Fitting of Mixtures with Unspecified Number of Components Using Cross Validation Distance Estimate

Estimation of the number of mixture components (k) is an unsolved problem. Available methods for estimation of k include bootstrapping the likelihood ratio test statistics and optimizing a variety of validity functionals such as AIC, BIC/MDL, and ICOMP. We investigate the minimization of distance between fitted mixture model and the true density as a method for estimating k. The distances considered are Kullback-Leibler (KL) and “L sub 2”. We estimate these distances using cross validation. A reliable estimate of k is obtained by voting of B estimates of k corresponding to B cross validation estimates of distance. This estimation methods with KL distance is very similar to Monte Carlo cross validated likelihood methods discussed by Smyth (2000). With focus on univariate normal mixtures, we present simulation studies that compare the cross validated distance method with AIC, BIC/MDL, and ICOMP. We also apply the cross validation estimate of distance approach along with AIC, BIC/MDL and ICOMP approach, to data from an osteoporosis drug trial in order to find groups that differentially respond to treatment.

Fitting ACE Structural Equation Models to Case-Control Family Data

Investigators interested in whether a disease aggregates in families often collect case-control family data, which consist of disease status and covariate information for families selected via case or control probands. Here, we focus on the use of case-control family data to investigate the relative contributions to the disease of additive genetic effects (A), shared family environment (C), and unique environment (E). To this end, we describe a ACE model for binary family data and then introduce an approach to fitting the model to case-control family data. The structural equation model, which has been described previously, combines a general-family extension of the classic ACE twin model with a (possibly covariate-specific) liability-threshold model for binary outcomes. Our likelihood-based approach to fitting involves conditioning on the proband’s disease status, as well as setting prevalence equal to a pre-specified value that can be estimated from the data themselves if necessary. Simulation experiments suggest that our approach to fitting yields approximately unbiased estimates of the A, C, and E variance components, provided that certain commonly-made assumptions hold. These assumptions include: the usual assumptions for the classic ACE and liability-threshold models; assumptions about shared family environment for relative pairs; and assumptions about the case-control family sampling, including single ascertainment. When our approach is used to fit the ACE model to Austrian case-control family data on depression, the resulting estimate of heritability is very similar to those from previous analyses of twin data.

Comparison of lactated Ringer's, gelatine and blood resuscitation on intestinal oxygen supply and mucosal tissue oxygen tension in haemorrhagic shock

OBJECTIVES: To evaluate the effects on intestinal oxygen supply, and mucosal tissue oxygen tension during haemorrhage and after fluid resuscitation with either blood (B; n=7), gelatine (G; n=8), or lactated Ringer's solution (R; n=8) in an autoperfused, innervated jejunal segment in anaesthetized pigs. METHODS: To induce haemorrhagic shock, 50% of calculated blood volume was withdrawn. Systemic haemodynamics, mesenteric venous and systemic acid-base and blood gas variables, and lactate measurements were recorded. A flowmeter was used for measuring mesenteric arterial blood flow. Mucosal tissue oxygen tension (PO(2)muc), jejunal microvascular haemoglobin oxygen saturation (HbO(2)) and microvascular blood flow were measured. Measurements were performed at baseline, after haemorrhage and at four 20 min intervals after fluid resuscitation. After haemorrhage, animals were retransfused with blood, gelatine or lactated Ringer's solution until baseline pulmonary capillary wedge pressure was reached. RESULTS: After resuscitation, no significant differences in macrohaemodynamic parameters were observed between groups. Systemic and intestinal lactate concentration was significantly increased in animals receiving lactated Ringer's solution [5.6 (1.1) vs 3.3 (1.1) mmol litre(-1); 5.6 (1.1) vs 3.3 (1.2) mmol litre(-1)]. Oxygen supply to the intestine was impaired in animals receiving lactated Ringer's solution when compared with animals receiving blood. Blood and gelatine resuscitation resulted in higher HbO(2) than with lactated Ringer's resuscitation after haemorrhagic shock [B, 43.8 (10.4)%; G, 34.6 (9.4)%; R, 28.0 (9.3)%]. PO(2)muc was better preserved with gelatine resuscitation when compared with lactated Ringer's or blood resuscitation [20.0 (8.8) vs 13.8 (7.1) mm Hg, 15.2 (7.2) mm Hg, respectively]. CONCLUSION: Blood or gelatine infusion improves mucosal tissue oxygenation of the porcine jejunum after severe haemorrhage when compared with lactated Ringer's solution.

