Inflammatory bowel disease associated with immune thrombocytopenic purpura in children.

OBJECTIVE: Previous reports suggest an association between inflammatory bowel disease (IBD) and immune thrombocytopenic purpura (ITP) in adults. To date, only five children with both diseases have been described. The aim of the study was to describe the characteristics of children with IBD and ITP. METHODS: Cases were obtained from the pediatric gastroenterology community by means of the pediatric gastroenterology internet bulletin board in June 1999. Eight cases were submitted from seven medical centers. Medical records were reviewed by two pediatric gastroenterologists and a pediatric hematologist. RESULTS: The age range of the patients was 2.1 to 16.5 years, with a mean age of 9.6 +/- 5.2 years. Four children had ulcerative colitis, three had Crohn disease, and one had indeterminate colitis. All had colonic involvement of IBD. Of eight patients, three had IBD first, three had ITP first, and two had both simultaneously. At ITP diagnosis, platelet count was less than 10,000/mL in five children, 17,000/mL in one child, and 50,000 to 60,000/mL in two children. Of the three children diagnosed with ITP first, two initially had rectal bleeding at the time of ITP diagnosis. Bone marrow evaluations, performed in six of eight children, were consistent with ITP. Six of the eight children had chronic ITP, including three children who were 5 years of age or younger. Therapy for ITP included steroids (n = 6), intravenous immunoglobulin (n = 6), Rh o (D) intravenous immunoglobulin (n = 2), and splenectomy (n = 1). CONCLUSIONS: The authors describe the largest pediatric case series of children with IBD and ITP. More than 50% of the children had the chronic form of ITP. Most patients responded to conventional therapy for ITP and IBD.

Transferrin is required for normal distribution of 59Fe and 54Mn in mouse brain.

Hypotransferrinemia (hpx/hpx) is a genetic defect in mice resulting in <1% of normal plasma transferrin (Tf) concentrations; heterozygotes for this mutation (+/hpx) have low circulating Tf concentrations. These mice provide a unique opportunity to examine the role of Tf in Fe and Mn transport in the brain. Twenty weanling wild-type BALB/cJ mice, 15 +/hpx mice, and 12 hpx/hpx mice of both sexes were injected i.v. with either 54MnCl(2) or 59FeCl(3) either 1 h or 1 week before killing at 12 weeks of age. Total brain counts of 54Mn and 59Fe were measured, and regional brain distributions were assessed by autoradiography. Hypotransferrinemia did not affect total brain Mn uptake. However, 1 week after i.v. injection, hpx/hpx mice had less 54Mn in forebrain structures including cerebral cortex, corpus callosum, striatum, and substantia nigra. The +/hpx mice had the highest total brain 59Fe accumulation 1 h after i.v. injection. A striking effect of regional distribution of 59Fe was noted 1 week after injection; in hpx/hpx mice, 59Fe was located primarily in choroid plexus, whereas in +/+ and +/hpx mice 59Fe was widely distributed, with relatively high amounts in cerebral cortex and cerebellum. We interpret these data to mean that Tf is necessary for the transport of Fe but not Mn across the blood-brain barrier, and that there is a Tf-independent uptake mechanism for iron in the choroid plexus. Additionally, these data suggest that endogenous synthesis of Tf is necessary for Fe transport from the choroid plexus.

Iron and manganese homeostasis in chronic liver disease: relationship to pallidal T1-weighted magnetic resonance signal hyperintensity.

The hyperintense signal in the globus pallidus of cirrhotic patients on T1-weighted magnetic resonance (MR) imaging has been postulated to arise from deposition of paramagnetic manganese2+ (Mn). Intestinal absorption of both iron and Mn are increased in iron deficiency; iron deficiency may therefore increase susceptibility to Mn neurotoxicity. To investigate the relationships between MR signal abnormalities and Mn and Fe status, 21 patients with chronic liver disease were enrolled (alcoholic liver disease, 5; primary biliary cirrhosis, 9; primary sclerosing cholangitis, 3; hepatitis B virus, 2; hepatitis C virus, 1; alpha1-antitrypsin deficiency, 1). Signal hyperintensity in the pallidum on axial T1 weighted images (repetition time/evolution time: 500 ms/15 ms) was observed in 13 of 21 subjects: four patients had mild hyperintensity, three moderate, and six exhibited marked hyperintensity. Erythrocyte Mn concentrations were positively correlated with the degree of the MR hyperintensity (Kendall's tau-b=0.52, P<0.005). The log of erythrocyte Mn concentration was also inversely correlated with all measures of iron status: hemoglobin (Pearson's R=-0.73, P<0.0005); hematocrit (R=-0.62, P<0.005); serum Fe concentrations (R=-0.65, P<0.005); and TIBC saturation (R=-0.62, P<0.005). These findings confirm the association of Mn with the development of pallidal hyperintensity in patients with liver disease. We further found that iron deficiency is an exacerbating factor, probably because of increased intestinal absorption of Mn. We therefore recommend that patients with chronic liver disease avoid Mn supplements without concurrent iron supplementation.

