Retinal toxicity related to hydroxychloroquine in patients with systemic lupus erythematosus and rheumatoid arthritis

Transcription

1 Doc Ophthalmol (2017) 135: DOI /s ORIGINAL RESEARCH ARTICLE Retinal toxicity related to hydroxychloroquine in patients with systemic lupus erythematosus and rheumatoid arthritis Hande Husniye Telek. Nilufer Yesilirmak. Gulten Sungur. Yaprak Ozdemir. Nesibe Karahan Yesil. Firdevs Ornek Received: 7 February 2017 / Accepted: 15 August 2017 / Published online: 29 August 2017 Ó Springer-Verlag GmbH Germany 2017 Abstract Purpose To compare the retinal toxicity due to hydroxychloroquine (HCQ) use in patients with systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA) using multifocal electroretinography (mferg), fundus autofluorescence (FAF) and optical coherence tomography (OCT). Methods Patients who were using HCQ due to SLE and RA, and healthy subjects evaluated in this study. Central foveal thickness (CFT), inner outer segment (IS OS) junction irregularity, retinal nerve fiber layer thickness, mferg and FAF measurements were performed to evaluate retinal toxicity. Results Study included 35 eyes of 35 SLE patients, 40 eyes of 40 RA patients and 20 eyes of 20 healthy subjects. In SLE group, retinal abnormality was found in three eyes with mferg, in one eye with FAF and in four eyes with OCT. In RA group, retinal abnormality was found in 10 eyes with mferg, in five eyes with H. H. Telek N. Yesilirmak (&) G. Sungur F. Ornek Department of Ophthalmology, Ankara Training and Research Hospital, Sukriye, Ulucanlar St. No. 89, Altindag, Ankara, Turkey dryesilirmak@gmail.com Y. Ozdemir Department of Statistics, Gazi University Faculty of Sciences, Ankara, Turkey N. K. Yesil Department of Rheumatology, Ankara Training and Research Hospital, Ankara, Turkey FAF and in nine eyes with OCT. A statistically significant difference was found with respect to mferg between eyes with abnormal responses and without abnormal responses and eyes with abnormal responses and controls (p \ 0.05). A statistically significant difference was found with respect to CFT between eyes with IS OS junction irregularities and without IS OS junction irregularities and eyes with/without IS OS junction irregularities and controls (p \ 0.05). Conclusions The use of HCQ seems to cause retinal toxicity more often in RA patients compared to SLE patients. For the early detection of retinal changes, OCT and mferg can be used as screening tools due to their higher sensitivity rates compared to other tests. Keywords Systemic lupus erythematosus Rheumatoid arthritis Hydroxychloroquine toxicity Optical coherence tomography Multifocal electroretinography Fundus autofluorescence Introduction Antimalarial drugs have been used for the treatment of rheumatic diseases from the second half of the twentieth century [1]. Particularly, chloroquine (CQ) and hydroxychloroquine (HCQ) sulfate have been used in systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA) as main treatment agents [2]. Although these agents are generally safe, some

