Prognostic significance of the CALLY index for mortality among patients with cardiovascular–kidney–metabolic syndrome stages 0–3 Authors: Mingshen Lin, Shuaiqing Chen, Jing Ye, Xiaolei Hu Published: 2025 ABSTRACT AbstractBackgroundThe C-reactive protein–albumin–lymphocyte (CALLY) index integrates inflammation, nutrition, and immune status, serving as a novel, cost-effective biomarker. However, its association with mortality risk among individuals in stages 0 to 3 of the cardiovascular–kidney–metabolic (CKM) syndrome remains poorly defined.MethodsData from 9,886 individuals with stage 0–3 CKM syndrome in NHANES were examined. Cox proportional hazards regression, accounting for various covariates, was employed to examine the association between ln CALLY and mortality. RCS and threshold analysis were applied to investigate possible non-linear associations. Survival distributions were visualized using Kaplan–Meier curves, and additional analyses included time-dependent ROC analyses, subgroup analyses, and interaction tests.ResultsA strong inverse relationship was observed between ln CALLY and both all-cause mortality (HR = 0.86, 95% CI: 0.82–0.89) and cardiovascular mortality (HR = 0.81, 95% CI: 0.75–0.89). Participants in the top tertile of ln CALLY showed a 34% reduction in all-cause mortality and a 37% decline in cardiovascular mortality relative to the bottom tertile. RCS indicated a distinct L-shaped curve linking ln CALLY with all-cause mortality. Across multiple subgroups, the relationships remained stable, with no meaningful interaction effects identified.ConclusionsThis analysis demonstrates that the CALLY index is inversely related to mortality in CKM stage 0–3 patients, indicating its value as a potential biomarker for early prognosis assessment.Supplementary InformationThe online version contains supplementary material available at 10.1186/s13098-025-02068-x.Keywords: CALLY index, CKM, Mortality, NHANES, Inflammation, Nutrition, Immune Introduction Defined as a multisystem disorder, cardiovascular–kidney–metabolic (CKM) syndrome captures the tightly woven relationship between cardiovascular, renal, and metabolic impairments [1]. The interaction among these networks expedites cardiovascular deterioration and amplifies the threat of multiple organ dysfunction and critical cardiac outcomes [2]. In its 2023 Presidential Advisory, the American Heart Association (AHA) emphasized that CKM syndrome should be viewed as a dynamic, progressive disease state rather than a simple sum of its constituent risk factors [3]. To support clinical appraisal and focused intervention, the American Heart Association introduced a five-stage classification spanning from stage 0 to stage 4, with the latter indicating established cardiovascular disease, while stages 0 to 3 describe the preclinical spectrum [4]. National surveillance data from 2011 to 2020 suggest that more than 90% of U.S. adults fall within CKM stages 0 to 3, a population already demonstrating a heightened predisposition to cardiovascular events [5]. Furthermore, over 25% of U.S. adults meet CKM diagnostic criteria, contributing significantly to healthcare expenditures [6, 7]. Among patients with diabetes, those with coexisting chronic kidney disease (CKD) face a markedly higher risk of mortality, with 10-year rates exceeding 30%, compared with less than 15% among those without CKD [8]. This contrast highlights the additive burden of multisystem involvement. For individuals in CKM Stages 0–3, the AHA advocates shifting preventive efforts upstream by emphasizing early risk identification and prioritizing interventions to slow disease progression and reduce cardiovascular disease (CVD)-related events [9]. Given its central role in the clinical trajectory of CKM syndrome, CVD warrants particular attention in efforts to understand and interrupt disease progression [10]. The pathophysiological progression of CVD is driven in part by systemic inflammation, nutritional compromise, and immune dysregulation [11]. Independent correlations have been identified between high C-reactive protein (CRP) and low serum albumin levels with heightened cardiovascular and mortality risks [12–15]. Lymphocytes, as key modulators of inflammation in atherosclerotic processes, have also emerged as potential indicators of disease progression [16]. Due to the complex interrelationship among these biological systems, the use of a single biomarker may inadequately capture overall disease risk. Iida et al. have recently proposed the C-reactive protein–albumin–lymphocyte (CALLY) index, a combined biomarker based on