OCT retinal angiography features in patients with rheumatoid arthritis: A pilot study

Introduction

Antimalarial drugs, specifically chloroquine and hydroxychloroquine, are currently used in the treatment of some autoimmune diseases including rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), Sjogren’s syndrome. ¹ This therapy often represents a first line treatment because the side effects are substantially infrequent and the drug is well tolerated.^2–5 In addition to the possible secondary effects on the gastro-enteric tract, skin as well as nervous system, there is a direct toxicity involving the central retina, positively correlated with the daily dose, the cumulative dose, and the duration of treatment. ⁶ The precise mechanisms underlying chloroquine/hydroxychloroquine toxicity are not fully understood. The most supported hypothesis is based on the high binding affinity of the drugs with the melanin in the retinal pigment epithelium (RPE) and the subsequent cell damage of RPE and photoreceptors.^7–9 However, some authors argue that the simple binding of the drug to melanin is not predictive of retinal toxicity. ¹⁰ Very recently, it has been hypothesized a role of the retinal microcirculation in the evolution of retinal damage induced by hydroxychloroquine.^11–15 In particular, patients with initial retinal toxicity related to the use of hydroxychloroquine would show a reduction, assessed by OCT angiography, in the vessel density (VD) of the retinal capillary plexuses and choriocapillary. Only at a later stage it would be possible to recognize retinal toxicity by means of instrumental tests, including visual field, structural OCT, autofluorescence, or multifocal electroretinogram examinations, as recommended by the American Academy of Ophthalmology for hydroxychloroquine retinal toxicity screening. ¹⁶

However, the latter hypothesis appears not entirely plausible. Autoimmune rheumatic diseases have a “vascular” component in their definition. Indeed, the alterations of the circulation of the large vessels and of the peripheral microvessels in patients suffering from SLE and AR are well known with implications of different pathogenic mechanisms and including cell-mediated inflammation, immune complex-mediated inflammation and autoantibody-mediated inflammation; SLE and RA share a common pathogenic pathway with endothelial cell activation and elevated levels of endothelin-1 potentially inducing vascular dysregulation.^17–20

Primary anomalies of the retinal microvascular system secondary to autoimmune rheumatic diseases could therefore be the normal expression of a general pathological vascular mechanism and not secondary to antimalarial drug toxicity. Our hypothesis assumes that there are alterations of the macular retinal microcirculation in RA patients and it is therefore of primary importance to define the vascular pattern of the macular superficial and deep retinal capillary plexuses before arguing that chloroquine induces a reduction in vascular density with a consequent pathogenetic mechanism in chloroquine retinal toxicity.

The aim of this study is the evaluation of the retinal superficial and deep capillary plexus by OCT angiography in treatment-naive subjects with recent diagnosis of RA.

Patients and methods

Patients affected by RA and treatment-naïve were prospectively enrolled in the current study. In adherence to the tenets of the Declaration of Helsinki, the research was carried out at the Department of Ophthalmology of the G. B Bietti Foundation in Rome, and the institutional review board approval was obtained before starting the study. Each patient was carefully informed about the purpose of the research and provided signed consent to all procedures and data analysis.

Inclusion criteria were as follows: recent diagnosis of “definite RA” based on 2010 Rheumatoid Arthritis Classification Criteria (patients were included with score >6 taking into consideration the following four main criteria: joint involvement, serology, acute-phase response measure, symptom duration, Table 1); ²¹ no previous or current treatment related to RA; no concomitant systemic disorders including diabetes, cardiovascular diseases, hypertension; no previous or concomitant ocular disease able to confound the appropriate examination including media opacity and pathologic myopia, and especially any type of retinal disorder; no history of ocular surgery. Lastly, patients with cylindrical refractor error more than 2.00 D or spherical refractor error more than 4.00 D were excluded to limit influence of axial length on proper assessment.

OPEN IN VIEWER

Table 1.

