Prevalence of diagnosed and undiagnosed cardiovascular disease burden in community dwelling 85+ year olds

Abstract

Objectives: The prevalence of cardiovascular diseases including heart failure (HF), atrial fibrillation (AF) and peripheral vascular disease (PVD) rises sharply among those aged 85 years and over, who now constitute the most rapidly increasing age group worldwide. Majority of this disease burden remains undiagnosed. Most previous community-based studies of left ventricular (LV) dysfunction and HF included only small numbers in this age group. We conducted a community-based study of 85+ year olds using domiciliary echocardiography, electrocardiography and ankle brachial index (ABI) assessments to estimate the prevalence of LV dysfunction, AF and PVD. We cross-referenced our findings to pre-existing HF, AF and PVD diagnoses present in general practice (GP) medical records to estimate the proportion of undiagnosed cardiovascular pathology. We also assessed to diagnostic performance of NT-proBNP to detect underlying LV dysfunction. Design: Cross-sectional analysis of data from Newcastle 85+ Cohort Study. Setting: Primary care, North-East England. Participants: 427 men and women (60.9% women) aged 85+ years and above, from Newcastle 85+ Study. Measurements: Assessment was conducted in home setting. 2-D and Doppler echocardiography was performed, with LV systolic and diastolic function graded according to American Society of Echocardiography guidelines. A dyspnoea questionnaire was used to assign New York Heart Association (NYHA) functional severity class. ABI measurement and other measures to assess arterial stiffness including pulse wave velocity and pulse wave analysis were carried out by portable sphygmoCor and vicorder devices. Bloods samples were taken for NT-proBNP levels. Previous diagnoses of HF, AF and PVD were abstracted from the GP medical records. Results: Normal LV function (ejection fraction greater than 55% and normal/mildly impaired diastolic function) was found in just 37.2% of participants. 48.4% had LV systolic dysfunction and 14.4% had isolated diastolic dysfunction. 66.1% people with underlying LV systolic or diastolic dysfunction, had symptoms of breathlessness (NYHA II or above). Overall 37.4% of participants had undiagnosed symptomatic significant LV dysfunction (29.5% systolic, 7.9% isolated diastolic). 23.8% of participants with pre existing diagnosis of HF, had no echocardiographic evidence of underling systolic or diastolic dysfunction. Markers of arterial stiffness were not significantly associated with LV dysfunction. Diagnostic performance of NT-proBNP to detect underlying symptomatic or asymptomatic dysfunction was not robust. Prevalence of peripheral vascular disease was 22.1%. 19.0% of participants who had no formal GP diagnosis of PVD had definite PVD on the basis of ABI assessment. Prevalence of atrial fibrillation was 25.5% in the entire cohort. Nearly half (53.2%) of these patients had had no existing GP diagnosis of AF. 87.2% participants with AF were CHA2DS2-VASc score 3 or above. Only 15.6% participants with AF were taking warfarin. In remaining 84.4% participants with AF, who were not on warfarin, only 42.4% participants were taking antiplatelet medications. Conclusions: Systolic and diastolic LV dysfunction was much commoner in 85+ year olds than most previous studies have suggested, affecting around half of a community-dwelling sample; the majority of cases were symptomatic. Despite a national initiative to improve heart failure management within primary care in England, over 80% of very old people with symptomatic significant LV dysfunction remained undetected. Prevalence of AF and PVD is also much common in that rapidly expanding fraction of population, majority of which remains undiagnosed. There is need to establish effective ways to detect this cardiovascular disease burden in very old population.