CRP, albumin, and lymphocyte count, which was originally created to predict outcomes in hepatocellular carcinoma patients receiving liver resection [17]. Unlike traditional markers, the CALLY index synthesizes information from multiple systemic domains, thereby offering a more integrated and potentially robust predictor of disease progression in CVD contexts [18]. The prognostic importance of the CALLY index has been widely confirmed through extensive investigations in multiple patient cohorts, revealing a significant link to mortality in those afflicted with heart failure, sepsis, rheumatoid arthritis, acute coronary syndrome, chronic obstructive pulmonary disease, and cancer [19–24]. Although the CALLY index has shown prognostic value in other populations, its link to mortality risk in CKM syndrome stages 0–3 is still not well defined and requires additional study. As the CALLY index integrates systemic inflammation, nutritional status, and immune competence, we hypothesized that higher CALLY values would be inversely associated with mortality among adults with CKM stages 0–3. Accordingly, our study aimed to test this hypothesis using data from the National Health and Nutrition Examination Survey (NHANES) and to clarify the potential role of the CALLY index in early mortality risk assessment. Methods Data source and participants Updated biennially, NHANES employs an innovative multimodal data collection framework comprising five key components: demographic characteristics, dietary intake, clinical examinations, laboratory analyses of biological specimens, and standardized questionnaire responses [25]. A probability-based sampling design ensures nationally representative coverage across diverse geographic regions, racial and ethnic groups, and socioeconomic strata within the United States [26]. The NCHS-IRB formally authorized this research, with mandatory written consent procedures completed for all enrollees during the pre-participation screening phase. As an open-access resource, NHANES provides a robust foundation for evidence-based public health policymaking and chronic disease prevention research. This study analyzed data from six successive two-year NHANES cycles between 1999 and 2010, including 62,160 participants. Based on predefined exclusion criteria, we excluded 29,696 participants aged less than 20 years, 19,538 participants not classified within CKM stages 0 to 3, 800 participants with missing CALLY index data, and 2,240 participants with missing or abnormal covariate data. The final analytic cohort consisted of 9,886 participants after exclusions, with the selection flow depicted in Fig. 1. In addition, we provided detailed information on the excluded participants and compared their baseline characteristics with those of the included population (Table S1). Flow chart of the study Definitions of CKM syndrome stages 0 to 3 Participants were stratified into stages 0 through 3 of CKM syndrome according to criteria established by AHA. Stage 0 denotes an optimal health state with no identifiable risk factors. Stage 1 is marked by adipose tissue dysfunction, typically indicated by obesity or impaired glucose regulation. Stage 2 encompasses individuals with confirmed metabolic disorders or moderate CKD. Stage 3 reflects progression to subclinical cardiovascular or renal damage, defined as a 10-year cardiovascular risk ≥ 20% (per Framingham Risk Score) or severe CKD. Comprehensive staging criteria for CKM syndrome are detailed in Supplementary Table S2. Outcome variables Mortality status was ascertained by probabilistically linking NHANES participant records to the National Death Index (NDI) database. Follow-up commenced at the baseline assessment—defined as the date of the interview and physical examination—and continued until December 31, 2019. Mortality causes were categorized according to ICD-10 criteria. Cardiovascular deaths encompassed those attributed to cardiac conditions (ICD-10: I00-I09, I11, I13, I20-I51) and cerebrovascular disorders (ICD-10: I60-I69). All-cause mortality encompassed all confirmed deaths from any cause. Exposed variables Blood samples were obtained from each participant upon recruitment. After preparation, the samples were preserved at − 20 °C and then sent to accredited laboratories for thorough analysis. The CALLY index is computed by the product of lymphocyte count (10⁹/L) and albumin (g/dL), divided by the CRP concentration (mg/dL) multiplied by 10 [27]. Covariate definitions This study included various covariates related to sociodemographic characteristics, lifestyle behaviors, and health