25-Hydroxyvitamin D Sufficiency Is Associated With Reduced Atelectasis in Common Variable Immunodeficiency: A Cross-Sectional Study

Abstract

Background

Common variable immunodeficiency (CVID) is an inborn error of immunity characterized by impaired antibody production, poor vaccine responses, and reduced serum immunoglobulin levels, resulting in respiratory manifestations that can progress to pulmonary complications. Although vitamin D has been associated with pulmonary outcomes in other populations, the association between serum 25-hydroxyvitamin D [25(OH)D] status and pulmonary manifestations, particularly atelectasis, in CVID remains unclear.

Material/Methods

We conducted a retrospective cross-sectional study of 48 adult patients with CVID stratified by serum 25(OH)D concentration (< 30 ng/mL [insufficient] vs ≥ 30 ng/mL [sufficient]). Pulmonary outcomes (atelectasis, bronchiectasis, recurrent pneumonia, and chronic obstructive pulmonary disease [COPD]) were obtained from medical records. Associations were assessed via Fisher’s exact test. Exploratory analyses included a longitudinal imaging review and descriptive Kaplan-Meier curves illustrating atelectasis-free disease duration by baseline 25(OH)D status.

Results

Among 48 patients, 35 (73%) had 25(OH)D concentrations below 30 ng/mL. Atelectasis was more frequent in the insufficient group (15/35; 43%) (odds ratio = 0.12; 95% confidence interval, 0.01–0.97; P = 0.02). No significant associations were observed for bronchiectasis (63% vs 46%; P = 0.53), recurrent pneumonia (91% vs 84%; P = 0.49), or COPD (3% vs 8%; P = 0.46). Kaplan-Meier curves did not show an apparent separation between groups.

Conclusions

Serum 25(OH)D sufficiency was associated with lower atelectasis prevalence in patients with CVID. This finding reflects a cross-sectional association and does not establish causality; it supports a potential link between 25(OH)D status and atelectasis. Prospective studies are needed to further evaluate this relationship.

Introduction

Common variable immunodeficiency (CVID) is classified as an inborn error of immunity, characterized by defective antibody production, low serum immunoglobulin levels, and impaired specific antibody responses [1]. CVID is among the most widespread symptomatic primary immunodeficiencies; its prevalence in developed countries ranges from 0.6 to 6.9 per 100 000 population, with lower rates in developing countries [1]. Diagnosis is often delayed until early or mid-adulthood [2].

Diagnostic criteria established by international guidelines incorporate hypogammaglobulinemia—typically reduced serum immunoglobulin G (IgG) levels, as well as low immunoglobulin A (IgA) and/or immunoglobulin M (IgM)—together with impaired vaccine responses and the exclusion of secondary causes [3,4]. Such diagnostic delays are consistently associated with worse outcomes, particularly involving the respiratory tract [1]. In addition to infectious susceptibility, the heterogeneity of CVID has become increasingly recognized, underscoring the importance of early diagnosis and phenotype characterization.

Large cohort studies have confirmed that the hallmark of CVID is defective antibody production, reflecting impaired B-cell differentiation; a substantial subset of patients also exhibit abnormalities in T-cell function, including defective T-cell help to B cells or altered T-cell signaling [5,6]. In clinical settings, most patients present with recurrent sinopulmonary infections, but many develop noninfectious complications, including autoimmunity, polyclonal lymphoproliferation, bronchiectasis, granulomatous/interstitial lung disease, enteropathy, and malignancy [6,7].

Among the noninfectious complications of CVID, atelectasis represents an underrecognized pulmonary manifestation. According to the 2024 Fleischner Society Glossary [8], atelectasis can arise from airway injury and chronic inflammation, in which recurrent infections promote air trapping, bronchial wall thickening, and the subsequent development of both reversible atelectasis and irreversible bronchiectasis. Disturbances in pulmonary surfactant homeostasis, which maintains alveolar stability by reducing surface tension, further contribute to its pathophysiology [9]. Vitamin D has emerged as a key regulator of immune and inflammatory pathways in the lung, influencing host defense, epithelial integrity, and the pulmonary microenvironment during respiratory injury [10,11].

Vitamin D exerts broad immunomodulatory effects. In innate immunity, it enhances antimicrobial peptide synthesis and phagocytic activity; in adaptive immunity, it modulates lymphocyte differentiation and B-cell activation, thus contributing to immune homeostasis [12]. Beyond immune regulation, experimental studies indicate that vitamin D influences surfactant synthesis [11], attenuates profibrotic signaling [13], and modulates macrophage-driven inflammatory responses within the lung microenvironment [14,15]. Collectively, these mechanisms provide a biological rationale for investigating whether vitamin D sufficiency can protect against pulmonary complications such as atelectasis in CVID.

