Subclinical retinal and choroidal microvascular involvement in pediatric asthma: an OCTA-based analysis of corticosteroid exposure

Abstract

Background

Childhood asthma is a chronic inflammatory and intermittently hypoxic condition that may exert systemic effects beyond the respiratory system. These processes may influence retinal and choroidal microvascular structures. Optical coherence tomography (OCT) and optical coherence tomography angiography (OCTA) allow non-invasive and quantitative evaluation of posterior segment neuronal and vascular structures. This study aimed to evaluate retinal and choroidal microvascular parameters in children with asthma without clinical ocular findings, to analyze differences according to inhaled corticosteroid (ICS) dose, and to compare the findings with healthy controls.

Methods

This cross-sectional study included 56 children with asthma and 30 age- and sex-matched healthy controls aged 5–18 years. Asthma diagnosis was established according to Global Initiative for Asthma (GINA) guidelines. Patients were categorized as low-dose or moderate-high-dose ICS users based on their treatment during the previous year. OCTA was used to measure vessel density (VD) in the superficial capillary plexus (SCP), deep capillary plexus (DCP), and choriocapillaris (CC), and OCT was used to measure central macular thickness (CMT). Group comparisons were performed. P values were corrected for multiple comparisons using the Benjamini-Hochberg false discovery rate (FDR) method. Discriminative parameters were evaluated using Receiver Operating Characteristic (ROC) analysis, and variables were included in univariate and multivariate logistic regression models.

Results

There were no significant differences between groups in terms of age and sex (p > 0.05). Vessel density values in the SCP, DCP, and CC were significantly lower in the asthma group compared with controls across all quadrants (all FDR-corrected p < 0.05). CMT was significantly thinner in asthmatic children (p < 0.001). In multivariate analysis, nasal CC vessel density remained statistically significant (p < 0.001). ROC analysis showed an area under the curve of 0.966 for nasal CC vessel density. In subgroup analysis according to ICS dose, no parameters remained statistically significant after FDR correction (all FDR-corrected p > 0.05).

Conclusions

Asthma may be associated with subclinical posterior segment microvascular alterations. The absence of significant differences according to ICS dose after multiple comparison correction suggests that these changes are primarily related to the disease itself rather than treatment exposure. OCT and OCTA may contribute to early detection and follow-up.

Introduction

Childhood asthma is a common chronic respiratory disease characterized by chronic airway inflammation and variable symptoms such as wheezing, dyspnea, cough, and chest tightness [1]. The prevalence of the disease ranges from 3% to 37.6% and varies according to countries and age groups [2, 3]. The main goals of asthma treatment are to achieve symptom control and reduce the frequency of exacerbations. Inhaled corticosteroids (ICS), which are widely used for this purpose, have been shown to significantly reduce airway hyperresponsiveness and inflammation [4, 5]. However, asthma is not limited to the respiratory system and is associated with various systemic effects and comorbidities [6–8].

Systemic inflammation, hypoxia, and particularly long-term use of ICS in asthma may lead to various ocular changes, especially in the pediatric population. While the anterior segment effects of corticosteroids, such as cataract and glaucoma, are well established, their impact on retinal structures has attracted increasing attention in recent years. Systemic or ICS have been reported to affect chorioretinal blood flow and may constitute a risk factor for retinal pathologies such as central serous chorioretinopathy [9–11]. In this context, investigating the subclinical effects of both the chronic inflammation associated with asthma and the dose of ICS used in its treatment on the neuronal and vascular structures of the posterior segment is of critical importance. Subclinical anterior segment changes, including increased meibomian gland atrophy and reduced tear film stability, have also been reported in children with newly diagnosed allergic disease, highlighting the relevance of evaluating ocular involvement in asymptomatic pediatric patients [12]. Optical coherence tomography (OCT) and optical coherence tomography angiography (OCTA) are rapid and non-invasive imaging modalities that allow detailed evaluation of these structures [13].

The aim of this study was to compare retinal and choroidal microvascular structures using OCT and OCTA among healthy controls and children with asthma without clinical ocular involvement receiving low and moderate-high doses of ICS, and to identify subclinical changes that may be associated with asthma itself or ICS dose.

