predictor of HCV infection in marrow recipients. The acute infection was subclinical in all patients. The long-term risk of chronic hepatitis C virus infection in these patients remains to be determined.

Involvement of Porphyromonas gingivalis fimbriae in adherence to Streptococcus gordonii.

Adherence of Porphyromonas gingivalis to early plaque bacteria, such as Streptococcus gordonii, is considered an important colonization mechanism. The molecules that mediate this interspecies binding have not been determined. Fimbriae were prepared from P. gingivalis 33277 by mild agitation, ammonium sulfate precipitation and DEAE-Sepharose chromatography. In a nitrocellulose blot adherence assay, purified fimbriae inhibited S. gordonii G9B-P. gingivalis 33277 binding by up to 54%. In addition, fimbriae bound to S. gordonii cells in a dot-blot assay. Incubation of fimbriae with S. gordonii cells followed by washing, sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), electroblotting and probing with P. gingivalis antibodies also revealed that the fimbriae bind to S. gordonii. In contrast, S. gordonii did not interact with fimbriae that were first subjected to SDS-PAGE and electroblotting or deposited on a nitrocellulose membrane, suggesting that conformational determinants of the fimbriae may be important in binding. The results indicate that binding between P. gingivalis and S. gordonii is mediated, at least in part, by the porphyromonads' fimbriae.

Effects of streptozotocin diabetes on amylase release and cAMP accumulation in rat parotid acinar cells.

Rat parotid responses to sympathetic nerve stimulation in vivo are impaired 2-4 weeks after the induction of streptozotocin diabetes. In this study, the effects of experimental diabetes of similar duration and severity on noradrenaline-stimulated amylase release and cAMP accumulation were examined in vitro. Amylase levels were significantly lower in acinar cells isolated from diabetic animals than in controls, and cellular amylase increased after treatment of the diabetic animals with either thyroxine (T4) or insulin. Diabetes and T4 had no apparent affect on amylase release measured as a percentage of the total. In contrast, giving insulin resulted in a significant reduction in maximal secretion (20.4 +/- 2.4% compared with 43.6 +/- 7.6%). Similar results were observed when amylase release was stimulated with forskolin. Basal cAMP levels were unaffected by diabetes or T4 (7.8 +/- 2.3 pmol/mg protein), but stimulated cAMP levels were significantly greater in diabetic acinar cells than in controls. Insulin reversed the effects of diabetes on cAMP accumulation, whereas T4 had no effect. Thus, diabetes (2-4 weeks) and insulin in vivo appear to have paradoxical effects on parotid amylase release and cAMP accumulation in vitro. Further, the effects of diabetes appear to be unrelated to thyroid status.