An association between low vitamin D status and childhood pneumonia severity in hospitalized bulgarian patients

Full Text

Introduction

Lower respiratory tract infections are among the most important causes of morbidity and mortality in the pediatric population worldwide. Despite advances in treatment and prevention, childhood pneumonia is a major reason for hospital admissions and remains a leading cause of death claiming an estimated 800 000 children’s lives in 2018. Globally, over 1.23 million children died of pneumonia before reaching their 5^th birthday — the equivalent of over 3400 deaths per day worldwide [27, 28].

There is growing evidence that vitamin D plays an important role in the immune system by modulating both innate and adaptive immunity. Vitamin D is an additional factor in the inflammatory response regulation [16]. Its action is mediated via the vitamin D receptor (VDR), which is present in almost all types of immune cells, including activated CD4⁺ and CD8⁺ cells, B cells, macrophages, neutrophils and dendritic cells. Vitamin D deficiency is associated with decreased host defenses against infections [6].

The relationship between vitamin D deficiency and the susceptibility to infections was investigated initially for tuberculosis patients [18]. Human alveolar macrophages stimulate the Toll-like receptors (TLR) in the presence of M. tuberculosis. TLR-activation upregulates VDR expression and vitamin D 1α-hydroxylase gene, thus increasing the local levels of 1,25(OH)₂D₃ [22]. A positive correlation between vitamin D levels in respiratory epithelial cells and antimicrobial peptide mRNA production has been reported [14]. Cathelicidine and beta-defensins are important components of the innate immunity in the lower respiratory tract. They inhibit pneumococcal, meningococcal, and group A streptococcal disease-causing agents [2]. These local vitamin D effects suggest the role of its deficiency in the development of acute lower respiratory tract infections. Some studies report that low vitamin D status is a risk factor for more severe disease among hospitalized pneumonia patients [15]. Recent reviews also supported the possible role of vitamin D in decreasing the risk of COVID-19 infections and mortality [7]. Adequate vitamin D concentrations can be a beneficial factor in preventing serious illness, faster recovery, and reduced hospital stays [21].

The relationship between serum vitamin D levels and the incidence and severity of pneumonia in hospitalized children has not been analyzed yet in Bulgaria. The aim of this study was to examine whether a low vitamin D status was a risk factor for complications of the disease, the use of multiple antibiotics and a long hospital stay in children with pneumonia. We hypothesized that lower levels of vitamin D contributed to a more severe clinical course of the disease.

Materials and methods

Patients characteristics. The study collection included 200 children aged 11 days to 17 years old from Pulmonology clinic with ICU at the University Children’s Hospital, Sofia, from January 2015 to January 2019. A written informed consent was signed by each participant’s parent upon enrollment. Ninety-eight children (48 male and 50 female) with severe pneumonia were chosen as study (pneumonia) group. Patients who developed complications, required admission to the ICU and/or surgical treatment were grouped into “complicated pneumonia”. All other patients were subgrouped to the “non-complicated-pneumonia”. One hundred and two healthy children (54 male and 48 female) selected in an outpatient setting were chosen as the control group. They were compared with the study group by sex and sampling season.

Immunoassay. Serum vitamin D levels were measured using Electrochemiluminescence immunoassay (ECLIA) for the in vitro determination of 25-hydroxyvitamin D. Information for vitamin D intake prior to measurement was obtained for all the patients. All of them underwent venous puncture and withdrawal of 2–3 ml of blood. Serum was separated and immediately frozen at –20°C until measurement. For determination of the vitamin D status, the following cutoffs were set: > 75 nmol/l — sufficiency; 50–75 nmol/l — insufficiency; 25–50 nmol/l — deficiency and < 25 nmol/l — severe deficiency. These cutoff values were set in accordance with the data, published by the Institute of Medicine (IOM) [23].

Statistical analysis. The Spearman correlation coefficient was calculated as described previously using SPSS v.23.0 software [17] and was used to indicate the direction of association. Testing for normality of variables was performed, using Kolmogorov–Smirnov and Shapiro–Wilk tests. The Mann–Whitney U-test and the Kruskal–Wallis test were used to evaluate quantitative data. Range values are presented in brackets.

Results

Patients characteristics

Cases with severe pneumonia were subdivided into groups with and without complications (36 and 62, respectively). Only 6 children in the complicated pneumonia subgroup were found to be vitamin D supplemented. The proportion of supplemented children in the non-complicated pneumonia subgroup was much higher — 44. Complications were mainly pulmonary with parapneumonic effusion (Table 1).

Only 19,6% (n = 20) of the healthy children and 32.6% (n = 32) of pneumonia patients were found to be vitamin D supplemented. Median vitamin D intake in the healthy children group was 200 IU/day, whereas patients in both subgroups were receiving 500 IU — a significantly higher dose than controls by the time of study measurement (p = 0.013).

