AVEC 2012.

Study characteristics
Methods	Single centre study. Three‐arm double‐blind randomised controlled trial. Randomisation: computer‐generated random allocation sequence. Study personnel and participants blinded to allocation. Study duration: 1 year.
Participants	Geographical region: USA. Setting: community. Aged ≥ 60 years. BP entry criteria: SBP > 140 mmHg or DBP > 90 mmHg or receiving antihypertensive therapy. Early cognitive impairment without dementia defined as executive impairment, as measured using the CLOX test or early memory impairment as measured using the Repeatable Battery for the Assessment of Neuropsychological Status.
Interventions	Three treatment arms with starting doses: lisinopril 10 mg once daily; candesartan 8 mg once daily; hydrochlorothiazide 12.5 mg daily. Participants seen fortnightly, average of two seated BP measurements taken, medications up‐titrated until BP < 140/90. Lisinopril, candesartan and hydrochlorothiazide titrated up through 20 mg and 40 mg, 16 mg then 32 mg and 25 mg respectively. Next step long acting nifedipine added at 30 mg titrated through 60 mg to 90 mg then long acting metoprolol added at 12.5 mg titrated through 25 mg to 50 mg. Once target BP reached, participants followed for 12 months.
Outcomes	Analysed in this review: Trail Making Test A and B; Hopkins Verbal Learning Test ‐ Revised; Digit Span Test Not analysed in this review: health questionnaire; instrumental activities of daily living; cerebral blood flow haemodynamics
Notes	Exclusion criteria: untreated SBP > 200 mmHg or DBP > 110 mmHg or treated SBP > 180 mmHg or DBP > 100 mmHg; MMSE < 20 or clinical diagnosis of dementia or Alzheimer's disease; raised serum creatinine or potassium; treatment with > 2 antihypertensive agents; congestive heart failure; diabetes; stroke. Of the n = 18 (ACE inhibitor), n = 20 (ARB) and n = 15 (diuretic) participants recruited, n = 11, n = 9 and n = 11, respectively, completed 12‐month follow‐up.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: 'this random allocation is performed at the central pharmacy...randomisation is performed using computer generated random numbers leading to a random allocation sequence'
Allocation concealment (selection bias)	Low risk	Quote: 'this random allocation is performed at the central pharmacy...randomisation is performed using computer generated random numbers leading to a random allocation sequence'
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: 'both study personnel and participants are blinded to the group assignment'
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: 'both study personnel and participants are blinded to the group assignment'
Incomplete outcome data (attrition bias) All outcomes	High risk	Baseline data reported for the 47 participants of the 53 participants with successful middle cerebral artery isolation at baseline. Only 47/53 and 31/53 completed 6‐month and 12‐month follow‐up, respectively (40/47 and 29/47 with reported outcomes). No analysis of influences on loss to follow‐up reported.
Selective reporting (reporting bias)	Unclear risk	Intention to treat analyses used. Only significant differences in cognitive outcomes between the groups reported.
Other bias	Unclear risk	Study protocol published, stated aiming for 100 participants, but only 63 eligible individuals screened. 53 randomised but results reported for n = 47. Described as a pilot study. Study aimed to recruit those with early cognitive impairment but only MMSE used to exclude participants with more advanced cognitive impairment, diagnosis of dementia given as exclusion criteria but no mention of criteria for making such a diagnosis. Mean (SD) MMSE are relatively high however. Add‐on antihypertensive therapy utilies for n = 1 each in ARB group and diuretic group and n = 3 in ACE inhibitor group.