Albuminuria is a common manifestation of diabetic kidney disease (DKD), a major complication of diabetes mellitus, and has been recognized as a marker not only of renal injury but also of systemic vascular pathology, including subclinical atherosclerosis (SA) (1–3). Large epidemiologic studies have further established chronic kidney disease (CKD) markers as risk factors for cardiovascular events; thus, both reduced glomerular filtration and elevated urinary albumin-to-creatinine ratio (UACR) confer significantly higher risks of cardiovascular morbidity and mortality (4, 5). Even low-grade albuminuria (10–30 mg/g) has been independently associated with early indicators of SA, such as increased carotid intima-media thickness (CIMT) and abnormal ankle-brachial index (ABI) (6–9). However, the mechanistic pathways linking albuminuria to vascular injury remain only partially elucidated.

Chronic low-grade inflammation is a well-established contributor to both DKD progression and atherogenesis (10–13). In type 2 diabetes, chronic low-grade inflammation arises from multiple metabolically active tissues. Expansion of visceral adipose tissue and the frequent coexistence of non-alcoholic fatty liver disease (NAFLD) contribute to systemic cytokine release and have been linked to early atherosclerotic changes, including increased carotid intima-media thickness (14). These metabolic inflammatory sources may interact with kidney-associated inflammatory pathways in diabetic kidney disease, amplifying vascular risk.

A growing body of evidence suggests that albuminuria may not merely reflect renal endothelial dysfunction but may actively amplify systemic inflammation through tubular cell activation, release of pro-inflammatory mediators, and recruitment of immune cells. Inflammatory cytokines such as tumor necrosis factor α (TNFα), interleukin (IL) IL6, and IL1β are upregulated in patients with renal impairment and directly contribute to vascular remodeling and plaque instability (15, 16). Circulating peripheral blood leukocytes represent a key source of these inflammatory mediators and have been implicated in the development of atherosclerotic lesions as well as the abnormal immune activation observed in CKD (17). Notably, pro-inflammatory monocyte subsets (CD14⁺CD16⁺) are expanded in T2DM patients with nephropathy and are associated with microinflammation in DKD, highlighting the link between systemic immune cell activation and renal injury (18).

Alterations in blood leukocyte RNA, particularly those involving inflammatory cytokines, pattern recognition receptors (e.g., toll-like receptor [TLR]), chemokines (e.g., C-C motif chemokine ligand 2 [CCL2]), and transcription factors (e.g., nuclear factor kappa B [NFκB]), have been reported in both DKD and cardiovascular disease (19–21). Nevertheless, few studies have simultaneously assessed the interplay between UACR, systemic inflammatory mediators, circulating leukocyte gene expression, and SA in a well-characterized population with DKD. This represents a critical knowledge gap in understanding the kidney–heart axis.

In this cross-sectional study, we investigated the hypothesis that albuminuria serves, at least in part, as a biological mediator linking systemic inflammation to early atherosclerosis in DKD. We analyzed a cohort of patients with T2DM and moderate CKD (stages G3a/b) to explore the associations between UACR, circulating inflammatory parameters (high-sensitivity C-reactive protein [hs-CRP], TNFα, IL6, IL1β, IL10), and gene expression profiles in peripheral blood leukocytes (including TNF, IL6, IL1B, IL10, CD36, CCL2, TLR2, TLR4, and NFKB). Vascular outcomes were evaluated using CIMT and ABI as markers of subclinical arterial damage. Our findings offer novel insights into the role of albuminuria as an active participant in the inflammatory-atherosclerotic axis in DKD, providing context for potential therapeutic interventions targeting inflammation.

In this study, we investigated the mediating role of inflammatory markers—both circulating proteins and circulating leukocyte gene expression—in the relationship between albuminuria and SA in patients with T2DM and moderate CKD. Our findings support the hypothesis that UACR is not merely a marker of glomerular injury but actively contributes to a pro-inflammatory state with systemic vascular consequences.

