Association of hemoglobin glycation index with the severity and mortality of cardiovascular-kidney-metabolic syndrome: a population-based cohort study Authors: Leiluo Geng, Xue Diao, Hong-jie Yu, Aimin Xu Published: 2025 ABSTRACT BackgroundCardiovascular-kidney-metabolic (CKM) syndrome is a recently proposed multisystem disorder with high prevalence and progressive stages from 0 to 4, characterized by uncertain risk factors. Hemoglobin glycation index (HGI), a simple marker reflecting individual variation in glycated hemoglobin, has been linked to diabetic and cardiovascular outcomes. However, its association with CKM syndrome severity and mortality remains unknown.MethodsThis study utilized data from the National Health and Nutrition Examination Survey (1999–2018) and included 14,680 adult participants. Weighted logistic regression was conducted to assess the association of HGI and CKM syndrome stage. Kaplan-Meier survival curves and weighted Cox proportional hazards models were adopted to explore the relationship between HGI and mortality by CKM syndrome stages, followed with analyses of restricted cubic spline and threshold effect.ResultsA positive correlation between HGI and CKM syndrome stage was observed after multivariate adjustment (OR 1.24, 95% CI 1.10–1.39). Over a median follow-up period of 7.5 years, a total of 1,111 all-cause deaths and 349 cardiovascular deaths were documented. U-shaped associations between HGI and all-cause mortality were observed, with inflection points at -0.280 (CKM syndrome stage 0–3) and − 0.077 (CKM syndrome stage 4). A one-unit increase in HGI was linked to a 33–39% reduction in all-cause mortality risk below the inflection point, and a 33–42% increase above it. In contrast, HGI showed a linear association with cardiovascular mortality in participants with CKM syndrome stage 4 (HR 1.34, 95% CI 1.08–1.65), however, no significant association was observed in CKM syndrome stage 0–3.ConclusionsA higher HGI was associated with increased severity of CKM syndrome. There was a U-shaped relationship between HGI and all-cause mortality while a linear association between HGI and cardiovascular mortality was observed depending on the CKM syndrome stage. HGI may serve as a valuable biomarker for assessing the risk of CKM syndrome and its related mortality.Supplementary InformationThe online version contains supplementary material available at 10.1186/s40842-025-00262-4. Introduction Given the complex interplay of cardiac, renal, and metabolic conditions, cardiovascular-kidney-metabolic (CKM) syndrome was proposed by the American Heart Association in 2023 [1]. Based on the conceptual framework, CKM syndrome encompasses 5 consecutive stages from 0 to 4, reflecting different severities of CKM multimorbidity [1]. Among the US adults, the estimated prevalence of CKM syndrome stage 0–4 was 10.6, 25.9, 49.0, 5.4, and 9.2%, varied by social determinants of health and sex [2, 3]. In a combination analysis of three Chinese cohorts, the overall prevalence was 6.6, 10.4, 47.2, 18.5, and 17.2% for CKM syndrome stage 0–4 [4]. Notably, a substantial rise in the prevalence of CKM syndrome was observed during the COVID-19 pandemic, with nearly half of the patients displaying one or more components of the syndrome [5]. Thus, CKM syndrome has been a great public health burden in both developed and developing countries. Advanced CKM syndrome stages are associated with higher risks of adverse health outcomes, including cardiovascular diseases and premature mortality [6–8]. Several biological risk factors were identified in relation to CKM syndrome and its associated outcomes, including triglyceride-derived metabolic indices and inflammatory markers [9–12]. Nonetheless, further research is necessary to elucidate additional risk factors, thereby enhancing the accuracy of risk stratification and facilitating the timely implementation of interventions for CKM syndrome. Glycated hemoglobin (HbA1c) serves as an indicator of average blood glucose levels over the preceding 2–3 months; however, individual variations in the relationship between HbA1c and mean blood glucose levels are consistently observed. The hemoglobin glycation index (HGI) was introduced to quantify both the magnitude and direction of the discrepancy between an individual’s observed HbA1c and the level predicted based on their blood glucose concentration [13]. HGI is affected by various factors, including individual differences in hemoglobin structure, red blood cell turnover, and the presence of certain medical conditions [14]. Previous studies have demonstrated that a higher HGI is associated with greater risk for chronic vascular diseases in normal, prediabetic, and diabetic populations [15–17]. Meanwhile, HGI has also been correlated with all-cause and cardiovascular mortality in general population or population with metabolic syndrome [18–21]. However, the association between HGI and CKM syndrome severity, as well as the relationship between HGI and mortality across different CKM syndrome stages, has not yet been studied. Given the substantial burden of CKM syndrome and its significant impact on mortality across different disease stages, this study aimed to examine the association between HGI and CKM syndrome stage, as well as to evaluate whether HGI possesses prognostic value for all-cause and cardiovascular mortality at both early and advanced stages of CKM syndrome. Methods Study design and participants This study utilized an integrated approach combining cross-sectional and cohort study designs. The association between HGI and CKM syndrome stage was evaluated at baseline using a cross-sectional analysis. Subsequently, the relationship between HGI and mortality, stratified by CKM syndrome stage, was examined through a prospective cohort analysis. The participants for this study were drawn from the National Health and Nutrition Examination Survey (NHANES), which is an ongoing nationally representative cross-sectional survey and conducted by the National Center for Health Statistics (NCHS). Approximately 5,000 people from 15 counties are surveyed each year. The survey protocol has been approved by the institutional review board of the NCHS. All participants or their legal representatives provided written informed consents before data collection. We combined data from ten cycles of NHANES (from 1999 to 2000 to 2017–2018). Out of the 101,316 participants, 46,235 participants of age < 20 were excluded. Participants with missing 2-year/4-year fasting subsample weights were excluded, as were subjects with unavailable follow-up time or death data, CKM syndrome staging data, and fasting glucose and HbA1c data. In addition, subjects with missing covariates were