Abstract
Introduction
This study focuses on investigating the relationship between serum neurofilament light chain (sNfL) and urinary albumin-to-creatinine ratio (uACR) among American adults aged 25–75.
Methods
An analysis was conducted on information gathered from 1741 individuals aged between 25 and 75 who participated in the National Health and Nutrition Examination Survey (NHANES) during the years 2013–2014. Generalized linear models were utilized, and restricted cubic spline (RCS) analysis was conducted to assess a non-linear relationship.
Results
Upon adjusting for multiple variables, a non-linear inverse J-shaped relationship was observed between sNfL and uACR. Compared with individuals in quartile 1 (Q1) of sNfL (2.8–8.3), those with quartile 4 (Q4) (≥19.1) had an adjusted β for uACR of 51.57.
Conclusions
The study found a J-shaped curve linking sNfL and uACR in American adults, with a turning point around log(sNfL) 2.928 pg/mL.
Keywords: Serum neurofilament light chain, albumin-to-creatinine ratio, J-shaped curve, NHANES
1. Introduction
Neurofilaments, unique to neurons, are type IV intermediate filaments that are heteropolymers [1]. They consist of three different chains: light, medium, and heavy. These proteins, which form the main elements of the neural cytoskeleton, are emitted into the surrounding milieu after neuro-axonal injuries. This release makes them promising biomarkers for various neurological conditions [1–3]. The neurofilament light chain (NfL), known for its high solubility, emerges as a significant biomarker. Its levels rise in blood and cerebrospinal fluid (CSF) during various neurological disorders that are characterized by neuro-axonal degeneration. Included in these conditions are multiple sclerosis, Alzheimer’s, Parkinson’s, and Huntington’s diseases, motor neuron disorders, cerebrovascular accidents, hereditary peripheral neuropathies, and injuries related to brain trauma [4–9]. It is important to note that in healthy individuals, levels of serum neurofilament light chain (sNfL) tend to rise naturally as they age, however, distinct reference ranges based on age are still not well-defined [10]. Recent research has revealed a notable correlation between the levels of NfL in CSF and blood, sparking increased interest in using NfL as a biomarker in various neurological conditions.
The use of the random urinary albumin-to-creatinine ratio (uACR) has become a common approach for evaluating and identifying albuminuria, offering significant benefits in terms of efficiency and ease [11]. Often described as increased urinary albumin excretion (UAE), albuminuria is generally identified by a uACR exceeding 30 mg/g [12–14]. It is often the sole marker in early-stage chronic kidney disease (CKD) patients who have a normal estimated glomerular filtration rate (eGFR), and is linked to negative health outcomes, which emphasizes the need for heightened clinical focus on albuminuria [15]. Therefore, it is essential to give significant clinical attention to the presence of albumin in urine.
Recent studies have established a correlation between plasma NfL levels and variables such as body mass index (BMI) and blood volume [16]. However, the connection between NfL levels and renal function remains a subject of debate [17,18]. A cross-sectional analysis of two distinct samples revealed a direct correlation between blood NfL and serum creatinine levels, lending evidence to the possibility that renal function partly influences blood NfL levels in older adults in Japan [19]. Additionally, conditions of renal dysfunction have been found to raise plasma NfL and total tau (t-tau) levels [20,21]. A previous study demonstrated that higher levels of phosphorylated tau at threonine 181 (p-tau181), NfL, t-tau, and ubiquitin carboxy-terminal hydrolase L1 (UCHL1) were present in individuals with cirrhosis than were in control subjects. In cases of cirrhosis, increased levels of p-tau181, NfL, and t-tau were associated with renal dysfunction and low serum albumin, indicating that liver and kidney functions play crucial roles in the interpretation of plasma neuropathological biomarkers [22].
The connection between sNfL and uACR in the adult U.S. population has apparently not been studied before. In the present cross-sectional study, our goal was to examine the relationship between sNfL and uACR, drawing on data from the National Health and Nutrition Examination Survey (NHANES).
2. Materials and methods
2.1. Study sample
Created by the National Center for Health Statistics, the NHANES utilizes a stratified, multistage probability clustering approach to select a representative sample, each year, from the noninstitutionalized U.S. population. The fundamental goal of the NHANES is to assess the health and nutritional status of Americans. The survey selects participants through a stratified, multi-tiered sampling technique. The NHANES gathers a broad range of health-related information, including demographic details, physical examination outcomes, laboratory results, and dietary patterns. Participants give written consent before their inclusion in the NHANES. The data, accessible on the NHANES website (http://www.cdc.gov/nchs/nhanes.htm), are public. The Tianjin Union Medical Center’s review board deemed this study exempt as it utilizes publicly accessible, anonymized data, negating the need for informed consent. This research complies with the guidelines established by the Strengthening the Reporting of Observational Studies in Epidemiology initiative.
