Ideal vitamin D and handgrip strength counteracts the risk effect of APOE genotype on dementia: a population-based longitudinal study

The data were derived from UK Biobank, a population-based cohort that recruited 502,412 participants aged 37–73 years who attended one of 22 assessment centers across the United Kingdom between 2006–2010 and were followed up until 2021 [19]. Participants provided completed touch-screen questionnaires, physical examination, and biological samples. We excluded participants with prevalent dementia or less than 60 years old or missing information on genetic factors, vitamin D, and grip strength. Data from 165,688 participants were available for analyses in the present study (Fig. 1). All the UK Biobank participants gave written informed consent before data collection.

Covariates were documented, including age, ethnicity (White, Black, Asian, and other), education level (upper secondary, lower secondary, vocational, and other), socioeconomic status (defined based on the Townsend deprivation index, encompassing information on social class, employment, car availability, and housing), alcohol consumption (never, former, and current), smoking status (never, former, and current), physical activity (active and inactive), diet (healthy and unhealthy), BMI (< 25 kg/m², 25 to 30 kg/m², and ≥ 30 kg/m²), heart disease (no, yes), stroke (no, yes), hypertension (no, yes), diabetes (no, yes), depression (no, yes; including depression symptoms, self-reported depression, and inpatient depression), cholesterol, high density lipoprotein cholesterol (HDL), and low density lipoprotein cholesterol (LDL). Regular physical activity was defined as engaging in moderate activity ≥ 150 min/week, vigorous activity ≥ 75 min/week, or moderate and vigorous activity ≥ 150 min/week [20]. A healthy diet was based on adequate intake of at least three of these five commonly eaten food groups (Vegetables ≥ 3 servings/day; Fruits ≥ 3 servings/day; Unprocessed red meats ≤ 1.5 servings/week; Processed meats ≤ 1 serving/week; Fish ≥ 2 servings/week) [21]. Depression symptoms were measured by 4 items from the Patient Health Questionnaire [22]: depressed mood, unenthusiasm/disinterest, tenseness/restlessness, and tiredness/lethargy.

All-cause dementia was ascertained based on a self-reported diagnosis of dementia (Data-Field 20,002, code: 1263), hospital records (ICD-10 codes: F00, F01, F02, F03, F05.1, G30, G31.1, G31.8), and death records (ICD-10 codes: F00, F01, F02, F03, F05.1, G30, G31.1, G31.8) from the Hospital Episode Statistics (England), the Scottish Morbidity Record (Scotland), and the Patient Episode Database (Wales).

Two single nucleotide polymorphism loci (SNPs), rs429358 and rs7412, determining the possible APOE isoforms, and forming three haplotypes (e2, e3, and e4) and six genotypes (e2/e2, e2/e3, e2/e4, e3/e3, e3/e4, e4/e4). APOE genotype was coded as APOE e4 non-carries (APOE e4−, low genetic risk) and APOE e4 carries (APOE e4+, high genetic risk).

Information on biochemistry makers were obtained from biological samples collected at study recruitment (2006–2010) [23, 24]. Serum 25(OH)D concentrations, a measure of vitamin D status, was measured by chemiluminescence technology analysis on a DiaSorin Ltd. LIASON XL [25]. Vitamin D was imputed with the minimum detectable value (10 nmol/L) if it was below the limit of detection [26], and the maximum detectable value (125 nmol/L) if too high for detection (< 0.1%). Vitamin D was categorized into three groups (low, middle, and high) according to tertiles.

Grip strength was measured using a Jamar J00105 hydraulic hand dynamometer. Isometric grip force was assessed from a single 3-s maximal grip effort of the right- and left-side arms with participants seated upright with their elbow by their side and flexed at 90° so that their forearm was facing forward and resting on an armrest. The mean of the right and left values was expressed in absolute units (kilograms), as reported elsewhere and was used in the current study. Due to biological differences in grip strength within sex, we divided participants into three groups according to sex-specific categories as low (the lowest tertile), middle (the middle tertile), and high (the highest tertile).

