Prevalence and molecular epidemiology of Staphylococcus aureus nasal colonization in four nursing home residents in Crete, Greece.

Nursing homes are considered as reservoirs for methicillin-resistant Staphylococcus aureus (MRSA). The present study investigated the point prevalence and molecular epidemiology of S. aureus colonization among nursing home residents. The study population comprised of 227 residents, living in four nursing homes of the Heraklion, Crete, Greece area, between January and December 2015. From each nursing home, swabs from the anterior nares of all eligible participants were obtained within a 2-week period. The isolated S. aureus strains were identified and screened by standard microbiological and molecular epidemiological methods. S. aureus carriage was found in 62 out of 227 participants (38.4%) with 33 out of 62 (53.2%) being MRSA. The median age was 83 years (range 52-103). Females were more frequently colonized [47 (75.8%)]. All 33 methicillin resistant Staphylococcus aureus (MRSA) isolates were mecA-positive carrying SCCmec type IV, 30 (91%) the fnbA, and 17 (51.5%) the PVL genes. Thirty-two (97%) belonged to a single pulsotype C; among them, the PVL-positives belonged to ST80 clone, whereas, the PVL-negatives to ST225. Among the 33 MRSA isolates, 32 (97%) were clindamycin-resistant, carrying the ermA gene. Methicillin-susceptible Staphylococcus aureus (MSSA) strains showed polyclonality and 76% were PVL-positive. In conclusion the present study has shown that nursing homes in our area can be regarded as important reservoirs for community-associated MRSA (CA-MRSA).

Metabolic profile of patients with isolated systolic hypertension.

PURPOSE/OBJECTIVE: While hypertension is an important contributor to cardiovascular disease (CVD) and its treatment has well-established mortality benefits, there is uncertainty as regards the management of isolated systolic hypertension (ISH). Furthermore, the association of ISH with CVD and mortality has been established, but the metabolic characteristics of the affected population have not as yet been adequately described. The aim of this study was to describe the metabolic profiles of patients with ISH. METHODS: An observational study of patients attending the Hypertension Unit of the University Hospital of Heraklion, Crete, Greece, was performed. RESULTS: In total, 809 hypertensive patients not on any antihypertensive treatment were identified. Among them, 44.7% were men, aged 55.6 ± 12.5 years, while 29.7% of both men and women were smokers. Systolic blood pressure was 161.3 ± 15.8 mmHg and diastolic blood pressure was 96.1 ± 11.3 mmHg. Body mass index (BMI) was 31 ± 5.3 kg/m2, while 9.6% had type 2 diabetes (T2D). A comparison of patients with ISH with those with hypertension, but not ISH, revealed that patients with ISH were older and had lower SBP and higher pulse pressure, while they also had lower total cholesterol and LDL and were more likely to have T2D, albeit they had a slightly lower BMI. On the other hand, they did not have any difference in terms of gender, smoking status, HDL, triglycerides, liver biochemistry, uric acid, or prevalence of impaired fasting glucose. CONCLUSION: Patients with ISH were older, with lower SBP, total cholesterol, and LDL and higher pulse pressure and higher prevalence of diabetes.

High prevalence of metabolic syndrome and cardiovascular risk factors in men with ankylosing spondylitis on anti-TNFalpha treatment: correlation with disease activity.

OBJECTIVE: Ankylosing spondylitis (AS) may be associated with an increased risk for cardiovascular diseases (CVD). We investigated the prevalence of cardiovascular risk factors and metabolic syndrome (MetS) in men with AS and assessed any correlation with AS-related factors. METHODS: This was a cross-sectional study of 63 men with AS, median age 40 (19-69) years, and 126 age-matched controls. Patients were on anti-TNFalpha treatment because of considerable disease activity at some time during the course of the disease. MetS was assessed according to the modified National Cholesterol Education Program Adult Treatment Panel III criteria. The risk for CVD event within the next 10 years was estimated using the Framingham equation. RESULTS: Patients had lower high-density lipoprotein cholesterol (HDL-C) (p<0.001), higher systolic (p=0.001) and diastolic (p<0.01) blood pressure compared with controls. The prevalence of the MetS was higher in patients compared to controls (34.9% vs. 19.0%; p<0.05). AS patients with MetS were older (p<0.01), with higher Framingham risk score (p=0.001), had longer disease duration (p<0.05) and higher BASDAI (5.1 vs. 3.7; p<0.05) than those without MetS, while both BASFI and CRP had an inverse correlation with HDL-C levels. CONCLUSIONS: Men with AS have a higher prevalence of cardiovascular risk factors and MetS compared with controls. The presence of MetS was associated with increased 10 year CVD risk in these patients. The association of AS disease activity with MetS suggests that CVD in AS patients may, at least in part, be attributed to the inflammatory burden of the disease.

