Abstract:AIM:To compare the consistency of macular microvascular parameters measured by three devices: frequency-domain optical coherence tomography angiography(FD-OCTA)with split-spectrum amplitude-decorrelation angiography(SSADA), FD-OCTA with optical microangiography(OMAG), and swept-source OCTA(SS-OCTA), in patients with long-term hydroxychloroquine(HCQ)administration for rheumatoid arthritis(RA).
METHODS:This prospective study enrolled patients treated with HCQ for 5 y due to RA who visited the Department of Ophthalmology of the hospital from January 2022 to November 2025. All subjects underwent a 6 mm×6 mm macular scan with three OCTA devices in random order, with a 5-minute interval between examinations by different devices. Images of the superficial capillary plexus(SCP)and deep capillary plexus(DCP)were acquired. The following microvascular parameters were compared: foveal superficial capillary plexus vessel density(F-SCP VD), parafoveal superficial capillary plexus vessel density(PF-SCP VD), foveal deep capillary plexus vessel density(F-DCP VD), parafoveal deep capillary plexus vessel density(PF-DCP VD), and the area of the foveal avascular zone(FAZ).
RESULTS:A total of 34 female patients(68 eyes)treated with HCQ for 5 y due to RA were enrolled in this study, aged 38-58 y with a mean age of 49.3±5.05 y. Statistically significant differences were found in F-SCP VD, PF-SCP VD, F-DCP VD, PF-DCP VD, and FAZ area among measurements obtained by SS-OCTA, SSADA-based FD-OCTA, and OMAG-based FD-OCTA(all P<0.05). No statistically significant differences in all above parameters were observed between SSADA-based FD-OCTA and OMAG-based FD-OCTA(all P>0.05). FAZ areas measured by the three devices were highly positively correlated(|r|≥0.8, P<0.01). The F-SCP VD, PF-SCP VD, F-DCP VD, and PF-DCP VD values obtained by SSADA-based FD-OCTA and OMAG-based FD-OCTA were all highly positively correlated(|r|≥0.8, P<0.01). The PF-SCP VD and PF-DCP VD values measured by SS-OCTA versus SSADA-based FD-OCTA and OMAG-based FD-OCTA showed moderate positive correlations(0.5≤|r|<0.8, P<0.01). The F-SCP VD and F-DCP VD values measured by SS-OCTA versus SSADA-based FD-OCTA and OMAG-based FD-OCTA exhibited low positive correlations(0.3≤|r|<0.5, P<0.05). Bland-Altman plots demonstrated good agreement between SSADA-based FD-OCTA and OMAG-based FD-OCTA for the measurements of F-SCP VD, PF-SCP VD, F-DCP VD and PF-DCP VD. By contrast, poor agreement was observed when comparing SS-OCTA with SSADA-based FD-OCTA or OMAG-based FD-OCTA for measurements of F-SCP VD, PF-SCP VD, F-DCP VD and PF-DCP VD.Satisfactory agreement was found among the three OCTA devices for FAZ area measurements.
CONCLUSION:SSADA-based FD-OCTA and OMAG-based FD-OCTA show favorable correlation and agreement in measuring macular vessel density of patients receiving HCQ, which allows cautious interchangeable use during clinical follow-up. The vessel density parameters measured by SS-OCTA were markedly different from those obtained by SSADA-based FD-OCTA and OMAG-based FD-OCTA; therefore, these devices cannot be directly interchanged.FAZ area measurements exhibit favorable inter-device agreement among the three systems, and interchangeability should be considered in the context of clinical practice.