We calculated incidence rates and 95% confidence intervals (95% CIs) per 1000 person-years for dementia among women and men. Multivariable restricted cubic regression splines were used to visual evaluate the relationship between vitamin D, grip strength and incidence of dementia, with four knots at the 25th, 50th, 75th, and 95th centiles. The P for overall model < 0.05 and P for non-linear < 0.05 indicated a non-linear relationship between vitamin D or grip strength and dementia, while P for overall model < 0.05 and P for non-linear > 0.05 showed a linear relationship. The cubic regression splines models were adjusted for age, ethnicity, education level, socioeconomic status, alcohol consumption, smoking status, BMI, heart disease, stroke, hypertension, diabetes, depression, TC, HDL, LDL, APOE e4 genotype, and adjusted for vitamin D in grip strength analysis or for grip strength in vitamin D analysis. Association of vitamin D or grip strength categories (tertiles 1 as reference) and prospective dementia incidence were investigated by using Cox proportional hazards regression models with follow-up year as the time scale. Hazard ratios (HRs) and 95% confidence interval (CIs) were used to reported the results. Follow-up time was calculated as the time from baseline assessment until the first event of dementia, death, or March 31, 2021, whichever occurred first. Schoenfeld residuals method was used to check the proportional hazards assumptions. The Kaplan–Meier survival curve were applied to assess the association between joint exposure of vitamin D and grip strength categories and APOE e4 genotype with risk of incident dementia. To examine whether APOE e4 genotype modified the associations of vitamin D and/or grip strength with the risk of dementia, additive and multiplicative interactions were assessed by fitting the relevant parameters into the models. All models were adjusted for age, ethnicity, education level, socioeconomic status, alcohol consumption, smoking status, BMI, heart disease, stroke, hypertension, diabetes, depression, TC, HDL, LDL, APOE e4 genotype, and adjusted for vitamin D in grip strength analysis or for grip strength in vitamin D analysis. If data were missing for a covariate, we used multiple imputations based on five replications and utilized a chained-equation method to account for the missing data [27].

Several additional analyses were performed to assess the robustness of our study results. First, we used stratification analysis to examine whether the association between dementia and the combined association of vitamin D/grip strength and APOE e4 genotype varied by age (< 60 vs. ≥ 60 years), ethnic background, education level, and socioeconomic status. Second, to address the role of potential reverse causality, we repeated the main analyses by excluding participants who developed dementia in the first 3-year follow-up period and participants who died within 3 years from baseline. Furthermore, we excluded participants with major chronic disease, such as heart disease, stroke, diabetes, or cancer at baseline to assess the robustness of our study results.

Of the 165,688 dementia-free participants, the mean age was 64.1 ± 2.9 years, and 81,078 (48.9%) were men. Over the follow-up (median: 12.0 years, IQR: 11.3 to 12.6), 3917 participants developed dementia. The baseline characteristics of the participants by incident dementia status among women and men are provided in Table 1. In women and men, respectively, participants who developed dementia were more likely to be older, excessive alcohol consumption, former or current smokers, obesity, APOE e4 carriers, had low education attainment and socioeconomic status, had high prevalence of heart disease, stroke, hypertension, diabetes, and depression, and less likely to have higher level of vitamin D and handgrip strength comparted to those who did not develop dementia. Additionally, there was a correlation between vitamin D and handgrip strength in both women (r = 0.011, P = 0.0018) and men (r = 0.051, P < 0.001). The scatter plot of the relationship between handgrip strength and vitamin D was shown in Additional file 1: Figure S1.

Restricted Cubic Spline models were used to evaluate the relationship between vitamin D and handgrip strength with dementia risk. In multi-adjusted models (Fig. 2), handgrip strength was negatively associated with incident dementia. The associations between handgrip strength and dementia were linear in women (P for overall model < 0.001 and P for non-linear = 0.340) and men (P for overall model < 0.001 and P for non-linear = 0.969). Per 1-SD increment in handgrip strength was associated with a 3% (HR = 0.97, 95% CI 0.96–0.98) and a 3% (HR = 0.97, 95% CI 0.97–0.98) lower risk of incident dementia among women and men, respectively. In addition, the association between vitamin D and dementia was linear among women (P for overall model < 0.001 and P for non-linear = 0.167), but the U shape relationship between vitamin D and dementia was found among men (P for overall model < 0.001 and P for non-linear = 0.013).

When the study categories handgrip strength into three groups according to tertiles, in women and men, respectively, compared with the lowest tertile of handgrip strength, the HRs (95% CIs) of dementia were lower in the middle [0.78 (0.70–0.87)/0.75 (0.68–0.82)] and the highest tertile [0.64 (0.56–0.75)/0.60 (0.52–0.68)]. Tertiles of vitamin D showed a similar pattern. In multi-adjusted Cox models, HRs and 95% CIs of dementia were 0.86 (0.76–0.97)/0.80 (0.72–0.90) for participants with middle vitamin D status and 0.81 (0.72–0.90)/0.73 (0.66–0.81) for those with high vitamin D status, compared to those with low vitamin D status (Table 2).

