Study population

This study is a part of the Białystok PLUS longitudinal general population‑based study from Poland that has been ongoing since 2018. Details of the study were described previously by Chlabicz et al.¹⁶ Briefly, citizens aged 20–79 years were randomly selected from the population of a mid‑sized (approximately 300 000 inhabitants) central European city (Bialystok, Poland) in such proportions as to obtain a distribution reflecting the city’s population. There are no exclusion criteria; however, the visits were delayed when a participant had a recent (up to 6 weeks) infection or medical procedure, and some tests have not been performed in pregnant women or in certain subpopulations, for example, oral glucose tolerance test (OGTT) in patients with diabetes.¹⁶ Our analysis included 2110 participants enrolled in the Białystok PLUS cohort study in the years 2017–2023.

Data collection

The data were collected through an analysis of medical interviews as well as laboratory and screening tests. The details, including patient medical history and lifestyle, were acquired through extensive questionnaires. Physical measurements of height and waist and hip circumference were taken with a SECA 201 tape (SECA, Hamburg, Germany). Waist‑to‑hip ratio (WHR) was defined as the relation between hip and waist circumference. Weight measurement was performed with a bioimpedance device InBody 770 (InBody, Eschborn, Germany). The World Health Organization guidelines¹⁷ were used to classify the population into body mass index (BMI) categories (Supplementary material, Table S1). Blood pressure (BP) measurements were taken with the oscillometric method applied by Omron Healthcare Co., Ltd, MG Comfort (Omron Healthcare Co., Ltd., Kyoto, Japan) after 5 minutes of rest. Percentage body mass composition was assessed using InBody 770 and Dual‑energy X‑ray absorptiometry (DEXA; Lunar iDXA, GE Healthcare, Fairfield, Connecticut, United States). The presence of any atherosclerotic plaques was evaluated in the 1) right common carotid artery (CCA), 2) left CCA, 3) right external carotid artery (ECA), 4) left ECA, 5) right internal carotid artery (ICA), 6) left ICA, 7) right bifurcation (BIF), and 8) left BIF. We assessed atherosclerotic plaques as binomial quality variables, and marked them as present when 1 of the following criteria was fulfilled: 1) local increase of intima media thickness (IMT) toward the lumen of the vessel, exceeding the surrounding IMT by over 0.5 mm, 2) local increase of IMT toward the lumen of the vessel, surpassing the surrounding IMT by 50%, 3) IMT increase by over 1.5 mm.²⁴

Moreover, peripheral venous blood samples were collected in the morning after at least 8 hours of fasting and declaration of appropriate sleep through the night. The samples were then prepared for further analysis by centrifugation and stored at −70 °C. A detailed description of the methods used to evaluate biochemical parameters in the blood samples is presented in our previous publication.¹⁸

Metabolic categorization

We assessed the occurrence of MetS in the general population of Bialystok residents, based on a recent statement by Polish experts.¹⁴ The 2022 definition states that obesity, represented by BMI of 30 kg/m² or higher or waist circumference of 88 cm or greater in women and 102 cm or greater in men, is a prerequisite. Additionally, obesity must be accompanied by at least 2 of the following: 1) systolic BP (SBP) of at least 130 mm Hg and / or diastolic BP (DBP) of at least 85 mm Hg on office measurement and / or antihypertensive treatment; 2) fasting glucose of at least 100 mg/dl and / or at least 140 mg/dl after 2 hours in OGTT and / or glycated hemoglobin (HbA_1c) level of at least 5.7% or glucose‑lowering treatment; 3) non–high‑density lipoprotein cholesterol (non–HDL‑C) level of at least 130 mg/dl or lipid‑lowering treatment.

Then, we assessed the occurrence of MetS based on the 2009 Joint Interim Statement.¹³ To diagnose MetS, the joint presence of a minimum 3 out of 5 of the following is necessary: 1) waist circumference of 80 cm or greater in women or 94 cm or greater in men; 2) SBP of at least 130 mm Hg and / or DBP of at least 85 mm Hg; 3) fasting glucose of at least 100 mg/dl or glucose‑lowering treatment; 4) HDL‑C below 50 mg/dl in women or below 40 mg/dl in men or lipid‑lowering treatment; 5) triglyceride (TG) level of at least 150 mg/dl or treatment with fibrates.

Additionally, we assessed the occurrence of CKM syndrome based on the AHA statement.¹⁵ A full description of the categorization is provided in Supplementary material.

