Cardiovascular risk factors in adult patients with multisystem Langerhans-cell histiocytosis: evidence of glucose metabolism abnormalities.

BACKGROUND: Langerhans-cell histiocytosis (LCH) is a rare disease with features of chronic inflammation and it may also induce hypopituitarism, conditions associated with an increased risk of cardiovascular diseases. AIM: Cardiovascular and metabolic risk profile investigation in multisystem LCH patients with and without anterior pituitary deficiency. DESIGN: Prospective, observational study. METHODS: Fourteen adult patients with LCH, 7 with and 7 without anterior pituitary deficiency, and 42 controls matched for age, body mass index (BMI) and smoking. Cardiovascular risk factors were estimated in all subjects: glucose and lipid profile, mathematical indices of insulin resistance (IR), blood pressure, structural arterial and functional endothelial properties (intima-media thickness, brachial artery flow-mediated dilatation). Cardiovascular risk factors were estimated in the three groups studied; the effect of disease activity and/or treatment was also determined in patients with LCH. RESULTS: Ten patients had diabetes insipidus, and 7 anterior pituitary hormone deficiencies: 8 patients had active disease and 11 had received systemic treatment. No difference was observed between the study groups in vascular parameters, in lipid profile or in blood pressure. However, the insulin resistance index GIR was decreased in patients with LCH without anterior pituitary deficiency compared to controls (P = 0.033). Three patients had impaired glucose tolerance and one diabetes mellitus type 2. These patients were older and had active disease; there was no association with hypopituitarism and/or previous treatment. CONCLUSION: Adults patients with LCH have abnormalities of glucose metabolism that tend to occur in patients with active disease, and may be a consequence of the pro-inflammatory state.

Acute effects of renin-angiotensin system blockade on arterial function in hypertensive patients.

The acute effects of the renin-angiotensin system (RAS) blockers may be important in some clinical settings. To assess the acute impact of such drugs on arterial function, we studied the effects of captopril 25 mg, quinapril 20 mg and telmisartan 80 mg on 100 hypertensive patients, according to a randomized, double-blind, placebo-controlled study. Central (aortic) blood pressure (BP) and augmentation index (AIx, a measure of wave reflections), as well as flow-mediated dilatation (FMD) of the brachial artery and forearm blood flow (FBF) (measures of conduit and resistance artery endothelial function, respectively), were evaluated before and 2 h after oral drug administration. Compared to placebo, captopril and quinapril decreased central systolic (by 7.5 mm Hg, P<0.05 and by 12.3 mm Hg, P<0.001) and diastolic BP (by 4.9 mm Hg, P<0.01 and by 8.4 mm Hg, P<0.001), whereas telmisartan had no significant effect (P=NS). Additionally, AIx was reduced after quinapril (absolute decrease of 7.2%, P<0.01) and marginally after captopril (decrease of 4.7%, P=0.07). Only quinapril led to a beneficial change of FMD (absolute increase of 2.7%, P<0.001). No treatment was related to significant changes of peak hyperaemic or 3-min hyperaemic FBF. In adjusted analyses, all the favourable alterations induced by quinapril were independent of potential confounding haemodynamic factors. Our data show that acute RAS inhibition with quinapril (20 mg) may be more beneficial in terms of arterial function and central haemodynamics compared to captopril (25 mg) or telmisartan (80 mg). Further studies are needed to investigate whether these acute arterial effects of quinapril are clinically significant.

The combined effect of augmentation index and carotid intima-media thickness on cardiovascular risk in young and middle-aged men without cardiovascular disease.

Carotid artery intima-media thickness (IMT) has been used as a surrogate marker of atherosclerosis and is related to cardiovascular risk. Indices of arterial stiffness are also associated with cardiovascular risk and atherosclerosis. The aim of this study was to assess the prognostic value of the combination of surrogate markers of cardiovascular disease measured non-invasively in subjects without cardiovascular disease. In this cross-sectional study, 81 young and middle aged males (39.2+/-6.3 years) without evidence of overt cardiovascular disease or diabetes mellitus were enrolled. High-resolution B-mode ultrasonography and pulse wave analysis were used to measure carotid artery IMT and augmentation index (AI), a measure of arterial stiffness. Framingham risk score (FRS) was used as an estimate of the risk for development of cardiovascular disease. Regional differences were observed in the carotid arteries' IMT regarding their relationship with FRS: combined (average from all sites) IMT and IMT in the carotid bulb (CB), but not in the common (CC) and internal carotid artery (IC), and AI showed significant increases of FRS by their tertiles. However, subjects with both AI and IMT at any site in the highest tertile (AI>15%, CC>0.65 mm, CB>0.8 mm, IC>0.65 mm) had an increased FRS compared to subjects with one or none of these parameters in the highest tertile. In conclusion, young and middle-aged men without overt cardiovascular disease with both high IMT and AI are in high cardiovascular risk, as assessed by FRS. Epidemiological studies are needed to further validate this combination.