Methods

Patients aged 5–18 years who were followed with a diagnosis of asthma at the Pediatric Allergy Clinic between August 2025 and January 2026, as well as a healthy control group within a similar age range, were included in the study. Clinical data, including demographic characteristics, allergen sensitization findings based on skin prick testing, and the presence of accompanying allergic diseases, along with laboratory parameters (complete blood count, total immunoglobulin E, eosinophil count and percentage, and inhalant-specific immunoglobulin E), were retrospectively obtained from electronic medical records. OCT and OCTA images were evaluated. These images were obtained as part of routine clinical ophthalmologic assessment of pediatric asthma patients receiving ICS therapy and were retrospectively analyzed within the scope of this study.

The diagnosis of asthma was established according to the Global Initiative for Asthma (GINA) guidelines [1], based on the presence of episodic and variable respiratory symptoms such as dyspnea, wheezing, cough, and chest tightness, together with variable expiratory airflow limitation. Early reversibility was defined as an increase of ≥ 12% in forced expiratory volume in one second (FEV₁) compared with baseline, or an increase of ≥ 15% in peak expiratory flow (PEF), measured by spirometry 10–15 min after inhalation of 200–400 µg salbutamol. The presence of similar improvements after four weeks of daily ICS therapy was defined as late reversibility and was considered supportive of the asthma diagnosis. When the follow-up data from the previous year were evaluated, 17 patients were classified as having partly controlled asthma, one patient as uncontrolled asthma, and the remaining patients as well-controlled asthma. Full asthma control was achieved in all patients during the last six months prior to OCT/OCTA imaging.

Atopic sensitization was confirmed by a positive response to at least one aeroallergen on skin prick testing or the presence of serum-specific immunoglobulin E (IgE). Specific IgE measurements were performed using the IMMULITE^® immunoassay method (Siemens, USA), and a cut-off value of 0.35 kU/L was used to define allergic sensitization. Skin prick testing was performed in the pediatric allergy clinic test room using standardized allergen extracts (Diater Laboratorios, Madrid, Spain). Histamine hydrochloride solution (10 mg/mL) was used as the positive control. Skin prick test results were evaluated 15 min after allergen application, and an induration of ≥ 3 mm compared with the negative control (saline) was considered a positive reaction. The tested allergens included Alternaria, Dermatophagoides farinae, Dermatophagoides pteronyssinus, grass pollen mix, Phleum pratense, cereal pollen mix, weed pollen mix, tree pollen mix, Cupressus sempervirens, olive tree pollen, Secale cereale, cat allergen, and dog allergen.

Asthmatic patients were classified into two groups according to their use of ICS during the previous year, including monotherapy with ICS (fluticasone propionate) and combination therapy (fluticasone propionate/salmeterol or budesonide/formoterol). Patients receiving ICS were further stratified into low-dose and moderate-high-dose groups based on treatment dosage, according to the stepwise treatment recommendations of the GINA guidelines [1]. In patients receiving combination therapy, the dose of long-acting β₂-agonists (LABA) was not independently standardized, as fixed-dose combination formulations were used; therefore, LABA dose varied proportionally with the ICS dose.

The control group consisted of age- and sex-matched children with no history of systemic inflammatory disease, respiratory disease, or parasitic infection.

Healthy controls were selected from children with no known history of allergic disease or clinical findings suggestive of atopy; however, no laboratory allergy testing was performed in this group due to the absence of clinical indication.

To minimize potential confounding effects, participants with BMI percentiles outside the normal range (5th-84th percentile) or abnormal blood pressure values were excluded from the study. BMI was evaluated according to age- and sex-specific percentiles. Blood pressure measurements were within normal limits in all included participants.

Participants with refractive errors greater than ± 3.00 diopters, amblyopia, corneal pathology, retinal diseases, glaucoma, optic disc anomalies, uveitis, or strabismus; those with a history of intraocular surgery or ocular trauma; and those with chronic systemic diseases or genetic syndromes were excluded from the study. All participants underwent a comprehensive ophthalmologic examination.

Optical coherence tomography and optical coherence tomography angiography

OCT and OCTA imaging were performed using a swept-source system (OCT-HS100, Canon Inc., Tokyo, Japan), and image analysis was conducted using the device software RX Capture for OCT-HS100 (Version 4.5.0.13). The macular region was evaluated using OCT, and central macular thickness (CMT) was recorded using the device’s automated segmentation algorithm. One randomly selected eye from each participant was included in the analysis. Using OCTA, vessel density of the superficial capillary plexus (SCP), deep capillary plexus (DCP), and choriocapillaris (CC) was measured in standard quadrants.

