Amlodipine and Cognitive Function: What the Evidence Actually Shows

What the label records

The Norvasc adverse reaction tables come from placebo-controlled trials of 1,730 patients on amlodipine up to 10 mg daily and 1,250 on placebo [1]. Somnolence was reported by 1.4% versus 0.6%, and fatigue by 4.5% versus 2.8% [1]. Dizziness was dose related: 1.1% at 2.5 mg, 3.4% at 5 mg and 3.4% at 10 mg, versus 1.5% on placebo [1]. Among events seen in under 1% of patients, where a causal relationship is uncertain, the label lists insomnia, nervousness, depression, abnormal dreams, anxiety and depersonalization [1]. These are reported symptoms, not scores on tests of attention or memory.

ASCOT-BPLA measured strokes, not memory

ASCOT-BPLA randomized 19,257 hypertensive adults aged 40 to 79 with at least three other cardiovascular risk factors to amlodipine 5 to 10 mg with perindopril 4 to 8 mg added as required (n=9,639) or atenolol 50 to 100 mg with bendroflumethiazide added as required (n=9,618), and stopped early at a median of 5.5 years [2]. Fatal and nonfatal stroke occurred in 327 versus 422 patients (unadjusted HR 0.77, 95% CI 0.66 to 0.89, p=0.0003); the primary endpoint was not significantly different, at 429 versus 474 (HR 0.90, 95% CI 0.79 to 1.02, p=0.1052) [2]. The endpoints reported were cardiovascular, not memory or dementia [2]. Perindopril was added as required in the amlodipine arm, so the comparison is regimen-level rather than amlodipine alone [2]. The label ties blood pressure lowering to fewer fatal and nonfatal cardiovascular events, primarily strokes and myocardial infarctions [1].

Cohort data on blood pressure drugs and dementia

The Rotterdam Study followed 6,249 participants (mean age 68.4) from 1990-1993 through 2005 using pharmacy records of filled prescriptions [3]. Compared with never-use, antihypertensive use was associated with an adjusted hazard ratio for all dementia of 0.95 per year of use (95% CI 0.91 to 0.99); the reduction was 8% per year of use for people 75 or younger (95% CI -15% to -1%) and 4% for people older than 75 (95% CI -11% to 4%) [3]. The investigators observed no apparent differences among antihypertensive drug types [3]. As an observational study, it cannot separate a drug effect from the reasons some people take antihypertensives and others do not [3].

Label cautions on blood pressure and titration

The label warns that symptomatic hypotension is possible, particularly in severe aortic stenosis, though acute hypotension is unlikely because the drug's onset is gradual [1]. The label says that in general one should wait 7 to 14 days between titration steps, and to titrate slowly in severe hepatic impairment, where the half-life is 56 hours [1].