Comparative peripheral inflammatory biomarker profiles in central and branch retinal artery occlusion: a retrospective study

Abstract

Background

Retinal artery occlusion (RAO), including central retinal artery occlusion (CRAO) and branch retinal artery occlusion (BRAO), is an ophthalmic emergency and an important marker of systemic vascular disease. Inflammation plays a key role in RAO pathogenesis; however, subtype-specific inflammatory profiles remain insufficiently characterized.

Method

We conducted a retrospective observational study of 363 patients diagnosed with RAO at Renmin Hospital of Wuhan University between January 2020 and April 2024 (CRAO, n = 320; BRAO, n = 43). Peripheral inflammatory biomarkers, including leukocyte count, neutrophil percentage, lymphocyte percentage, monocyte percentage, eosinophil percentage, C-reactive protein (CRP, mg/L), and neutrophil-to-lymphocyte ratio (NLR), were compared between CRAO and BRAO using independent-sample t tests or Mann–Whitney U tests, as appropriate. Sex- and age-stratified analyses were performed to explore subgroup patterns.

Results

A modest shift was observed between BRAO, CRAO patients in term leukocyte counts (6.57 ± 2.19 vs. 5.95 ± 1.66 × 10⁹/L; p = 0.03) and lymphocyte (30.75 ± 8.62% vs. 27.32 ± 7.52%; p = 0.01), but lower neutrophil percentages (58.33 ± 9.08% vs. 61.92 ± 8.07%; p = 0.01). These between-group differences were small, and group-level means largely remained within conventional reference ranges. Monocyte and eosinophil percentages did not differ significantly between groups. CRP levels (mg/L) showed no statistically significant difference, although greater variability was observed in the CRAO group. In patients with available absolute counts, BRAO demonstrated a higher NLR than CRAO. Sex- and age-stratified analyses revealed consistent subtype-related differences, with more pronounced neutrophil–lymphocyte shifts in males and steeper age-related changes in CRAO.

Conclusions

CRAO and BRAO exhibit distinct systemic inflammatory profiles. CRAO is characterized by higher leukocyte and lymphocyte levels, suggesting broader systemic inflammatory involvement, whereas BRAO demonstrates relative neutrophil predominance and higher NLR, consistent with a more localized inflammatory response. These findings support the concept that CRAO and BRAO are immunologically distinct entities and may benefit from subtype-specific evaluation and management strategies. Prospective, multi-center studies are warranted to validate these observations and clarify their clinical implications.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12886-026-04715-x.

Introduction

Retinal artery occlusion (RAO), characterized by a sudden interruption of retinal blood flow, is a vision-threatening ophthalmic emergency with significant systemic implications. RAO is clinically categorized into central (CRAO) and branch (BRAO) subtypes, with each subtype leading to varying degrees of visual morbidity. CRAO typically causes severe, irreversible vision loss due to occlusion of the central retinal artery, while BRAO results in partial visual deficits from blockage of smaller retinal branches [1, 2]. Both subtypes are strongly associated with systemic vascular risk factors such as hypertension, diabetes, and atherosclerosis, highlighting their role as indicators of underlying cardiovascular disease [3, 4]. Despite these shared risk factors, emerging evidence suggests that CRAO and BRAO may have divergent pathophysiological mechanisms, particularly in terms of their inflammatory responses, which remain inadequately understood [5, 6]. Systemic inflammation is increasingly recognized as a key driver of ischemic vascular events, including RAO [7]. Elevated leukocyte counts, neutrophil-to-lymphocyte ratios (NLR), and C-reactive protein (CRP) levels have been linked to poorer outcomes in both cerebral and retinal ischemia, reflecting endothelial dysfunction and thrombotic cascades [8, 9]. However, most prior studies on RAO have aggregated cases without distinguishing between CRAO and BRAO subtypes, which limits understanding of potential differences in inflammatory profiles [10–12]. For example, while leukocytosis has been associated with poor visual recovery in mixed RAO cohorts, its specific role in CRAO versus BRAO has not been explored [13]. Similarly, NLR, a well-established prognostic marker in ischemic stroke, has not been rigorously evaluated in RAO subtypes [14]. This knowledge gap hampers the development of targeted therapies, as anti-inflammatory strategies may need to be optimized for each RAO subtype. The primary objective of this study was to investigate and compare the inflammatory biomarker profiles in CRAO and BRAO, aiming to better understand their distinct pathophysiological mechanisms. Through the analysis of inflammatory responses in a cohort of RAO patients, we sought to identify specific biomarkers that could serve as key indicators for diagnosis, prognosis, and targeted treatment. Specifically, we aimed to determine whether CRAO and BRAO exhibit different inflammatory profiles, particularly regarding leukocytes, neutrophil-to-lymphocyte ratios, and C-reactive protein levels. The findings from this study are intended to inform the development of personalized therapeutic strategies for RAO, grounded in subtype-specific inflammatory biomarker profiles.

