Relationship between systolic blood pressure and mortality in patients with severe cognitive impairment

Life expectancy is increasing worldwide, and particularly in developed countries, because of rapid and dramatic progress in medical sciences, diagnostic procedures, preventive measures, and disease treatments. The immediate consequence has been a parallel increase in the prevalence of age-related disorders, including hypertension, which affects more than 60% of people aged at least 65 years, and more than 80% of individuals aged more than 85 years [1]. Hypertension is the most frequent modifiable risk factor for cardiovascular and cerebrovascular diseases such as stroke, cognitive decline, and dementia. The correlation between vascular dementia and hypertension has been extensively reported [2–6]: cognitive decline begins during mid-life and gradually becomes more pronounced and less reversible over time [7]. It may coexist with clinical cardiovascular and renal disease induced by hypertension and the commonly associated cardiovascular risk factors, which seem to play a major role in the development and progress of cognitive decline and dementia. For this reason, cardiovascular disease is also the most common cause of death in people with vascular dementia, Alzheimer's disease and mixed dementia [8,9]. However, aging is a lengthy process in which the relationships between blood pressure (BP) and vascular damage change over time. In middle age, the closest relationships between hypertension and arterial lesions are established by the use of DBP. However, between 60 and 80 years of age, the closest relationships are between SBP, as a consequence of the stiffness of the large arteries, whereas in people in their 80s and 90s, the fragility and bed-bound situation of many study participants means that standing SBP tends to decrease. For this reason, the relationship between BP and cognitive decline in the very elderly is represented as a J curve in which cognitive decline and dementia are observed in patients with either very low or very high SBP. As both low SBP and cognitive impairment are separately associated with an increased risk of mortality [10], the mortality risk associated with BP may differ with respect to the level of cognitive function. In this issue of the Journal, Weidung et al.[11] report interesting results on this topic. The authors investigated the association between SBP and all-cause mortality in 1115 individuals aged more than 85 years with differing Mini Mental State Examination (MMSE) scores. Study participants were recruited from a representative sample of the Swedish and Finnish populations participating in the GERDA (prospective Umeå 85+/GErontological Regional DAtabase) study. After 2 years of follow-up, 26% of these very elderly study participants died and the highest mortality risk was observed among those with the most severe dementia. Interestingly, BP was independently associated with the mortality risk in participants with MMSE scores of 0–10, but the relationship in these study participants with severe cognitive impairment was not linear and increased in patients with SBP values less than 125 mmHg or more than 165 mmHg compared with intermediate SBP (126–139 mmHg) values, thus following a J curve. The association between BP and mortality has not previously been analyzed separately in individuals with differing MMSE scores, and particularly in those with severe cognitive impairment with MMSE 10 or less, and this is one of the major strengths of the study. In previous studies, an independent association between low SBP and 5-year mortality was described in very old individuals with moderate cognitive impairment and MMSE scores less than 24, but the results of the study by Weidung et al.[11] suggest that the association between low SBP and mortality is found in individuals with much greater cognitive impairment than previously known and, therefore, with much greater fragility. Atrial fibrillation, depression, and dependency, which are very common in these very old individuals, seem to influence the mortality risk associated with low SBP. The increased mortality risk observed with high and low SBP in individuals with very severe cognitive impairment is more closely associated with intense structural damage of the very small cerebral arteries, in the form of white matter lesions, lacunar infarcts, microbleeds, and enlarged perivascular spaces, and with large artery stiffness [12]. These silent lesions are clearly associated not only with high SBP levels but also with excessive nocturnal systolic reduction and arterial stiffness. In addition, elderly study participants with very severe cognitive impairment are more likely to have reduced cerebral BP autoregulation because of severe dementia [13], which may enhance cerebrovascular damage caused by the dissociation between cerebral blood flow and oxygen demand. Therefore, older individuals with high or low SBP in combination with reduced cerebral BP autoregulation may be at greater risk of neuronal damage. The study has several limitations: BP was measured in the supine position, which overestimates BP values compared to the seated position in elderly study participants in whom orthostatic hypotension is very frequent. The 24-h BP monitoring could have provided valuable information about BP variability, nocturnal BP, and episodic hypotension, but was not obtained. Women were underrepresented in the study sample, preventing the generalizability of the results. Finally, MMSE is much more sensitive in detecting severe cognitive decline than in detecting mild cognitive impairment. Despite these limitations, the findings of a nonlinear relationship between SBP and mortality in very old study participants with severe dementia have implications for the treatment of these patients. In this population, which is especially susceptible to adverse drug reactions and polypharmacy, treatment individualization is particularly important. The Secondary Prevention of Small Subcortical Strokes trial in patients after a lacunar stroke has suggested that the lowest risk of recurrent stroke occurred at a nadir of about 124 mmHg SBP [14]. This lower limit value for secondary prevention of small-vessel stroke coincides, curiously, with the SBP values below which the mortality risk increases in patients with severe cognitive impairment [11]. The results of the SPRINT trial [15], which showed that a more intensive SBP reduction to less than 125 mmHg was associated with a significant reduction in mortality in hypertensive patients aged more than 75 years, must be interpreted with special caution in patients with severe cognitive impairment, in whom the paper by Weidung et al.[11], has shown a J-shaped curve between mortality risk and SBP. ACKNOWLEDGEMENTS Conflicts of interest There are no conflicts of interest.