Glomerular monocyte/macrophage influx correlates strongly with complement activation in 1-week protocol kidney allograft biopsies.

BACKGROUND: The specific role of monocytes/macrophages (MO) in kidney graft rejection is not yet fully elucidated. In a recent protocol biopsy study of living-donor recipients, we demonstrated massive capillary influx of MO, associated with severe complement activation and acute rejection (AR) 1 week after transplantation [Sund et al.]. To gain further insight into the possible relationship between MO and complement activation, we analyzed glomerular and interstitial MO in these biopsies. METHODS: Twenty-seven protocol biopsies were stained with antibodies to calprotectin (L1) and CD68 as markers for MO. Cells were counted as an average number per glomerulus and as an average number per defined visual field in the interstitium. Polymorphonuclear leukocytes (PMN) were counted in glomeruli and interstitium by light microscopy. Baseline specimens from 10 of the patients served as controls. The results were compared with data on deposition of complement from the foregoing study, and with histopathologic and clinical data on AR. RESULTS: Cases with diffuse C4d deposition in peritubular capillaries consistent with acute antibody-mediated rejection (AbAR) (n = 4) had significantly higher numbers of intraglomerular MO than the other protocol biopsies (L1: median 20.7 vs 3.6, p = 0.0002; CD68: median 10.1 vs. 2.0, p = 0.0008). With a cut-off of 10 L1-positive and 6 CD68-positive MO, the specificity for the diagnosis of AbAR was 96% and 91%, respectively. The number of interstitial MO was significantly higher in patients with AR than in those without, but in contrast to glomerular MO, interstitial MO could not discriminate between complement positive and negative AR. The number of glomerular and interstitial PMNs was significantly higher in the AbAR group than in the other protocol biopsies. CONCLUSIONS: The strong correlation between complement activation and early glomerular influx of MO in the kidney allograft suggests a causal relationship between these 2 events. At 1 week after transplantation, a number of 10 L1-positive and 6 CD68-positive MO per glomerulus indicates AbAR.

Narthecium ossifragum (L.) huds. causes kidney damage in goats: morphologic and functional effects.

We studied the effects of Narthecium ossifragum on goat kidneys. Twenty-five Norwegian dairy goats, 5 weeks to 4 months of age, were orally dosed with an aqueous extract from N. ossifragum. In experiment 1, we studied microscopic and functional changes in 12 animals that were euthanatized 2, 3, 4, 5, and 6 days after treatment. In experiment 2, we included ultrastructural studies on serial renal biopsies and urine analysis from five extract-treated animals and two controls. In addition, urine samples were collected from four dosed and two control goats. Ultrasonography revealed perirenal and retroperitoneal fluids. Microscopic changes were observed after 6 hours. The findings, most obvious in the inner cortex and the outer medulla, consisted of cytoplasmic vacuolization, interstitial edema, and focal necrosis of tubular epithelial cells. Ultrastructurally, the tubules had loss of microvilli, irregular cytoplasmic vacuolization, mitochondrial swelling with loss of cristae, and irregular but continuous basement membranes even with necrosis. In the glomeruli, there were occasional endothelial damage and shortening and swelling of the foot processes. Peritubular capillaries had breaks in the vessel walls and irregular endothelial cell edema, and the interstitium had marked edema. The functional lesions included elevated serum urea, creatinine, and magnesium concentrations, a slight decrease in serum calcium concentration, elevated urine protein and urine protein-creatinine ratio, and increased activities of urine alkaline phosphatase and gamma glutamyl transferase. Our findings indicate a fast-acting toxic principle inducing damage by both direct toxic and secondary ischemic effects.

Hepatocellular damage in porcine endotoxemia: beneficial effects of selective versus non-selective nitric oxide synthase inhibition?

