Analysis of differences in age, gender, visual acuity, and intraocular pressure between the two groups

A total of 44 participants (87 eyes) were included: 12 males (27.3%) and 32 females (72.7%). The average age was 48.77 ± 15.59 years. The study comprised 22 participants (44 eyes) in the healthy control (HC) Group, 14 in the antibody-positive group (27 eyes), and 30 in the antibody-negative group (60 eyes). There was a statistically significant difference in gender among the three groups (P = 0.049), with a higher proportion of females in the antibody-positive group compared to the other two groups. No statistically significant differences were observed in age among the three groups (P > 0.05). Table 1 provides basic information for the three groups of study subjects.

Table 1 Comparison of data between neuromyelitis Optica spectrum disorder (NMOSD) AQP4-Ab+, AQP4-Ab- groups, and normal control groupComparison of macular thickness, ganglion cell thickness, and retinal nerve fiber layer thickness in different regions among the three groups

Results indicate significant differences in 18 variables among the three groups, including ganglion cell thickness in various macular regions, inner and outer macular thickness, and RNFL thickness (all P < 0.05). All observed indicators differed significantly between the AQP4-Ab + and healthy control groups (P < 0.05). Significant differences were found in all variables between the AQP4-Ab- and healthy control groups except for macular central inner limiting membrane to retinal pigment epithelium thickness (ILM-RPE), RNFL upper thickness, and temporal thickness (P < 0.05). Significant differences were observed in all variables between the AQP4-Ab + and AQP4-Ab- groups except for macular central ILM-RPE thickness (P < 0.05), as shown in Table S1.

Comparison of macular OCTA data among the three groups

The results indicated that there were no statistically significant differences in macular inner ring superior perfusion density (P = 0.305 and P = 0.304) and foveal avascular zone (FAZ) area in the central fovea (P = 0.083 and P = 0.789). The healthy control group showed significantly higher macular central vessel density, inner ring vessel density, outer ring vessel density, overall vessel density, inner ring perfusion density, outer ring perfusion density, and overall perfusion density compared to the other two groups, with all differences being statistically significant (all P < 0.05), as shown in Table 2.

Table 2 Comparison of OCTA data in the macular region between the NMOSD and healthy control groupsComparison of optic disc OCTA data among the three groups

The healthy control group exhibited significantly higher optic disc central vessel density, inner ring vessel density, outer ring vessel density, overall vessel density, optic disc central perfusion density, outer ring perfusion density, and overall perfusion density compared to the other two groups, with all differences being statistically significant between the HC and AQP4-Ab + groups (all P < 0.05); all differences were statistically significant between the HC and AQP4-Ab- groups (all P < 0.05) except for optic disc central vessel density, overall optic disc vessel density, inner optic disc ring temporal vessel density, outer optic disc ring inferior vessel density, outer optic disc ring nasal vessel density, inner optic disc ring inferior perfusion density, and outer optic disc ring nasal perfusion density (all P > 0.05), as shown in Table 3.

Table 3 Comparison of OCTA data in the optic disc region between the NMOSD and healthy control groupsComparison of macular and optic disc OCTA data between AQP4-Ab + and double-seronegative NMOSD subgroups

Direct comparisons of OCTA parameters between AQP4-Ab + and double-seronegative patients revealed statistically significant differences in several key parameters:

Significant differences were observed in outer macular ring nasal vessel density (P = 0.042). However, no statistically significant differences were observed in most other macular parameters between the subgroups (all P > 0.05).Several parameters differed significantly, notably optic disc central vessel density (P < 0.001), inner optic disc ring vessel density (P = 0.010), inner optic disc ring superior vessel density (P = 0.008), inner optic disc ring nasal vessel density (P = 0.043), inner optic disc ring perfusion density (P = 0.009), and inner optic disc ring nasal perfusion density (P = 0.047). The remaining optic disc OCTA parameters did not demonstrate statistically significant subgroup differences (P > 0.05).These findings indicate that while certain vascular changes around the optic disc and specific macular regions may distinguish AQP4-Ab + patients from double-seronegative patients, these differences were not universally present across all OCTA.

Analysis of AUC values for diagnostic indicators in the macular and optic disc regions

The diagnostic efficacy of various parameters, including macular thickness, GCC thickness, RNFL thickness measured by OCT, and superficial vessel density and perfusion density measured by OCTA in the optic disc and macular regions, was assessed using the area under the curve (AUC) values for distinguishing AQP4-Ab + NMOSD. The results are presented in Table 4; Fig. 1, and Figure S1. From Table 4, it is evident that, except for the RNFL average thickness (all P > 0.05), all other parameters showed statistically significant differences among the groups (all P < 0.05) and could effectively differentiate AQP4-Ab + NMOSD. Among these parameters, optic disc vessel density demonstrated the highest diagnostic efficacy, with AUC values exceeding 0.8. The diagnostic efficacy of macular perfusion density was similar to macular vessel density (excluding the central macula), both with AUC values exceeding 0.7. Among all evaluated OCTA parameters, nasal optic disc vessel density demonstrated the highest area under the curve (AUC = 0.829) in distinguishing AQP4-Ab + NMOSD patients from healthy controls.

Table 4 Comparison of AUC data for various diagnostic indicatorsFig. 1

Receiver operating characteristics (ROC) curves for the accuracy of vessel density and perfusion density in different parts of the eye measured by optical coherence tomographic angiography (OCTA) for distinguishing AQP4 antibody-positive neuromyelitis optica spectrum disorder

Correlations between vision and OCTA parameters

As shown in Table 5, BCVA is correlated with outer macular ring nasal vessel density (r = 0.246, P = 0.033), optic disc central vessel density (r = 0.260, P = 0.023), inner optic disc ring vessel density (r = 0.353, P = 0.002), inner optic disc ring superior vessel density (r = 0.250, P = 0.030), overall optic disc vessel density (r = 0.247, P = 0.031), inner optic disc ring nasal vessel density (r = 0.333, P = 0.003), inner optic disc ring temporal vessel density (r = 0.361, P = 0.001), outer optic disc ring superior vessel density (r = 0.309, P = 0.007), outer optic disc ring nasal vessel density (r = 0.255, P = 0.026), outer optic disc ring perfusion density (r = 0.328, P = 0.004), overall optic disc perfusion density (r = 0.312, P = 0.006), inner optic disc ring nasal perfusion density (r = 0.328, P = 0.004), and inner optic disc ring temporal perfusion density (r = 0.297, P = 0.009).

Table 5 The correlations between vision and OCTA