Previous longitudinal studies have identified albuminuria and reduced eGFR as independent risk factors for cardiovascular events in individuals with diabetes (24). The increased cardiovascular risk in CKD may be partially explained by endothelial dysfunction and accelerated large-vessel atherosclerosis. Consistent with previous research (25, 26), we observed significant associations between UACR and both CIMT and ABI, indicating that even in the absence of overt cardiovascular disease, higher albuminuria correlates with early vascular alterations. The direct association between UACR and CIMT observed in our study, alongside the inverse association with ABI, supports the notion that albuminuria reflects early vascular injury that may precede more advanced atherosclerotic disease in this population. Importantly, our data suggest that elevated UACR is linked to SA independently of traditional risk factors, reinforcing the importance of albuminuria monitoring in diabetic CKD patients as part of cardiovascular risk assessment.

A central finding of our study is the association between elevated UACR and systemic inflammation. Notably, serum IL6 and leukocyte IL6 mRNA accounted for roughly 40% of the association between albuminuria and SA, underscoring IL6 as a pivotal inflammatory mediator of vascular injury linked to albuminuria. This is consistent with the established pro-atherogenic role of IL6, which includes stimulating hepatic CRP production, endothelial activation, and leukocyte recruitment, key events in vascular injury and plaque formation (26–29). This aligns with the broader literature establishing IL6 as both a marker and driver of cardiovascular risk. Indeed, recent evidence from the MESA cohort indicates that IL6 levels are an even stronger predictor of cardiovascular events than hs-CRP (30), emphasizing its central pathogenic role. Moreover, a recent Phase 2 trial of IL6 inhibition in CKD patients with elevated inflammation (the RESCUE trial) demonstrated that antagonizing IL6 signaling (using the monoclonal antibody ziltivekimab) can dramatically reduce systemic inflammatory biomarkers (31). Similarly, IL1β contributed significantly as a mediator between UACR and SA. IL1β is a major downstream effector of the NLRP3 inflammasome, whose upregulation in CKD has been linked to renal and vascular inflammation, plaque destabilization and atherogenesis. In our analysis, IL1β explained 17.4% of the effects of albuminuria on SA. Notably, targeting of IL1β with canakinumab has shown cardiovascular benefit in individuals with elevated CRP (as evidenced by the CANTOS trial) (32). It is also worth noting that residual inflammatory risk related to upstream mediators such as IL6 persists even after IL1β inhibition (33), highlighting the need to target multiple points in the inflammatory cascade. In any case, RESCUE and CANTOS trials underscore the causal role of inflammation in atherosclerosis and suggest potential therapeutic avenues for high-risk groups including those with DKD. Our findings solidify the concept that specific innate immune cytokines like IL6 and IL1β are not just markers but active drivers of cardiovascular risk in CKD population.

Mediation analysis also revealed a significant, though smaller, mediating effect of TLR2 gene expression in leukocytes, accounting for approximately 7.2% of the association between UACR and SA. LR2 is an innate immune receptor that recognizes endogenous damage-associated molecular patterns (DAMPs) released during tissue injury. In the context of DKD, excessive filtered albumin can act as a DAMP that activates tubular cells and monocytes via TLRs. TLR engagement triggers NF-κB signaling and induces transcription of cytokines like IL6 and IL1β (34, 35). The upregulation of TLR2 in our patients with higher albuminuria and SA suggests that albuminuria-induced renal inflammation may initiate systemic immune activation through TLR pathways. This is consistent with experimental data showing that TLR activation in the kidney exacerbates inflammation and injury (36). Our findings thus support a model wherein albuminuria is associated with TLR signaling, and may contribute to systemic vascular inflammation. Notably, increased expression of TLR2 and TLR4 – receptors known to mediate responses to DAMPs – may further amplify the inflammatory burden in patients with DKD and SA (37).