A randomized, blinded, multicenter trial of lipid-associated amphotericin B alone versus in combination with an antibody-based inhibitor of heat shock protein 90 in patients with invasive candidiasis

BACKGROUND: Mycograb (NeuTec Pharma) is a human recombinant monoclonal antibody against heat shock protein 90 that, in laboratory studies, was revealed to have synergy with amphotericin B against a broad spectrum of Candida species. METHODS: A double-blind, randomized study was conducted to determine whether lipid-associated amphotericin B plus Mycograb was superior to amphotericin B plus placebo in patients with culture-confirmed invasive candidiasis. Patients received a lipid-associated formulation of amphotericin B plus a 5-day course of Mycograb or placebo, having been stratified on the basis of Candida species (Candida albicans vs. non-albicans species of Candida). Inclusion criteria included clinical evidence of active infection at trial entry plus growth of Candida species on culture of a specimen from a clinically significant site within 3 days after initiation of study treatment. The primary efficacy variable was overall response to treatment (clinical and mycological resolution) by day 10. RESULTS: Of the 139 patients enrolled from Europe and the United States, 117 were included in the modified intention-to-treat population. A complete overall response by day 10 was obtained for 29 (48%) of 61 patients in the amphotericin B group, compared with 47 (84%) of 56 patients in the Mycograb combination therapy group (odds ratio [OR], 5.8; 95% confidence interval [CI], 2.41-13.79; P<.001). The following efficacy criteria were also met: clinical response (52% vs. 86%; OR, 5.4; 95% CI, 2.21-13.39; P<.001), mycological response (54% vs. 89%; OR, 7.1; 95% CI, 2.64-18.94; P<.001), Candida-attributable mortality (18% vs. 4%; OR, 0.2; 95% CI, 0.04-0.80; P = .025), and rate of culture-confirmed clearance of the infection (hazard ratio, 2.3; 95% CI, 1.4-3.8; P = .001). Mycograb was well tolerated. CONCLUSIONS: Mycograb plus lipid-associated amphotericin B produced significant clinical and culture-confirmed improvement in outcome for patients with invasive candidiasis.

Early resuscitation with polymerized bovine hemoglobin reverses acidosis, but not peripheral tissue oxygenation, in a severe hamster shock model

Awake hamsters equipped with the dorsal window chamber preparation were subjected to hemorrhage of 50% of the estimated blood volume. Initial resuscitation (25% of estimated blood volume) with polymerized bovine hemoglobin (PBH) or 10% hydroxyethyl starch (HES) occurred in concert with an equivolumetric bleeding to simulate the early, prehospital setting (exchange transfusion). Resuscitation (25% of estimated blood volume) without bleeding was performed with PBH, HES, or autologous red blood cells (HES-RBCs). Peripheral microcirculation, tissue oxygenation, and systemic hemodynamic and blood gas parameters were assessed. After exchange transfusion, base deficit was -8.6 +/- 3.7 mmol/L (PBH) and -5.1 +/- 5.3 mmol/L (HES) (not significant). Functional capillary density was 17% +/- 6% of baseline (PBH) and 31% +/- 11% (HES) (P < 0.05) and arteriolar diameter 73% +/- 3% of baseline (PBH) and 90% + 5% (HES) (P < 0.01). At the end, hemoglobin levels were 3.7 +/- 0.3 g/dL with HES, 8.2 +/- 0.6 g/dL with PBH, and 10.4 +/- 0.8 g/dL with HES-RBCs (P < 0.01 HES vs. PBH and HES-RBCs, P < 0.05 PBH vs. HES-RBCs). Base excess was restored to baseline with PBH and HES-RBCs, but not with HES (P < 0.05). Functional capillary density was 46% +/- 5% of baseline (PBH), 62% + 20% (HES-RBCs), and 36% +/- 19% (HES) (P < 0.01 HES-RBCs vs. HES). Peripheral oxygen delivery and consumption was highest with HES-RBCs, followed by PBH (P < 0.05 HES-RBCs vs. PBH, P < 0.01 HES-RBCs and PBH vs. HES). In conclusion, the PBH led to a correction of base deficit comparable to blood transfusion. However, oxygenation of the peripheral tissue was inferior with PBH. This was attributed to its negative impact on the peripheral microcirculation caused by arteriolar vasoconstriction.

