Introduction
Vitamin D (VD) has been described as an essential element in the maintenance of mineral homeostasis and bone architecture (X); however, the vision on the physiological functions of said nutrient has changed radically, the discovery of the VD receptor in immune system cells initiated a series of investigations related to the regulatory activity of this vitamin toward inflammatory phenomena and the association between low levels and development of various infectious and degenerative pathologies. Interest in VD has increased considerably because of multiple studies show that between 30 and 50% of the general population have serum VD deficiency1,2. In Mexico, Flores et al. report a prevalence of VD deficiency of 10-24% and 18-30% of insufficiency in children3. Ninety percent of VD requirements are synthesized through proper skin exposure to sunlight; the remaining 10% is obtained from food4,5.
The main cause of VD deficiency is inadequate sunlight exposure. Sunscreen use reduces its synthesis more than 95%. People with dark skin tone require at least 3-5 times more sun exposure to produce the same amount of VD that a person with white skin tone. There is an inverse association with body mass index (BMI) because more than 30 kg/m2 values are associated with deficiency. Other causes include malabsorption syndromes, nephrotic syndrome (a case in which one loses VD bound proteins through urine), anticonvulsants, antiretrovirals, and chronic use of systemic corticosteroids6. Endocrinology guidelines for assessment, treatment, and prevention of deficiency suggest that 25 (OH) VD can be measured to evaluate blood levels of VD6. An optimal level is defined between 40 and 60 ng/ml, sufficiency between 30 and 39 ng/ml, insufficiency between 21 and 29 ng/ml, and deficiency < 20 ng/ml7.
The association between VD levels and respiratory allergic diseases such as asthma and rhinitis has been investigated. Some studies have associated levels < 30 ng/ml with severe asthma, impaired lung function, frequent exacerbations, uncontrollable asthma, and increased use of inhaled steroids and oral steroids cycles in children, such effect implies a wear and tear both economically and in the quality of life8. Gupta et al. described VD insufficiency prevalence in 47% and deficiency in 17% of asthmatic patients9. Regarding allergic rhinitis, even though it shares similar inflammatory mechanisms with allergic asthma, its association with VD levels is less consistent10,11.
The objective of this study was to evaluate the prevalence of VD deficiency in patients with respiratory allergies and describe the associated factors.
Material and methods
Study population
We conducted a transversal and descriptive study in patients between 3 and 14 years old who attended the allergy and immunology department, who had a recent clinical diagnosis of allergic rhinitis and asthma following ARIA and GINA guidelines. These patients were sensitized to one or more aeroallergen identified by a skin prick tests (Greer® allergens). As immunotherapy guidelines indicate, a positive skin prick test was considered if the wheal was 3 mm bigger than the negative control. All patients and their parents signed assent and informed consent. The protocol was approved by the Research and Ethics Committee of the Hospital (Approval number DI/6/309/04/013). This study was performed according to the Mexican General Health Law regarding Investigation in human beings and the Helsinki Declaration.
After a confirmed diagnosis of allergic rhinitis or asthma was made, we measured serum VD levels and we investigated the different variables considered to be related to VD levels. The following variables were evaluated: age, gender, BMI, type of birth, and season in which VD levels were taken, skin type according to Fitzpatrick, time of sun exposure, sunscreen use, VD fortified food consumption, drugs that interfere with the metabolism and absorption of VD (systemic steroids, anticonvulsants, and antiretrovirals), allergic rhinitis and asthma severity, associated diseases (Sinusitis, allergic conjunctivitis, atopic dermatitis, and food Allergy), total serum immunoglobulin E (IgE), peripheral blood eosinophil count, and 25 (OH) VD levels.
VD levels
VD levels were evaluated by the electrochemiluminescence method. Values were reported in ng/ml. Normal values were between 30 and 100 ng/ml. Patients were divided into three groups according to VD levels: ≥ 30 ng/ml sufficiency, 21-29 ng/ml insufficiency, and < 20 ng/ml deficiency.
Statistical analysis
It was performed with SPSS v. 24.0. For descriptive statistics of quantitative variables, mean, and standard deviation were used. For categorical variables, frequencies and percentages were used. The mean comparison was performed using the Kruskal-Wallis test, and the Chi-square test was used for categorical variables. The odds ratio (OR) with confidence intervals (CI) of 95% with p < 0.05 was estimated.