excluded. In the end, the study included 14,680 qualified adult participants (Figure S1). Calculation of HGI HGI was calculated by subtracting the predicted HbA1c from the measured HbA1c (HGI = measured HbA1c - predicted HbA1c), and the predicted HbA1c was calculated by inserting the values of corresponding fasting plasma glucose (FPG) into the simple linear regression equation [predicted HbA1c (%) = 0.444 × FPG (mmol/L) + 3.015, R2 = 0.685, P < 0.001, Figure S2]. The levels of HbA1c and FPG were measured using high-performance liquid chromatography systems and hexokinase methods, respectively. The HGI was then categorized into quartiles as categorical variables and remained prototypes as continuous variables simultaneously. CKM syndrome staging To account for the diverse clinical severity of CKM syndrome, participants were categorized into five distinct stages based on the 2023 American Heart Association Presidential Advisory on CKM Health [1]. The definition and staging of CKM syndrome in this study were based on the framework proposed by Aggarwal et al. [2] which provides detailed information on its diagnostic criteria, stages, and associated risk factors. In brief, for each subject, we calculated the CKM syndrome stage: stage 0 (no CKM risk factors, such as hypertension), 1 (excess or dysfunctional adiposity), 2 (additional metabolic risk factors or moderate- or high-risk chronic kidney disease), 3 (very high-risk chronic kidney disease or high predicted 10-year cardiovascular disease risk), or 4 (established cardiovascular disease, such as coronary artery disease). A detailed description of the CKM syndrome definition and staging framework is provided in the Supplementary Methods. In the following association analyses, CKM syndrome stages were further grouped into stage 0–3 (without cardiovascular disease) and stage 4 (with established cardiovascular disease). Ascertainment of mortality We extracted potential leading causes of death from the mortality profile and used all-cause mortality and cardiovascular mortality as endpoints. Mortality information was ascertained from the National Death Index public access files to which NHANES is linked. Specific deaths were determined using The International Classification of Diseases, Tenth Revision (ICD-10). Deaths from all causes were included in all-cause mortality. Cardiovascular mortality was defined using codes of ICD-10 (I00-I09, I11, I13, I20-I51) and the NCHS classification of heart disease (054–068) [19]. The median follow-up time for the entire population was 7.50 years, with durations of 7.58 years for the CKM syndrome stage 0–3 group and 6.42 years for the stage 4 group. Covariate Race was categorized into five groups: Mexican American, other Hispanic, Non-Hispanic Black, Non-Hispanic White, and other race. Annual household income was divided into less than $20,000 and greater than or equal to $20,000. Education level was divided into under high school graduation, high school graduation or equivalent, and above high school graduation. Each participant completed a physical activity questionnaire, recording the type, duration, intensity, and frequency of activities performed over the past 30 days. Activity-specific metabolic equivalent (MET) scores were assigned according to activity type and intensity. These MET scores were multiplied by the average duration and frequency to obtain MET minutes over 30 days (MET min/30 days) for each activity. The total MET min/30 days was calculated by summing the individual activity scores and dividing by 4.29 to estimate weekly MET minutes. Participants were then classified into low or high physical activity groups based on national guidelines: low physical activity (< 600 MET min/week) and high physical activity (≥ 600 MET min/week) [22]. Participants who had smoked less than 100 cigarettes in their lifetime were labeled “never smokers”. Former smokers were those who had smoked more than 100 cigarettes in their lifetime but had quit, and current smokers were those who were currently smoking. Regarding alcohol consumption, participants who had consumed < 12 alcoholic drinks in the past 12 months were classified as “no” and participants who had consumed at least 12 drinks in the past 12 months were classified as “yes”. BMI was defined as the body weight divided by the square of the body height and expressed in units of kg/m2. Participants were classified as having hypertension if they self-reported a diagnosis of hypertension or high blood pressure, were currently taking prescribed antihypertensive medication, or exhibited elevated biological measurements, defined as a systolic blood pressure of ≥ 140 mmHg and/or a diastolic blood pressure of ≥ 90 mmHg. Participants were classified as having diabetes if they exhibited FPG levels of ≥ 7.0 mmol/L (126 mg/dL), a 2-hour plasma glucose level of ≥ 11.1 mmol/L (200 mg/dL) during the oral glucose tolerance test, HbA1c levels of ≥ 6.5%, self-reported a diagnosis of diabetes, or were currently receiving glucose-lowering treatment. Statistical analysis The baseline characteristics were presented according to the quartile of HGI or the CKM syndrome stage. Survey-weighted skewness and weighted histograms were used to assess the distribution of continuous variables [23]. Data are presented as mean ± standard error (SE) for symmetric continuous variables and compared by one-way analysis of variance. Data are presented as median (Q₁, Q₃) for skewed continuous variables and compared by Kruskal-waills tests. Categorical variable was expressed as count (weighted proportion) and compared by Rao-Scott chi-square test, a design-adjusted version of the Pearson chi-square test for analyzing data collected with complex sampling designs [24]. To explore the association between HGI and baseline CKM syndrome stage, a weighted logistic regression model was used with results expressed as odds ratios (ORs) with 95% confidence intervals (CIs). The mortality rates between different HGI quartile groups were presented as Kaplan-Meier curves and compared by log-rank test. Weighted Cox proportional hazards regression models were used to estimate the hazard ratios (HRs) and 95% CIs for the association of HGI and mortality. The confounding factors were gradually controlled by constructing three models (model 1, unadjusted; model 2, adjusted for sex, race, age, education, income, and marital status; model 3, adjusted for sex, race, age, education, income, marital status, smoking, drinking, body mass index, physical activity, hypertension, and diabetes). In addition, to study the potential nonlinear association between HGI and outcomes, restricted cubic splines (RCS) were introduced in each model. Furthermore, a threshold effect analysis