Methods

Study design, setting, and duration

A retrospective observational study was conducted among patients diagnosed with retinal artery occlusion (RAO) between January 2020 and April 2024 at Renmin Hospital of Wuhan University. Demographic, clinical, and laboratory data were extracted from the hospital’s electronic medical record system. Cases were identified using the International Classification of Diseases, 10th Revision (ICD-10), including central retinal artery occlusion (CRAO; ICD-10 code H34.1) and branch retinal artery occlusion (BRAO; ICD-10 code H34.8).

Study participants

Patients were eligible for inclusion if they: (1) were aged 18 years or older; (2) had a confirmed diagnosis of CRAO or BRAO; (3) had available inflammatory biomarker measurements at presentation; and (4) had complete demographic and clinical information. Exclusion criteria were: (1) missing laboratory data for key inflammatory biomarkers; (2) evidence of active infectious disease at presentation; (3) Patient with active maliganancy; or (4) severe hepatic or renal dysfunction. Details of patient screening and selection are presented in Fig. 1.

Fig. 1.

Flowchart of patient selection and study cohort formation

Patients with confirmed retinal artery occlusion were identified from the electronic medical record system. After exclusion of patients with missing inflammatory biomarker data or incomplete diagnostic information, 363 patients were included in the final analytic cohort, comprising 320 patients with central retinal artery occlusion (CRAO) and 43 patients with branch retinal artery occlusion (BRAO).

Study variables

The primary exposure variable was RAO subtype (CRAO vs. BRAO). Secondary clinical variables included age, sex, history of systemic diseases (e.g., hypertension, diabetes), medication use (e.g., anticoagulants), smoking status, alcohol consumption, and comorbid conditions.

Inflammatory biomarkers were analyzed as continuous variables and included leukocyte count, neutrophil percentage, lymphocyte percentage, monocyte percentage, eosinophil percentage, C-reactive protein (CRP, mg/L), and the neutrophil-to-lymphocyte ratio (NLR). Categorical variables included sex (male/female), diagnosis type (CRAO/BRAO), smoking status, and alcohol use.

Data collection

All clinical and laboratory data were collected retrospectively from the hospital information system by two independent investigators. Laboratory values were obtained from routine blood tests performed at the time of each patient’s initial presentation for RAO.

According to standard laboratory protocols during the study period, venous blood samples for hematologic analysis were processed within approximately 1 h of collection using an automated hematology analyzer (Sysmex XN-3000). Absolute leukocyte and differential counts were used for the calculation of inflammatory indices, including the neutrophil-to-lymphocyte ratio (NLR), when available.

For CRP measurement, serum samples were centrifuged immediately after collection and either analyzed promptly or stored at 2–8 °C for up to 24 h, in accordance with routine laboratory practice, before analysis using high-sensitivity immunoassays. CRP concentrations were recorded in milligrams per liter (mg/L). As all laboratory testing followed standardized pre-analytical and analytical procedures at the time of care, biomarker values were considered comparable across patients despite the retrospective study design. Any discrepancies in data extraction were resolved by consensus or adjudication by a senior investigator.

Study ethics

This study was approved by the Ethics Committee of Renmin Hospital of Wuhan University (approval number: 2020-X-58) and was conducted in accordance with the Declaration of Helsinki. Given the retrospective observational design, the requirement for informed consent was waived by the Ethics Committee. Patient confidentiality was maintained through anonymization and secure handling of de-identified data.

Statistical analysis

Continuous variables are presented as mean ± standard deviation (SD) or median (interquartile range), as appropriate, and categorical variables are presented as frequencies and percentages. Normality was assessed using the Shapiro–Wilk test. Between-group comparisons of inflammatory biomarkers between CRAO and BRAO were performed using independent-sample t tests or Mann–Whitney U tests, depending on data distribution. Categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate. Because the CRAO and BRAO groups were imbalanced in size, we complemented unadjusted comparisons with (i) covariate-adjusted regression models and (ii) propensity score–based inverse probability of treatment weighting (IPTW) as a sensitivity analysis. All statistical analyses were conducted using R software (version 4.3.2), with a two-sided p value < 0.05 considered statistically significant.

