Parkinson’s disease (PD) is the second most common neurodegenerative disease affecting both dopaminergic-mediated motor and non-dopaminergic-mediated non-motor systems (Yoo et al., 2019). Cognitive impairment is one of the most prevalent non-motor symptoms especially in the late stages of PD, which involves impairments in specific domains such as visuospatial, attentional, executive, and memory functions (Weintraub et al., 2015). The cognitive decline will reduce treatment compliance, increase extra nursing time, and worsen the quality and prognosis of life for PD patients (Leroi et al., 2012). Moreover, previous research has suggested that PD dementia and Alzheimer’s disease (AD) may share parallel changes in both symptoms and protein profiles (Gomperts et al., 2013), indicating that AD-related proteins have significant roles in PD-related cognitive decline.

Chronic kidney disease (CKD) is characterized by impaired renal function, which is associated with accelerated cognitive decline (Helmer et al., 2011). The renal indices include serum creatinine (Scr), uric acid (UA), and blood urea nitrogen (BUN), among which estimated glomerular filtration rates (eGFR) based on Scr levels are the most widely used to assess renal function, with diagnostic thresholds of less than 60 mL/min/1.73m² for eGFR. Previous studies have reported that the cognitive-related pathological changes in the brain are exacerbated by reduced renal function in those with cognitive impairment (Lopez et al., 2018; Petersen et al., 2019). However, the pathophysiological mechanism linking cognition and kidney function in PD patients remains unclear.

The commonly used tools to detect cognitive impairment include neuropsychological tests, neuroimaging examinations, and fluid biomarkers. Blood biomarkers have the advantages of objectivity and accessibility compared to others and are more suitable for the early identification of cognitive impairment. Likewise, it lacks and urgently needs valid intervention for PD comorbid with cognitive decline. Therefore, this study aims to investigate whether there is an association between eGFR and cognitive function in PD patients from the open-source database Parkinson’s Progression Marker Initiative (PPMI). We hypothesize that the reduced eGFR, which reflects impaired kidney function, would correlate with cognitive decline in PD and can be used as a potential predictive and prognostic tool to evaluate PD progression in clinical practice.

This prospective study demonstrated significant associations between impaired kidney function and global cognitive impairment in patients with de novo PD independent of confounding variables, as well as specific cognitive domains such as episodic memory, language, processing speed, and attention. Remarkably, decreased eGFR was related to lower levels of CSF Aβ_1-42 and α-syn and higher levels of serum NfL at baseline. Meanwhile, decreased eGFR predicted a greater accumulation of CSF T-tau, P-tau, and serum NfL in up-to-seven-year follow-ups. These results suggested that eGFR might be a potential blood biomarker for predicting and monitoring cognitive changes in patients with de novo PD compared with other renal indicators.

The relationship of kidney function with subsequent cognitive decline in early PD patients remains unclear in previous studies although it has been studied that poor performances of kidney function are associated with an increased risk of cognitive impairment in the elderly population (Miyazawa et al., 2018; Fujiyoshi et al., 2020). These associations may be related to atherosclerosis, cerebral small vessel diseases, hematopoietic function impairment, or methylation metabolism that are secondary to CKD. Impaired kidney function is a very common condition in the general population, which can be easily monitored using blood tests. In this context, GFR, calculated by Scr, is useful in reflecting glomerular function and employed to manage CKD over the past two decades (Levey et al., 2007). There is a long-term process for eGFR affecting cognition. Previous studies found that reduced eGFR was linked to cognitive impairment in the population with normal renal function (Davey et al., 2013), and the one unit decrease of eGFR per year indicated a higher risk of cognitive decline (Barzilay et al., 2013). Additionally, another study showed that elevated Scr levels were correlated with the severity and duration of PD (Zhong et al., 2018), and patients with CKD had an increased risk of PD, suggesting a common pathophysiological process between CKD and PD (Nam et al., 2019). This might be because the decreased eGFR could lead to toxic accumulation, resulting in neurotoxic effects or neuroinflammatory reactions (Zhong et al., 2018). In the current study, we found that PD patients had worse cognitive performances; however, no obvious kidney impairment occurred in patients with de novo PD. The main reason might be attributed to the short disease duration of the included PD participants.