However, the study and control groups differed in age. The average age of the pneumonia group was 4 (2–8) years and the median age of the healthy controls was 7 (4–8) years, p = 0.002. Exclusion criteria for the children in the control group were history of respiratory symptoms one month prior to enrollment as well as any accompanying chronic disease.

We evaluated the relationship between the inflammatory markers (CRP, erythrocyte sedimentation rate, white blood count) and vitamin D levels in the pneumonia group. We also evaluated the length of hospital stay and duration of antibiotic treatment and looked for correlations with vitamin D status.

Overall vitamin D status

All children included in the study had a median vitamin D level of 52.4 (36.7–72.7) nmol/l, typically in the insufficient range. In the control group half of the children had deficiency. Only 16.7% of the children had sufficient vitamin D concentrations and 33.3% had insufficiency. Surprisingly, sufficient levels were more frequent for children with pneumonia, accounting 33.8%. In the pneumonia group 31.7% had insufficiency and 34.5% had deficiency.

In the pneumonia subgroups the larger proportion of the non-complicated pneumonia patients had sufficiency — 40.3%; deficiency was found in 27.4% and insufficiency was present in 32.3%. In the complicated pneumonia group 47.2% had deficiency, 30.6% had insufficiency, and 22.2 had sufficiency.

No gender relationship between vitamin D levels in the pneumonia and control groups was found in this study.

Vitamin D levels

The pneumonia subgroup median 25-hydroxyvitamin D level was 61.3 (40.9–82.0) nmol/l, whereas in the control subgroup it was 49.5 (33.1–65.8) (Fig. 1). Using the Mann–Whitney U test we found significantly lower vitamin D levels in the control group (mean ranks: 111.75 and 98.6, U = 9147.5, Z = –2.269, p = 0.007, r = 0.19).

 

Figure 1. 25-hydroxyvitamin D level comparison between pneumonia and control group

 

Table 1. Types of complications in pneumonia subgroup

Complication		Number of cases	%
Pulmonary	Parapneumonic effusion	24	66.7
	Respiratory failure	6	16.7
	Pneumothorax	1	2.7
	Hydrothorax	1	2.7
	Atelectasis	1	2.7
	Necrotizing pneumonia	1	2.7
Extrapulmonary	Sepsis	1	2.7
	Hemolytic uremic syndrome (HUS)	1	2.7
Total		36	100

 

Pneumonia subgroups comparison

Vitamin D levels in the non-complicated and complicated pneumonia subgroups were 63.8 (49.1–88.6) nmol/l and 50.8 (29.7–68.0) nmol/l, respectively (Fig. 2). Patients who developed complications showed significantly lower levels than patients with no complications (mean ranks 57.02 and 36.54, U = 649.5, Z = –3.43, p = 0.001, r = 0.32).

 

Figure 2. 25-hydroxyvitamin D levels comparison in pneumonia subgroups

 

All complicated pneumonia patients were treated in an intensive care setting. Patients with surgical complications and mechanical ventilation were found to have significantly lower vitamin D concentrations in comparison to non-complicated pneumonia patients (Table 2).

 

Table 2. Comparison between vitamin D serum levels in children, receiving (yes) and not receiving (no) ICU procedures

Procedure		n	Median vitamin D, nmol/l	P
ICU admission	Yes	36	50.85	0.001
	No	62	63.85
Oxygen supplementation	Yes	21	50.90	0.066
	No	77	62.15
Mechanical ventilation	Yes	6	21.16	0.0002
	No	92	62.27
Surgery	Yes	11	48.19	0.029

 

Median vitamin D levels in all types of complications found in the study group are shown on Fig. 3.

 

Figure 3. Vitamin D levels (median values) related to pneumonia complications

 

ОR for developing complicated pneumonia if 25-hydroxyvitamin D level was below 51 nmol/l was 1.925 times higher (CI 95% 1.15–3.20). Mortality rate due to pulmonary complications in the complicated pneumonia subgroup was 8.3%, accounting for 3 cases. Their vitamin D levels were within the deficiency and severe deficiency state — 9.0 nmol/l, 9.2 nmol/l and 35.3 nmol/l.

This study was not conducted to evaluate seasonal vitamin D variations. However, we compared vitamin D levels in different seasons for the main groups — pneumonia and control, but not for pneumonia subgroups. Statistically, there was no significant difference between vitamin D levels in pneumonia and cotrol groups across seasons (not shown). However, this might be due to the small sample sizes.