Of the initial 2071 participants, aged between 25 and 75 years, with available sNfL data from the 2013–2014 dataset, 20 were excluded due to missing uACR data. An additional 310 were excluded for incomplete covariate information. As a result, the final analysis was performed on 1741 subjects, as illustrated in Figure 1. Each of these voluntarily enrolled in the study and gave their written informed consent. Approval for the study was granted by the Institutional Review Board at the National Center for Health Statistics.
Figure 1.
Flow diagram of the screening and enrollment of study participants.
2.2. Measurement of serum neurofilament light chain
In the NHANES 2013–2014 dataset, the measurement of sNfL levels was performed using the Siemens Healthineers’ Atellica Immunoassay System, which accomplished a detection rate of 98.4%. Detailed information about this analytical method can be found on the NHANES website, and additional elaboration is provided in the supplementary materials of our study.
2.3. Measurement of urinary albumin-to-creatinine ratio
The estimated eGFR was calculated using the formula developed by the CKD Epidemiology Collaboration [23]. CKD is diagnosed when the eGFR drops below 60 mL/min/1.73 m2. For NHANES participants, urine samples were collected initially in Mobile Examination Center. The measurements of urinary albumin and creatinine were conducted using a single-spot urine sample via a solid-phase fluorescence immunoassay and a modified Jaffe kinetic method. The uACR was determined by dividing the concentration of urinary albumin (mg) by that of urinary creatinine (g). Definitions for microalbuminuria and macroalbuminuria correspond to spot-urine uACR values between 30 and 300 mg/g and those ≥300 mg/g, respectively [24]. Albuminuria is categorized by a uACR >30 mg/g [25,26].
2.4. Covariates
In the present study, we examined various potential covariates based on previous research [27], which included age, sex, marital status, ethnicity, education level, family income, smoking and drinking habits, BMI, hypertension, diabetes, and a history of cardiovascular disease (CVD). Participants self-reported demographic information, whereas medical history, blood pressure, glucose levels, and physical measurements were collected by trained technicians. Smoking and drinking status were determined by questionnaire. The definition of smoking status included never smoker, current smoker, or former smoker; alcohol-drinking status was categorized as never, former, or current. CVD history encompassed conditions such as coronary heart disease, congestive heart failure, heart attack, stroke, or angina.
2.5. Statistical analysis
Characteristics of the participants are depicted as means with 95% confidence intervals (CIs) for variables that are continuous, and as proportions with 95%CIs for those that are categorical. The t-test was used for comparing continuous variables; the Chi-square (χ2) test was used for categorical variables.
Differences between groups were assessed using one-way ANOVA for variables with normal distribution, the Kruskal–Wallis test for those with non-normal distribution, and the χ2 test for categorical data. In linear-trend tests, categorical variables were treated as continuous.
We constructed a multivariate-adjusted, restricted cubic spline (RCS) model with three knots, to explore the potential non-linear dose-response relationship between the log(sNfL) and log(uACR).
The association threshold between log(sNfL) and log(uACR) was analyzed using a two-piecewise linear-regression model with a smoothing function, adjusting for variables in model 3. We identified inflection points using the likelihood-ratio test and bootstrap resampling.
A sensitivity analysis was conducted for enhanced result reliability, treating sNfL as a categorical variable in linear regression models and trend tests.
Furthermore, potential modifications of the relationship between sNfL and uACR were assessed, including the following variables: sex, age (<60 vs. ≥60 years), ethnicity (Hispanic vs. non-Hispanic), BMI (<30 vs. ≥30 kg/m2), hypertension, diabetes, and CVD. Heterogeneity among subgroups was assessed by multivariate linear regression, and interactions between subgroups and sNfL were examined by likelihood ratio testing.
As the study was based on available data, no preliminary statistical power calculation was done. Analyses were performed using R software (version 4.2.1; R Foundation for Statistical Computing, R Foundation for Statistical Computing, Vienna, Austria, http://www.Rproject.org), and Free Statistics software (version 1.9.1; Beijing Free Clinical Medical Technology Co., Ltd., Beijing, China). All tests were two-tailed, with a p ≤ .05 indicating statistical significance.