status. NHANES collected data on gender, age, race, marital status, education level, poverty-to-income ratio (PIR), smoking and drinking status, and history of hypertension and diabetes through household interview questionnaires. At the mobile examination center, body mass index (BMI) was measured. Laboratory measures included serum alanine aminotransferase (ALT) and serum creatinine (SCr). Smoking, alcohol use, diabetes, and hypertension were defined based on standardized questionnaires (SMQ020, ALQ101, DIQ010, BPQ020). Statistical analysis To address the pronounced asymmetry in the CALLY index distribution, a natural logarithmic transformation (ln CALLY) was implemented. Statistical analyses incorporated NHANES complex survey design elements including examination weights (WTMEC2YR), stratification (SDMVSTRA), and cluster variables (SDMVPSU) to maintain national representativeness. Continuous data are reported as means with standard deviations, while categorical data are shown as counts with proportions. Weighted rank-sum and chi-square tests were employed for continuous and categorical variable comparisons respectively. We employed Cox regression to examine the association between ln CALLY and all-cause and cardiovascular mortality, with results expressed as hazard ratios (HR) and corresponding 95% confidence intervals (CI). ln CALLY was analyzed as both a continuous variable and in tertile categories (T1–T3, reference = T1). Multivariable analyses included: Model 1 (crude); Model 2 (sex/age/race-adjusted); Model 3 (fully adjusted for sociodemographic, lifestyle, and clinical factors). To explore the dose-response association, including both linear and nonlinear relationships, we conducted a restricted cubic spline (RCS) analysis with the 5th, 50th, and 95th percentiles of ln CALLY as knot points, adjusting for confounders in Model 3. The optimal cutoff value of ln CALLY was identified through the RCS model. To further evaluate its prognostic relevance, a segmented logistic regression analysis was conducted to clarify the association between ln CALLY and mortality based on this threshold. Survival curves were generated using the Kaplan-Meier method to estimate survival probabilities among ln CALLY tertile groups, with statistical significance assessed by log-rank testing. In addition, to evaluate the ability of ln CALLY to predict mortality risk over both the short and long term, we performed time-dependent receiver operating characteristic (ROC) analyses and generated corresponding time-varying area under the curve (AUC) curves. We evaluated the incremental prognostic value of ln CALLY by comparing a base model with a new model. The base model included all covariates from Model 3 as well as estimated Glomerular Filtration Rate (eGFR) and Urine Albumin-to-Creatinine Ratio (UACR). The new model consisted of the base model plus ln CALLY. We compared the C-statistic (Harrell’s C, calculated using the survival package in R) and used the likelihood ratio test to assess statistical significance. We also calculated the Net Reclassification Improvement (NRI) and Integrated Discrimination Improvement (IDI) for 10-year risk using the survIDINRI package. Finally, to evaluate and compare the clinical utility of the models, Decision Curve Analysis (DCA) was performed for the 10-year risk of all-cause and cardiovascular mortality. After comprehensive covariate adjustment using the primary analytical framework (Model 3), we explored subgroup analyses by sex, age group, race, smoking and alcohol use, as well as diabetes, hypertension, and CKM. Interaction tests were incorporated to assess potential effect modification in the association. Furthermore, we performed three sensitivity analyses to evaluate the robustness of our findings. First, the potential influence of acute inflammation on baseline measurements was addressed by repeating the primary Cox regression analysis after excluding participants with a CRP level > 10 mg/L. Second, potential bias from excluding participants with missing covariate data was handled by performing multiple imputation using chained equations (MICE) to create 5 imputed datasets; the results from the Cox models were then pooled according to Rubin’s rules. Finally, for the cardiovascular mortality analysis, we utilized a Fine-Gray subdistribution hazard model to account for the impact of competing risks, treating non-cardiovascular death as a competing event. The entirety of the statistical workflow was implemented in R Studio, with significance determined at a two-tailed P-value of 0.05. Results Baseline characteristics