Definitions of serum 25(OH)D sufficiency remain inconsistent across international guidelines, particularly regarding whether concentrations of at least 30 ng/mL should be considered an appropriate threshold in at-risk populations. The 2024 Endocrine Society Guideline does not endorse a universal serum cutoff; it emphasizes that optimal 25(OH)D concentrations should be individualized according to the clinical context, rather than based on routine population-wide screening [16]. In contrast, the 2022 Central and Eastern European Expert Consensus explicitly defines at least 30 ng/mL as the desirable sufficiency threshold for individuals with chronic or immune-mediated conditions [17]. The Brazilian Society of Endocrinology and Metabolism (SBEM) recommends maintaining serum 25(OH)D concentrations of at least 30 ng/mL in individuals with inflammatory diseases or immunosuppression [18]. In the present study, a threshold of at least 30 ng/mL was adopted as the operational definition of sufficiency for stratification purposes, in accordance with SBEM recommendations.

Evidence regarding 25(OH)D status in CVID is scarce. To date, a single observational study has shown frequent and often severe vitamin D deficiency in these patients, particularly among those with autoimmune disease and bronchiectasis [19]. However, that study was descriptive and did not assess radiologic outcomes, leaving it unclear whether 25(OH)D status is associated with specific pulmonary manifestations, particularly atelectasis. Therefore, the present study aimed to investigate the association between baseline serum 25(OH)D status and pulmonary manifestations in CVID. We hypothesized that baseline 25(OH)D sufficiency (≥ 30 ng/mL) would be associated with a lower prevalence of atelectasis documented on radiologic examinations (available in medical records) throughout the disease course. Atelectasis was defined as the primary outcome of interest; secondary analyses explored associations with other pulmonary manifestations, including bronchiectasis, recurrent pneumonia, and chronic obstructive pulmonary disease (COPD).

Material and Methods

Ethics Statement

The study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Institutional Review Board of the Hospital das Clínicas, Faculty of Medicine, University of São Paulo (HC-FMUSP) (approval number CAAE: 93846318.2.0000.0068). Written informed consent was obtained from all participants, in accordance with institutional requirements for retrospective chart review studies.

Study Design and Population

This retrospective, cross-sectional study was conducted at the Division of Clinical Immunology and Allergy, Faculty of Medicine, University of São Paulo (FMUSP), between March 2018 and December 2019. Baseline 25(OH)D—defined as the earliest available serum measurement within the study period—was used to stratify patients into below 30 ng/mL (insufficient) and at least 30 ng/mL (sufficient) groups. When multiple 25(OH)D measurements were available, the baseline value was used for group classification. Additional measurements were reviewed descriptively to assess consistency but were not used for longitudinal exposure modeling. A longitudinal review of available imaging data was performed as an exploratory, descriptive analysis; Kaplan-Meier curves were constructed for a similar purpose.

In total, 71 adult patients previously diagnosed with CVID were initially assessed for eligibility. Of these, 48 patients aged 18 to 50 years met the inclusion criteria and fulfilled the 2019 European Society for Immunodeficiencies diagnostic criteria. The diagnostic criteria required (1) serum IgG levels less than 2 SD below age-matched controls plus either low IgA (< 0.07 g/L) or IgM (< 0.04 g/L); (2) at least 1 clinical feature (recurrent infections, impaired vaccine response, or < 2% switched memory B cells); and (3) exclusion of secondary causes of hypogammaglobulinemia [4]. Exclusion criteria were current smoking; pregnancy; untreated or active malignancy; diabetes mellitus; chronic systemic corticosteroid use; vitamin D supplementation within the previous 6 months; or incomplete clinical, laboratory, or imaging records. Figure 1 shows the patient selection flowchart.

Figure 1.

Flowchart illustrating the selection process of adult patients with common variable immunodeficiency (CVID) followed at the Department of Allergy and Immunology, Faculty of Medicine, University of São Paulo (FMUSP), between March 2018 and December 2019. Among 71 patients assessed for eligibility, 23 were excluded due to vitamin D supplementation, corticosteroid use, lack of thoracic imaging, or incomplete records. Overall, 48 patients met all inclusion criteria and were included in the study. A subset of 16 patients with paired 25(OH)D and imaging data was included in the exploratory longitudinal sub-analysis. ESID, European Society for Immunodeficiencies.

Pulmonary Manifestations

Atelectasis was defined in accordance with the 2024 Fleischner Society Glossary of Thoracic Imaging Terms as partial or complete lung collapse evidenced by displacement of interlobar fissures, vascular crowding, or parenchymal opacification with volume loss in the absence of obstructing masses or mucus plugs [8]. Data were obtained from routine thoracic imaging reports (computed tomography or chest radiography) retrieved from electronic medical records. All imaging analyses were based on radiology reports issued by board-certified thoracic radiologists at the Institute of Radiology, HC-FMUSP; no independent reinterpretation of imaging studies was performed.