was performed on exposure and outcome, if there was a nonlinear association. To better understand the relationship between exposure and outcomes, we performed crude subgroup analyses by stratifying the population based on factors such as age, gender, race, education level and other relevant variables without adjustment for potential confounders. Finally, sensitivity analysis was performed to check the robustness of the findings. Data processing and analysis were performed using R (version 4.3.2), along with Zstats v1.0 (www.zstats.net). R packages used in this study were listed in Supplementary Methods. P < 0.05 was considered statistically significant. Results Baseline characteristics The baseline characteristics of participants stratified by four HGI quartiles were shown in Table 1. A total of 14,680 US adults were included in this study, with a weighted mean age of 47.85 years and a weighted mean BMI of 29.06 kg/m2. Approximately half of the population were male, married, and never smokers (50.17–56.60%). The majority of participants identified as non-Hispanic white, possessed at least a high school education, reported an annual household income of $20,000 or higher, and engaged in drinking and high physical activity (64.13–86.33%). Participants with higher HGI values were more likely to be older, obese, female, hypertensive, and diabetic, and they also demonstrated lower annual household income, educational attainment, and levels of physical activity compared to those in the lowest quartile (P < 0.001). There were no statistically significant differences in marital and smoking status among different HGI quartiles (P > 0.05). Table 1Baseline characteristics of participants classified by HGI quartilesVariableTotal(n = 14680)Q1(HGI<-0.303)Q2(-0.303 ≤ HGI<-0.047)Q3(-0.047 ≤ HGI < 0.199)Q4(HGI ≥ 0.199)P-valueAge, years47.85 ± 0.2543.47 ± 0.3845.32 ± 0.4348.44 ± 0.3854.16 ± 0.33< 0.001Sex< 0.001 Male7399 (50.17)1933 (60.60)1708 (50.47)1686 (46.37)2072 (43.24) Female7281 (49.83)1292 (39.40)1666 (49.53)1855 (53.63)2468 (56.76)Race< 0.001 Mexican American2273 (7.78)488 (7.62)584 (8.65)511 (6.95)690 (7.91) Other Hispanic1288 (5.04)268 (4.47)304 (5.44)324 (5.18)392 (5.08) Non-Hispanic White6884 (69.76)1759 (75.11)1750 (72.85)1717 (70.92)1658 (60.20) Non-Hispanic Black2774 (10.48)425 (6.48)399 (6.27)594 (9.61)1356 (19.56) Other race1461 (6.93)285 (6.33)337 (6.79)395 (7.34)444 (7.25)Education< 0.001 Under high school3352 (14.98)663 (12.84)715 (13.30)773 (14.66)1201 (19.12) High school3378 (23.49)718 (22.61)739 (21.87)779 (22.61)1142 (26.87) Above high school7950 (61.53)1844 (64.55)1920 (64.83)1989 (62.73)2197 (54.01)Marital status0.571 Unmarried6781 (43.40)1512 (44.64)1560 (43.31)1627 (42.92)2082 (42.74) Married7899 (56.60)1713 (55.36)1814 (56.69)1914 (57.08)2458 (57.26)Annual household income< 0.001 <20,000$2990 (13.67)585 (11.59)659 (12.88)698 (13.00)1048 (17.20) ≥20,000$11,690 (86.33)2640 (88.41)2715 (87.12)2843 (87.00)3492 (82.80)Smoking0.115 Never7956 (54.24)1717 (54.39)1892 (56.61)1916 (53.56)2431 (52.38) Former3890 (26.32)894 (27.23)832 (24.71)936 (26.63)1228 (26.70) Current2834 (19.45)614 (18.37)650 (18.68)689 (19.82)881 (20.91)Drinking< 0.001 No3755 (20.78)621 (15.49)728 (16.92)934 (21.52)1472 (29.17) Yes10,925 (79.22)2604 (84.51)2646 (83.08)2607 (78.48)3068 (70.83)Hypertension< 0.001 No8103 (59.96)1907 (63.22)2106 (66.17)2054 (61.98)2036 (48.50) Yes6577 (40.04)1318 (36.78)1268 (33.83)1487 (38.02)2504 (51.50)Diabetes< 0.001 No11,537 (83.55)2567 (84.93)2998 (91.60)3066 (89.70)2906 (67.97) Yes3143 (16.45)658 (15.07)376 (8.40)475 (10.30)1634 (32.03)Physical activity< 0.001 Low5833 (35.87)1150 (32.15)1251 (34.08)1391 (35.54)2041 (41.72) High8847 (64.13)2075 (67.85)2123 (65.92)2150 (64.46)2499 (58.28)BMI, kg/m228.00 (24.10, 32.58)27.80 (24.15, 32.00)27.80 (23.95, 32.20)27.44 (23.70, 32.07)29.10 (24.91, 34.10)< 0.001BMI< 0.001 <254226 (29.84)982 (29.59)1042 (31.32)1121 (33.25)1081 (25.22) 25–304867 (32.63)1127 (35.33)1146 (32.97)1158 (31.87)1436 (30.37) ≥305587 (37.52)1116 (35.07)1186 (35.71)1262 (34.89)2023 (44.40)CKM syndrome stage< 0.001 01232 (9.99)271 (9.60)350 (12.01)409 (12.85)202 (5.51) 12251 (18.07)604 (21.96)602 (20.77)503 (16.78)542 (12.76) 28754 (57.40)1891 (56.36)1975 (55.10)2083 (57.53)2805 (60.60) 3836 (5.58)176 (5.14)165 (5.12)177 (4.62)318 (7.44) 41607 (8.96)283 (6.94)282 (7.00)369 (8.22)673 (13.70)HbA1c, %5.40 (5.20, 5.80)5.10 (4.90, 5.30)5.30 (5.20, 5.50)5.50 (5.30, 5.70)5.90 (5.70, 6.40)< 0.001FPG, mmol/L5.55 (5.16, 6.05)5.72 (5.38, 6.27)5.55 (5.22, 5.94)5.44 (5.11, 5.83)5.50 (5.05, 6.33)< 0.001Data are presented as mean ± standard error for symmetric continuous variables or median (Q₁, Q₃) for skewed continuous variables, and as n (%) for categorical variables. BMI, body mass index; CKM, cardiovascular-kidney-metabolic; FPG, fasting plasma glucose; HbA1c, hemoglobin A1c; HGI, hemoglobin glycation index Table 1Baseline characteristics of participants classified by HGI quartilesVariableTotal(n = 14680)Q1(HGI<-0.303)Q2(-0.303 ≤ HGI<-0.047)Q3(-0.047 ≤ HGI < 0.199)Q4(HGI ≥ 0.199)P-valueAge, years47.85 ± 0.2543.47 ± 0.3845.32 ± 0.4348.44 ± 0.3854.16 ± 0.33< 0.001Sex< 0.001 Male7399 (50.17)1933 (60.60)1708 (50.47)1686 (46.37)2072 (43.24) Female7281 (49.83)1292 (39.40)1666 (49.53)1855 (53.63)2468 (56.76)Race< 0.001 Mexican American2273 (7.78)488 (7.62)584 (8.65)511 (6.95)690 (7.91) Other Hispanic1288 (5.04)268 (4.47)304 (5.44)324 (5.18)392 (5.08) Non-Hispanic White6884 (69.76)1759 (75.11)1750 (72.85)1717 (70.92)1658 (60.20) Non-Hispanic Black2774 (10.48)425 (6.48)399 (6.27)594 (9.61)1356 (19.56) Other race1461 (6.93)285 (6.33)337 (6.79)395 (7.34)444 (7.25)Education< 0.001 Under high school3352 (14.98)663 (12.84)715 (13.30)773 (14.66)1201 (19.12) High school3378 (23.49)718 (22.61)739 (21.87)779 (22.61)1142 (26.87) Above high school7950 (61.53)1844 (64.55)1920 (64.83)1989 (62.73)2197 (54.01)Marital status0.571 Unmarried6781 (43.40)1512 (44.64)1560 (43.31)1627 (42.92)2082 (42.74) Married7899 (56.60)1713 (55.36)1814 (56.69)1914 (57.08)2458 (57.26)Annual household income< 0.001 <20,000$2990 (13.67)585 (11.59)659 (12.88)698 (13.00)1048 (17.20) ≥20,000$11,690 (86.33)2640 (88.41)2715 (87.12)2843 (87.00)3492 (82.80)Smoking0.115 Never7956 (54.24)1717 (54.39)1892 (56.61)1916 (53.56)2431 (52.38) Former3890 (26.32)894 (27.23)832 (24.71)936 (26.63)1228 (26.70) Current2834 (19.45)614 (18.37)650 (18.68)689 (19.82)881 (20.91)Drinking< 0.001 No3755 (20.78)621 (15.49)728 (16.92)934 (21.52)1472 (29.17) Yes10,925 (79.22)2604 (84.51)2646 (83.08)2607 (78.48)3068 (70.83)Hypertension< 0.001 No8103 (59.96)1907 (63.22)2106 (66.17)2054 (61.98)2036 (48.50) Yes6577 (40.04)1318 (36.78)1268 (33.83)1487 (38.02)2504 (51.50)Diabetes< 0.001 No11,537 (83.55)2567 (84.93)2998 (91.60)3066 (89.70)2906 (67.97) Yes3143 (16.45)658 (15.07)376 (8.40)475 (10.30)1634 (32.03)Physical activity< 0.001 Low5833 (35.87)1150 (32.15)1251 (34.08)1391 (35.54)2041 (41.72) High8847 (64.13)2075 (67.85)2123 (65.92)2150 (64.46)2499 (58.28)BMI, kg/m228.00 (24.10, 32.58)27.80 (24.15, 32.00)27.80 (23.95, 32.20)27.44 (23.70, 32.07)29.10 (24.91, 34.10)< 0.001BMI< 0.001 <254226 (29.84)982 (29.59)1042 (31.32)1121 (33.25)1081 (25.22) 25–304867 (32.63)1127 (35.33)1146 (32.97)1158 (31.87)1436 (30.37) ≥305587 (37.52)1116 (35.07)1186 (35.71)1262 (34.89)2023 (44.40)CKM syndrome stage< 0.001 01232 (9.99)271 (9.60)350 (12.01)409 (12.85)202 (5.51) 12251 (18.07)604 (21.96)602 (20.77)503 (16.78)542 (12.76) 28754 (57.40)1891 (56.36)1975 (55.10)2083 (57.53)2805 (60.60) 3836 (5.58)176 (5.14)165 (5.12)177 (4.62)318 (7.44) 41607 (8.96)283 (6.94)282 (7.00)369 (8.22)673 (13.70)HbA1c, %5.40 (5.20, 5.80)5.10 (4.90, 5.30)5.30 (5.20, 5.50)5.50 (5.30, 5.70)5.90 (5.70, 6.40)< 0.001FPG, mmol/L5.55 (5.16, 6.05)5.72 (5.38, 6.27)5.55 (5.22, 5.94)5.44 (5.11, 5.83)5.50 (5.05, 6.33)< 0.001Data are presented as mean ± standard error for symmetric continuous variables or median (Q₁, Q₃) for skewed continuous variables, and as n (%) for categorical variables. BMI, body mass index; CKM, cardiovascular-kidney-metabolic; FPG, fasting plasma glucose; HbA1c, hemoglobin A1c; HGI, hemoglobin glycation index Baseline characteristics of participants classified by HGI quartiles Data are presented as mean ± standard error for symmetric continuous variables or median (Q₁, Q₃) for skewed continuous variables, and as n (%) for categorical variables. BMI, body mass index; CKM, cardiovascular-kidney-metabolic; FPG, fasting plasma glucose; HbA1c, hemoglobin A1c; HGI, hemoglobin glycation index The weighted median values of HbA1c and FPG were 5.40% and 5.55 mmol/L for the whole population. The frequency distribution of FPG was similar among the four HGI subgroups (Fig. 1A). In contrast, the distribution of HbA1c varied obviously among the subgroups, exhibiting a trend toward increased HGI and elevated HbA1c levels (Fig. 1B). Consistent with previous reports [2, 3], the weighted proportion of participants at CKM syndrome stage 2 was the highest, accounting for 57.40%, while stage 3 was the lowest at 5.58%. The distribution of CKM syndrome stages was comparable across the four HGI subgroups (Fig. 1C). The baseline characteristics grouped by CKM syndrome stage were also provided in Table S1. The weighted mean value of HGI was − 0.02 for the whole population. Notably, with the progression of CKM syndrome stages, levels of FPG, HbA1c, and HGI exhibited increasing trends (Fig. 1D), indicating a potential amplification of glucose dysregulation as the syndrome advances. Fig. 1Assessment of FPG, HbA1c, HGI, and CKM syndrome stage at baseline. (A-B) Distribution of fasting plasma glucose (FPG, A) and HbA1c (B) by hemoglobin glycation index (HGI) quartile subgroups. (C) Weighted proportions of cardiovascular-kidney-metabolic (CKM) syndrome stages in total population and HGI quartile subgroups. (D) Levels of FPG, HbA1c, and HGI in different CKM syndrome stages Assessment of FPG, HbA1c, HGI, and CKM syndrome stage at baseline. (A-B) Distribution of fasting plasma glucose (FPG, A) and HbA1c (B) by hemoglobin glycation index (HGI) quartile subgroups. (C) Weighted proportions of cardiovascular-kidney-metabolic (CKM) syndrome stages in total population and HGI quartile subgroups. (D) Levels of FPG, HbA1c, and HGI in different CKM syndrome stages Association between HGI and CKM syndrome severity In weighted logistic regression analyses, with CKM syndrome stage dichotomized into stages 0–3 versus 4 (Table 2), each unit increase in HGI was associated with a 24% higher odds of more severe CKM syndrome after full adjustment for demographic, lifestyle and comorbidity factors (OR 1.24, 95% CI 1.10–1.39, P < 0.001). Meanwhile, HGI quartile analyses demonstrated significant dose-dependent associations in both unadjusted and adjusted models (P-trend < 0.001). In addition, HGI showed a significant association with more severe CKM syndrome in restricted cubic spline analysis (P-overall < 0.001), without evidence of non-linearity (P-nonlinearity = 0.079) after full adjustments (Fig. 2). Table 2Regression analysis on the association between HGI and CKM syndrome stage (0–3 vs. 4)Model 1Model 2Model 3OR (95% CI)P-valueOR (95% CI)P-valueOR (95% CI)P-valueHGI continuous1.69 (1.49–1.92)< 0.0011.66 (1.44–1.91)< 0.0011.24 (1.10–1.39)< 0.001HGI-Q2Reference-Reference-Reference-HGI-Q10.99 (0.81–1.22)0.9280.98 (0.81–1.20)0.8810.84 (0.68–1.03)0.092HGI-Q31.19 (0.97–1.46)0.0951.16 (0.95–1.43)0.1461.06 (0.86–1.32)0.578HGI-Q42.11 (1.73–2.57)< 0.0011.99 (1.62–2.44)< 0.0011.23 (0.99–1.52)0.067P for trend< 0.001< 0.001< 0.001Model 1: no adjustment; Model 2: adjusted for sex, race, age, income, marital status, and education; Model 3: adjusted for sex, race, age, body mass index, income, marital status, education, hypertension, smoking, drinking, diabetes, and physical activity. OR, odds ratio; CI, confidence interval; HGI, hemoglobin glycation index; CKM, cardiovascular-kidney-metabolic Table 2Regression analysis on the association between HGI and CKM syndrome stage (0–3 vs. 4)Model 1Model 2Model 3OR (95% CI)P-valueOR (95% CI)P-valueOR (95% CI)P-valueHGI continuous1.69 (1.49–1.92)< 0.0011.66 (1.44–1.91)< 0.0011.24 (1.10–1.39)< 0.001HGI-Q2Reference-Reference-Reference-HGI-Q10.99 (0.81–1.22)0.9280.98 (0.81–1.20)0.8810.84 (0.68–1.03)0.092HGI-Q31.19 (0.97–1.46)0.0951.16 (0.95–1.43)0.1461.06 (0.86–1.32)0.578HGI-Q42.11 (1.73–2.57)< 0.0011.99 (1.62–2.44)< 0.0011.23 (0.99–1.52)0.067P for trend< 0.001< 0.001< 0.001Model 1: no adjustment; Model 2: adjusted for sex, race, age, income, marital status, and education; Model 3: adjusted for sex, race, age, body mass index, income, marital status, education, hypertension, smoking, drinking, diabetes, and physical activity. OR, odds ratio; CI, confidence interval; HGI, hemoglobin glycation index; CKM, cardiovascular-kidney-metabolic Regression analysis on the association between HGI and CKM syndrome stage (0–3 vs. 4) Model 1: no adjustment; Model 2: adjusted for sex, race, age, income, marital status, and education; Model 3: adjusted for sex, race, age, body mass index, income, marital status, education, hypertension, smoking, drinking, diabetes, and physical activity. OR, odds ratio; CI, confidence interval; HGI, hemoglobin glycation index; CKM, cardiovascular-kidney-metabolic Fig. 2Restricted cubic spline plot for the association of HGI with CKM syndrome stage after full adjustments. The model was adjusted for sex, race, age, body mass index, income, marital