Data-driven machine-learning analysis of potential embolic sources in embolic stroke of undetermined source.

BACKGROUND AND PURPOSE: Hierarchical clustering, a common 'unsupervised' machine-learning algorithm, is advantageous for exploring potential underlying aetiology in particularly heterogeneous diseases. We investigated potential embolic sources in embolic stroke of undetermined source (ESUS) using a data-driven machine-learning method, and explored variation in stroke recurrence between clusters. METHODS: We used a hierarchical k-means clustering algorithm on patients' baseline data, which assigned each individual into a unique clustering group, using a minimum-variance method to calculate the similarity between ESUS patients based on all baseline features. Potential embolic sources were categorised into atrial cardiopathy, atrial fibrillation, arterial disease, left ventricular disease, cardiac valvulopathy, patent foramen ovale (PFO) and cancer. RESULTS: Among 800 consecutive ESUS patients (43.3% women, median age 67 years), the optimal number of clusters was four. Left ventricular disease was most prevalent in cluster 1 (present in all patients) and perfectly associated with cluster 1. PFO was most prevalent in cluster 2 (38.9% of patients) and associated significantly with increased likelihood of cluster 2 [adjusted odds ratio: 2.69, 95% confidence interval (CI): 1.64-4.41]. Arterial disease was most prevalent in cluster 3 (57.7%) and associated with increased likelihood of cluster 3 (adjusted odds ratio: 2.21, 95% CI: 1.43-3.13). Atrial cardiopathy was most prevalent in cluster 4 (100%) and perfectly associated with cluster 4. Cluster 3 was the largest cluster involving 53.7% of patients. Atrial fibrillation was not significantly associated with any cluster. CONCLUSIONS: This data-driven machine-learning analysis identified four clusters of ESUS that were strongly associated with arterial disease, atrial cardiopathy, PFO and left ventricular disease, respectively. More than half of the patients were assigned to the cluster associated with arterial disease.

Focal Hypoperfusion in Acute Ischemic Stroke Perfusion CT: Clinical and Radiologic Predictors and Accuracy for Infarct Prediction.

BACKGROUND AND PURPOSE: Perfusion CT may improve the diagnostic performance of noncontrast CT in acute ischemic stroke. We assessed predictors of focal hypoperfusion in acute ischemic stroke and perfusion CT performance in predicting infarction on follow-up imaging. MATERIALS AND METHODS: Patients from the Acute STroke Registry and Analysis of Lausanne data base with acute ischemic stroke and perfusion CT were included. Clinical and radiologic data were collected. We identified predictors of focal hypoperfusion using multivariate analyses. RESULTS: From the 2216 patients with perfusion CT, 38.2% had an acute ischemic lesion on NCCT and 73.3% had focal hypoperfusion on perfusion CT. After we analyzed 104 covariates, high-admission NIHSS, visual field defect, aphasia, hemineglect, sensory deficits, and impaired consciousness were positively associated with focal hypoperfusion. Negative associations were pure posterior circulation, lacunar strokes, and anticoagulation. After integrating radiologic variables into the multivariate analyses, we found that visual field defect, sensory deficits, hemineglect, early ischemic changes on NCCT, anterior circulation, cardioembolic etiology, and arterial occlusion were positively associated with focal hypoperfusion, whereas increasing onset-to-CT delay, chronic vascular lesions, and lacunar etiology showed negative association. Sensitivity, specificity, and positive and negative predictive values of focal hypoperfusion on perfusion CT for infarct detection on follow-up MR imaging were 66.5%, 79.4%, 96.2%, and 22.8%, respectively, with an overall accuracy of 76.8%. CONCLUSIONS: Compared with NCCT, perfusion CT doubles the sensitivity in detecting acute ischemic stroke. Focal hypoperfusion is independently predicted by stroke severity, cortical clinical deficits, nonlacunar supratentorial strokes, and shorter onset-to-imaging delays. A high proportion of patients with focal hypoperfusion developed infarction on subsequent imaging, as did some patients without focal hypoperfusion, indicating the complementarity of perfusion CT and MR imaging in acute ischemic stroke.