of participants stratified by Ln CALLY tertiles With sampling weights considered, the 9,886 subjects were extrapolated to represent represented approximately 70,923,014 adults in the U.S. adults diagnosed with CKM stages 0 through 3. Baseline demographic data of participants stratified by all-cause and cardiovascular mortality are presented in Table 1. The mean age recorded was 44.7 ± 15.75 years; males comprised 48.3%, and non-Hispanic Whites accounted for 72.6%. Over the 13.53-year average observation period, 1,570 deaths occurred due to all causes, with 365 resulting specifically from cardiovascular conditions. The mean CALLY index was 10.3 ± 17.33. Compared with individuals in the lowest tertile, those in the highest ln CALLY tertile tended to be younger males with higher education and income levels and better weight management. Smoking prevalence was reduced among them, whereas alcohol use was comparatively more common. Additionally, they had a lower prevalence of diabetes and hypertension and experienced fewer deaths from all causes and cardiovascular causes. Table 1Participant baseline characteristics stratified by all-cause and cardiovascular mortality Characteristics Overall N = 70,923,014 n = 9,886 All-cause mortality P -value Cardiovascular mortality P -value Yes (n = 1,570) No (n = 8,316) Yes (n = 365) No (n = 9,521) Age (years) 44.7 ± 15.7563.5 ± 15.7142.3 ± 13.99<0.00165.3 ± 15.5344.2 ± 15.39<0.001 Gender (%) 0.0240.019Male48.3%51.8%47.9%56.3%48.1%Female51.7%48.2%52.1%43.7%51.9% Race (%) <0.001<0.001Non-Hispanic White72.6%80.2%71.6%79.0%72.4%Non-Hispanic Black10.1%10.1%10.1%13.0%10.0%Others17.4%9.8%18.3%8.0%17.6% Marital status (%) 0.0130.059Married59.2%55.6%59.7%53.5%59.3%Unmarried and others40.8%44.4%40.3%46.5%40.7% Education level (%) <0.001<0.001Less than high school17.2%27.7%15.8%25.6%17.0%High school or GED24.9%29.0%24.4%32.8%24.7%Above high school57.9%43.3%59.8%41.7%58.3% PIR (%) <0.001<0.001˂ 1.522.1%28.4%21.3%31.3%21.9%1.5 - 3.532.6%39.0%31.8%36.5%32.5%˃ 3.545.2%32.6%46.9%32.2%45.6%BMI (kg/m2, %)0.3230.008˂ 2534.3%32.3%34.5%25.8%34.5%25 - 3034.1%34.7%34.0%33.8%34.1%˃ 3031.6%33.0%31.5%40.5%31.4% Smoking (%) <0.0010.012Yes47.7%60.3%46.1%55.6%47.5%No52.3%39.7%53.9%44.4%52.5% Alcohol use (%) <0.001<0.001Yes75.4%68.3%76.4%64.1%75.7%No24.6%31.7%23.6%35.9%24.3% Diabetes (%) <0.001<0.001Yes6.0%15.7%4.7%18.7%5.6%No94.0%84.3%95.3%81.3%94.4% Hypertension (%) <0.001<0.001Yes26.1%46.8%23.4%53.4%25.4%No73.9%53.2%76.6%46.6%74.6% C KM (%) <0.001<0.001Stage 014.4%3.7%15.8%1.5%14.8%Stage 120.3%6.9%22.0%5.4%20.7%Stage 261.3%63.7%60.9%64.7%61.2%Stage 34.0%25.6%1.2%28.4%3.4% ALT (U/L) 26.1 ± 19.4025.9 ± 24.6626.1 ± 18.600.78227.8 ± 33.3426.0 ± 18.900.495 SCr (umol/L) 75.6 ± 24.8083.8 ± 50.2774.5 ± 18.81<0.00183.4 ± 32.3675.4 ± 24.54<0.001 CALLY index 10.3 ± 17.336.7 ± 13.9110.8 ± 17.68<0.0015.3 ± 10.9410.5 ± 17.45<0.001Continuous variables were showed as mean ± SD, categorical variables were showed as percentagePIR, poverty index ratio; BMI, body mass index; CKM, cardiovascular-kidney-metabolic syndrome; ALT, serum alanine aminotransferase; SCr, serum creatinine Participant baseline characteristics stratified by all-cause and cardiovascular mortality Continuous variables were showed as mean ± SD, categorical variables were showed as percentage PIR, poverty index ratio; BMI, body mass index; CKM, cardiovascular-kidney-metabolic syndrome; ALT, serum alanine aminotransferase; SCr, serum creatinine Association between Ln CALLY and mortality The Cox proportional hazards models in Table 2 demonstrate the association between ln CALLY and mortality outcomes in patients with CKM stages 0 to 3. Elevated ln CALLY levels showed statistically significant inverse relationships with mortality. For all-cause mortality, ln CALLY as a continuous variable was significantly associated with decreased risk in Model 1 (HR = 0.79, 95% CI: 0.76–0.82) and Model 2 (HR = 0.85, 95% CI: 0.82–0.89), with the association persisting in the fully adjusted Model 3 (HR = 0.86, 95% CI: 0.82–0.89). When stratified into tertiles, participants in T3 exhibited a 34% lower risk of all-cause mortality compared to T1 (HR = 0.66, 95% CI: 0.57–0.75, P for trend < 0.001). A similar pattern was observed for cardiovascular mortality, where ln CALLY remained significantly inversely associated across all models (Model 1: HR = 0.74, 95% CI: 0.69–0.80; Model 2: HR = 0.79, 95% CI: 0.73–0.86; Model 3: HR = 0.81, 95% CI: 0.75–0.89). Notably, participants in T3 had a 37% lower risk of cardiovascular mortality compared to T1 (HR = 0.63, 95% CI: 0.47–0.84, P for trend < 0.05). Table 