UNLABELLED: While nitric oxide (NO) is implicated as an important mediator of hypotension in sepsis and endotoxemia, its role as a mediator of tissue injury in shock is controversial. During porcine endotoxemia (lipopolysaccharide (LPS) 1.7 microg kg(-1) x h(-1) i.v. for 6 h), we compared circulatory and morphological changes in the liver induced by two different NO synthase inhibitors (N(G)-nitro-L-arginine methyl ester, L-NAME, 25 mg x kg(-1) i.v. and aminoethyl-isothiourea, AE-ITU, 10 mg x kg(-1) i.v.), both given after 3 h. LPS induced time-dependent tissue reactions with edema, sinusoidal dilation, packing of red cells and leukocyte infiltration, progressing to endothelial cell and hepatocyte damage, formation of thrombi, and at 6 h widespread necrosis. These changes were similar in all pigs receiving LPS, regardless of treatment with NOS inhibitors. LPS caused significant increases in aspartate aminotransferase (AST), alkaline phosphatase (ALP) and alpha glutathione S-transferase (alpha-GST), L-NAME caused further increases in AST, ALP and alpha-GST, while AE-ITU prevented the late increase in ALP and alpha-GST observed in the other LPS groups. LPS reduced liver blood flow by approximately 40%. L-NAME further reduced flow by approximately 50%, while AE-ITU restored liver blood flow to baseline values. CONCLUSION: L-NAME in endotoxemia had detrimental effects on liver circulation, while AE-ITU improved liver blood flow and attenuated the late increase in liver enzymes. Liver morphology was unaffected within the 3-h observation time after NOS inhibition.

Familial mydriasis, cardiac arrhythmia, respiratory failure, muscular weakness and hypohidrosis.

OBJECTIVES: To describe a family with some sort of progressive autonomic failure in one generation (2 affected of a sibship of 7 sisters). The main features were: mydriasis, cardiac arrhythmia, cardiomegaly, hypohidrosis, respiratory failure, and muscular weakness. METHODS: Pupillometry, evaporimetry, and isokinetic power measurements were carried out. RESULTS: The autonomic dysfunction pattern (mainly cardiac abnormalities, mydriasis) seems to differ somewhat from that of progressive autonomic failure (Shy-Drager syndrome). "Lewy body-like" inclusions were present, in particular in substantia nigra, but also in locus ceruleus and raphe nuclei (cell loss only in locus ceruleus). There were no oligodendroglial, cytoplasmatic inclusions, apparently a marker in multiple system atrophy. Proper Lewy bodies were also present. Differences seemed to prevail vs the Shy-Drager syndrome. Various traits: muscular weakness pattern (e.g. preferential peroneal distribution), minor elbow contractures, and arrhythmia were reminiscent of Emery-Dreifuss muscle dystrophy (E-D). Distinguishing features included: hereditary pattern, mydriasis, and hypohidrosis. CONCLUSION: Conceivably, this disorder is close to, but still not identical with E-D.

Inhibition of platelet aggregation by the GPIIb/IIIa antagonist reopro does not significantly prolong xenograft survival in an ex vivo model.

Effects on hyperacute rejection were studied in a discordant model with the platelet GPIIb/IIIa antagonist Reopro. Pig kidneys perfused with human blood survived median 118 min in the Reopro group and 103 min in the controls (P = 0.22). Platelet and leukocyte counts decreased, whereas plasma thrombospondin and soluble as well as platelet membrane P-selectin increased significantly in both groups without significant intergroup differences. beta-Thromboglobulin and myeloperoxidase increased significantly more in the control group than in the Reopro group (P = 0.009 and P = 0.02, respectively). The classical complement pathway was substantially and similarly activated in both groups. Light and electron microscopy revealed arterial thrombi and numerous glomerular platelet aggregates in the control group in contrast to the Reopro group. In conclusion, Reopro reduced platelet aggregation, and platelet and leukocyte activation to some extent, but had no effect on complement activation and did not significantly prolong xenograft survival, even though better preservation of morphology was shown.

Living donor kidney transplants: a biopsy study 1 year after transplantation, compared with baseline changes and correlation to kidney function at 1 and 3 years.