Inflammatory markers

Significant negative correlation between serum vitamin D values and levels of the inflammatory markers — CRP and erythrocyte sedimentation rate (ESR) was found. White blood count (WBC) did not correlate with vitamin D concentrations, though there was a tendency towards higher leucocyte number at lower 25-hydroxyvitamin D level (Table 3).

 

Table 3. Correlation between vitamin D concentrations and inflammatory markers in all pneumonia patients

Value	CRP, mg/L	ESR, mm/h	WBC, n × 109/l
Median	32.1	32.0	11.2
Range	0.38–390.0	2–15	0.8–40.5
Spearman correlation coefficient	–0.357	–0.237	–0.045
p-value	0.001	0.047	0.674

 

Hospital stay and duration of antibiotic treatment

In the study group, significant negative correlation between serum vitamin D levels and lenght of hospital stay, as well as duration of intravenous antibiotic treatment was established (Table 4).

 

Table 4. Vitamin D levels and duration of hospital stay and intravenous antibiotic therapy in all pneumonia patients

Correlation between vitamin D levels	Hospital stay (days)	AB-treatment i.v. (days)
Median	7.00	5.50
Range	5–30	0–21
Spearman correlation coefficient	–0.238	–0.254
p-value	0.018	0.013

 

Complicated pneumonia patients had significantly longer hospital stay and thus longer antibiotic therapy than non-complicated pneumonia cases.

Patients in the complicated pneumonia subgroup had significantly longer hospital stay of 12 days (8–20), compared to non-complicated pneumonia patients, who were hospitalized for 5 days (5–7.5), p = 0.0001.

Median duration of antibiotic treatment in non-complicated pneumonia patients was 5 days (5–6), whereas complicated pneumonia patients were treated significantly longer — 10 days (7–15), p = 0.0001.

Number of antibiotics

In the pneumonia group 56% (n = 55) received 1 intravenous antibiotic for treatment, whereas 44% (n = 43) needed multiple antibiotic treatment.

In the non-complicated pneumonia subgroup 76% (n = 47) were treated with monotherapy, whereas 24% (n = 15) received multiple antibiotic therapy. In the complicated pneumonia subgroup 22% (n = 8) were treated with monotherapy and 78% (n = 28) received multiple antibiotic therapy.

Median vitamin D levels were significantly lower in children, receiving multiple intravenous antibiotic therapy at 51.6 nmol/l, compared to those requiring monotherapy at 64.0 nmol/l (p = 0.004).

Frequent respiratory illness was reported for 45% (n = 44) of the study group and for 21% (n = 21) of the control group.

Discussion

All 200 children who participated in this study had overall deficient and nsufficient vitamin D levels. This result confirms previously published data by Holick and Palacios on the high prevalence of inadequate vitamin D status among children and adolescents worldwide [9, 10, 19]. Since subjects of this study were chosen not to have underlying conditions known to affect vitamin D production, we assume limited sun exposure and reduced dietary intake and supplementation to be the main causes of this result. It has been recognized hat using sunscreen with SPF 30 might inhibit up to 95% of vitamin D skin production [8].

It has been found that pneumonia occurs throughout the year with a peak frequency in winter, when serum vitamin D concentrations are naturally depleted. In comparison with other studies, vitamin D levels in our healthy subjects are generally low [5, 20, 25]. This might be explained with the fact, that determination of vitamin D status was based mainly on winter serum concentrations. Much of our studied controls, higher than 50%, showed deficient levels. Different studies estimated large variation of vitamin D status within different European countries. In details, Lips et al. concluded that deficiency of vitamin D levels in healthy children in Europe were present in almost as high proportion in Germany (44.5%), Greece (40.5%) and UK (56.4%), but not in Romania (29%) may be due to the low age range of the examined children (0–3 years), where supplementation was wide spread and a major factor for maintaining adequate levels [4, 13].

Supplemented children in the study group were a larger number than supplemented controls. Thus, vitamin D levels in patients were significantly higher than healthy controls. Median vitamin D intake was 500 IU/day for the pneumonia patients vs only 200 IU/day for controls. The latter doses are lower than recommended by the IOM [23]. This result highlights the importance of supplementation in elevating and maintaining adequate vitamin D status. In this study, however, supplementation proves not to be sufficient.

Since vitamin D serum levels were higher in patients, compared to controls, we assumed that vitamin D status was not a major factor for pneumonia incidence. Vitamin D status, however might be a factor for disease severity, since cases with pneumonia complications had significantly lower vitamin D levels, than non-complicated pneumonia patients. Complicated pneumonia patients had insufficient vitamin D levels along with significantly elevated inflammatory markers, causing usage of multiple antibiotics and prolonged length of hospital stay for adequate treatment and observation.

