Introduction
Inflammation, an essential risk factor of neurodevelopmental impairment during early to middle childhood,1 played a key role in various chronic diseases, such as cardiovascular diseases and type 2 diabetes mellitus.2 3 Previous study pointed out that nutrition had the potential value of nutritional therapy by exerting crucial anti-inflammatory effects.4 Therefore, the exploration of novel modified nutritional factors for controlling systematic inflammation would be of high priority.
25-hydroxyvitamin D (25(OH)D), the main form of circulating vitamin D, is characteristics of long half-life and relatively stable.5 It has been considered as the optimal indicator to reflect the nutritional status of circulating vitamin D in the body. Sîrbe et al indicated that 25(OH)D exerted pleiotropic effects on inflammation and related diseases.6 Hypersensitivity C reactive protein (hsCRP) was considered as a representative biomarker of systemic inflammation, which might be modulated by 25(OH)D.7 Prior studies aimed at exploring the associations of 25(OH)D or vitamin D supplementation with hsCRP have produced inconsistent results, with inverse8–10 and null associations11 in different adults. Vitamin D supplementation is the only source of circulating vitamin D for children, and the body metabolism level of children is significantly different from that of adults. Nevertheless, only few data assessed the association between 25(OH)D and hsCRP in children (positive12 and inverse13–15 associations). Additionally, red blood cell distribution width–platelet count ratio (RDWPCR), mean platelet volume–platelet count ratio (MPVPCR), neutrophil–lymphocyte ratio (NLR) and white blood cell–neutrophil ratios (WBCNR) were widely known as the haematological biomarkers of systemic inflammation. The stability, sensitivity and specificity of these comprehensive indices were better than those of a single indicator.16–21 The emerging evidences demonstrated the key roles of these novel inflammatory markers in some chronic diseases (eg, cardiovascular disease,22 hypertension,23 decompensation24 and type 2 diabetes nephropathy25). Nonetheless, no study has yet evaluated the associations of 25(OH)D with RDWPCR, MPVPCR, NLR and WBCNR.
To address the above questions, this hospital-based cross-sectional study examined the associations of 25(OH)D with hsCRP and four novel inflammatory markers (RDWPCR, MPVPCR, NLR and WBCNR) in children.
Materials and methods
Study participants
We conducted a hospital-based cross-sectional study. The medical records of 10 141 children with inflammatory marker measurements (The age range was 1–53 months, mean age 14.6 months) were analysed between January and December 2021 from the department of Child Healthcare, Nantong Maternal and Child Healthcare Hospital. Children aged 1 month to <5 years were included in this study, regardless of gender. Subjects were excluded according to the following predefined criteria: (1) suffering from mental disorders, cognitive impairment, blindness, deafness and other diseases; (2) receiving or having previously received treatment for major illnesses, such as radiotherapy and chemotherapy; (3) having a history of severe congenital diseases, such as congenital heart disease, congenital biliary atresia and congenital pulmonary dysplasia.
Epidemiological data and laboratory measurements
Sex, age, weight, height, vitamin D intake (eg, 400, 600 or 800 IU/day), feeding patterns (breastfeeding, artificial feeding or mixed feeding) and complementary feeding (yes or no) were collected. Height and weight were measured using an electronic column scale or electronic baby scale (Saikang Medical Measurement System (Pinghu), China) by medical workers in accordance with the standard protocol, and then all readings were accurate to 1 cm and 0.01 kg, respectively. Height and weight were measured twice, and the interval between two measurements were less than 2 cm and 0.20 kg, respectively.
Serum 25(OH)D concentration was measured using ELISA. Briefly, 10 µL serum sample were mixed with 200 µL biotin. The mixture was vortexed for 30 s (200–400 rpm), incubated at room temperature for 60 min, washed the plate with washing solution (five times) and then added 200 µL enzyme conjugates. Keep for 30 min at room temperature. The board was washed, added 100 µL substrate and sealed well. After stand at room temperature for 15 min, 100 µL termination liquid was added into the mixture, and then slightly oscillate for 20–30 s. The Rayto RT-6100 ELISA analyzer was applied to assess the serum concentration of 25(OH)D. Red blood cell distribution width, platelet count, mean platelet volume, neutrophil, lymphocyte and white blood cell were measured using automatic haematology analyser BC-7500CRP (Mindray, Shenzhen, China). The RDWPCR, MPVPCR, NLR and WBCNR were calculated.