status, education, smoking, drinking, physical activity, hypertension, and diabetes. OR, odds ratio; CI, confidence interval; CKM, cardiovascular-kidney-metabolic; HGI, hemoglobin glycation index Restricted cubic spline plot for the association of HGI with CKM syndrome stage after full adjustments. The model was adjusted for sex, race, age, body mass index, income, marital status, education, smoking, drinking, physical activity, hypertension, and diabetes. OR, odds ratio; CI, confidence interval; CKM, cardiovascular-kidney-metabolic; HGI, hemoglobin glycation index Weighted ordinal logistic regression analysis was also conducted with CKM syndrome stage treated as an ordinal variable with categories 0, 1, 2, 3, and 4 (Table S2). Similar results were observed. Each unit increase in HGI was associated with a 22% higher odds of more severe CKM syndrome after comprehensive adjustments (OR 1.22, 95% CI 1.12–1.32, P < 0.001). Collectively, higher HGI is significantly associated with more severe CKM syndrome. Association of HGI with all-cause and cardiovascular mortality by CKM syndrome stage Among the total participants, 1,111 and 349 individuals experienced all-cause and cardiovascular mortality during a median follow-up of 7.5 years, respectively. Kaplan-Meier survival analyses revealed that all-cause mortality and cardiovascular mortality were significantly different across HGI quartiles (all log rank P < 0.001, Fig. 3). Fig. 3Kaplan-Meier analyses of all-cause (A, B, C) or cardiovascular (D, E, F) mortality in total participants (A, D), participants with CKM syndrome stage 0–3 (B, E), and participants with CKM syndrome stage 4 (C, F) based on HGI quartiles. CKM, cardiovascular-kidney-metabolic; HGI, hemoglobin glycation index Kaplan-Meier analyses of all-cause (A, B, C) or cardiovascular (D, E, F) mortality in total participants (A, D), participants with CKM syndrome stage 0–3 (B, E), and participants with CKM syndrome stage 4 (C, F) based on HGI quartiles. CKM, cardiovascular-kidney-metabolic; HGI, hemoglobin glycation index Table 3 demonstrates the results of the Cox regression models between the HGI and mortality. HGI, as a continuous variable, was not significantly associated with all-cause mortality after adjustment for all confounding factors. However, quartile analyses showed that HGI-Q1 and Q4 groups demonstrated significantly higher risks of all-cause mortality than the reference Q2 group (Q1 vs. Q2: HR 1.33, 95% CI 1.10–1.60, P = 0.003; Q4 vs. Q2: HR 1.29, 95% CI 1.10–1.51, P = 0.002). The relationships were more pronounced in participants with CKM syndrome stage 0–3 (Q1 vs. Q2: HR 1.57, 95% CI 1.26–1.94, P < 0.001; Q4 vs. Q2: HR 1.39, 95% CI 1.14–1.70, P = 0.001). With regard to cardiovascular mortality, in the overall population each unit increase in HGI level was associated with a 25% higher risk (HR 1.25, 95% CI 1.06–1.47, P = 0.007). A similar result was observed in participants with CKM syndrome stage 4 (HR 1.34, 95% CI 1.08–1.65, P = 0.007). Table 3Hazard ratios for all-cause and cardiovascular mortality according to the HGIModel 1Model 2Model 3HR (95% CI)P-valueHR (95% CI)P-valueHR (95% CI)P-value All-cause Mortality All participants HGI continuous1.36 (1.22–1.51)< 0.0011.10 (0.98–1.24)0.0911.02 (0.92–1.12)0.759 HGI-Q2Reference-Reference-Reference- HGI-Q11.34 (1.10–1.63)0.0031.45 (1.22–1.73)< 0.0011.33 (1.10–1.60)0.003 HGI-Q31.20 (1.02–1.41)0.0301.06 (0.89–1.25)0.5201.04 (0.88–1.23)0.663 HGI-Q42.21 (1.89–2.60)< 0.0011.50 (1.28–1.76)< 0.0011.29 (1.10–1.51)0.002 P for trend< 0.0010.1960.813 CKM syndrome stage 0–3 HGI continuous1.24 (1.08–1.43)0.0021.02 (0.87–1.19)0.8530.96 (0.85–1.09)0.528 HGI-Q2Reference-Reference-Reference- HGI-Q11.57 (1.27–1.94)< 0.0011.72 (1.41–2.10)< 0.0011.57 (1.26–1.94)< 0.001 HGI-Q31.36 (1.09–1.70)0.0071.24 (0.98–1.56)0.0701.22 (0.97–1.54)0.089 HGI-Q42.25 (1.85–2.73)< 0.0011.59 (1.29–1.96)< 0.0011.39 (1.14–1.70)0.001 P for trend< 0.0010.8900.661 CKM syndrome stage 4 HGI continuous1.17 (1.02–1.35)0.0291.09 (0.94–1.26)0.2431.03 (0.88–1.21)0.680 HGI-Q2Reference-Reference-Reference- HGI-Q11.08 (0.74–1.56)0.6900.98 (0.70–1.37)0.9000.95 (0.68–1.33)0.768 HGI-Q30.75 (0.57–0.99)0.0450.63 (0.48–0.84)0.0020.59 (0.44–0.79)< 0.001 HGI-Q41.45 (109 − 1.93)0.0101.13 (0.86–1.48)0.3901.01 (0.75–1.37)0.933 P for trend0.0170.3080.792 Cardiovascular Mortality All participants HGI continuous1.59 (1.38–1.84)< 0.0011.40 (1.20–1.64)< 0.0011.25 (1.06–1.47)0.007 HGI-Q2Reference-Reference-Reference- HGI-Q11.13 (0.83–1.53)0.4391.21 (0.91–1.62)0.1901.05 (0.78–1.41)0.760 HGI-Q31.09 (0.82–1.44)0.5560.95 (0.71–1.28)0.7460.92 (0.68–1.25)0.604 HGI-Q42.38 (1.73–3.26)< 0.0011.59 (1.14–2.22)0.0061.28 (0.92–1.78)0.136 P for trend< 0.0010.0420.174 CKM syndrome stage 0–3 HGI continuous1.43 (1.15–1.77)0.0011.23 (0.90–1.68)0.1981.11 (0.85–1.45)0.436 HGI-Q2Reference-Reference-Reference- HGI-Q11.51 (1.08–2.12)0.0161.68 (1.19–2.36)0.0031.43 (1.01–2.03)0.042 HGI-Q31.17 (0.80–1.72)0.4221.07 (0.71–1.61)0.7351.05 (0.70–1.59)0.802 HGI-Q42.29 (1.56–3.36)< 0.0011.60 (1.06–2.39)0.0241.32 (0.88–1.98)0.179 P for trend0.0240.8910.875 CKM syndrome stage 4 HGI continuous1.45 (1.20–1.75)< 0.0011.40 (1.16–1.70)< 0.0011.34 (1.08–1.65)0.007 HGI-Q2Reference-Reference-Reference- HGI-Q10.75 (0.43–1.32)0.3160.66 (0.39–1.10)0.1080.60 (0.36–0.99)0.048 HGI-Q30.83 (0.55–1.24)0.3540.67 (0.44–1.00)0.0510.58 (0.38–0.90)0.014 HGI-Q41.67 (1.09–2.57)0.0201.30 (0.85–1.98)0.2341.03 (0.65–1.65)0.886 P for trend< 0.0010.0040.063Model 1: no adjustment; Model 2: adjusted for sex, race, age, education, income, and marital status; Model 3: adjusted for sex, race, age, education, income, marital status, smoking, drinking, body mass index, physical activity, hypertension, and diabetes. HR, hazard ratio; CI, confidence interval; HGI, hemoglobin glycation index; CKM, cardiovascular-kidney-metabolic Table 3Hazard ratios for all-cause and cardiovascular mortality according to the HGIModel 1Model 2Model 3HR (95% CI)P-valueHR (95% CI)P-valueHR (95% CI)P-value All-cause Mortality All participants HGI continuous1.36 (1.22–1.51)< 0.0011.10 (0.98–1.24)0.0911.02 (0.92–1.12)0.759 HGI-Q2Reference-Reference-Reference- HGI-Q11.34 (1.10–1.63)0.0031.45 (1.22–1.73)< 0.0011.33 (1.10–1.60)0.003 HGI-Q31.20 (1.02–1.41)0.0301.06 (0.89–1.25)0.5201.04 (0.88–1.23)0.663 HGI-Q42.21 (1.89–2.60)< 0.0011.50 (1.28–1.76)< 0.0011.29 (1.10–1.51)0.002 P for trend< 0.0010.1960.813 CKM syndrome stage 0–3 HGI continuous1.24 (1.08–1.43)0.0021.02 (0.87–1.19)0.8530.96 (0.85–1.09)0.528 HGI-Q2Reference-Reference-Reference- HGI-Q11.57 (1.27–1.94)< 0.0011.72 (1.41–2.10)< 0.0011.57 (1.26–1.94)< 0.001 HGI-Q31.36 (1.09–1.70)0.0071.24 (0.98–1.56)0.0701.22 (0.97–1.54)0.089 HGI-Q42.25 (1.85–2.73)< 0.0011.59 (1.29–1.96)< 0.0011.39 (1.14–1.70)0.001 P