Effect of parenteral antibiotic administration on the establishment of colonization with vancomycin-resistant Enterococcus faecium in the mouse gastrointestinal tract.

A mouse model of intestinal colonization with vancomycin-resistant enterococci (VRE) was used to study the effect of different beta-lactam antibiotics on establishment of VRE colonization. A clinical VanB VRE isolate, Enterococcus faecium C68 (102 or 104 cfu), was inoculated by gastric gavage in conjunction with subcutaneous administration of antibiotics. The MIC of ceftriaxone and ticarcillin against VRE strain C68 is >10,000 microg/mL, and the MIC of piperacillin is 1250 microg/mL. Ceftriaxone and ticarcillin-clavulanate treatment groups developed persistently high levels of stool VRE compared with both the saline and the piperacillin-tazobactam (Pip-Taz) groups (P<.008). The level of stool VRE in the Pip-Taz group did not differ from that for the saline group. Thus, in this mouse model, beta-lactam antibiotics with minimal anti-enterococcal activity promoted establishment of high-level VRE colonization, but Pip-Taz (a beta-lactam antibiotic with more potent activity against VRE) did not.

Disseminated intravascular coagulation in liver cirrhosis: fact or fiction?

OBJECTIVE: Cirrhosis is commonly associated with haemostatic dysfunction. The similarities of laboratory tests of disseminated intravascular coagulation (DIC) to those found in cirrhosis has led to the belief that DIC is a feature of the haemostatic failure of cirrhosis. METHODS: The aim of this study was to determine whether DIC is part of the coagulopathy of cirrhosis by applying quantitative tests for prothrombin fragment 1 + 2, antithrombin III, thrombin-antithrombin complex, and specific fribrinogen degradation products levels (XDP), as well as the thrombelastograph for detecting the Clot Lysis Index. RESULTS: Fifty-two stable cirrhotic patients (33 men, 19 women; mean age, 58.8 yr; range, 24-72 yr) with differing etiologies were studied. On tests of thrombin generation: thrombin-antithrombin complexes, fibrin(ogen) degradation products, and prothrombin fragments 1 + 2 were not found to be significantly different from an age- and gender-matched control group (p = 0.18, 0.3, and 0.67, respectively), whereas albumin, Factor V, fibrinogen, antithrombin III, and alpha2-antiplasmin were all significantly low (p = 0.0004, 0.002, 0.06, 0.004, and 0.004, respectively), reflecting reduced synthetic function and correlation in ascitic and non-ascitic patients. There was no correlation between impaired synthesis (antithrombin III and alpha2-antiplasmin) and indices of DIC (prothrombin fragment 1 + 2, thrombin-antithrombin complexes, and XDP) (p = not significant). The percentage of patients with high prothrombin fragments 1 + 2 and thrombin antithrombin levels in each Child grade group was similar. Thrombin time was significantly elevated in the cirrhotic group (a manifestation of low fibrinogen levels). The Clot Lysis Index as measured by thrombelastography was significantly abnormal, indicating mild hyperfibrinolysis. CONCLUSION: We conclude that DIC is not part of the coagulopathy in stable liver cirrhosis without recent complications.

Effect of parenteral antibiotic administration on persistence of vancomycin-resistant Enterococcus faecium in the mouse gastrointestinal tract.

A mouse model of vancomycin-resistant Enterococcus faecium (VRE) intestinal colonization was used to study the effect of different subcutaneous antibiotics on persistence and density of VRE colonization. Gastric inoculation of a clinical VanB VRE isolate, in conjunction with oral vancomycin in drinking water (250 microgram/mL), resulted in high-level VRE colonization (mean, 9.5 log10 cfu/g) in all 169 experimental mice. After discontinuation of oral vancomycin, the level of VRE in the stool specimens of mice receiving subcutaneous saline steadily decreased (mean, 3.59 log10 cfu/g at day 19). Subcutaneous vancomycin, clindamycin, piperacillin-tazobactam, ticarcillin-clavulanic acid, metronidazole, cefotetan, ampicillin, and ampicillin-sulbactam all promoted persistent high levels of stool VRE. Subcutaneous ceftriaxone, cefepime, ciprofloxacin, and aztreonam promoted increased VRE density to a lesser degree or not at all. Thus, in a mouse model, vancomycin and antibiotics with potent antianaerobic activity promoted persistent high-density intestinal VRE colonization, whereas antibiotics lacking potent antianaerobic activity did not.