Peripheral vascular endothelial dysfunction in patients with angina pectoris and normal coronary arteriograms.

OBJECTIVES: We sought to determine endothelium-dependent vasodilator function in the brachial artery of patients with microvascular angina pectoris. BACKGROUND: Previous studies suggest the presence of endothelial dysfunction of the coronary microcirculation in patients with microvascular angina pectoris. It is not known whether endothelial dysfunction in these patients is a generalized process or whether it is confined to the coronary microcirculation only. METHODS: In 11 women (mean [+/-SD] age 60.1 +/- 7.8 years) with microvascular angina (anginal pain, normal epicardial coronary arteries, positive exercise stress test), endothelium-dependent vasodilation was assessed in the brachial artery by measuring the change in brachial artery diameter in response to hyperemic flow. Results were compared with 11 age- and gender-matched patients with known three-vessel coronary artery disease and 11 age- and gender-matched healthy control subjects. In all subjects, the intima-media thickness (IMT) of the common carotid artery was also measured. RESULTS: Flow-mediated dilation (FMD) was comparable in patients with microvascular angina and coronary artery disease (1.9 +/- 2.5% vs. 3.3 +/- 3.3%, p = NS) but was significantly lower in patients with microvascular angina than in healthy control subjects (1.9 +/- 2.5% vs. 7.9 +/- 3%, p < 0.05). IMT was significantly lower in patients with microvascular angina than in those with coronary artery disease (0.64 +/- 0.08 vs. 1.0 +/- 0.28 mm, p < 0.05) and was comparable between patients with microvascular angina pectoris and healthy control subjects (0.64 +/- 0.08 vs. 0.56 +/- 0.14 mm, p = NS). IMT > or = 0.8 mm was observed in 1 of 11 patients with microvascular angina, 1 of 11 control subjects and 10 of 11 patients with coronary artery disease. CONCLUSIONS: These findings suggest that endothelial dysfunction in microvascular angina is a generalized process that also involves the peripheral conduit arteries and is similar to that observed in atherosclerotic disease. IMT could be helpful in discriminating patients with microvascular angina and atherosclerotic coronary artery disease.

Common carotid artery intima-media thickness for the risk assessment of lacunar infarction versus intracerebral haemorrhage.

BACKGROUND AND PURPOSE: Arterial hypertension is the major risk factor for intracerebral haemorrhage (ICH) and lacunar infarction (LI) and both types of cerebral lesions originate from pathology of the same deep perforating small arteries. We aimed to evaluate the relationship between vascular risk factors including common carotid artery intima-media thickness (CCA-IMT) with LI versus ICH. METHODS: We prospectively collected data from 159 first ever stroke patients (67 cases with ICH and 92 cases with LI) with documented history of hypertension. All subjects underwent B-mode ultrasonographic measurements of the CCA-IMT. Logistic regression modelling was used to determine the factors (established vascular risk factors, severity and duration of hypertension, concomitant medications and CCA-IMT) that may significantly differentiate LI from ICH. RESULTS: Patients with LI had significantly (p=0.002) larger CCA-IMT values (0.926 mm, 95% CI: 0.881-0.971) than subjects with ICH (0.815 mm, 95% CI: 0.762-0.868) even after adjusting for baseline characteristics and cardiovascular medications. The multivariate logistic regression procedure selected CCA-IMT, diabetes mellitus and hypercholesterolaemia as the only independent factors able to discriminate between LI and ICH. The risk for LI versus ICH increased continuously with increasing CCA-IMT. For each increment of 0.1 mm in CCA-IMT the probability of suffering from LI versus ICH increased by 36.6% (95 % CI: 13%-65.2%, p=0.001) even after adjustment for cardiovascular risk factors. CONCLUSIONS: Increased CCA-IMT values are a factor favouring LI over ICH in hypertensive patients. The measurement of CCA-IMT may be a useful non-invasive diagnostic tool for the risk assessment of LI with respect to ICH in such patients.

