Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Association between remnant cholesterol and verbal learning and memory function in the elderly in the US

Association between remnant cholesterol and verbal learning and memory function in the elderly in... Background: The relationship between remnant cholesterol (RC) and atherosclerotic cardiovascular risk has been given increasing attention in recent years. However, its association with verbal learning and memory performance has not been reported. Methods: Data were extracted from the National Health and Nutrition Examination Survey (NHANES) 2011–2014 database. Participants aged ≥60 years with available fasting lipid data were included. Verbal learning and memory performance were evaluated using the Consortium to Establish a Registry for Alzheimer’s Disease Word List Memory Task (CERAD‑ WL) subtest. The CERAD total score was calculated as the mean of three immediate recalls and a delayed recall. RC was calculated as total cholesterol ( TC) minus the sum of low‑ density lipoprotein cholesterol (LDL‑ C) and high‑ density lipoprotein cholesterol (HDL‑ C). Multivariate ordinal logistic regression was performed to evaluate the association between RC, as well as its derived marker, the TC/RC ratio, and age‑stratified quartiles of the CERAD total score. Results: A total of 1377 participants were analysed. On a continuous scale, per 1 mmol/L increase in RC and per 1 unit increase in the TC/RC ratio were associated with multivariable adjusted odds ratios (95% CI) of 0.74 (0.58–0.94) and 1.45 (1.13–1.87), respectively, for having a CERAD total score in a higher quartile. On a categorical scale, higher RC quartiles were associated with a CERAD total score in a lower quartile; in contrast, the higher TC/RC quartile was associated with a CERAD total score in a higher quartile (all P for trend < 0.05). Conclusions: The current study suggests that lower RC levels and a higher TC/RC ratio are associated with better verbal learning and memory function, which indicates that lowering RC levels could be beneficial for preventing cognitive impairment in elderly individuals. Further research is needed to validate the causal roles of RC and the TC/RC ratio in cognition. Keywords: Remnant cholesterol, Total cholesterol/remnant cholesterol, Cognition, Verbal learning and memory function, Elderly, Cross‑sectional study Introduction Dementia is a vital health problem in the global growing *Correspondence: [email protected]; [email protected] aging population. The estimated dementia population Cardio‑Metabolic Medicine Center, National Center for Cardiovascular will increase from 57.4 (50.4–65.1) million cases globally Diseases, Fuwai Hospital, Chinese Academy of Medical Sciences, Peking Union in 2019 to 152.8 (130.8–175.9) million cases in 2050 [1]. Medical College, Beijing, China 3 Given the lack of effective pharmacological therapies for Division of Cardiology, The First Affiliated Hospital of Soochow University, Suzhou, Jiangsu Province, China dementia thus far, the primary prevention of modifiable Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Xie et al. Lipids in Health and Disease (2022) 21:120 Page 2 of 10 risk factors has become crucial in addressing the rising nutritional status of civilian, noninstitutionalized resi- epidemic of cognitive impairment. dent populations in the United States. A complex, mul- Dyslipidemia has been found to be independently tistage sampling procedure was implemented to select associated with cognitive impairment. In observational eligible representative participants. The present study studies, higher serum TC levels in mid-life were associ- made use of the 2011–2012 and 2013–2014 cycles with ated with a higher risk of developing dementia in late life 19,931 individuals, of which participants aged ≥60 years [2], but the impact of serum TC levels in late life on the were eligible for cognitive function assessment and development of dementia remains unclear. A meta-anal- enrolled for analysis. Participants without fasting plasma ysis synthesizing 34 cohort studies found no association lipid data or complete cognitive assessment results were between serum TC levels and mild cognitive impair- excluded. Those without complete measurements of TC, ment, Alzheimer’s disease (AD), vascular dementia, any LDL-C, and HDL-C values for the calculation of RC were dementia, or cognitive decline [3]. Studies on serum also excluded. LDL-C levels and dementia or cognitive decline also dis- The NHANES protocols were approved by the National played differential results and seemed to vary by sex and Center for Health Statistics Ethics Review Board of the the presence of cardiovascular risk factors [4]. However, U.S. CDC. All participants received and completed writ- a Mendelian randomization study showed that geneti- ten informed consent during the survey. All material for cally low serum levels of LDL-C reduced the risk of AD analysis can be accessed on the NHANES official website [5]. Overall, inconsistent results have been found in the [11]. association between serum lipid profiles and cognitive impairment, limiting the clinical utility of lipid biomark- Measurement of blood lipids and the definition ers. In addition to further research on traditional blood of exposure variables lipid parameters, the exploration of new lipid parameters TC and triglycerides (TGs) were measured using an to predict cognitive function is urgently needed, espe- enzymatic assay method. HDL-C was measured using cially in the elderly population. the heparin-manganese precipitation method or a direct Remnant cholesterol (RC) represents the amount of immunoassay technique. LDL-C was calculated from cholesterol in the remnant lipoproteins transformed from measured values of TC, TGs, and HDL-C according to triglyceride-rich lipoproteins in the blood [6]. Recently, the Friedewald formula as follows: [LDL-C] = [TC] – RC has been shown to equally or even more precisely [HDL-C] – [TG/5]. The formula is valid for TG values predict cardio-cerebrovascular outcomes compared to less than or equal to 400  mg/dL. The exposure variables LDL-C or HDL-C [7–9]. Elevated RC levels are closely were remnant cholesterol (RC) and its derivative, the related to triglyceride metabolism disorders and insuf- TC/RC ratio. RC was calculated as TC minus the sum of ficient APOE-mediated remnant lipoprotein clearance LDL-C and HDL-C [RC = TC-(LDL-C + HDL-C)], and by the liver, such as type III hyperlipidemia [10]. Mean- the TC/RC ratio was calculated as TC divided by RC. while, APOE variants are closely related to the risk of AD. Therefore, RC has the biological possibility of affect - Assessment of verbal learning and memory function ing cognitive function. However, no study has reported and the definition of the outcome variable whether or how RC is associated with cognitive function. In NHANES 2011–2014, a series of evaluations of cog- In seeking a more plausible and reliable lipid marker of nitive functioning were conducted during the MEC sur- late-life cognitive impairment, the present study explored vey, including the word learning and recall modules from the relationship of RC with cognitive function utilizing the Consortium to Establish a Registry for Alzheimer’s data from a cohort of American elderly individuals in Disease (CERAD W-L). The CERAD W-L assesses the the National Health and Nutrition Examination Survey immediate and delayed learning ability for new verbal (NHANES) conducted between 2011 and 2014. Given information (memory subdomain), consisting of three the complexity of the metabolism of remnant lipopro- sequential learning trials and a delayed recall trial. For tein in human plasma, the study also explored a new lipid learning trials, participants were tasked with reading marker derived from RC, the total cholesterol to remnant 10 unrelated words displayed on a screen aloud, one at cholesterol (TC/RC) ratio, to further investigate the rela- a time, and were asked to recall as many words as pos- tionship of RC with cognitive function. sible immediately after their presentation. A delayed recall test was administered after the other two cogni- Methods tive tests were completed. The outcome variable was the Study design and population CERAD total score, which was calculated as the average The NHANES database is a cross-sectional program with of all three immediate recalls after each learning trial and a 2-year-cycle design that aims to assess the health and the delayed recall. Given the significant effect of age on Xie  et al. Lipids in Health and Disease (2022) 21:120 Page 3 of 10 cognitive function, the age- stratified (≥60 to 70, ≥70 survey design of the NHANES was taken into account to 80, and ≥80 years) quartiles of the CERAD total test by specifying primary sampling units (PSUs), strata, score were used [12, 13] to evaluate verbal learning and and sampling weights in the software’s svyset module. memory function. Specifically, the age-stratified quar - Sampling weights were constructed for the combined tiles of the CERAD total test score were calculated by 2011–2012 and 2013–2014 4-year cycles by dividing first dividing the studypopulation into 3 age groups (≥60 the individual sampling weights by 2, according to the to 70, ≥70 to 80, and ≥80 years) and then categorizing NHANES analytic tutorials [16]. participants into 4 grades of verbal learning and memory Baseline characteristics were grouped by age-strati- function (1 to 4) by the weighted quartiles of CERAD fied quartiles of the CERAD total test score. Data are total score separately in the 3 age groups. The lowest described as the mean (± standard deviation) for con- quartile indicated impaired verbal learning and memory tinuous variables, median (interquartile