Lipids and stroke: neglect of a useful preventive measure?

The epidemiological studies linking lipid variables and stroke are reviewed. These studies indicate that serum total cholesterol (TC) levels are associated positively with thrombotic and negatively with haemorrhagic strokes. Relationships for other lipid fractions are not as clearly defined. The results of trials with lipid lowering drugs suggest that only statins effectively reduce the incidence of stroke. Differences between trial results may be due to variation in the extent of reduction of TC levels. Possible underlying mechanisms for benefit and the apparent superiority of statins are also discussed. The reduction in the risk of thrombotic stroke with statins is most evident through meta-analyses (p < 0.001) and less impressive in individual trials (p < 0.03). This difference is largely attributable to the small number of events in trials primarily aimed at evaluating ischaemic heart disease (IHD) reduction. This also means that benefit may be limited to those with established IHD. IHD and thrombotic stroke share common risk factors and patients with one condition are at high risk of developing the other. Therefore, one additional reason for using statins in these patients is that these drugs can effectively prevent IHD-related events including deaths.

The effects of antihypertensive therapy on haemostatic parameters.

There is extensive trial-based evidence showing that antihypertensive drugs reduce the risk of vascular events (e.g. stroke and myocardial infarction) as well as target organ damage (e.g. left ventricular hypertrophy and microalbuminuria). However, some of these benefits appear to be, at least partially, independent of the extent of blood pressure (BP) lowering. It is also evident that in certain clinical situations some antihypertensive drugs are more effective than others. In this review we discuss the effects of antihypertensive drugs on the endothelium, platelets, fibrinolysis and coagulation. These properties may account for the observed BP-independent actions. Antihypertensive drugs exert multiple effects on the vascular endothelium. These include effects on nitric oxide (NO) and angiotensin II-mediated actions. Many BP lowering drugs can inhibit platelet activity, although the relevance of this property is unknown, especially if patients are also taking platelet inhibitors (e.g. aspirin). Antihypertensive drugs also influence fibrinolysis and coagulation. These effects may be mediated by a variety of mechanisms, including altering insulin sensitivity. The haemostatic actions of antihypertensive drugs deserve greater recognition and further investigation.

Effect of hypertension and its treatment on lipid, lipoprotein(a), fibrinogen, and bilirubin levels in patients referred for dyslipidemia.

We measured the serum lipid profile, together with plasma fibrinogen and serum lipoprotein(a) (Lp[a]), glucose, bilirubin, and albumin levels in 491 patients (310 men) who were referred for the management of primary dyslipidemia. All these variables have been shown to predict vascular events. The patients were not taking lipid-lowering drugs; hypertension was present in 156 (31.7%) of them. Of the hypertensive patients, 52 (33%) were not receiving any treatment to control their blood pressure. This omission was not due to a lower prevalence of established vascular disease. The treated hypertensives were divided into three groups according to their treatment: 62 were taking lipid-hostile antihypertensives (beta-blockers, thiazides), 37 were taking lipid-neutral antihypertensives (angiotensin converting enzyme inhibitors, Ca-channel blockers, angiotensin II receptor blockers, indapamide sustained release), and five were taking lipid-friendly antihypertensives (doxazosin). Lipid-hostile antihypertensive drugs were associated with a significantly higher fibrinogen concentration when compared with untreated hypertensives or those taking lipid-neutral/lipid-friendly drugs (median values: 383, 353, and 336 mg/dL, respectively; P < .01). Lipid-neutral/lipid-friendly antihypertensive drugs were associated with lower Lp(a) levels when compared with untreated hypertensives (median values: 22 and 45 mg/dL, respectively; P < .05). The serum bilirubin level was significantly lower in the untreated hypertensives when compared with normotensives or the treated hypertensives. There were no significant differences in lipids, glucose, or albumin among the groups of hypertensives or normotensives. The influence of antihypertensive drugs on additional cardiovascular risk factors should be considered when selecting medication to reduce blood pressure.

Prevalence study of hospital-acquired infections in 14 Greek hospitals: planning from the local to the national surveillance level.