Impaired endothelium-dependent vasodilatation in women with previous gestational diabetes.

OBJECTIVE: To assess whether otherwise healthy women with a history of gestational diabetes mellitus (GDM) may have abnormalities in endothelial function at a very early stage, before glucose intolerance occurs. RESEARCH DESIGN AND METHODS: A total of 33 women with previous GDM (17 nonobese [BMI < 27] and 16 obese [BMI > or = 27]) and 19 healthy nonobese women were examined. A 75-g oral glucose tolerance test was performed, and insulin levels and biochemical parameters were also measured. Using high-resolution ultrasound, we measured vasodilatory responses of the brachial artery during reactive hyperemia (endothelium-dependent vasodilatation), and after nitroglycerin administration, an endothelium-independent vasodilator. RESULTS: Flow-mediated dilatation (FMD) was significantly and equally decreased in both groups of women with previous GDM, compared with control subjects (1.6 +/- 3.7% in the nonobese GDM group and 1.6 +/- 2.5% in the obese GDM group vs. 10.3 +/- 4.4% in control subjects, P < 0.001). FMD correlated inversely with serum uric acid levels, BMI, serum total cholesterol, and basal insulin resistance (homeostasis model assessment). Nitrate-induced dilatation was significantly decreased only in the obese GDM group compared with control subjects, (21.4 +/- 5.1 vs. 27.9 +/- 9.5, P < 0.05). CONCLUSIONS: Endothelial dysfunction, which is considered as a very early index of atherogenesis, is already present in both obese and nonobese women with a history of GDM, even when they have normal glucose tolerance.

Ankle-brachial index as a predictor of the extent of coronary atherosclerosis and cardiovascular events in patients with coronary artery disease.

Resting ankle-brachial pressure index (ABI) is a noninvasive method to assess the patency of the lower extremity arterial system. This study aimed to examine the relation between ABI and the extent of coronary atherosclerosis, the extracoronary atherosclerosis lesions, and the prognosis of patients referred for elective coronary angiography. One hundred sixty-five consecutive patients underwent coronary angiography, ultrasound imaging for intima-media thickness measurement of carotid and femoral arteries and ABI evaluation; subjects were followed up for 14.5 +/- 2.4 months. With regard to vascular risk factors, only smoking (p = 0.025) and diabetes (p = 0.01) were related to ABI in the multiple regression analysis. ABI was independently and inversely related to carotid bifurcation (p = 0.0002) and common femoral artery intima-media thickness (p = 0.018). ABI was related to the extent of coronary artery disease as measured by number of coronary arteries diseased (analysis of variance, p = 0.04) and Gensini angiographic score (p = 0.01). In the follow-up study ABI < 0.90 was a univariate predictor of cardiovascular events (cardiac death, nonfatal myocardial infarction, unstable angina) and revascularization procedures. The estimated cumulative rate free of cardiovascular events was 90% for ABI > 0.90 and 73% for ABI < 0.90 (p = 0.02). In logistic regression analysis, ABI < 0.90 was an independent predictor for cardiovascular events after adjustment for age, low-density lipoprotein cholesterol, carotid and femoral intima-media thickness, and Gensini score. Further adjustment for the confounding effect of insulin weakened the relation between ABI and cardiovascular events (p = 0.1). In conclusion, ABI is a simple index related to the extent of atherosclerosis in coronary and noncoronary arterial beds, reflecting generalized atherosclerosis. ABI could be useful in assessing the risk for cardiovascular events in patients with coronary artery disease.

Atherosclerotic changes of extracoronary arteries are associated with the extent of coronary atherosclerosis.

The aim of the present study was to examine the association between carotid and femoral artery intima media thickness (IMT) and the extent and severity of coronary artery disease (CAD) as well as the effects of traditional vascular risk factors on the atherosclerotic changes in the carotid and femoral arteries. Two hundred twenty-four patients who underwent coronary angiography for suspected CAD were evaluated by B-mode ultrasound imaging of the common carotid, internal carotid, carotid bifurcation, and femoral artery for measurement of IMT; traditional vascular risk factors were also evaluated in these patients. CAD extent was evaluated by the number of diseased vessels and by Gensini score. Age, male gender, and diabetes were common risk factors for higher CAD extent and higher carotid and femoral IMT. Insulin levels were correlated with femoral IMT and CAD extent, whereas blood lipids were correlated predominantly with carotid IMT. IMT from carotid and femoral arteries increased significantly with an increase in CAD extent. Using multiple stepwise regression analysis, the following parameters were found to be independent predictors of CAD extent: male gender (p<0.0001), common femoral artery IMT (p = 0.0028), common carotid artery IMT (p = 0.015), age (p = 0.02), diabetes mellitus (p = 0.035), and carotid artery bulb IMT (p = 0.04). Common femoral IMT was the only independent parameter for predicting Gensini score (p<0.0001). In conclusion, there are territorial differences in the various arterial beds regarding their response to risk factors. Femoral artery and carotid bulb are independent predictors of CAD extent and the inclusion of these measurements would add information to that provided by the common carotid artery.