range) for function. The higher the quartile, the better the learning skewed variables, and sample counts (weighted per- and memory function. centage) for categorical variables. A weighted trend test was performed across age-stratified quartile groups for each continuous and categorical variable. Covariates In the statistical analysis of lipid-cognition asso- Potential confounding factors were investigated, includ- ciations, RC and the TC/RC ratio were treated as both ing age, sex (male and female), race, BMI status, educa- continuous and categorical variables. Continuous RC tional level, smoking status, and drinking status. Race and TC/RC variables were natural log-transformed to was defined as follows: 1) non-Hispanic white; 2) non- fit regression models due to their right-skewed distri - Hispanic black; and 3) other races, including Mexican bution [7]. Categorical RC and TC/RC variables with American, other Hispanic, non-Hispanic Asian and mul- four grades were created using quartiles as cutoffs. tiracial. Educational level was defined as follows: 1) less th Multivariate ordinal logistic regression analysis was than high school: less than 9 grade or no high school performed to evaluate the relationship among RC, the diploma; 2) high school graduate: high school graduate/ TC/RC ratio and grades of verbal learning and memory GED or equivalent; and 3) college or above: some college functioning. Models were further adjusted for poten- or AA degree, college graduate or above. BMI was cat- tial risk factors or confounders based on prior stud- egorized into 1) normal, 2) overweight and 3) obese using 2 2 ies. Model 1 included univariate analysis. Model 2 was 25 kg/m and 30 kg/m as the cutoffs. Smoking status was adjusted for age, sex, race, and education level. Model defined as follows: 1) current smokers: had smoked more 3 was further adjusted for BMI status, smoking sta- than 100 cigarettes (including hand rolled cigarettes, tus and drinking status plus the adjustments in Model cigars, and cigarillos) in their lifetime and had smoked in 2. Subgroup analysis was performed to evaluate the the last 28  days; 2) ex-smokers: had smoked more than robustness of the results in a diversity of demographic, 100 cigarettes in their lifetime but had not smoked in the disease-specific and lipid-stratified subgroups. A two- last 28 days; and 3) never smokers: had not smoked more sided P < 0.05 was considered statistically significant. than 100 cigarettes in their lifetime and did not cur- rently smoke [14]. Drinking status was defined as follows: 1) current drinker: had drunk at least 12 drinks in their Results lifetime and had drunk at least 1 drink in the past year; Characteristics of the study population 2) former drinker: had drunk at least 12 drinks in their Finally, according to the study population selection pro- lifetime and had drunk no drinks in past year; and 3) life- cess, 1377 participants aged ≥60 years were included time abstainer: had drunk fewer than 12 drinks in their in the study (Fig.  1). The mean age of the participants lifetime [15]. was 69.1 ± 6.5 years, and 55% were male. A total of 79% Other variables included in the baseline characteristics of the participants were non-Hispanic white, 61% had and subgroup analysis were marital status, statin use, dia- an educational level of college or above, and 32% were betes, and hypertension. Marital status was defined as 1) not married or never married. Thirty-eight percent of married/cohabitating and 2) not married/never married. the participants were obese (BMI ≥28 kg/m2), 19%had Diabetes and hypertension were based on self-report a history of diabetes, and 59% had hypertension. questionnaire data. For the baseline lipid profile, the mean TC level was 4.97 ± 1.07  mmol/L, and the median RC level was 0.54 Statistical analyses (0.39, 0.78) mmol/L. For CERAD test performance, 26% All statistical analyses were conducted by Stata SE 16.0 of the participants had scores in the lowest age-strati- (Stata Corporation, College Station, TX). The complex fied quartile. Details are depicted in Table 1 . Xie et al. Lipids in Health and Disease (2022) 21:120 Page 4 of 10 Table 1 Baseline characteristics of the study population grouped by age‑stratified quartiles of the CERAD total score Groups by age-stratified quartiles of CERAD total score st nd rd th Total (n = 1377) 1 quartile 2 quartile 3 quartile 4 quartile P for trend (n = 458) (n = 381) (n = 305) (n = 233) Age, yrs 69.1 ± 6.5 69.6 ± 7.4 69.3 ± 6.3 68.8 ± 6.3 68.6 ± 6.9 < 0.001 Gender, n(%) Female 705(55) 184(12) 197(14) 178(16) 146(14) < 0.001 Male 672(45) 274(14) 184(13) 127(9.4) 87(7.5) Race, n(%) White 696(79) 180(18) 185(21) 171(21) 160(19) < 0.001 Black 281(8.5) 110(3) 83(2.6) 57(1.8) 31(1) Other races 400(12) 168(4.8) 113(3.6) 77(2.4) 42(1.2) Education level, n(%) Less than high school 356(16) 183(7.5) 110(5.4) 45(2.3) 18(1.1) < 0.001 High school graduate 329(23) 113(6.8) 94(6.8) 72(5.5) 50(3.9) College or above 690(61) 161(11) 176(15) 188(17) 165(16) Marital status, n(%) Married/Cohabiting 855(68) 294(18) 225(19) 190(17) 146(15) < 0.001 Not married/Never married 522(32) 164(8) 156(9.2) 115(8.1) 87(6.3) BMI, n(%) Normal 379(27) 122(6.3) 98(6.6) 88(7.2) 71(7.1) < 0.001 Overweight 465(35) 155(9.3) 128(9.8) 100(8.4) 82(7.1) Obese 533(38) 181(10) 155(11) 117(9.4) 80(7.4) Drinking status, n(%) Lifetime abstainer 215(14) 68(3.3) 69(4) 45(3.4) 33(2.8) < 0.001 Former drinker 368(23) 136(7.4) 109(6.8) 77(5.2) 46(3.7) Current drinker 766(63) 239(15) 195(17) 180(16) 152(15) Smoking status, n(%) Never smoker 682(49) 222(12) 179(13) 149(12) 132(12) < 0.001 Former smoker 527(40) 173(10) 154(12) 122(11) 78(7.2) Current smoker 166(11) 63(3.7) 46(2.7) 34(2.5) 23(1.9) Diabetes, n(%) 302(19) 108(6.3) 99(6.2) 61(3.7) 34(2.5) < 0.001 Hypertension, n(%) 859(59) 290(16) 241(17) 189(14) 139(12) < 0.001 Stroke, n(%) 99(6.7) 35(2) 34(2) 15(1.3) 15(1.4) < 0.001 Statin use, n(%) Statin user 850(60) 276(14) 223(16) 195(15) 156(15) < 0.001 Non statin user 527(40) 182(12) 158(12) 110(9.6) 77(6.5) TC, mmol/L 4.97 ± 1.07 4.77 ± 1.12 4.91 ± 1.02 5.08 ± 1.07 5.17 ± 0.99 < 0.001 LDL-C, mmol/L 2.87 ± 0.93 2.73 ± 1.00 2.81 ± 0.89 2.95 ± 0.91 3.02 ± 0.85 < 0.001 HDL-C, mmol/L 1.47 ± 0.43 1.38 ± 0.45 1.46 ± 0.38 1.54 ± 0.41 1.54 ± 0.42 < 0.001 TG, mmol/L 1.19 (0.85,1.71) 1.38 (0.90,1.75) 1.21 (0.89,1.75) 1.12 (0.78,1.64) 1.10 (0.81,1.68) < 0.001 RC, mmol/L 0.54 (0.39,0.78) 0.63 (0.41,0.81) 0.55 (0.41,0.80) 0.51 (0.36,0.75) 0.50 (0.36,0.77) < 0.001 * a b c Data were shown as mean (± standard deviation), median (interquartile range), unweighted counts (weighted percentage) Analysis of the association among RC, the TC/RC ratio odds ratio (95% CI) was 0.66 (0.46–0.95) for the fourth and the CERAD total score quartile, 0.61 (0.44–0.86) for the third quartile, and 0.89 On a continuous scale, each 1 mmol/L increase in RC was (0.67–1.17) for the second quartile. The P for trend test associated with a multivariable adjusted odds ratio (95% was 0.011 across the quartiles. Details are provided in CI) of 0.74 (0.58–0.94) for having a CERAD total score Table 2. in a higher quartile. On a categorical scale, compared On a continuous scale, each 1 unit increase in the with that of the first quartile, the multivariable adjusted TC/RC ratio was associated with a multivariable Xie  et al. Lipids in Health and Disease (2022) 21:120 Page 5 of 10 Subgroup analysis Table 2 Weighted multivariable ordinal logistic regression analysis between RC levels (log‑transformed) and age ‑stratified Stratified analyses were conducted for the relationship quartiles of the CERAD total score between RC levels, the TC/RC ratio, and age-stratified quartiles of the CERAD total score (Figs.  2  and  3). After RC, mmol/L OR (95%CI) adjusting for confounders, the effect size of the RC level Model 1 Model 2 Model 3 as well as the TC/RC ratio on quartiles of the CERAD total score were almost all consistent across the sub- Continuous, 0.72(0.56–0.93)* 0.74(0.58–0.94)* 0.77(0.61–0.96)* mmol/L groups, including sex, age, education level, BMI status, Categorical (quartiles) diabetes, hypertension, and TC level (P for interaction ≤0.39 mmol/L Ref Ref Ref ≥0.05). Only the race of black was an exception, which 0.39 to ≤ 0.87(0.67–1.13) 0.91(0.68–1.21)* 0.93(0.70–1.25) deserves further study. 