What's new?

This study is the first in Poland to comprehensively analyze and describe lipid disorders among older adults aged 65 and above. It provides detailed epidemiological data on key lipid fractions across distinct age groups of seniors, highlighting significant sex- and age‑related differences. The work also explores the relationships between hypercholesterolemia and other cardiovascular risk factors. These novel observations emphasize the importance of addressing lipid disorders in the elderly population to improve cardiovascular outcomes.

Introduction

According to the World Health Organization, cardiovascular disease (CVD) accounts for nearly 31% of all deaths worldwide, making it the leading cause of death. Each year, CVD kills more people than any other disease, a statistic that has remained consistent for nearly a decade.¹

Despite significant progress in preventing premature mortality, CVD remains the leading cause of death in Poland. In 2013, data from the Polish Central Statistical Office showed that CVD was responsible for approximately 46% of all deaths in the general population and 53.1% among individuals over 65 years old.² By comparison, the average CVD‑related mortality rate in the European Union stands at 40%.³

To understand the morbidity and mortality associated with CVD, it is necessary to assess the distribution of its key risk factors. Well‑established contributors, such as hypertension, lipid disorders, abdominal obesity, diabetes, and smoking are the main causes of CVD in adult populations.⁴ The prevalence of these risk factors increases with age, and their combination significantly accelerates atherosclerosis, leading to a logarithmic growth in CVD risk.

Total cholesterol (TC) level is an independent predictor of coronary heart disease (CHD) in men over 65 years old.⁵ A study by Benfante et al,⁵ involving 1480 men over 65 years of age, demonstrated a progressive increase in CHD risk with rising TC levels. Furthermore, a meta‑analysis by the Prospective Studies Collaboration⁶ showed that reducing TC concentration by 1 mmol/l decreased CHD‑related mortality by one‑sixth in those aged 70 to 89 years, irrespective of sex.

The relationship between cardiovascular risk and individual risk factors in adults is well‑documented in the literature. However, understanding these dynamics in older adult populations is equally critical. This paper presents the data on the prevalence of lipid disorders among older adults in the Polish population, a representative cohort of Central and Eastern Europe, which is characterized by some of the highest CVD mortality rates in the world.^7,8

Patients and methods

The design, inclusion criteria, examination procedures, and study cohort structure for the PolSenior survey have been described in detail elsewhere.⁹ Briefly, the PolSenior survey was a cross‑sectional study aimed at assessing the prevalence of major medical, psychological, and socioeconomic issues in a representative group of older adults in Poland. Participants were selected using a 3‑stage stratified proportional draw.⁹ The overall response rate for the survey was 43%. The study included 4101 respondents (2136 men and 1965 women) who provided blood samples for lipid measurements. The sample was divided into 6 equally sized age groups: 65–69, 70–74, 75–79, 80–84, 85–89, and ≥90 years.

Data collection for the PolSenior survey took place between 2008 and 2009. The participants underwent examinations during 3 separate home visits conducted by well‑trained nurses. These visits included completing questionnaires, measuring blood pressure (BP), performing anthropometric assessment (described in detail elsewhere⁹), and collecting blood and urine samples. The methodology for BP measurements is outlined in a dedicated publication.¹⁰

Hypertension was diagnosed according to the 2018 European Society of Hypertension / European Society of Cardiology guidelines for the management of arterial hypertension.¹¹ This included cases where the average systolic and diastolic BP values from 2 measurements during each visit were equal or higher than 140 mm Hg and / or 90 mm Hg, respectively, or if the patient had been using antihypertensive medication within the past 2 weeks due to a prior hypertension diagnosis.

Overweight was defined as body mass index (BMI) of 25 to 29.9 kg/m², while obesity was classified as BMI of at least 30 kg/m². Visceral obesity was defined as a waist circumference of at least 80 cm in women and at least 94 cm in men.

Blood samples were collected at participants’ homes following 10 to 12 hours of fasting. Serum was separated and frozen at a local laboratory within 2 hours of collection. Serum TC, high‑density lipoprotein cholesterol (HDL‑C), and triglyceride concentrations were measured using a Modular PPE Analyzer (Roche Diagnostics GmbH, Mannheim, Germany).

Serum TC concentrations were determined using an enzymatic method with cholesterol esterase and cholesterol oxidase. HDL‑C levels were measured using accelerator selective detergent, which facilitates accelerated non–HDL‑C oxidation and HDL‑C dissolving. The concentrations of low‑density lipoprotein cholesterol (LDL‑C) were calculated using the Friedewald formula: LDL‑C [mg/dl] = TC [mg/dl] – HDL‑C [mg/dl] – triglycerides [mg/dl]/5. This formula was not applied for triglyceride concentrations exceeding 400 mg/dl. Triglyceride concentrations were determined by an enzymatic method using glycerol kinase and glycerol‑3‑phosphate oxidase.