Statistical analysis
Means (±SD) and median (IQR) were used to present normally distributed and skewed continuous variables, respectively. All continuous variables were examined for normality distribution by Kolmogorov-Smirnov test. Continuous variables were rank transformed. The original data and the rank transformation of data were included in analysis, respectively. Categorical variables were presented as the frequencies (per cent). The t-test and χ2 test were used to analyse continuous and categorical variables, respectively. Partial correlation was performed to examine the correlations between serum 25(OH)D and vitamin D intake and inflammatory markers (hsCRP, RDWPCR, MPVPCR, NLR and WBCNR). Analysis of covariance was performed to compare the mean differences in the concentrations of inflammatory markers (eg, hsCRP, RDWPCR, MPVPCR, NLR and WBCNR) among sex-stratified quartiles classified by serum 25(OH)D. Linear trends in continuous variables across quartiles of serum25(OH)D were assessed with a generalised linear model. Moreover, logistic regression model was performed to determine the associations between serum 25(OH)D and elevated inflammatory biomarkers levels. Elevated inflammatory biomarkers levels were determined by hsCRP≥1.0mg/L26 and inflammatory biomarkers (arbitrarily defined as top quartile by sex). Model 1 was univariate analysis without adjustment; and model 2 adjusted for potential covariates, including sex, height, weight, ages, vitamin D intake, feeding patterns and complementary feeding. Stratified analyses were conducted in boys and girls, respectively. All analyses were conducted using SPSS V.21.0 software package (IBM SPSS Statistics, Armonk, New York, USA). Two-sided p values<0.05 were considered statistically significant.
Patient and public involvement
Patients were not involved in the design of this study.
Results
Characteristics of subjects
The subjects were divided into quartiles according to serum 25(OH)D (n=10 141) level (table 1). The means according to the quartiles of 25(OH)D were as follows: 46.46 ng/mL (quartile (Q) 1 group), 71.75 ng/mL (Q2 group), 91.55 ng/mL (Q3 group) and 117.47 ng/mL (Q4 group). The subjects with higher 25(OH)D concentration had an older age, higher height and weight.
Baseline characteristics of children by quartiles of serum 25(OH)D
Correlation of serum 25(OH)D and vitamin D intake with hsCRP and novel inflammatory biomarkers
Table 2 showed an inverse correlation of serum 25(OH)D and vitamin D intake with hsCRP and novel inflammatory biomarkers in partial correlation analyses adjusted for sex and age. Mean partial correlation coefficients of 25(OH)D and vitamin D intake were −0.564 and −0.041 for hsCRP, −0.541 and −0.043 for RDWPCR, −0.539 and −0.043 for NLR and −0.223 and −0.041 for WBCNR, respectively. Besides, only 25(OH)D was inversely correlated with MPVPCR (r: −0.080), whereas null correlation between vitamin D intake and MPVPCR was observed.
Relationships of serum novel inflammatory biomarkers with 25(OH)D and vitamin D intake*
Associations of serum 25(OH)D with hsCRP and novel inflammatory biomarkers
As shown in table 3 and online supplemental table 1, serum 25(OH)D was inversely associated with hsCRP and novel inflammatory biomarkers in both univariable and multivariable analysis according to the rank transformation of the data. After adjustment for the potential covariates, serum 25(OH)D was inversely associated with hsCRP and novel inflammatory biomarkers (Q4 vs Q1: 1129.75 vs 2090.99 for hsCRP; 4246.94 vs 6829.89 for RDWPCR; 4863.57 vs 5545.66 for MPVPCR; 4345.76 vs 6507.46 for NLR; 2418.84 vs 2868.39 for WBCNR). In the stratified analyses by sex (boys and girls), respectively. In boys, serum 25(OH)D was inversely associated with hsCRP and novel inflammatory biomarkers (Q4 vs Q1: 1146.23 vs 2098.25 for hsCRP; 4495.58 vs 7105.08 for RDWPCR; 5085.07 vs 5680.90 for MPVPCR; 4277.01 vs 6537.81 for NLR; 2498.13 vs 2983.22 for WBCNR). Similar inverse associations were also observed in girls (Q4 vs Q1: 1088.60 vs 2079.05 for hsCRP; 3938.40 vs 6587.51 for RDWPCR; 4564.79 vs 5407.63 for MPVPCR; 4355.82 vs 6554.33 for NLR; 2344.05 vs 2731.31 for WBCNR).
Supplemental material
Multivariable analysis of hsCRP and novel inflammatory biomarkers level according to quartiles of 25(OH)D.
Additionally, similar results were also found in both univariable and multivariable analysis based on the original data (online supplemental tables 2 and 3).