for trend< 0.0010.8900.661 CKM syndrome stage 4 HGI continuous1.17 (1.02–1.35)0.0291.09 (0.94–1.26)0.2431.03 (0.88–1.21)0.680 HGI-Q2Reference-Reference-Reference- HGI-Q11.08 (0.74–1.56)0.6900.98 (0.70–1.37)0.9000.95 (0.68–1.33)0.768 HGI-Q30.75 (0.57–0.99)0.0450.63 (0.48–0.84)0.0020.59 (0.44–0.79)< 0.001 HGI-Q41.45 (109 − 1.93)0.0101.13 (0.86–1.48)0.3901.01 (0.75–1.37)0.933 P for trend0.0170.3080.792 Cardiovascular Mortality All participants HGI continuous1.59 (1.38–1.84)< 0.0011.40 (1.20–1.64)< 0.0011.25 (1.06–1.47)0.007 HGI-Q2Reference-Reference-Reference- HGI-Q11.13 (0.83–1.53)0.4391.21 (0.91–1.62)0.1901.05 (0.78–1.41)0.760 HGI-Q31.09 (0.82–1.44)0.5560.95 (0.71–1.28)0.7460.92 (0.68–1.25)0.604 HGI-Q42.38 (1.73–3.26)< 0.0011.59 (1.14–2.22)0.0061.28 (0.92–1.78)0.136 P for trend< 0.0010.0420.174 CKM syndrome stage 0–3 HGI continuous1.43 (1.15–1.77)0.0011.23 (0.90–1.68)0.1981.11 (0.85–1.45)0.436 HGI-Q2Reference-Reference-Reference- HGI-Q11.51 (1.08–2.12)0.0161.68 (1.19–2.36)0.0031.43 (1.01–2.03)0.042 HGI-Q31.17 (0.80–1.72)0.4221.07 (0.71–1.61)0.7351.05 (0.70–1.59)0.802 HGI-Q42.29 (1.56–3.36)< 0.0011.60 (1.06–2.39)0.0241.32 (0.88–1.98)0.179 P for trend0.0240.8910.875 CKM syndrome stage 4 HGI continuous1.45 (1.20–1.75)< 0.0011.40 (1.16–1.70)< 0.0011.34 (1.08–1.65)0.007 HGI-Q2Reference-Reference-Reference- HGI-Q10.75 (0.43–1.32)0.3160.66 (0.39–1.10)0.1080.60 (0.36–0.99)0.048 HGI-Q30.83 (0.55–1.24)0.3540.67 (0.44–1.00)0.0510.58 (0.38–0.90)0.014 HGI-Q41.67 (1.09–2.57)0.0201.30 (0.85–1.98)0.2341.03 (0.65–1.65)0.886 P for trend< 0.0010.0040.063Model 1: no adjustment; Model 2: adjusted for sex, race, age, education, income, and marital status; Model 3: adjusted for sex, race, age, education, income, marital status, smoking, drinking, body mass index, physical activity, hypertension, and diabetes. HR, hazard ratio; CI, confidence interval; HGI, hemoglobin glycation index; CKM, cardiovascular-kidney-metabolic Hazard ratios for all-cause and cardiovascular mortality according to the HGI Model 1: no adjustment; Model 2: adjusted for sex, race, age, education, income, and marital status; Model 3: adjusted for sex, race, age, education, income, marital status, smoking, drinking, body mass index, physical activity, hypertension, and diabetes. HR, hazard ratio; CI, confidence interval; HGI, hemoglobin glycation index; CKM, cardiovascular-kidney-metabolic The relationships between HGI and mortality were further assessed by RCS curves. HGI was positively associated with an increased adjusted risk of all-cause mortality in the whole participants (P for overall < 0.001; P for nonlinearity < 0.001. Figure 4A). Similar U-shaped associations were also observed in the subgroups with CKM syndrome stage 0–3 or stage 4 (P for overall < 0.001; P for nonlinearity < 0.001. Figure 4B and C). In addition, HGI was positively correlated with an increased adjusted risk of cardiovascular mortality in the whole population (P for overall = 0.002; P for nonlinearity = 0.101; Fig. 4D) and population with CKM syndrome stage 4 (P for overall = 0.015; P for nonlinearity = 0.215; Fig. 4F), but not in population with CKM syndrome stage 0–3 (P for overall = 0.157; P for nonlinearity = 0.147; Fig. 4E). These results suggest there is a linear relationship between the HGI and cardiovascular death, varying in cohorts with different CKM syndrome stages. Fig. 4Restricted cubic spline plots for the associations of HGI with all-cause (A, B, C) or cardiovascular (D, E, F) mortality in total participants (A, D), participants with CKM syndrome stage 0–3 (B, E), and participants with CKM syndrome stage 4 (C, F). All the models were adjusted for sex, race, age, body mass index, income, marital status, education, smoking, drinking, physical activity, hypertension, and diabetes. HGI, hemoglobin glycation index; CKM, cardiovascular-kidney-metabolic; HR, hazard ratio; CI, confidence interval Restricted cubic spline plots for the associations of HGI with all-cause (A, B, C) or cardiovascular (D, E, F) mortality in total participants (A, D), participants with CKM syndrome stage 0–3 (B, E), and participants with CKM syndrome stage 4 (C, F). All the models were adjusted for sex, race, age, body mass index, income, marital status, education, smoking, drinking, physical activity, hypertension, and diabetes. HGI, hemoglobin glycation index; CKM, cardiovascular-kidney-metabolic; HR, hazard ratio; CI, confidence interval Two-piecewise regression analysis was further conducted to investigate the threshold effect of HGI on all-cause mortality (Table 4). When the HGI was below the inflection point − 0.307 in the whole population, the risk of all-cause mortality decreased by 39% for every one unit increase in HGI level (HR 0.61, 95% CI 0.55–0.68, P < 0.001). When the HGI was above − 0.307, the risk of all-cause mortality increased by 48% for every one unit increase in HGI level (HR 1.48, 95% CI 1.39–1.57, P < 0.001). The threshold model significantly improved the fit compared with the linear model (P for likelihood ratio test < 0.001). Similar findings were observed in subgroups with CKM syndrome stage 0–3 and stage 4, each with different inflection points at -0.280 and − 0.077, respectively. Table 4Threshold effect analyses of HGI on all-cause mortalityHR (95% CI)P-value All participants Model 1 Fitting model by standard linear regression1.30 (1.22–1.39)< 0.001Model 2 Fitting model by two-piecewise linear regressionInflection point-0.307 <-0.3070.61 (0.55–0.68)< 0.001 ≥-0.3071.48 (1.39–1.57)< 0.001P for likelihood ratio test< 0.001 CKM syndrome stage 0–3 Model 1 Fitting model by standard linear regression1.20 (1.10–1.32)< 0.001Model 2 Fitting model by two-piecewise linear regressionInflection point-0.280 <-0.2800.61 (0.54–0.69)< 0.001 ≥-0.2801.42 (1.30–1.54)< 0.001P for likelihood ratio test< 0.001 CKM syndrome stage 4 Model 1 Fitting model by standard linear regression1.16 (1.04–1.30)0.006Model 2 Fitting model by two-piecewise linear regressionInflection point-0.077 <-0.0770.67 (0.50–0.89)0.006 ≥-0.0771.33 (1.18–1.49)< 0.001P for likelihood ratio test< 0.001HR, hazard ratio; CI, confidence interval; CKM, cardiovascular-kidney-metabolic; HGI, hemoglobin glycation index Table 4Threshold effect analyses of HGI on all-cause mortalityHR (95% CI)P-value All participants Model 1 Fitting model by standard linear regression1.30 (1.22–1.39)< 0.001Model 2 Fitting model by two-piecewise linear regressionInflection point-0.307 <-0.3070.61 (0.55–0.68)< 0.001 ≥-0.3071.48 (1.39–1.57)< 0.001P for likelihood ratio test< 0.001 CKM syndrome stage 0–3 Model 1 Fitting model by standard linear regression1.20 (1.10–1.32)< 0.001Model 2 Fitting model by two-piecewise linear regressionInflection point-0.280 <-0.2800.61 (0.54–0.69)< 