2Association between Ln CALLY and all-cause and cardiovascular mortalityCharacteristicModel 1Model 2Model 3HR95% CIP-valueHR95% CIP-valueHR95% CIP-value All-cause mortality ln CALLY (continuous) 0.790.76, 0.82< 0.0010.850.82, 0.89< 0.0010.860.82, 0.89< 0.001 ln CALLY T1 RefRefRefRefRefRef T2 0.800.72, 0.90< 0.0010.750.67, 0.84< 0.0010.790.70, 0.88< 0.001 T3 0.500.44, 0.57< 0.0010.660.58, 0.75< 0.0010.660.57, 0.75< 0.001P for trend< 0.001< 0.001< 0.001 Cardiovascular mortality ln CALLY (continuous) 0.740.69, 0.80< 0.0010.790.73, 0.86< 0.0010.810.75, 0.89< 0.001 ln CALLY T1 RefRefRefRefRefRef T2 0.830.66, 1.050.1180.780.62, 0.980.0340.850.67, 1.070.166 T3 0.440.33, 0.58< 0.0010.590.44, 0.78< 0.0010.630.47, 0.840.002P for trend< 0.001< 0.0010.002Model 1: UnadjustedModel 2: Adjusted for Gender, Age, and RaceModel 3: Adjusted for Gender, Age, Race, Marital status, Education level, PIR, BMI, Smoking, Alcohol use, Diabetes, Hypertension, ALT, and SCr Association between Ln CALLY and all-cause and cardiovascular mortality Model 1: Unadjusted Model 2: Adjusted for Gender, Age, and Race Model 3: Adjusted for Gender, Age, Race, Marital status, Education level, PIR, BMI, Smoking, Alcohol use, Diabetes, Hypertension, ALT, and SCr RCS analysis provides further insights into the association between ln CALLY and all-cause mortality and cardiovascular mortality among patients with CKM stages 0 to 3 (Fig. 2). The solid line represents the estimated HR, with the shaded region indicating the 95% CI. The analysis revealed an L-shaped association between ln CALLY and all-cause mortality (P-nonlinear < 0.05), while a linear relationship was observed with cardiovascular mortality (P-nonlinear > 0.05). As shown in Table 3, threshold effect analysis identified 1.7 as a key inflection point (P for log likelihood ratio < 0.05). ln CALLY levels below 1.7 were markedly linked to a decreased all-cause mortality risk (HR = 0.81, 95% CI: 0.76–0.85, P < 0.001), whereas no meaningful association was noted when ln CALLY reached or exceeded 1.7 (HR = 1.01, 95% CI: 0.90–1.13, P > 0.05). RCS fitting for the association between ln CALLY and all-cause (A) and cardiovascular (B) mortality Table 3Threshold effect analysis of Ln CALLY on all-cause mortalityCharacteristicHR95% CIP-value Break-point = 1.7 ln CALLY < 1.7 0.810.76, 0.85< 0.001 ln CALLY ≥ 1.7 1.010.90, 1.130.879 Log likelihood ratio 0.003Adjusted for Gender, Age, Race, Marital status, Education level, PIR, BMI, Smoking, Alcohol use, Diabetes, Hypertension, ALT, and SCr Threshold effect analysis of Ln CALLY on all-cause mortality Adjusted for Gender, Age, Race, Marital status, Education level, PIR, BMI, Smoking, Alcohol use, Diabetes, Hypertension, ALT, and SCr Prognostic value of Ln CALLY Significant disparities in both all-cause and cardiovascular mortality were observed among ln CALLY tertiles based on Kaplan–Meier survival curves (Fig. 3). Participants in T3 exhibited significantly higher long-term survival rates compared to those in T1 (P < 0.05, log-rank). Kaplan–Meier survival analysis plot for all-cause (A) and cardiovascular (B) mortality with tertile groups of ln CALLY In addition, time-dependent ROC analysis demonstrated that ln CALLY consistently maintained moderate and stable discriminatory power in predicting both short- and long-term mortality (Fig. 4). The AUCs were 0.632, 0.618, and 0.601 at 3, 5, and 10 years for all-cause mortality, and 0.602, 0.615, and 0.636 for cardiovascular mortality. Time-dependent ROC curves and time-dependent AUC values of ln CALLY for predicting all-cause (A-B) and cardiovascular (C-D) mortality Finally, we performed several analyses to assess the incremental prognostic value of adding ln CALLY to the base model for both all-cause and cardiovascular mortality. For all-cause mortality, the addition of ln CALLY resulted in a significant improvement in model discrimination, with the C-index increasing from 0.8623 to 0.8659 (P < 0.001). Similarly, for cardiovascular mortality, the C-index increased from 0.8778 to 0.8836 (P < 0.001). Furthermore, ln CALLY showed significant incremental value in reclassification metrics. For all-cause mortality, NRI was 0.119 (P < 0.001) and IDI was 0.011 (P < 0.001). Similar results were observed for cardiovascular mortality (NRI = 0.128, P < 0.001; IDI = 0.007, P < 0.001), indicating a significant improvement in risk prediction and reclassification (Table S3). However, DCA for 10-year mortality suggested that this statistical improvement did not translate into a substantial gain in clinical utility. As shown in Figure S1, the net benefit