Results
Demographic characteristics of the population
Sixty-three patients were included with a mean age of 7.25 years (± 2.84); 34 (54%) were females. The mean weight was 26.7 kg (± 10.89), mean BMI of 16.78 kg/m2 (± 2.76), and VD mean levels of 23.66 ng/ml (±7.28), mean total serum IgE was 378.2 UI/mL (± 547.65), and eosinophils 371.1 cells/mL (± 311.6). Ninety-five percent of the samples were taken in winter and spring, and the skin type most commonly found was type IV in 61.9%. Twenty-seven percent of the population used some type of sunscreen. All the subjects consumed some food containing VD, milk being the most frequent.
According to inclusion criteria, all the patients had allergic rhinitis, 46% had asthma, and 32% had a concomitant disease. Of the patients with concomitant diseases, 50% had chronic rhinosinusitis, 30% had allergic conjunctivitis, 15% had atopic dermatitis, and 5% had food allergy, as described in table 1.
Females n (%) | 34 (54) |
Age in years, mean (SDa) | 7 (IQR: 5-9) |
Weight in kg, mean (SD) | 23.05 (IQR: 19.8-34.5) |
Body mass index (Kg/m2), mean (SD) | 16.78 (± 2.76) |
Vitamin D ng/ml, mean (SD) | 23.66 (± 7.28) |
Sufficient levels (≥ 30) n (%) | 14 (22.2) |
Insufficient levels (21-29) n (%) | 30 (47.6) |
Deficient levels (< 20) n (%) | 19 (30.2) |
Total serum immunoglobulin E UI/l, mean (SD) | 154 (IQR: 65.1-440) |
Eosinophils, mean (SD) | 230 (IQR: 200-500) |
Type of birth, n (%) | |
Cesarean section | 35 (55.6) |
Vaginal delivery | 28 (44.6) |
Season, n (%) | |
Winter | 30 (47.6) |
Spring | 30 (47.6) |
Autumn | 3 (4.8) |
Skin type (According to Fitzpatrick) n (%) | |
I | 0 |
II | 2 (3.2) |
III | 21 (33.3) |
IV | 39 (61.9) |
V | 1 (1.6) |
Sunscreen use, n (%) | 17 (27) |
Consumption of food rich in Vitamin D, n (%) | 63 (100) |
Cod liver supplements | 4 (4.8) |
Fish | 44 (69) |
Mushrooms | 9 (14.3) |
Egg | 55 (87.3) |
Dairy products | 61 (96.8) |
Cereals | 55 (87.3) |
Allergic rhinitis n (%) | 63(100) |
Intermittent mild | 18 (28.6) |
Intermittent moderate/severe | 2 (3.2) |
Persistent mild | 6 (9.5) |
Persistent moderate/severe | 37 (58.7) |
Asthma, n (%) | 29 (46) |
Controlled | 5 (8.0) |
Partially controlled | 12 (19.0) |
Uncontrolled | 12 (19.0) |
Comorbidities n (%) | 20 (32) |
Sinusitis | 10 (50) |
Allergic conjunctivitis | 6 (30) |
Atopic dermatitis | 3 (15) |
Food allergy | 1 (5) |
aAverages and standard deviation are shown; However, in the variables whose distributions were not normal, medians, and interquartile range are shown.
As shown in table 2, they were divided into three groups according to VD levels: 14 (22.2%) in Group I with sufficient levels (≥30), 30 (47.6 %) in Group II with insufficient levels (21-29), and 19 (30.2%) in Group III with deficient levels (< 20).