Atelectasis was considered present if documented on any available thoracic imaging report during the disease course, regardless of whether the finding preceded or followed the baseline 25(OH)D measurement. Thus, the analysis evaluated the association between baseline 25(OH)D status and the presence or history of atelectasis documented on available imaging, rather than incident atelectasis prospectively observed after 25(OH)D assessment.

In addition to the presence or absence of atelectasis, each case was classified morphologically and anatomically according to radiologic patterns documented in the reports, distinguishing laminar, fibroatelectatic, segmental/lobar, and tractional atelectasis, as well as the predominant lobar distribution (lingula, lower lobes, or other regions), in accordance with radiologic terminology standardized by the Fleischner Society [8]. These qualitative radiologic features are summarized in Table 1, which describes the radiologic patterns and predominant lobar distribution of atelectasis according to baseline serum 25(OH)D status. Atelectasis patterns were classified based on their radiologic appearance as fibrotic/laminar, segmental/lobar, or tractional. Serum 25(OH)D categories were defined according to baseline measurements.

Table 1.

Radiologic patterns and lobar distribution of atelectasis in patients with CVID according to serum 25(OH)D status.

Baseline 25(OH)D category	n	Atelectasis history	Pattern type	Main lobar distribution
25(OH)D < 30 ng/mL at baseline	15	Atelectasis documented throughout disease course	Fibrotic/laminar (65%); Segmental/lobar (29%); Tractional (6%)	Lingula or left lower lobe (~65%)
25(OH)D ≥ 30 ng/mL at baseline	1	Atelectasis documented prior to or throughout disease course	Laminar	Left lower lobe

Notes: 25(OH)D, 25-hydroxyvitamin D; CVID, common variable immunodeficiency. Atelectasis may have occurred before or after the baseline 25(OH)D measurement.

Secondary pulmonary manifestations included bronchiectasis, recurrent pneumonia (≥ 2 episodes within 12 months [4]), and chronic obstructive pulmonary disease (COPD), confirmed by imaging and clinical criteria. Imaging studies were reviewed at the Institute of Radiology, HC-FMUSP. Additional exploratory variables included systemic manifestations frequently described in CVID: lymphadenopathy, splenomegaly, hepatomegaly, portal hypertension, and autoimmunity.

Laboratory Analysis

Serum 25(OH)D levels were measured by chemiluminescent immunoassay using an automated analyzer (Cobas e601, Roche Diagnostics) at the Central Laboratory Division, HC-FMUSP (São Paulo, Brazil). Immunoglobulin (IgG, IgA, and IgM) levels were quantified by immunoturbidimetric assay (BN ProSpec, Siemens Healthcare), with blood samples obtained immediately before intravenous immunoglobulin (IVIg) infusion. Lymphocyte subsets (CD3⁺, CD4⁺, CD8⁺, and CD19⁺) were enumerated by flow cytometry (FACSCanto II, BD Biosciences).

Exploratory Longitudinal Sub-Analysis

An exploratory longitudinal sub-analysis was performed to descriptively compare available thoracic imaging studies with the nearest corresponding 25(OH)D measurements over time. All imaging and laboratory data from patients who developed atelectasis during follow-up were reviewed. Because 25(OH)D sufficiency was maintained or achieved in only 1 patient, transitions between sufficient and insufficient status were extremely limited. Accordingly, this component of the analysis was considered descriptive rather than inferential. Atelectasis was classified as present or absent at each time point, and 25(OH)D status was categorized as insufficient or sufficient. Given that discordant sufficient/insufficient 25(OH)D pairs were rare, McNemar’s test was not applied. Instead, paired observations were descriptively summarized to characterize stability or variation over time.

Kaplan-Meier curves were constructed for the full 48-patient cohort as an exploratory descriptive analysis to illustrate the distribution of atelectasis-free disease duration according to baseline 25(OH)D category. Baseline 25(OH)D was defined as the earliest available serum measurement within the study period. Atelectasis events documented at any time in available medical records, including those occurring before or after the baseline 25(OH)D measurement, were included to characterize the overall disease burden. Disease duration (years from CVID symptom onset to the first documented radiologic evidence of atelectasis or censoring at the last follow-up) was regarded as the time variable. The numbers of events and censored observations were recorded for descriptive reporting.

Because baseline 25(OH)D measurements were not necessarily obtained before the first occurrence of atelectasis, and because the timing of events was not consistently aligned with exposure assessment, the Kaplan-Meier analysis was not intended to establish temporal sequence, causality, or prognostic effects; it should be interpreted as descriptive only. This approach reflects the cross-sectional nature of the study and the lack of temporal alignment between exposure and outcome, precluding time-to-event inference.