OCTA analyses were performed using 3 × 3 mm² en-face scans centered on the macula. The acquired images were analyzed using the device’s built-in software, and VD measurements were obtained at the levels of the SCP, DCP, and CC. Vessel density was defined as the percentage of the area occupied by blood vessels. Analyses were conducted at the foveal center and in the parafoveal quadrants (nasal, temporal, superior, and inferior).

To ensure image quality, scans with low signal strength, segmentation errors, or significant motion artifacts were excluded from the analysis.

Ethical statement

The study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Bursa City Hospital Scientific Research Ethics Committee (approval no.: 2025/12 − 8). Due to the retrospective design of the study, informed consent was not obtained from the participants.

Statistics

Statistical analyses were performed using SPSS software (IBM SPSS Statistics for Windows, Version 27.0; IBM Corp., Armonk, NY, USA). Descriptive statistics were presented as mean ± standard deviation for normally distributed variables, median (minimum-maximum) for non-normally distributed variables, and number (%) for categorical variables.

The normality of continuous variables was assessed using the Kolmogorov-Smirnov and Shapiro-Wilk tests. Comparisons of independent continuous variables were performed using the independent samples t-test for normally distributed data and the Mann-Whitney U test for non-normally distributed data. Categorical variables were compared using the chi-square test.

The discriminative potential of parameters to distinguish between asthma and control groups was evaluated using in-sample receiver operating characteristic (ROC) curve analysis, and cut-off values were calculated. Because the ROC analysis was performed on the same dataset used for model development, the reported AUC values are in-sample estimates and may overestimate the true discriminative performance. The effects of variables on group discrimination were first analyzed using univariate logistic regression, followed by multivariate logistic regression (Forward LR method) for variables found to be significant.

To minimize Type I error due to multiple comparisons of OCTA and OCT parameters, p-values were corrected using the Benjamini-Hochberg False Discovery Rate (FDR) procedure. For all comparisons, an FDR-corrected p-value of < 0.05 was considered statistically significant.

Results

A total of 56 pediatric patients with asthma and 30 healthy controls were included in the study. The mean age was 12.4 ± 3.2 years in the asthma group and 12.6 ± 3.1 years in the control group (p = 0.888). There were no significant differences between the groups in terms of age or sex distribution (p = 0.983). The demographic and clinical characteristics of the participants are presented in Table 1.

Table 1.

Demographic and clinical characteristics of the participants

Variable	Control Group (n = 30)	Asthma Group (n = 56)	p
Age (years)	12.6 ± 3.1	12.4 ± 3.2	0.888
Sex (Female), n (%)	16 (53.3%)	30 (53.6%)	0.983
Intraocular pressure (mmHg) BMI percentile	14.6 ± 3 -	14.7 ± 3 50.0 ± 18.0	0.898 -
Systolic blood pressure (mmHg)	-	101.0 ± 7.0	-
Diastolic blood pressure (mmHg)	-	65.0 ± 5.0	-
Age at asthma diagnosis (years)	-	9.2 ± 4.0	-
Duration of asthma (months)	-	35.0 (6.0-168.0)	-
Total IgE (IU/mL)	-	178.0 (11.0-2480.0)	-
Eosinophil count (/µL)	-	230.0 (10.0-1200.0)	-
FEV₁ (%)	-	86.0 (51.0-113.0)	-
FEV₁/FVC (%)	-	95.0 (46.0-113.0)	-
PEF (%)	-	75.0 (15.0-128.0)	-
FEF 25–75 (%)	-	77.0 (43.0-132.0)	-
WBC (/µL)	-	7690 (1177–17000)	-
SII	-	469.0 (172.0-1501.0)	-
Asthma control - well-controlled, n (%)	-	38 (67.9%)	-
Asthma control - partly controlled, n (%)	-	17 (30.4%)	-
Asthma control - uncontrolled, n (%)	-	1 (1.8%)	-

Data are presented as mean ± standard deviation or median (minimum-maximum), as appropriate

Abbreviations: BMI body mass index, IgE immunoglobulin E, FEV₁ forced expiratory volume in 1 s, FVC forced vital capacity, PEF peak expiratory flow, FEF 25–75 forced expiratory flow between 25% and 75% of forced vital capacity, WBC white blood cell count, SII systemic immune-inflammation index

SCP, DCP, and CC VD values were significantly lower in all quadrants in the asthma group compared with the control group (all FDR-corrected p < 0.05). CMT was also significantly thinner in the asthma group (all FDR-corrected p < 0.05). Quantitative OCTA findings are summarized in Table 2. Representative OCTA images obtained from a healthy control and an asthmatic child are shown in Figs. 1 and 2.