Cardiovascular risk assessment

The Systematic Coronary Risk Estimation 2 (SCORE2) and Systematic Coronary Risk Estimation 2‑Older Persons (SCORE2‑OP) were used to calculate cardiovascular risk in primary prevention.²¹ The SCORE2 and SCORE2‑OP estimate an individual 10‑year risk of fatal and nonfatal CVD events (myocardial infarction, stroke) in apparently healthy people. The SCORE2 and SCORE2‑OP were calculated, excluding the participants who were prequalified into the high and very high cardiovascular risk classes based on the abovementioned guidelines, that is, the participants with CVD or diabetes (diagnosed previously or at the time of the study entry), moderate or severe CKD at the time of the study entry, and younger than 40 years old. The SCORE2 was calculated for the individuals aged 40–69 years, while the SCORE2‑OP was calculated for those aged 70–89 years. Poland belongs to high CVD risk countries, and an appropriate calculator was used.²¹ The study population was divided according to the latest recommendations listed in the 2021 European Society of Cardiology Guidelines on CVD prevention in clinical practice²¹ into low‑to‑moderate, high, and very high CVD risk categories. The participants younger than 40 years who were not prequalified were excluded from the cardiovascular risk assessment.

Statistical analysis

We utilized the complete available dataset. Some data were missing due to technical reasons. These included lipoprotein(a) (Lp[a]) for 279 idividuals and active smoking for 52 individuals, and no imputation was performed for these missing values. Data on 2‑hour glucose levels (n = 232) and body fat percentage assessed by DEXA (n = 256) were unavailable due to contraindications that precluded these measurements.

Continuous variables were described by mean (SD) and median (interquartile range [IQR]). Categorical variables of baseline characteristics were shown as frequency (percentage). The baseline characteristics of all participants based on BMI and MetS categories were compared. Statistical analysis for continuous and categorical variables was performed using the Kruskal–Wallis or the χ² test, respectively. The comparisons between the groups were made using the post hoc Dwass–Steel–Critchlow–Fligner (DSCF) test for continuous variables. Comparisons of proportions between the groups provided as post hoc tests for χ² tests were assessed using tests for 2 proportions with the Bonferroni adjustment for multiple comparisons. Classification value of selected parameters with both definitions was assessed with the area under the curve (AUC) of the receiver operating characteristic (ROC). The statistical analyses were performed with SPSS sotware, version 27.0 (IBM, Armonk, New York, United States). Calculation of the DSCF tests was done using R package, version 4.4.2 (R Foundation for Statistical Computing, Vienna, Austria) based on the methodology described in the literature.³¹ All P values were 2‑sided, and deemed significant when below 0.05.

Ethics

Ethical approval for the study was given by the local Ethics Committee (R‑I‑002/108/2016). Written informed consent was obtained from all participants. The study identifier in the Trial Registration and Ethical Issues ClinicalTrials.gov database is NCT03197363.

Study population and the prevalence of metabolic syndrome in the general population

The cross‑sectional analysis included 2110 participants (mean [SD] age, 49.3 [15.3] y; 44.3% men), with mean (SD) BMI of 27.1 (5) kg/m². Figure 1 illustrates the relationships between the diagnosis of MetS based on different criteria. Among the 2110 study participants, 466 (22.1%) met the diagnostic criteria for MetS according to both definitions. Additionally, 109 individuals (5.1%) were identified solely by the 2022 guidelines, while 276 (13.1%) met only the 2009 criteria. Notably, 1259 participants (59.7%) did not meet the MetS criteria under either definition. Specifically, the updated Polish criteria, which include obesity as a prerequisite, resulted in fewer overall MetS diagnoses, but also identified a subset of individuals at a metabolic risk (n = 109; 5.2%) not captured by the 2009 criteria. Conversely, 276 participants (13.1%) met only the older criteria, typically due to differences in waist circumference or lipid thresholds.

Table 1 and Supplementary material Table S2 present baseline characteristics of the study group, stratified by the presence of MetS. The individuals diagnosed with MetS based on either definition were significantly older than those without MetS. The highest mean (SD) age (58.6 [12.5] y) was observed in the participants fulfilling both criteria, followed by those meeting only the 2022 (57.5 [11.9] y) or 2009 (55 [13.4] y) criteria, in contrast to the non‑MetS group (43.9 [14.5] y). The distribution of sex also differed significantly across the groups: men were markedly overrepresented in the 2009‑only MetS group (67.8%), and underrepresented in the 2022‑only MetS group (27.5%). Anthropometric variables, including BMI, WHR, and body fat percentage, were higher in the participants with MetS, reaching the highest levels among those meeting both definitions. The 2022‑only MetS group demonstrated the highest mean BMI and body fat percentage among all subgroups.