Figure 3 shows the association between dementia and the joint exposures of handgrip strength, vitamin D, and APOE e4 genotype. After fully-adjusted for confounding factors, compared to participants with APOE e4 non-carries and low handgrip strength, the HRs (95% CIs) of dementia were 3.65 (3.22–4.15) and 2.84 (2.54–3.19) for those with APOE e4 carries plus low handgrip strength among women and men, respectively, and 2.45 (2.01–2.99) and 1.67 (1.39–2.01) for those with APOE e4 carries and high handgrip strength among women and men, respectively. Similar, compared to participants with APOE e4 non-carries and low vitamin D status, the HRs (95% CIs) of dementia were 4.21 (3.62–4.90) and 2.54 (2.22–2.92) for those with APOE e4 carries plus low vitamin D status, and 3.23 (2.75–3.79) and 2.08 (1.82–2.39) for those with APOE e4 carries and high vitamin D status among women and men, respectively.

Furthermore, there was a significant additive interaction between lowest tertile of both handgrip strength and vitamin D and APOE e4 carries that associated with dementia among women (RERI: 2.868, 95% CI 1.048–4.688; AP: 0.352, 95% CI 0.169–0.537; SI: 1.671, 95% CI 1.175–2.375) and men (RERI: 2.020, 95% CI 0.460–3.581; AP: 0.262, 95% CI 0.085–0.439; SI: 1.431, 95% CI 1.069–1.915. In joint effect analysis, compared to participants with both low handgrip strength/low vitamin D and APOE e4 non-carries, the HRs (95% CIs) of dementia were 4.29 (3.52–5.22) and 2.62 (2.2.20–3.14) in those with APOE e4 carries plus low handgrip strength/low vitamin D profiles among women and men, respectively, and 2.41 (1.77–2.33) and 1.26 (0.94–1.68) in those with APOE e4 carries and both high handgrip strength and vitamin D profiles among women and men, respectively. In women and men, respectively, participants who had both high vitamin D and grip strength was associated with lower risk of dementia compared to those with both low vitamin D and grip strength among APOE e4 genotype carries (HR = 0.56, 95% CI 0.42–0.76, and HR = 0.48, 95% CI 0.36–0.64) and APOE e4 genotype non-carries (HR = 0.56, 95% CI 0.38–0.81, and HR = 0.34, 95% CI 0.24–0.47).

The Kaplan–Meier survival curves showed that the risk of incident dementia was highest for APOE e4 carries with both lowest tertile of handgrip strength and vitamin D profiles among women and men (Fig. 4A, B). Of the APOE e4 carries with both low vitamin D and grip strength, 6.4% (95% CI 5.9–7.0%) incident dementia compared to 2.4% (95% CI 2.1–2.7%) of APOE e4 carries with both high vitamin D and grip strength (Fig. 4C, D).

Association between dementia and the combined association of vitamin D/grip strength and APOE ε4 genotype did not meaningfully differ by age (Additional file 1: Table S1), education level (Additional file 1: Table S2), and socioeconomic status (Additional file 1: Table S3; all P for interaction > 0.05). The results were not much altered compared with those from initial analyses when we repeated analyses by excluding participants who developed dementia in the first 3-year follow-up period and participants who died within 3 years from baseline (Additional file 1: Table S4), or by excluding participants with major chronic disease, such as heart disease, stroke, diabetes, or cancer at baseline (Additional file 1: Table S5). Additionally, we estimated the association between vitamin D and the incidence of dementia stratified by APOE e4 genotype. In women, participants who had highest tertile of vitamin D was associated with lower risk of dementia compared to those with lowest tertile of vitamin D among APOE e4 carries (HR = 0.73; 95% CI 0.63–0.85), but not among APOE e4 non-carries (HR = 0.91; 95% CI 076–1.08). In men, compared to participants with lowest tertile of vitamin D, those with the highest tertile of vitamin D had a lower risk of dementia in both APOE e4 non-carries (HR = 0.65, 95% CI 0.56–0.75) and APOE e4 carries (HR = 0.80, 95% CI 0.69–0.92) (Additional file 1: Table S6).