Rapid restoration of microcirculatory blood flow with hyperviscous and hyperoncotic solutions lowers the transfusion trigger in resuscitation from hemorrhagic shock

Resuscitation from hemorrhagic shock relies on fluid retransfusion. However, the optimal properties of the fluid have not been established. The aim of the present study was to test the influence of the concentration of hydroxyethyl starch (HES) solution on plasma viscosity and colloid osmotic pressure (COP), systemic and microcirculatory recovery, and oxygen delivery and consumption after resuscitation, which were assessed in the hamster chamber window preparation by intravital microscopy. Awake hamsters were subjected to 50% hemorrhage and were resuscitated with 25% of the estimated blood volume with 5%, 10%, or 20% HES solution. The increase in concentration led to an increase in COP (from 20 to 70 and 194 mmHg) and viscosity (from 1.7 to 3.8 and 14.4 cP). Cardiac index and microcirculatory and metabolic recovery were improved with HES 10% and 20% when compared with 5% HES. Oxygen delivery and consumption in the dorsal skinfold chamber was more than doubled with HES 10% and 20% when compared with HES 5%. This was attributed to the beneficial effect of restored or increased plasma COP and plasma viscosity as obtained with HES 10% and 20%, leading to improved microcirculatory blood flow values early in the resuscitation period. The increase in COP led to an increase in blood volume as shown by a reduction in hematocrit. Mean arterial pressure was significantly improved in animals receiving 10% and 20% solutions. In conclusion, the present results show that the increase in the concentration of HES, leading to hyperoncotic and hyperviscous solutions, is beneficial for resuscitation from hemorrhagic shock because normalization of COP and viscosity led to a rapid recovery of microcirculatory parameters.

NONLINEAR TUBE-FITTING FOR THE ANALYSIS OF ANATOMICAL AND FUNCTIONAL STRUCTURES

We are concerned with the estimation of the exterior surface of tube-shaped anatomical structures. This interest is motivated by two distinct scientific goals, one dealing with the distribution of HIV microbicide in the colon and the other with measuring degradation in white-matter tracts in the brain. Our problem is posed as the estimation of the support of a distribution in three dimensions from a sample from that distribution, possibly measured with error. We propose a novel tube-fitting algorithm to construct such estimators. Further, we conduct a simulation study to aid in the choice of a key parameter of the algorithm, and we test our algorithm with validation study tailored to the motivating data sets. Finally, we apply the tube-fitting algorithm to a colon image produced by single photon emission computed tomography (SPECT)and to a white-matter tract image produced using diffusion tensor `imaging (DTI).

