In this 15 year follow-up study of an unselected cohort of adults with type 1 diabetes, greater baseline CAN severity was independently associated with both accelerated decline in eGFR and a marked increase in UACR over time. These associations remained significant after adjustment for age, sex, diabetes duration, HbA1c, systolic BP, triglycerides and ACEi/ARB therapy.

Participants with CAN exhibited a faster rate of eGFR decline than those without CAN, alongside a substantially greater progression of albuminuria, whereas baseline renal function did not differ between two groups. Among participants with preserved renal function at baseline (G1A1), CAN was not associated with eGFR decline but remained independently associated with progression of UACR following full covariate adjustment. These findings are consistent with prior reports linking CAN to increased albuminuria and extend existing evidence by demonstrating that CAN is also associated with accelerated decline in kidney function over time [19, 22, 28,29,30].

A key strength of this study is the use of an unselected, population-based cohort of adults with type 1 diabetes, enhancing the representativeness and generalisability of our findings to the broader clinical population. At baseline, the prevalence of CAN was 41%, consistent with previously reported estimates in cohorts with type 1 diabetes [31]. Participants with established CAN were older, had a longer duration of diabetes and lower eGFR than those with no CAN or early CAN. No statistically significant differences were observed across groups in HbA1c, systolic or diastolic BP, retinopathy or hypertension prevalence, lipid profiles, or reno-protective medication use such as ACEi/ARB. Consistent with their more advanced clinical phenotype, participants with CAN were more frequently stratified into higher KDIGO CKD risk categories at study entry.

Over the 15 year follow-up, the presence and severity of CAN were significantly associated with an increased risk of adverse renal outcomes, including progression to severely increased albuminuria (UACR >30 mg/mmol), transition to a higher KDIGO risk category and ≥30% decline in eGFR. These associations reinforce CAN as a clinically meaningful marker of heightened renal risk. However, the associations observed may reflect shared underlying microvascular pathology common to multiple diabetes complications rather than a strictly causal relationship between CAN and kidney function decline.

Several biological mechanisms have been proposed to explain the link between CAN and DKD. Early parasympathetic dysfunction leading to relative sympathetic overactivity is thought to contribute to systolic hypertension, resting tachycardia and increased cardiac output, haemodynamic changes that promote glomerular membrane injury through excessive filtration of sodium and albumin [13, 16, 32, 33]. Impaired parasympathetic regulation may also lead to a loss of nocturnal BP dipping, resulting in higher nocturnal intraglomerular pressure and increased sodium and albumin excretion [11, 19,20,21, 34]. In addition, relative increase in sympathetic tone as a result of early parasympathetic denervation may also activate the intrarenal RAAS, leading to renal vasoconstriction, ischaemia and structural injury, including glomerular basement membrane thickening and mesangial matrix expansion, thereby accelerating DKD progression [9, 31].

Autonomic dysfunction has additionally been linked to increased inflammation and oxidative stress. Impairment of the sympathetic nervous system may disrupt the activity of inflammatory cytokines and chemokines [15, 35], several of which have been identified as potential therapeutic targets and biomarkers for DKD [15, 36]. Oxidative stress may be further amplified by impaired antioxidant defence mechanisms, including reduced activity of enzymatic scavengers and increased production of reactive oxygen species, contributing to progressive renal injury [34, 35]. In advanced stages of CAN, sympathetic denervation may exacerbate renal dysfunction by impairing sodium handling, reducing renal blood flow and promoting vasoconstriction, ultimately leading to a decline in eGFR [11, 12].

The absence of a nocturnal decline in both systolic and diastolic BP among individuals with CAN has been consistently reported in multiple studies using 24 h ambulatory BP monitoring [19,20,21, 37,38,39]. In contrast, evidence for other neuropathic mechanisms affecting renal vascular haemodynamic remains limited. Future research employing advanced imaging techniques, such as MRI, is needed to better characterise renal haemodynamic alterations in this context [40].

