Frailty is a multideterminant syndrome in which muscle function appears to play a central role. Muscle function depends on brain control, nutrition and perfusion. We hypothesized that multiparametric MRI assessment combined with comprehensive gerontological assessment (CGA) and routine biological assessment of inflammation in a sample of older people with and without diabetes will allow to explore on one side the possibilities of multi-parametric MRI muscle and brain imaging to describe the correlates of frailty and on the other side will describe the different muscle/brain alterations due to diabetes in frailty.

The main objective is to compare the lipid percent of the rectus femoris in frail and pre-frail older subjects and in non-frail older subjects.

Frailty concept has been created to screen the older people at risk for dependency and to propose preventive intervention. Muscle function is at the centre of the concept and the majority of interventions proposed to reverse or to prevent frailty have targeted physical function. Anatomical and functional alteration of muscle, called sarcopenia is defined as a low skeletal muscle mass, a decrease in strength (dynapenia or sarcopenia is the age-associated loss of muscle strength that is not caused by neurologic or muscular diseases) and functional consequences such as low gait speed. Qualitative analysis should be associated with quantitative (mass) analysis in older subjects assessed for frailty. Muscle architecture, lipid and active tissue muscle content should be measured. Proton NMR imaging (MRI) can be used for this purpose. Brain changes were also reported to be associated with frailty. The study of structural changes associated with brain MRI alterations may better explain the frailty process.

Robust, frail and pre-frail subjects will be compared for clinical and MRI data. Grey matter volumes, white matter hyperintensities, diffusion tensor imaging data and muscle assessments relationships will be described After baseline assessment follow-up will be performed by phone calls after one month and after six months to record the number of falls and severity, the number of unscheduled hospitalization, the admission in institution for older people and death.

Inclusion Criteria:

• Robust or frail or pre-frail with at least 25% frail and 25% pre-frail according to Frieds criteria
• Barthel index > or = 60/100
• With or without diabetes mellitus, 45 to 55 % with known diabetes mellitus
• With no contraindication to undergo an MRI examination

Exclusion Criteria:

• not willing to participate
• not able to give informed consent or to understand basic instruction due to any problem (sensorial, educational, language)
• without social insurance
• with a legal protection
• with significant cognitive alteration (MMSe<21/30 or in case of low literacy <19/30)
• with a recent (2 month period) severe event: hospitalization, sepsis, stroke even with complete recovery, trauma
• with stroke sequelae (motor, speech)