Hypercholesterolemia was defined as either TC concentrations greater than 190 mg/dl (>5 mmol/l) or the use of statins or fibrates. Elevated LDL‑C was diagnosed for concentrations equal to or greater than 115 mg/dl (≥3 mmol/l). Abnormal serum HDL‑C levels were defined as less than 45 mg/dl (<⁠1.2 mmol/l) in women and less than 40 mg/dl (<⁠1 mmol/l) in men, excluding those on statin therapy. Triglyceride concentrations were considered elevated if equal to or greater than 150 mg/dl (≥1.7 mmol/l). The criteria for diagnosing abnormal lipid and lipoprotein concentrations are in line with the 2016 European Society of Cardiology / European Atherosclerosis Society guidelines for the management of dyslipidemias.¹²

To address additional study objectives, a questionnaire survey was conducted, recording responses related to hypercholesterolemia awareness and the use of statins or fibrates in the preceding 2 weeks. The questionnaire also collected demographic and socioeconomic data, including the level of education, occupation, place of residence, and economic status.

For analyses utilizing multivariable logistic regression models, diabetes diagnosis was defined as a prior diagnosis and / or the use of glucose‑lowering therapy. The study protocol was approved by the Bioethics Committee at the Medical University of Silesia in Katowice, Poland (KNW‑6501‑38/I/08). All participants provided written consent.

Results

Discussion

This study shows the prevalence of hypercholesterolemia in the population of individuals aged 65 years and older, categorized into 6 age groups (65–69,70–74, 75–79, 80–84, 85–89, and ≥90 years). To the best of our knowledge, the PolSenior survey is the first large‑scale, population‑based study in Central and Eastern Europe that provides detailed insight into the prevalence of hypercholesterolemia across the older population, including those aged 90 years and older.

Despite the growing body of research on lipid disorders, data specific to seniors remain relatively limited compared with younger populations, posing challenges in developing evidence‑based recommendations, particularly for primary prevention in this demographic.¹³

The mean serum TC concentration in women increased in all age groups, whereas in men aged over 80 years, these levels were within the reference range (Table 2). In both sexes, TC levels showed a steady decline from the youngest to the oldest age. Notably, in each age category, women exhibited higher mean TC levels than men (Table 2). This discrepancy in TC levels may partially contribute to the earlier excess mortality in men, typically occurring before the age of 65 years.

The Dubbo study,¹⁴ conducted on a cohort of 1237 elderly individuals in Australia (divided into 3 age groups: 60–69, 70–79, and ≥80 years), similarly showed higher concentrations of cholesterol in women compared with men, with levels declining with age, particularly among men.

When comparing absolute values, the Polish population over 80 years old demonstrated significantly lower mean TC values than the Australian cohort—approximately 30 mg/dl lower in men and 40 mg/dl lower in women. Trends in serum LDL‑C levels mirrored those of TC, showing a general decline with advancing age, with mean LDL‑C levels consistently lower in men, as compared with women.

Comparable findings were reported in the Framingham Study over a 30‑year follow‑up,¹⁵ where cholesterol values in women were markedly higher than in men, a trend apparent in populations over 50 years old. In both sexes, TC and LDL‑C concentrations decreased in individuals over 65 years of age.

It is important to note that serum LDL‑C levels calculated using the Friedewald formula may underrepresent actual concentrations. Balder et al¹⁶ observed that LDL‑C levels measured directly were 7 mg/dl higher in men and 6 mg/dl higher in women, as compared with values calculated using the Friedwald formula.

In our study, HDL‑C levels tended to remain stable across age groups in men, while in women a minor decline with age was observed. However, the difference in mean HDL‑C values between sexes persisted. Comparable results were seen in the Australian population studied in the Dubbo cohort. In contrast, data from the Framingham cohort revealed that HDL‑C levels remained constant until the advanced age in both sexes.¹²

In the Irish population, sex‑related differences in lipid profiles emerged before the age of 20 years and persisted throughout life.¹⁷ In both sexes, TC and LDL‑C levels increased until middle age, then gradually decreased toward old age. The concentrations were higher in men than in women up to middle age, after which the trend was reversed, with women showing higher concentrations.

Median triglyceride values showed similar age- and sex‑related distribution to TC, increasing with age until middle age and then declining, though the distribution showed a marked positive skew, particularly among middle‑aged men. The levels of HDL‑C, in contrast, remained relatively stable throughout life, with men having lower levels than women.¹⁷

Balder et al¹⁶ in their analysis of the Dutch population, observed a gradual decrease in serum LDL‑C concentrations in men with age, consistent with trends noted in our study. However, the pattern in women was less definitive, with no clear trend for LDL‑C levels decreasing with age.

The multivariable logistic regression analysis showed that the probability of hypercholesterolemia decreases with age among men, with a similar trend observed in women, but only among the oldest age groups. The interpretation of this pattern poses a challenge and is not well‑supported by the current literature.

One possible explanation could be the higher cardiovascular risk in men, which may lead to earlier mortality among those with hypercholesterolemia, effectively “selecting out” individuals with higher lipid levels from the older male population. For women, the delayed onset of this trend might reflect their comparatively lower cardiovascular risk at younger ages, allowing the survival of individuals with hypercholesterolemia into older age.

The fact that hypercholesterolemia is less likely to be diagnosed in both sexes in the presence of diabetes and in men who are smokers can be interpreted in a similar way. A review of the literature revealed no reports on this phenomenon, highlighting the need for further in‑depth research.