2010 Rheumatoid arthritis (RA) classification criteria. (American College of Rheumatology).

Classification criteria for rheumatoid arthritis (a total score of >6/10 is needed for classification of a patient as having definite RA)	Score
A. Joint involvement
1 large joint	0
2–10 large joints	1
1–3 small joints (with or without involvement of large joints)	2
4–10 small joints (with or without involvement of large joints)	3
>10 joints (at least one small joint)	5
B. Serology (at least one test result is needed for classification)
Negative RF and negative ACPA	0
Low-positive RF or low-positive ACPA	2
High-positive RF or high-positive ACPA	3
C. Acute-phase reactants (at least one test result is needed for classification)
Normal CRP and normal ESR	0
Abnormal CRP or abnormal ESR	1
D. Duration of symptoms
<6 weeks	0
>6 weeks	1

RF: rheumatoid factor; ACPA: anti-citrullinated protein antibody; CRP: C-reactive protein; ESR: erythrocyte sedimentation.

All patients underwent a comprehensive ophthalmological examination inclusive of the best-corrected visual acuity (BCVA) assessment on standard ETDRS LogMAR charts at 4 m, slit lamp examination, tonometry, dilated fundus examination, structural, and angiographic optical coherence tomography (OCT/OCTA). The OCTA technology is based on reconstructing the retinal and subretinal vascular network by using a source signal derived from the movement of erythrocytes inside the vessels, which after appropriate encoding makes it possible to trace the trajectories of vascular flows and consequently map the vascular network. Retinal microvascular imaging was evaluated by mean of OCTA, Optovue XR Avanti, Fremont, CA, and was performed on all patients. The scans were centered on the fovea and the images were acquired with protocol 6 mm× 6-mm. After (image) acquisition, the internal software automatically delineated the superficial capillary plexus (SCP) and the deep capillary plexus (DCP). The AngioVue software detects the SCP as the layer included between approximately 3 μm below the internal limiting membrane and 15 μm below the inner plexiform layer whereas the DCP was determined from 15 to 70 μm below the inner plexiform layer. Values of the vessel density (VD) of SCP and DCP, central foveal thickness (CFT), and of the foveal avascular zone (FAZ) were collected after automatic analysis provided by AngioAnalytics software. In more detail, VD of the global area, foveal area (circle 1 mm in diameter centered on the fovea) and parafoveal region (between two concentric circles of 1 and 3 mm diameter) of the SCP e DCP were collected, providing specific values of the temporal, nasal, superior, and inferior regions of the parafoveal area. OCTA scans with poor quality (signal strength below index 50, motion artifacts, and inappropriate segmentation) were excluded from the analysis.

Healthy subjects matched for age and sex were enrolled as control group and underwent a complete ophthalmological examination including OCTA imaging with the same protocol. Detailed clinical history examination was carried out to exclude any type of systemic disorders, drug assumption, or previous ocular disease.

Primary outcome measure was the assessment of the VD alterations in the retinal capillary plexus in subjects affected by RA with recent diagnosis and treatment-naïve, and the comparison of the data with a healthy control group.

The Shapiro–Wilk Test was used to check the assumption of normality of the variables. Parametric (t-test) and non-parametric (Mann–Whitney) tests were applied according to variable distribution. Demographic and clinical data were summarize for the descriptive statistic presenting the mean value as well as the standard deviation. All statistical comparisons were performed using two-tailed significance tests with p < 0.05 considered statistically significant. One eye was randomly selected for each patient according to a computer-generated random sequence. MedCalc Statistical Software (version 19.8, Ostend, Belgium) was used for all analyses.

Results

Sixteen eyes of 16 RA patients fully matched inclusion and exclusion criteria thus resulting eligible for the study purposes and being included in the final analysis. As per protocol, 16 healthy subjects, age- and sex-matched, were identified and recruited in the study. Signal strength index was 76 ± 7 in the RA group and 78 ± 5 in the healthy control group (p: 0.44). Demographic and clinical data are summarized in the Table 2.

OPEN IN VIEWER

Table 2.

Demographics of rheumatoid arthritis patients and controls.

	RA (mean ± SD); ratio	Healthy control (mean ± SD); ratio	p
Age, years	61.2 ± 12	59.2 ± 12	0.63
Gender (female/male)	8/8	8/8
RA Symptoms duration (months)	5.6 ± 2.7
BCVA, letter score (Snellen)	84 ± 1.8 (20/20−)	84.6 ± 1.2 (20/20−)	0.26
CFT (μm)	268 ± 19	269 ± 15	0.88

RA: rheumatoid arthritis; BCVA: best-corrected visual acuity; CFT: central foveal thickness.