Thrombotic vascular risk factors in inflammatory bowel disease.

BACKGROUND: Thrombosis may be an important effector mechanism in the pathogenesis of Crohn's disease. METHODS: This study therefore investigated the prevalence of independent thrombotic risk factors (factor VII coagulant activity, lipoprotein (a), fibrinogen, plasma triglycerides, and smoking) in patients with Crohn's disease, ulcerative colitis, and normal controls. RESULTS: In Crohn's disease (n = 75), the mean plasma VII:C, lipoprotein (a) and fibrinogen concentrations were significantly greater than in the normal population (n = 85). In ulcerative colitis (n = 35), only the mean factor VII:C concentration was significantly higher than normal. Ninety three per cent of patients with Crohn's disease and 86% of those with ulcerative colitis had at least one risk factor for thrombotic vascular disease, compared with 61% of the normal population (p < 0.001). CONCLUSIONS: In many young patients with inflammatory bowel disease, plasma concentrations of these prothrombotic factors were in excess of the limits that are regarded as posing an increased risk for the development of occlusive vascular disease.

Neurotoxicity, anticoagulant activity and evidence of rhabdomyolysis in patients bitten by death adders (Acanthophis sp.) in southern Papua New Guinea.

Thirty-two patients with enzyme-immunoassay-proven death adder (Acanthophis sp.) bites were studied in Port Moresby, Papua New Guinea. Eighteen were envenomed; local signs were rare and none had incoagulable blood, but all except one had signs of neurotoxicity. Five (27.7%) envenomed patients required intubation and ventilation. One patient developed renal failure, previously undescribed following death adder bites. Laboratory investigations showed mild prolongation of prothrombin and partial thromboplastin times in some patients. In vitro studies showed that the venom contains anticoagulant activity, but does not cause fibrinogenolysis. In contrast to taipan envenoming, neurotoxicity did not progress after antivenom administration, and there was reversal of neurotoxicity, evident within 6 h, in three severely envenomed patients treated less than 12 h after the bite. One patient treated with antivenom and anticholinesterases had the most dramatic response to treatment; the optimum management of bites by this species may include prompt treatment with both antivenom and anticholinesterases in addition to effective first aid.

Coagulopathy following bites by the Papuan taipan (Oxyuranus scutellatus canni).

The mechanisms of haemostatic failure were studied in 87 patients bitten by the Papuan taipan (Oxyuranus scutellatus canni). Eighty (92%) had evidence of a coagulopathy on laboratory testing; 36 (41.4%) developed spontaneous systemic bleeding, although this was rarely of clinical significance. Coagulation assays in 48 completely defibrinated patients showed marked reductions in factors V and VIII and reductions in factors II, IX, XI, XII and XIIIA. There was a reduction in plasminogen and alpha 2-antiplasmin levels and both total and cross-linked fibrin(ogen) degradation products (FDP) levels were elevated. The mean platelet count was initially decreased and fell further during admission. Similar but less severe changes were seen in patients who were mildly defibrinated. Following treatment with antivenom, fibrinogen levels rose rapidly and coagulability was restored within 6-12 h in 93% of patients. These abnormalities may be primarily attributable to the prothrombin activator present in taipan venom, but it is likely that other uncharacterized venom components contributed.

Aspirin does not affect the flow cytometric detection of fibrinogen binding to, or release of alpha-granules or lysosomes from, human platelets.

1. Aspirin inhibits the conversion of arachidonic acid to thromboxane A2 which reinforces the effects of weak agonists such as ADP in platelets. 2. In this study the effect of aspirin (300 mg/day) on platelet agonist response was measured by whole blood flow cytometry of unfixed blood samples from normal subjects (n = 10), an assay that investigates aggregation-independent changes in the platelet. 3. Fibrinogen binding to unstimulated platelets or to platelets stimulated with ADP or thrombin was unaffected by aspirin. 4. Under the conditions of this assay, platelets undergo a partial degranulation of alpha-granules and lysosomes (evidenced by expression of P-selectin and CD63, respectively) in response to ADP, and full degranulation in response to thrombin. P-selectin expression was paralleled by release of beta-thromboglobulin. None of these events was affected by aspirin. 5. Thromboxane formation was totally prevented by the aspirin treatment, as shown by Born aggregometry in which the platelet aggregatory response to arachidonic acid was abolished and secondary aggregation by ADP was inhibited. 6. The flow cytometric assay can therefore be used to investigate platelets in patients, regardless of aspirin therapy. 7. These findings suggest that platelet fibrinogen binding and the release of platelet alpha-granule and lysosomal contents, in response to stimulation with physiological agonists, can continue in patients despite aspirin therapy. This may help to explain why aspirin is only partially effective in preventing thrombotic events.