Lipid profiles varied significantly across the groups. HDL‑C levels were consistently the lowest in all MetS groups, whereas total cholesterol, low‑density lipoprotein cholesterol (LDL‑C), and TG levels were the highest, particularly among the individuals meeting only the 2022 definition. In contrast, glycemic parameters followed a different pattern. The highest fasting glucose levels were observed in the participants who fulfilled both MetS definitions. A significant difference in the prevalence of impaired fasting glucose was noted between the former and updated criteria. We also observed that the TG/HDL‑C ratio was lower in the individuals meeting only the 2022 criteria than in those identified based on the 2009 definition. The ratio was significantly higher in the group meeting only the 2009 criteria than in the individuals fulfilling both definitions, and significantly lower in the group classified solely according to the 2022 criteria. Among current smokers, the majority were classified as not having MetS. A significant difference was also observed in medication use between the analyzed groups; the group meeting both MetS criteria had the highest proportion of individuals receiving both treatment for diabetes and lipid‑lowering therapy.

Association of metabolic syndrome and atherosclerosis

Subsequently, we evaluated the association of MetS with preclinical atherosclerosis and cardiovascular risk, as shown in Table 1. The prevalence of carotid atherosclerotic plaques was substantially higher among the individuals with MetS, regardless of the definition applied. Moreover, ICA thickening over 50% was most frequently observed in the participants who fulfilled both the 2009 and 2022 diagnostic criteria for MetS. Across all cardiovascular risk categories, the individuals without MetS differed significantly from those with MetS, independent of the definition used. Cardiovascular risk stratification showed that the majority of individuals in the non‑MetS group were at a low‑to‑moderate risk (78.9%), whereas this proportion declined markedly in the MetS groups. Conversely, the prevalence of very high cardiovascular risk was increasing among the participants who met both MetS definitions (20%), as compared with only 5% in the non‑MetS group. Significant differences between the MetS definitions were noted in the low‑to‑moderate and high cardiovascular risk categories, whereas no significant difference was observed in the very high risk category. Notably, the group fulfilling both criteria, as well as the group defined by the 2022 recommendations, contained the largest proportions of individuals in the higher cardiovascular risk categories. No significant differences were observed in Lp(a) concentrations across the MetS or cardiovascular risk definitions, as presented in Table 1 and Supplementary material, Table S3. Lp(a) concentrations varied significantly across CKM stages, with post hoc comparisons indicating that stage 4 levels were significantly higher than those observed at stages 1–3. Interestingly, a significant difference between the analyzed populations across the MetS definitions was found among the individuals with normal BMI; however, this difference was not observed in the participants with BMI values above the normal range (Supplementary material, Table S4).

Comparison of previous and new guidelines of metabolic syndrome definition

To evaluate the effectiveness of various indicators in diagnosing MetS according to both definitions, we performed the ROC curve analysis. Supplementary material, Tables S5 and S6, present the ROC curves generated using the presence of MetS as the categorical outcome variable, defined according to the former definition. Among the evaluated parameters, waist circumference (AUC = 0.8778; 95% CI, 0.863–0.892), BMI (AUC = 0.8488; 95% CI, 0.832–0.865), fasting glucose (AUC = 0.842; 95% CI, 0.825–0.859), and TG/HDL‑C ratio (AUC = 0.8261; 95% CI, 0.807–0.845) exhibited the highest diagnostic accuracy, as indicated by the largest AUC (Figure 2). In comparison, LDL‑C and average body fat percentage showed markedly lower discriminative power when considered as independent diagnostic markers. Nonetheless, no single indicator emerged as a definitive marker with superior diagnostic accuracy over the others. Figure 3 and Supplementary material, Table S6 demonstrate ROC curves for MetS defined according to the updated 2022 consensus criteria as the categorical variable. BMI (AUC = 0.941; 95% CI, 0.931–0.951) and waist circumference (AUC = 0.912; 95% CI, 0.9–0.924) exhibited the greatest diagnostic utility in the detection of MetS of all the indicators evaluated. Both LDL‑C and average fat percentage showed reduced diagnostic relevance in this context.

Cardiovascular‑kidney‑metabolic syndrome: its prevalence and comparative assessment with metabolic syndrome