Variables | Group I sufficient levels (≥ 30) (n = 14) | Group II insufficient levels (20-29) (n = 30) | Group III deficient levels (< 20) (n = 19) | p* |
---|---|---|---|---|
Age, n (%) | 0.032 | |||
3-5 years old | 5 (35.7) | 11 (40) | 2 (10.5) | |
6-11 years old | 8 (57.1) | 17 (56.7) | 12 (63.2) | |
12-14 years old | 1 (7.2) | 1 (3.3) | 5 (26.3) | |
Sex, n (%) | 0.63 | |||
Male | 5 (35.7) | 14 (46.7) | 10 (52.6) | |
Female | 9 (64.3) | 16 (53.3) | 9 (47.4) | |
BMIa (Kg/m2) medium (Minimun-maximun) | 15.80 (13.52-21.66) | 16.05 (13.22-20.98) | 17.39 (12.77-26.19) | 0.29 |
Type of birth, n (%) | 0.95 | |||
Caesarean section | 8 (57.1) | 17 (56.7) | 10 (52.6) | |
Normal | 6 (42.9) | 13 (43.3) | 9 (47.4) | |
Season, n (%) | 0.85 | |||
Winter | 6(42.9) | 14 (46.7) | 10 (52.6) | |
Spring | 7 (50) | 15 (50) | 8 (42.1) | |
Autumn | 1 (7.1) | 1 (3.3) | 1 5.3) | |
Skin Type, n (%) | 0.06 | |||
II | 1 (7.1) | 1 (3.3) | 0 | |
III | 7 (50) | 10 (33.3) | 4 (21) | |
IV | 6 (42.9) | 19 (63.4) | 14 (73.7) | |
V | 0 | 0 | 1 (5.3) | |
Using sunscreen n (%) | 3 (31.4) | 9 (52.9) | 5 (26.3) | 0.83 |
Medium SPF 15-29 | 0 | 3 (10) | 1 (5.3) | |
High SPF 30-50 | 0 | 4 (13.3) | 2 (10.5) | |
Very high SPF > 50 | 3 (21.4) | 2 (6.7) | 2 (10.5) | |
IgEb UI/ml, mean (SD) | 304 (297.02) | 360.64 (570.45) | 460.57 (658.15) | 0.57 |
Eosinophils, mean (SD) | 392.85(350.85) | 307.67(282.57) | 455.26 (320.12) | 0.23 |
Allergic rhinitis n (%) | 1.0 | |||
Intermittent mild | 4 (28.6) | 8 (26.7) | 6 (31.6) | |
Intermittent moderate/severe | 1 (7.1) | 1 (3.3) | 0 | |
Persistent mild | 2 (14.3) | 0 | 4 (21.1) | |
Persistent moderate/severe | 7 (50) | 21 (70) | 9 (47.4) | |
Asthma, n (%) | 0.226 | |||
Controlled | 2 (14.3) | 3 (10) | 0 | |
Partially controlled | 2 (14.3) | 6 (20) | 4 (21.1) | |
Uncontrolled | 2 (14.3) | 8 (26.7) | 2 (10.5) | |
Comorbidities, n (%) | 2 (14.3) | 10 (33.3) | 8 (42.1) | 0.235 |
*p values were calculated using Kruskal-Wallis test, differences were considered statistically significant with p ≤ 0.05 Abbreviations: aBody mass index was estimated using the WHO Child Growth Standards,
bImmunoglobulin E
Table 2 describes each one of the groups' characteristics. VD values show that the patients in Group III had higher BMI values, total serum IgE, and eosinophils. Group III also had more comorbidities, but only observing statistically significant differences in the age and skin type.
Table 3 divides the population into the following categories based on VD enriched food consumption: no consumption versus consumption. Cod liver supplement consumption was common in five patients who had higher VD levels. Table 4 describes the frequency of sun exposure in minutes per day. No statistically significant differences between the groups were observed.
Variables | Group I (n = 14) | Group II (n = 30) | Group III (n = 19) | p* |
---|---|---|---|---|
Cod liver supplements | 0.02 | |||
No | 11 (78.6) | 29 (96.7) | 29 (96.7) | |
Yes | 3 (21.4) | 1 (3.3) | 1 (3.3) | |
Fish | 0.23 | |||
No | 2 (14.3) | 9 (30) | 8 (42.1) | |
Yes | 12 (85.7) | 21 (70) | 11 (57.9) | |
Mushrooms | 0.56 | |||
No | 12 (85.7) | 27 (90) | 15 (78.9) | |
Yes | 2 (14.3) | 3 (10) | 4 (21.1) | |
Egg | 0.94 | |||
No | 2 (14.3) | 4 (13.3) | 2 (10.5) | |
Yes | 12 (85.7) | 26 (86.7) | 17 (89.5) | |
Dairy products | 0.69 | |||
No | 0 | 1 (3.3) | 1 (5.3) | |
Yes | 14 (100) | 29 (96.7) | 18 (94.7) | |
Cereals | 0.27 | |||
No | 0 | 5 (16.7) | 3 (15.8) | |
Yes | 14 (100) | 25 (83.3) | 16 (84.2) |
*The differences were considered statistically significant with p ≤ 0.05, these
P values were calculated using the Chi-square test.
Variables (min) | Group I sufficient levels (≥ 30) (n = 14) | Group II insufficient levels (21-29) (n = 30) | Group III deficient levels (< 20) (n = 19) | p* |
---|---|---|---|---|
0-15 | 2 (14.3) | 4 (13.3) | 0 | 0.24 |
16-30 | 2 (14.3) | 1 (3.3) | 2 (10.5) | 0.60 |
31-45 | 1 (7.1) | 0 | 1 (5.3) | 0.12 |
46-60 | 1 (7.1) | 6 (20) | 2 (10.5) | 0.55 |
61-75 | 0 | 1 (3.3) | 1 (5.3) | 0.69 |
76-90 | 1 (7.1) | 0 | 0 | 0.17 |
91-105 | 1 (7.1) | 0 | 0 | 0.17 |
106-120 | 1 (7.1) | 5 (16.3) | 7 (36.8) | 0.19 |
> 120 | 5 (35.7) | 13 (43.3) | 6 (31.6) | 0.70 |
*p values were calculated using Kruskal-Wallis test, differences were considered statistically significant with p ≤ 0.05.