In terms of laboratory tests, the most important for assessing the severity of pneumonia are C-reactive protein (CRP) and the parallel leukocyte count, DCC and ESR. CRP is one of the generally accepted markers in clinical practice, reflecting the magnitude of the activity of inflammatory processes and tissue damage. According to Van den Berghe et al., the inflammatory markers, among which CRP, ESR and WBC levels were negatively correlated with vitamin D concentrations. In the conditions of active inflammation, vitamin depletion, as well as increase or decrease of the inflammatory response parallel processes are completely possible [26].

High dose supplementation in the course of treatment of pneumonia did not lower inflammatory markers, but reduced disease relapses. Although the degree of inflammatory response is known to vary individually due to factors unrelated to vitamin D status, we hypothesize that our results are consistent with the immunomodulatory role of vitamin D in infectious diseases. 25-hydroxyvitamin D increases the antimicrobial peptide synthesis in lungs. In addition, it induces the switch of Th1 to a more regulatory Th2 type of immune response [18]. Thus, adequate 25-hydroxyvitamin D concentration could be protective against the adverse physiologic effects that occur in excessive inflammatory areas. [3, 12, 22, 26]

Our observational study has indicated that there is an association between low serum vitamin D levels and elevated markers of inflammation at presentation of the disease acute pneumonia but we cannot answer the question of whether decreased vitamin D levels lead to or cause increased inflammatory activity.

We also found high incidence of life-threatening complications, requiring ICU-treatments, surgical interventions and mechanical ventilation to be significantly correlated to vitamin D insufficient levels. Patients with pulmonary complications – parapneumonic effusion, hydrothorax, pneumothorax and necrotizing pneumonia had profound vitamin D deficiency, suggesting that there might be association between low vitamin D status and impairment of the local immune defense in lung parenchyma. Patients developed respiratory failure did not show lower vitamin D levels. This might be due to other factors, such as anatomical features of pediatric patients, including low chest wall compliance – important factor in the development of respiratory failure. Correlation between vitamin D deficiency and severity in hospitalized pneumonia patients was found by Sakka, De Oliveira and Inamo, showing similar results [1, 11, 24]

The chances of developing complicated pneumonia when serum levels are below 51 nmol/l are quite small in this study. However, complications were major and associated with multiple use of antibiotics, longer therapy, and prolonged hospital stay.

Pneumonia is widely recognized as a leading cause of death among the pediatric population under the age of 5, especially in developing countries. Vitamin D may be important for an adequate immune response in developing lungs, especially in the setting of infection. Children with low levels of vitamin D were likely to have higher inflammatory markers and might be at higher risk of developing life-threatening complications. Whether vitamin D supplementation is helpful in preventing complications of pneumonia or in reducing inflammatory markers in children is a matter of further large scale studies.

We found high rates of insufficient and deficient vitamin D levels in all participants. Attention should be paid to supplementation of healthy children. Serum 25-hydroxyvitamin D concentrations were not related to pneumonia incidence. However, low vitamin D levels were associated with disease severity. Children with low levels of vitamin D might be at higher risk for developing life-threatening complications, higher inflammatory response at presentation, ICU-admission, usage of more than one antibiotics and prolonged hospital stay.

Acknowledgements

The authors thank the Health Insurance Fund Doverie and Medcross Outpatient Care Centers for providing the controls group. We are grateful to the medical personnel at the Children’s Hospital of the Medical University of Sofia, Pediatric Surgery Clinic at the N.I. Pirogov University Hospital for Emergency Medicine.

About the authors

N. Rimpova

Children’s Hospital of the Medical University of Sofia

Email: nadiarimpova@abv.bg

MD, Assistant Professor, Department of Paediatrics

Bulgaria, 1113, Sofia, Acad. Georgi Bonchev str., 26

V. Valcheva

Stephan Angeloff Institute of Microbiology of the Bulgarian Academy of Sciences

Email: violeta_valcheva@mail.bg

PhD, Associate Professor, Department of Infectious Microbiology

Bulgaria, 1113, Sofia, Acad. Georgi Bonchev str., 26

A. Tsakova

Aleksandrovska Hospital of the Medical University of Sofia

Email: adelina_d@abv.bg

Head Assistant Professor, Central Clinical Laboratory

Bulgaria, 1113, Sofia, Acad. Georgi Bonchev str., 26

H. Shivachev

N.I. Pirogov University Hospital for Emergency Medicine

Email: hshivachev@gmail.com

PhD, Associate Professor

Bulgaria, 1113, София, ул. академика Георги Бончева, 26

D. Iliev

Children’s Hospital of the Medical University of Sofia

Author for correspondence.
Email: daniel_iliev@gmx.net

PhD, Associate Professor, Department of Paediatrics

Bulgaria, 1113, Sofia, Acad. Georgi Bonchev str., 26

References

Supplementary files