Pulse pressure in normotensives: a marker of cardiovascular disease.

The purpose of the present study was to evaluate the relation of the systemic arterial pulse pressure and other parameters derived from the 24-h arterial blood pressure (BP) monitoring to the severity of coronary artery disease, carotid lesions, and left ventricular (LV) mass index in patients without arterial hypertension. One hundred ten patients with known coronary artery disease underwent coronary arteriography, 24-h arterial BP monitoring, and ultrasound imaging of the carotid arteries and the myocardium. Measurements of 24-h arterial BP monitoring (systolic, diastolic, and average BP, pulse pressure, abnormal values of systolic and diastolic BP, and heart rate), the severity of coronary heart disease (Gensini score), intima-media thickness (IMT) of the common carotid artery and LV mass index were determined in all patients. By univariate analysis, only 24-h pulse pressure was significantly related to the severity of coronary artery disease (P < .01), carotid IMT(P < .01), and LV mass index (P < .01). In a multivariate analysis, 24-h pulse pressure was also the best predictor of the severity of coronary lesions (P = .009), carotid IMT (P = .003), and LV mass index (P = .009). Gensini score was related (P < .01) to LV mass index and not to carotid IMT. In conclusion, systemic arterial pulse pressure derived from 24-h arterial BP monitoring is related to coronary artery disease, carotid IMT, and LV mass index independently of age or any other derivative of 24-h arterial BP monitoring, indicating that this parameter could be a marker of global cardiovascular risk.

Systemic sclerosis is not associated with clinical or ambulatory blood pressure.

BACKGROUND: Raynaud's phenomenon is often the first symptom and occurs eventually in more than 95% of patients with systemic sclerosis (SSc). Angiographic studies disclose narrowing and obstruction of the digital arteries, which on autopsy histologic study show prominent subintimal connective tissue proliferation without inflammation, as well as adventitial fibrosis. It is also known that primary cardiac problems include pericarditis, left ventricular or biventricular failure, serious supraventricular or ventricular arrhythmias emerge in patients with SSc. It is not known if these patients present hypertension or hypotension and which parameter of the ambulatory blood pressure may influence such a disease course. METHODS: A total of 85 subjects underwent clinical blood pressure (BP) readings, 24-hour ambulatory BP monitoring, left ventricular assessment by echocardiography and measurement of intima media thickness (IMT) of the right-left internal carotid arteries (RICA and LICA) and right-left common carotid arteries (RCCA and LCCA). The population consisted of 40 subjects with SSc according to the criteria of the American College of Rheumatology (SCL-group) who were not receiving any antihypertensive treatment and 45 healthy volunteers (control group). The two groups did not differ in age. RESULTS: Clinical systolic and diastolic blood pressure, clinical heart rate, mean 24 h systolic blood pressure, SD systolic blood pressure, mean 24 h diastolic blood pressure, SD 24 h diastolic blood pressure, mean 24 h heart rate, SD 24 h heart rate, pulse pressure 24 h, serum glucose, cholesterol, triglycerides, HDL, LDL, creatinine, urea, potassium and natrium did not statistically significant differ between the two groups. Furthermore, the left ventricular mass/BSA and IMT of both carotid arteries did not show a statistically significant difference between the groups. CONCLUSION: Systemic sclerosis is not associated with clinical blood pressure or the parameter of 24 h blood pressure monitoring.

Cholesterol and unsaturated fat diets influence lipid and glucose concentrations in rats.