0.001 ≥-0.2801.42 (1.30–1.54)< 0.001P for likelihood ratio test< 0.001 CKM syndrome stage 4 Model 1 Fitting model by standard linear regression1.16 (1.04–1.30)0.006Model 2 Fitting model by two-piecewise linear regressionInflection point-0.077 <-0.0770.67 (0.50–0.89)0.006 ≥-0.0771.33 (1.18–1.49)< 0.001P for likelihood ratio test< 0.001HR, hazard ratio; CI, confidence interval; CKM, cardiovascular-kidney-metabolic; HGI, hemoglobin glycation index Threshold effect analyses of HGI on all-cause mortality HR, hazard ratio; CI, confidence interval; CKM, cardiovascular-kidney-metabolic; HGI, hemoglobin glycation index Subgroup analyses In subgroup analyses for the association of HGI and CKM syndrome stage (Figure S3), significant interactions were detected for sex (P for interaction = 0.007) and diabetes (P for interaction = 0.047). The relationship was more pronounced in males (OR 1.45, 95% CI 1.23–1.70) and non-diabetic participants (OR 1.59, 95% CI 1.17–2.15) compared to females (OR 1.02, 95% CI 0.87–1.20) and those with diabetes (OR 1.16, 95% CI 1.03–1.32). Among racial groups, Non-Hispanic White participants showed the strongest association (OR 1.30, 95% CI 1.10–1.54), while Non-Hispanic Black participants showed no significant association (OR 1.03, 95% CI 0.88–1.20). In general, higher HGI is associated with more severe CKM syndrome across different subgroups. In subgroup analyses for all-cause mortality (Figure S4), significant interaction was only found stratified by diabetes (P for interaction < 0.001). In the participants without diabetes, HGI was positively associated with all-cause mortality (HR 1.92, 95% CI 1.50–2.47, P < 0.001), while a negative association was detected in participants with diabetes despite marginal statistical significance (HR 0.91, 95% CI 0.82-1.00, P = 0.057). For participants with age below 60, HGI was positively associated with all-cause mortality (HR 1.33, 95% CI 1.04–1.70, P = 0.022), which was not observed in participants with age above 60 (HR 1.00, 95% CI 0.88–1.13, P = 0.986). Similarly, significant interactions were detected for age (P for interaction = 0.036) and diabetes (P for interaction = 0.002) in the subgroup analyses for cardiovascular mortality (Figure S5). Notably, HGI was positively associated with cardiovascular mortality in the participants without diabetes (HR 2.45, 95% CI 1.50-4.00, P < 0.001), while no significant association was detected in participants with diabetes (HR 1.09, 95% CI 0.92–1.29, P = 0.332). Collectively, these results suggest that diabetes and age may exert remarkable effect modifications on the positive correlations of HGI with all-cause and cardiovascular mortality. Sensitivity analyses To assess the robustness of the findings, we imputed missing values ​​for the covariates and expanded the sample size to 20,248 subjects (Table S3), followed with analyses of the associations of HGI with CKM syndrome stage and mortality using the same models. Similar results were obtained (Table S4 and S5, Figure S6 and S7). The consistent direction and significance of the associations across all models and imputed datasets suggest that the relationships between HGI and the severity and mortality of CKM syndrome were robust, with minimal influence from potential confounders or the handling of missing data. Discussion This large, nationally representative cohort study investigated the relationship between HGI and CKM syndrome severity, as well as related all-cause and cardiovascular mortality in the US general population. Key findings include a linear positive association between HGI and CKM syndrome severity, U-shaped relationships between HGI and all-cause mortality across CKM syndrome stages, and a linear positive link between HGI and cardiovascular mortality in CKM syndrome stage 4. Subgroup and sensitivity analyses confirmed these results. To our knowledge, this is the first study to explore the association of HGI with CKM syndrome severity and mortality. CKM syndrome is classified into five consecutive stages and assessed through various components, including excessive or dysfunctional adiposity, hyperglycemia, prediabetes, diabetes, hypertension, chronic kidney disease, cardiovascular disease, and others [1, 2]. Several previous studies have examined the association of HGI with these individual risk components. In a community-based cross-sectional survey in China, high HGI was independently associated with the risk of hypertension [25]. Two Chinese studies found that nondiabetic adults with high HGI displayed higher risks of insulin resistance and incident diabetes [26, 27]. With regard to vascular diseases, two cross-sectional studies conducted in Japan and Italy showed that high HGI was associated with increased prevalence of systemic arterial stiffening and carotid atherosclerosis [26, 28]. In a prospective Chinese cohort, high HGI was associated with an increased risk of incident stroke in adults [29]. A cross-sectional analysis performed in Korea showed that high HGI was independently associated with overall and individual cardiovascular diseases in patients with prediabetes or diabetes [30]. Additionally, HGI was shown to be associated with renal function decline or diabetic nephropathy, in spite of heterogenous dose-response results between the general population and patients with different types of diabetes [31–34]. Taken together, these studies demonstrate the association between HGI and the individual risk factors used in CKM syndrome staging. In our current study, we identified a linear relationship between HGI and CKM syndrome stage in the general population, highlighting the significant role of HGI in transitioning from assessing individual risk factors to evaluating overall integrated risk for CKM syndrome. Over the past few years, several studies have explored the utility of HGI as a predictor of all-cause and cardiovascular mortality in different populations. Consistent with previous studies in general population [19, 21, 35], U-shaped relationships between HGI and all-cause mortality were also observed among participants with CKM syndrome stages 0–3 and 4, indicating that HGI is a robust and universal predictor of all-cause death regardless of CKM syndrome stage. Notably, the inflection points differ between participants with CKM syndrome stages 0–3 (-0.280) and stage 4 (-0.077), which may help identify individuals at higher risk of death at different CKM syndrome stages. Previous studies have reported a U-shaped association between HGI and cardiovascular or cardio-cerebrovascular mortality in the general population [19, 21] and individuals with metabolic syndrome [18]. In contrast, we identified a lineal positive relationship between HGI and cardiovascular