Reason of mummies associated of VD deficiency and insufficiency
We found that 47.6% of patients had insufficient VD levels, while 30.15% had deficient levels, according to the classification mentioned previously.
RISK FACTORS ASSOCIATED WITH VD DEFICIENCY
The only group that showed a lower risk of VD deficiencies was 3-5 years old (OR 0.18, 95% IC 0.03-0.91). However, it can be recognized that patients at risk of VD deficiency were those aged 12-14 years old (OR 7.5, 95% IC 1.306-43.066), those with skin type V (OR 3.444, IC 2.334-5.083 95%, P 0.125), moderate-severe intermittent rhinitis (OR 1.452, 95% IC 1.227-1.719), and controlled asthma (OR 1.487, 95% CI 1.243-1.780). The lack of consumption of cod liver supplements was associated with deficiency (OR 1.475, 95% CI 1.237-1.759, p: 0.1), table 5.
Variable | OR | 95% IC | p‡ |
---|---|---|---|
Age, n (%) | |||
3-5 years old | 0.18 | 0.03-0.91 | 0.02 |
6-11 years old | 1.30 | 0.43-3.94 | 0.63 |
12-14 years old | 7.50 | 1.30-43.06 | 0.03 |
Sex | |||
Male | 1.46 | 0.49-4.30 | 0.49 |
Female | 0.68 | 0.23-2.01 | 0.49 |
Body mass index percentile | |||
Under weight | 0.91 | 0.16-5.20 | 0.92 |
Normal | 0.71 | 0.23-2.23 | 0.56 |
Overweight | 0.91 | 0.16-5.20 | 0.92 |
Obesity | 0.91 | 0.16-5.20 | 0.92 |
Type of birth, n (%) | |||
Caesarean section | 1.11 | 0.40-3.48 | 0.75 |
Vaginal delivery | 0.84 | 0.28-2.48 | 0.75 |
Season, n (%) | |||
Winter | 1.33 | 0.45-3.92 | 0.60 |
Spring | 0.72 | 0.24-2.15 | 0.56 |
Autumn | 1.16 | 0.09-13.69 | 0.90 |
Skin type | |||
II | 1.45 | 1.22-1.71 | 0.35 |
III | 0.42 | 0.12-1.49 | 0.17 |
IV | 2.12 | 0.65-6.94 | 0.20 |
V | 3.44 | 2.33-5.08 | 0.12 |
Sunscreen use | 1.05 | 0.31-3.54 | 0.93 |
Immunoglobulin E (UI/l) | |||
0-165 | 0.55 | 0.17-1.72 | 0.45 |
≥166 | 1.80 | 0.58-5.61 | 0.30 |
Eosinophils | |||
0-300 | 0.46 | 0.15-1.39 | 0.16 |
≥300 | 2.14 | 0.71-6.42 | 0.16 |
Allergic rhinitis n (%) | |||
Intermittent mild | 1.23 | 0.38-3.97 | 0.72 |
Intermittent moderate/severe | 1.45 | 1.22-1.71 | 0.34 |
Persistent mild | 5.60 | 0.92-33.77 | 0.04 |
Persistent moderate/severe | 0.51 | 0.17-1.53 | 0.22 |
Asthma, n (%) | |||
Controlled | 1.48 | 1.24-1.78 | 0.12 |
Partially controlled | 1.20 | 0.31-4.59 | 0.79 |
Uncontrolled | 0.40 | 0.07-2.03 | 0.25 |
Comorbidities | 1.93 | 0.62-5.98 | 0.24 |
Cod liver supplements | |||
No | 1.47 | 1.23-1.75 | 0.17 |
Yes | 0.67 | 0.56-0.80 | 0.17 |
Sun exposure | |||
< 60 min | 0.56 | 0.17-1.86 | 0.34 |
61-120 min | 2.82 | 0.87-9.10 | 0.07 |
Over 120 min | 0.66 | 0.21-2.08 | 0.48 |
Odds Ratio (OR), confidence interval (CI),
‡differences were considered statistically significant with p ≤ 0.05, p values were calculated using the Chi-square test.
We did not find other conditions associated with low levels of VD in any of the patients, such as the use of systemic corticosteroids, antiretrovirals, anticonvulsants, or protein-losing syndromes.