INTRODUCTION: Chronic changes in biopsies from long-term stable kidney allografts have been reported to correlate with graft prognosis. Morphological changes in baseline ('zero-hour') biopsies have been described as well, but their importance for long-term prognosis have been less clear. The aim of the present study was to evaluate biopsy changes from baseline to 1 year after transplantation in patients receiving kidneys from living donors, and to assess the possible prognostic implications of these findings. METHODS: Light microscopical changes in 18 gauge full-core biopsies were scored semi-quantitatively in 33 patients 1 year after transplantation, and compared to baseline changes previously reported [1]. All cases were also examined with transmission electron microscopy. The semi-quantitative data from baseline and at 1 year were correlated with kidney function 1 and 3 years after transplantation. The reproducibility of baseline findings regarding arteriosclerosis and arteriolar hyalinosis was tested by comparison with biopsies 1 week after transplantation (n = 43). RESULTS: We found a significant increase in mesangial glomerular sclerosis (P<0.001), interstitial fibrosis/tubular atrophy (if/ta) (P = 0.002), and mononuclear cell interstitial infiltration (P = 0.003) after 1 year, compared to baseline changes. There was an increase of arteriosclerosis (P = 0.028) and arteriolar hyalinosis (P = 0.006) when compared to biopsies taken 1 week after transplantation, but not when compared to the 'zero-hour' findings. Electron microscopy revealed one case of recurrent immune-complex glomerulonephritis and another case of recurrent light chain deposition kidney disease. Comparing 1-week vascular findings with baseline gave a low level of reproducibility, probably due to sampling error. Baseline biopsy findings could not predict long-term kidney function. In the 1-year biopsy, if/ta was significantly correlated with serum creatinine (P = 0.007) and glomerular filtration rate (GFR) (P<0.001) at 1 year, with serum creatinine at 3 years (P = 0.011), and with the first-year cumulative dose of methylprednisolone (P = 0.004). Serum creatinine at 1 year, however, was found to be the most accurate predictor of 3-year kidney function (P<0.001). Donor age was correlated to kidney function at 3 years (P = 0.013) but not at 1 year after transplantation. CONCLUSION: Morphological changes in baseline biopsies of living donor kidneys tend to become more pronounced in well-functioning allografts during the first year after transplantation. In the 1 year biopsy, if/ta seems to be the most reliable variate for grading of chronic changes. However, 1-year serum creatinine predicted long-term kidney function more precisely than did the biopsy scores. Based on the results of the present study, a protocol 1-year biopsy does not seem warranted in the management of the graft recipient with a stable kidney function.

Compstatin, a peptide inhibitor of C3, prolongs survival of ex vivo perfused pig xenografts.

Compstatin, a newly described C3-binding peptide, inhibits complement activation by blocking C3 convertase-mediated cleavage of C3. As the complement activation is an essential part of the rejection reaction, we evaluated the ability of Compstatin to delay or prevent hyperacute rejection in an ex vivo xenograft model. Pig kidneys were perfused with fresh human blood containing either Compstatin (n=6) or a control agent (n=6). Graft survival and activation of complement, leukocytes and platelets both in the fluid-phase and in the tissue were examined. The survival of the Compstatin-perfused kidneys (median, 380 min) was significantly (P=0.0036) longer than that of the controls (median, 90 min). The classical complement pathway (C1rs-C1inhibitor and C4bc) was significantly and equally activated in both groups during the first 60 min. C3 activation products increased fivefold and terminal complement complex eightfold in the control group, but no increase occurred in the Compstatin group during this period. Immunohistochemistry showed less C3 and fibrin deposition and immune electron microscopy showed less terminal SC5b-9 complement complex deposition in the Compstatin group. A significant change in total white cells, neutrophils, myeloperoxidase, and expression of the surface activation markers CD11b (CR3) and CD35 (CR1) and CD62L (L-selectin) was observed in both groups. Leukocyte activation was lower in the Compstatin group but the difference was not statistically significant. There were no differences in platelet counts, thrombospondin, soluble P-selectin or beta-thromboglobulin between the groups. We conclude that Compstatin prolongs graft survival and suggest that it may be a useful agent for attenuating hyperacute rejection by inhibiting C3 and thus terminal complement pathway activation.

Platelet compatibility of an artificial surface modified with functionally active heparin.

Platelet compatibility after coating an artificial material with functionally active heparin was investigated. Blood was circulated in uncoated or heparin coated PVC tubing. In one hour platelet counts decreased from 155 (113-184)x10(9)/l to 124 (100-148)x10(9)/l with uncoated compared to 164 (132-192)x10(9)/l with heparin coated tubing (intergroup p = 0.02). Beta-thromboglobulin increased from 116 (80-148) microg/l to 1039 (757-1298) microg/l with uncoated and to 352 (229-638) microg/l with heparin coated tubing (intergroup p = 0.005). Platelet counts and beta-thromboglobulin correlated closely with complement activation. Solid-phase enzyme immunoassay demonstrated substantial deposition of CD42a/GPIbIX and CD61/GPIIIa on uncoated, but not on heparin coated tubing (intergroup p<0.0005). Scanning electron microscopy demonstrated activated platelets and aggregates on uncoated in contrast to heparin coated tubing, where scattered, unactivated platelets were found. Changes in P-selectin and microparticles were minor. In conclusion, this heparin surface substantially improved platelet compatibility. Markers of choice for in vitro evaluation were platelet counts, beta-thromboglobulin and platelet deposition.