3. Results
3.1. Baseline features
Table 1 contains the baseline characteristics of the subjects included in the study. From the NHANES 2013–2014 dataset, 1741 participants were incorporated into our analysis. Of these, 846 participants (48.60%) were male, with an average age of 46.9 (15.4) years. The M (IQR) of sNfL was 12.3 (8.3–19.0) pg/mL. The prevalence of albuminuria was 11.1% (microalbuminuria, 9.3%; macroalbuminuria, 1.8%).
Table 1.
Baseline characteristics of the study population according to serum neurofilament light chain quartiles.
| Serum neurofilament light chain quartiles, pg/mL |
||||||
|---|---|---|---|---|---|---|
| Variables | Total | 2.8–8.3 | 8.4–12.3 | 12.4–19.0 | ≥19.1 | p |
| No. of participants | 1741 | 435 | 422 | 448 | 436 | |
| Age, mean (SD), years | 46.9 (15.4) | 35.1 (10.6) | 44.3 (13.5) | 51.6 (14.2) | 56.5 (13.9) | <.001 |
| Sex, n (%) | .003 | |||||
| Male | 846 (48.6) | 179 (41.1) | 216 (51.2) | 220 (49.1) | 231 (53) | |
| Female | 895 (51.4) | 256 (58.9) | 206 (48.8) | 228 (50.9) | 205 (47) | |
| Ethnicity, n (%) | <.001 | |||||
| Non-Hispanic White | 803 (46.1) | 174 (40) | 179 (42.4) | 222 (49.6) | 228 (52.3) | |
| Non-Hispanic Black | 305 (17.5) | 79 (18.2) | 86 (20.4) | 60 (13.4) | 80 (18.3) | |
| Mexican American | 241 (13.8) | 82 (18.9) | 60 (14.2) | 44 (9.8) | 55 (12.6) | |
| Other Hispanic | 155 (8.9) | 35 (8) | 44 (10.4) | 50 (11.2) | 26 (6) | |
| Others | 237 (13.6) | 65 (14.9) | 53 (12.6) | 72 (16.1) | 47 (10.8) | |
| Marital status, n (%) | <.001 | |||||
| Married | 940 (54.0) | 220 (50.6) | 231 (54.7) | 251 (56) | 238 (54.6) | |
| Never married | 335 (19.2) | 129 (29.7) | 87 (20.6) | 71 (15.8) | 48 (11) | |
| Living with partner | 133 (7.6) | 44 (10.1) | 38 (9) | 19 (4.2) | 32 (7.3) | |
| Other | 333 (19.1) | 42 (9.7) | 66 (15.6) | 107 (23.9) | 118 (27.1) | |
| PIR, mean (SD) | 2.5 (1.7) | 2.4 (1.6) | 2.6 (1.7) | 2.6 (1.7) | 2.5 (1.6) | .285 |
| Education, n (%) | .343 | |||||
| <9 | 348 (20.0) | 85 (19.5) | 83 (19.7) | 80 (17.9) | 100 (22.9) | |
| 9–12 | 367 (21.1) | 87 (20) | 92 (21.8) | 89 (19.9) | 99 (22.7) | |
| >12 | 1026 (58.9) | 263 (60.5) | 247 (58.5) | 279 (62.3) | 237 (54.4) | |
| BMI (kg/m2), mean (SD) | 29.3 (7.4) | 29.8 (7.6) | 29.0 (7.3) | 28.7 (6.9) | 29.9 (7.7) | .022 |
| Smoke, n (%) | <.001 | |||||
| Never | 969 (55.7) | 279 (64.1) | 248 (58.8) | 233 (52) | 209 (47.9) | |
| Former | 397 (22.8) | 65 (14.9) | 91 (21.6) | 121 (27) | 120 (27.5) | |
| Now | 375 (21.5) | 91 (20.9) | 83 (19.7) | 94 (21) | 107 (24.5) | |
| Drink, n (%) | <.001 | |||||
| Never | 213 (12.2) | 62 (14.3) | 37 (8.8) | 65 (14.5) | 49 (11.2) | |
| Former | 272 (15.6) | 39 (9) | 53 (12.6) | 81 (18.1) | 99 (22.7) | |
| Now | 1256 (72.1) | 334 (76.8) | 332 (78.7) | 302 (67.4) | 288 (66.1) | |
| Hypertension, n (%) | <.001 | |||||
| No | 1040 (59.7) | 346 (79.5) | 273 (64.7) | 241 (53.8) | 180 (41.3) | |
| Yes | 701 (40.3) | 89 (20.5) | 149 (35.3) | 207 (46.2) | 256 (58.7) | |
| Diabetes, n (%) | <.001 | |||||
| No | 1430 (82.1) | 404 (92.9) | 374 (88.6) | 360 (80.4) | 292 (67) | |
| Yes | 311 (17.9) | 31 (7.1) | 48 (11.4) | 88 (19.6) | 144 (33) | |
| CVD, n (%) | <.001 | |||||
| No | 1596 (91.7) | 431 (99.1) | 403 (95.5) | 406 (90.6) | 356 (81.7) | |