Associations between serum 25(OH)D and elevated inflammation levels
Additionally, we further explored the associations between serum 25(OH)D and elevated inflammation levels (table 4). Generally, serum 25(OH)D was inversely associated with elevated inflammation levels. In model 1 without adjustment, the ORs (95% CI) in the fourth quartile were 0.10 (0.08, 0.13) for hsCRP, 0.18 (0.16, 0.21) for RDWPCR, 0.71 (0.63, 0.81) for MPVPCR, 0.33 (0.29, 0.38) for NLR and 0.39 (0.34, 0.44) for WBCNR compared with the first quartile, respectively. In model 2 with adjustment for other potential covariates, inverse associations between serum 25(OH)D and elevated inflammation levels were still observed. The corresponding ORs (95% CI) were 0.05 (0.04, 0.06) for hsCRP, 0.13 (0.11, 0.15) for RDWPCR, 0.74 (0.64, 0.85) for MPVPCR, 0.11 (0.09, 0.13) for NLR and 0.57 (0.49, 0.66) for WBCNR in the fourth quartile compared with the first quartile, respectively. Table 5 and online supplemental table 4 showed the stratified analyses by sex. Inverse associations between serum 25(OH)D and elevated inflammation levels were observed in both boys and girls. In multivariable analysis, the corresponding ORs (95% CI) were 0.05 (0.03, 0.07) and 0.05 (0.03, 0.07) for hsCRP, 0.12 (0.10, 0.15) and 0.13 (0.10, 0.16) for RDWPCR, 0.78 (0.64, 0.95) and 0.71 (0.57, 0.87) for MPVPCR, 0.09 (0.07, 0.12) and 0.12 (0.09, 0.15) for NLR and 0.49 (0.40, 0.61) and 0.65 (0.52, 0.80) for WBCNR in the fourth quartile compared with the first quartile, respectively. Consistent to these findings from the rank transformation of the data, there were the inverse associations between serum 25(OH)D and elevated inflammation levels according to the original data (online supplemental tables 5–7).
The associations of 25(OH)D with the presence of high inflammation determined by inflammatory biomarkers
Multivariable analysis of the associations of 25(OH)D with the presence of high inflammation determined by inflammatory biomarkers by sex (OR, 95% CI)
Discussion
In this hospital-based cross-sectional study of children, we observed graded and inverse associations of serum 25(OH)D with hsCRP and four novel inflammatory markers (RDWPCR, MPVPCR, NLR and WBCNR). Consistently, serum 25(OH)D was inversely associated with elevated inflammation levels.
Elevated serum hsCRP level, a biomarker of asthma control in children,27 was also linked to some chronic diseases, such as gestational diabetes mellitus,28 hyperlipidemia29 and cardiovascular diseases.30 These studies indicated that lowering serum hsCRP level was a rationale for therapies and prevention of asthma and inflammatory related chronic diseases. Cojic et al pointed out that vitamin D exerted anti-inflammatory and antioxidative effects via influencing myeloperoxidase, xanthine oxidase and catalase activity in type 2 diabetic patients with metformin therapy.31 However, the association between circulating 25(OH)D and hsCRP has not been extensively characterised in children. Brustad et al found a negative association between plasma 25(OH)D and hsCRP in Childhood aged 6 months (β: −0.004).14 Lower hsCRP value was observed in ≥25 ng/mL 25(OH)D group than that in <25 ng/mL 25(OH)D group among obese children (0.053 vs 0.356 mg/dL).13 Similarly, our study also found that serum 25(OH)D was inversely associated with hsCRP level. Nevertheless, the negative association between plasma 25(OH)D and hsCRP was not replicated in previous study (β: 0.003).14 The inverse correlation between 25(OH)D and CRP might be attenuated to null in infants with 25(OH)D above 50 nmol/L,32 and even reversed in asymptomatic adults.33 Besides, a positive correlation between cord blood 25(OH)D and hsCRP (B=1.003) in the newborn was observed, with a 10-unit increase in 25(OH)D level resulting in a 10-unit increase in hsCRP level.12 Previous study also observed a U-shaped association between 25(OH)D and hsCRP in the general population,34 indicating the anti-inflammatory and proinflammatory effects of 25(OH)D. Different subjects (a larger population-based cohort (n=633) vs a smaller high-risk cohort (n=264)) and different measurement methods of 25(OH)D might provide explanation for inconsistent results in the two cohorts.14 Multivariate linear regression models were calculated in healthy newborns without considering total vitamin D supplement of pregnant mothers,12 which might partly overestimate the underlying associations. Additionally, the differences in sample size (nthis study=10141 vs nCOPSAC2000=264,14) the subjects (newborn12 vs obese children13 vs 6 months infants14 vs 14.6 months children (this study)), statistic methods (linear regression analysis12 vs analysis of covariance (this study)) and potential covariates adjusted might also partly contribute to the heterogeneity in above studies. Although no data exist concerning mechanism of 25(OH)D and hsCRP, some cell culture models had indicated the biological background of 25(OH)D.35–39 As one major metabolites of 25(OH)D, 1,25(OH)2D could modulate inflammatory processes by decreasing proinflammatory cytokines (eg, interleukin 6 (IL-6), IL-23 or tumour necrosis factor)36 37 and impairing the proliferation of dendritic cells38 and B-cells.39