Haemostatic dysfunction and acute renal failure following envenoming by Merrem's hump-nosed viper (Hypnale hypnale) in Sri Lanka: first authenticated case.

A five years old boy was bitten by a Merrem's hump-nosed viper (Hypnale hypnale) in Central Province, Sri Lanka. He developed local swelling, incoagulable blood, thrombocytopenia, bleeding into the gastrointestinal tract, and acute renal failure. Treatment with Serum Institute of Indian polyspecific antivenom (specific for venoms of cobra, common krait, Russell's viper and saw-scaled viper) had no effect on the coagulopathy, which persisted for more than a week. The boy recovered after 27 d in hospital, during which he was treated with peritoneal dialysis for renal failure. Laboratory studies demonstrated that the venom of H. hypnale was procoagulant, fibrinolytic and aggregated platelets. This first authenticated case of life-threatening acute renal failure and haemostatic disturbances caused by H. hypnale, a species responsible for 27% of snake bites in Sri Lanka, demonstrates the need for a new antivenom with specific activity against the venom of this species.

Action of snake venom components on the haemostatic system.

Among the components in snake venom are a number which have profound effects (either stimulatory or inhibitory) on haemostatic mechanisms, including coagulation, fibrinolysis, platelet function and vascular integrity. As a consequence, human victims of snakebite may suffer severe and sometimes fatal haemorrhagic and/or thrombotic sequelae. Many of these venom components have been isolated and their precise mechanisms of action established. Apart from direct fibrinolysins, procoagulants predominate, most of these exerting their effect late in the clotting cascade, activating factor X or prothrombin or directly converting fibrinogen to fibrin. Some of the procoagulants are, or have the potential to be, used as therapeutic agents. Some venom components have been put to use as laboratory reagents for diagnostic purposes or for characterising molecular defects of haemostasis, although because they often have unphysiological actions, results must be interpreted with caution. These and other useful constituents e.g. protein C activator and platelet aggregating agents are discussed.

Factor XIIIA subunit and Crohn's disease.

Factor XIIIA is the active subunit of plasma factor XIII that is responsible for cross linking fibrin into a stable clot. Sixteen patients with Crohn's disease were studied prospectively from relapse (Crohn's disease activity index > 150) into remission. Plasma factor XIIIA concentrations were significantly lower in active disease (median 63 (95% CI 46-72) U/dl) than remission (median 90 (95% CI 60-112) U/dl; p = 0.002). Plasma factor XIIIA concentrations correlated positively with the activity index (p = 0.005) and platelet count (p = 0.003), and negatively with serum albumin (p = 0.006). In five patients with persistent aggressive disease, the factor XIIIA concentration remained below the lower range of normal despite apparent clinical improvement in response to medical treatment. Tissues from three patients who underwent surgical resection during the study were immunostained for factor XIIIA. Gut mucosal and submucosal macrophages stained strongly for factor XIIIA. In one patient, capillary thrombi near superficial mucosal erosions immunostained for factor XIIIA in macroscopically normal mucosa. Similar changes were identified in more severely inflamed sections of intestine from the other two patients. The demonstration of significantly low plasma factor XIIIA concentrations in active Crohn's disease, and the immunostaining of factor XIIIA in capillary thrombi in the bowel wall, suggest that activation of coagulation may be involved in the pathogenesis of Crohn's disease. The plasma factor XIIIA concentration may prove a useful laboratory marker of disease activity.

Evidence for activation of coagulation in Crohn's disease.

