Application of Optical Coherence Tomography Angiography for hydroxychloroquine-induced microvascular changes: a systematic review and meta-analysis

doi:10.21203/rs.3.rs-2222786/v1

Download PDF

Research Article

Application of Optical Coherence Tomography Angiography for hydroxychloroquine-induced microvascular changes: a systematic review and meta-analysis

https://doi.org/10.21203/rs.3.rs-2222786/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 08 Mar, 2023

Read the published version in Graefe's Archive for Clinical and Experimental Ophthalmology →

Version 1

posted

You are reading this latest preprint version

Background

Retinal toxicity with long-term hydroxychloroquine (HCQ) treatment is a major concern. This systematic review aims to assess the application of Optical Coherence Tomography Angiography (OCTA) to detect HCQ-induced microvascular alterations.

Methods

PubMed, Scopus, Web of Science, and Cochrane Library databases were systematically searched until November 1st, 2021. Studies using OCTA as a primary diagnostic method to evaluate the macular microvasculature of HCQ users were included. Primary outcomes were macular vessel density (VD) and foveal avascular zone (FAZ) at the superficial (SCP) and deep (DCP) capillary plexus. Meta-analysis was performed using a random-effects model.

Results

Of 159 screened abstracts, seven were found eligible, enrolling 424 eyes from 335 patients. High-risk patients due to longer duration of treatment presented lower VD in the retinal microvasculature than those with low-risk in SCP (P = 0.01 in fovea; P = 0.002 in parafovea) and in DCP (P = 0.03 in fovea). When compared with healthy controls, HCQ users had lower VD in both plexus and a wider FAZ area – no quantitative synthesis was presented.

Conclusions

Microvascular changes were found in autoimmune patients under HCQ treatment without any documented retinopathy. However, the evidence produced so far does not allow to draw conclusion concerning the effect of drug as studies were not controlled for disease duration.

Autoimmune diseases

Hydroxychloroquine

Macula Lutea

Microvasculature

Optical Coherence Tomography Angiography

Hydroxychloroquine (HCQ) is an antimalarial drug widely used by many specialties for the management of systemic lupus erythematosus (SLE), rheumatoid arthritis (RA), and other autoimmune conditions. It is being considered for broader applications[1], including cancers, multiple sclerosis, primary antiphospholipid antibody syndrome, hypertension[2] and diabetes mellitus[3]. More recently, it was used in the first months of the COVID-19 pandemy[4,5]. Despite the few systemic adverse effects, long-term use of HCQ may lead to irreversible and vision-threatening toxic retinopathy[1]. Although it is considered rare, it may occur in up to 7.5% of patients[6]. The pathophysiology is not fully understood, but it is thought to induce retinal damage by inhibiting retinal pigment epithelial (RPE) cells autophagy, increasing their concentration of lipofuscin, which is linked to photoreceptor degradation. In addition, it also disrupts the recycling of all-trans-retinol in RPE cells, an important step in the visual cycle[1]. It is known that a long duration of exposure (> five years) confers a higher risk of retinal toxicity[6]. Additional risk factors include high dose (>5 mg/kg/day), concomitant tamoxifen therapy, and renal failure[7].

Early detection of retinal toxicity is crucial to avoid irreversible damage and vision loss. Nowadays, the methods used to screen HCQ-induced retinopathy include Spectral Domain Optical Coherence Tomography (SD-OCT), visual field testing, fundus autofluorescence, and multifocal electroretinography (mERG). OCT Angiography (OCTA) is a novel non-invasive imaging modality based on the variation in the signal caused by moving particles, such as red blood cells, allowing the assessment of retinal and choroidal microvasculature without the need of dye injection. Several recent studies detected microvascular changes in patients using HCQ without evident microangiopathy or retinopathy on ophthalmoscopy and complementary exams such as SD-OCT and fluorescein angiography[8-10]. In addition, OCTA may help elucidate the mechanisms of HCQ-induced retinal toxicity.

Several observational publications exist on this subject, but a systematic review and a meta-analysis are yet to be published to our best knowledge. We sought to conduct a systematic review and meta-analysis of the OCTA technique to detect HCQ-induced microvascular alterations and further characterize them.

We conducted a systematic review with meta-analysis following the guidelines presented by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement[11] and the Meta-Analyses and Systematic Reviews of Observational Studies (MOOSE) statement[12]. The protocol was not registered.

2.1. Search strategy

Two independent reviewers (A.F. and R.A.) searched PubMed, Web of Science, Cochrane Library, and Scopus databases for all relevant studies published from inception to November 1^st, 2021. Search terms used are detailed in Supplementary Table 1. References of relevant articles were hand-searched, and a forward citation search was performed.