Statistical Analysis

Analyses were performed using IBM SPSS Statistics (version 18.0; IBM Corp., Armonk, NY, USA) and GraphPad Prism (version 10.6.1; GraphPad Software, San Diego, CA, USA) for graphical output. Data normality was assessed via the Shapiro-Wilk test. Continuous variables were compared using Student’s t-test or the Mann-Whitney U test, as appropriate; categorical variables were compared using Fisher’s exact test or the chi-square test.

IVIg dose was compared between 25(OH)D groups using the Mann-Whitney U test. Correlations between IVIg dose and serum IgG levels were assessed using Spearman’s rank correlation coefficient. Longitudinal imaging and 25(OH)D data were summarized descriptively. Kaplan-Meier curves were generated for exploratory descriptive visualization, without formal time-to-event inference (eg, log-rank testing or hazard ratio estimation) due to the lack of temporal alignment between exposure and outcome. Two-tailed P-values < 0.05 were considered statistically significant for the primary cross-sectional analyses.

Results

Baseline Characteristics of the CVID Cohort

The study included 48 adults with CVID, of whom 29 (60.4%) were women and 19 (39.6%) were men. The mean age was 37.15 ± 7.65 years. The mean age at symptom onset was 15.92 ± 10.15 years, and the mean age at CVID diagnosis was 24.04 ± 9.24 years, resulting in a mean diagnostic delay of 9.42 ± 9.01 years. The mean disease duration was 21.00 ± 9.78 years. Patients were stratified according to serum 25(OH)D levels: 35 (73%) had concentrations below 30 ng/mL, and 13 (27%) had concentrations of at least 30 ng/mL (Table 2).

Table 2.

Baseline demographic, clinical, and immunological characteristics of patients with CVID according to serum 25(OH)D status.

Demographic / clinical parameter	CVID (n = 48)	25(OH)D < 30 ng/mL (n = 35)	25(OH)D ≥ 30 ng/mL (n = 13)	P
Age, years	37.15 ± 7.65	36.83 ± 7.50	38.00 ± 8.00	0.16
Age at symptom onset, years	15.92 ± 10.15	15.20 ± 8.90	18.10 ± 7.40	0.24
Age at diagnosis, years	24.04 ± 9.24	23.90 ± 9.30	24.50 ± 9.80	0.84
Diagnostic delay, years	9.42 ± 9.01	9.29± 7.92	10.31 ± 11.36	0.73
Disease duration, years	21.00 ± 9.78	21.00 ± 10.00	17.00 ± 9.00	0.08
Female sex, n (%)	29 (60.42)	22 (62.86)	7 (53.85)	0.72
Male sex, n (%)	619 (39.58)	13 (37.14)	6 (46.15)	0.72
Immunoglobulins
IgG, mg/dL	491 ± 215	465 ± 220	561 ± 191	0.04*
IgA, mg/dL	11.60 ±30.90	13.00 ± 34.00	8.00 ± 14.00	0.55
IgM, mg/dL	13.00 ±18.40	11.00 ± 20.00	16.00 ± 16.00	0.47
Lymphocyte subsets
CD3+, cells/μL	1483 ± 546	1378 ± 457	1262 ± 551	0.41
CD4+, cells/μL	679 ± 302	660 ± 305	599 ± 314	0.52
CD8+, cells/μL	736 ± 352	706 ± 584	637 ± 353	0.60
CD19+, cells/μL	174 ± 115	151 ± 105	134 ± 95	0.48

Notes:

^*P < 0.05.

Values are presented as mean ± standard deviation. Reference ranges for adults: IgG, 700–1600 mg/dL; IgA, 70–400 mg/dL; IgM, 40–230 mg/dL; lymphocyte subsets (cells/μL): CD3⁺, 605–2460; CD4⁺, 493–1666; CD8⁺, 224–1112; CD19⁺, 100–500. P-values refer to comparisons between groups.

Comparison of Patients by 25(OH)D Status

Patients with 25(OH)D concentrations below 30 ng/mL and those with concentrations of at least 30 ng/mL were comparable in age (36.83± 7.50.00 vs 38.00 ± 8.00 years, P = 0.16), age at symptom onset (15.20 ± 8.90 vs 18.10 ± 7.40 years, P = 0.24), age at diagnosis (23.90 ± 9.30 vs 24.50 ± 9.80 years, P = 0.84), and diagnostic delay (9.29 ± 7.92 vs 10.31 ± 11.36 years, P = 0.73). Disease duration showed a nonsignificant trend toward being longer in the 25(OH)D-insufficient group (21.00 ± 10.00 vs 17.00 ± 9.00 years, P = 0.08). Sex distribution was similar between groups (female: 62.9% vs 53.8%, P = 0.72) (Table 2).