Table 2.

Comparison of superficial and deep macular and choriocapillaris vascular plexus densities between groups

Variable	Control Group (n = 30) Mean ± SD	Asthma Group (n = 56) Mean ± SD	p value	Adjusted p (FDR)*
SCP VD (%)
Foveal	22.4 ± 0.8	20.6 ± 1.9	0.001	0.002
Nasal	44.6 ± 0.6	42.0 ± 2.1	< 0.001	<0.001
Temporal	45.4 ± 0.6	43.7 ± 1.6	< 0.001	<0.001
Inferior	43.7 ± 0.8	42.7 ± 1.5	< 0.001	<0.001
Superior	44.7 ± 0.6	43.5 ± 1.3	< 0.001	<0.001
DCP VD (%)
Foveal	25.6 ± 0.6	24.5 ± 2.8	0.007	0.010
Nasal	44.5 ± 1.1	42.6 ± 2.0	0.002	0.003
Temporal	44.4 ± 0.5	44.1 ± 1.3	0.016	0.021
Inferior	44.6 ± 0.4	43.2 ± 2.1	< 0.001	<0.001
Superior	45.3 ± 0.6	43.8 ± 2.2	< 0.001	<0.001
CC VD (%)
Foveal	54.9 ± 1.7	53.4 ± 2.0	0.046	0.049
Nasal	56.1 ± 0.6	53.4 ± 1.0	< 0.001	< 0.001
Temporal	56.0 ± 0.8	53.5 ± 1.0	< 0.001	< 0.001
Inferior	56.0 ± 0.9	54.0 ± 1.3	< 0.001	< 0.001
Superior	56.0 ± 0.5	55.0 ± 1.6	< 0.001	< 0.001
CMT	255.0 ± 1.5	249.0 ± 5.5	< 0.001	< 0.001

Mean ± standard deviation values are presented. P values were adjusted for multiple comparisons using the Benjamini-Hochberg false discovery rate (FDR) method. Statistically significant results were defined as adjusted p < 0.05 and are indicated in bold

Abbreviations: VD vessel density, SCP superficial capillary plexus, DCP deep capillary plexus, CC choriocapillaris, F foveal, N nasal, T temporal, I inferior, S superior, FDR false discovery rate

Fig. 1.

Optical coherence tomography angiography (OCTA) images of a healthy control subject. (a) Superficial capillary plexus (SCP), (b) Deep capillary plexus (DCP), (c) Choriocapillaris (CC)

Fig. 2.

Optical coherence tomography angiography (OCTA) images of a pediatric patient with asthma showing (a) the superficial capillary plexus (SCP), (b) the deep capillary plexus (DCP), and (c) the choriocapillaris (CC)

In univariate logistic regression analysis, all SCP, DCP, CC, and CMT parameters were statistically significant in distinguishing between asthma and control groups (all FDR-corrected p < 0.05). When included in a multivariate logistic regression model, only nasal CC VD remained independently significant (p < 0.001). The results of the multivariate analysis are presented in Table 3.

Table 3.

Logistic regression analysis for discrimination between control and asthma groups

Parameter	Univariate Model OR (95% CI)	p value	Adjusted p (FDR)*	Multivariate Model OR (95% CI)	p value
SCP F	1.967 (1.325–2.920)	0.001	0.002	-	-
SCP N	2.710 (1.718–4.273)	< 0.001	< 0.001	-	-
SCP T	2.556 (1.581–4.132)	< 0.001	< 0.001	-	-
SCP I	4.381 (2.164–8.871)	< 0.001	< 0.001	-	-
SCP S	2.180 (1.307–3.635)	0.003	0.005	-	-
DCP F	1.648 (1.144–2.375)	0.007	0.010	-	-
DCP N	1.569 (1.150–2.140)	0.004	0.006	-	-
DCP T	1.866 (1.053–3.307)	0.032	0.043	-	-
DCP I	10.20 (3.366–30.91)	< 0.001	< 0.001	-	-
DCP S	5.875 (2.494–13.84)	< 0.001	< 0.001	-	-
CC F	1.308 (1.016–1.684)	0.037	0.047	-	-
CC N	26.01 (5.342–126.6)	< 0.001	< 0.001	26.01 (5.342–126.6)	< 0.001
CC T	10.91 (4.089–29.11)	< 0.001	< 0.001	-	-
CC I	> 100 (3.069->100)	0.029	0.042	-	-
CC S	5.133 (2.166–12.163)	< 0.001	< 0.001	-	-
CMT	1.415 (1.193–1.679)	< 0.001	< 0.001	-	-