2.2. Study selection

The following inclusion criteria were applied: a) Type of studies: Observational studies. Articles in English, Portuguese or Spanish were included; b) Type of participants: i) studies with the primary focus on patients treated with HCQ; ii) studies with a healthy control group or comparing patients with high-risk versus low-risk of HCQ-induced retinopathy; iii) population > 18 years old; c) Type of interventions: Studies where OCTA was performed; d) Type of outcome: Studies including measurements performed in at least one of the following: i) foveal avascular zone (FAZ); ii) vessel density (VD) at the macular superficial capillary plexus (SCP); iii) VD at the macular deep capillary plexus (DCP). No restrictions regarding time or country of origin were placed during the search process.

The exclusion criteria were as follows: a) duplicate publications, conference presentations, summaries; b) studies in patients with ocular diseases; c) studies in patients with other relevant systemic co-morbidities and nonhuman studies; d) qualitative studies; e) review studies and meta-analysis were excluded although bibliographic references were manually reviewed.

2.3. Data collection process

Two authors (A.F. and R.A.) independently reviewed each title and study to exclude duplicates and irrelevant studies. The two authors performed a full-text review to assess study eligibility, and disagreements were resolved by consensus. Authors were contacted to acquire any missing information. Data was extracted using Microsoft® Word. Our primary outcomes were: FAZ area and VD at the macular SCP and DCP. Whenever possible, the analysis of choriocapillaris, FAZ perimeter, and flow area was performed. OCTA parameters were only synthesized for analysis if they were reported in a minimum number of three papers. When data were reported in subgroups, combination onto a single group was performed using a methodology described elsewhere[13]. Data extracted included study setting (first author, publication, year); participant characteristics (age, gender, eyes included, daily doses, duration of disease, and duration of treatment); study design; OCTA characteristics (model, scan size, and scan definition); image analysis (software) and main outcomes.

2.4. Quantitative analysis

Meta-analyses of continuous outcomes were conducted with the Cochrane Collaboration's Review Manager software (Version 5.3) using an inverse variance in the model. Two consecutive analyses were performed: i) comparing patients with high-risk versus low-risk of HCQ-induced retinopathy, based on the duration of exposure using five years as cut-off; ii) comparing patients using HCQ versus healthy controls. Random-effects models were chosen for both analyses.

Cochran's Chi-squared test for homogeneity (Chi2) was used to assess heterogeneity between studies, and variation due to heterogeneity was estimated by calculating the I² and interpreted as recommended[14,15]. Following Cochrane Guidelines[16], meta-regression was not performed as there were fewer than ten studies for each OCTA parameter that was meta-analyzed. Funnel plots were used to assess publication bias.

Due to unacceptable heterogeneity (ranging from 79% to 96%) not solved with a sensitivity analysis, the quantitative analysis of HCQ patients versus healthy controls comparison was excluded.

2.5. Risk of bias in individual studies and sensitivity analysis

Two reviewers (AF and R.J.-V.) independently evaluated the risk of bias for each eligible study using a modified Moskalewicz and coworkers.[17] Seven-Question Newcastle-Ottawa Scale (NOS) questionnaire to our study-specific language as detailed in Supplementary Methods 1. A sensitivity analysis was conducted for analysis with an I² > 60%.

3.1. Literature search

The study selection process is shown in a flow diagram (shown in Fig. 1). The whole search process retrieved 159 original abstracts, 11 of which met the inclusion criteria. After full-text review, four were excluded due to patient enrollment with mERG changes[18], clinical case[19] or conference abstract[20] structure, or lack of OCTA parameters defined in the inclusion criteria[21]. Eight studies were included in qualitative and quantitative syntheses[8-10,22-25]. No previous systematic reviews with meta-analysis on this specific issue were found.

3.2. Study characteristics

Tables 1 and 2 summarize the characteristics of the eight studies included. All studies were cross-sectional and published between 2018 and 2021. A total of 424 eyes from 335 patients were enrolled, including 184 patients with autoimmune diseases. All studies used age and sex-matched groups. The overall mean age varied between 38.4 and 55.6 years, and the proportions of women spanned between 66.7% and 100%. The OCTA devices used were AngioVue (Optovue Inc., Fremont, CA, USA) in five studies[8,9,22,24,25] and Cirrus HD-OCT 5000 (Carl Zeiss Meditec, Jena, Germany)[23] and DRI-OCT Triton Plus (Topcon, Tokyo, Japan)[10] in one study each. All devices are Spectral Domain OCTA with the exception of the Triton Plus, which uses the Swept Source technology. The OCTA parameters analyzed in each study are depicted in Supplementary Table 2.