Existing and emerging mechanisms for transport of iron and manganese to the brain.

The metals iron (Fe) and manganese (Mn) are essential for normal functioning of the brain. This review focuses on recent developments in the literature pertaining to Fe and Mn transport. These metals are treated together because they appear to share several transport mechanisms. In addition, several neurological diseases such as Alzheimer's Disease, Parkinson's Disease, and Huntington's Disease are all associated with Fe mismanagement in the brain, particularly in the striatum and basal ganglia. Similarly, Mn accumulation in brain also appears to target the same brain regions. Therefore, stringent regulation of the concentration of these metals in the brain is essential. The homeostatic mechanisms for these metals must be understood in order to design neurotoxicity prevention strategies.

Transferrin response in normal and iron-deficient mice heterozygotic for hypotransferrinemia; effects on iron and manganese accumulation.

Hypotransferrinemia is a genetic defect in mice resulting < 1% of normal plasma transferrin (Tf) concentrations; heterozygotes for this mutation (+/hpx) have low circulating Tf concentrations. These mice provide a unique opportunity to examine the developmental pattern and response of Tf to iron-deficient diets, and furthermore, to address the controversial role of Tf in Mn transport. Twenty-three weanling +/hpx mice and forty-five wild-type BALB/cJ mice were either killed at weaning or fed diets containing either 13 or 72 mg kg-1 Fe, and killed after four or eight weeks. Plasma Tf concentrations were lower in +/hpx mice, plasma Tf nearly doubled and liver Tf was only 50% of normal in response to iron deficiency. Brain iron concentration did not correlate significantly with either plasma Tf or TIBC. However, iron accumulation into brain continued with iron deficiency whereas most other organs had less iron. These results imply that either there is a selected targeting of iron to the brain by plasma Tf or there is an alternative iron delivery system to the brain. Furthermore, we observed no differences in tissue distribution of 54Mn despite the differences in circulating Tf concentrations and body iron stores; this suggests that there are non-Tf dependent mechanisms for Mn transport.

Evidence for non-transferrin-mediated uptake and release of iron and manganese in glial cell cultures from hypotransferrinemic mice.

Transferrin (Tf) is accepted as the iron mobilization protein, but its role in transport of other metals is controversial. In this study, we used mixed glial cultures from hypotransferrinemic (Hp) mice to determine the dependence of these cells on transferrin for iron and manganese delivery and release. Hp mice have a splicing defect in the transferrin (Tf) gene, resulting in < 1% of the normal plasma levels of Tf. Cellular iron and manganese uptake increases over 24 hr in cultures of normal and Hp glial cells in the presence of standard concentrations of Tf in the media; although total 59iron uptake in the Hp mouse cultures was 2X greater than normal, 54Mn uptake was similar between the two groups. The absence of Tf in the media resulted in a significant increase in 59iron uptake in both normal and Hp glial but did not affect Mn uptake. Elevated Tf (10X normal) in the media reduced both 59iron and 54Mn uptake. Efflux of 59Iron and 54Mn occurred in normal and Hp cultures, indicating the existence of a dynamic exchange of metals, and that intracellular Tf is not necessary for metal release. However, in the absence of Tf in the media, significantly more iron was retained in the cells than if Tf were present in both normal and Hp glial cultures. 54Mn release was minimally affected by extracellular Tf. The data demonstrate that Tf is not required for iron and Mn uptake into glial cells. These data further demonstrate a dynamic metal exchange system for glial cells which is not dependent on intracellular Tf.

Distribution of injected iron 59 and manganese 54 in hypotransferrinemic mice.

Transferrin has been proposed as the mobilization protein for iron and manganese. To better understand the role of transferrin in the transport of these metals, we studied the tissue distribution of injected iron 59 and manganese 54 in the hypotransferrinemic (Hp) mouse mutant. The Hp mouse has a mutation in the transferrin gene and produces < 1% of normal transferrin levels. The tissue distribution of 59Fe and 54Mn in Hp mice was compared with that in animals heterozygous for the Hp mutation (50% transferrin levels) and wild-type animals. Formed elements in the brain, liver, spleen, heart, sternum/rib, plasma, and blood were analyzed for isotope incorporation at 24 hours, 7 days, and 4 weeks after injection. Tissue distribution of both 59Fe and 54Mn was similar in wild-type and heterozygote animals, indicating that decreased transferrin concentration and increased saturation did not influence the tissue distribution of the injected metals. The absence of transferrin in the Hp mutant was associated with abnormal tissue distribution of radiolabeled iron; there was 4 times more 59Fe than normal in the Hp liver and 10 times less 59Fe in the spleen and blood formed elements than normal. Injected manganese also accumulated at abnormally high levels in the Hp mouse liver. Distribution of either metal to the brain, heart, and sternum/rib was not affected by the absence of plasma transferrin. These results reveal that transferrin is required for proper targeting of manganese and iron, especially from the liver to other organs, but further indicate that nontransferrin transport mechanisms for iron and manganese must exist.