Our findings align with previous studies demonstrating a strong correlation between albuminuria and systemic inflammation and support the role of albuminuria as a biological amplifier of systemic inflammation, contributing to vascular remodeling and early atherosclerosis (38–40). Albuminuria appears to function as more than a passive marker, instead acting as a biological amplifier of inflammation that promotes vascular remodeling. The mechanistic basis for this likely involves multiple interrelated pathways. Excessive filtered albumin acts as a DAMP, activating inflammatory pathways such as the NLRP3 inflammasome in tubular epithelial cells, and stimulating the release of cytokines that can enter systemic circulation, contributing to low-grade systemic inflammation and vascular dysfunction (27, 40, 41). Moreover, tubular epithelial cells exposed to albumin release extracellular vesicles, including exosomes carrying mRNA and protein cargo such as CCL2. These can be taken up by immune cells, including macrophages, enhancing their activation and promoting renal inflammation (42). Our results are in accordance with this mechanism supporting for the hypothesis that systemic inflammation through both humoral (cytokine) and cellular (gene expression) pathways mediates the contribution of albuminuria to early atherogenesis. The graded mediation effects for IL6, IL1β, and TLR2 suggest the involvement of multiple branches of the inflammatory cascade – humoral (cytokine-mediated), cellular (leukocytes activation), and innate immune receptor pathways – likely interacting synergistically.

A novel aspect of our study is the integrated analysis of both serum cytokine levels and gene expression in peripheral blood leukocytes. Changes in peripheral leukocyte gene expression reflect systemic immune alterations and have been implicated in the pathogenesis of both DKD and atherosclerosis (43–48). Altered mRNA expression in peripheral leukocytes—including lymphocytes and mononuclear cells—has been reported in patients with diabetes, diabetic nephropathy, CKD, metabolic syndrome, and advanced atherosclerosis (19, 20, 49, 50). Thus, in patients with carotid atherosclerosis, pro-inflammatory gene signatures in circulating monocytes mirror those found in plaque-resident immune cells. Our finding that leukocyte expression of genes like TNF, TLR2, and CD36 correlates with albuminuria and SA supports the concept of a systemic “inflammatory fingerprint” linking kidney and vascular pathology. Notably, Sternberg et al. demonstrated that immune cells isolated from atherosclerotic plaques and peripheral blood share common inflammatory activation patterns, underscoring the interplay between circulating mononuclear cells and vascular inflammation (21). By concurrently assessing soluble biomarkers and circulating leukocyte mRNA, our study provides a more comprehensive view of the inflammatory state in DKD and its contribution to atherogenesis.

CD36 is mechanistically linked to oxidized LDL uptake, foam-cell formation, and early atherogenesis, which makes it a biologically plausible marker. However, in our dataset, although leukocyte CD36 expression was higher in individuals with SA and macroalbuminuria, its association with SA was attenuated and no longer significant after multivariable adjustment. CD36 also did not emerge as a significant mediator in our exploratory mediation analyses, particularly after FDR correction. These findings suggest that its independent contribution beyond IL6, IL1β, and TLR2 is limited, possibly reflecting collinearity with related innate immune pathways or insufficient power to detect a modest effect. Therefore, no firm mechanistic conclusions can be drawn regarding CD36 in this cohort, and its role warrants further investigation.

Beyond IL6 and IL1β, other cytokines such as IL-18 have also been implicated in atherosclerosis and vascular inflammation and may represent additional mechanistic links between metabolic disease and vascular injury, although these pathways were not directly assessed in the present study (51). In addition to kidney-derived inflammatory signals, obesity- and NAFLD-associated inflammation represents an important upstream contributor to systemic inflammatory burden in T2DM. Hepatic steatosis and visceral adipose tissue dysfunction have been associated with increased carotid intima-media thickness and with cytokines such as IL-18, IL-15, and IL-17 in prior studies, suggesting that vascular injury in diabetes reflects the convergence of multiple metabolic inflammatory pathways rather than a single organ-specific mechanism (14, 51–53).