Perfusion computed tomography in posterior circulation stroke: predictors and prognostic implications of focal hypoperfusion.

BACKGROUND AND PURPOSE: The aim was to determine the predictors of focal hypoperfusion on computed tomography (CT) perfusion (CTP) in patients with acute posterior circulation stroke and its association with long-term outcome. METHODS: Patients with posterior circulation stroke were selected from the Acute Stroke Registry and Analysis of Lausanne (ASTRAL) who underwent CTP within 24 h of stroke onset as part of the stroke imaging protocol. Hypoperfusion was defined as an area of visually well demarcated mean transit time prolongation corresponding to an arterial territory on standard reconstruction CTP imaging maps. Areas of hypoperfusion were assessed with the posterior circulation Acute Stroke Prognosis Early CT Score. Clinical and imaging associations with focal hypoperfusion were identified using multiple imputation analyses, and the adjusted functional outcome measured by the modified Rankin Scale at 3 and 12 months was determined. RESULTS: Of the 3595 consecutive patients from the ASTRAL registry between 2003 and 2014, 1070 (29.7%) had a posterior circulation stroke and 436 of these (40.7%) patients had a good quality baseline CTP. 23.1% had early ischaemic changes and 37.4% had focal hypoperfusion. In multiple imputation analysis, visual field deficits, reduced level of consciousness, cardiac and multiple stroke mechanisms, significant vessel pathology and ischaemic changes visible on plain CT were associated with focal hypoperfusion. Focal hypoperfusion was independently associated with outcome at 12 months (odds ratio 2.04, 95% confidence interval 1.22-3.42, P < 0.01). CONCLUSIONS: In posterior circulation stroke patients undergoing acute CTP, multiple clinical, aetiological and radiological variables were associated with focal hypoperfusion. Patients with focal hypoperfusion had a worse 12-month outcome.

Improving prediction of recanalization in acute large-vessel occlusive stroke.

BACKGROUND: Recanalization in acute ischemic stroke with large-vessel occlusion is a potent indicator of good clinical outcome. OBJECTIVE: To identify easily available clinical and radiologic variables predicting recanalization at various occlusion sites. METHODS: All consecutive, acute stroke patients from the Acute STroke Registry and Analysis of Lausanne (2003-2011) who had a large-vessel occlusion on computed tomographic angiography (CTA) (< 12 h) were included. Recanalization status was assessed at 24 h (range: 12-48 h) with CTA, magnetic resonance angiography, or ultrasonography. Complete and partial recanalization (corresponding to the modified Treatment in Cerebral Ischemia scale 2-3) were grouped together. Patients were categorized according to occlusion site and treatment modality. RESULTS: Among 439 patients, 51% (224) showed complete or partial recanalization. In multivariate analysis, recanalization of any occlusion site was most strongly associated with endovascular treatment, including bridging therapy (odds ratio [OR] 7.1, 95% confidence interval [CI] 2.2-23.2), and less so with intravenous thrombolysis (OR 1.6, 95% CI 1.0-2.6) and recanalization treatments performed beyond guidelines (OR 2.6, 95% CI 1.2-5.7). Clot location (large vs. intermediate) and tandem pathology (the combination of intracranial occlusion and symptomatic extracranial stenosis) were other variables discriminating between recanalizers and non-recanalizers. For patients with intracranial occlusions, the variables significantly associated with recanalization after 24 h were: baseline National Institutes of Health Stroke Scale (NIHSS) (OR 1.04, 95% CI 1.02-1.1), Alberta Stroke Program Early CT Score (ASPECTS) on initial computed tomography (OR 1.2, 95% CI 1.1-1.3), and an altered level of consciousness (OR 0.2, 95% CI 0.1-0.5). CONCLUSIONS: Acute endovascular treatment is the single most important factor promoting recanalization in acute ischemic stroke. The presence of extracranial vessel stenosis or occlusion decreases recanalization rates. In patients with intracranial occlusions, higher NIHSS score and ASPECTS and normal vigilance facilitate recanalization. Clinical use of these predictors could influence recanalization strategies in individual patients.