The RA patients did not differ with the control group with regard to age, gender, BCVA, and CFT. Symptoms’ duration related to RA lasted for 5.6 months before recruited patients were included in the study.

In the RA patient, OCT-A data analysis showed a statistically significant reduction in the VD in the mean global area, inner ring, especially in the superior quadrant of the SCP in comparison with healthy control group (Table 3).

OPEN IN VIEWER

Table 3.

Vessel density analysis of superficial capillary plexus (SCP), deep capillary plexus (DCP) and foveal avascular zone (FAZ) values.

	RA (mean ± SD)	Healthy control (mean ± SD)	p
Vessel density (%)
SCP Global area	45.3 ± 4.6	48.5 ± 2.8	0.02
SCP CSF	25.2 ± 7.5	22.9 ± 5.0	0.30
SCP Inner ring	47.8 ± 6.3	51.5 ± 3.6	0.04
SCP Temporal	47.9 ± 6.9 (median:48.8)	50.4 ± 3.7 (median:49.7)	0.21 (0.50, M–W test)
SCP Superior	47.2 ± 6.1	52.0 ± 3.9	0.01
SCP Nasal	46.9 ± 6.8	50.2 ± 4.2	0.11
SCP Inferior	49.1 ± 6.7	52.0 ± 4.1	0.16
DCP Global area	45.7 ± 6.3	48.6 ± 3.1	0.10
DCP CSF	37.3 ± 8.9	35.7 ± 6.7	0.55
DCP Inner ring	51.2 ± 5.0	50.8 ± 3.1	0.75
DCP Temporal	52.1 ± 5.7	51.9 ± 3.1	0.89
DCP Superior	50.4 ± 5.9	50.6 ± 2.7	0.90
DCP Nasal	51.7 ± 6.0	50.3 ± 3.6	0.44
DCP Inferior	50.8 ± 4.5	50.0 ± 4.3	0.63
FAZ Value mm2	0.22 ± 0.07	0.25 ± 0.09	0.27

RA: rheumatoid arthritis; SCP: superficial capillary plexus; DCP: deep capillary plexus; FAZ: foveal avascular zone; CSF: central subfield.

Data of temporal SCP was not normally distributed in RA group and Mann–Whitney test was applied. Comparisons showing a statistically significant difference appear in bold.

Images of representative cases are illustrated in the Figure 1.

OPEN IN VIEWER

Figure 1.

Optical coherence tomography angiography (OCT-A) and capillary perfusion density analysis.

A trend of VD reduction was also registered in temporal, nasal, and inferior quadrants respectively, although it did not reach a statistically significant value. Assessment of VD of the DCP revealed similar outcomes in the two groups. No significant difference between the two groups was measured considering the FAZ value although it was slightly lower in the RA group.

Discussion

The main finding in this pilot study was that OCT-A analysis in eyes of treatment-naive patients and with recent diagnosis of RA outlined a statistically significant reduction in VD in the superior area, in the inner ring as well as in the whole image of the SCP when compared to a healthy control group. A trend of VD reduction was also registered in almost all parameters of the SCP although it did not reach a statistically significant value. No meaningful difference between the two groups was measured considering the parameters of the DCP.

Similar to other retinal vascular disorders – including diabetic retinopathy – RA manifests through cellular and molecular pathogenetic mechanisms involving the vascular wall. Elevated levels of circulating immune complexes have been demonstrated in patients with rheumatoid vasculitis often in association with elevated serum rheumatoid factor levels. The deposits of immune complexes contribute to the damage mechanism of the vascular wall and pathological specific feature of RA is the infiltration of mononuclear and neutrophil cells of the wall of large and small vessels.^18,22,23 Peripheral ischemia with a reduction in capillary density have been demonstrated in the peripheral capillary circulation of the extremities of the limbs and in the oral and labial mucosa.^17,24 The current study shows that the vascular remodeling is also detectable in the retinal capillary circulation and further extends the concept of peripheral vascular damage in addition to the well-defined mixed model damage – vascular and inflammatory – in the joints.