Haemostatic changes in 16 patients with Crohn's disease were studied from active disease into clinical remission and beyond. Elevated concentrations of fibrinopeptide A (FpA) and prothrombin fragments F1 + 2 (F1 + 2) were found at times of both active (FpA median 3.2, range [0.3-40] ng/ml and F1 + 2 median 2.3, range [0.3-18] nm/l) and inactive disease (FpA median 2, range [0.4-40] ng/ml and F1 + 2 median 1.3, range [0.2-20) nm/l]. We also measured the physiological inhibitors of coagulation and fibrinolysis; there was no significant difference in the levels of antithrombin III, protein C or the Exner ratio between active and inactive disease. Free protein S levels were significantly lower in active disease (median 34, range 9-54 U/dl) than in remission (median 40, range 12-65 U/dl). Plasminogen activator inhibitor type 1 (PAI-1) was significantly raised in remission (median 11, range 3-32 ng/ml) when compared to active disease (median 7, range 3-42 ng/ml). The D-dimer correlated significantly with fibrinopeptide A (P < 0.001), suggesting reactive fibrinolysis in some patients. Most (35/52, 67%) samples showed evidence of persistent haemostatic activation (elevated FpA and/or F1 + 2) during phases of apparent clinical remission in Crohn's disease, a factor that is not reflected by clinical activity scores. This study supports the hypothesis that coagulation is activated in the mesenteric vasculature of patients with Crohn's disease.

Development and evaluation of ELISAs for factor XIIIA and XIIIB subunits in plasma.

Severe, congenital deficiency of factor XIII is extremely rare. However, a moderate reduction in the plasma level of the functional subunit (factor XIIIA) and also to a lesser extent of the carrier subunit (factor XIIIB), and a decrease in the XIIIA:B subunit ratio, have recently been reported in patients with the inflammatory bowel disorder Crohn's disease, particularly during clinical relapse. In order to accurately monitor patients, sensitive, reliable assays for the two subunits of factor XIII are required. We report here the development and validation of ELISAs for these components. The assays are identical except in respect of the specificity of the polyclonal antiserum used as starting material, both of which are commercially available. The antisera are purified by n-octanoic acid precipitation and portions of these purified immunoglobulins are used as coating antibodies. The remaining portions are biotinylated and used with streptavidin and horse-radish peroxidase as tracer antibodies. A normal range (n = 24) was established for factor XIIIA (mean 95 range 60-130 U/dl) and for factor XIIIB (mean 99 range 60-130 U/dl). There were no significant differences between the ELISA and electroimmunodiffusion assays either for factor XIIIA (means +/- 1 standard deviation 95 +/- 15.9 and 89 +/- 22.7 respectively) or for factor XIIIB (99 +/- 18.3 and 106 +/- 23.4 respectively). These assays have been in routine use for six months, during which time two further antisera purifications and biotinylations have been carried out without significant problems of reproducibility or stability.

Rediscovery and redefinition of Malcolm Smith's Trimeresurus kanburiensis in Thailand, with a report of envenoming.

Three specimens of an apparently rare pit viper, Trimeresurus kanburiensis, previously known only from the holotype collected in 1928, were found near Kanchanaburi in western Thailand. One of the snakes had bitten a young woman on the foot. She experienced severe local pain, swelling that involved the whole of the bitten limb and beyond, local bruising, recurrent shock, peripheral leucocytosis and a mild coagulopathy, but she recovered despite the lack of specific treatment. The severity of envenoming augurs ill for a young or debilitated patient bitten by this species. During the last 20 years, the name T. kanburiensis has been used incorrectly for the medically more important species, T. purpureomaculatus. Conversely, the name T. purpureomaculatus has been misapplied to specimens of a species of viper from southern Thailand which we consider very similar to T. kanburiensis, but for which a new specific name, 'T. venustus', has been suggested recently. The rediscovery and redefinition of T. kanburiensis should prevent further confusion.

Systemic haemorrhage in rats induced by a haemorrhagic fraction from Bothrops jararaca venom.

The main haemorrhagic fraction of Bothrops jararaca venom, showing in vitro fibrinogenolytic activity and an inhibitory effect on platelets aggregation induced by collagan, was studied in rats. Development of coagulopathy and/or haemorrhage was studied 2 hr after s.c. injection of batroxobin, B. jararaca whole venom and its haemorrhagic fraction. Incoagulable blood, together with low fibrinogen levels, were found only in rats injected with batroxobin and whole venom; thrombocytopenia alone was detected in rats given s.c. injections of haemorrhagin. Intravenous injection of low doses of haemorrhagin (less than 15 micrograms) resulted in significant thrombocytopenia, without any alterations in the blood coagulation mechanism. Severe damage to the vascular endothelium and skeletal muscle following s.c. injection of haemorrhagin together with signs of systemic haemorrhage in the kidneys, lungs and liver occurred. Levels of factor VIII and von Willebrand factor antigen were within the normal range in all animals. Serum levels of both whole venom and haemorrhagin were significantly correlated. This study confirms that B. jararaca haemorrhagin plays a vital role in systemic bleeding.