Early major worsening in ischemic stroke: predictors and outcome.

INTRODUCTION: We aimed to investigate the characteristics and outcome of patients suffering early major worsening (EMW) after acute ischemic stroke (AIS) and assess the parameters associated with it. METHODS: All consecutive patients with AIS in the ASTRAL registry until 10/2010 were included. EMW was defined as an NIHSS increase of ≥8 points within the first 24 h after admission. The Bootstrap version of the Kolmogorov-Smirnov test and the χ(2)-test were used for the comparison of continuous and categorical covariates, respectively, between patients with and without EMW. Multiple logistic regression analysis was performed to identify independent predictors of EMW. RESULTS: Among 2155 patients, 43 (2.0 %) had an EMW. EMW was independently associated with hemorrhagic transformation (OR 22.6, 95 % CI 9.4-54.2), cervical artery dissection (OR 9.5, 95 % CI 4.4-20.6), initial dysarthria (OR 3.7, 95 % CI 1.7-8.0), and intravenous thrombolysis (OR 2.1, 95 % CI 1.1-4.3), whereas a negative association was identified with initial eye deviation (OR 0.4, 95 % CI 0.2-0.9). Favorable outcome at 3 and 12 months was less frequent in patients with EMW compared to patients without (11.6 vs. 55.3 % and 16.3 vs. 50.7 %, respectively), and case fatality was higher (53.5 vs. 12.9 % and 55.8 vs. 16.8 %, respectively). Stroke recurrence within 3 months in surviving patients was similar between patients with and without EMW (9.3 vs. 9.0 %, respectively). CONCLUSIONS: Worsening of ≥8 points in the NIHSS score during the first 24 h in AIS patients is related to cervical artery dissection and hemorrhagic transformation. It justifies urgent repeat parenchymal and arterial imaging. Both conditions may be influenced by targeted interventions in the acute phase of stroke.

Blood pressure changes in acute ischemic stroke and outcome with respect to stroke etiology.

OBJECTIVE: Previous research suggested that proper blood pressure (BP) management in acute stroke may need to take into account the underlying etiology. METHODS: All patients with acute ischemic stroke registered in the ASTRAL registry between 2003 and 2009 were analyzed. Unfavorable outcome was defined as modified Rankin Scale score >2. A local polynomial surface algorithm was used to assess the effect of baseline and 24- to 48-hour systolic BP (SBP) and mean arterial pressure (MAP) on outcome in patients with lacunar, atherosclerotic, and cardioembolic stroke. RESULTS: A total of 791 patients were included in the analysis. For lacunar and atherosclerotic strokes, there was no difference in the predicted probability of unfavorable outcome between patients with an admission BP of <140 mm Hg, 140-160 mm Hg, or >160 mm Hg (15.3 vs 12.1% vs 20.8%, respectively, for lacunar, p = 015; 41.0% vs 41.5% vs 45.5%, respectively, for atherosclerotic, p = 075), or between patients with BP increase vs decrease at 24-48 hours (18.7% vs 18.0%, respectively, for lacunar, p = 0.84; 43.4% vs 43.6%, respectively, for atherosclerotic, p = 0.88). For cardioembolic strokes, increase of BP at 24-48 hours was associated with higher probability of unfavorable outcome compared to BP reduction (53.4% vs 42.2%, respectively, p = 0.037). Also, the predicted probability of unfavorable outcome was significantly different between patients with an admission BP of <140 mm Hg, 140-160 mm Hg, and >160 mm Hg (34.8% vs 42.3% vs 52.4%, respectively, p < 0.01). CONCLUSIONS: This study provides evidence to support that BP management in acute stroke may have to be tailored with respect to the underlying etiopathogenetic mechanism.

Persistent hyperglycemia at 24-48 h in acute hyperglycemic stroke patients is not associated with a worse functional outcome.