The present study investigated the effects of dietary cholesterol and monounsaturated and polyunsaturated fatty acids on plasma lipids and glucose concentrations. Four groups of ten male Wistar albino rats were fed diets of different fatty acid composition for 40 days. The control group consumed nonpurified diet (containing fat 3.7 g/100 g diet), and cholesterol, olive oil, and safflower oil groups consumed the nonpurified diet enriched with 14 g fat/100 g diet with egg yolk, olive oil, or safflower oil, respectively. Compared with the control, the diet enriched with cholesterol significantly increased fasting plasma cholesterol (P < 0.01), triacylglycerol (P < 0.01), total lipid (P < 0.01) and glucose (P < 0.05) concentrations; in the olive oil group, cholesterol and triacylglycerol levels were significantly increased compared with control group (P < 0.01 in both instances). In safflower oil group, triacylglycerol levels were also significantly increased (P < 0.05) compared with the controls. After comparing diets providing the same amount of fat (cholesterol, olive oil, and safflower oil groups), higher cholesterol, triacylglycerol and total lipid levels were observed in the cholesterol group than in the olive oil group (P < 0.01, P < 0.05 and P < 0.01, respectively), and safflower oil group (P < 0.01 in all instances). High-density lipoprotein-cholesterol concentrations were significantly lower in the cholesterol group than in the olive oil and safflower oil groups (P < 0.05 in both instances) and fasting plasma glucose levels were higher in the cholesterol than in the olive oil (P < 0.05) and safflower oil groups (P < 0.01). Finally, after comparing lipid and glucose levels in the unsaturated fatty acids-enriched diets, higher plasma cholesterol concentrations were observed in the olive oil than in the safflower oil group (P < 0.05). These data suggest that not only the amount but also the type of dietary fat can influence serum lipid levels.

The role of mitral annulus motion in left ventricular function.

The dynamics between mitral annulus motion and left ventricular function were studied. Thirty subjects (10 normals, 10 with severe cardiac failure and 10 with mitral regurgitation) were examined. Left ventricular long and short-axis dimensions were measured in a two-dimensional apical four-chamber image. A volume was calculated as defined by cardiac cyclic changes of mitral annulus in size and motion. This volume had the approximate geometric configuration of a truncuated cone and was covered by mitral annulus motion at enddiastole and endsystole. Left ventricular ejection fraction was calculated echocardiographically using the prolate ellipsoid/area-length method. The volume defined by mitral annulus motion was 24 ml in normals, 16 ml in patients with cardiac failure and 48 ml in patients with mitral regurgitation while the ejection fraction was 51%, 19% and 53% respectively. Comparison between the volume of the truncuated cone and the ejection fraction in all 30 subjects revealed a statistically significant correlation (p less than 0.01). Thus, the bigger the volume determined by the mitral annulus motion during systole, the larger portion of the left ventricle is "atrialized", and a higher ejection fraction follows. Mitral annulus motion may provide new information about cardiac mechanics in normal and abnormal conditions.

Effect of acute blood pressure changes on signal averaged electrocardiogram.

An acute blood pressure elevation may cause ventricular ectopic rhythms, while its reduction may alleviate them. It is studied whether the blood pressure exerts some effect on parameters obtained by the signal averaged electrocardiogram. In 25 patients with either hypertension (8 cases) or ventricular ectopic rhythms (10 cases) or both (7 cases) the blood pressure was reduced by sodium nitroprusside (24 cases) and/or elevated with metaraminol (10 cases) and the signal averaging electrocardiogram was recorded under 2 or 3 pressure values on each patient. During the high pressure (193.6 +/- 20.1 mm Hg) the following differences were noted compared to the low pressure (77.4 +/- 15.2 less): longer QRS duration in all 25 patients (+9.92 +/- 10.51 ms, P < 0.001); longer low (< 40 microV) amplitude signals (LAS) in 18 patients (+6.94 +/- 10.93 ms, P < 0.005); lower root mean square voltage of the terminal 40 ms of the QRS in 22 patients (-15.73 +/- 21.60 microV, P < 0.005); and ventricular ectopic beat incidence higher in 8, lower in 1, and equal in 2 cases (with no arrhythmia in the other 14). The generally and focally reduced conduction, as suggested by the QRS and LAS prolongation, might contribute to the proarrhythmic effect of acute blood pressure elevation.

Diurnal variation of endothelial function and arterial stiffness in hypertension.