Four studies[8,9,24,25] compared patients with high-risk versus low-risk of HCQ-induced retinopathy using five years as the cut-off. No differences were found in the mean daily doses of HCQ between groups in the three studies[8,24,25] reporting this parameter. However, two studies[8,24] reported significant differences between groups in the cumulative dose of HCQ. The mean duration of the treatment ranged from 1.69 to 10.33 years and was found in all studies. Only one study[8] reported that the disease duration was significantly different between groups.

Five studies[9,10,22-24] compared HCQ patients versus healthy controls. The mean duration of treatment was superior to five years in all reports, with two studies[10,22] having this threshold as inclusion criteria. Goker and coworkers[22] included only patients with a dose inferior to 5 mg/kg/day, whereas Lopes and coworkers[23] reported a mean dose of 5.81±1.37 mg/kg/day.

3.3. Quantitative analysis of high-risk versus low-risk HCQ patients

High-risk versus low-risk HCQ patients analysis is shown in the forest plot of Figure 2.

3.3.1. Superficial capillary plexus (shown in Fig. 2A)

Regarding the fovea, a significant reduction of VD (SMD -0.53 [-0.93, -0.12], P = 0.01) in high-risk HCQ patients with moderate heterogeneity (I² = 47% and P = 0.130). In the parafovea, a significant reduction of VD (SMD -0.44 [-0.73, -0.16], P = 0.002) in high-risk HCQ patients with low heterogeneity (I² = 0% and P = 0.590).

3.3.2. Deep capillary plexus (shown in Fig. 2B)

In the fovea, the meta-analysis revealed a significant reduction of VD (SMD -0.53 [-1.01, -0.06], P = 0.03) in high-risk HCQ patients with substantial heterogeneity (I² = 61% and P = 0.05). The sensitivity analysis revealed a range of 57%-65% for heterogeneity with the exclusion of one study at a time. The lower value was found with the exclusion of Bulut and coworkers[8], leading to a non-significant the overall effect (SMD -0.39 [-0.94, 0.16], P = 0.16).

Considering the parafoveal zone, high-risk HCQ patients presented a trend for reduction of VD (SMD -0.64 [-1.33, 0.06], P = 0.07) with substantial heterogeneity (I² = 81% and P = 0.001). The sensitivity analysis revealed a heterogeneity range of 60%-85% with the exclusion of one study at a time. The lower value was found with the exclusion of Cinar and coworkers[26], leading to a non-significant the overall effect (SMD -0.31 [-0.80, 0.18], P = 0.22).

3.3.3. FAZ area (shown in Fig. 2C)

The study by Tarakcioglu and coworkers[25] was not included in this analysis as FAZ was reported as a whole instead of discriminated by plexuses. Regarding the SCP, no difference was found between groups (P = 0.22). In the DCP, high-risk HCQ patients displayed a trend for foveal avascularity increase (SMD 0.74 [-0.02, 1.50], P = 0.06) with substantial heterogeneity (I² = 75% and P = 0.02). The sensitivity analysis of the DCP revealed a heterogeneity range of 0%-84% with the exclusion of one study at a time. The lower value was found with the exclusion of Ozek and coworkers[24] leading to a significant the overall effect (SMD 1.10 [0.65, 1.56], P < 0.001).

3.4. Results of HCQ patients versus healthy controls comparison (Table 3)

3.4.1. Superficial capillary plexus

Regarding the fovea, a significant reduction of VD in HCQ patients was reported by three studies[26,22,23] while only two studies[26,23] found the same difference in the parafoveal zone.

3.4.2. Deep capillary plexus

A significant reduction of VD in HCQ patients was found at fovea in four studies[26,10,22,23] and at parafovea in two studies[26,24] The study by Lopes and coworkers[23] found a trend toward a decrease of VD in the parafovea.

3.4.3. FAZ area

Regarding the SCP, a significant increase of FAZ in HCQ patients was described in three studies[26,10,22] while in the DCP only two studies[26,10] reported this difference.

3.5. Bias assessment

Publication biases were investigated by plotting funnel plots (shown in Supplementary Figures 1), which revealed a symmetrical distribution of studies about the SMD, indicating little to neither publication bias nor small study bias. Bias analysis was performed in all seven studies that were considered eligible and are shown in Supplementary Table 3, Supplementary Figure 2, and Supplementary Graph 1.

OCTA is a non-invasive method that allows for visualization and quantification of the microvascular structure of the retina and choroid. In this systematic review with meta-analysis, we summarized the evidence on the use of OCTA to assess HCQ-induced microvascular retinal changes. Patients with a high-risk of HCQ-induced retinopathy due to the longer duration of treatment (> five years) presented lower VD in the retinal microvasculature in both plexuses and a trend for wider FAZ in DCP compared with those of low-risk. HCQ users had lower VD in both retinal plexuses and presented a wider FAZ than healthy controls. As already mentioned, a meta-analysis of this comparison could not be performed, and this conclusion results from an appreciation of the results without the support of a formal analysis. This work summarized studies enrolling only patients without any retinopathy.