A prevalence study of hospital-acquired infections (HAI) was carried out in 14 of 112 Greek hospitals (15.7%), scattered throughout Greece. Five of seven Greek university hospitals and nine regional hospitals participated in the one-day study, and 3925 hospitalized patients (10.5% of the total hospital beds in Greece) were recorded. The aim of this project was to organize a surveillance of HAI with the participation of the greatest possible number of Greek hospitals, transferring the experience from the local Cretan infection control network in an effort to create a nationwide network. Special attention was paid to recruit all Greek university hospitals in our attempt to expand the study base. Co-ordination of the participating centres, education of the infection control teams on surveillance methods, preparation of agreed definitions, and elaboration of the protocol for the collection of the data were the major objectives of this study. The difficulties, however, were limited resources and the lack of skilled personnel. The overall prevalence of HAI was found to be 9.3%. The most common HAI recorded involved lower respiratory tract infections (30.3%), followed by urinary tract infections (22.7%), bloodstream infections (15.8%), and surgical site infections (14.8%). The greatest prevalence rate was found in the adult ICU (48.4%), followed by the neonatal ICU (30.3%). The duration of hospitalization, the number of operations, the total number of used devices and invasive procedures were significantly correlated with HAI. Positive cultures were found in 51.5% of the cases. The most frequently isolated micro-organisms were: Pseudomonas aeruginosa (16.6%), Escherichia coli (10.8%), Klebsiella pneumoniae (10.3%), Staphylococcus epidermidis (8.1%) and Staphylococcus aureus (7.6%). The administration of antibiotics was also recorded. The prevalence of antibiotic use was 51.4%.

Treating dyslipidaemia in non-insulin-dependent diabetes mellitus -- a special reference to statins.

Patients with non-insulin-dependent diabetes (NIDDM) have an increased incidence of ischaemic heart disease (IHD) when compared with nondiabetic subjects. In addition, they have a worse prognosis after their first myocardial infarction (MI). According to the recent USA recommendations, the threshold for initiation of dietary intervention in diabetic subjects is an LDL greater than 2.6 mmol/l, with the goal to achieve levels less than 2.6 mmol/l (100 mg/dl). This is also the threshold for initiation and treatment goal for pharmacological intervention in diabetic subjects, unless they are completely free of IHD, peripheral vascular disease or cerebrovascular disease and have no other IHD risk factors. In the latter circumstances, the threshold for treatment is an LDL greater than 3.38 mmol/l (130 mg/dl), with the goal to achieve levels less than 3.38 mmol/l. The HMG-CoA reductase inhibitors (statins) can improve the lipid profile effectively and safely in NIDDM. Results from post hoc analyses of diabetic subgroups in the large intervention trials suggest that some statins significantly reduce the risk for IHD-related mortality/morbidity. However, because these results are derived from secondary prevention trials, we cannot be sure if these benefits apply to all diabetic subjects or only to those who already have IHD. Nevertheless, it seems logical to assume that this benefit also applies to NIDDM patients who do not have IHD because they share a similar vascular risk as nondiabetic subjects who have IHD. Intervention trials using statins and fibrates, alone or in combination, in NIDDM are under way. In a few years these trials will provide definitive end-point-based evidence in this high-risk group of patients.

Statin + fibrate combination therapy fluvastatin with bezafibrate or ciprofibrate in high risk patients with vascular disease.

We evaluated the use of combination therapy (ciprofibrate 100 mg or bezafibrate 400 mg plus fluvastatin 40 mg) in 23 patients (n = 13 in the ciprofibrate group) with established cardiovascular disease. Both treatments achieved a significant (P< or =0.01) decrease in the total cholesterol (TC) (32 and 21%), triglycerides (TG) (53 and 46%) and low-density lipoprotein (LDL) (36 and 26%) levels and the TC/high-density lipoprotein (HDL) (42 and 31%) and LDL/HDL (46 and 35%) ratios. HDL levels were increased (19% for both treatment groups), but this rise only achieved significance (P=0.01) in the ciprofibrate group. Although the two patient groups were not strictly matched, the reduction in serum TC and LDL levels was greater with ciprofibrate (32 and 36%, respectively; P< or =0.001) than with bezafibrate (21 and 26%, respectively; P< or =0.01). There was a significant reduction in plasma fibrinogen levels (36.4 and 13.5% in the ciprofibrate and bezafibrate group, respectively). None of the patients reported myalgia or had abnormal creatine kinase activity or liver function tests. Combination therapy is worth considering in high-risk patients because of the advantages associated with this option. Combination therapy is competitively priced when compared with high doses of statins. An end-point-based trial is needed.