Discussion
This study shows a VD deficiency of 30.15%, and insufficiency of 47.1%, indicating that only 22.75% of the study population has sufficient VD levels. These data are superior to those showed by the national health and nutrition survey (ENSANUT) 2006 that reported a VD deficiency of 24.1% and insufficient of 30% in children between 2 and 5 years old, and a deficiency of 10% and insufficiency of 18% in children between 6 and 12 years old. This survey was conducted in the general population, unlike our study that included only allergic patients. We found that patients between 12 and 14 years old showed an increased risk of deficiency, unlike the ENSANUT survey, which reported a higher deficiency in the preschool group3. BMI was higher in the group of patients with deficiency, but there was no statistically significant difference between the groups and no association, unlike what is reported in the literature, in which it is stated that patients with a higher BMI have lower VD levels. There are different hypotheses that explain deficiency in these subjects such as low sun exposure, poor supplement intake, accumulation into adipose tissue, or more bone mass needed to support the heavy weight12.
The season in which VD levels were measures does not show significant differences between groups, and no risk of significant association was found, despite what Webb et al. report, as they observed lower VD levels in the winter months caused by differences in solar radiation during those months. Although, this study was carried out in a Canadian population where solar exposure may be lower in the winter months than in our population13.
Significant association was found in types II and V according to Fitzpatrick being greater in the latter. However, we did not explain why skin types being these extremes were associated with the deficiency, different from what Clemens et al. who describe that higher levels of melanin are associated with lower levels of VD due to a decrease in solar absorption in hyper-pigmented skin14. Regarding the use of sunscreen, a greater association with deficiency was found with its use, but the OR did not show a statistically significant association. Matsuoka et al. described that using a sunscreen with a sun protection factor > 15 decreases VD absorption in the skin up to 99%15.
Higher levels of IgE and eosinophils were found in patients with VD deficiency, but the OR did not show a significant association. Muehleisen describes the role of VD in the regulation of IgE production, which may explain that low levels of VD produce an imbalance of regulatory T cells and IgE-producing B cells. However, other authors have not been able to find this association16.
Regarding the association with allergic rhinitis, there are controversial data some authors have documented a strong relationship between levels of VD and allergic process, for example: Vasallo and Camargo demonstrated that VD deficiencies are associated with low tolerance to environmental allergens17. On the other hand, Mai et al. documented that in adult Norwegian patients there is a significant relationship between deficient levels of VD and the development of allergic rhinitis18. Rothers et al. found that deficient levels of VD in umbilical cord blood are associated with aeroallergens and predispose to the development of rhinitis and asthma and even Gupta et al. described that low levels of this vitamin are associated with severe asthma, uncontrolled disease, and airway remodeling19,20. By another hand, others researchers found no significative associations between rhinitis and lower levels of VD8. In our case, those patients with persistent disease were associated with the lower VD values without statistical significance, so we cannot establish a proper association between rhinitis severity and VD deficiency. We did not find an association between VD deficiency and asthma severity and in similar way, VD deficiency seems to be associated with higher comorbidities, such as rhinosinusitis21. However, we did not find a statistically significant association.
With respect to foods enriched with VD, we found that cod liver supplement consumption is a protective factor for deficiency, which is due to its high content of this vitamin22. Holick et al. described that exposure to sunlight is a protective factor for VD deficiency, but we were unable to demonstrate this association6.
This study has some limiting, including a small sample size and that it only included children. We could not find a strong association with some variables considered to be important regarding VD levels, but this could have been more evident with a bigger sample. One of problem in this work is that it only included allergic patients so we have an idea of VD deficiency repercussion in this population but the comparison is not complete without a control group. In the future, it could be convenient to include a control group to establish if this condition is related to the allergic conditions in these patients.
VD deficiency is a global public health problem, and patients with allergic rhinitis and asthma show a high either deficiency or insufficiency - even more than the general population. Different studies have shown an association between VD deficiency and severe manifestations of allergic diseases. However, we could not demonstrate that relationship with statistical significance. It seems that consumption of cod liver supplements is a protective factor for deficiency.
The deficiency of VD is high in allergic patients; the factors associated with the deficiency were little consumption of cod liver, age, skin photo type, and moderate-severe rhinitis. It is necessary more works to confirm our findings and to reveal the mechanisms underlying these observations.
Conclusions
Vitamin D deficiency is frequent in allergic patients due to several factors such as cod liver consumption, age, skin phototype and the presence of moderate-severe allergic rhinitis.
There is a need for similar studies, not only to confirm our findings, but also, those investigating the underlying mechanism by which allergic patients have lower vitamin D levels compared to the rest of the population.