Result

Baseline clinical and demographic characteristics

A total of 363 patients were included in the final analytic cohort, comprising 320 patients with CRAO and 43 patients with BRAO. Baseline clinical and demographic characteristics are summarized in Table 1. There were no statistically significant differences between the CRAO and BRAO groups with respect to age (mean ± SD: 57.72 ± 12.39 vs. 55.33 ± 12.75 years; p = 0.23) or sex distribution (p = 0.43). Similarly, no significant between-group differences were observed in the use of systemic medications, prevalence of malignant tumors, history of systemic disease, drinking history, or history of ocular disease (all p > 0.05).

Table 1.

Distribution of clinico-demographic characteristics across CRAO and BRAO

Characteristic	CRAO (n = 320)	BRAO (n = 43)	P-value
Age, years (mean ± SD)	57.72 ± 12.39	55.33 ± 12.75	0.23
Sex, n (%)			0.43
Male	220 (68.8)	27 (62.8)
Female	100 (31.3)	16 (37.2)
Systemic medication use, n (%)			0.69
Anticoagulants	11 (3.5)	1 (2.3)
None	306 (96.5)	42 (97.7)
Malignant tumors, n (%)			0.69
Lung cancer	11 (3.5)	1 (2.3)
No	309 (96.5)	42 (97.7)
History of systemic disease, n (%)			0.95
Yes	251 (78.7)	34 (79.1)
No	69 (21.3)	9 (20.9)
Other diseases, n (%)			0.29
Yes	7 (2.8)	0 (0.0)
No	313 (97.8)	43 (100.0)
Smoking history, n (%)			0.08
Current smoker	98 (31.0)	7 (16.3)
Former smoker	5 (1.6)	0 (0.0)
Never smoker	217 (67.8)	36 (83.7)
Alcohol consumption, n (%)			0.24
Current drinker	64 (20.3)	5 (11.6)
Former drinker	6 (1.9)	0 (0.0)
Non-drinker	250 (78.1)	38 (88.4)
History of eye disease, n (%)			0.15
Yes	47 (14.7)	10 (23.3)
No	273 (85.3)	33 (76.7)

A higher prevalence of current smoking was observed among patients with CRAO compared with those with BRAO (31.0% vs. 16.3%); however, this difference did not reach statistical significance (p = 0.08).

Comparison of inflammatory biomarkers between CRAO and BRAO

Between-group comparisons of inflammatory biomarkers demonstrated distinct inflammatory profiles between CRAO and BRAO (Table 2; Fig. 2). Given the marked imbalance between CRAO and BRAO (320 vs. 43), we evaluated subtype–biomarker associations using both covariate-adjusted regression and propensity score IPTW (Supplementary Table S1). Patients with CRAO exhibited significantly higher leukocyte counts compared with those with BRAO (6.57 ± 2.19 vs. 5.95 ± 1.66 × 10⁹/L; mean difference 0.62, 95% CI 0.03–1.21; p = 0.03). Lymphocyte percentages were also higher in the CRAO group (30.75 ± 8.62% vs. 27.32 ± 7.52%; mean difference 3.42, 95% CI 0.82–6.02; p = 0.01). The values were within commonly used laboratory reference ranges. Accordingly, these findings should be interpreted as phenotype-associated shifts in peripheral leukocyte distribution rather than evidence of clinically overt leukocytosis or lymphocytosis. In contrast, neutrophil percentages were significantly lower in CRAO compared with BRAO (58.33 ± 9.08% vs. 61.92 ± 8.07%; mean difference − 3.59, 95% CI − 6.37 to − 0.81; p = 0.01).

Table 2.

Comparison of inflammatory biomarkers across CRAO and BRAO

Biomarker	CRAO (n = 320) Mean ± SD	BRAO (n = 43) Mean ± SD	Mean difference (SE)	95% CI (Lower–Upper)	P-value
Leukocyte count (×10⁹/L)	6.57 ± 2.19	5.95 ± 1.66	0.62 (0.29)	0.03 to 1.21	0.03
Neutrophils (%)	58.33 ± 9.08	61.92 ± 8.07	−3.59 (1.38)	−6.37 to − 0.81	0.01
Lymphocytes (%)	30.75 ± 8.62	27.32 ± 7.52	3.42 (1.29)	0.82 to 6.02	0.01
Monocytes (%)	7.69 ± 2.50	7.65 ± 2.56	0.03 (0.43)	−0.82 to 0.90	0.92
Eosinophils (%)	2.57 ± 2.77	2.38 ± 1.55	0.19 (0.29)	−0.39 to 0.78	0.50
CRP (mg/L)	2.59 ± 6.89	1.59 ± 2.85	0.99 (0.67)	−0.33 to 2.32	0.14
Neutrophil-to-lymphocyte ratio (NLR)*	2.27 ± 2.52	2.59 ± 1.32	—	—	0.02

* NLR was calculated in patients with available absolute neutrophil and lymphocyte counts (CRAO n = 277; BRAO n = 39) and compared using the Mann–Whitney U test

Fig. 2.