| Yes | 145 (8.3) | 4 (0.9) | 19 (4.5) | 42 (9.4) | 80 (18.3) | |
| uACR, median (IQR) | 7.1 (4.7, 12.9) | 6.8 (4.4, 11.4) | 6.4 (4.4, 11.0) | 7.0 (4.9, 12.1) | 8.5 (5.1, 20.0) | <.001 |
| Urine albumin, n (%) | <.001 | |||||
| Normal (<30) | 1548 (88.9) | 392 (90.1) | 393 (93.1) | 411 (91.7) | 352 (80.7) | |
| Microalbuminuria (30–299) | 162 (9.3) | 42 (9.7) | 27 (6.4) | 31 (6.9) | 62 (14.2) | |
| Macroalbuminuria (≥300) | 31 (1.8) | 1 (0.2) | 2 (0.5) | 6 (1.3) | 22 (5) | |
| eGFR, median (IQR) | 94.7 (80.1, 108.5) | 108.2 (96.9, 117.9) | 97.4 (86.7, 109.4) | 90.2 (77.3, 101.4) | 83.7 (69.0, 96.7) | <.001 |
BMI: body mass index; PIR: poverty income ratio; CVD: coronary vascular disease; eGFR: estimated glomerular filtration rate; uACR: urinary albumin-to-creatinine ratio.
Data are presented as count for categorical variables, and as weighted means (standard deviation (SD)) for continuous variables. To calculate p values for trends in participant characteristics, linear regression was used for continuous variables, and logistic regression for categorical ones.
3.2. Association between sNfL and uACR
The univariate analysis demonstrated that age, PIR, BMI, habits of smoking and drinking, and the incidence of hypertension, diabetes, eGFR were associated with uACR (all p < .05), as depicted in Table 2.
Table 2.
Association of covariates and uACR.
| Variable | β (95%CI) | p (t-test) | p (F-test) |
|---|---|---|---|
| Age (years) | 1.23 (0.35, 2.12) | .006 | .006 |
| Sex, n (%) | |||
| Male | 1 (ref.) | ||
| Female | −23.55 (−50.81, 3.71) | .09 | .090 |
| Ethnicity, n (%) | |||
| Non-Hispanic White | 1 (ref.) | .052 | |
| Non-Hispanic Black | 46.56 (8.37, 84.76) | .017 | |
| Mexican American | 37.35 (−4.36, 79.06) | .079 | |
| Other Hispanic | 41.4 (−8.42, 91.22) | .103 | |
| Others | 42.67 (0.69, 84.65) | .046 | |
| Marital status, n (%) | |||
| Married | 1 (ref.) | .052 | |
| Never married | −18.4 (−54.54, 17.74) | .318 | |
| Living with partner | −27.95 (−80.57, 24.67) | .298 | |
| Other | 36.08 (−0.14, 72.31) | .051 | |
| PIR, n(%) | −10.06 (−18.24, −1.88) | .016 | .016 |
| Education, n (%) | |||
| <9 | 1 (ref.) | .482 | |
| 9–12 | −9.8 (−52.36, 32.76) | .652 | |
| >12 | −20.86 (−56.15, 14.43) | .246 | |
| BMI, n (%) | 2.43 (0.58, 4.27) | .01 | .01 |
| Smoke, n (%) | |||
| Never | 1 (ref.) | .024 | |
| Former | 43.43 (9.59, 77.27) | .012 | |
| Now | −5.71 (−40.25, 28.82) | .746 | |
| Drink, n (%) | |||
| Never | 1 (ref.) | .005 | |
| Former | 13.09 (−38.82, 65) | .621 | |
| Now | −42.81 (−84.85, −0.77) | .046 | |
| Hypertension, n (%) | |||
| No | 1 (ref.) | ||
| Yes | 74.5 (46.92, 102.08) | <.001 | <.001 |
| Diabetes, n (%) | |||
| No | 1 (ref.) | ||
| Yes | 138.14 (103.14, 173.14) | <.001 | <.001 |
| CVD, n (%) | |||
| No | 1 (ref.) | ||
| Yes | 40.79 (−8.52, 90.1) | .105 | .105 |
| eGFR (mL/min/1.73 m2) | −3.23 (−3.89, −2.57) | <.001 | <.001 |
| sNfL (pg/mL) | 4.47 (3.83, 5.12) | <.001 | <.001 |
In the multivariable linear regression analysis, a higher level of sNfL was linked to an increase in uACR (β, 3.90; 95%CI, 3.24–4.57; p < .001). This association remained after adjusting for factors such as age, sex, marital status, ethnicity, education level, PIR, BMI, smoking and drinking habits, hypertension, diabetes, CVD, and eGFR.