Selective inhibition of inducible nitric oxide synthase maintains haemodynamic stability without untoward consequences for hepatic function or morphology.

OBJECTIVE: To examine the effects of the inducible nitric oxide synthase inhibitor aminoethyl-isothiourea (AE-ITU) on haemodynamic measurements, and correlate these with hepatic morphology and function in a porcine model of endotoxaemia. DESIGN: Experimental study. ANIMALS: 15 juvenile pigs. INTERVENTIONS: Flow probes were placed around the hepatic artery and portal vein. Catheters were introduced into the portal and hepatic veins, pulmonary artery, and aorta. Infusion of AE-ITU was started one hour before that of endotoxin (study group n = 6); thereafter both substances were infused simultaneously until the end of the study (6 hours). The controls (n = 9) had endotoxin alone. MAIN OUTCOME MEASURES: Hepatic morphology assessed by light and electron microscopy; and hepatic integrity and function by transaminase activities and oxygen consumption. Systemic, pulmonary, and hepatic blood flow and pressure. RESULTS: AE-ITU maintained systemic blood pressure (p < 0.05 compared with controls) without causing pulmonary hypertension. Neither hepatic morphology nor function were adversely influenced. CONCLUSION: In endotoxaemia AE-ITU has a favourable haemodynamic profile which is achieved without impairment of hepatic function or morphology.

Postpartum renal failure: a complex case with probable coexistence of hemolysis, elevated liver enzymes, low platelet count, and hemolytic uremic syndrome.

BACKGROUND: Postpartum renal failure in previously healthy subjects is associated most often with preeclampsia and/or hypertension; hemolysis, elevated liver enzymes, low platelets (HELLP) syndrome, hemolytic uremic syndrome; or thrombotic thrombocytopenic purpura. Transient oliguria associated with preeclampsia is common, but renal failure is rare. Coexistence of HELLP and hemolytic uremic syndromes has been suggested, but histopathologic documentation of this combination has been scarce. CASE: A 30-year-old primigravida with severe preeclampsia at 35 weeks and 3 days' gestation presented with the development of HELLP syndrome and renal failure postpartum. Histopathologic lesions characteristic of hemolytic uremic syndrome were present in the kidney. CONCLUSION: Probable overlapping of HELLP and hemolytic uremic syndromes in pregnancy or postpartum should be taken into consideration when treating patients with these syndromes and associated complications, such as renal failure.

In situ complement activation in porcine membranoproliferative glomerulonephritis type II.

Pigs genetically deficient in complement factor H all develop lethal membranoproliferative glomerulonephritis (MPGN) type II characterized by massive glomerular deposits of complement, intramembranous dense deposits, and mesangial hypercellularity. To elucidate the chronological relationship between these glomerular changes, and to precisely determine the localization of glomerular complement deposits, we studied kidney specimens from factor H-deficient piglets at different ages from fetal life until terminal kidney failure had developed. Deposits of C3 and the terminal complement complex localized within the glomerular basement membrane (GBM) were present already in factor H-deficient fetuses, without concurrent intramembranous dense deposits or mesangial hypercellularity. Incipient subendothelial dense deposits containing complement appeared no earlier than four days after birth, and intramembranous dense deposits in older piglets with established MPGN type II also contained large amounts of complement as detected by immune electron microscopy. Onset of kidney failure coincided with pronounced mesangial hypercellularity and expansion, compromising glomerular capillary patency. Formation of glomerular capillary wall double contours coincided with electron microscopic evidence of laminar disintegration of intramembranous dense deposits. Complement was also deposited in the mesangial matrix, but not on glomerular cells. We conclude that all components of the alternative and terminal pathways of complement have access into the GBM and the mesangial matrix. In the absence of factor H, complement is spontaneously activated and deposited in situ in these locations resulting in dense deposit formation. It is proposed that factor H dysfunction may play an essential role even in human MPGN type II.

Morphological studies of baseline needle biopsies from living donor kidneys: light microscopic, immunohistochemical and ultrastructural findings.