BACKGROUND: Recently, it was shown that the relation between admission glucose and functional outcome after ischemic stroke is described by a J-shaped curve, with a glucose range of 3.7-7.3 mmol/l associated with a favorable outcome. We tested the hypothesis that persistence of hyperglycemia above this threshold at 24-48 h after stroke onset impairs 3-month functional outcome. METHODS: We analyzed all patients with glucose >7.3 mmol/l on admission from the Acute STroke Registry and Analysis of Lausanne (ASTRAL). Patients were divided into two groups according to their subacute glucose level at 24-48 h after last well-being time (group 1: ≤7.3 mmol/l, group 2: >7.3 mmol/l). A favorable functional outcome was defined as a modified Rankin Score (mRS) ≤2 at 3 months. A multiple logistic regression analysis of multiple demographic, clinical, laboratory and neuroimaging covariates was performed to assess predictors of an unfavorable outcome. RESULTS: A total of 1,984 patients with ischemic stroke were admitted between January 1, 2003 and October 20, 2009, within 24 h after last well-being time. In the 421 patients (21.2%) with admission glucose >7.3 mmol/l, the proportion of patients with a favorable outcome was not statistically significantly different between the two groups (59.2 vs. 48.7%, respectively). In multiple logistic regression analysis, unfavorable outcome was significantly associated with age (odds ratio, OR: 1.06, 95% confidence interval, 95% CI: 1.03-1.08 for every 10-year increase), National Institute of Health Stroke Score, NIHSS score, on admission (OR: 1.16, 95% CI: 1.11-1.21), prehospital mRS (OR: 12.63, 95% CI: 2.61-61.10 for patients with score >0), antidiabetic drug usage (OR: 0.36, 95% CI: 0.15-0.86) and glucose on admission (OR: 1.16, 95% CI: 1.02-1.31 for every 1 mmol/l increase). No association was found between persistent hyperglycemia at 24-28 h and outcome in either diabetics or nondiabetics. CONCLUSIONS: In ischemic stroke patients with acute hyperglycemia, persistent hyperglycemia (>7.3 mmol/l) at 24-48 h after stroke onset is not associated with a worse functional outcome at 3 months whether the patient was previously diabetic or not.

Probiotic cheese production using Lactobacillus casei cells immobilized on fruit pieces.

Lactobacillus casei cells were immobilized on fruit (apple and pear) pieces and the immobilized biocatalysts were used separately as adjuncts in probiotic cheese making. In parallel, cheese with free L. casei cells and cheese only from renneted milk were prepared. The produced cheeses were ripened at 4 to 6 degrees C and the effect of salting and ripening time on lactose, lactic acid, ethanol concentration, pH, and lactic acid bacteria viable counts were investigated. Fat, protein, and moisture contents were in the range of usual levels of commercial cheeses. Reactivation in whey of L. casei cells immobilized on fruit pieces after 7 mo of ripening showed a higher rate of pH decrease and lower final pH value compared with reactivation of samples withdrawn from the remaining mass of the cheese without fruit pieces, from cheese with free L. casei, and rennet cheese. Preliminary sensory evaluation revealed the fruity taste of the cheeses containing immobilized L. casei cells on fruit pieces. Commercial Feta cheese was characterized by a more sour taste, whereas no significant differences concerning cheese flavor were reported by the panel between cheese containing free L. casei and rennet cheese. Salted cheeses scored similar values to commercial Feta cheese, whereas unsalted cheese scores were significantly lower, but still acceptable to the sensory panelists.

Active matrix metalloproteinases are present in cartilage during immune complex-mediated arthritis: a pivotal role for stromelysin-1 in cartilage destruction.