The exact mechanism underlying the association of kidney function with subsequent cognitive impairment in PD is unknown but some plausible explanations have been suggested. We observed lower levels of CSF Aβ_1-42, T-tau, and P-tau, as well as higher serum NfL in patients with PD compared with controls. Amyloid plaques and neurofibrillary tangles coexist in PD brains (Jellinger, 2008), and both amyloid and tau pathology can dynamically interact with the α-synuclein misfolding process according to previous studies (Irwin et al., 2017), which may explain our findings. Nevertheless, controversies still persist about the relationships between fluid biomarkers and PD (Parnetti et al., 2019). AD pathology may also be an important co-pathology and driver of adverse disease outcomes. To the best of our knowledge, this is the first time there is a report on the associations between eGFR and AD-related biomarkers in patients with PD. The results indicated that PD patients with worse kidney function had lower levels of CSF Aβ_1-42 and α-syn, as well as higher levels of serum NfL at baseline, while they had greater accumulation of CSF T-tau, P-tau, and serum NfL at 7-year follow-up. Some scholars argue that the correlation between AD pathology and renal function can be explained by tau protein being transported from the brain to the peripheral blood and cleared in the liver and kidney (Liu et al., 2015). Another study also proposes that the increased clearance rate of peripheral tau will reduce tau accumulation and neurodegeneration in the brain, which may be a potential treatment method for PD (Shi et al., 2016). Alternatively, an impaired renal function could accelerate the pathological process of Aβ and tau. CKD can cause oxidative stress and hyperhomocysteinemia via vascular injury and endothelial dysfunction (Etgen et al., 2012; An et al., 2017). Abnormal levels of serum homocysteine are involved in the pathology of cognitive impairment as they may enhance Aβ production and mediate its neurotoxicity (Vinothkumar et al., 2017). These factors may jointly contribute to the increased risk of cognitive impairment in PD. Furthermore, severe CKD and uremia may cause metabolic derangements in the basal ganglia, and inadequate dialysis has been reported to result in basal ganglia injury (Cupidi et al., 2006). Patients with CKD are particularly susceptible to aluminum and manganese toxicity (Ohtake et al., 2005). Therefore, PD-related exacerbation caused by CKD can aggravate cognitive impairment through disease progression. However, our findings did not show an association between PD and renal indicators, which did not support this hypothesis.

It was noted that the mechanisms of synergetic effect on neurodegenerative and vascular damage in the brain may contribute to cognitive decline in patients with poor kidney function (Guerville et al., 2021). According to neuroimaging results, cognitive impairment associated with lower eGFR is involved in several brain functional domains and is responsible for the damage of multiple cortical regions (especially the frontal lobe) and subcortical modulatory neurons, particularly adrenergic neurons in the mesencephalon and cholinergic neurons in the nucleus basalis of Meynert (Viggiano et al., 2020). Our analyses indicated that worse renal function was associated with poorer performances in both global and specific cognitive domains, which reflect the functional degeneration of different brain regions. Consistent with previous research, these observations suggest that renal impairment in PD is more closely related to multiple dysfunctions among cognitive domains, which mainly involved executive function, delayed recall, language, and attention (Kurella et al., 2005; Lee et al., 2015). In addition, our finding showed that PD patients with higher UA levels had a faster progression of executive function, which was similar to previous results (Vannorsdall et al., 2008; Latourte et al., 2018), whereas those patients with higher BUN levels were also predicted to have a larger decline in language and attention. The mechanism of this relationship with PD is still unclear, and further studies are needed to clarify whether kidney function affects cognitive impairment in patients with PD from the neuroimaging perspective.

The subgroup results indicate that there are significant associations between eGFR and cognitive impairment in older and male populations. We, therefore, propose that elderly people and men with low levels of eGFR have a higher risk of cognitive impairment and cognitive-related pathology accumulation in de novo PD and suggest that improving eGFR will provide greater benefits for those populations at high risk of cognitive impairment. Besides, the obvious relationships also emerge in APOE ε4 negative carriers, implying that the results can be further generalized to the community- and population-based studies.

The main strengths of this study were that both cross-sectional and prospective study designs were adopted, and the associations between eGFR and biomarkers of AD pathology were reported for the first time. However, it also had some limitations that need to be addressed. First, the study only observed a correlation between reduced eGFR and cognitive impairment but did not elucidate the underlying mechanism, which warrants further investigation. Second, the study population consisted mostly of middle-aged adults (50–70 years), whereas dementia and CKD are more prevalent in older age groups. Therefore, future studies should include older participants to examine the association more comprehensively. Third, the study relied on eGFR as the sole indicator of renal function, without considering other measures such as albuminuria and cystatin C, which were not available in our cohort. Additionally, lifestyle factors and cholesterol metabolites may also affect renal function and should be taken into account in future analyses.

To sum up, this study demonstrated that eGFR levels are reduced in de novo PD subjects with cognitive impairment. As well, decreased eGFR and UA/Scr predicted worsening cognitive function over time in PD patients. Notably, there were cross-sectional associations of eGFR with the levels of CSF Aβ_1–42, α-syn, and serum NfL, as well as longitudinal relationships of eGFR with the levels of CSF T-tau, P-tau, and serum NfL. These findings suggest that eGFR and UA/Scr ratio can act as potential diagnostic and prognostic biomarkers of cognitive impairment in de novo PD. It can assist clinicians and researchers in identifying patients with PD at risk of rapid cognitive decline and may have the potential for monitoring responses to therapy in future clinical practice.

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. The clinical data used in this study from the PPMI cohort are available at the PPMI website (https://www.ppmi-info.org/access-data-specimens/download-data/).

ZX and YQ designed and conceptualized the study. YQ, J-TL, KA, J-YL, Q-XQ, D-LW, and J-WZ conducted the study. YQ, Z-JM, Y-JX, Z-JM, and ZX analyzed and extracted the data. YQ, J-TL, KA, and ZX wrote the first draft of the manuscript. All authors contributed to the article and approved the submitted version.

This study was funded by the National Natural Science Foundation of Hubei Province (grant number 2020CFB590) and the Fundamental Research Funds for the Central Universities (YCJJ202201020). PPMI (a public-private partnership) is funded by the Michael J. Fox Foundation for Parkinson’s Research and multiple funding partners.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.