The effect of peripheral vascular disease on the serum levels of natural anti-oxidants: bilirubin and albumin.

BACKGROUND: Bilirubin and albumin may act as antioxidants. Their circulating levels are lower in those patients with ischemic heart disease (IHD) and could be further reduced by more extensive atherosclerosis, i.e. peripheral vascular disease (PVD). METHODS: Serum bilirubin and albumin were measured in 456 patients classified into 3 groups: 1) no clinically evident cardiovascular disease (CVD), 2) IHD present and 3) PVD present. Smoking status and gender (which affect bilirubin and albumin levels in healthy individuals) were considered separately. RESULTS: Bilirubin was lower in smoking men without CVD or with PVD than in non-smokers (p=0.02 and p=0.04, respectively) in the same groups. Non-smoking women without CVD had significantly (p=0.004) lower bilirubin levels than the corresponding group of men. Frequency analysis of male non-smokers revealed significantly (p=0.04) more patients with a lower bilirubin (<6.5 mmol/l) in the PVD compared with the no CVD group. Albumin levels showed the same trends. CONCLUSIONS: Prospective studies should consider smoking and gender when assessing the relevance of bilirubin and/or albumin levels in patients with vascular disease. Our findings support those of other studies that show that low serum bilirubin and albumin levels are associated with the presence of vascular disease.

Beneficial effect of moderate alcohol consumption on vascular disease: myth or reality?

Moderate ethanol consumption (1-3 drinks/day on 5-6 days/week) has a favourable effect on vascular disease-related mortality and morbidity [especially ischaemic heart disease (IHD)]. This cardioprotective effect may be due to significant effects on cardiovascular risk factors such as high density cholesterol (HDL) concentration (HDL protects from IHD) and an inhibition of platelet aggregation (increased platelet aggregability predicts coronary events). In contrast, alcoholics and problem drinkers have an excess of IHD-related, and possibly stroke-related, mortality. Excessive alcohol intake may raise the blood pressure. Prolonged alcohol abuse can also result in alcoholic heart muscle disease. Alcohol is the major cause of non-ischaemic cardiomyopathy in Western society. Although there is a widespread belief that red wine protects more than other alcoholic beverages, several studies do not support this interpretation.

Smoking and urological disease.

It is important to realise that virtually every part of the body, including the urological system, is adversely affected by smoking. Smoking is the most important known preventable cause of urinary bladder cancer and is also associated with a risk of prostatic and renal cancer. The exact mechanism by which smoking increases the incidence of urological malignancy is not known. One possibility is that chemicals in cigarette smoke inhibit the synthesis of cytoprotective eicosanoids. Deficient local protection, against the hostile environment caused by the presence of urine, could then encourage the process of carcinogenesis. Smoking is a powerful predictor of erectile dysfunction; cessation may restore normal function. Cigarette smoke also exerts adverse effects on sperm motility and count. Although there is no convincing evidence of reduced fertility in male smokers, it is advisable for men to quit smoking should they have marginal semen quality and wish to start a family. Smoking causes substantial urological pathology; these facts can be used to convince patients with urological problems to quit smoking.

The application of hazard analysis critical control point (HACCP) in a flight catering establishment improved the bacteriological quality of meals.

Following an outbreak of salmonellosis affecting 415 passengers on flights in 1991, the associated flight catering establishment located on a Greek island was surveyed for two years. During the first year of the survey, the bacteriological quality of food was not satisfactory. In an attempt to maximize food safety for crew and passengers the HACCP approach was implemented in 1993. Since its application, greatly supported by the management and staff, the bacteriological quality of aircraft meals was considerably improved.

Smoking, diabetes and hyperlipidaemia.

The epidemiological evidence linking smoking with insulin resistance is considerable. This evidence is even more convincing because there is a dose response relationship between smoking and the risk of non-insulin dependent diabetes (NIDDM). Similarly, there is a time-dependent decrease in risk of NIDDM for those who quit smoking. Insulin resistance (in the form of impaired glucose tolerance, IGT) may precede the development of NIDDM. There is a biochemical basis for the smoking-IGT/NIDDM relationship. Smoking increases the risk of developing diabetic complications like nephropathy, neuropathy and retinopathy Smoking is also an independent risk factor for myocardial infarction and all-cause mortality in NIDDM. Smokers are both insulin resistant and lipid intolerant. Smoking cessation increases circulating high density lipoprotein (HDL) and reduces low density lipoprotein (LDL) levels, despite weight gain. Those providing advice or treatment to improve cardiovascular risk factors should be aware of these smoking-related harmful effects. This is especially true if IGT is underdiagnosed despite the fact that this condition increases the risk of vascular events. Explaining that smoking increases the chance of developing diabetes as well as raising 'blood fat' levels may convince more smokers to quit.