Antibiotic-induced colitis.

Most cases of antibiotic-associated diarrhea can be classified into two categories: cases in which Clostridium difficile is implicated and cases in which no putative agent or recognized pathophysiological mechanism is recognized. C difficile colonization produces a spectrum of conditions, ranging from asymptomatic carriage to fatal pseudomembranous colitis: it is implicated in virtually all cases of pseudomembranous colitis and up to 25% of cases of antibiotic-associated diarrhea without colitis. Clindamycin is notorious for its propensity to induce C difficile colitis, but in current practice, broad-spectrum penicillins and cephalosporins are the most frequently implicated agents, reflecting their widespread use. Treatment includes cessation or alteration of antimicrobial therapy when possible, provision of supportive care, and specific therapy aimed at eradicating C difficile if symptoms are severe or persistent or if antibiotic therapy cannot be safely discontinued. Oral metronidazole or vancomycin is equally efficacious in uncomplicated disease; optimal therapy for severe disease has not been established. Relapses occur in up to 15% of cases and cannot be accurately predicted or prevented.

Dystonia, hyperintense basal ganglia, and high whole blood manganese levels in Alagille's syndrome.

Hyperintensity of the globus pallidus on T1-weighted magnetic resonance imaging (MRI) has been reported in patients with chronic liver disease. This abnormality has been associated with the severity of liver disease and tremor, but its cause is unknown. Similar MRI signal abnormalities have been reported in experimental models of manganese neurotoxicity. This case report describes a child with Alagille's syndrome and end-stage liver disease who developed dystonia and tremor associated with an elevated whole blood manganese level and symmetric hyperintense globus pallidi and subthalamic nuclei on T1-weighted but not T2-weighted MRI. Liver transplantation was performed; 2 months later, neurological function was improved, manganese levels were normal, and the MRI signal abnormality had completely resolved. This child had neurological findings described in manganese neurotoxicity with compatible laboratory and radiological findings. Manganese is excreted by the liver in bile, and toxicity may have resulted from the inadequacy of this mechanism, subsequently corrected by liver transplantation.

Symptomatic manganese neurotoxicity in a patient with chronic liver disease: correlation of clinical symptoms with MRI findings.

Hyperintense symmetric pallidal lesions have been described in chronic hepatic failure. Similar lesions are reported in experimental models of manganese neurotoxicity. We describe an 8-year-old girl with chronic hepatic failure and dystonia in association with an elevated whole blood manganese level and symmetric hyperintense pallidal lesions on magnetic resonance imaging. After hepatic transplantation, her symptoms and signs resolved with normalization of magnetic resonance imaging and the whole blood manganese suggesting that in chronic hepatic failure, the pallidal lesions may be secondary to manganese deposition.

Post-infectious human serum antibodies inhibit IgA1 proteinases by interaction with the cleavage site specificity determinant.

Bacterial pathogens of the genera Neisseria and Haemophilus secrete IgA1 proteinases which cleave human IgA1 in the heavy chain hinge region. The exact peptide bond cleaved is strain-dependent, but remains invariant despite repeated subculture. Haemophilus influenzae and Neisseria meningitidis produce proteinases of two cleavage site specificities (type 1 and type 2). We examined serial acute and convalescent sera from patients recovering from meningitis due to N. meningitidis or H. influenzae, and found a significant rise in serum titer of inhibitory antibodies against these enzymes. In each case the proteinase from the infecting organism was more susceptible to inhibition than were proteinases from that genus that had different cleavage specificity. Inhibition of sixteen type 1-type 2 hybrid H. influenzae IgA1 proteinases revealed complete concordance between inhibitory titer and cleavage site specificity. Inhibition of hybrid proteinases differing in a 123 amino acid segment known to determine cleavage site specificity (termed the CSD) further localized the site of antibody action to this site. These results from a limited number of patients with natural infections suggest that inhibiting antibody recognizes epitopes within the CSD. Alternatively, antibody may bind to epitopes outside the CSD and inhibit via steric hindrance.

Qualitative testing for circulating immune complexes by use of zone electrophoresis on agarose.

On binding of antibody to antigen an immune complex is formed that has a net surface charge different from that of either of the two components. This, together with clonal restriction of the antibody response, gives rise to distinctive patterns that are readily apparent in stained agarose gels after routine zone electrophoresis. Most circulating immune complexes appear as a rectangular pattern, with well-defined edges, located in the gamma-region. The identity of the material responsible for these patterns has been established by three different experimental approaches: analysis of tetanus/anti-tetanus complexes formed in vitro, analysis of sera from rabbits with experimental immune complex disease, and analysis of human type II and type III cryoglobulins. Studies of reproducibility, interfering substances, and correlation with other assays for detecting immune complexes indicate that zone electrophoresis in agarose gel is a sensitive, highly specific technique for immune complex detection, of potential value as a screening tool.