2 188 Doc Ophthalmol (2017) 135: serious adverse events including cardiomyopathy, depression and irreversible retinal toxicity have been reported [3]. CO and HCQ bind to melanin pigments in the retinal pigment epithelium (RPE). This binding may serve to concentrate the agents in the cell and contribute to their long-term effects. Although CO and HCQ have similar pharmacology and equipotentiality, it has been shown that HCQ has a lower tissue accumulation and a lower melanin affinity compared to CO, which is considered as the reason of lower degree of retinal damage seen with HCQ than seen with CO [4, 5]. The classic pattern of retinal toxicity of HCQ is RPE depigmentation with foveal sparing, known as bull s-eye maculopathy. Although visual acuity in these patients seems intact, patients complain from paracentral scotomas associated with reading difficulties. Besides, central vision loss, alteration in the visual field and reduced color perception can be seen as retinopathy symptoms. Prior to the treatment, complete ophthalmological examination including best-corrected visual acuity and dilated fundus examination is necessary. It is important to diagnose the disease before the recognition of RPE changes on slit-lamp examination. In recent years, visual field testing, optical coherence tomography (OCT), fundus autofluorescence (FAF) and multifocal electroretinography (mferg) are used for this purpose [6 8]. The retinal layers can be observed in more detail by new high-resolution OCT technology, and it has been shown that parafoveal inner segment the outer segment (IS OS) junction irregularity may be considered as the early sign of retinal toxicity [7 10]. To assess subtle RPE changes, FAF imaging can be used. A reduction of autofluorescence may show the early involvement of RPE, and an increased autofluorescence may show the involvement of photoreceptor cells (with the accumulation of outer segment debris) [7 12]. Multifocal electroretinography can be used to demonstrate the localized paracentral depression by generating local ERG responses in all retinal areas that can help for the early recognition of toxicity [6, 8, 11 14]. In this study, we aimed to evaluate sensitivity and specificity of mferg, FAF and OCT [IS OS junction irregularity, central foveal thickness (CFT) and retinal nerve fiber layer (RNFL) thickness] measurements in the detection of HCQ-related renal toxicity in SLE and RA patients. Patients and methods Patient population This prospective and comparative study included patients who were using HCQ for the treatment of SLE and RA and healthy subjects who were not using any drugs at the Ankara Training and Research Hospital, Ankara, Turkey, between June 2015 and March Prior to the study, an institutional review board approval was obtained and adhered to the tenets of the Declaration of Helsinki. Informed consent was obtained from all individual participants included in the study. Exclusion criteria Patients who were having a diagnosis of SLE or RA for \2 years, refractive error greater than ±3 diopters, history of prior ocular surgery, serious coexistent ocular disease, systemic or ocular disease that may lead to retinopathy, nerve fiber damage and altered vision field, active ocular inflammation, corneal opacities, pseudo-exfoliation syndrome, glaucoma, cataract and chronic use of topical or systemic steroids or nonsteroidal anti-inflammatory drugs were excluded. Ocular examinations All subjects underwent a complete ophthalmological examination. Refractive errors were measured by an autorefractometer. The best-corrected visual acuity of each eye was determined by Snellen chart, separately. The anterior segment was evaluated by slit-lamp examination. Intraocular pressures (IOP) were measured by applanation method, and indirect ophthalmoscopy was performed with a noncontact fundus lens (Super Field Volk Lens) after maintaining mydriasis. The 10-2 visual field testing was performed by Humphrey visual field test. Foveal images were taken by Spectralis OCT (Heidelberg, Germany); IS OS junction irregularity, CFT and RNFL analyses were also performed with the same device. Fast RNFL thickness protocol was used for this measurement. In accordance with the protocol, 360 circular OCT scans with a diameter of 3.4 mm centered on the optic disk were obtained. Temporal, superior, nasal and inferior RNFL results were recorded separately. FAF imaging

3 Doc Ophthalmol (2017) 135: was also performed by Spectralis OCT device. mferg was performed with Roland Consult Electrophysiological Diagnostic Systems device. Prior to examination, active electrodes were placed on the nasal side of eyes, reference electrodes were placed on the temporal side, and a ground electrode was placed in the middle of the forehead. Patients were stimulated while looking at the screen 20 cm away. This procedure was performed in eight cycles, each lasting 45 s. Statistical analysis SPSS 20.0 software package was used to calculate means and standard deviations for all variables. Two independent t tests were conducted to compare the groups. Receiver-operating characteristic (ROC) curves were used to determine the predictive accuracy of the abnormalities by the area under the curve. Differences were considered statistically significant when p values were\0.05. Results Thirty-five eyes of 35 female SLE patients with a mean age of ± 4.33 years (SLE group), 40 eyes of 40 female RA patients with a mean age of ± 3.24 years (RA group) and 20 eyes of 20 healthy female subjects with a mean age of ± 4.13 years (control group) were included in this study. There was no statistically significant difference in mean age and gender between groups (p [ 0.05). In SLE and RA groups, all patients were receiving the safe dose of HCQ, 6.5 mg/kg/day. The mean HCQ exposure time in SLE patients was 6.9 ± 5.1 months (3 40 months), and it was 7.6 ± 6.0 months (4 42 months) in RA patients. There was no abnormality on the visual field and slit-lamp fundus examinations of both groups. In SLE group, mferg showed an abnormality (Fig. 1) in three eyes (4.29% ± 0.20), FAF results revealed pathology (Fig. 2) in one eye (1.43% ± 0.12), and IS OS junction irregularity (Fig. 3) was found in four eyes (5.71% ± 0.23). In RA group, mferg showed an abnormality in 10 eyes (12.5% ± 0.33), FAF results revealed pathology in five eyes (6.25% ± 0.24), and IS OS junction irregularity was found in nine eyes (11.30% ± 0.32). There were no OCT, FAF, mferg abnormalities in the control group. Table 1 shows the mferg responses following a distribution in rings in all groups. In both SLE and RA groups, statistically significant differences were found between eyes with abnormal responses and without abnormal responses and eyes with abnormal responses and controls (p \ 0.05). ROC analysis results were in the good diagnostic range (good sensitivity and specificity) in determining the predictive accuracy of the abnormalities by the area under the curve (p \ 0.01) (Figs. 4, 5). Table 2 shows the CFT and RNLF results of all groups. In both SLE and RA groups, statistically significant CFT differences were found between eyes with IS OS junction irregularities and without IS OS junction irregularities and eyes with/without IS OS junction irregularities and controls (p \ 0.05). There was no RNLF difference between groups (p [ 0.05). Discussion Retinal toxicity is a potentially severe side effect of HCQ use, and risk factors for its development include high daily drug dose, long duration of treatment, high level of body fat, concomitant renal or liver disease and age older than 60 years [1]. The incidence of retinal toxicity ranged from 0 to 4%. Mavrikakis et al. [15] followed 526 patients who were using HCQ for 15 years, and they found the retinal toxicity incidence as 0.38%. In a more recent study of Wolke and Marmor, the risk of toxicity has been found to be associated with the duration of HCQ use. It was shown that the toxicity incidence was very low within the first 5 years of use and could increase rapidly after 5 years to 1% [16]. However, some studies found the toxicity incidence higher (ranged from 20 to 60%) even \5 years of CQ/HCQ usage when using mferg as a screening tool and they explained this as a result of the on- and off-bipolar cellular contribution to the mferg response [6, 12, 13, 17, 18]. In our study, we found the incidence rate as 5.71% with OCT, 4.29% with mferg, 1.43% with FAF in SLE group, whereas the incidence rate was 11.3% with OCT, 12.5% with mferg, 6.25% with FAF in RA group. These results show that diagnosis can depend on the imaging technique when screening HCQ-related retinal toxicity.