mortality in the overall population and participants with CKM syndrome stage 4. This discrepancy may be due to variations in participant characteristics. Compared to the previous general population [19, 21], our cohort had a higher proportion of males and individuals with hypertension. In comparison to individuals with metabolic syndrome [18], other than a higher proportion of males and individuals with hypertension, our cohort not only had a higher proportion of males and individuals with hypertension but was also generally younger and had a lower BMI. In addition, we observed divergent associations of HGI with mortality between diabetic and non-diabetic subgroups, which may be influenced by the cross-sectional nature of the study, unmeasured confounding factors, or potential survival bias. Specifically, individuals with advanced CKM syndrome or extreme HGI values may have been underrepresented due to premature mortality prior to study inclusion, which could bias the observed associations toward the null. Several potential mechanisms may explain the association between HGI and CKM syndrome severity. First, individuals with high HGI are more susceptible to glucose dysregulation, such as hypoglycemia and glycemic variability, which can impair vascular endothelial function and promote CKM syndrome progression [14, 36, 37]. Second, high HGI is associated with elevated levels of inflammatory biomarkers, including C-reactive protein, fibrinogen, complement C3, neutrophils, and monocytes, indicating that increased HGI may reflect heightened inflammatory burden and diminished anti-inflammatory capacity [18, 26, 30, 38, 39]. Given the critical role of inflammation in the development of cardiometabolic and vascular diseases, it is plausible that HGI contributes to CKM syndrome progression via pro-inflammatory signaling and oxidative stress. Third, a positive correlation between HGI and advanced glycation end-products (AGEs) has been observed in diabetic patients [40, 41]. AGEs can induce metabolic disturbances and endothelial dysfunction by binding to their receptors, leading to the production of pro-inflammatory factors and reactive oxygen species [41]. The pathways described above continue to be relevant in explaining the association between high HGI and increased all-cause and cardiovascular mortality [42–44]. Interestingly, our study also found that lower HGI is associated with higher mortality, displaying a U-shaped relationship in the restricted cubic spline analysis. Lower HGI may indicate a shorter red blood cell lifespan, which can result from conditions such as hemolytic anemia, chronic infections, hereditary disorders, or hematologic malignancies, thereby increasing mortality risk [14, 45]. However, the exact mechanisms underlying these associations are still unclear and warrant further research. According to a presidential advisory from the American Heart Association [1], there is a critical need for an approach to CKM syndrome staging that promotes prevention throughout the life course. Our data demonstrated a positive linear correlation between HGI and CKM syndrome stages in the general population, suggesting that routine monitoring of HGI early in life may enable clinicians to identify high-risk patients. This could provide opportunities for preventive interventions to halt or reverse disease progression and improve long-term outcomes. In this study, we observed a U-shaped relationship between HGI and all-cause mortality among patients with non-advanced and advanced stages of CKM syndrome, each with distinct inflection points. This indicates that HGI may serve as a valuable clinical biomarker for mortality risk assessment across CKM syndrome stages and underscores the importance of targeted public health strategies and clinical management tailored to specific stages of the syndrome. Furthermore, a linear association between HGI and cardiovascular mortality was observed in patients with stage 4 CKM syndrome, highlighting that healthcare professionals should consider individual HGI levels when managing these patients. Collectively, these findings offer novel insights into biological markers for CKM syndrome, enhancing risk stratification, informing clinical decision-making, and supporting the development of targeted therapeutic strategies. Our study has several strengths. First, it was conducted using a large, nationally representative sample of adults across different CKM syndrome stages, supporting the robustness of our findings. Second, it contributes new evidence on the relationship between HGI and mortality at both early and advanced CKM syndrome stages. Third, sensitivity analyses confirmed our main results, enhancing their credibility. However, there are limitations to consider. First, the cross-sectional design at baseline prevents us from establishing causality or understanding the temporal relationship between HGI and CKM syndrome severity. Future prospective studies are needed to evaluate whether HGI can predict CKM syndrome development over time, as well as the clinical relevance and underlying mechanisms. Second, there is a potential for misclassification due to the adapted CKM syndrome staging, which relies on limited NHANES variables and may not fully align with AHA criteria. To mitigate such bias, a well-designed prospective cohort study is necessary. Third, HGI was calculated from a single measurement of fasting plasma glucose and HbA1c, which does not account for long-term variability or dynamic changes over time, potentially affecting prognosis. Fourth, as an observational study, residual confounding may still exist despite adjustments for multiple covariates. Future studies incorporating detailed data on medication regimens, dietary patterns, inflammatory markers, and comprehensive renal function assessments are warranted to further elucidate these relationships and minimize residual confounding. Fifth, the imputation of missing data for sensitivity analyses may introduce bias. Finally, since our population was drawn from the US, the applicability of our findings to other ethnic groups or populations may be limited. Conclusions The HGI was positively correlated with more advanced stages of CKM syndrome. Furthermore, HGI demonstrated a U-shaped association with all-cause mortality in both early and late CKM syndrome stages, with different inflection points. In participants with advanced CKM syndrome, HGI was linearly linked to cardiovascular mortality. These complex and varying relationships between HGI and mortality highlight the need for personalized risk assessment strategies based on CKM syndrome stage. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2 Supplementary Material 3 Supplementary Material 3