When sNfL levels were categorized into quartiles, the association persisted. Compared with individuals in quartile 1 (Q1) of sNfL (2.8–8.3), those in quartile 4 (Q4) (≥19.1) had an adjusted β for uACR of 51.57 (95%CI, 7.76–95.38; p = .021). As shown in Table 3.
Table 3.
Association between sNfL and uACR.
| Non-adjusted model |
Model 1 |
Model 2 |
Model 3 |
||||||
|---|---|---|---|---|---|---|---|---|---|
| sNfL, pg/mL | N = 1741 | β (95%CI) | p | β (95%CI) | p | β (95%CI) | p | β (95%CI) | p |
| Continuous | 4.47 (3.83–5.12) | <.001 | 4.59 (3.92–5.26) | <.001 | 4.55 (3.88–5.22) | <.001 | 3.9 (3.24–4.57) | <.001 | |
| Categorical | |||||||||
| Q1 (2.8–8.3) | 435 | 0 (ref.) | 0 (ref.) | 0 (ref.) | 0 (ref.) | ||||
| Q2 (8.4–12.3) | 422 | 0.17 (−38.28 to 38.62) | .993 | −2.0 (−41.7 to 37.69) | .921 | 2.58 (−37.09 to 42.26) | .898 | −7.28 (−45.51 to 30.96) | .709 |
| Q3 (12.4–19.0) | 448 | 9.95 (−27.93 to 47.83) | .607 | 7.04 (−34.88 to 48.97) | .742 | 9.93 (−32 to 51.87) | .643 | −17.51 (−58.15 to 23.12) | .398 |
| Q4 (≥19.1) | 436 | 100.94 (62.8–139.07) | <.001 | 98.83 (54.27–143.39) | <.001 | 101.71 (57.19–146.23) | <.001 | 51.57 (7.76–95.38) | .021 |
| p for trend | 1741 | <.001 | <.001 | <.001 | .037 | ||||
95%CI: 95% confidence interval.
Model 1 adjusted for age, sex, education level, ethnicity, marital status, and PIR. Model 2 further adjusted for BMI, smoking habits, and drinking habits plus model 1. Model 3 further adjusted for hypertension, diabetes, CVD, and eGFR plus model 2.
The relationship between log(sNfL) and log(uACR) displayed a J-shaped curve, indicating a nonlinear correlation (p for non-linearity < .001), as depicted in the RCSs analysis shown in Figure 2. In this figure, RCSs were utilized to flexibly illustrate the association between the predicted values of log(sNfL) and log(uACR). The log(uACR) remained relatively stable until reaching a predicted log(sNfL) value of approximately 2.928, beyond which it showed a sharp increase . The two-piecewise linear-regression model identified the inflection point at 2.928 pg/mL, based on the nonlinear analysis, as detailed in Table 4.
Figure 2.
The nonlinear relationship between log(sNfL) and log(uACR). The solid lines depict the projected values, and the dashed lines indicate the 95% confidence intervals. These values were calibrated considering factors such as age, sex, marital status, ethnicity, educational level, PIR, BMI, smoking and drinking habits, as well as the presence of hypertension, diabetes, CVD, and eGFR. Only 99.9% of the data is shown.
Table 4.
Association between log(sNfL) and log(uACR) using two-piecewise regression models.
| Log(sNfL), pg/mL | Adjusted model |
|
|---|---|---|
| β (95%CI) | p Value | |
| <2.928 | −0.04 (−0.17 to 0.09) | .537 |
| ≥2.928 | 0.60 (0.32–0.87) | <.001 |
| Log-likelihood ratio test | <.001 | |
CI: confidence interval.
Adjusted for sociodemographic (age, sex, ethnicity, marital status, PIR, and education level), BMI, smoking habits, drink habits, hypertension, diabetes, CVD, and eGFR. Only 99.9% of the data is displayed.