Statistically significant values after Benjamini-Hochberg (FDR) correction (adjusted p < 0.05) are shown in bold. A forward stepwise logistic regression model was used for the multivariate analysis. Adjusted p-values for univariate models were calculated using the Benjamini-Hochberg (FDR) procedure to account for multiple comparisons

OR odds ratio, CI confidence interval, SCP superficial capillary plexus, DCP deep capillary plexus, CC choriocapillaris, F foveal, N nasal, T temporal, I inferior, S superior, CMT central macular thickness, FDR false discovery rate

In exploratory ROC curve analysis, nasal CC VD as a continuous variable yielded an AUC of 0.966 (95% CI: 0.935–0.997; p < 0.001). When an optimal cut-off value of 54.7 was applied, the AUC was 0.930 (95% CI: 0.869–0.991; p < 0.001), providing a sensitivity of 89.3% and a specificity of 96.7%. The ROC curve is shown in Fig. 3, and the diagnostic performance parameters are summarized in Table 4.

Fig. 3.

Receiver operating characteristic (ROC) curve showing the diagnostic performance of nasal choriocapillaris (CC) vessel density in distinguishing children with asthma from healthy controls (AUC = 0.966; 95% CI: 0.935–0.997)

Table 4.

Diagnostic performance of nasal choriocapillaris vessel density

Parameter		AUC	95% CI	p value
Nasal CC VD (continuous variable)		0.966	0.935–0.997	< 0.001
Nasal CC VD (cut-off: 54.7)		0.930	0.869–0.991	< 0.001
Diagnostic Index (at 54.7 Cut-off)	Value (%)
Sensitivity	89.3
Specificity	96.7
Positive Predictive Value	98.0
Negative Predictive Value	82.9

The AUC value for the continuous variable (0.966) represents the overall exploratory discriminative potential of nasal CC VD across all possible thresholds. The AUC value for the cut-off (0.930) represents the practical diagnostic performance specifically at the 54.7 threshold, which yielded the optimal balance of sensitivity and specificity

Abbreviations: AUC area under the curve, CC choriocapillaris, CI confidence interval, NPV negative predictive value, PPV positive predictive value, VD vessel density

When patients were compared according to ICS dose, a statistically significant difference was initially observed in the superior quadrant of the DCP (p = 0.010); however, this difference was no longer statistically significant after FDR correction. No significant differences were observed in OCTA parameters according to ICS dose (all FDR-corrected p > 0.05). Subgroup analysis results are presented in Table 5.

Table 5.

Comparison of OCTA parameters between low-dose and moderate-to-high–dose inhaled corticosteroid users

Parameter	Low-dose ICS (n = 25) Mean ± SD / Median	Moderate-to-high-dose ICS (n = 31) Mean ± SD / Median	p value	Adjusted p (FDR)*
SCP F	20.7 ± 2.1 / 20.6	20.7 ± 1.7 / 20.6	0.987	0.993
SCP N	41.4 ± 1.9 / 41.1	42.0 ± 2.2 / 42.1	0.338	0.538
SCP T	43.4 ± 1.6 / 43.5	44.0 ± 1.5 / 43.7	0.156	0.416
SCP I	42.2 ± 1.4 / 42.4	42.6 ± 1.5 / 42.7	0.302	0.538
SCP S	43.3 ± 1.3 / 43.4	43.5 ± 1.2 / 43.5	0.454	0.650
DCP F	22.8 ± 3.0 / 24.1	23.8 ± 2.6 / 24.5	0.070	0.320
DCP N	43.3 ± 1.8 / 43.5	42.8 ± 2.1 / 42.5	0.405	0.650
DCP T	43.8 ± 1.4 / 44.1	44.0 ± 1.1 / 44.1	0.741	0.846
DCP I	42.4 ± 2.2 / 43.1	43.0 ± 2.0 / 43.4	0.229	0.488
DCP S	42.4 ± 2.4 / 42.6	43.9 ± 1.9 / 44.3	0.010	0.160
CC F	52.7 ± 2.2 / 52.6	53.4 ± 1.8 / 53.9	0.238	0.488
CC N	53.5 ± 1.1 / 53.3	53.4 ± 0.9 / 53.4	0.538	0.717
CC T	53.6 ± 1.1 / 53.5	53.5 ± 0.8 / 53.5	0.796	0.849
CC I	53.9 ± 1.2 / 54.0	53.9 ± 1.4 / 54.0	0.993	0.993
CC S	54.4 ± 1.7 / 55.0	54.5 ± 1.5 / 55.0	0.813	0.849
CMT	249.8 ± 5.3 / 249.0	248.1 ± 5.7 / 248.0	0.246	0.488

