Poor nutrition during fetal development can permanently alter growth, cardiovascular physiology and metabolic function. Animal studies have shown that prenatal undernutrition followed by balanced postnatal nutrition alters deoxyribonucleic acid (DNA) methylation of gene promoter regions of candidate metabolic control genes in the liver. The aim of this study was to investigate whether methylation status of the proximal promoter regions of four candidate genes differed between individuals exposed to the Dutch famine in utero. In addition, we determined whether methylation status of these genes was associated with markers of metabolic and cardiovascular disease and adult lifestyle. Methylation status of the GR1-C (glucocorticoid receptor), PPARγ (peroxisome proliferator-activated receptor gamma), lipoprotein lipase and phosphatidylinositol 3 kinase p85 proximal promoters was investigated in DNA isolated from peripheral blood samples of 759 58-year-old subjects born around the time of the 1944–45 Dutch famine. We observed no differences in methylation levels of the promoters between exposed and unexposed men and women. Methylation status of PPARγ was associated with levels of high-density lipoprotein cholesterol and triglycerides as well as with exercise and smoking. Hypomethylation of the GR promoter was associated with adverse adult lifestyle factors, including higher body mass index, less exercise and more smoking. The previously reported increased risk of cardiovascular and metabolic disease after prenatal famine exposure was not associated with differences in methylation status across the promoter regions of these candidate genes measured in peripheral blood. The adult environment seems to affect GR and PPARγ promoter methylation.

Background. We studied whether heterogeneous profiles of cognitive function are relevant to survival in patients with early Alzheimer's disease.

Methods. CAMCOG subscales of cognitive function were used as predictors of survival, together with gender in 157 consecutively referred patients with early Alzheimer's disease. Statistical analysis was performed with Cox proportional hazards analysis and Kaplan–Meier survival curves. Survival rates were compared with those in the general population.

Results. Eighty patients (51%) died during the follow-up that extended to 5·7 years, with a median survival of 4·4 years after entry. Only the praxis subscore was statistically significant related to survival (P<0·0001). Its predictive power was based on only two items, including copying ability for a spiral and a three-dimensional house, independent of age, sex, education, overall CAMCOG score, dementia severity and symptom duration. Kaplan–Meier curves for the combined score of these items (0, 1, or 2) showed three groups with significantly different survival rates for both men and women. Comparison of gender specific survival rates with data from the general population showed that excess mortality was statistically significant (P<0·01) higher in men (51%) than in women (21%) after follow-up extending to 5 years.

Conclusions. A simple test of copying ability defines subgroups of AD patients with large differences in survival rates. This suggests that parietal lobe impairment is an important predictor of mortality in AD. Also, the course of AD may be more benign in women than in men.