A case of necrotizing fasciitis with septic shock in a cat caused by Acinetobacter baumannii

A 4-year-old, neutered female, domestic shorthair cat admitted to the animal hospital for recurrent constipation presumed to be due to post-traumatic injuries, went into shock with signs including fever and ataxia followed by stupor. On the fifth day of hospitalization, the cat developed severe, diffuse oedema of the ventral abdomen with multifocal to coalescing erythematous areas and small vesicle formation. The results of bacteriological cultures of liver, spleen and kidney specimens led to the diagnosis of Acinetobacter baumannii sepsis. Histopathological findings of skin samples taken during necropsy showed an extensive epidermal and dermal necrosis with septic vasculitis and numerous intralesional gram-negative bacteria. Detection of the bla(OXA-51-like) gene specific for A. baumannii by PCR, performed retrospectively on samples of the deep layers of the skin, confirmed the presence of A. baumannii also in the cutaneous lesions. To our knowledge this is the first report of a necrotizing fasciitis with septic shock in a cat caused by A. baumannii.

ADAMTS-13, von Willebrand factor and related parameters in severe sepsis and septic shock

BACKGROUND: Insufficient control of von Willebrand factor (VWF) multimer size as a result of severely deficient ADAMTS-13 activity results in thrombotic thrombocytopenic purpura associated with microvascluar thrombosis and platelet consumption, features not seldom seen in severe sepsis and septic shock. METHODS: ADAMTS-13 activity and VWF parameters of 40 patients with severe sepsis or septic shock were compared with those of 40 healthy controls of the same age and gender and correlated with clinical findings and sepsis outcome. RESULTS: ADAMTS-13 activity was significantly lower in patients than in healthy controls [median 60% (range 27-160%) vs. 110% (range 63-200%); P < 0.001]. VWF parameters behaved reciprocally and both VWF ristocetin cofactor activity (RCo) and VWF antigen (VWF:Ag) were significantly (P < 0.001) higher in patients compared with controls. Neither ADAMTS-13 activity nor VWF parameters correlated with disease severity, organ dysfunction or outcome. However, a contribution of acute endothelial dysfunction to renal impairment in sepsis is suggested by the significantly higher VWF propeptide and soluble thrombomodulin levels in patients with increased creatinine values as well as by their strong positive correlations (creatinine and VWF propeptide r(s) = 0.484, P < 0.001; creatinine and soluble thrombomodulin r(s) = 0.596, P < 0.001). CONCLUSIONS: VWF parameters are reciprocally correlated with ADAMTS-13 activity in severe sepsis and septic shock but have no prognostic value regarding outcome.

Factor XIII in severe sepsis and septic shock

OBJECTIVE: In sepsis, activation of coagulation and inhibition of fibrinolysis lead to microvascular thrombosis. Thus, clot stability might be a critical issue in the development of multiple organ dysfunction syndrome. Activated FXIII (FXIIIa) forms stable fibrin clots by covalently cross-linking fibrin monomers. Therefore, we investigated the impact of FXIII antigen and activity levels on disease severity and fatality in sepsis patients. PATIENTS AND METHODS: FXIII subunit A (FXIIIA) and FXIII cross-linking activity (FXIIICA) were measured in 151 controls, in 32 patients with severe sepsis and 8 with septic shock. In addition, FXIII subunit B (FXIIIB) was measured in the sepsis patients. Moreover, clotting parameters were determined. RESULTS: Patients suffering from sepsis (n=40) had significantly (p<0.005) lower FXIIIA levels (median [range]: 36.5% [8.8-127.4%]) and FXIIICA levels (76.5% [9.4-266%]) as compared to healthy controls (n=151, 119% [31.3-283.2] and 122.4% [40.6-485.3], respectively). No difference in FXIIIA, FXIIIB and FXIIICA levels between survivors and non-survivors, nor between patients with severe sepsis and septic shock was found. The specific activity of FXIII (FXIIICA/FXIIIA, SA(FXIII)) was significantly (p<0.001) higher in sepsis patients (2.0 [0.8-5.3]) as compared to healthy controls (1.0 [0.4-5.1]). SA(FXIII) significantly (p<0.05) increased with fatality (non-survivors [n=13] vs. survivors [n=27]: 3.3 [1.2-5.0] vs. 1.9 [0.8-5.3]) and disease severity (septic shock vs. severe sepsis: 3.4 [1.8-4.3] vs. 1.9 [0.8-5.3]). CONCLUSION: We show decreased FXIIICA and FXIIIA levels, but higher SA(FXIII) in sepsis as compared to controls. Increased SA(FXIII) correlates with disease severity and fatality in sepsis patients.