Immunoglobulin and Lymphocyte Profiles

Patients with sufficient 25(OH)D levels had significantly higher serum IgG concentrations than those with insufficient levels (561 ± 191 vs 465 ± 220 mg/dL; P = 0.04). No significant differences were observed in IgA or IgM levels between groups. All patients received regular immunoglobulin replacement therapy every 28 days. The IVIg dose did not significantly differ between 25(OH)D groups (U = 204, P = 0.42), and no significant correlation was observed between IVIg dose and serum IgG levels (Spearman’s rho = −0.25, P = 0.09). Lymphocyte subsets (CD3⁺, CD4⁺, CD8⁺, and CD19⁺) also were comparable between 25(OH)D groups (all P > 0.40) (Table 2).

Pulmonary Manifestations

Patients with sufficient 25(OH)D levels exhibited a significantly lower prevalence of atelectasis than those with insufficient levels (8% vs 43%; odds ratio [OR] = 0.12; 95% confidence interval [CI], 0.01–0.97; P = 0.02). The OR compares the odds of atelectasis in patients with 25(OH)D concentrations of at least 30 ng/mL relative to those with concentrations below 30 ng/mL, corresponding to an absolute difference of 35%.

No significant differences were observed for bronchiectasis, recurrent pneumonia, or COPD (Table 3). These differences in pulmonary manifestations according to serum 25(OH)D status are visually summarized in Figure 2, which highlights the pronounced contrast in atelectasis prevalence between groups.

Table 3.

Pulmonary manifestations in patients with CVID according to serum 25(OH)D status.

Outcome	25(OH)D < 30 ng/mL (n = 35)	25(OH)D ≥ 30 ng/mL (n = 13)	Odds ratio [95% Confidence interval]	P
Atelectasis	43% (15)	8% (1)	0.12 [0.01–0.97]	0.02*
Bronchiectasis	63% (22)	46% (6)	0.51 [0.14–1.84]	0.53#
Recurrent pneumonia	91% (32)	84% (11)	0.52 [0.08–3.26]	0.49#
Chronic obstructive pulmonary disease	3% (1)	8% (1)	2.80 [0.16–48.9]	0.46

Notes:

^*P < 0.05;

^#Fisher’s exact test.

Odds ratios compare the odds of each outcome in patients with 25(OH)D ≥ 30 ng/mL relative to those with 25(OH)D < 30 ng/mL.

Figure 2.

(A) Distribution of pulmonary manifestations (atelectasis, bronchiectasis, recurrent pneumonia, and chronic obstructive pulmonary disease [COPD]) according to 25(OH)D status. (B) Atelectasis prevalence was significantly lower among patients with sufficient 25(OH)D levels (≥ 30 ng/mL) than among those with insufficient levels (< 30 ng/mL) (P = 0.02; odds ratio = 0.12; 95% confidence interval, 0.01–0.97).

Bronchiectasis prevalence was similar between groups (P = 0.53, Fisher’s exact test), and its radiologic characteristics were heterogeneous, without consistent predominance across 25(OH)D categories (Table 4). No significant association was observed for recurrent pneumonia (91% vs 84%; P = 0.49, Fisher’s exact test).

Table 4.

Radiologic patterns and anatomic distribution of bronchiectasis in patients with CVID according to serum 25(OH)D status.

Variable	25(OH)D < 30 ng/mL (n = 35)	25(OH)D ≥ 30 ng/mL (n = 13)
Bronchiectasis (any)	22 (63%)	6 (46%)
Cylindrical	3 (9%)	3 (23%)
Traction	3 (9%)	0 (0%)
Varicose/cystic	2 (6%)	0 (0%)
Incipient/mild	0 (0%)	1 (8%)
Not specified	7 (20%)	1 (8%)
Associated with atelectasis	12 (34%)	1 (8%)
Lower lobe predominance	3 (9%)	0 (0%)
Bilateral involvement	1 (3%)	0 (0%)

Notes: Values are presented as number (%). Bronchiectasis patterns and anatomic distribution were classified according to radiologic descriptions documented in the medical records. Categories are not mutually exclusive. This table is descriptive and intended to characterize structural lung disease as a potential confounding factor in the interpretation of atelectasis.

Extrapulmonary Manifestations

No significant differences were observed between the 25(OH)D-sufficient and 25(OH)D-insufficient groups in lymphadenopathy (22% vs 46%), splenomegaly (37% vs 53%), hepatomegaly (34% vs 23%), portal hypertension (22% vs 23%), or autoimmunity (31% vs 15%) (all P > 0.05) (Table 5).

Table 5.

Extrapulmonary manifestations in patients with CVID according to serum 25(OH)D status.

Manifestations	25(OH)D < 30 ng/mL (n = 35)	25(OH)D ≥ 30 ng/mL (n = 13)	P
Lymphadenopathy	22.9% (8)	46.2% (6)	0.12
Splenomegaly	37.1% (13)	53.8 (7)	0.30
Hepatomegaly	34.3% (12)	23.1% (3)	0.46
Portal hypertension	22.9% (8)	23.1% (3)	0.99
Autoimmunity	31.4% (11)	15.4% (2)	0.27

Notes: Values are presented as number (%). Comparisons between groups were performed using Fisher’s exact test. P < 0.05 was considered statistically significant. 25(OH)D, 25-hydroxyvitamin D; CVID, common variable immunodeficiency.