Fifty-seven consecutive living donors (31 women and 26 men aged 20.7-72.3 years, mean 50.6 years) were subjected to needle biopsy during nephrectomy, immediately before removal of the kidney. By light microscopy, semiquantitative scoring (0-3) was performed for arteriosclerosis, arteriolar hyalinosis (hyalin arteriolosclerosis), glomerulosclerosis, interstitial mononuclear cell infiltration, and interstitial fibrosis/tubular atrophy. Whereas vascular changes were striking in many biopsies (arteriosclerosis grades 2-3: 28/54 cases, arteriolar hyalinosis grades 2-3: 15/55 cases), glomerular and tubulointerstitial changes were mostly low grade. The morphological changes tended to be more pronounced in middle-aged and older individuals, but, in particular, vascular changes were seen also in the younger age group. Immunofluorescence microscopy revealed glomerular granular staining for IgM in 52.7% of the cases, IgA in 9.1%), IgG in 1.8%, and C3 in 12.7%. The main ultrastructural finding was glomerulosclerosis; one case with diffuse glomerular IgA showed distinct dense deposits, and one case showed similar dense deposits without IgA deposition. Arteriolar wall deposition of C3 was found in 58.2% of the cases, and IgM in 10.9%. Especially C3 occurred both with arteriolar hyalinosis and in arterioles without light microscopic alterations. The observation of significant vascular changes in baseline biopsies is relevant especially in the differential diagnosis of chronic rejection and cyclosporine nephropathy. The immunohistochemical findings strongly indicate the occurrence of immunoglobulins and complement factor C3 in both glomeruli and arterioles without clinical or morphological signs of renal disease. The possible pathophysiological significance of such deposits remains, however, uncertain.

Colony variation of Helicobacter pylori: pathogenic potential is correlated to cell wall lipid composition.

BACKGROUND: Differences in expression of disease after infection with Helicobacter pylori have so far been connected with host factors and bacterial interstrain variation. In this study, spontaneous and ecology-mediated intrastrain variation was examined. METHODS: Four clinical isolates of H. pylori were shown to give rise to two colony forms. Bacterial morphology was examined by electron microscopy. Bacterial fractions were examined for proteins using ion exchange chromatography and SDS-PAGE; for lipids using thin-layer chromatography, lipid anion-exchange chromatography, column chromatography on silica gel, 31P-NMR, gas chromatography and mass spectrometry. Bacterial in vitro invasiveness and adhesiveness were examined in two different systems, and urease and VacA toxin were assayed by Western blot analysis. RESULTS: H. pylori was shown to give rise to two colony forms: at normal pH the population was dominated by L colonies. One strain was chosen for further studies. Bacteria from L colonies retained VacA toxin and urease, did not invade or adhere to epithelial cells, and contained normal quantities of phosphatidylethanolamine. In a small frequency, spontaneous S colonies were formed. Bacteria from these colonies released VacA and urease, adhered to and invaded epithelial cells and contained increased amounts of lysophosphatidyl ethanolamine and phosphatidyl serine. After addition of HCl to the culture medium (pH6), almost only S colonies were formed. The results demonstrate that environmental factors, such as HCl, can change the bacterial cell wall, and thereby enhance expression of virulence factors of H. pylori in vitro. A similar in vivo variation would have implications for our understanding of the interaction between HCl secretion in the gastric mucosa and H. pylori in the development of peptic ulcer disease.

[May-Hegglin's syndrome. Hereditary macrothrombocytopenia with inclusions in neutrophil granulocytes]. / May-Hegglins syndrom. Hereditaer makrotrombocytopeni med inklusjoner i nøytrofile granulocytter.

May-Hegglin's anomaly is a rare, autosomally dominant inherited syndrome not previously described in Norway. We report two cases, sisters with characteristic light and electron microscopic inclusions in the granulocytes, thrombocytopenia and giant platelets. The relevant literature is briefly reviewed.

Encephaloneuropathy with lysosomal zebra bodies and GM2 ganglioside storage.