Sex-stratified distributions of inflammatory biomarkers in CRAO and BRAO. Boxplots with overlaid individual data points show the distributions of leukocytes, neutrophils, lymphocytes, monocytes, eosinophils, and C-reactive protein (CRP, mg/L), stratified by sex and retinal artery occlusion subtype. CRAO and BRAO are displayed side-by-side to facilitate direct comparison

No statistically significant differences were observed between groups for monocyte or eosinophil percentages (both p > 0.05). C-reactive protein (CRP, mg/L) levels did not differ significantly between CRAO and BRAO, although greater variability was observed in the CRAO group (SD 6.89 vs. 2.85).

The neutrophil-to-lymphocyte ratio (NLR) was analyzed in patients with available absolute neutrophil and lymphocyte counts (CRAO n = 277; BRAO n = 39). BRAO patients demonstrated a significantly higher NLR than CRAO patients (mean ± SD: 2.59 ± 1.32 vs. 2.27 ± 2.52; median: 2.36 vs. 1.96), as determined by the Mann–Whitney U test (p = 0.02).

Stratified analyses by sex and age

Sex-stratified analyses demonstrated that differences in neutrophil and lymphocyte distributions between CRAO and BRAO were consistent across both sexes, with more pronounced contrasts observed among male patients (Fig. 2). Age-stratified analyses revealed steeper age-related increases in neutrophil percentages and sharper declines in lymphocyte percentages among patients with CRAO compared with those with BRAO (Fig. 3). No consistent age- or sex-related patterns were observed for CRP.

Fig. 3.

Age-stratified distributions of inflammatory biomarkers in CRAO and BRAO

Discussion

Our comparative analysis of CRAO and BRAO revealed distinct inflammatory profiles, particularly in leukocyte, lymphocyte, and neutrophil dynamics. CRAO patients exhibited significantly higher leukocyte and lymphocyte levels compared with BRAO, but paradoxically lower neutrophil percentages. These findings suggest that CRAO and BRAO may reflect divergent systemic and local inflammatory processes.

Notably, while our direction of association is consistent with prior reports describing altered leukocyte distributions in RAO and other ischemic vascular events [15, 16]. Leukocytosis has been associated with endothelial dysfunction, microvascular occlusion, and plaque instability [19], supporting the interpretation of CRAO as a systemic vascular warning sign rather than an isolated ocular event. Elevated lymphocyte percentages in CRAO may indicate an adaptive immune response to chronic vascular injury. Similar trends have been observed in peripheral blood indices, where lymphocyte levels correlated with CRAO severity [17, 20]. However, it is noteworthy to mention that the absolute differences observed were modest and the group-level means largely remained within commonly used laboratory reference ranges. This limits immediate clinical interpretability and suggests these markers should be viewed as phenotype-associated shifts in peripheral leukocyte composition rather than evidence of clinically overt leukocytosis/lymphocytosis or directly actionable thresholds. Accordingly, the present findings are best interpreted as hypothesis-generating signals that warrant confirmation in prospectively phenotyped cohorts with standardized sampling and more granular vascular/etiologic characterization.

In contrast, BRAO patients demonstrated relatively higher neutrophil proportions and NLR compared with CRAO. Neutrophils are well-recognized mediators of ischemia–reperfusion injury and tissue damage [18], and NLR has emerged as a sensitive marker of CRAO severity [19, 20] and a predictive biomarker in meta-analyses of RAO [25]. Our findings suggest that BRAO may be characterized by a more localized neutrophil-driven inflammatory response, whereas CRAO reflects broader systemic leukocyte activation. However, because this study did not include healthy controls, it remains uncertain whether neutrophil levels in BRAO are elevated relative to baseline populations. Comparisons with conditions such as CRVO [21, 22] should therefore be interpreted with caution.

