Publications & Posters

Association of Chronic Kidney Disease With Plasma NfL and Other Biomarkers of Neurodegeneration: The H70 Birth Cohort Study in Gothenburg

Neurology | May 24, 2023

Dittrich A, Ashton NJ, Zetterberg H, Blennow K, Zettergren A, Simrén J, Skillbäck T, Shams S, Machado A, Westman E, Schöll M, Skoog I, Kern S. 

Neurology. 2023

This study was performed using Simoa Homebrew assay(s).


Background and objectives: Studies associate chronic kidney disease (CKD) with neurodegeneration. This study investigated the relation between kidney function, blood, cerebrospinal fluid (CSF), and structural brain MRI markers of neurodegeneration, in a sample including individuals with and without CKD.

Methods: Participants from the Gothenburg H70 Birth Cohort Study, with data on plasma-neurofilament light (P-NfL), estimated glomerular filtration rate (eGFR) and structural brain MRI were included. Participants were invited to also have CSF collected. The primary endpoint of the present study was to determine any association between CKD and P-NfL. Secondary endpoints included cross-sectional associations between CKD, eGFR and cerebrospinal fluid (CSF)- and MRI-derived markers of neurodegeneration and Alzheimer’s disease (AD) pathology (MRI: cortical thickness, hippocampal volume, lateral ventricle volume, white matter lesion volume; CSF: β-amyloid (Aβ) 42, Aβ42/40, Aβ42/p-tau, t-tau, p-tau, NfL). Participants with P-NfL and eGFR at baseline were re-examined on eGFR, 5.5 (5.3; 6.1) years (median; IQR) after the first visit, and the predictive value of P-NfL levels on incident CKD was estimated longitudinally, using a Cox proportional hazards model.

Results: We included 744 participants, 668 without CKD (Age 71 (70; 71) years, 50% males) and 76 with CKD (age 71 (70;71) years, 39% males). Biomarkers from cerebrospinal fluid (CSF) were analysed in 313 participants. 558 individuals returned for a re-examination of eGFR (75% response rate, age 76 (76; 77), 48% males, 76 new cases of CKD). Participants with CKD had higher P-NfL levels than those with normal kidney function (median; 18.8 versus 14.0 pg/mL, p<0.001), while MRI and CSF markers were similar between the groups. P-NfL was independently associated with CKD after adjustment for confounding variables, including hypertension and diabetes (OR; 3.231, p<0.001), in a logistic regression model. eGFR, and CSF Aβ 42/40: R=0.23, p=0.004 correlated in participants with Aβ42 pathology. P-NfL levels in the highest quartile were associated with incident CKD at follow-up (HR; 2.08 (1.14: 4.50)).

Discussion: In a community-based cohort of 70-year olds, P-NfL was associated with both prevalent and incident CKD, while CSF and/or imaging measures did not differ by CKD status. Participants with CKD and dementia presented similar levels of P-NfL.