An 11-year-old girl died of a neuronal storage disorder that clinically was characterized by failure to thrive and muscular hypotonia from birth, with the subsequent evolution of motor neuron disease, epilepsy, and dementia. A wide range of metabolic disorders, including all forms of GM2 gangliosidosis, could be excluded. Electron microscopy demonstrated neuronal zebra body inclusions, and immunohistochemistry demonstrated that GM2 ganglioside was a major constituent of the storage material. We suggest that the patient died of a lysosomal storage disease that is clinically and biochemically different from Tay-Sachs disease, Sandhoff disease, and other GM2 gangliosidoses described previously. This case also further demonstrates that significant accumulation of GM2 ganglioside, which is crucial for dendritic formation, may occur in neuronal storage diseases lacking known defects in ganglioside catabolism.

Intravascular lung macrophages play an essential role in lipid entrapment and the inflammatory tissue reaction seen after long-term lipid-based parenteral nutrition in pigs. An ultrastructural study.

To detect possible interactions between lipid-based total parenteral nutrition (TPN) substrates and mononuclear phagocytes, ultrastructural in vitro and in vivo studies were carried out on material from pigs. Mononuclear phagocytes isolated from peripheral blood, phagocytosed lipid after incubation with 1 mg/ml Intralipid for 24 h. Similarly, lipid was taken up by intravascular macrophages in the lungs and liver after central venous administration of TPN containing 2.3 g/kg body weight/day of Intralipid for 5-7 weeks. Lipid accumulation was almost exclusively found intravascularly in the lungs and liver, and not in macrophages obtained from bronchoalveolar lavage fluid. A morphometric study of the lung capillaries showed that the macrophages in TPN animals had increased in size and number, and occupied a larger portion of the capillary lumina. The macrophages appeared to be activated, but the endothelial lining was well preserved. Free intravascular lipid droplets had a diameter both in vitro and in vivo of about 0.5 micron, indicating good stability of the emulsion. We suggest that the lipid uptake stimulates the macrophages and thereby plays a role in the lung tissue inflammation seen in response to long-term lipid-based TPN in pigs.

Glycoprotein IIb-IIIa on platelet-derived microparticles, and microparticle structures studied by electron microscopy, confocal laser microscopy and crossed radio-immunoelectrophoresis.

Shedding of microparticles from the platelet surface is usually associated with exposure of platelet procoagulant activity. Platelet-derived microparticles have been detected in blood in various disease states. In vitro, platelet stimulation with a number of different agonists results in formation of microparticles. In the present study, microparticles induced by platelet stimulation by calcium ionophore or by membrane incorporation of the terminal complement complex C5b-9 were studied using electron microscopy, confocal laser microscopy, flow cytometry and radio-immunoelectrophoresis. When studied by electron microscopy, microparticle morphology was found to be dependent upon the induction method. Platelet stimulation with the calcium ionophore resulted in smaller, more homogeneous and electron dense microparticles than those induced by insertion of the terminal complement complex. With flow cytometry and confocal laser immunofluorescence microscopy, microparticle GPIIb-IIIa was demonstrated using a FITC-conjugated antibody to GPIIIa. Surface-bound GPIIb-IIIa was demonstrated on the microparticles by immunoelectron microscopy. Crossed immunoelectrophoresis of detergent-solubilized microparticles visualized a very prominent GPIIb-IIIa immunoprecipitate arc, and binding of [(125)1]fibrinogen to microparticle GPIIb-IIIa was demonstrated by radio-immunoelectrophoresis. This suggests that the activated GPIIb-IIIa complex is preserved intact during the shedding of microparticles from the platelet surface.

Studies on the haemostatic defect in a complicated syndrome. An inverse Scott syndrome platelet membrane abnormality?

The Stormorken syndrome is a multifacetted syndrome including a bleeding tendency. No deviations were found in the coagulation- or fibrinolytic systems. Platelet number was low normal, and size abnormal, whereas EM findings were unremarkable. Survival time was half normal. Clot retraction was initially rapid, but clearly decreased, whereas prothrombin consumption was also initially rapid, but complete. Membrane GP's were normal, so was AA metabolism, PI-cycle, granule storage and secretion, and c-AMP function, whereas 5-HT uptake and storage was decreased. Optical platelet aggregation was low normal with all physiological agonists. The only clearly abnormal finding was that coagulant activity was present on non stimulated platelets at the same level as kaolin-stimulated normal platelets. This indicated a platelet abnormality which should lead to a thrombogenic, not to a haemorrhagic trait. This paradox may have its origin in rheology, because when challenged with in vivo shear rates in an ex vivo perfusion chamber, platelet cohesion was abnormally low. Further studies to better delineate the membrane abnormality are underway.