Vasopressin in septic shock: effects on pancreatic, renal, and hepatic blood flow

INTRODUCTION: Vasopressin has been shown to increase blood pressure in catecholamine-resistant septic shock. The aim of this study was to measure the effects of low-dose vasopressin on regional (hepato-splanchnic and renal) and microcirculatory (liver, pancreas, and kidney) blood flow in septic shock. METHODS: Thirty-two pigs were anesthetized, mechanically ventilated, and randomly assigned to one of four groups (n = 8 in each). Group S (sepsis) and group SV (sepsis/vasopressin) were exposed to fecal peritonitis. Group C and group V were non-septic controls. After 240 minutes, both septic groups were resuscitated with intravenous fluids. After 300 minutes, groups V and SV received intravenous vasopressin 0.06 IU/kg per hour. Regional blood flow was measured in the hepatic and renal arteries, the portal vein, and the celiac trunk by means of ultrasonic transit time flowmetry. Microcirculatory blood flow was measured in the liver, kidney, and pancreas by means of laser Doppler flowmetry. RESULTS: In septic shock, vasopressin markedly decreased blood flow in the portal vein, by 58% after 1 hour and by 45% after 3 hours (p < 0.01), whereas flow remained virtually unchanged in the hepatic artery and increased in the celiac trunk. Microcirculatory blood flow decreased in the pancreas by 45% (p < 0.01) and in the kidney by 16% (p < 0.01) but remained unchanged in the liver. CONCLUSION: Vasopressin caused marked redistribution of splanchnic regional and microcirculatory blood flow, including a significant decrease in portal, pancreatic, and renal blood flows, whereas hepatic artery flow remained virtually unchanged. This study also showed that increased urine output does not necessarily reflect increased renal blood flow.

Effects of vasopressin on microcirculatory blood flow in the gastrointestinal tract in anesthetized pigs in septic shock

BACKGROUND: Vasopressin increases arterial pressure in septic shock even when alpha-adrenergic agonists fail. The authors studied the effects of vasopressin on microcirculatory blood flow in the entire gastrointestinal tract in anesthetized pigs during early septic shock. METHODS: Thirty-two pigs were intravenously anesthetized, mechanically ventilated, and randomly assigned to one of four groups (n=8 in each; full factorial design). Group S (sepsis) and group SV (sepsis-vasopressin) were made septic by fecal peritonitis. Group C and group V were nonseptic control groups. After 300 min, group V and group SV received intravenous infusion of 0.06 U.kg.h vasopressin. In all groups, cardiac index and superior mesenteric artery flow were measured. Microcirculatory blood flow was recorded with laser Doppler flowmetry in both mucosa and muscularis of the stomach, jejunum, and colon. RESULTS: While vasopressin significantly increased arterial pressure in group SV (P<0.05), superior mesenteric artery flow decreased by 51+/-16% (P<0.05). Systemic and mesenteric oxygen delivery and consumption decreased and oxygen extraction increased in the SV group. Effects on the microcirculation were very heterogeneous; flow decreased in the stomach mucosa (by 23+/-10%; P<0.05), in the stomach muscularis (by 48+/-16%; P<0.05), and in the jejunal mucosa (by 27+/-9%; P<0.05), whereas no significant changes were seen in the colon. CONCLUSION: Vasopressin decreased regional flow in the superior mesenteric artery and microcirculatory blood flow in the upper gastrointestinal tract. This reduction in flow and a concomitant increase in the jejunal mucosa-to-arterial carbon dioxide gap suggest compromised mucosal blood flow in the upper gastrointestinal tract in septic pigs receiving low-dose vasopressin.