The reductions in VD and enlargement of FAZ observed can result from either structural (absence of capillaries) or functional (non-perfusion) changes. As OCTA relies on comparing consecutive scans, flow is only detected above a minimum threshold, leaving the impression of non-perfusion for regions with flow below the detectable treshold[27]. Coupling OCTA with adaptive optics may help to elucidate further the status of retinal microvasculature as this technology provides in vivo ultra-high-resolution imaging of retinal vessel morphology[28].

4.1. Comparison of high-risk versus low-risk HCQ patients

The subjects enrolled in this comparison presented significant differences in mean treatment duration and cumulative dose, as expected, but not in the mean daily dose, which suggests that the potential microvascular changes induced by the drug depend on exposure time. In fact, in two studies[8,9], the cumulative dose and duration of exposure had a significant negative correlation with VD and a significant positive correlation with FAZ parameters. A recent non-comparative longitudinal study found no significant change in VD among RA patients during the first year of HCQ treatment, suggesting that a long exposure period is required to elicit microvascular alterations[29]. Another central point deserving consideration is the effect of disease duration on the microvasculature, primarily when considering immunologic diseases. Indeed, some authors detected an early reduction in the VD of SCP among naïve patients with recent onset of RA symptoms[29]. Only one study[8] reported that the mean disease duration was significantly different between groups, but no correlation was found between disease duration and OCTA parameters.

High-risk patients presented a trend for FAZ enlargement in the DCP but not in the SCP. However, the results of the three studies analyzed were not in agreement, with one study[8] demonstrating FAZ enlargement in both plexuses and another in none[24]. The work by Tarakcioglu and coworkers[25], excluded from the analysis as already explained, did not find a significant difference in FAZ area and perimeter between groups. The reason for these differences is unclear. A considerable variation in the FAZ size among healthy individuals has been previously observed[30]. These facts prevent us from recommending FAZ as a parameter for assessing microvascular changes.

4.2. Comparison of HCQ patients versus healthy controls

In this comparison, patients under HCQ presented a mean treatment duration superior to five years, conferring these patients a high-risk of HCQ-induced retinopathy. Two studies[9,24] presented a sub-analysis of low-risk patients versus controls, but the parameters reported were not coincident among them to permit a quantitative synthesis. These studies reported a reduction of VD in both plexuses of retinal[9,24], and choriocappilaris[9] microvasculature but not in the FAZ zone.

The quantitative synthesis of the comparison between HCQ patients and healthy controls displayed considerable heterogeneity that could not be solved with a sensitivity analysis. For this reason, this analysis was excluded. This high heterogeneity may be related to the utilization of three devices of different technologies, in-built and external software to perform the analysis, and the inclusion criteria for the treatment group.

4.3. Choriocapillaris

The choriocapillaris quantitative analysis was not performed due to an insufficient number of studies reporting this parameter. Bulut and coworkers[8] described that choroidal flow rate and thickness were reduced in high-risk patients versus low-risk ones. In addition, other two reports[9,25] found a decreased choroidal flow area in the same group. One study[23] reported a reduction of VD in choriocapillaris in HCQ patients compared with healthy subjects. Likewise, a significant reduction in choroidal thickness was described for HCQ patients, both high-risk and low-risk, compared with healthy subjects[9]. These facts suggest that the choroid is also affected and involved in the mechanism of retinal toxicity induced by HCQ.

4.4. Critical appraisal and recommendations for future research

One major limitation the studies hereby reviewed is the lack of control for the disease. Considering the immunologic diseases with , one cannot exclude the effect of the disease on the retinal microvasculature either directly or synergistically with the drug. In fact, the comparison of high-risk versus low-risk HCQ patients applied treatment duration as a unique criteria, ignoring disease duration. The other analysis compared healthy subjects with patients whose microvascular alterations could be due to two potential sources the drug and the disease. In addition, the retinal capillary changes hereby described do not adequately explain the typical alterations in outer retinal and RPE of HCQ-induced retinopathy and do not correlate with any inner retinal change. As choroidal vasculature supplies the outer retina, changes in that sector may be an essential element for that occurrence. Thus, future research must focus on comparing 1) autoimmine patients with low to no-risk of drug toxicity versus healthy subjects, to clarify the disease effect; 2) autoimmune patients under vs no HCQ treatment, adjusted for disease duration and comobilities, aiming to clarify a potential effect of the drug. In addition, the choroidal vasculare beyong choriocappilaris must also be studied. Some authors have already shown changes in choroidal thickness between healthy subjects and autoimmune patients, with opposite effects for different autoimmune diseases[31-33]. Thus, future studies should also focus on one disease instead of consider several autoimmune diseases as a unique entity.