CRP levels did not differ significantly between groups, but greater variability in CRAO suggests heterogeneity in the acute-phase response. This observation is consistent with evidence that CRAO involves systemic and heterogeneous inflammatory activation [23], CRP is a non-specific marker, and we could not reliably ascertain or exclude arteritic CRAO from structured retrospective records, which may contribute to dispersion and limits etiologic inference. Additional cytokines, such as interleukin-6, have also been implicated in retinal ischemic diseases [24], and should be prioritized in future biomarker studies.

These findings highlight the need for a subtype-specific approach to RAO management. CRAO, characterized by systemic leukocytosis and relative lymphocytosis, may warrant comprehensive cardiovascular risk assessment and consideration of systemic anti-inflammatory strategies [15–17, 26]. In contrast, BRAO, with higher neutrophil proportions and NLR, may reflect a localized neutrophilic response, suggesting potential benefit from therapies aimed at limiting microvascular injury [18, 25]. Emerging transcriptomic evidence further supports the involvement of immune and inflammatory pathways in CRAO [26], underscoring the potential for biomarker-guided, individualized treatment strategies.

This study has several limitations. First, its retrospective, single-center design precludes assessment of temporal biomarker dynamics and limits generalizability. Second, the relatively small BRAO sample (n = 43) reduced statistical power. To address this, we supplemented unadjusted comparisons with covariate-adjusted regression and propensity score IPTW. These findings suggest that the reported differences are not solely an artifact of unequal group sizes, although residual confounding cannot be excluded. Third, the absence of non-RAO controls prevents determination of whether leukocyte or lymphocyte elevations represent disease-specific changes or background variation. Fourth, advanced inflammatory markers such as interleukin-6, tumor necrosis factor-α, and lipid–immune interactions [24, 27] were not evaluated, restricting mechanistic insights. Fifth, we could not reliably identify or exclude arteritic CRAO due to giant cell arteritis (GCA) from structured retrospective records. This may contribute to the greater dispersion of CRP observed in the CRAO group and limits etiologic interpretation of CRP findings. Lastly, detailed vascular risk factors and etiologic characterization (e.g., hypertension, atrial fibrillation, carotid stenosis severity, hyperlipidemia/lipid profile, and results of cardiac/carotid investigations) were not consistently available as structured variables in this retrospective dataset and therefore could not be reported or adjusted for. This limitation introduces potential residual confounding and restricts mechanistic inference; our findings should be viewed as exploratory and hypothesis-generating. Future multi-center, prospective studies with larger sample sizes, longitudinal sampling, and broader immune profiling and disease status are needed to validate these findings and clarify their prognostic and therapeutic relevance.

Conclusion

This study identified distinct inflammatory signatures in RAO subtypes. CRAO was associated with elevated leukocyte and lymphocyte levels, consistent with systemic vascular inflammation, whereas BRAO showed relatively higher neutrophil proportions and NLR, suggestive of a localized neutrophilic response. These findings indicate that CRAO and BRAO are not immunologically equivalent and may benefit from subtype-specific evaluation and management strategies. Larger, prospective, multi-center studies with advanced immune profiling are needed to validate these observations and clarify their prognostic and therapeutic relevance.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

RAO

Retinal artery occlusion

CRAO

Central retinal artery occlusion

BRAO

Branch retinal artery occlusion

WBC

White blood cell (leukocyte) count

NEU

Neutrophils

LYM

Lymphocytes

MON

Monocytes

EOS

Eosinophils

CRP

C-reactive protein

ICD-10

International classification of diseases, 10th revision

SD

Standard deviation

NLR

Neutrophil-to-lymphocyte ratio

PCI

Percutaneous coronary intervention

RVO

Retinal vein occlusion

CRVO

Central retinal vein occlusion

Author contributions

Al Asadi Hussein Mohammed Mahdi: Conceptualization, Methodology, Investigation, Data curation, Writing – Original Draft. Xuan Xiao: Supervision, Conceptualization, Writing – Review & Editing. All authors read and approved the final manuscript.

Funding

This research received no external funding.

Data availability

The datasets generated and/or analyzed during the current study are not publicly available due to institutional data protection policies but are available from the corresponding author on reasonable request and with permission of Renmin Hospital of Wuhan University.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Renmin Hospital of Wuhan University (approval number: 2020-X-58) and was conducted in accordance with the Declaration of Helsinki. Given the retrospective observational design, the requirement for informed consent was waived by the Ethics Committee.

Consent for publication

Not applicable. The manuscript does not contain any individual person’s identifiable data in any form.

Competing interests

The authors declare no competing interests.