It is important to note that several observed correlations, although statistically significant, were of modest magnitude. Such small effect sizes suggest limited standalone biological impact and highlight that inflammation likely acts through multiple overlapping pathways rather than a single dominant mediator. Accordingly, these correlations should be interpreted cautiously and not overstated.

While our findings are compelling, several limitations warrant consideration. The pattern of associations suggests that inflammation may help explain why higher albuminuria coincides with a greater vascular disease burden. However, because the study is cross-sectional, these observations are not evidence of causality and we cannot definitively prove that albuminuria causes inflammation-mediated atherosclerosis, only that they are strongly associated. Thus, all associations reported here should be interpreted as non-causal and hypothesis-generating. Prospective studies are needed to confirm temporal relationships and causality. Second, unmeasured confounders may be present despite multivariable adjustment; for instance, we did not have data on dietary factors or gut microbiota-derived toxins that might influence inflammation. Third, this was a single-center study involving a relatively homogenous population in which all participants received uniform background therapy with RAS blockers and statins, which may limit generalizability of our findings to patients not receiving these treatments or to broader ethnic or geographic groups. Fourth, SA was assessed by CIMT and ABI rather than direct imaging of plaques; although these are established markers, they are surrogate measures of atherosclerosis. Mediation analysis assumptions should be acknowledged – we modeled the mediating effect of each inflammatory marker independently, although these pathways overlap and may interact. Fifth, our analysis focused on select inflammatory pathways; other mediators such as oxidative stress, adipokines, or components of adaptive immunity, as well as residual confounding factors like homocysteine, were not measured and may also contribute to the albuminuria–CVD link. In addition, other unmeasured clinical or lifestyle factors may also have influenced the inflammatory and vascular profiles observed. Finally, although we excluded patients with acute illness or recent nephrotoxic drug exposure, some residual short-term variability in UACR is possible and may have attenuated certain associations. Moreover, as the study was conducted before the introduction of SGLT2 inhibitors and finerenone, our findings reflect a pre–SGLT2/finerenone therapeutic era. Despite these limitations, our study has notable strengths. To our knowledge, this is one of the first studies to integrate circulating cytokine profiling with circulating leukocytetranscriptomic data in DKD patients to elucidate links with atherosclerosis. All patients were on uniform background therapies (RAS blockade and statins), which may reduce confounding by treatment differences. We rigorously adjusted for key clinical covariates and employed bootstrapped mediation analysis to better quantify indirect effects. The consistency of findings across serum and cellular markers (e.g. serum and leukocyte mRNA IL-6 levels) strengthens the biological plausibility of our conclusions.

In conclusion, this study provides new insights into how albuminuria is associated with systemic inflammation and early vascular injury in patients with DKD. We demonstrate that elevated albuminuria is associated with a heightened inflammatory state – characterized by increased IL6, IL1β, and innate immune receptor expression – which in turn contributes to the development of SA. Our findings support the concept of the kidney as a source and target of inflammatory signaling in DKD, potentially contributing to systemic inflammatory activation. Identifying IL6 and TLR2 as key mediators suggests potential therapeutic avenues. For example, interventions that reduce albuminuria (such as SGLT2 inhibitors or optimized RAS blockade) or that target inflammation (e.g. IL1β or IL6 inhibitors, or novel TLR antagonists) might attenuate the propagating cycle of renal and vascular injury (32). Indeed, broad anti-inflammatory approaches like low-dose colchicine have recently shown efficacy in reducing cardiovascular events in non-CKD populations (32), highlighting inflammation as a tractable target. Future studies should explore whether anti-inflammatory therapies can specifically improve outcomes in patients with DKD and whether biomarkers such as IL6 or TLR2 could help stratify cardiovascular risk in CKD. As the prevalence of diabetes and CKD continues to rise, a deeper understanding of the kidney–heart connection will be essential for developing strategies to prevent cardiovascular complications in this high-risk group.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.