Staphylococcus aureus nasal carriage might be associated with vitamin D receptor polymorphisms in type 2 diabetes.

Vitamin D receptor (VDR) gene polymorphisms have been associated with susceptibility to several diseases, including type 1 diabetes (T1D), type 2 diabetes (T2D), and various infections. The study investigated whether VDR gene polymorphisms influence nasal carriage of Staphylococcus aureus in individuals with T2D, an important source for bloodstream, surgical site and other nosocomial infections. In 173 patients with T2D genotyped for the VDR gene polymorphisms on FokI (rs10735810) F>f, BsmI (rs1544410) B>b, ApaI (rs7975232) A>a, and TaqI (rs731236) T>t, a nasal swab was obtained to detect colonization by S. aureus. A repeat swab was obtained in 162/173 subjects for the estimation of persistent S. aureus carriage. The prevalence of S. aureus nasal colonization was 19.7% and of persistent carriage was 8.6%. Nasal colonization by S. aureus was more common in individuals with FokI f allele than F allele (p 0.05; OR 1.69, 95% CI 1.00-2.89) and individuals with FokI ff genotypes were more frequently colonized than those with FokI FF and Ff genotypes combined (p 0.03; OR 2.61, 95% CI 1.14-5.99). The presence of the FokI f allele was related to higher rates of S. aureus persistent nasal colonization (p 0.002; OR 3.53, 95% CI 1.56-7.98), and individuals with a FokI ff genotype were more often persistent carriers than those with FokI FF and Ff genotypes combined (p <0.001; OR 7.32, 95% CI 2.39-22.41). This study is the first, to our knowledge, to show an association between FokI polymorphism in the VDR gene and nasal carriage of S. aureus in individuals with T2D.

Infections in traumatic brain injury patients.

Traumatic brain injury (TBI) victims are considered to be at high risk for infection. The purpose of this cohort study was to delineate the rates, types and risk factors for infection in TBI patients. Retrospective surveillance of infections was conducted for all TBI patients, aged ≥18 years, cared for at the Department of Neurosurgery of the University Hospital of Heraklion, Greece, between 1999 and 2005. A total of 760 patients (75% men) with a median age of 41 years were included. Most (59%) were injured in a motor vehicle accident. One third of them underwent a surgical procedure. Two hundred and fourteen infections were observed. The majority were infections of the lower respiratory tract (47%), followed by surgical site infections (SSI) (17%). Multivariate analysis showed that SSI development was independently associated with the performance of ≥2 surgical procedures (OR 16.7), presence of concomitant infections, namely VAP (OR 5.7) and UTI (OR 8.8), insertion of lumbar (OR 34.5) and ventricular drains (OR 4.0), and cerebrospinal fluid (CSF) leak (OR 3.8). Development of meningitis was associated with prolonged hospitalization (OR 1.02), especially >7 days ICU stay (OR 25.5), and insertion of lumbar (OR 297) and ventricular drains (OR 9.1). There was a notable predominance of Acinetobacter spp. as a VAP pathogen; gram-positive organisms remained the most prevalent in SSI cases. Respiratory tract infections were the most common among TBI patients. Device-related communication of the CSF with the environment and prolonged hospitalization, especially in the ICU setting, were independent risk factors for SSIs and meningitis cases.

Predicting coronary heart disease risk using the Framingham and PROCAM equations in dyslipidaemic patients without overt vascular disease.