Exploratory Longitudinal Sub-Analysis

All available thoracic imaging studies were reviewed along with the corresponding 25(OH)D measurements in patients who had documented atelectasis. Sixteen patients were evaluable; 15 exhibited persistent 25(OH)D insufficiency, and 1 maintained sustained sufficiency throughout follow-up. Atelectasis remained radiologically persistent in all cases. Predominant patterns included fibrotic/laminar atelectasis (65%), followed by segmental or lobar collapse (29%) and tractional atelectasis (6%). The lingula and left lower lobe were the most frequently affected regions (~65%) (Table 1). Because transitions between 25(OH)D categories were exceedingly rare and atelectasis remained chronically stable, this longitudinal assessment was considered descriptive rather than inferential. This sub-analysis was not intended to infer temporal relationships between 25(OH)D status and atelectasis but rather to provide descriptive insight into the available paired data.

Kaplan-Meier Survival Analysis

A Kaplan-Meier analysis was performed in the full 48-patient cohort as an exploratory descriptive approach to illustrate the distribution of atelectasis-free disease duration according to baseline 25(OH)D status. No statistically significant separation between curves was observed.

Overall, 15 atelectasis events occurred in the group with 25(OH)D concentrations below 30 ng/mL, whereas 1 event occurred in the group with concentrations of at least 30 ng/mL; 20 and 12 patients, respectively, were censored at the last follow-up. Although most atelectasis events occurred in patients with 25(OH)D concentrations below 30 ng/mL, only a single event was observed in the sufficient group, limiting meaningful comparisons between strata.

Given the small number of events in the 25(OH)D-sufficient group and the lack of temporal alignment between exposure and outcome, this analysis is presented as a descriptive visualization and does not support inferences regarding disease progression or temporal relationships. Corresponding Kaplan-Meier curves are shown in Figure 3.

Figure 3.

Kaplan-Meier curves of atelectasis-free survival in patients with common variable immunodeficiency (CVID) according to serum 25(OH)D levels (≥ 30 ng/mL vs < 30 ng/mL). Tick marks indicate censored observations, and the table shows the number of patients at risk over time.

Discussion

The cohort analyzed in this study included 48 patients with CVID, with a mean age of 37 years and a predominance of women (60%). The upper age limit of 50 years was intentionally defined to maintain cohort homogeneity and minimize potential confounding by age-related comorbidities [20]. The long mean disease duration (21 years) and diagnostic delay (9 years) reflect the chronic nature of CVID, as well as the well-recognized gap between symptom onset and definitive diagnosis [4]. Importantly, age, diagnostic delay, and disease duration did not significantly differ according to baseline serum 25(OH)D status (Table 2), indicating that baseline 25(OH)D groups were comparable with respect to key disease chronology parameters.

Most patients presented with recurrent pneumonia (43/48, 90%; Table 3), consistent with the clinical spectrum described in other CVID cohorts, in which respiratory manifestations reflect the combined effects of chronic infection, immune dysfunction, airway remodeling, and—in some cases—systemic involvement [21]. These manifestations often persist despite conventional therapy and represent major pulmonary comorbidities in this population. Within this clinical context, the identification of modifiable factors that could mitigate respiratory damage, such as 25(OH)D status, becomes particularly relevant.

To our knowledge, this is the first study to show that sufficient serum 25(OH)D levels (≥ 30 ng/mL) are significantly associated with a lower prevalence of atelectasis in patients with CVID. These findings support emerging evidence linking 25(OH)D insufficiency to impaired pulmonary function and heightened inflammatory activity, mechanisms that can compromise alveolar stability and thereby contribute to atelectasis [10,13].

Atelectasis is not only a structural consequence of chronic lung disease but also a dynamic and potentially reversible condition, as demonstrated in other clinical settings. In a prospective study, Yang et al showed that lung recruitment maneuvers significantly reduced ultrasound-detected atelectasis in older patients undergoing laparoscopic surgery [22]. Although the population studied by Yang et al differs from patients with CVID, these findings emphasize that alveolar collapse may be modifiable rather than solely a consequence of irreversible structural damage.