Mean ± standard deviation / Median values are shown

p-values were adjusted using the Benjamini-Hochberg (FDR) procedure. No parameters remained statistically significant after FDR adjustment (adjusted p > 0.05)

Abbreviations: SCP superficial capillary plexus, DCP deep capillary plexus, CC choriocapillaris, F foveal, N nasal, T temporal, I inferior, S superior, CMT central macular thickness, ICS inhaled corticosteroid, FDR false discovery rate

Additionally, no statistically significant differences were observed between monotherapy and combination therapy groups within either ICS dose subgroup after FDR correction (all FDR-corrected p > 0.05) (Tables 6 and 7).

Table 6.

Comparison of OCTA parameters between monotherapy and combination therapy groups in low-dose ICS users

Parameter	Monotherapy (n = 11) Mean ± SD	Combination Therapy (n = 14) Mean ± SD	p value	Adjusted p (FDR)*
SCP F	21.03 ± 1.57	20.38 ± 2.51	0.438	0.888
SCP N	42.09 ± 2.01	40.90 ± 1.70	0.122	0.488
SCP T	43.67 ± 1.66	43.27 ± 1.62	0.548	0.888
SCP I	42.23 ± 1.66	42.14 ± 1.30	0.888	0.888
SCP S	43.49 ± 0.94	43.10 ± 1.56	0.447	0.888
DCP F	23.42 ± 3.18	22.26 ± 2.79	0.342	0.888
DCP N	43.86 ± 1.47	42.83 ± 1.98	0.132	0.488
DCP T	44.28 ± 0.94	43.42 ± 1.67	0.539	0.888
DCP I	43.05 ± 2.10	41.82 ± 2.14	0.185	0.592
DCP S	43.23 ± 2.27	41.71 ± 2.37	0.120	0.488
CC F	53.18 ± 2.16	52.27 ± 2.30	0.330	0.888
CC N	53.75 ± 1.05	53.34 ± 1.13	0.363	0.888
CC T	53.76 ± 1.00	53.44 ± 1.17	0.468	0.888
CC I	53.88 ± 1.35	53.86 ± 1.10	0.960	0.960
CC S	54.43 ± 1.29	54.36 ± 1.96	0.928	0.960
CMT (µm)	251.09 ± 4.61	248.86 ± 5.68	0.301	0.888

Table 7.

Comparison of OCTA parameters between monotherapy and combination therapy groups in moderate-to-high-dose ICS users

Parameter	Monotherapy (n = 4) Mean ± SD	Combination Therapy (n = 27) Mean ± SD	p value	Adjusted p (FDR)*
SCP F	21.33 ± 2.44	20.57 ± 1.58	0.411	0.904
SCP N	41.18 ± 2.01	42.08 ± 2.25	0.454	0.904
SCP T	44.20 ± 0.94	44.02 ± 1.57	0.828	0.915
SCP I	42.65 ± 0.90	42.55 ± 1.56	0.904	0.915
SCP S	43.33 ± 1.43	43.56 ± 1.22	0.728	0.915
DCP F	23.33 ± 3.01	23.93 ± 2.60	0.674	0.915
DCP N	41.28 ± 2.14	43.06 ± 2.08	0.291	0.904
DCP T	43.68 ± 0.31	44.04 ± 1.18	0.349	0.904
DCP I	43.45 ± 0.68	42.92 ± 2.10	0.884	0.915
DCP S	43.95 ± 1.62	43.84 ± 1.92	0.915	0.915
CC F	54.15 ± 1.49	53.29 ± 1.81	0.601	0.915
CC N	53.58 ± 0.75	53.32 ± 0.92	0.601	0.915
CC T	53.88 ± 0.89	53.46 ± 0.84	0.365	0.904
CC I	53.78 ± 1.86	53.89 ± 1.34	0.884	0.915
CC S	55.55 ± 0.90	54.34 ± 1.55	0.141	0.488
CMT (µm)	243.00 ± 3.56	248.89 ± 5.61	0.033	0.488