3.3. Stratified analyses based on additional variables
In several subgroups, stratified analysis was performed to assess potential effect modifications on the relationship between sNfL and uACR. No significant interactions in age, sex, ethnicity, BMI, and CVD. There are significant interactions in hypertension and diabetes. But the specific mechanism needs further research, as shown in supplementary materials (SFigure 1).
4. Discussion
To our understanding, this research is the first exploration of the association between sNfL and uACR in a comprehensive U.S. national cohort. Our study, a cross-sectional analysis of American adults, revealed a J-shaped correlation between sNfL and uACR, pinpointing a critical inflection point at nearly sNfL log-transformed 2.928 pg/mL. Sensitivity tests further confirmed the robustness of this relationship.
Previous studies have indicated a positive association between blood NfL levels and serum creatinine, indicating that renal function might influence blood NfL concentrations to some extent [19]. One other study revealed that among children suffering from CKD, the levels of NfL increased more rapidly as they age [28]. The relationship between NfL levels and renal function in adults with acquired kidney diseases continues to be a contentious subject. Some studies have suggested associations with aging and cardiovascular conditions, yet not consistently with reduced kidney function [29]. In older individuals suffering from atrial fibrillation, Polymeris et al. observed a stronger correlation between glomerular filtration rate and NfL levels compared to BMI, highlighting the glomerular filtration rate as a key factor influencing NfL levels [30].
Diverging from earlier research, our study utilized NHANES data and conducted adjustments for various potential confounders through multivariate regression analysis. This approach was designed to extend the applicability of our findings to the broader U.S. adult population. Our dose–response analysis disclosed a non-linear association between sNfL and uACR. Specifically, we observed that log(uACR) did not increase with increasing log(sNfL) in individuals whose log(sNfL) was below 2.928 pg/mL. In contrast, for those with log(sNfL) above this threshold, log(uACR) increased alongside an increase in log(sNfL).
The mechanistic relationship between sNfL and kidney function remains unclear. Several potential mechanisms have been proposed: (i) renal clearance of NfL: as a biomarker of neural damage, NFL might be cleared from the body via the kidneys. This suggests that in individuals with impaired kidney function, the clearance of NfL is reduced, leading to elevated levels of NfL in the blood. This hypothesis is bolstered by research demonstrating a positive relationship between blood NfL concentrations and serum creatinine levels, which serve as an indicator of renal function [19]. (ii) Shared pathological processes: the relationship between NfL and kidney function might also reflect underlying shared pathological mechanisms. For example, ailments that lead to neurodegeneration might also concurrently affect renal function [31]. This is implied by genetic and environmental influences on the association between plasma NFL and creatinine, suggesting that factors affecting both the nervous system and the kidneys might contribute to the observed correlation. (iii) Inflammatory processes: chronic inflammation is a common factor in many neurodegenerative diseases and may concurrently impact the nervous system and the kidneys. Inflammatory processes could lead to neural damage (reflected in increased NfL) and to impaired kidney function.
The present study is not without its limitations. First, owing to its cross-sectional design, it is not possible to determine the cause-and-effect relationship between sNfL and uACR. Future longitudinal cohort studies are essential to explore this temporal link. Second, despite the use of regression models, stratified analyses, and sensitivity testing, the possibility of residual confounding by unaccounted or unknown variables cannot be completely ruled out. Third, since our findings are based on a survey conducted among U.S. adults, extrapolating these results to other populations warrants additional research and validation.
5. Conclusions
The data of the present study revealed a J-shaped correlation between sNfL and uACR within the adult population of the U.S., with a significant inflection observed at approximately log(sNfL) 2.928 pg/mL. These findings highlight the significant link between sNfL and uACR, underscoring its importance for further research and clinical attention.
Supplementary Material
Acknowledgements
We acknowledge all of the participants and staff involved in NHANES for their valuable contributions.
Funding Statement
This work was supported by Tianjin Health Research Project (Grant No. TJWJ2022ZD005) and Tianjin Union Medical Center Research Project (Grant Nos. 2021YJ012 and 2021YJ013).
Author contributions
YTL analyzed and interpreted the data and results, and drafted the manuscript. MYZ revised the manuscript. JNL and QY proposed the concept and design of the study and revised manuscript for critical intellectual content. MYZ revised manuscript for critical intellectual content. YTL and QY contributed equally to this work. All authors had access to the data. All authors read and approved the final manuscript.