Coagulopathy and haemorrhage in human victims of Bothrops jararaca envenoming in Brazil.

Thirty-four patients envenomed by Bothrops jararaca in Brazil were studied. Of these, 20 (59%) had incoagulable blood associated with local and/or systemic bleeding and 10 of the 20 were thrombocytopenic. Among 14 patients with coagulable blood, 6 (43%) had bleeding symptoms and 3 (21%) were thrombocytopenic. High levels of von Willebrand factor (vWF), plasminogen activator inhibitor type 1 (PAI-1) and tissue type plasminogen activator (t-PA) antigens were also recorded in some patients with systemic bleeding with or without incoagulable blood. These substances may have been released from endothelial cells. Admission serum venom antigen levels were similar in both groups. The study indicated that systemic haemorrhage may occur in patients with coagulable blood and thrombocytopenia and that coagulopathy is not therefore the primary cause of haemorrhage.

Arboreal green pit vipers (genus Trimeresurus) of South-East Asia: bites by T. albolabris and T. macrops in Thailand and a review of the literature.

In Thailand 29 patients were proved to have been bitten by arboreal green pit vipers: 24 by Trimeresurus albolabris and 5 by T. macrops. They were studied in order to define the clinical effects of envenoming, to characterize the haemostatic abnormalities and assess the efficacy of Thai Red Cross antivenom. T. macrops caused only local painful swelling, neutrophil leucocytosis and thrombocytopenia. T. albolabris caused more severe envenoming with local blistering and necrosis, shock, spontaneous systemic bleeding, defibrination, thrombocytopenia and leucocytosis. There was no evidence of disseminated intravascular coagulation, but fibrinolytic activity was increased. Platelet function was normal. The product of admission venom antigen concentration and the delay between bite and admission was significantly higher in defibrinated patients than in those without severe coagulopathy. Antivenom (5 ampoules intravenously) restored blood coagulability, but there was persistent venom antigenaemia, associated in some cases with recurrent coagulopathy. The literature on bites by south Asian green pit vipers of the genus Trimeresurus is reviewed; these bites are common medical problems and causes of morbidity. The identification of individual species is difficult, but may be important if antivenom is to be improved and used appropriately.

Differences between the venoms of two sub-species of Russell's viper: Vipera russelli pulchella and Vipera russelli siamensis.

Ion exchange chromatography was carried out using venoms obtained from two sub-species of Russell's viper; V. russelli siamensis from Burma and V. russelli pulchella from Sri Lanka. Differences were observed in the elution position of venom components having haemolytic and procoagulant activity but not those causing fibrinolysis. Only the V. russelli siamensis venom exhibited any platelet aggregating activity. The Indian (Haffkine) polyspecific and the Burmese (Burma Pharmaceutical Industries) monospecific antivenoms, when used in cross immunoelectrophoresis against the two venoms, revealed differences in the number and/or intensity of the precipitin bands present. An important functional consequence of this was that the Burmese antivenom did not neutralize the haemolytic activity of the V. russelli pulchella venom in vitro and would thus probably not be effective in treating this consequence of envenoming by Russell's viper in Sri Lanka. Differences in the composition and the clinical effects of the two venoms emphasizes the importance of using venom from the local snake for antivenom production if optimal clinical efficacy is to be achieved.

Changes in protein C and free protein S during pregnancy and following hysterectomy.

Two longitudinal studies were undertaken to measure the naturally occurring coagulation inhibitors, protein C and protein S, in females who were either pregnant or who were undergoing hysterectomy. Functional and immunological protein C and free protein S were assayed. During pregnancy, protein C levels remained unchanged except for a small increase in protein C antigen at 28-32 weeks gestation. The free protein S fell significantly and progressively during pregnancy, although only in about one-third of patients did the level fall below the normal range. In the hysterectomy study, a significant fall in protein C occurred on days one and three after surgery but had returned to normal by the time of discharge (days 7-10). A small rise in free protein S was observed at time of discharge, but this factor was otherwise unchanged. We conclude that the changes observed may contribute to the hypercoagulable state which is associated with pregnancy or major surgery.