0.54 mmol/L 0.54 to ≤ 0.56(0.40–0.80)** 0.62(0.44–0.88)** 0.64(0.47–0.88)** 0.78 mmol/L Discussion > 0.78 mmol/L 0.66(0.44–0.98)* 0.67(0.46–0.98)* 0.70(0.50–0.98)* To our knowledge, this is the first study to explore the P for linear trend 0.040 0.015 0.013 association between plasma RC levels, as well as the TC Model 1 univariate analysis to RC ratio, and cognitive function. The present study Model 2 adjusted for age, gender, education level, and race had the following important findings. First, a higher level Model 3 further adjusted for BMI status, smoking status, and drinking status plus of RC was associated with a higher risk of verbal learn- model 2 ing and memory function impairment. Second, a higher P < 0.05, **P < 0.01 TC to RC ratio was associated with a lower risk of ver- bal learning and memory function impairment. Third, in comparison with RC levels, the TC/RC ratio showed a steadier relationship with verbal learning and memory Table 3 Weighted multivariable ordinal logistic regression function under multiple analytic approaches. Fourth, the analysis between the TC/RC ratio (log‑transformed) and age ‑ effect sizes of RC levels as well as the TC/RC ratio on ver - stratified quartiles of the CERAD total score bal learning and memory function were consistent across TC/RC OR (95% CI) almost all subgroup analyses. RC is defined as the cholesterol composition of rem - Model 1 Model 2 Model 3 nants that are metabolized from TG-rich lipoproteins, Continuous 1.68(1.29–2.19)** 1.46(1.12–1.90)** 1.40(1.10–1.78)** including chylomicron and very low-density lipopro- Categorical (quartiles) tein (VLDL) [17]. Quite a few observational studies have ≤6.08 Ref Ref Ref investigated the relationship between classical lipid com- > 6.08 to ≤8.75 1.12(0.80–1.56) 1.14(0.83–1.56) 1.09(0.79–1.51) ponents, such as TC, LDL-C, and HDL-C, and cogni- > 8.75 to ≤12.94 1.38(1.09–1.76)** 1.42(1.10–1.84)* 1.33(1.03–1.71)* tive function. However, there is no current laboratory > 12.94 1.87(1.25–2.80)** 1.51(1.02–2.22)* 1.41(1.00–1.97)* or epidemiological evidence available concerning the P for linear trend 0.002 0.020 0.028 association between RC and cognitive function and the Model 1 univariate analysis underlying mechanisms. Here, for the first time, this Model 2 adjusted for age, gender, education level, and race study provides a glimpse of the relationship between RC Model 3 further adjusted for BMI status, smoking status, and drinking status plus and cognitive function, suggesting that a lower RC level is model 2 associated with better verbal learning and memory func- P < 0.05, **P < 0.01 tion defined by the CERAD total score. The CERAD WL subtest is part of the CERAD Neuropsychology battery, which was originally designed to permit staging of AD. Utilizing a multiple analytic approach, the study dem- onstrated that a higher level of RC was correlated with adjusted odds ratio (95% CI) of 1.45 (1.13–1.87) for worse verbal learning and memory function in American having a CERAD total score in a higher quartile. On a elderly individuals aged ≥60 years. categorical scale, compared with that of the first quar- Although no relevant study is currently available, the tile, the multivariable adjusted odds ratio (95% CI) was study results have biologically plausible explanations. 1.50 (1.03–2.18) for the fourth quartile, 1.38 (1.07– Remnants with diameters less than 70 nm can penetrate 1.78) for the third quartile, and 1.12 (0.81–1.55) for the the intima of the arterial wall, leading to atherosclerosis second quartile. The P for trend test was 0.020 across [18]. The relationship between RC and atherosclerotic the quartiles. Details are provided in Table 3. Xie et al. Lipids in Health and Disease (2022) 21:120 Page 6 of 10 Fig. 1 Flowchart of the sample selection process of the study population diseases [19], as well as its association with cardiovas- clearance by the liver. Combined with the results of the cular outcomes independent of LDL-C [20], has been present study, disordered APOE-mediated clearance established in previous studies. Due to its atherogenic of remnant lipoprotein might partly participate in the properties, RC can contribute to atherosclerosis in both development of cognitive impairment. Further research the carotid artery [21] and arterioles in the brain. A should be conducted to determine the underlying Mendelian randomization study confirmed the causal mechanisms and their clinical significance. relationship of remnant lipoprotein-associated genes The present study proposed a new blood lipid index, and ischemic stroke [22]. Prospective cohort studies the TC/RC ratio. The adverse effect of high TC levels on symptomatic intracranial atherosclerotic stenosis on cognitive function has been abundantly investi- and ischemic stroke indicated that RC causes cerebral gated in previous studies. However, serum TC includes hypoperfusion [23]. In addition, population-based stud- cholesterol molecules from a variety of subtypes of ies showed that a reduction in cerebral perfusion was lipoprotein, and lowering cholesterol of different lipo - associated with an increased risk of dementia [24, 25]. protein subtypes might produce differential outcomes. Previous studies also reported that RC enhanced oxi- Therefore, the beneficial effect of cholesterol-lowering dative stress and proinflammatory effects on vascular therapy on cognitive function is still controversial, endothelial and smooth muscle cells [26], which might especially in elderly individuals [29]. Based on previ- damage the blood‒brain barrier, subsequently altering ous studies and the relationship between RC levels and amyloid degradation and cholesterol homeostasis [27] the CERAD total score found in the present study, it is in the brain. A recent Mendelian randomization study supposed that TC, in combination with RC, might be a on the risk factors for AD found that genetically ele- better bioindex for the prediction of cognitive function vated TC and LDL-C levels increased neurotic plaque than TC or RC separately. In light of this, the present burden, but the effects were driven by single nucleotide study examined and demonstrated that a higher TC/RC polymorphisms of APOE [28], whose genetic product ratio is suggestive of better verbal learning and memory is known to be the key ligand for remnant lipoprotein function assessed by CERAD tests. Xie  et al. Lipids in Health and Disease (2022) 21:120 Page 7 of 10 Fig. 2 Eec ff t size of RC levels on the age ‑stratified quartiles of the CERAD total score in subgroups. Notes: Model adjusted for age, sex, educational level, race, BMI status, smoking status, and drinking status In addition, the present study found that the TC/RC by previous studies. For future clinical application, the ratio showed a significant positive association not only present study provided evidence for the utilization of with the CERAD total score but also with the CERAD RC levels and the TC/RC ratio in the evaluation of ver- delayed trial test score [see Additional file  1, Table  S1], bal learning and memory function, which might assist while RC levels showed no association with delayed trial in risk stratification for cognitive function impairment test score [see Additional file  1, Table S2]. Previous stud- or AD susceptibility in elderly individuals. Consider- ies have found that some of the measures in the CERAD ing both cardiovascular and cognitive benefits implied WL subtest, particularly delayed recall of a word list, by the present study, the lower the RC level, the better. could more efficiently distinguish persons with demen - Regarding the benefit of a higher TC/RC ratio for verbal tia from those with normal cognition [30]. Therefore, learning and memory function, there might be a poten- the results of the present study indicated that the TC/RC tial atherogenic risk when the higher ratio is mainly ratio might have better value for predicting verbal learn- attributed to a relatively high TC level rather than a ing and memory function than RC levels. low RC level. However, in the context of this study, the mean TC level was 4.97 ± 1.07 mmol/L, which was not Comparisons with other studies and what the current work a significant atherogenic level. adds to the existing knowledge Conclusively, the current study examined two available Study strengths and limitations blood lipid indices for the assessment of verbal learning There are several strengths in the present study. First, and memory function, which have not been reported the study observed a known cardiovascular risk factor, Xie et al. Lipids in Health and Disease (2022) 21:120 Page 8 of 10 Fig. 3 Eec ff t size of the TC/RC ratio on the age ‑stratified quartiles of the CERAD total score in subgroups. Notes: Model adjusted for age, sex, educational level, race, BMI status, smoking status, and drinking status RC, and its negative relationship with verbal learning function. Second, in this study, LDL-C was calcu- and memory function. Moreover, the study utilized lated based on the Friedewald equation, which is not data extracted from the NHANES database, which applicable when TG levels > 400  mg/dl. Therefore, used complex, multistage sampling, and relatively the results of the study should not be interpreted in convincing results could be drawn with a proper ana- the situation of very high TG levels. Third, the pre- lytic approach. Additionally, the study proposed an sent study used calculated fasting RC to represent the RC derivative, the TC/RC ratio, and found its positive “remnant” cholesterol level; however, this calculated association with verbal learning and memory function, fasting RC includes not only cholesterol from rem- which had not been mentioned previously. Addition- nants but also cholesterol from newly formed VLDL ally, the results were estimated in several subpopula- particles, which in fact overestimates the cholesterol tions, verifying its authenticity. levels of actual remnants. However, RC from the indi- The study has several limitations. First, this was rect formula has been widely used in numerous clini- an observational study with a cross-sectional design, cal studies and has been proven to be a convenient and and the causal relationship could not be determined reliable risk predictor. Finally, although the present between RC levels or the TC/RC ratio and cognitive study included many important potential covariates Xie  et al. Lipids in Health and Disease (2022) 21:120 Page 9 of 10 Declarations previously reported to affect cognition in the statisti- cal models, the possibility that residual confounding Ethics approval and consent to participate factors remain could not be ruled out. Future large- The NHANES protocols were approved by the National Center for Health Statistics Ethics Review Board of the US CDC, and written informed consent scale, prespecified trials are needed to further explore from all the participants was provided during the survey. this subject. Consent for publication Not applicable. Conclusions Competing interests Utilizing a cohort of American elderly individuals aged The authors declare that they have no competing interests. over 60  years from the NHANES database, the pre- Author details sent study found that in the context of a TG level below Cardio‑Metabolic Medicine Center, National Center for Cardiovascular 400  mg/dl, a lower RC level and a higher TC/RC ratio Diseases, Fuwai Hospital, Chinese Academy of Medical Sciences, Peking Union Medical College, Beijing, China. Coronary Heart Disease Center, were associated with better verbal learning and memory National Center for Cardiovascular Diseases, Fuwai Hospital, Chinese Academy function. The present study indicated that lowering RC of Medical Sciences, Peking Union Medical College, Beijing, China. Division levels or increasing the TC/RC ratio when the TG level of Cardiology, The First Affiliated Hospital of Soochow University, Suzhou, Jiangsu Province, China. is below 400  mg/dl could possibly be beneficial for pre - venting cognitive impairment in elderly individuals. The Received: 3 October 2022 Accepted: 1 November 2022 study results, which need to be verified in future larger cohort studies, might help in guiding risk prediction and primary prevention of cognitive impairment in elderly individuals. References 1. GBD 2019 Dementia Forecasting Collaborators. Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: Abbreviations an analysis for the Global Burden of Disease Study 2019. Lancet Public RC: Remnant cholesterol; TC/RC: Total cholesterol/remnant cholesterol; CERAD: Health. 2022;7(2):c105–25 Consortium to Establish a Registry for Alzheimer’s Disease Word List Memory 2. Liu Y, Zhong X, Shen J, Jiao L, Tong J, Zhao W, et al. Elevated serum TC Task; NHANES: National Health and Nutrition Examination Survey; TC: Total and LDL‑ C levels in Alzheimer’s disease and mild cognitive impairment: A cholesterol; LDL: Low‑ density lipoprotein; HDL: High‑ density lipoprotein; meta‑analysis study. Brain Res. 2020;1727. LDL‑ C: Low‑ density lipoprotein cholesterol; HDL‑ C: High‑ density lipoprotein 3. Anstey KJ, Ashby‑Mitchell K, Peters R. Updating the Evidence on the cholesterol; TG: Triglycerides; VLDL: Very low‑ density lipoprotein; BMI: Body Association between Serum Cholesterol and Risk of Late‑Life Dementia: mass index. Review and Meta‑Analysis. J Alzheimers Dis. 2017;56(1):215–28. 4. Ding D, Zhou F, Cao Y, Liang X, Wu W, Xiao Z, et al. Cholesterol profiles and incident cognitive decline among older adults: the Shanghai Aging Supplementary Information Study. Age Ageing. 2021;50(2):472–9. The online version contains supplementary material available at https:// doi. 5. Benn M, Nordestgaard BG, Frikke‑Schmidt R, Tybjærg‑Hansen A. Low LDL org/ 10. 1186/ s12944‑ 022‑ 01729‑4. cholesterol, PCSK9 and HMGCR genetic variation, and risk of Alzheimer’s disease and Parkinson’s disease: Mendelian randomisation study. Bmj. 2017;357:1648. Additional file 1: Table S1. Weighted multivariable ordinal logistic regres‑ 6. Bruemmer D, Cho L. Remnant Cholesterol: The Leftovers and Their sion analysis between RC levels (log‑transformed) and age ‑stratified quar ‑ Contribution to Atherosclerotic Cardiovascular Disease. Circ Cardiovasc tiles of the CERAD delayed score. Table S2. Weighted multivariable ordinal Imaging. 2021;14(4). logistic regression analysis between the TC/RC ratio (log‑transformed) and 7. Quispe R, Martin SS, Michos ED, Lamba I, Blumenthal RS, Saeed A, et al. age‑stratified quartiles of the CERAD delayed score. Remnant cholesterol predicts cardiovascular disease beyond LDL and ApoB: a primary prevention study. Eur Heart J. 2021;42(42):4324–32. 8. Varbo A, Benn M, Tybjærg‑Hansen A, Jørgensen AB, Frikke ‑Schmidt R, Acknowledgements Nordestgaard BG. Remnant cholesterol as a causal risk factor for ischemic Not applicable. heart disease. J Am Coll Cardiol. 2013;61(4):427–36. 9. Burnett JR, Hooper AJ, Hegele RA. Remnant Cholesterol and Atheroscle‑ Authors’ contributions rotic Cardiovascular Disease Risk. J Am Coll Cardiol. 2020;76(23):2736–9. Ying‑ Yi Xie: Methodology implementation, Formal analysis, Writing – original 10. Paquette M, Bernard S, Paré G, Baass A. Dysbetalipoproteinemia: Differ ‑ draft, Writing – review & editing. Liang Zhao: Validation. Li‑ Jian Gao: Validation. entiating Multifactorial Remnant Cholesterol Disease From Genetic ApoE Rui‑ Xia Xu: Validation. Ying Gao: Validation. Ke‑Fei Dou: Validation. Yuan‑Lin Deficiency. J Clin Endocrinol Metab. 2022;107(2):538–48. Guo: Conceptualization, Methodology guidance, Project administration, 11. National Health and Nutrition Examination Survey. National Center for Validation, Writing – review & editing. Yong‑Ming He: Methodology guidance, Health Statistics (NCHS), Hyattsville. 2011–2014. https:// wwwn. cdc. gov/ Project administration, Validation, Writing – review & editing. The author(s) Nchs/ Nhanes/. Accessed 12 Mar 2022. read and approved the final manuscript. 12. Dong X, Li S, Sun J, Li Y, Zhang D. Association of Coffee, Decaffeinated Coffee and Caffeine Intake from Coffee with Cognitive Performance Funding in Older Adults: National Health and Nutrition Examination Survey This study was supported by the Chinese Academy of Medical Science Inno‑ (NHANES) 2011–2014. Nutrients. 2020;12(3):840. vation Fund for Medical Sciences (2021‑I2M ‑1–009). 13. Li S, Sun W, Zhang D. Association of Zinc, Iron, Copper, and Selenium Intakes with Low Cognitive Performance in Older Adults: A Cross‑ Availability of data and materials Sectional Study from National Health and Nutrition Examination Survey The datasets analyzed during the current study are available on the NHANES (NHANES). J Alzheimers Dis. 2019;72(4):1145–57. official website, https:// wwwn. cdc. gov/ Nchs/ Nhanes/. Xie et al. Lipids in Health and Disease (2022) 21:120 Page 10 of 10 14. Cantini L, Mentrasti G, Russo GL, Signorelli D, Pasello G, Rijavec E, et al. Evaluation of COVID‑19 impact on DELAYing diagnostic‑therapeutic pathways of lung cancer patients in Italy (COVID‑DELAY study): fewer cases and higher stages from a real‑ world scenario. ESMO Open. 2022;7(2). 15. National Center for Health Statistics (NCHS). Adult Alcohol Use Informa‑ tion: Glossary ‑ Alcohol. 2018. Available from: https:// www. cdc. gov/ nchs/ nhis/ alcoh ol/ alcoh ol_ gloss ary. htm. Accessed 12 Mar 2022. 16. National Center for Health Statistics (NCHS). Module 3: Weighting. National Health and Nutrition Examination Survey: Tutorials. 2021. Avail‑ able from: https:// wwwn. cdc. gov/ nchs/ nhanes/ tutor ials/ modul e3. aspx. Accessed 12 Mar 2022. 17. Varbo A, Benn M, Nordestgaard BG. Remnant cholesterol as a cause of ischemic heart disease: evidence, definition, measurement, atherogenic‑ ity, high risk patients, and present and future treatment. Pharmacol Ther. 2014;141(3):358–67. 18. Nordestgaard BG. Triglyceride‑Rich Lipoproteins and Atherosclerotic Cardiovascular Disease: New Insights From Epidemiology, Genetics, and Biology. Circ Res. 2016;118(4):547–63. 19. Lin A, Nerlekar N, Rajagopalan A, Yuvaraj J, Modi R, Mirzaee S, et al. Rem‑ nant cholesterol and coronary atherosclerotic plaque burden assessed by computed tomography coronary angiography. Atherosclerosis. 2019;284:24–30. 20. Castañer O, Pintó X, Subirana I, Amor AJ, Ros E, Hernáez Á, et al. Remnant Cholesterol, Not LDL Cholesterol, Is Associated With Incident Cardiovas‑ cular Disease. J Am Coll Cardiol. 2020;76(23):2725–35. 21. Qian S, You S, Sun Y, Wu Q, Wang X, Tang W, et al. Remnant Cholesterol and Common Carotid Artery Intima‑Media Thickness in Patients With Ischemic Stroke. Circ Cardiovasc Imaging. 2021;14(4). 22. Si S, Hou L, Chen X, Li W, Liu X, Liu C, et al. Exploring the Causal Roles of Circulating Remnant Lipid Profile on Cardiovascular and Cerebro ‑ vascular Diseases: Mendelian Randomization Study. J Epidemiol. 2022;32(5):205–14. 23. Kim DE, Kim JY, Jeong SW, Cho YJ, Park JM, Lee JH, et al. Association between changes in lipid profiles and progression of symptomatic intrac‑ ranial atherosclerotic stenosis: a prospective multicenter study. Stroke. 2012;43(7):1824–30. 24. Wolters FJ, Zonneveld HI, Hofman A, van der Lugt A, Koudstaal PJ, Vernooij MW, et al. Cerebral Perfusion and the Risk of Dementia: A Population‑Based Study. Circulation. 2017;136(8):719–28. 25. Wolters FJ, de Bruijn RF, Hofman A, Koudstaal PJ, Ikram MA. Cerebral Vasoreactivity, Apolipoprotein E, and the Risk of Dementia: A Population‑ Based Study. Arterioscler Thromb Vasc Biol. 2016;36(1):204–10. 26. Nakajima K, Tanaka A. Atherogenic postprandial remnant lipoproteins; VLDL remnants as a causal factor in atherosclerosis. Clin Chim Acta. 2018;478:200–15. 27. Saeed AA, Genové G, Li T, Lütjohann D, Olin M, Mast N, et al. Eec ff ts of a disrupted blood‑brain barrier on cholesterol homeostasis in the brain. J Biol Chem. 2014;289(34):23712–22. 28. Andrews SJ, Fulton‑Howard B, O’Reilly P, Marcora E, Goate AM. Causal Associations Between Modifiable Risk Factors and the Alzheimer’s Phe ‑ nome. Ann Neurol. 2021;89(1):54–65. 29. Power MC, Rawlings A, Sharrett AR, Bandeen‑Roche K, Coresh J, Ballan‑ tyne CM, et al. Association of midlife lipids with 20‑ year cognitive change: A cohort study. Alzheimers Dement. 2018;14(2):167–77. 30. Fillenbaum GG, van Belle G, Morris JC, Mohs RC, Mirra SS, Davis PC, et al. Consortium to Establish a Registry for Alzheimer’s Disease (CERAD): the Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : first twenty years. Alzheimers Dement. 2008;4(2):96–109. fast, convenient online submission Publisher’s Note thorough peer review by experienced researchers in your field Springer Nature remains neutral with regard to jurisdictional claims in pub‑ rapid publication on acceptance lished maps and institutional affiliations. support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Lipids in Health and Disease Springer Journals

Association between remnant cholesterol and verbal learning and memory function in the elderly in the US

Download PDF
10 pages

Loading next page...