Ethics statement
The survey was performed by the National Center for Health Statistics (NCHS) and approved by the NCHS Institutional Review Board (IRB). All informed consents had been obtained from the eligible subjects before initiating data collection and NHANES health examinations. All authors confirmed that all methods were carried out in accordance with relevant NHANES Analytic Guidelines.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
These survey data are free and publicly available, and can be downloaded directly from the NHANES website (http://www.cdc.gov/nchs/nhanes.htm) by users and researchers worldwide (accessed on 1 November 2023).
References
- 1.Petzold A. Neurofilament phosphoforms: surrogate markers for axonal injury, degeneration and loss. J Neurol Sci. 2005;233(1–2):183–198. doi: 10.1016/j.jns.2005.03.015. [DOI] [PubMed] [Google Scholar]
- 2.Bacioglu M, Maia LF, Preische O, et al. Neurofilament light chain in blood and CSF as marker of disease progression in mouse models and in neurodegenerative diseases. Neuron. 2016;91(1):56–66. doi: 10.1016/j.neuron.2016.05.018. [DOI] [PubMed] [Google Scholar]
- 3.Lépinoux-Chambaud C, Eyer J.. Review on intermediate filaments of the nervous system and their pathological alterations. Histochem Cell Biol. 2013;140(1):13–22. doi: 10.1007/s00418-013-1101-1. [DOI] [PubMed] [Google Scholar]
- 4.Gunnarsson M, Malmeström C, Axelsson M, et al. Axonal damage in relapsing multiple sclerosis is markedly reduced by natalizumab. Ann Neurol. 2011;69(1):83–89. doi: 10.1002/ana.22247. [DOI] [PubMed] [Google Scholar]
- 5.Disanto G, Adiutori R, Dobson R, et al. Serum neurofilament light chain levels are increased in patients with a clinically isolated syndrome. J Neurol Neurosurg Psychiatry. 2016;87(2):126–129. doi: 10.1136/jnnp-2014-309690. [DOI] [PubMed] [Google Scholar]
- 6.Weydt P, Oeckl P, Huss A, et al. Neurofilament levels as biomarkers in asymptomatic and symptomatic familial amyotrophic lateral sclerosis. Ann Neurol. 2016;79(1):152–158. doi: 10.1002/ana.24552. [DOI] [PubMed] [Google Scholar]
- 7.Byrne LM, Rodrigues FB, Blennow K, et al. Neurofilament light protein in blood as a potential biomarker of neurodegeneration in Huntington’s disease: a retrospective cohort analysis. Lancet Neurol. 2017;16(8):601–609. doi: 10.1016/S1474-4422(17)30124-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.De Marchis GM, Katan M, Barro C, et al. Serum neurofilament light chain in patients with acute cerebrovascular events. Eur J Neurol. 2018;25(3):562–568. doi: 10.1111/ene.13554. [DOI] [PubMed] [Google Scholar]
- 9.Olsson B, Portelius E, Cullen NC, et al. Association of cerebrospinal fluid neurofilament light protein levels with cognition in patients with dementia, motor neuron disease, and movement disorders. JAMA Neurol. 2019;76(3):318–325. doi: 10.1001/jamaneurol.2018.3746. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Disanto G, Barro C, Benkert P, et al. Serum neurofilament light: a biomarker of neuronal damage in multiple sclerosis. Ann Neurol. 2017;81(6):857–870. doi: 10.1002/ana.24954. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Miller WG, Bruns DE, Hortin GL, et al. Current issues in measurement and reporting of urinary albumin excretion. Clin Chem. 2009;55(1):24–38. doi: 10.1373/clinchem.2008.106567. [DOI] [PubMed] [Google Scholar]
- 12.Matsushita K, van der Velde M, Astor BC, et al. Association of estimated glomerular filtration rate and albuminuria with all-cause and cardiovascular mortality in general population cohorts: a collaborative meta-analysis. Lancet. 2010;375(9731):2073–2081. doi: 10.1016/S0140-6736(10)60674-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Tonelli M, Muntner P, Lloyd A, et al. Using proteinuria and estimated glomerular filtration rate to classify risk in patients with chronic kidney disease: a cohort study. Ann Intern Med. 2011;154(1):12–21. doi: 10.7326/0003-4819-154-1-201101040-00003. [DOI] [PubMed] [Google Scholar]
- 14.Astor BC, Matsushita K, Gansevoort RT, et al. Lower estimated glomerular filtration rate and higher albuminuria are associated with mortality and end-stage renal disease. A collaborative meta-analysis of kidney disease population cohorts. Kidney Int. 2011;79(12):1331–1340. doi: 10.1038/ki.2010.550. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Fernández-Fernández B, Valiño-Rivas L, Sánchez-Niño MD, et al. Albuminuria downregulation of the anti-aging factor klotho: the missing link potentially explaining the association of pathological albuminuria with premature death. Adv Ther. 2020;37(Suppl. 2):62–72. doi: 10.1007/s12325-019-01180-5. [DOI] [PubMed] [Google Scholar]