AIM: To compare the Framingham and Prospective Cardiovascular Munster (PROCAM) risk calculations. METHODS: We calculated the risk in 234 dyslipidaemic patients without overt vascular disease and in different subgroups. For example, the proportion of patients with coronary heart disease (CHD) risk >or= 20%, the effect of including the family history (FaHist) and of adjusting raised triglyceride (TG) levels. RESULTS: The Framingham risk was significantly (p < 0.0001) higher than the PROCAM risk (with and without including the FaHist) in different subgroups and when the TGs were adjusted to 1.7 mmol/l. The percentage of patients with CHD risk >or= 20% calculated by the Framingham (based on systolic or diastolic blood pressure) and PROCAM equations was 21.4% or 23.1% and 16.2% respectively. In the tertile with the highest PROCAM risk, the Framingham score was significantly greater than the PROCAM risk only when the FaHist was included in the Framingham calculation. When we analysed risk by gender, the Framingham score did not differ but the PROCAM risk was significantly (p < 0.0001) greater in men. When TG values were adjusted to 1.7 mmol/l, the predicted risk using PROCAM changed by 0% to -2% in all subgroups. CONCLUSIONS: In dyslipidaemic patients without overt vascular disease the Framingham model predicted a higher risk than PROCAM. Thus, the Framingham equation probably leads to substantial overtreatment compared with PROCAM. However, according to the literature, even the PROCAM equation may overestimate risk. This has considerable cost implications. New more accurate risk engines are needed to calculate risk in dyslipidaemic patients without overt vascular disease.

Metabolic syndrome is common among middle-to-older aged Mediterranean patients with rheumatoid arthritis and correlates with disease activity: a retrospective, cross-sectional, controlled, study.

OBJECTIVES: Patients with rheumatoid arthritis have an increased risk for cardiovascular disease (CVD). The prevalence of metabolic syndrome (MetS)-a major contributor to CVD-in a cohort of patients with rheumatoid arthritis and its relationship with rheumatoid arthritis related factors is investigated here. METHODS: 200 outpatients with rheumatoid arthritis (147 women and 53 men), with a mean (standard deviation (SD)) age of 63 (11) years, and 400 age and sex-matched controls were studied. MetS was assessed according to the adult treatment panel III criteria and rheumatoid arthritis disease activity by the disease activity score of 28 joints (DAS28). A standard clinical evaluation was carried out, and a health and lifestyle questionnaire was completed. RESULTS: The overall prevalence of MetS was 44% in patients with rheumatoid arthritis and 41% in controls (p = 0.5). Patients with rheumatoid arthritis were more likely to have low high-density lipoprotein cholesterol compared with controls (p = 0.02), whereas controls were more likely to have increased waist circumference or raised blood pressure (p = 0.001 and 0.003, respectively). In multivariate logistic regression analysis adjusting for demographics and rheumatoid arthritis treatment modalities, the risk of having moderate-to-high disease activity (DAS28>3.2) was significantly higher in patients with MetS compared with those with no MetS components (OR 9.24, 95% CI 1.49 to 57.2, p = 0.016). CONCLUSION: A high, albeit comparable to the control population, prevalence of MetS was found in middle-to-older aged patients with rheumatoid arthritis. The correlation of rheumatoid arthritis disease activity with MetS suggests that the increased prevalence of coronary heart disease in patients with rheumatoid arthritis may, at least in part, be attributed to the inflammatory burden of the disease.

Serum homocysteine concentration as a marker of nutritional status of healthy subjects in Crete, Greece.

BACKGROUND: Dietary habits are an important determinant of serum homocysteine (tHcy), which may be a marker rather than a cause of progression of the atherosclerotic process. The aim of the present study was to evaluate the nutritional status, and to determine the serum tHcy concentrations in healthy subjects who live in rural areas of Crete, and who theoretically follow a contemporary Mediterranean-style diet. METHODS: Serum tHcy, folate, vitamin B(12), creatinine, glucose, and the lipid profile, were measured in 203 (141 men and 62 women) healthy subjects, aged 33-78 years. The major risk factors for cardiovascular disease such as age, gender, cigarette smoking, obesity were recorded and dietary data were assessed using a 3-day weighed food intake record. RESULTS: Our population had high serum tHcy, low serum folate concentrations and lower than the traditional Cretan dietary folate intake [median (range): 12.0 (3.6-44.7) micromol L(-1), 7.9 (1.9-15.5) ng mL(-1) and 241 (68-1106) microg, respectively]. Dietary intake of fibre, omega-3, and mono- or/ polyunsaturated fatty acids was also low. An inverse relation was observed between serum tHcy concentrations and serum folate (r = -0.28; P < 0.01) and vitamin B(12) levels (r = -0.33; P < 0.001). CONCLUSIONS: Nowadays, the Cretan diet has changed towards a more westernized eating pattern. Given the analytic difficulties in determining the amount of folate in food and the inverse correlation between serum tHcy and folate levels, serum tHcy concentrations may be a useful marker for nutritional status, especially folate deficiency, in healthy subjects.