 
/lp/springer-journals/association-between-remnant-cholesterol-and-verbal-learning-and-memory-JyVOIdVqAi

References (32)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2022
eISSN
1476-511X
DOI
10.1186/s12944-022-01729-4
Publisher site
See Article on Publisher Site

Abstract

Background: The relationship between remnant cholesterol (RC) and atherosclerotic cardiovascular risk has been given increasing attention in recent years. However, its association with verbal learning and memory performance has not been reported. Methods: Data were extracted from the National Health and Nutrition Examination Survey (NHANES) 2011–2014 database. Participants aged ≥60 years with available fasting lipid data were included. Verbal learning and memory performance were evaluated using the Consortium to Establish a Registry for Alzheimer’s Disease Word List Memory Task (CERAD‑ WL) subtest. The CERAD total score was calculated as the mean of three immediate recalls and a delayed recall. RC was calculated as total cholesterol ( TC) minus the sum of low‑ density lipoprotein cholesterol (LDL‑ C) and high‑ density lipoprotein cholesterol (HDL‑ C). Multivariate ordinal logistic regression was performed to evaluate the association between RC, as well as its derived marker, the TC/RC ratio, and age‑stratified quartiles of the CERAD total score. Results: A total of 1377 participants were analysed. On a continuous scale, per 1 mmol/L increase in RC and per 1 unit increase in the TC/RC ratio were associated with multivariable adjusted odds ratios (95% CI) of 0.74 (0.58–0.94) and 1.45 (1.13–1.87), respectively, for having a CERAD total score in a higher quartile. On a categorical scale, higher RC quartiles were associated with a CERAD total score in a lower quartile; in contrast, the higher TC/RC quartile was associated with a CERAD total score in a higher quartile (all P for trend < 0.05). Conclusions: The current study suggests that lower RC levels and a higher TC/RC ratio are associated with better verbal learning and memory function, which indicates that lowering RC levels could be beneficial for preventing cognitive impairment in elderly individuals. Further research is needed to validate the causal roles of RC and the TC/RC ratio in cognition. Keywords: Remnant cholesterol, Total cholesterol/remnant cholesterol, Cognition, Verbal learning and memory function, Elderly, Cross‑sectional study Introduction Dementia is a vital health problem in the global growing *Correspondence: [email protected]; [email protected] aging population. The estimated dementia population Cardio‑Metabolic Medicine Center, National Center for Cardiovascular will increase from 57.4 (50.4–65.1) million cases globally Diseases, Fuwai Hospital, Chinese Academy of Medical Sciences, Peking Union in 2019 to 152.8 (130.8–175.9) million cases in 2050 [1]. Medical College, Beijing, China 3 Given the lack of effective pharmacological therapies for Division of Cardiology, The First Affiliated Hospital of Soochow University, Suzhou, Jiangsu Province, China dementia thus far, the primary prevention of modifiable Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Xie et al. Lipids in Health and Disease (2022) 21:120 Page 2 of 10 risk factors has become crucial in addressing the rising nutritional status of civilian, noninstitutionalized resi- epidemic of cognitive impairment. dent populations in the United States. A complex, mul- Dyslipidemia has been found to be independently tistage sampling procedure was implemented to select associated with cognitive impairment. In observational eligible representative participants. The present study studies, higher serum TC levels in mid-life were associ- made use of the 2011–2012 and 2013–2014 cycles with ated with a higher risk of developing dementia in late life 19,931 individuals, of which participants aged ≥60 years [2], but the impact of serum TC levels in late life on the were eligible for cognitive function assessment and development of dementia remains unclear. A meta-anal- enrolled for analysis. Participants without fasting plasma ysis synthesizing 34 cohort studies found no association lipid data or complete cognitive assessment results were between serum TC levels and mild cognitive impair- excluded. Those without complete measurements of TC, ment, Alzheimer’s disease (AD), vascular dementia, any LDL-C, and HDL-C values for the calculation of RC were dementia, or cognitive decline [3]. Studies on serum also excluded. LDL-C levels and dementia or cognitive decline also dis- The NHANES protocols were approved by the National played differential results and seemed to vary by sex and Center for Health Statistics Ethics Review Board of the the presence of cardiovascular risk factors [4]. However, U.S. CDC. All participants received and completed writ- a Mendelian randomization study showed that geneti- ten informed consent during the survey. All material for cally low serum levels of LDL-C reduced the risk of AD analysis can be accessed on the NHANES official website [5]. Overall, inconsistent results have been found in the [11]. association between serum lipid profiles and cognitive impairment, limiting the clinical utility of lipid biomark- Measurement of blood lipids and the definition ers. In addition to further research on traditional blood of exposure variables lipid parameters, the exploration of new lipid parameters TC and triglycerides (TGs) were measured using an to predict cognitive function is urgently needed, espe- enzymatic assay method. HDL-C was measured using cially in the elderly population. the heparin-manganese precipitation method or a direct Remnant cholesterol (RC) represents the amount of immunoassay technique. LDL-C was calculated from cholesterol in the remnant lipoproteins transformed from measured values of TC, TGs, and HDL-C according to triglyceride-rich lipoproteins in the blood [6]. Recently, the Friedewald formula as follows: [LDL-C] = [TC] – RC has been shown to equally or even more precisely [HDL-C] – [TG/5]. The formula is valid for TG values predict cardio-cerebrovascular outcomes compared to less than or equal to 400  mg/dL. The exposure variables LDL-C or HDL-C [7–9]. Elevated RC levels are closely were remnant cholesterol (RC) and its derivative, the related to triglyceride metabolism disorders and insuf- TC/RC ratio. RC was calculated as TC minus the sum of ficient APOE-mediated remnant lipoprotein clearance LDL-C and HDL-C [RC = TC-(LDL-C + HDL-C)], and by the liver, such as type III hyperlipidemia [10]. Mean- the TC/RC ratio was calculated as TC divided by RC. while, APOE variants are closely related to the risk of AD. Therefore, RC has the biological possibility of affect - Assessment of verbal learning and memory function ing cognitive function. However, no study has reported and the definition of the outcome variable whether or how RC is associated with cognitive function. In NHANES 2011–2014, a series of evaluations of cog- In seeking a more plausible and reliable lipid marker of nitive functioning were conducted during the MEC sur- late-life cognitive impairment, the present study explored vey, including the word learning and recall modules from the relationship of RC with cognitive function utilizing the Consortium to Establish a Registry for Alzheimer’s data from a cohort of American elderly individuals in Disease (CERAD W-L). The CERAD W-L assesses the the National Health and Nutrition Examination Survey immediate and delayed learning ability for new verbal (NHANES) conducted between 2011 and 2014. Given information (memory subdomain), consisting of three the complexity of the metabolism of remnant lipopro- sequential learning trials and a delayed recall trial. For tein in human plasma, the study also explored a new lipid learning trials, participants were tasked with reading marker derived from RC, the total cholesterol to remnant 10 unrelated words displayed on a screen aloud, one at cholesterol (TC/RC) ratio, to further investigate the rela- a time, and were asked to recall as many words as pos- tionship of RC with cognitive function. sible immediately after their presentation. A delayed recall test was administered after the other two cogni- Methods tive tests were completed. The outcome variable was the Study design and population CERAD total score, which was calculated as the average The NHANES database is a cross-sectional program with of all three immediate recalls after each learning trial and a 2-year-cycle design that aims to assess the health and the delayed recall. Given the significant effect of age on Xie  et al. Lipids in Health and Disease (2022) 21:120 Page 3 of 10 cognitive function, the age- stratified (≥60 to 70, ≥70 survey design of the NHANES was taken into account to 80, and ≥80 years) quartiles of the CERAD total test by specifying primary sampling units (PSUs), strata, score were used [12, 13] to evaluate verbal learning and and sampling weights in the software’s svyset module. memory function. Specifically, the age-stratified quar - Sampling weights were constructed for the combined tiles of the CERAD total test score were calculated by 2011–2012 and 2013–2014 4-year cycles by dividing first dividing the studypopulation into 3 age groups (≥60 the individual sampling weights by 2, according to the to 70, ≥70 to 80, and ≥80 years) and then categorizing NHANES analytic tutorials [16]. participants into 4 grades of verbal learning and memory Baseline characteristics were grouped by age-strati- function (1 to 4) by the weighted quartiles of CERAD fied quartiles of the CERAD total test score. Data are total score separately in the 3 age groups. The lowest described as the mean (± standard deviation) for con- quartile indicated impaired verbal learning and memory tinuous variables, median (interquartile