Additionally, previous studies indicated that novel inflammatory biomarkers (RDWPCR, MPVPCR, NLR or WBCNR) were proposed as a representation of inflammation in some chronic diseases.40–42 However, scarce data examined the associations of 25(OH)D and novel inflammatory biomarkers. Based on data from 208 patients with end-stage renal disease receiving maintenance dialysis, a significant inverse association of serum 25(OH)D and NLR was observed (25(OH)D>10 vs <10 ng/dL: 2.96 vs 3.49).43 Subgroup analysis found that NLR was still significantly lower in 25(OH)D>10 ng/dL group than that in 25(OH)D<10 ng/dL group among haemodialysis patients (8.97 vs 14.88); but the difference was not significant in peritoneal dialysis patients.43 An observed study including 100 newborns pointed out that serum 25(OH)D3 was negatively associated with NLR.44 The corresponding NLR values were 1.8 and 1.3 in the insufficient and sufficient groups, respectively.44 In line with these findings, our study also observed an inverse association between serum 25(OH)D and NLR in children. In vivo and vitro studies demonstrated that vitamin D might reduce NLR by increasing T-regulatory cell populations,45 inhibiting the production of proinflammatory cytokines and increasing the production of anti-inflammatory cytokines.46–51 Moreover, our data revealed inverse associations of 25(OH)D with RDWPCR, MPVPCR and WBCNR. Although no study had yet assessed these associations until now, some biological mechanisms supported our findings. Agh et al found that vitamin D supplementation significantly reduced white blood cell counts and mean platelet volume in male rats,51 which was in agreement with the results of the present study. Toxqui et al pointed out that vitamin D-fortified flavoured skim milk could reduce red blood cell distribution width in iron-deficient menstruating women.52 This study indicated vitamin D might modulate RDWPCR via reducing red blood cell distribution width. Nonetheless, a larger sample prospective study is needed to confirm our findings.
There were several strengths of our study. We first identified the associations of serum 25(OH)D with hsCRP and four novel inflammatory markers (RDWPCR, MPVPCR, NLR and WBCNR) in children. Emerging evidences reported the important role of hsCRP and four novel inflammatory markers in chronic diseases.40–42 The present study revealed that 25(OH)D might be the modulating indicator for hsCRP and four novel inflammatory markers levels, and elucidate the health benefits of 25(OH)D on growth and development of children through improving hsCRP and four novel inflammatory markers levels. Besides, serum 25(OH)D could accurately and objectively reflect the metabolites of vitamin D without being influenced by random errors from dietary or supplemental assessment. Moreover, a large sample size allowed us to determine the associations of serum 25(OH)D with hsCRP and four novel inflammatory markers in children and ensured the stability and reliability of the results. Finally, the seasonal variation of vitamin D level might be considered due to subjects were recruited over all seasons. Nevertheless, some limitations were still considered. First, the causality could not be inferred from the cross-sectional study. Although the age range of included participants was relatively large, serum 25(OH)D could accurately and objectively reflect the metabolites of vitamin D without being influenced by random errors in dietary or supplemental assessment. Besides, 25(OH)D, a major metabolite of vitamin D, is produced by hydroxylation in liver,53 which is mainly influenced by diets or vitamin D supplements rather than inflammatory cytokines. Moreover, circulating 25(OH)D level was relatively steady with a half-life of 2–3 w.5 Therefore, cause-and-effect correlations were less likely to be overturned. Second, 25(OH)D level was measured only once in our study, but it was stable in serum and hence was generally used to estimate vitamin D level.54 Third, the underlying associations in the present study might be overestimated due to the lack of information on other drugs using (eg, docosahexaenoic acid, calcium and vitamins). Finally, the make-up of the study sample in this study had a relatively narrow population range, thus caution is needed when extrapolating the findings of this study.
Generally, the graded and inverse associations of serum 25(OH)D with hsCRP and four novel inflammatory markers (RDWPCR, MPVPCR, NLR and WBCNR) were observed. The present study provided further support for the anti-inflammatory effects of 25(OH)D. Further prospective studies are needed to replicate these findings in different populations.
This post was originally published on https://bmjopen.bmj.com