4.5. Strengths and limitations

This review presents several limitations. Albeit all studies enrolled age- and sex-matched groups, the difference in the mean age among different studies spanned 15 years. This fact prevented us from conducting a fixed effect analysis for the first comparison where all studies used the same device. On the other hand, this age range is representative of most patients with these diseases, increasing the generability of our results. Different devices, image analysis software, and other parameters can difficult direct comparisons, but the methodology was similar to some degree. We preferred a standardized mean difference in the meta-analysis to minimize this problem. The difference in segmentation of retinal capillary plexuses may include the intermediate capillary plexus in the measurement of DCP in some OCTA devices. However, all studies used the same device for all patients, excluding a differential bias.

This systematic review presents several strengths. A significant number of papers that encompassed a representative number of patients were summarized. Our analysis was comprehensive and approached several OCTA parameters.

In summary, microvascular retinal and choroidal changes were found in patients under HCQ treatment without any documented retinopathy. Given the limitations in study design of the evidence produced so far, one cannot conclude if those alterations are a consequence of the drug or the disease. Thus, further studies controlling for disease duration and severity and drug exposure are warranted.

Acknowledgements: none

Statement of Ethics: An ethics statement is not applicable because this study is based exclusively on published literature

Conflict of Interest Statement: The authors have no conflicts of interest to declare.

Funding: None

Authors' contributions: All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising, or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Availability of data and materials: all data gathered for this review is presented in the manuscript or as supplementary files