Our findings suggest that the influence of CAN on DKD progression may be dependent on when CAN develops in the course of kidney disease. Specifically, the presence of CAN prior to the onset of DKD appeared to have a minimal impact on subsequent eGFR decline. In contrast, CAN occurring in the setting of established or more advanced DKD may act as an accelerator of renal function decline. However, due to the relatively small number of participants with advanced DKD at baseline, we were unable to formally test this temporal interaction. Larger studies with sufficient representation across DKD stages are needed to determine whether CAN contributes causally to kidney disease progression and to clarify its role as a potential therapeutic target.

Evidence from the ADDITION study demonstrates that early identification and management of cardiometabolic risk factors in individuals with or at risk of cardiovascular disease [41] can significantly reduce all-cause mortality, emphasising the value of timely intervention. Within this context, our findings highlight autonomic dysfunction, specifically CAN, as a potential early marker and contributor to DKD progression. This reinforces the rationale for routine, systematic screening for CAN in clinical practice. Although the ADA currently recommends CAN screening, implementation remains inconsistent. Incorporating regular CAN assessment into diabetes care pathways could facilitate earlier identification of high-risk individuals, support more targeted risk-reduction strategies, and ultimately improve long-term renal outcomes.

Strengths of this study include the recruitment of participants with relatively preserved renal function (eGFR >30 ml/min per 1.73 m2), with most participants exhibiting normal renal parameters at baseline (eGFR >90 ml/min per 1.73 m2 and normoalbuminuria). This facilitated the examination of early predictors of renal decline over a 15 year follow-up period. Longitudinal data collection at 4 year intervals enabled robust analysis of renal endpoints, and the low attrition rate enhanced the validity of our findings. Crucially, all participants underwent standardised gold-standard assessments for CAN, ensuring consistency in the assessment of autonomic function across the cohort.

Limitations of this study include its single-centre design and the relatively small number of participants with lower baseline eGFR or established albuminuria, which limited subgroup analyses of moderate- and high-risk individuals. Additionally, blood samples were collected at 4 year intervals during the 15 year follow-up. While this allowed for longitudinal assessment, more frequent sampling, such as annual measurements, would have enabled a more precise determination of the timing and trajectory of changes in eGFR and UACR. While the sample size may limit detection of small effects, it was sufficient to identify clinically meaningful differences in renal outcomes and to explore associations across CAN severity stages. These findings provide a foundation for hypothesis generation and inform the design of future, larger-scale prospective studies. Another potential limitation of this study is the exclusion of mortality from the composite renal endpoint. Participants who had died before follow-up assessment were excluded, and cause-of-death data were not available. Consequently, some individuals with more advanced disease may have died from renal or competing non-renal causes before reaching recorded renal endpoints. This may introduce survivorship bias and could have led to underestimation of the true burden of progressive kidney disease.

Additionally, medications are a recognised cause of altered autonomic responses and may contribute to false-positive CART findings. Agents most commonly implicated include cardiovascular drugs (e.g. diuretics, nitrates, β-blockers and α-blockers) and non-cardiovascular drugs (e.g. antidepressants, antipsychotics and benzodiazepines) [42]. In this study, participants were instructed to withhold β-blockers on the day of testing; however, the potential influence of other medications cannot be entirely excluded. A further limitation is that individual CART components were not available for separate analysis; therefore, we were unable to determine whether specific parasympathetic or sympathetic abnormalities were differentially associated with renal outcomes. Future studies examining individual autonomic test measures may provide greater mechanistic insight into the relationship between autonomic dysfunction and DKD.

Despite its limitations, this study provides novel evidence supporting a potential role for CAN as an independent risk factor for renal decline in individuals with type 1 diabetes. To our knowledge, it is among the first to examine the relationship between CAN and KDIGO prognostic risk categories in the context of DKD. We demonstrated that greater CAN severity has significant association with higher KDIGO CKD risk and that CAN predicts renal function decline independently of traditional risk factors. These findings offer insight into the haemodynamic mechanisms by which autonomic dysfunction may contribute to DKD progression and support further investigation into CAN as a target for early intervention strategies to delay or prevent kidney disease.