- 16.Manouchehrinia A, Piehl F, Hillert J, et al. Confounding effect of blood volume and body mass index on blood neurofilament light chain levels. Ann Clin Transl Neurol. 2020;7(1):139–143. doi: 10.1002/acn3.50972. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Korley FK, Goldstick J, Mastali M, et al. Serum NfL (neurofilament light chain) levels and incident stroke in adults with diabetes mellitus. Stroke. 2019;50(7):1669–1675. doi: 10.1161/STROKEAHA.119.024941. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Hermansson L, Yilmaz A, Price RW, et al. Plasma concentration of neurofilament light chain protein decreases after switching from tenofovir disoproxil fumarate to tenofovir alafenamide fumarate. PLOS One. 2019;14(12):e226276. doi: 10.1371/journal.pone.0226276. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Akamine S, Marutani N, Kanayama D, et al. Renal function is associated with blood neurofilament light chain level in older adults. Sci Rep. 2020;10(1):20350. doi: 10.1038/s41598-020-76990-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Benkert P, Meier S, Schaedelin S, et al. Serum neurofilament light chain for individual prognostication of disease activity in people with multiple sclerosis: a retrospective modelling and validation study. Lancet Neurol. 2022;21(3):246–257. doi: 10.1016/S1474-4422(22)00009-6. [DOI] [PubMed] [Google Scholar]
- 21.Syrjanen JA, Campbell MR, Algeciras-Schimnich A, et al. Associations of amyloid and neurodegeneration plasma biomarkers with comorbidities. Alzheimers Dement. 2022;18(6):1128–1140. doi: 10.1002/alz.12466. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Berry K, Asken BM, Grab JD, et al. Hepatic and renal function impact concentrations of plasma biomarkers of neuropathology. Alzheimers Dement. 2022;14(1):e12321. doi: 10.1002/dad2.12321. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604–612. doi: 10.7326/0003-4819-150-9-200905050-00006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Levin A, Stevens PE.. Summary of KDIGO 2012 CKD Guideline: behind the scenes, need for guidance, and a framework for moving forward. Kidney Int. 2014;85(1):49–61. doi: 10.1038/ki.2013.444. [DOI] [PubMed] [Google Scholar]
- 25.Cho Y-T, Chen C-W, Chen M-P, et al. Diagnosis of albuminuria by tryptic digestion and matrix-assisted laser desorption ionization/time-of-flight mass spectrometry. Clin Chim Acta. 2013;420:76–81. doi: 10.1016/j.cca.2012.12.016. [DOI] [PubMed] [Google Scholar]
- 26.Standards of Medical Care in Diabetes-2016: summary of revisions. Diabetes Care. 2016;39 (Suppl. 1):S4–S5. doi: 10.2337/dc16-S003. [DOI] [PubMed] [Google Scholar]
- 27.Ciardullo S, Muraca E, Bianconi E, et al. Diabetes mellitus is associated with higher serum neurofilament light chain levels in the general US population. J Clin Endocrinol Metab. 2023;108(2):361–367. doi: 10.1210/clinem/dgac580. [DOI] [PubMed] [Google Scholar]
- 28.van der Plas E, Lullmann O, Hopkins L, et al. Associations between neurofilament light-chain protein, brain structure, and chronic kidney disease. Pediatr Res. 2022;91(7):1735–1740. doi: 10.1038/s41390-021-01649-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Barro C, Chitnis T, Weiner HL.. Blood neurofilament light: a critical review of its application to neurologic disease. Ann Clin Transl Neurol. 2020;7(12):2508–2523. doi: 10.1002/acn3.51234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Polymeris AA, Helfenstein F, Benkert P, et al. Renal function and body mass index contribute to serum neurofilament light chain levels in elderly patients with atrial fibrillation. Front Neurosci. 2022;16:819010. doi: 10.3389/fnins.2022.819010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Tang R, Panizzon MS, Elman JA, et al. Association of neurofilament light chain with renal function: mechanisms and clinical implications. Alzheimers Res Ther. 2022;14(1):189. doi: 10.1186/s13195-022-01134-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
These survey data are free and publicly available, and can be downloaded directly from the NHANES website (http://www.cdc.gov/nchs/nhanes.htm) by users and researchers worldwide (accessed on 1 November 2023).