Jorge A, Ung C, Young LH, Melles RB, Choi HK (2018) Hydroxychloroquine retinopathy - implications of research advances for rheumatology care. Nat Rev Rheumatol 14 (12):693-703. doi:10.1038/s41584-018-0111-8
Anjos R, Ferreira A, Barkoudah E, Claggett B, Pinto LA, Miguel A (2021) Application of optical coherence tomography angiography macular analysis for systemic hypertension. A systematic review and meta-analysis. Am J Hypertens. doi:10.1093/ajh/hpab172
Braslow RA, Shiloach M, Macsai MS (2017) Adherence to Hydroxychloroquine Dosing Guidelines by Rheumatologists: An Electronic Medical Record-Based Study in an Integrated Health Care System. Ophthalmology 124 (5):604-608. doi:10.1016/j.ophtha.2016.12.021
Singh B, Ryan H, Kredo T, Chaplin M, Fletcher T (2021) Chloroquine or hydroxychloroquine for prevention and treatment of COVID-19. The Cochrane database of systematic reviews 2:CD013587. doi:10.1002/14651858.CD013587.pub2
Martins-Filho PR, Ferreira LC, Heimfarth L, Araujo AAS, Quintans-Junior LJ (2021) Efficacy and safety of hydroxychloroquine as pre-and post-exposure prophylaxis and treatment of COVID-19: A systematic review and meta-analysis of blinded, placebo-controlled, randomized clinical trials. Lancet Reg Health Am 2:100062. doi:10.1016/j.lana.2021.100062
Melles RB, Marmor MF (2014) The risk of toxic retinopathy in patients on long-term hydroxychloroquine therapy. JAMA Ophthalmol 132 (12):1453-1460. doi:10.1001/jamaophthalmol.2014.3459
Yusuf IH, Foot B, Lotery AJ (2021) The Royal College of Ophthalmologists recommendations on monitoring for hydroxychloroquine and chloroquine users in the United Kingdom (2020 revision): executive summary. Eye (Lond) 35 (6):1532-1537. doi:10.1038/s41433-020-01380-2
Bulut M, Akidan M, Gozkaya O, Erol MK, Cengiz A, Cay HF (2018) Optical coherence tomography angiography for screening of hydroxychloroquine-induced retinal alterations. Graefes Arch Clin Exp Ophthalmol 256 (11):2075-2081. doi:10.1007/s00417-018-4117-3
Cinar E, Yuce B, Asian F (2021) Evaluation of retinal and choroidal microvascular changes in patients who received hydroxychloroquine by optical coherence tomography angiography. Arquivos Brasileiros de Oftalmologia 84 (1):1-10. doi:10.5935/0004-2749.20210001
Forte R, Haulani H, Dyrda A, Jürgens I (2021) Swept source optical coherence tomography angiography in patients treated with hydroxychloroquine: correlation with morphological and functional tests. Br J Ophthalmol 105 (9):1297-1301. doi:10.1136/bjophthalmol-2018-313679
Moher D, Liberati A, Tetzlaff J, Altman DG, Group P (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol 62 (10):1006-1012. doi:10.1016/j.jclinepi.2009.06.005
Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D, Moher D, Becker BJ, Sipe TA, Thacker SB (2000) Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 283 (15):2008-2012. doi:10.1001/jama.283.15.2008
Higgins JP, Deeks JJ (2008) Selecting studies and collecting data. Cochrane handbook for systematic reviews of interventions: Cochrane book series:151-185
Higgins JP, Thompson SG (2002) Quantifying heterogeneity in a meta-analysis. Stat Med 21 (11):1539-1558. doi:10.1002/sim.1186
Deeks JJ, Higgins JP, Altman DG, Group CSM (2019) Analysing data and undertaking meta‐analyses. Cochrane handbook for systematic reviews of interventions:241-284
Sterne JA, Egger M, Moher D (2008) Addressing reporting biases. Cochrane handbook for systematic reviews of interventions: Cochrane book series:297-333
Moskalewicz A, Oremus M (2020) No clear choice between Newcastle-Ottawa Scale and Appraisal Tool for Cross-Sectional Studies to assess methodological quality in cross-sectional studies of health-related quality of life and breast cancer. J Clin Epidemiol 120:94-103. doi:10.1016/j.jclinepi.2019.12.013
Akhlaghi M, Kianersi F, Radmehr H, Dehghani A, Beni AN, Noorshargh P (2021) Evaluation of optical coherence tomography angiography parameters in patients treated with Hydroxychloroquine. BMC OPHTHALMOLOGY 21 (1). doi:10.1186/s12886-021-01977-5
Kam J, Zhang Q, Lin J, Liu J, Wang RK, Rezaei K (2016) Optical coherence tomography based microangiography findings in hydroxychloroquine toxicity. Quant Imaging Med Surg 6 (2):178-183. doi:10.21037/qims.2016.01.01
Sharma P, Pefkianaki M, Adam MK, Ho A, Hsu J, Maguire J, Regillo CD, Affel CG, Affel E, Haller JA (2017) Optical Coherence Tomography Angiography in Hydroxychloroquine Toxicity. Investigative ophthalmology & visual science 58 (8)
Mihailovic N, Leclaire MD, Eter N, Brücher VC (2020) Altered microvascular density in patients with systemic lupus erythematosus treated with hydroxychloroquine-an optical coherence tomography angiography study. Graefes Arch Clin Exp Ophthalmol 258 (10):2263-2269. doi:10.1007/s00417-020-04788-4
Goker YS, Atilgan CU, Tekin K, Kiziltoprak H, Yetkin E, Karahan NY, Koc M, Kosekahya P (2019) The Validity of Optical Coherence Tomography Angiography as a Screening Test for the Early Detection of Retinal Changes in Patients with Hydroxychloroquine Therapy. Curr Eye Res 44 (3):311-315. doi:10.1080/02713683.2018.1545912
Lopes D, Loureiro T, Carreira AR, Miranda A, Pereira M, Machado I, Campos N (2020) Oct angiography reveals early changes with hydroxychloroquine therapy. Retina Today 2020 (November-December):46-49
Ozek D, Onen M, Karaca EE, Omma A, Kemer OE, Coskun C (2019) The optical coherence tomography angiography findings of rheumatoid arthritis patients taking hydroxychloroquine. Eur J Ophthalmol 29 (5):532-537. doi:10.1177/1120672118801125
Tarakcioglu HN, Ozkaya A, Yigit U (2021) Is optical coherence tomography angiography a useful tool in the screening of hydroxychloroquine retinopathy? Int Ophthalmol 41 (1):27-33. doi:10.1007/s10792-020-01549-4
Cinar E, Yuce B, Asian F (2021) Evaluation of retinal and choroidal microvascular changes in patients who received hydroxychloroquine by optical coherence tomography angiography. Arq Bras Oftalmol 84 (1):1-10. doi:10.5935/0004-2749.20210001
de Carlo TE, Romano A, Waheed NK, Duker JS (2015) A review of optical coherence tomography angiography (OCTA). Int J Retina Vitreous 1:5. doi:10.1186/s40942-015-0005-8
Bakker E, Dikland FA, van Bakel R, Andrade De Jesus D, Sanchez Brea L, Klein S, van Walsum T, Rossant F, Farias DC, Grieve K, Paques M (2021) Adaptive optics ophthalmoscopy: a systematic review of vascular biomarkers. Surv Ophthalmol. doi:10.1016/j.survophthal.2021.05.012
Iacono P, Da Pozzo S, Bedendo A, Arrigo A, Parravano M, Varano M, Battaglia Parodi M (2021) OCT retinal angiography features in patients with rheumatoid arthritis: A pilot study. Eur J Ophthalmol:11206721211035626. doi:10.1177/11206721211035626
Chui TY, Zhong Z, Song H, Burns SA (2012) Foveal avascular zone and its relationship to foveal pit shape. Optom Vis Sci 89 (5):602-610. doi:10.1097/OPX.0b013e3182504227
Ferreira CS, Beato J, Falcao MS, Brandao E, Falcao-Reis F, Carneiro AM (2017) Choroidal Thickness in Multisystemic Autoimmune Diseases without Ophthalmologic Manifestations. Retina 37 (3):529-535. doi:10.1097/IAE.0000000000001193
Dias-Santos A, Tavares Ferreira J, Pinheiro S, Cunha JP, Alves M, Papoila AL, Moraes-Fontes MF, Proenca R (2019) Choroidal thickness changes in systemic lupus erythematosus patients. Clin Ophthalmol 13:1567-1578. doi:10.2147/OPTH.S219347
Altinkaynak H, Duru N, Uysal BS, Erten S, Kurkcuoglu PZ, Yuksel N, Duru Z, Cagil N (2016) Choroidal Thickness in Patients with Systemic Lupus Erythematosus Analyzed by Spectral-domain Optical Coherence Tomography. Ocul Immunol Inflamm 24 (3):254-260. doi:10.3109/09273948.2015.1006790