range) for function. The higher the quartile, the better the learning skewed variables, and sample counts (weighted per- and memory function. centage) for categorical variables. A weighted trend test was performed across age-stratified quartile groups for each continuous and categorical variable. Covariates In the statistical analysis of lipid-cognition asso- Potential confounding factors were investigated, includ- ciations, RC and the TC/RC ratio were treated as both ing age, sex (male and female), race, BMI status, educa- continuous and categorical variables. Continuous RC tional level, smoking status, and drinking status. Race and TC/RC variables were natural log-transformed to was defined as follows: 1) non-Hispanic white; 2) non- fit regression models due to their right-skewed distri - Hispanic black; and 3) other races, including Mexican bution [7]. Categorical RC and TC/RC variables with American, other Hispanic, non-Hispanic Asian and mul- four grades were created using quartiles as cutoffs. tiracial. Educational level was defined as follows: 1) less th Multivariate ordinal logistic regression analysis was than high school: less than 9 grade or no high school performed to evaluate the relationship among RC, the diploma; 2) high school graduate: high school graduate/ TC/RC ratio and grades of verbal learning and memory GED or equivalent; and 3) college or above: some college functioning. Models were further adjusted for poten- or AA degree, college graduate or above. BMI was cat- tial risk factors or confounders based on prior stud- egorized into 1) normal, 2) overweight and 3) obese using 2 2 ies. Model 1 included univariate analysis. Model 2 was 25 kg/m and 30 kg/m as the cutoffs. Smoking status was adjusted for age, sex, race, and education level. Model defined as follows: 1) current smokers: had smoked more 3 was further adjusted for BMI status, smoking sta- than 100 cigarettes (including hand rolled cigarettes, tus and drinking status plus the adjustments in Model cigars, and cigarillos) in their lifetime and had smoked in 2. Subgroup analysis was performed to evaluate the the last 28  days; 2) ex-smokers: had smoked more than robustness of the results in a diversity of demographic, 100 cigarettes in their lifetime but had not smoked in the disease-specific and lipid-stratified subgroups. A two- last 28 days; and 3) never smokers: had not smoked more sided P < 0.05 was considered statistically significant. than 100 cigarettes in their lifetime and did not cur- rently smoke [14]. Drinking status was defined as follows: 1) current drinker: had drunk at least 12 drinks in their Results lifetime and had drunk at least 1 drink in the past year; Characteristics of the study population 2) former drinker: had drunk at least 12 drinks in their Finally, according to the study population selection pro- lifetime and had drunk no drinks in past year; and 3) life- cess, 1377 participants aged ≥60 years were included time abstainer: had drunk fewer than 12 drinks in their in the study (Fig.  1). The mean age of the participants lifetime [15]. was 69.1 ± 6.5 years, and 55% were male. A total of 79% Other variables included in the baseline characteristics of the participants were non-Hispanic white, 61% had and subgroup analysis were marital status, statin use, dia- an educational level of college or above, and 32% were betes, and hypertension. Marital status was defined as 1) not married or never married. Thirty-eight percent of married/cohabitating and 2) not married/never married. the participants were obese (BMI ≥28 kg/m2), 19%had Diabetes and hypertension were based on self-report a history of diabetes, and 59% had hypertension. questionnaire data. For the baseline lipid profile, the mean TC level was 4.97 ± 1.07  mmol/L, and the median RC level was 0.54 Statistical analyses (0.39, 0.78) mmol/L. For CERAD test performance, 26% All statistical analyses were conducted by Stata SE 16.0 of the participants had scores in the lowest age-strati- (Stata Corporation, College Station, TX). The complex fied quartile. Details are depicted in Table 1 . Xie et al. Lipids in Health and Disease (2022) 21:120 Page 4 of 10 Table 1 Baseline characteristics of the study population grouped by age‑stratified quartiles of the CERAD total score Groups by age-stratified quartiles of CERAD total score st nd rd th Total (n = 1377) 1 quartile 2 quartile 3 quartile 4 quartile P for trend (n = 458) (n = 381) (n = 305) (n = 233) Age, yrs 69.1 ± 6.5 69.6 ± 7.4 69.3 ± 6.3 68.8 ± 6.3 68.6 ± 6.9 < 0.001 Gender, n(%) Female 705(55) 184(12) 197(14) 178(16) 146(14) < 0.001 Male 672(45) 274(14) 184(13) 127(9.4) 87(7.5) Race, n(%) White 696(79) 180(18) 185(21) 171(21) 160(19) < 0.001 Black 281(8.5) 110(3) 83(2.6) 57(1.8) 31(1) Other races 400(12) 168(4.8) 113(3.6) 77(2.4) 42(1.2) Education level, n(%) Less than high school 356(16) 183(7.5) 110(5.4) 45(2.3) 18(1.1) < 0.001 High school graduate 329(23) 113(6.8) 94(6.8) 72(5.5) 50(3.9) College or above 690(61) 161(11) 176(15) 188(17) 165(16) Marital status, n(%) Married/Cohabiting 855(68) 294(18) 225(19) 190(17) 146(15) < 0.001 Not married/Never married 522(32) 164(8) 156(9.2) 115(8.1) 87(6.3) BMI, n(%) Normal 379(27) 122(6.3) 98(6.6) 88(7.2) 71(7.1) < 0.001 Overweight 465(35) 155(9.3) 128(9.8) 100(8.4) 82(7.1) Obese 533(38) 181(10) 155(11) 117(9.4) 80(7.4) Drinking status, n(%) Lifetime abstainer 215(14) 68(3.3) 69(4) 45(3.4) 33(2.8) < 0.001 Former drinker 368(23) 136(7.4) 109(6.8) 77(5.2) 46(3.7) Current drinker 766(63) 239(15) 195(17) 180(16) 152(15) Smoking status, n(%) Never smoker 682(49) 222(12) 179(13) 149(12) 132(12) < 0.001 Former smoker 527(40) 173(10) 154(12) 122(11) 78(7.2) Current smoker 166(11) 63(3.7) 46(2.7) 34(2.5) 23(1.9) Diabetes, n(%) 302(19) 108(6.3) 99(6.2) 61(3.7) 34(2.5) < 0.001 Hypertension, n(%) 859(59) 290(16) 241(17) 189(14) 139(12) < 0.001 Stroke, n(%) 99(6.7) 35(2) 34(2) 15(1.3) 15(1.4) < 0.001 Statin use, n(%) Statin user 850(60) 276(14) 223(16) 195(15) 156(15) < 0.001 Non statin user 527(40) 182(12) 158(12) 110(9.6) 77(6.5) TC, mmol/L 4.97 ± 1.07 4.77 ± 1.12 4.91 ± 1.02 5.08 ± 1.07 5.17 ± 0.99 < 0.001 LDL-C, mmol/L 2.87 ± 0.93 2.73 ± 1.00 2.81 ± 0.89 2.95 ± 0.91 3.02 ± 0.85 < 0.001 HDL-C, mmol/L 1.47 ± 0.43 1.38 ± 0.45 1.46 ± 0.38 1.54 ± 0.41 1.54 ± 0.42 < 0.001 TG, mmol/L 1.19 (0.85,1.71) 1.38 (0.90,1.75) 1.21 (0.89,1.75) 1.12 (0.78,1.64) 1.10 (0.81,1.68) < 0.001 RC, mmol/L 0.54 (0.39,0.78) 0.63 (0.41,0.81) 0.55 (0.41,0.80) 0.51 (0.36,0.75) 0.50 (0.36,0.77) < 0.001 * a b c Data were shown as mean (± standard deviation), median (interquartile range), unweighted counts (weighted percentage) Analysis of the association among RC, the TC/RC ratio odds ratio (95% CI) was 0.66 (0.46–0.95) for the fourth and the CERAD total score quartile, 0.61 (0.44–0.86) for the third quartile, and 0.89 On a continuous scale, each 1 mmol/L increase in RC was (0.67–1.17) for the second quartile. The P for trend test associated with a multivariable adjusted odds ratio (95% was 0.011 across the quartiles. Details are provided in CI) of 0.74 (0.58–0.94) for having a CERAD total score Table 2. in a higher quartile. On a categorical scale, compared On a continuous scale, each 1 unit increase in the with that of the first quartile, the multivariable adjusted TC/RC ratio was associated with a multivariable Xie  et al. Lipids in Health and Disease (2022) 21:120 Page 5 of 10 Subgroup analysis Table 2 Weighted multivariable ordinal logistic regression analysis between RC levels (log‑transformed) and age ‑stratified Stratified analyses were conducted for the relationship quartiles of the CERAD total score between RC levels, the TC/RC ratio, and age-stratified quartiles of the CERAD total score (Figs.  2  and  3). After RC, mmol/L OR (95%CI) adjusting for confounders, the effect size of the RC level Model 1 Model 2 Model 3 as well as the TC/RC ratio on quartiles of the CERAD total score were almost all consistent across the sub- Continuous, 0.72(0.56–0.93)* 0.74(0.58–0.94)* 0.77(0.61–0.96)* mmol/L groups, including sex, age, education level, BMI status, Categorical (quartiles) diabetes, hypertension, and TC level (P for interaction ≤0.39 mmol/L Ref Ref Ref ≥0.05). Only the race of black was an exception, which 0.39 to ≤ 0.87(0.67–1.13) 0.91(0.68–1.21)* 0.93(0.70–1.25) deserves further study. 0.54 mmol/L 0.54 to ≤ 0.56(0.40–0.80)** 0.62(0.44–0.88)** 0.64(0.47–0.88)** 0.78 mmol/L Discussion > 0.78 mmol/L 0.66(0.44–0.98)* 0.67(0.46–0.98)* 0.70(0.50–0.98)* To our knowledge, this is the first study to explore the P for linear trend 0.040 0.015 0.013 association between plasma RC levels, as well as the TC Model 1 univariate analysis to RC ratio, and cognitive function. The present study Model 2 adjusted for age, gender, education level, and race had the following important findings. First, a higher level Model 3 further adjusted for BMI status, smoking status, and drinking status plus of RC was associated with a higher risk of verbal learn- model 2 ing and memory function impairment. Second, a higher P < 0.05, **P < 0.01 TC to RC ratio was associated with a lower risk of ver- bal learning and memory function impairment. Third, in comparison with RC levels, the TC/RC ratio showed a steadier relationship with verbal learning and memory Table 3 Weighted multivariable ordinal logistic regression function under multiple analytic approaches. Fourth, the analysis between the TC/RC ratio (log‑transformed) and age ‑ effect sizes of RC levels as well as the TC/RC ratio on ver - stratified quartiles of the CERAD total score bal learning and memory function were consistent across TC/RC OR (95% CI) almost all subgroup analyses. RC is defined as the cholesterol composition of rem - Model 1 Model 2 Model 3 nants that are metabolized from TG-rich lipoproteins, Continuous 