The onset of cardiovascular events presents a circadian variation that may be mediated by similar temporal patterns of vascular function. Blood pressure also follows circadian variation. We investigated the possible diurnal variation of endothelial function and arterial stiffness in patients with hypertension. Thirty-five individuals with recently diagnosed hypertension (mean age 48.3 years, range 30-60 years, 14 men) were examined. Flow-mediated vasodilatation (FMD), nitrate-mediated vasodilatation (NMD) and carotid-femoral (cf) pulse wave velocity (PWV) were measured at three different occasions: at 0700 hours immediately after awaking, at 1200 hours and at 2100 hours. FMD was markedly lower in the morning (0700 hours, 2.22+/-1.58%; 1200 hours, 4.37+/-2.25%; 2100 hours, 4.28+/-2.12%; P<0.001), whereas NMD was similar at the same time points. This difference remained significant after adjustment for baseline brachial artery diameter and reactive hyperaemia. PWVcf progressively increased from morning to evening (0700 hours, 9.8+/-1.9 m s(-1); 1200 hours, 10.2+/-2.2 m s(-1); 0900 hours, 10.5+/-1.9 m s(-1); P=0.013 for linear trend). Similar temporal patterns were observed in systolic and diastolic blood pressures peaking in the evening. PWVcf changes lost significance after adjustment for changes in mean blood pressure. Endothelial function is decreased in the early morning in hypertensive patients, whereas arterial stiffness is increased in the evening. Changes in BP-dependent passive artery distension may be involved in this phenomenon.

Cardiac involvement in collagen diseases.

The purpose of this study is to evaluate the early morphological and functional abnormalities of the heart in patients with collagen disease. The study population was free of risk factors for coronary artery disease and without any clinically evident cardiac manifestations. In 62 patients with collagen disease (25 with progressive systemic sclerosis, 19 with systemic lupus erythematosus, 15 with rheumatoid arthritis, three with dermatomyositis) and in 40 healthy subjects an echocardiographic study was performed. Echocardiographic examination from the apical four-chamber view was performed at rest and during the end of a 3 min isometric exercise with handgrip. Global and regional ejection fraction of the left ventricle were calculated. In the group with progressive systemic sclerosis the left ventricular mass index was significantly higher than in the control group (110.78 +/- 48.61 vs 82.18 +/- 28.46 g.m-2) and the ejection fraction (53.61 +/- 7.95%) was the lowest of all groups (control: 61.47 +/- 8.52%, systemic lupus erythematosus: 59.04 +/- 8.58%, rheumatoid arthritis: 62.38 +/- 6.88%). Regional ejection fraction analysis revealed a major dysfunction of the proximal segment of the interventricular septum, in all groups. During isometric exercise, the global and regional ejection fraction did not change significantly, although differences between groups disappeared. In rheumatoid arthritis, mitral and aortic valve leaflet separation appeared to be reduced. In the group with systemic lupus erythematosus, mild abnormalities were noticed, although the mean age and duration of the disease were the smallest compared with the other groups. In conclusion, patients with progressive systemic sclerosis mainly present left ventricular hypertrophy with a reduced ejection fraction while rheumatoid arthritis patients show a predominant valve dysfunction.(ABSTRACT TRUNCATED AT 250 WORDS)

The Athens stroke registry: results of a five-year hospital-based study.

The advent and wide application of new technology, especially noninvasive techniques, has enabled physicians to more completely investigate and clarify the etiopathogenic mechanisms of stroke. Such data have not been available until recently for Southeastern Europe. In addition, during the last decades, strategies for the modification of risk factors and primary prevention may have changed the prevalence of each subgroup of stroke as well. We investigated 1, 042 consecutive patients who had first strokes, during a period of 5 years (from June 1992 to May 1997) and classified them prospectively based on etiopathogenic mechanisms. Patients with transient ischemic attacks and subarachnoid hemorrhage were excluded. There were 613 male and 429 female patients, with a mean age of 70.2 +/- 11.9 years. Forty-six percent of the patients arrived within 3 h from stroke onset. The probable mechanisms were: large-artery atherosclerosis, 156 (15%); lacunes, 177 (17%); cardioembolic, 335 (32.1%); infarct of unknown cause, 182 (17.5%); miscellaneous causes, 35 (3.3%), and intracerebral hemorrhage (ICH), 157 (15.1%). In the cardioembolic group, nonvalvular atrial fibrillation (NVAF) was the probable cause in 225 patients, especially in patients older than 75 years (65%). The overall hospital mortality was 15.2% (from 0.6% for lacunar stroke to 34% for ICH). In our population, cardioembolism is the most frequent subtype of stroke. NVAF is the most likely source, especially in older patients.