Table 1 – Study design and patient characteristics of included studies that compared high-risk vs low-risk HCQ patients

Author

Year

Country

Design

Study summary/HCQ groups

N of subjects (eyes)

Mean age ± SD (years)

Female (%)

Mean dose ± SD

daily (mg/day)
daily (mg/kg)
total (g)

Mean duration ± SD (years)

- disease

- treatment

High-Risk

Low-Risk

High-Risk

Low-Risk

High-Risk

Low-Risk

High-Risk

Low-Risk

High-Risk

Low-Risk

Bulut et al

2018

Turkey

≥5y vs <5y of HCQ use

T – 30 (30)

RA – 10

SS – 12

SLE – 5

CTD - 3

T – 30 (30)

RA – 11

SS – 11

SLE – 5

CTD - 3

54.60 ± 7.42

53.07 ± 5.59

100%

268.97 ± 96.75

3.60 ± 1.03

746.73 ± 446.52

246.67 ± 86.04

3.50 ± 0.85

137.67 ± 104.56*

8.43 ± 3.29

7.74 ± 3.58

4.78 ± 2.89*

1.69 ± 1.37*

Cinar et al

2021

Turkey

HCQ use >12mo

≥5y vs <5y of HCQ use

14 (14)

Ozek et al

2019

Turkey

RA patients

≥5y vs <5y of HCQ use

24 (24)

16 (16)

45.47±12.4

38.38±15.4

83.3%

87.5%

203.23±55.32

520.34±112.71

218.34±101.12

330.12±138.56*

10.33 ± 1.79

1.90 ± 0.88*

Tarakcioglu et al

2020

Turkey

>5y vs ≤5y of HCQ use

41 (41)

AD - NR

29 (29)

AD - NR

49.2 ± 10.5

228 ± 45

221 ± 41

8.52 ± 2.69

2.48 ± 1.19*

* statistically significant compared with the high-risk group

Abbreviations: HCQ – Hydroxychloroquine; CS – Cross-sectional; NR – not reported; RA - rheumatoid arthritis; SS - Sjögren’s syndrome; SLE - systemic lupus erythematosus; CTD - connective tissue disease; AD – autoimmune disease;

Table 2 – Study design and patient characteristics of included studies that compared HCQ patients vs controls.

Author

Year

Country

Design

Study summary

N of patients (eyes)

Mean age ± SD (years)

Female (%)

Mean dose ± SD

daily (mg/day)
daily (mg/kg)
total (g)

Mean duration ± SD (years)

- disease

- treatment

HCQ

Control

HCQ

Control

HCQ

Control

HCQ

Control

HCQ

Control

Cinar et al

2021

Turkey

HCQ use >12mo

vs healthy controls

T– 28 (28)

RA – 5

SS – 5

SLE – 10

CTD - 8

28 (28)

45.5 ± 11.1

44.5 ± 13.9

85.7%

292.857 ± 85

593.714 ± 450

5.25 ± 0.93

Forte et al

2019

Spain

HCQ use >5y vs healthy controls

T – 10 (20)

All SLE

9 (18)

38.87±8.6

41.11±8.12

80%

77.8%

4.46 ± 1.47

937.59 ± 332.37

10.25 ± 3.28

10.0 ± 3.25

Goker et al

2019

Turkey

HCQ ≤ 5 mg/kg/day for ≥5y vs healthy controls

T – 20 (40)

SLE – NR

RA – NR

20 (40)

55.58 ± 8.33

54.61 ± 8.62

90%

3.08 mg/kg/day

5.85 ± 0.85

Lopes et al

2020

Portugal

HCQ users vs healthy controls

T – 15 (30)

RA – 4

SS – 4

SLE – 7

15 (30)