Within this framework, the association between vitamin D sufficiency and reduced atelectasis in CVID appears biologically plausible and may be explained through 3 converging pathways. First, mechanistically, the active metabolite 1,25(OH)₂D₃ induces transcription of antimicrobial peptides such as cathelicidin (LL-37) and β-defensins in epithelial cells and macrophages, enhancing mucosal barrier integrity and accelerating pathogen clearance [23]. These molecular effects are supported by clinical evidence indicating that individuals with low 25(OH)D levels exhibit increased susceptibility to respiratory infections, whereas maintenance of sufficiency is associated with a lower risk of infection in large population-based studies [24]. In CVID, where recurrent bacterial infections and impaired mucosal defense directly contribute to airway injury and remodeling [25], these mechanisms could plausibly reduce mucus plugging and small-airway obstruction, both recognized precursors of resorptive atelectasis [8].

Second, at the alveolar level, 1,25(OH)₂D₃ has been shown in experimental models to stimulate surfactant-related pathways and phospholipid synthesis in type II pneumocytes, mechanisms that reduce surface tension and may decrease the propensity for alveolar collapse [13]. Although these data are derived from preclinical rather than CVID-specific settings, they provide biologically plausible support for a potential protective effect. This interpretation is consistent with emerging evidence linking 25(OH)D insufficiency to systemic inflammatory dysregulation [26] and with well-established evidence demonstrating inflammation-mediated pulmonary vulnerability in CVID [27].

Third, through vitamin D receptor-mediated transcriptional signaling, vitamin D may mitigate profibrotic remodeling, including transforming growth factor-β-driven pathways described in experimental studies [13]. In the CVID context, inflammatory and fibrotic processes have been described in granulomatous-lymphocytic interstitial lung disease and in cases complicated by chronic cytomegalovirus infection [7,28]. Within this framework, 25(OH)D may help preserve alveolar stability and limit atelectasis secondary to inflammatory remodeling. In our cohort, laminar and fibroatelectatic patterns were predominant (approximately 65%), consistent with nonobstructive or tractional forms of atelectasis rather than acute obstructive collapse [8,28]. These lesions were most frequently localized to the lingula and left lower lobe, regions classically associated with gravity-dependent ventilation-perfusion imbalance and chronic inflammatory remodeling [29].

Bronchiectasis, a common structural complication of CVID, may contribute to the development of atelectasis through airway distortion, mucus retention, and regional ventilation impairment [30]. In the present study, however, the prevalence and radiologic characteristics of bronchiectasis were similar across 25(OH)D groups (Tables 3, 4). Therefore, although bronchiectasis may contribute to atelectasis in CVID, the observed association between 25(OH)D sufficiency and lower atelectasis prevalence does not appear to be explained solely by differences in concomitant bronchiectasis. Other mechanisms might also contribute, including infection-related airway inflammation and alterations in alveolar stability and surfactant function, which have been implicated in atelectasis pathophysiology under chronic inflammatory lung conditions [13].

Recent translational studies suggest that atelectasis is not merely a mechanical consequence of regional volume loss but also a site of active inflammatory signaling. In a comprehensive proteomic analysis of human lung tissue, Rashid et al demonstrated that atelectatic regions exhibit heightened activation of alveolar macrophages and neutrophils, upregulation of cytokine-related pathways, and enrichment of immune response proteins [31]. These findings align with the morphological and topographic predominance of laminar and fibroatelectatic forms observed in our cohort, suggesting that apparently passive structural changes might reflect localized inflammatory and immunoregulatory processes driven by macrophage-neutrophil crosstalk within areas of impaired alveolar ventilation.

This chronic alveolar inflammatory microenvironment, dominated by activated macrophages and neutrophils, may represent the context in which vitamin D exerts its modulatory effects. Experimental analyses have shown that vitamin D enhances macrophage regulatory programs, promoting anti-inflammatory gene expression and microRNA-mediated pathways that limit excessive activation [15]. Through these vitamin D receptor-dependent mechanisms, vitamin D can reduce the production of proinflammatory cytokines (eg, tumor necrosis factor-α, interleukin [IL]-6, and IL-1β) and promote a more regulated macrophage phenotype, thereby attenuating the persistent inflammatory milieu that may contribute to fibroatelectatic remodeling [14].

Additionally, excessive T helper type 1 and T helper type 17 activities, driven by increased interferon-γ and IL-17 release, promote persistent neutrophilic infiltration and tissue injury [32]. These processes may compromise alveolar stability and predispose the lung to collapse, further supporting the biological plausibility of our radiologic findings. Conversely, vitamin D-mediated expansion of regulatory T cells, together with increased IL-10 production, exerts a counterregulatory effect that mitigates these injurious pathways [33].

Another parameter of interest was the immunological profile. Prior evidence indicated that abnormalities in B- and T-cell compartments, particularly imbalances between CD4⁺ and CD8⁺ subsets, are strongly associated with pulmonary involvement [5,34]. For this reason, such subgroups were prioritized in our analysis. Additionally, we assessed CD3⁺ (total T lymphocytes) and CD19⁺ (total B lymphocytes) to broaden the immunological characterization—alterations within these compartments have been linked to noninfectious complications in CVID [35].