The involvement of immune complexes during experimental arthritis in induction of metalloproteinases (MMP)-induced neoepitopes in aggrecan in cartilage, as well as the role of stromelysin-1 (SLN-1) in the induction of this neoepitope, was investigated. Passive immune complex arthritis was induced, and generation of the MMP-specific cleavage product (VDIPEN) was studied by immunolocalization. The role of SLN-1 was studied with use of SLN-1-deficient (SLN-1KO) mice. VDIPEN expression was studied in vitro by exposing the cartilage to IL-1 and subsequent activation of latent MMPs. Immune complex arthritis was characterized by an acute inflammation, with influx of mainly polymorphonuclear cells into the joint cavity. Expression of VDIPEN neoepitopes was consistently found in areas extensively depleted from proteoglycans. SLN-1KO mice did not show expression of the VDIPEN neoepitope, although inflammation and proteoglycan depletion was comparable to wild-type mice. In addition, erosions of cartilage were absent in SLN-1KO mice, but were present in wild-type mice, suggesting an important role for SLN-1 in cartilage destruction. In vitro studies showed that SLN-1 is also pivotally involved in IL-1-induced MMP activity. Stimulated polymorphonuclear neutrophils were able to activate latent MMPs present in the cartilage. Neutrophil elastase was also capable of activating IL-1-induced latent MMPs, which identifies elastase as a possible activator for latent VDIPEN-inducing MMPs. This study suggests that IC are important in the activation of latent MMPs in cartilage, possibly through polymorphonuclear neutrophil activation on the cartilage edge. SLN-1 is a pivotal enzyme in overall MMP-activity in cartilage during immune complex-mediated arthritis.

Robust and realistic approaches to carry-over.

The relationship between choice of model for carry-over and choice of efficient cross-over design is studied in particular with reference to two treatment designs in four periods and two sequences. The effect of model miss-specification is also examined. It is concluded that previous claims regarding efficient designs are not necessarily reasonable.

Fluctuation of fluoride concentrations in drinking waters: a collaborative study.

The aim of this study was to describe the variations with time in the concentrations of fluoride in drinking water sources in Greenland, Kenya, Greece, Denmark and Ireland. Water samples were collected monthly and shipped to laboratories in Aarhus for electrometric analyses. In Narssaq, Greenland the fluoride concentration of a single piped water supply ranged from 0.3 to 2.8 ppm, the variations being related to climate, precipitation and temperature over the year. Water from the Athi River, Kenya had a fluoride content ranging from 0.3 to 1.2 ppm, the higher concentrations being associated with the dry seasons. The fluoride concentration in piped water from mountain rivers in Mourjes, Greece, ranged over the year between 1.3 to 2.0 ppm, the changes being apparently unrelated to rainfall. Marked variations in fluoride concentrations from 0.5 to over 3.5 ppm were observed in water from artesian wells in Assiros, Greece. In drinking waters from boreholes in Boennerup Strand, Denmark, fluoride concentrations ranged with time from 1.4 to 2.4 ppm, the variations being unrelated to climate or precipitation, while little variation in fluoride concentrations was found in water from boreholes in either Roedvig or Egens, Denmark. Water obtained from two sources of artificially fluoridated water supplies from Ireland showed considerable variations with time, although pooled samples indicated relatively constant levels over the year. The study indicates that the results of single fluoride ion measurement from any given source should not be taken as being a reliable indicator of fluoride exposure from drinking water.(ABSTRACT TRUNCATED AT 250 WORDS)

Fluoride in mixed human saliva after different topical fluoride treatments and possible relation to caries inhibition.

Fluoride concentrations in mixed saliva were measured at different intervals following topical treatments with various fluoride preparations and procedures such as dentifrice, tablets, mouthrinse and topical solution. The results were related to available caries reduction data from corresponding clinical trials previously published. Salivary fluoride levels varied widely reflecting the different dosages applied, whereas the caries reduction data irrespective of type of treatment appear very similar, with a magnitude of about 30%. When combining salivary fluoride data with recent understanding of enamel-fluoride kinetics it was possible to explain the clinical caries reductions.

In vitro studies of the phenomenon of tetracycline incorporation into enamel.

Incorporation of tetracyline into dental enamels was studied by exposing presoftened enamel slabs to tetracycline-containing mineralizing solutions. Tetracycline was incorporated only when remineralization (assessed by hardness) occurred and the experimental conditions favored the formation of a soluble tetracycline Ca complex. The fluorescence induced by tetracycline incorporation into the enamel slabs was comparable to that reported under in vivo conditions.