54.3 ± 19.5

52.4 ± 1.0

100%

5.81±1.37 mg/kg/day

1,231 ± 619.3 (cumulative)

8.9 ± 4.0

Ozek et al

2019

Turkey

HCQ users vs healthy controls

T – 40 (40)

All RA

20 (20)

vide table 1

42.06 ± 8.31

85%

80%

vide table 1

Abbreviations: HCQ –Hydroxychloroquine; AD – autoimmune disease; CS – Cross-sectional; T – Total; NR – not reported; RA - rheumatoid arthritis; SS - Sjögren’s syndrome; SLE - systemic lupus erythematosus; CTD - connective tissue disease;

Table 3 – Optical Coherence Tomography Angiography parameters for hydroxychloroquine patients vs. healthy controls. For Ozek et al., data from high-risk and low-risk HCQ patients were combined into a single group as described in the Methods and compared using an independent samples t-test using Stata software (StataCorp. 2015. Stata Statistical Software: Release 14. College Station, TX: StataCorp LP).

Author

Year

Country

Region for VD

VD at SCP

VD at DCP

Plexus for FAZ

FAZ area

HCQ

Controls

P-value

HCQ

Controls

P-value

HCQ

Controls

P-value

Cinar et al

2021

Turkey

Fovea

Parafovea

34.053 ± 1.83

53.520 ± 1.27

36.635 ± 1.22

55.771 ± 1.28

0.013

0.011

36.157 ± 0.71

55.446 ± 1.01

37.978 ± 0.55

57.285 ± 0.56

0.032

0.030

Superficial

Deep

0.331 ± 0.014

0.357 ± 0.010

0.310 ± 0.018

0.309 ± 0.018

0.034

0.013

Forte et al

2019

Spain

Fovea

22.65±4.97

23.66±3.78

0.2

19.89±4.45

22.06±6.03

0.03

Superficial

Deep

250.25±134.36

295.4±139.9

213.65±76.92

236.84±91.76

0.03

0.04

Goker et al

2019

Turkey

Fovea

Parafovea

18.02 ± 6.15

53.04 ± 3.86

23.43 ± 6.80

53.53 ± 2.94

0.012

0.685

36.02 ± 6.75

56.30 ± 4.74

40.28 ± 8.63

56.25 ± 4.44

0.022

0.968

Superficial

Deep

0.578 ± 0.152

0.450 ± 0.121

0.007

Lopes et al

2020

Portugal

Fovea

Parafovea

8.1 ± 6.2

16.8 ± 6.7

12.3 ± 4.8

22.4 ± 6.7

<0.001

0.001

12.6 ± 6.0

34.0 ± 7.4

15.6 ± 4.6

36.7 ± 6.8

0.001

0.061

Superficial

Deep

Ozek et al

2019

Turkey

Fovea

Parafovea

20.70±7.96

43.44±10.21

20.43 ± 8.23

47.40 ± 14.2

0.902

0.220

21.01±17.68

45.82±23.49

21.77 ± 5.43

59.53 ± 18.54

0.852

0.026

Superficial

Deep

0.213±0.137

0.404±0.606

0.227±0.141

0.400±0.88

0.713

0.983

Abbreviations: HCQ – Hydroxychloroquine; FAZ – foveal avascular zone; (S/D)CP – superficial/deep capillary plexus; VD – vessel density;

HCQSuppFig1.pdf
Supplementary Figure 1 – Funnel plot of vessel density at A) superficial capillary plexus, B) deep capillary plexus and C) foveal avascular area for the comparison between patients at high versus low risk of HCQ-induced maculopathy.
HCQSuppFig2.pdf
Supplementary Figure 2 – Risk bias summary for included primary studies.
HCQSuppGraph1.pdf
Supplementary Graph 1 – Risk bias graph for included primary studies.
HCQSuppMethods1.pdf
Supplementary Methods 1 – Bias analysis questionnaires.
HCQSuppTable1.pdf
Supplementary Table 1 – Search query
HCQSuppTable2.pdf
Supplementary Table 2 - OCTA devices, parameters, and calculation method as defined by the authors.
HCQSuppTable3.pdf
Supplementary Table 3 – Bias analysis for cross-sectional studies

Download PDF

Journal Publication

published 08 Mar, 2023

Read the published version in Graefe's Archive for Clinical and Experimental Ophthalmology →

Version 1

posted

You are reading this latest preprint version

Application of Optical Coherence Tomography Angiography for hydroxychloroquine-induced microvascular changes: a systematic review and meta-analysis

Status:

Journal Publication

Version 1

Abstract

Figures

1. Introduction

2. Methods

3. Results

4. Discussion

5. Conclusion

Declarations

References

Tables

Supplementary Files

Status:

Journal Publication

Version 1