of the new model was only marginally higher than that of the base model, and their 95% CI largely overlapped across the range of risk thresholds. For instance, at a 20% risk threshold for all-cause mortality, this translated to a modest net identification of only 0.267 additional true positives per 100 patients (Table S4). Subgroup and sensitivity analyses We employed forest plots to illustrate the findings from subgroup analyses and interaction tests for ln CALLY and all-cause mortality and cardiovascular mortality (Fig. 5). The findings revealed no significant heterogeneity among subgroups (all P for interaction > 0.05), further affirming the robustness and consistency of these associations across diverse populations. Subgroup analysis of ln CALLY with all-cause (A) and cardiovascular (B) mortality The results of our sensitivity analyses confirmed the robustness of our primary findings. After excluding participants with CRP levels > 10 mg/L, the inverse association between ln CALLY and both all-cause mortality (HR = 0.85, 95% CI: 0.81–0.89) and cardiovascular mortality (HR = 0.82, 95% CI: 0.74–0.90) remained significant and consistent with our main analysis (Table S5). Furthermore, in the analysis using multiple imputation for missing covariates, the associations were also materially unchanged for both all-cause mortality (HR = 0.87, 95% CI: 0.84–0.91) and cardiovascular mortality (HR = 0.83, 95% CI: 0.76–0.90), further strengthening the credibility of our results (Table S6). Finally, after accounting for non-cardiovascular deaths as competing events using a Fine-Gray model, the inverse association between ln CALLY and cardiovascular mortality remained significant and robust. In the fully adjusted model, higher ln CALLY was associated with a lower risk of cardiovascular mortality, both as a continuous variable (subdistribution HR = 0.86, 95% CI: 0.79–0.94) and when comparing the highest to the lowest tertile (subdistribution HR = 0.71, 95% CI: 0.53–0.95), further supporting our primary findings (Table S7). Discussion This study retrospectively analyzed NHANES data to assess how the CALLY index correlates with mortality among adults in the United States diagnosed with CKM syndrome stages 0 to 3. After comprehensive multivariable adjustment, ln CALLY was found to be significantly inversely associated with mortality. Further analysis revealed an L-shaped association between ln CALLY and all-cause mortality, with an inflection point at 1.7. Below this threshold, increases in the CALLY index were strongly associated with reduced mortality risk; above it, the association plateaued, suggesting a potential saturation effect. Survival analysis demonstrated that individuals in the highest tertile of ln CALLY had significantly better survival outcomes compared with those in the lowest tertile. Moreover, ln CALLY showed good predictive performance for 3-, 5-, and 10-year mortality. Stratified subgroup analyses confirmed the robustness and consistency of these associations across various demographic and clinical subgroups. These findings underscore the potential utility of the CALLY index as a simple, integrative biomarker for early mortality risk assessment in individuals with preclinical CKM syndrome. To our knowledge, this is the first study to demonstrate a significant association between the CALLY index and mortality outcomes among patients in CKM stages 0 to 3. The CALLY index has garnered increasing attention for its prognostic utility across various disease settings. Han et al., analyzing 2,183 participants with preexisting cardiovascular disease from NHANES (2001–2010), found that higher CALLY scores were significantly associated with lower risks of all-cause (HR = 0.58, 95% CI: 0.48–0.71) and cardiovascular mortality (HR = 0.54, 95% CI: 0.38–0.76), with a nonlinear association confirmed by RCS analysis [28]. He et al. found that in a sample of 320 older adults diagnosed with Heart failure with preserved ejection fraction (HFpEF), the CALLY index served as an independent prognostic indicator for long-term mortality (HR = 0.811, 95% CI: 0.714–0.921) [29]. Similarly, in renal populations, a study involving 436 maintenance hemodialysis (MHD) patients showed the index to be a significant protective factor for mortality. When incorporated into a predictive nomogram, the model achieved excellent discrimination for all-cause mortality (AUC = 0.821, 95% CI: 0.778–0.861) [30]. In the metabolic domain, a recent NHANES-based study (2003–2010) observed that higher CALLY index were inversely associated with the prevalence of metabolic