Values are presented as mean ± standard deviation. P values were adjusted using the Benjamini–Hochberg false discovery rate (FDR) procedure. No parameter remained statistically significant after FDR adjustment (adjusted p > 0.05)

Abbreviations: SCP superficial capillary plexus, DCP deep capillary plexus, CC choriocapillaris, CMT central macular thickness (µm), ICS inhaled corticosteroid, FDR false discovery rate

Discussion

Asthma is not limited to the respiratory system and, due to its systemic inflammatory and intermittent hypoxic nature, may also affect ocular tissues [9, 10]. In this study, vessel density values in SCP, DCP, and CC were found to be significantly lower in all quadrants in asthmatic children compared to healthy controls. In addition, CMT was significantly reduced in the asthma group. In multivariate analysis, only nasal CC vessel density remained independently significant. No significant differences were observed in OCTA parameters according to ICS dose or between monotherapy and combination therapy.

Hypoxia has been reported to influence retinal microcirculation and to induce changes in capillary-level perfusion [14]. Supporting this, OCTA studies in children with obstructive sleep apnea (OSA), in which similar hypoxic processes are observed, have demonstrated that vessel density and perfusion parameters in the macular region were significantly lower compared to the control group [15]. Following adenotonsillectomy, these parameters increased and approached control levels, indicating that hypoxia-related microcirculatory changes may be at least partially reversible. Similarly, a significant increase in vessel density and perfusion indices in the parafoveal region has been reported after adenoidectomy [16].

In a study investigating the effects of maternal asthma on retinal microvascular structure in children, global differences in retinal microvascular parameters were identified in 4-6-year-old children born to asthmatic mothers compared to controls, and increased retinal venular tortuosity was demonstrated [17]. In addition, the increase in placental growth factor levels during pregnancy was found to be lower in the asthma group and was associated with microvascular parameters such as arteriolar branching and venular tortuosity. These findings suggest that exposure to maternal asthma may lead to persistent microvascular changes in childhood through intrauterine angiogenic mechanisms.

Studies evaluating OCTA parameters in healthy pediatric subjects have demonstrated that macular microvascular measurements are influenced by various demographic and anatomical factors [18, 19]. In the study by Borrelli et al., it was reported that the foveal avascular zone area increased with age and was associated with thinner foveal macular thickness; additionally, male sex was associated with increased perfusion density, whereas CC perfusion density was higher at younger ages [18]. Turan et al. [19] reported that OCT parameters exhibit a wide physiological variability and are associated with factors such as age, axial length, and refractive status, with high interobserver agreement in choroidal thickness measurements. These findings indicate that retinal and choroidal parameters in pediatric populations are influenced by individual variability, which should be taken into account when interpreting OCTA measurements.

Retinal microvascular changes have also been reported in adult diseases associated with chronic hypoxia. In a study conducted in adults with obstructive sleep apnea syndrome, increased Apnea-Hypopnea Index values were found to be associated with reduced vessel density, with significant decreases observed in the parafoveal and perifoveal regions compared to controls [20]. Similarly, in chronic obstructive pulmonary disease, it has been shown that as disease severity increases, vessel density decreases, and these changes are associated with pulmonary function parameters [21]. In addition, in a study evaluating retinal and posterior segment vascular changes related to systemic hypoxia using OCTA, no significant differences were observed in retinal capillary plexuses, whereas significant changes were reported at the level of the optic nerve head and especially the CC, and this layer was suggested to be more sensitive to systemic hypoxia [22]. These findings indicate that chronic hypoxia and systemic inflammation may affect retinal microvascular perfusion and that microcirculatory changes may be associated with pulmonary function.

The structural effects of asthma on the posterior segment have also been demonstrated using different imaging modalities. In a study conducted in patients with allergic asthma, CMT and choroidal vascularity index were found to be significantly lower compared to the control group, suggesting that asthma may lead to structural changes in the choroidal vasculature [23]. Similarly, in a swept-source OCT analysis conducted in pediatric asthma patients receiving ICS treatment, superior, inferior, and mean retinal nerve fiber layer thickness values were found to be significantly lower compared to healthy controls, whereas no significant difference was observed in central retinal thickness and central choroidal thickness [24]. In addition, the presence of a negative correlation between corticosteroid dose and retinal nerve fiber layer thickness and an association between IgE levels and ganglion cell layer thickness suggests that structural retinal changes may be related to both treatment and inflammatory burden. When these findings are evaluated together, it is evident that asthma is a systemic disease that may affect not only retinal microvascular structures but also choroidal and neurosensory retinal tissues.