In Table 2, we present the clinical and biochemical characteristics of the study participants stratified by CKM syndrome, as defined by the AHA. These stages represent a continuum of cardiometabolic risk, ranging from metabolically healthy individuals (stage 0) to those with clinically manifested CVD (stage 4). A gradual increase in age was observed across the CKM stages (P <⁠0.001). The proportion of men was the lowest at the stage 0 (23.9%), peaked at the stage 2 (61.5%), and followed a moderate decline at the stages 3 (43.3%) and 4 (52.2%). There was an increase in BMI, WHR, and waist circumference across CKM stages (P <⁠0.001). A similar pattern was observed for arterial BP. Both SBP and DBP increased with advancing CKM stage, with DBP reaching its highest median value at the stage 2, and remaining elevated at the subsequent stages. Moreover, glycemic parameters also showed a consistent deterioration across CKM progression. Notably, 2‑hour postload insulin levels nearly doubled between the stages 0 and 3, indicating impaired glucose metabolism among the individuals with more advanced stages of CKM. Lipid abnormalities were most pronounced at the intermediate stages. TG concentrations increased markedly from the stage 1 (median [IQR], 81 [61–103] mg/dl) to stage 2 (147 [97–187] mg/dl) and remained elevated thereafter. Mean (IQR) LDL‑C level peaked at the stage 2 (136.2 [112.5–162.7] mg/dl) and declined slightly at the stage 4 (110.1 [79.2–136.3] mg/dl). Similarly, total cholesterol concentrations were the highest at the stage 2, and decreased at later stages, potentially reflecting the impact of lipid‑lowering therapies in the individuals with established CVD. The burden of subclinical atherosclerosis increased sharply across CKM stages. While only 11.6% of the participants at the stage 0 exhibited carotid atherosclerotic plaques, the prevalence rose to 81.6% at the stage 3 and 80.6% at the stage 4. ICA thickening over 50% was not observed at the stages 0 through 2, but was present in 2% of the individuals at the stage 3 and 5.4% at the stage 4. For low and high cardiovascular risk, the stage 0 was significantly different from all other CKM stages. At each successive stage, a significant increase was observed in the prevalence of individuals classified as the cardiovascular risk stages 2 and 3, alongside a noticeable decrease in the proportion of participants remaining at the stage 1. A significantly higher proportion of those with high and very high cardiovascular risk qualified for the stage 3 and 4, although there was an upward trend from the stage 2 onward. Figure 4 presents distribution of CKM stages across different age groups stratified by sex. Among men, we saw a clear age‑related increase in the prevalence of higher CKM stages. In the youngest group (20–34 y), the majority of men were at the stage 0 (29%) or stage 1 (33%), while only 10% were at the stage 3, and even fewer at the stage 4. In contrast, in the oldest male group (65–80 y), the majority were at the stages 2 (62%) and 3 (36%), with a notable proportion (22%) reaching the stage 4. For women, a similar age‑related pattern was observed, though with notable differences in the distribution. In the youngest women group, a substantial proportion were at the stage 0 (60%), with only small fractions at the stages 3 and 4. In the oldest women group, most individuals were at the stage 2 (66%), with decreasing proportions at the higher stages, as compared with their male counterparts. Figure 5 and Supplementary material, Table S7 compare CKM and MetS prevalence considering all definitions. Notably, CKM stages 3 and 4 were strongly associated with MetS diagnosed by either the 2009 or 2022 criteria. However, a substantial subset of high‑risk individuals (n = 193 at stage 3) did not fulfill the criteria for MetS, suggesting that CKM staging captures additional risk dimensions, particularly related to subclinical cardiovascular and renal dysfunction, not encompassed by traditional MetS definitions. To gain a more comprehensive understanding of the cardiometabolic profile of high‑risk individuals who do not meet the formal diagnostic thresholds for MetS, we examined the distribution of MetS components, according to both the 2009 and 2022 definitions, across CKM stages (Supplementary material, Tables S8 and S9). Among the patients classified as CKM stage 3 (n = 471), only 3.2% met none of the 2009 MetS criteria, while 13.2% met 1 criterion and 24.6% met 2 criteria. Consequently, 41% of the patients at the stage 3 did not fulfill the diagnostic threshold for MetS but exhibited 1 or 2 metabolic abnormalities. A similar pattern was observed using the 2022 definition, with 2.3%, 9%, and 23.7% of the patients meeting 0, 1, or 2 criteria, respectively. To further characterize this subgroup, we conducted a comparative analysis of the stage 3 patients with (n = 278) and without (n = 193) MetS, as defined by the 2009 criteria (Supplementary material, Table S10). Despite not meeting the formal diagnostic threshold, the individuals without MetS demonstrated clear metabolic abnormalities, including elevated mean (SD) fasting glucose (101.7 [15.1] mg/dl) and TG levels (94.9 [42.9] mg/dl), and BMI (26.9 [4.1] kg/m²). Differences between MetS‑positive and MetS‑negative individuals were observed for glucose metabolism parameters, lipid profile parameters, BP, and adiposity indices (P <⁠0.05). Notably, while these differences were both consistent and clinically meaningful, the mean values in the MetS‑negative group nonetheless deviated from standard reference ranges, suggesting the presence of subclinical or emerging cardiometabolic risk.

Conclusions

Our findings indicate a very high prevalence of MetS, regardless of the evaluation criteria used. The results show the distribution of the individuals meeting each stage of the CKM syndrome diagnostic criteria. The outcomes confirmed the high burden of ASCVD risk factors in the general population, including the participants who have not yet developed clinical CVD. The choice of MetS definition significantly impacts both prevalence estimates and cardiometabolic risk stratification. Notably, the updated criteria effectively identify patients at a higher cardiovascular risk, while CKM staging further captures high‑risk individuals who yet do not meet the MetS criteria, highlighting its broader applicability and added value as an integrative tool for comprehensive cardiometabolic risk assessment.