Three quarters of patients with Alzheimer's disease have at least one vascular risk factor (VRF). Vascular brain lesions are present in most Alzheimer's patients (especially older ones). This cerebrovascular disease potentiates Alzheimer's lesions in early-stage disease. Many research studies have shown that VRFs are also risk factors for Alzheimer's disease; this is true for arterial hypertension and dyslipidaemia in particular and, to a lesser extent, diabetes and cardiopathy. Moreover, recent drug trials (SYST-EUR, PROGRESS and HOPE) have indicated that antihypertensive medications can prevent the appearance of dementia (and notably Alzheimer's disease) in over-60 hypertensive subjects. An observational study of 233 Alzheimer's patients with an average follow-up period of 4 years has shown that the annual decline in the Mini-Mental State Examination (MMSE) score was lower in patients in whom all the VRFs were being treated than in patients in whom no VRFs were being treated (1.5 ± 2.5 points versus 2.5 ± 2 points, respectively; p<0.04).1 However, it is not currently known whether optimal treatment of VRFs can influence the progression and prognosis of Alzheimer's disease. Answering this question could have a significant impact on public health.

It is not currently known whether the optimum treatment of VRFs influences the progression and prognosis of Alzheimer's disease. Our starting hypothesis is that VRF control in Alzheimer's patients is associated with slower cognitive decline, less intense loss of personnel independence and fewer adverse events over the course of the disease (cardiovascular or cerebrovascular events, behavioural disorders, caregiver burden, hospitalization and death).

COVARAD study is a randomized, controlled, multicentre study comparing 2 VRF care strategies in mild-to-moderate (MMSE > 18) Alzheimer's disease patients with at least one VRF. The objective of this work is to evaluate the effect of "optimal" care strategy, in strict compliance with the French HAS guidelines concerning targets for blood pressure, glycaemia and blood lipid levels, on the cognitive function in mild-to-moderate Alzheimer's patients (MMSE score > 18), in comparison with a control group (i.e. receiving standard care from a primary care physician). The study test the hypothesis whereby "optimal" care of the 3 main modifiable VRFs is associated with slower cognitive decline in Alzheimer's disease patients (evaluated on the ADAS-cog score), when compared with standard care and to compare the MMSE, MoCA and VADAS-cog scores, mood and behaviour (MADRS and NPI), loss of independence (ADCS-ADL), the occurrence of cardiovascular or cerebrovascular events, the number and length of hospitalisations, caregiver burden (on the Zarit scale), institutionalization and survival in the two groups (i.e. depending whether VRFs are managed optimally or not).

This study could influence clinical practice. If VRF control does have an influence on the progression of Alzheimer's disease, an information campaign could modify practice and have a significant impact on public health.

An independent Data and Safety Monitoring Board will be set up to monitor the diabetic patients, in view of the risks related to "optimal" care (ACCOR and ADVANCE studies). Nevertheless, the risk of adverse events will be limited by raising the threshold value for glycated haemoglobin to 8%.

Inclusion criteria

• Subjects aged 60 or over
• Subjects with Alzheimer's disease, according to the NINCDS/ADRDA diagnostic criteria 71
• MMSE > 18
• Subjects with at least one VRF (whether treated or not): arterial hypertension (defined as SBP/DBP ≥ 140/90 mmHg in at least three different consultations or, for ambulatory measurements, > 130/80 mmHg with a Holter recorder or > 135/85 mmHg with a self-measurement device), type 2 diabetes (defined as a glycaemia value over 1.26 g/l (7 mmol/l) after an 8-hour fast (confirmed on two occasions), dyslipidaemia (defined as an LDL cholesterol level > 1.6 g/l or 1.3 or 1 g/l, depending on the patient's risk level)
• Subjects having agreed to participate in the study (provision of informed consent).
• Subjects accompanied by a person likely to provide information on the patient (during the visit or over the phone).

Exclusion criteria

• Any other disease that might interfere with the evaluation of cognitive disorders.
• No formal education or a poor understanding of French (interfering with administration of the neuropsychological tests).
• Major physical problems likely to interfere with administration of the tests (poor eyesight, hearing, etc.).
• Non-Alzheimer's dementia (isolated vascular dementia, Lewy body dementia, frontotemporal dementia, etc.)
• Psychotropic drugs likely to modify the patient's non-stabilized cognitive state.
• Patients with a history of cardiovascular events can be included (randomization will be balanced in terms of this criterion).
• Participation in a therapeutic clinical trial during the study period.