syndrome, with the highest quartile showing markedly reduced odds compared to the lowest (OR = 0.25, 95% CI: 0.20–0.32) [31]. In contrast to prior investigations focused on individuals with established disease, the current study is the first to validate the association between the CALLY index and mortality among individuals across CKM stages 0 to 3, representing a continuum from no evident risk to subclinical pathology. Taken together, despite variability in study design, sample size, and outcome definitions, the body of evidence consistently supports a robust inverse association between CALLY index and mortality, reinforcing its potential clinical value as a composite prognostic biomarker. CRP, serum albumin, and peripheral lymphocyte count play integral roles in modulating clinical outcomes in cardiovascular, renal, and metabolic diseases through interrelated pathways of inflammation, nutritional status, and immune function. CRP, in particular, is a potent pro-inflammatory mediator that exacerbates vascular inflammation via activation of the complement system and upregulation of cytokines such as interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α) [32]. These effects contribute to endothelial dysfunction and destabilization of atherosclerotic plaques [33]. Additionally, CRP has been shown to promote renal tubular injury and interstitial fibrosis through activation of the NF-κB and Smad3 signaling pathways [34]. In the context of metabolic regulation, CRP, in concert with IL-6, enhances inflammatory signaling in adipose tissue, thereby worsening insulin resistance and impairing metabolic homeostasis [35, 36]. In contrast, serum albumin is a well-established indicator of both nutritional status and chronic systemic inflammation [37]. In patients with CKD, hypoalbuminemia often signifies the presence of malnutrition-inflammation complex syndrome, characterized by protein loss due to persistent proteinuria and reduced hepatic synthesis under inflammatory conditions [38]. This condition has been strongly associated with accelerated atherosclerotic progression and increased all-cause mortality [39]. Additionally, low albumin levels are linked to elevated oxidative stress and coagulation disturbances, both of which may exacerbate vascular injury and atherosclerotic risk [40]. Lymphocyte count is a vital biomarker reflecting the integrity of adaptive immune function and the capacity to regulate systemic inflammation [41]. Adequate lymphocyte levels are indispensable for effective immune surveillance and protection against disease progression. In contrast, lymphopenia typically indicates an immunosuppressed state, commonly observed in chronically malnourished individuals [42]. Persistent systemic inflammation and elevated stress hormone levels can lead to increased lymphocyte apoptosis and aberrant redistribution, ultimately compromising immune efficacy [43]. Additionally, protein-energy malnutrition may suppress lymphopoiesis and attenuate Th1-type cytokine production, particularly interferon-γ, thereby exacerbating vulnerability to disease progression [44]. The CALLY index, derived from C-reactive protein, serum albumin, and lymphocyte count, effectively integrates markers of inflammation, nutrition, and immune function, thereby offering a holistic snapshot of systemic health. As a prognostic tool, it has demonstrated significant value in the early risk stratification of cardiovascular, renal, and metabolic diseases. Findings from this study extend its relevance to individuals in stages 0 to 3 of CKM syndrome, highlighting its promising clinical applicability even in subclinical disease states. Beyond systemic inflammation, nutrition, and immunity, adipose tissue itself actively contributes to cardiometabolic risk through secretion of bioactive molecules. Dysregulated adipokines such as leptin and adiponectin, together with inflammatory mediators including complement factor H, serum amyloid A, and pigment epithelium-derived factor, have been shown to modulate insulin resistance, vascular inflammation, and atherosclerotic progression [45–50]. These adipose-derived factors represent additional pathways of dysfunctional adiposity that may influence CKM progression and mortality. Although the CALLY index provides an integrative reflection of inflammation, nutrition, and immune status, it cannot fully capture these adipose-specific mechanisms. Future studies should therefore consider incorporating both systemic and adipose tissue-derived biomarkers to improve early risk stratification. One of the key findings of this study is the