1.68(1.29–2.19)** 1.46(1.12–1.90)** 1.40(1.10–1.78)** including chylomicron and very low-density lipopro- Categorical (quartiles) tein (VLDL) [17]. Quite a few observational studies have ≤6.08 Ref Ref Ref investigated the relationship between classical lipid com- > 6.08 to ≤8.75 1.12(0.80–1.56) 1.14(0.83–1.56) 1.09(0.79–1.51) ponents, such as TC, LDL-C, and HDL-C, and cogni- > 8.75 to ≤12.94 1.38(1.09–1.76)** 1.42(1.10–1.84)* 1.33(1.03–1.71)* tive function. However, there is no current laboratory > 12.94 1.87(1.25–2.80)** 1.51(1.02–2.22)* 1.41(1.00–1.97)* or epidemiological evidence available concerning the P for linear trend 0.002 0.020 0.028 association between RC and cognitive function and the Model 1 univariate analysis underlying mechanisms. Here, for the first time, this Model 2 adjusted for age, gender, education level, and race study provides a glimpse of the relationship between RC Model 3 further adjusted for BMI status, smoking status, and drinking status plus and cognitive function, suggesting that a lower RC level is model 2 associated with better verbal learning and memory func- P < 0.05, **P < 0.01 tion defined by the CERAD total score. The CERAD WL subtest is part of the CERAD Neuropsychology battery, which was originally designed to permit staging of AD. Utilizing a multiple analytic approach, the study dem- onstrated that a higher level of RC was correlated with adjusted odds ratio (95% CI) of 1.45 (1.13–1.87) for worse verbal learning and memory function in American having a CERAD total score in a higher quartile. On a elderly individuals aged ≥60 years. categorical scale, compared with that of the first quar- Although no relevant study is currently available, the tile, the multivariable adjusted odds ratio (95% CI) was study results have biologically plausible explanations. 1.50 (1.03–2.18) for the fourth quartile, 1.38 (1.07– Remnants with diameters less than 70 nm can penetrate 1.78) for the third quartile, and 1.12 (0.81–1.55) for the the intima of the arterial wall, leading to atherosclerosis second quartile. The P for trend test was 0.020 across [18]. The relationship between RC and atherosclerotic the quartiles. Details are provided in Table 3. Xie et al. Lipids in Health and Disease (2022) 21:120 Page 6 of 10 Fig. 1 Flowchart of the sample selection process of the study population diseases [19], as well as its association with cardiovas- clearance by the liver. Combined with the results of the cular outcomes independent of LDL-C [20], has been present study, disordered APOE-mediated clearance established in previous studies. Due to its atherogenic of remnant lipoprotein might partly participate in the properties, RC can contribute to atherosclerosis in both development of cognitive impairment. Further research the carotid artery [21] and arterioles in the brain. A should be conducted to determine the underlying Mendelian randomization study confirmed the causal mechanisms and their clinical significance. relationship of remnant lipoprotein-associated genes The present study proposed a new blood lipid index, and ischemic stroke [22]. Prospective cohort studies the TC/RC ratio. The adverse effect of high TC levels on symptomatic intracranial atherosclerotic stenosis on cognitive function has been abundantly investi- and ischemic stroke indicated that RC causes cerebral gated in previous studies. However, serum TC includes hypoperfusion [23]. In addition, population-based stud- cholesterol molecules from a variety of subtypes of ies showed that a reduction in cerebral perfusion was lipoprotein, and lowering cholesterol of different lipo - associated with an increased risk of dementia [24, 25]. protein subtypes might produce differential outcomes. Previous studies also reported that RC enhanced oxi- Therefore, the beneficial effect of cholesterol-lowering dative stress and proinflammatory effects on vascular therapy on cognitive function is still controversial, endothelial and smooth muscle cells [26], which might especially in elderly individuals [29]. Based on previ- damage the blood‒brain barrier, subsequently altering ous studies and the relationship between RC levels and amyloid degradation and cholesterol homeostasis [27] the CERAD total score found in the present study, it is in the brain. A recent Mendelian randomization study supposed that TC, in combination with RC, might be a on the risk factors for AD found that genetically ele- better bioindex for the prediction of cognitive function vated TC and LDL-C levels increased neurotic plaque than TC or RC separately. In light of this, the present burden, but the effects were driven by single nucleotide study examined and demonstrated that a higher TC/RC polymorphisms of APOE [28], whose genetic product ratio is suggestive of better verbal learning and memory is known to be the key ligand for remnant lipoprotein function assessed by CERAD tests. Xie  et al. Lipids in Health and Disease (2022) 21:120 Page 7 of 10 Fig. 2 Eec ff t size of RC levels on the age ‑stratified quartiles of the CERAD total score in subgroups. Notes: Model adjusted for age, sex, educational level, race, BMI status, smoking status, and drinking status In addition, the present study found that the TC/RC by previous studies. For future clinical application, the ratio showed a significant positive association not only present study provided evidence for the utilization of with the CERAD total score but also with the CERAD RC levels and the TC/RC ratio in the evaluation of ver- delayed trial test score [see Additional file  1, Table  S1], bal learning and memory function, which might assist while RC levels showed no association with delayed trial in risk stratification for cognitive function impairment test score [see Additional file  1, Table S2]. Previous stud- or AD susceptibility in elderly individuals. Consider- ies have found that some of the measures in the CERAD ing both cardiovascular and cognitive benefits implied WL subtest, particularly delayed recall of a word list, by the present study, the lower the RC level, the better. could more efficiently distinguish persons with demen - Regarding the benefit of a higher TC/RC ratio for verbal tia from those with normal cognition [30]. Therefore, learning and memory function, there might be a poten- the results of the present study indicated that the TC/RC tial atherogenic risk when the higher ratio is mainly ratio might have better value for predicting verbal learn- attributed to a relatively high TC level rather than a ing and memory function than RC levels. low RC level. However, in the context of this study, the mean TC level was 4.97 ± 1.07 mmol/L, which was not Comparisons with other studies and what the current work a significant atherogenic level. adds to the existing knowledge Conclusively, the current study examined two available Study strengths and limitations blood lipid indices for the assessment of verbal learning There are several strengths in the present study. First, and memory function, which have not been reported the study observed a known cardiovascular risk factor, Xie et al. Lipids in Health and Disease (2022) 21:120 Page 8 of 10 Fig. 3 Eec ff t size of the TC/RC ratio on the age ‑stratified quartiles of the CERAD total score in subgroups. Notes: Model adjusted for age, sex, educational level, race, BMI status, smoking status, and drinking status RC, and its negative relationship with verbal learning function. Second, in this study, LDL-C was calcu- and memory function. Moreover, the study utilized lated based on the Friedewald equation, which is not data extracted from the NHANES database, which applicable when TG levels > 400  mg/dl. Therefore, used complex, multistage sampling, and relatively the results of the study should not be interpreted in convincing results could be drawn with a proper ana- the situation of very high TG levels. Third, the pre- lytic approach. Additionally, the study proposed an sent study used calculated fasting RC to represent the RC derivative, the TC/RC ratio, and found its positive “remnant” cholesterol level; however, this calculated association with verbal learning and memory function, fasting RC includes not only cholesterol from rem- which had not been mentioned previously. Addition- nants but also cholesterol from newly formed VLDL ally, the results were estimated in several subpopula- particles, which in fact overestimates the cholesterol tions, verifying its authenticity. levels of actual remnants. However, RC from the indi- The study has several limitations. First, this was rect formula has been widely used in numerous clini- an observational study with a cross-sectional design, cal studies and has been proven to be a convenient and and the causal relationship could not be determined reliable risk predictor. Finally, although the present between RC levels or the TC/RC ratio and cognitive study included many important potential covariates Xie  et al. Lipids in Health and Disease (2022) 21:120 Page 9 of 10 Declarations previously reported to affect cognition in the statisti- cal models, the possibility that residual confounding Ethics approval and consent to participate factors remain could not be ruled out. Future large- The NHANES protocols were approved by the National Center for Health Statistics Ethics Review Board of the US CDC, and written informed consent scale, prespecified trials are needed to further explore from all the participants was provided during the survey. this subject. Consent for publication Not applicable. Conclusions Competing interests Utilizing a cohort of American elderly individuals aged The authors declare that they have no competing interests. over 60  years from the NHANES database, the pre- Author details sent study found that in the context of a TG level below Cardio‑Metabolic Medicine Center, National Center for Cardiovascular 400  mg/dl, a lower RC level and a higher TC/RC ratio Diseases, Fuwai Hospital, Chinese Academy of Medical Sciences, Peking Union Medical College, Beijing, China. Coronary Heart Disease Center, were associated with better verbal learning and memory National Center for Cardiovascular Diseases, Fuwai Hospital, Chinese Academy function. The present study indicated that lowering RC of Medical Sciences, Peking Union Medical College, Beijing, China. Division levels or increasing the TC/RC ratio when the TG level of Cardiology, The First Affiliated Hospital of Soochow University, Suzhou, Jiangsu Province, China. is below 400  mg/dl could possibly be beneficial for pre - venting cognitive impairment in elderly individuals. The Received: 3 October 2022 Accepted: 1 November 2022 study results, which need to be verified in future larger cohort studies, might help in guiding risk prediction and primary prevention of cognitive impairment in elderly individuals. References 1. GBD 2019 Dementia Forecasting Collaborators. Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: Abbreviations an analysis for the Global Burden of Disease Study 2019. Lancet Public RC: Remnant cholesterol; TC/RC: Total cholesterol/remnant cholesterol; CERAD: Health. 2022;7(2):c105–25 Consortium to Establish a Registry for Alzheimer’s Disease Word List Memory 2. Liu Y, Zhong X, Shen J, Jiao L, Tong J, Zhao W, et al. Elevated serum TC Task; NHANES: National Health and Nutrition Examination Survey; TC: Total and LDL‑ C levels in Alzheimer’s disease and mild cognitive impairment: A cholesterol; LDL: Low‑ density lipoprotein; HDL: High‑ density lipoprotein; meta‑analysis study. Brain Res. 2020;1727. LDL‑ C: Low‑ density lipoprotein cholesterol; HDL‑ C: High‑ density lipoprotein 3. Anstey KJ, Ashby‑Mitchell K, Peters R. Updating the Evidence on the cholesterol; TG: Triglycerides; VLDL: Very low‑ density lipoprotein; BMI: Body Association between Serum Cholesterol and Risk of Late‑Life Dementia: mass index. Review and Meta‑Analysis. J Alzheimers Dis. 2017;56(1):215–28. 4. Ding D, Zhou F, Cao Y, Liang X, Wu W, Xiao Z, et al. Cholesterol profiles and incident cognitive decline among older adults: the Shanghai Aging Supplementary Information Study. Age Ageing. 2021;50(2):472–9. The online version contains supplementary material available at https:// doi. 5. Benn M, Nordestgaard BG, Frikke‑Schmidt R, Tybjærg‑Hansen A. Low LDL org/ 10. 1186/ s12944‑ 022‑ 01729‑4. cholesterol, PCSK9 and HMGCR genetic variation, and risk of Alzheimer’s disease and Parkinson’s disease: Mendelian randomisation study. Bmj. 2017;357:1648. Additional file 1: Table S1. Weighted multivariable ordinal logistic regres‑ 6. Bruemmer D, Cho L. Remnant Cholesterol: The Leftovers and Their sion analysis between RC levels (log‑transformed) and age ‑stratified quar ‑ Contribution to Atherosclerotic Cardiovascular Disease. Circ Cardiovasc tiles of the CERAD delayed score. Table S2. Weighted multivariable ordinal Imaging. 2021;14(4). logistic regression analysis between the TC/RC ratio (log‑transformed) and 7. Quispe R, Martin SS, Michos ED, Lamba I, Blumenthal RS, Saeed A, et al. age‑stratified quartiles of the CERAD delayed score. Remnant cholesterol predicts cardiovascular disease beyond LDL and ApoB: a primary prevention study. Eur Heart J. 2021;42(42):4324–32. 8. Varbo A, Benn M, Tybjærg‑Hansen A, Jørgensen AB, Frikke ‑Schmidt R, Acknowledgements Nordestgaard BG. Remnant cholesterol as a causal risk factor for ischemic Not applicable. heart disease. J Am Coll Cardiol. 2013;61(4):427–36. 9. Burnett JR, Hooper AJ, Hegele RA. Remnant Cholesterol and Atheroscle‑ Authors’ contributions rotic Cardiovascular Disease Risk. J Am Coll Cardiol. 2020;76(23):2736–9. Ying‑ Yi Xie: Methodology implementation, Formal analysis, Writing – original 10. Paquette M, Bernard S, Paré G, Baass A. Dysbetalipoproteinemia: Differ ‑ draft, Writing – review & editing. Liang Zhao: Validation. Li‑ Jian Gao: Validation. entiating Multifactorial Remnant Cholesterol Disease From Genetic ApoE Rui‑ Xia Xu: Validation. Ying Gao: Validation. Ke‑Fei Dou: Validation. Yuan‑Lin Deficiency. J Clin Endocrinol Metab. 2022;107(2):538–48. Guo: Conceptualization, Methodology guidance, Project administration, 11. National Health and Nutrition Examination Survey. National Center for Validation, Writing – review & editing. Yong‑Ming He: Methodology guidance, Health Statistics (NCHS), Hyattsville. 2011–2014. https:// wwwn. cdc. gov/ Project administration, Validation, Writing – review & editing. The author(s) Nchs/ Nhanes/. Accessed 12 Mar 2022. read and approved the final manuscript. 12. Dong X, Li S, Sun J, Li Y, Zhang D. Association of Coffee, Decaffeinated Coffee and Caffeine Intake from Coffee with Cognitive Performance Funding in Older Adults: National Health and Nutrition Examination Survey This study was supported by the Chinese Academy of Medical Science Inno‑ (NHANES) 2011–2014. Nutrients. 2020;12(3):840. vation Fund for Medical Sciences (2021‑I2M ‑1–009). 13. Li S, Sun W, Zhang D. Association of Zinc, Iron, Copper, and Selenium Intakes with Low Cognitive Performance in Older Adults: A Cross‑ Availability of data and materials Sectional Study from National Health and Nutrition Examination Survey The datasets analyzed during the current study are available on the NHANES (NHANES). J Alzheimers Dis. 2019;72(4):1145–57. official website, https:// wwwn. cdc. gov/ Nchs/ Nhanes/. Xie et al. Lipids in Health and Disease (2022) 21:120 Page 10 of 10 14. Cantini L, Mentrasti G, Russo GL, Signorelli D, Pasello G, Rijavec E, et al. Evaluation of COVID‑19 impact on DELAYing diagnostic‑therapeutic pathways of lung cancer patients in Italy (COVID‑DELAY study): fewer cases and higher stages from a real‑ world scenario. ESMO Open. 2022;7(2). 15. National Center for Health Statistics (NCHS). Adult Alcohol Use Informa‑ tion: Glossary ‑ Alcohol. 2018. Available from: https:// www. cdc. gov/ nchs/ nhis/ alcoh ol/ alcoh ol_ gloss ary. htm. Accessed 12 Mar 2022. 16. National Center for Health Statistics (NCHS). Module 3: Weighting. National Health and Nutrition Examination Survey: Tutorials. 2021. Avail‑ able from: https:// wwwn. cdc. gov/ nchs/ nhanes/ tutor ials/ modul e3. aspx. Accessed 12 Mar 2022. 17. Varbo A, Benn M, Nordestgaard BG. Remnant cholesterol as a cause of ischemic heart disease: evidence, definition, measurement, atherogenic‑ ity, high risk patients, and present and future treatment. Pharmacol Ther. 2014;141(3):358–67. 18. Nordestgaard BG. Triglyceride‑Rich Lipoproteins and Atherosclerotic Cardiovascular Disease: New Insights From Epidemiology, Genetics, and Biology. Circ Res. 2016;118(4):547–63. 19. Lin A, Nerlekar N, Rajagopalan A, Yuvaraj J, Modi R, Mirzaee S, et al. Rem‑ nant cholesterol and coronary atherosclerotic plaque burden assessed by computed tomography coronary angiography. Atherosclerosis. 2019;284:24–30. 20. Castañer O, Pintó X, Subirana I, Amor AJ, Ros E, Hernáez Á, et al. Remnant Cholesterol, Not LDL Cholesterol, Is Associated With Incident Cardiovas‑ cular Disease. J Am Coll Cardiol. 2020;76(23):2725–35. 21. Qian S, You S, Sun Y, Wu Q, Wang X, Tang W, et al. Remnant Cholesterol and Common Carotid Artery Intima‑Media Thickness in Patients With Ischemic Stroke. Circ Cardiovasc Imaging. 2021;14(4). 22. Si S, Hou L, Chen X, Li W, Liu X, Liu C, et al. Exploring the Causal Roles of Circulating Remnant Lipid Profile on Cardiovascular and Cerebro ‑ vascular Diseases: Mendelian Randomization Study. J Epidemiol. 2022;32(5):205–14. 23. Kim DE, Kim JY, Jeong SW, Cho YJ, Park JM, Lee JH, et al. Association between changes in lipid profiles and progression of symptomatic intrac‑ ranial atherosclerotic stenosis: a prospective multicenter study. Stroke. 2012;43(7):1824–30. 24. Wolters FJ, Zonneveld HI, Hofman A, van der Lugt A, Koudstaal PJ, Vernooij MW, et al. Cerebral Perfusion and the Risk of Dementia: A Population‑Based Study. Circulation. 2017;136(8):719–28. 25. Wolters FJ, de Bruijn RF, Hofman A, Koudstaal PJ, Ikram MA. Cerebral Vasoreactivity, Apolipoprotein E, and the Risk of Dementia: A Population‑ Based Study. Arterioscler Thromb Vasc Biol. 2016;36(1):204–10. 26. Nakajima K, Tanaka A. Atherogenic postprandial remnant lipoproteins; VLDL remnants as a causal factor in atherosclerosis. Clin Chim Acta. 2018;478:200–15. 27. Saeed AA, Genové G, Li T, Lütjohann D, Olin M, Mast N, et al. Eec ff ts of a disrupted blood‑brain barrier on cholesterol homeostasis in the brain. J Biol Chem. 2014;289(34):23712–22. 28. Andrews SJ, Fulton‑Howard B, O’Reilly P, Marcora E, Goate AM. Causal Associations Between Modifiable Risk Factors and the Alzheimer’s Phe ‑ nome. Ann Neurol. 2021;89(1):54–65. 29. Power MC, Rawlings A, Sharrett AR, Bandeen‑Roche K, Coresh J, Ballan‑ tyne CM, et al. Association of midlife lipids with 20‑ year cognitive change: A cohort study. Alzheimers Dement. 2018;14(2):167–77. 30. Fillenbaum GG, van Belle G, Morris JC, Mohs RC, Mirra SS, Davis PC, et al. Consortium to Establish a Registry for Alzheimer’s Disease (CERAD): the Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : first twenty years. Alzheimers Dement. 2008;4(2):96–109. fast, convenient online submission Publisher’s Note thorough peer review by experienced researchers in your field Springer Nature remains neutral with regard to jurisdictional claims in pub‑ rapid publication on acceptance lished maps and institutional affiliations. support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

Lipids in Health and Disease – Springer Journals

Published: Nov 14, 2022

Keywords: Remnant cholesterol; Total cholesterol/remnant cholesterol; Cognition; Verbal learning and memory function; Elderly; Cross-sectional study

There are no references for this article.