Alterations of the retinal macrocirculation in association with RA have been previously demonstrated. ²⁵ Anyfanti et al. analyzed the arterial and venous retinal vascular diameters and the corresponding arteriovenous ratio in a sample study of 113 patients and correlated the morphometry of the retinal vessels with subclinical arteriosclerosis and cardiovascular risk. The authors demonstrated a reduction in the caliber of the arterial vessels and a reduction in the AVR; an association between the morphological alterations of the retinal vessels and the index of atherosclerosis of the large carotid vessels was also evidenced, suggesting a role for the systemic inflammatory processes. Finally, it was hypothesized that the narrowing of the retinal arterial vessels could have an independent predictive role for cardiovascular risk.

In addition, changes in peripheral microcirculation of the extremities and labial mucosa have been closely correlated with an increased cardiovascular risk.^26,27 Therefore, it would be relevant to investigate the analogies of changes within the retinal capillary circulation and the role of the retinal vascular density index as biomarkers of vascular remodeling and the possible correlation with cardiovascular risk.

The anatomic changes of retinal microcirculation found in this specific population of RA patients suggests to reconsider the evaluation studies on the retinal toxic effects of therapy with antimalarials. Recent studies have shown through OCTA a reduction in the vascular density of the retinal capillary plexuses and an enlargement of the FAZ in patients suffering from rheumatic disease including RA, SLE, Sjogren’s syndrome, treated with chloroquine/hydroxy chloroquine, when compared to a group of healthy subjects.^11–15 The results of the studies also seem to underline that the drug cumulative dose and the duration of the treatment correlate with the degree of vascular density reduction suggesting the OCTA as a screening method for the early detection of retinal toxicity.

On the contrary, data collected in this study suggest that the reduction in the VD is probably the effect of inflammatory processes related to RA rather than a secondary modification to chloroquine/hydroxychloroquine therapy. Therefore, the role of antimalarial drug therapy could be differently interpreted. The treatment may have any effects on the retinal microcirculation or it may have a protective effect still preserving a good circulatory function. Otherwise it could have an additional negative effect. Longitudinal studies are needed to validate the real effects of antimalarial therapy on retinal microcirculation. The vascular remodeling of the retinal microcirculation could be used as a novel biomarker for the evaluation of the efficacy of the therapies or – in addition and as assumed in previous studies – could be investigated as a screening test for early identification of retinal toxicity.

Our study has some limitations that are common to pilot studies like the sample study which can be considered small in size. However, the study population was well selected to reduce confounding factors including systemic diseases often associated with RA patients, systemic therapies. The healthy control group was also well matched. Additional limitation is related to the instrumentation in use namely the OCTA Optovue XR Avanti as different devices may have provided different results. However, in this specific case the results of the study can be easily compared with most of the previous studies which used the same instrument and the same acquisition protocols. Furthermore, the data collected in the current study find application in the analysis of retinal microcirculation in RA and further studies must be conducted before exporting these concepts to different autoimmune diseases such as Sjogren’s syndrome or SLE. Additional studies are also needed to explain and confirm the apparent greater susceptibility of SCP to change in RA patients compared to DCP and the variability in the different quadrants. We can assume that the early anatomical alterations of SCP found in this preliminary study are attributable to RA and they could be included in the capillaropathy characterizing this rheumatic disease.

The results of this study offer a new area of application of OCTA and seem to expand the potential of this tool also to other fields of medicine. The instrument could be used to define a staging of retinal microvascular damage in RA and potentially provide classification indexes for vascular pathology. Vascular remodeling could be used as index of disease progression and response to systemic therapy not only with antimalarial drugs but also with immunomodulatory or biologic drugs. Similarly to peripheral vascular changes associated with RA including nail and labial flow changes, vascular modification in the retinal capillary circulation could be included in new cardiovascular risk assessment methodologies. Lastly, it would be very interesting to investigate the existence of a parallelism between the retinal microcirculation variations and the severity of joint changes.

In conclusion, OCTA allows to highlight alterations of the retinal microcirculation in subjects suffering from RA even in the early stages of the disease, and they are well characterized before starting therapy with antimalarial drugs. The data obtained offer new perspectives in the pathogenetic definition of RA and broaden the horizon on the diagnostic potential of this instrument.