identification of a statistical inflection point at ln CALLY = 1.7 in the association with all-cause mortality. This data-driven value marks a threshold beyond which the inverse association between the CALLY index and mortality began to plateau. From a methodological standpoint, this finding highlights the nonlinear nature of the relationship. We caution that this threshold is derived from our specific dataset and requires external validation before being considered for any clinical application. Furthermore, a deeper evaluation of the prognostic value of the index revealed a nuanced distinction between its statistical improvement and clinical utility. Although adding the CALLY index yielded statistically significant improvements in reclassification metrics, this did not translate into a substantial increase in net benefit on DCA. This was assessed for a risk horizon of 10 years, a standard timeframe for preventive strategies in CKM syndrome. Within the clinically relevant threshold range of 5% to 40%, where decisions about intensifying lifestyle counseling or initiating pharmacotherapy are typically made, the 95% CI for the new model’s net benefit largely overlapped with those of the base model. This suggests that while the CALLY index provides novel prognostic information from a statistical standpoint, its added clinical utility may be modest when incorporated into a comprehensive risk model that already includes robust predictors of clinical and kidney health. This finding underscores the high bar for new biomarkers to demonstrate clear clinical benefit over existing models. Despite these promising findings, several limitations and potential biases should be acknowledged. A primary limitation concerns the generalizability of our findings. As our new analysis in Table S1 demonstrates, our final cohort differs systematically from the excluded population in characteristics such as age, socioeconomic status, and comorbidity burden. The excluded group is a heterogeneous population that includes individuals with established clinical CVD (CKM Stage 4) as well as those who could not be staged due to missing key health data. Consequently, our conclusions are most applicable to adults within CKM stages 0–3 with relatively complete health records, and caution is warranted when extrapolating these results to a broader, more complex real-world population. Additionally, although extensive confounder adjustment was performed, residual confounding from unmeasured or uncontrolled factors cannot be fully excluded. The CALLY index was also measured solely at baseline without accounting for changes over time, leaving the influence of such dynamics on mortality risk uncertain. Finally, given the characteristics of the NHANES dataset, our findings are specific to the U.S. population, and further studies are needed to confirm their external applicability. Conclusion In summary, our study demonstrates a significant inverse relationship between the CALLY index and both all-cause and cardiovascular mortality among patients with stage 0 to 3 CKM syndrome. These findings emphasize the potential of the CALLY index as a tool for early identification of high-risk individuals in this population. Future large-scale, multicenter prospective studies are essential to confirm the robustness and external validity of these results. Supplementary Information Supplementary Material 1 Author contributions Mingshen Lin and Xiaolei Hu participated in data collection and drafting and editing of the paper. Shuaiqing Chen and Jing Ye participated in data analysis and revision of the paper. All authors saw and approved the final version and no other person made a substantial contribution to the paper. Funding This work was supported by the Zhejiang Province Traditional Chinese Medicine Science and Technology Plan Project (project number 2025ZL616). The funding body did not participate in the design of the study, data collection, analysis, interpretation of data or in writing the manuscript. Data availability The datasets generated and analysis during the current study are available in the NHANES, www.cdc.gov/nchs/NHANEs/. Declarations Ethics approval and consent to participate In accordance with NHANES protocol, all participants in the NHANES study provided written informed consent. As this study utilized publicly available data from the NHANES database, no additional ethical approval or consents were required. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement The datasets generated and analysis during the current study are available in the NHANES, www.cdc.gov/nchs/NHANEs/.