In patients with CVID, IVIg therapy is the standard of care, providing passive IgG replacement and reducing the frequency and severity of recurrent respiratory infections, thus limiting chronic airway inflammation and slowing the progression of structural lung damage (eg, bronchiectasis), which may contribute to the development of atelectasis [4]. This pattern underscores the importance of considering underlying structural lung disease when interpreting atelectasis in immunodeficient patients.

In our cohort, 25(OH)D sufficiency was associated with higher serum IgG concentrations (Table 2). The IVIg dose did not differ between 25(OH)D groups; no significant correlation was observed between IVIg dose and serum IgG levels, suggesting that differences in IgG concentrations are unlikely to be explained by variations in immunoglobulin replacement therapy. This finding should be interpreted with caution because the analysis was exploratory and not a primary objective of the study, which focused on pulmonary outcomes. Therefore, the observed difference in IgG concentrations should be considered hypothesis-generating. Although previous research suggests that vitamin D can influence B-cell differentiation and antibody production [32], this biological plausibility was not directly assessed in the present cohort. Moreover, recent evidence suggests that systemic inflammation acts as a mediator linking vitamin D status to immune competence, including humoral responses [26]. Although this finding should be interpreted cautiously, it raises the possibility that maintenance of 25(OH)D sufficiency may contribute both to alveolar stability and to broader aspects of immune function in CVID.

We also evaluated extrapulmonary manifestations commonly associated with pulmonary disease progression in CVID—including lymphadenopathy, splenomegaly, hepatomegaly, portal hypertension, and autoimmunity—but found no associations with 25(OH)D status in our cohort (Table 5).

Our findings of a significant association between 25(OH)D sufficiency and reduced atelectasis should be interpreted within the broader clinical spectrum of CVID. Earlier cohort studies established the high prevalence of chronic lung disease and extrapulmonary complications in this population [2]. These findings were subsequently confirmed and refined by systematic evidence [1]; more recently, multicenter studies have further emphasized the strong associations of noninfectious pulmonary manifestations with autoimmunity and impaired survival [6]. In this context, the absence of an association with extrapulmonary manifestations in our cohort likely reflects the limited sample size and phenotypic variability, potentially indicating that the effect of 25(OH)D is more specific to atelectasis than to systemic complications.

Consistent with the exploratory Kaplan-Meier curves (Figure 3), no clear separation between groups was observed. Although most atelectasis events occurred among patients with 25(OH)D concentrations below 30 ng/mL, only a single event was observed in the sufficient group, limiting meaningful comparisons between strata. Given the lack of temporal alignment between exposure and outcome, along with the small number of events in the sufficient group, these curves should be strictly interpreted as descriptive. They provide a graphical illustration consistent with the cross-sectional association but do not support inferences regarding disease progression, temporal relationships, or protective effects.

The near absence of atelectasis events in the 25(OH)D-sufficient group, in conjunction with the small number of patients at risk during later disease-duration intervals, limited the statistical power for inferential testing. Importantly, serum 25(OH)D was assessed only at baseline. The single atelectasis event observed in the sufficient group occurred in a patient whose 25(OH)D concentration subsequently declined below 30 ng/mL, highlighting the potential temporal variability of 25(OH)D status and the possibility of exposure misclassification. Such misclassification would be expected to bias associations toward the null.

Collectively, our findings indicate a consistent association between 25(OH)D sufficiency and a lower prevalence of atelectasis in CVID, reinforcing the observed cross-sectional patterns. Although these results do not support causal or temporal inferences, they provide a rationale for further investigation in prospective studies incorporating longitudinal imaging, detailed immune phenotyping, and immunometabolic profiling.

Limitations

This study has some limitations. Quantitative radiologic measures of atelectasis severity were not available, and imaging studies were not centrally reviewed, limiting the assessment of dose-response relationships. Key inflammatory and molecular pathways potentially linking 25(OH)D to lung injury were not directly measured; instead, they were inferred from existing literature. The small number of patients with sufficient 25(OH)D levels and the single-center design limit statistical power and generalizability. Additionally, IVIg dosing was recorded only in grams per infusion (g/28 days); weight and height data were incomplete in medical records, precluding weight-adjusted analyses. Although longitudinal review of the medical records allowed estimation of average IgG levels over time, unmeasured factors (eg, variability in treatment adherence and potential constraints regarding access to immunoglobulin therapy in public health care settings) may have influenced serum IgG levels and cannot be excluded.

Conclusions

In patients with CVID, serum 25(OH)D levels of at least 30 ng/mL were associated with a lower prevalence of atelectasis; no associations were observed with bronchiectasis, recurrent pneumonia, or COPD. These findings reflect a cross-sectional association within this cohort and should not be interpreted as evidence of temporal or causal relationships. Prospective longitudinal studies are needed to clarify the role of 25(OH)D status in the development and progression of atelectasis in CVID.