In OCTA-based studies, Karaaslan et al. reported no significant differences in macular SCP, DCP, and foveal avascular zone (FAZ) parameters, while demonstrating decreased retinal nerve fiber layer thickness in the temporal quadrant, reduced outer retinal flow area, and increased intradisc radial peripapillary capillary density [25]. In contrast, Kıvrak et al., in a study comparing pediatric asthma patients receiving and not receiving ICS treatment with healthy controls, demonstrated that in asthma groups, central choroidal thickness was reduced and vessel density was significantly decreased in the superior and nasal quadrants of the SCP, in all quadrants of the DCP, and at the level of the foveal CC; additionally, differences in inferior and nasal peripapillary vascular density were reported [26]. Moreover, correlations between eosinophil and CRP levels and certain vascular parameters suggested that microvascular changes may be related to inflammatory processes. In both studies, no significant differences were found between patients receiving and not receiving ICS.

Consistent with these findings, our study also demonstrated widespread reduction in vessel density across all quadrants in SCP, DCP, and CC, along with decreased CMT in asthmatic children. In particular, the finding that only nasal CC vessel density remained independently significant in multivariate analysis may suggest that microvascular involvement could be more pronounced at this level. Similarly, the absence of significant differences according to ICS dose may indicate a possible association of these retinal changes with the underlying disease process; however, this finding should be interpreted with caution, as the study may be underpowered to detect treatment-related differences.

Several limitations should be acknowledged. This study was conducted at a single center with a retrospective design, and the sample size was relatively limited. The cross-sectional nature of the study precludes causal inference; therefore, the observed differences in vessel density cannot be directly attributed to asthma itself, ICS use, or any single factor. OCTA imaging was performed using a 3 × 3 mm macular scan area, and thus wider retinal regions were not evaluated. FAZ parameters were not analyzed. Although OCTA images were automatically generated by the device, quantitative measurements were manually recorded from the displayed output, as the system did not provide exportable numerical data, which may have introduced measurement variability. In addition, OCTA-related artifacts and segmentation errors cannot be completely excluded. OCTA-derived vessel density measurements are device-specific and may not be directly comparable across different imaging platforms or normative datasets.

Although clinical history was used to screen for atopy in the control group, subclinical allergic sensitization could not be excluded through laboratory testing. Furthermore, while all patients achieved full asthma control during the six-month follow-up period prior to OCTA imaging, the potential cumulative effects of earlier periods of uncontrolled disease could not be isolated. Although participants with refractive errors greater than ± 3.00 diopters were excluded and intraocular pressure measurements were within normal limits, axial length and spherical equivalent measurements were not routinely available and therefore could not be analyzed. Because retinal and choroidal measurements may be influenced by these ocular biometric parameters, the absence of direct axial length and spherical equivalent data should be considered a limitation. The absence of an a priori sample size calculation should also be considered a limitation, particularly for the ICS dose subgroup analysis, which may have been underpowered to detect modest between-group differences.

Finally, as the ROC analysis was performed on the same dataset used for model development, the reported AUC values represent in-sample estimates and may overestimate the true discriminative performance. Larger, prospective studies are required to confirm these findings and to better define the temporal course of the observed microvascular changes.

Conclusion

In conclusion, this study demonstrated that macular microvascular structures at the superficial, deep, and CC levels were altered in pediatric patients with asthma compared with healthy controls, along with reduced CMT. The identification of nasal CC vessel density as an independently associated parameter in multivariate analyses may suggest that regional microvascular alterations could reflect the systemic impact of asthma. No consistent differences were observed according to ICS dose; however, a potential treatment effect cannot be excluded. Although a high AUC value was observed, this finding is based on in-sample estimates and cannot be directly translated into clinical practice. Further large-scale prospective studies with external validation are required to confirm these findings and clarify the long-term clinical implications of these microvascular alterations.

Authors’ contributions

All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by G.K. and Z.K. The first draft of the manuscript was written by G.K., and both authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

There is no funding.

Data availability

The data that support the study findings are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This retrospective study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Bursa City Hospital Scientific Research Ethics Committee (approval no.: 2025/12 − 8). The requirement for informed consent was waived by the Ethics Committee due to the retrospective design of the study. For participants under the age of 18, parental consent was not required as the study involved retrospective analysis of anonymized clinical data.

Consent for publication

Not Applicable.

Competing interests

The authors declare no competing interests.