4 190 Doc Ophthalmol (2017) 135: Fig. 1 Wave forms and 3-D density plots of 61 local responses, exploring rings 1 5 (from zero to 25 central), demonstrate mferg responses in all groups. a Reduced mferg responses of SLE patient, b reduced mferg responses of RA patient, c normal mferg responses of healthy control Periodic screening is recommended for patients within the first 6 months of HCQ treatment and yearly afterward. Amsler grid testing, color vision testing and automated visual fields (AVFs) have been using as most common screening tools to detect the early retinal toxicity. However, these tests are subjective and cannot objectively document visual function. For instance, AVFs are used in the early stages of the toxicity; however, the initial scotomas may be subtle and may not be detected. Initial scotomas are often noticed in the superior field within 10 from the fixation. Over time, these scotomas enlarge and reduce the visual acuity. Even after the development of central scotoma, slit-lamp examination may remain normal [19]. The other hindrances of this test include the experience of the user and the patient s cooperation [20]. Recently, the American Academy of Ophthalmology recommended a shift toward more objective testing modalities such as fundus autofluorescence (FAF), SD-OCT and mferg for the screening of CQ/HCQ retinopathy in addition to previous subjective screening methods [21]. With the introduction of FAF, OCT and mferg for detecting retinal toxicity, patients started to get early diagnosis of toxicity. FAF shows areas of hyper- and hypoautofluorescence corresponding to photoreceptor/rpe cellular stress and photoreceptor/rpe cell loss, respectively [7]. With OCT, early toxicity findings may be related to the localized thinning in the parafoveal region and loss of the IS OS line. Recent studies have shown that SD-OCT (highresolution OCT) displays a localized thinning of the retinal layers and confirms the toxicity in early stages, before the visual field testing [9, 22]. Also it has been reported that early stage of HCQ toxicity causes localized defects of perifoveal IS OS junction, also

5 Doc Ophthalmol (2017) 135: Fig. 2 FAF image demonstrates mottled hypofluorescence in a circular parafoveal distribution known as flying saucer sign, and it makes the diagnosis even before the RPE changes on biomicroscopy [23]. Therefore, studies concluded that SD- OCT is a sensitive examination tool, detecting the disruption of IS OS junction even before the RPE changes [10]. On the other hand, there are some studies indicating that mferg is the key test to show toxicity in the early stages [6, 12, 13, 17, 24]. It may display a significant localized depression at the onset of the renal toxicity by detecting functional electrophysiological changes in bipolar cells, photoreceptors and to some extent inner retinal cells [13]. Also AAO recommends mferg as a first choice test in patients with high risk of retinal toxicity and in patients who develop visual complaints [21]. According to our results, most of the abnormalities related to HCO toxicity were found when using mferg and OCT (IS OS junction irregularities and CFT decreases) and some of them were found when using FAF in both SLE and RA groups. With respect to mferg (abnormal responses) and OCT (decrease in CFT) results, there was a significant difference between affected eyes, unaffected eyes and controls. However, with respect to RNLF and FAF results, there was no difference found between groups. Therefore, we thought that mferg and OCT are more sensitive tests than FAF and other tests in the early detection of HCQ-related retinal toxicity in both RA and SLE patients. On the other hand, our results also revealed that HCQ use caused retinal toxicity more frequently in RA patients compared to SLE patients. This result may relate to the mean HCQ exposure time, which is 7 months longer in RA patients. In the literature, no study investigated whether such drugs cause different toxic effects between differently diagnosed patients. Felix et al. [25] demonstrated that minor alleles of common genetic variants in ABCA4 significantly reduce susceptibility to develop toxic maculopathy under CQ or HCQ treatment. This study and our Fig. 3 OCT image demonstrates IS OS junction irregularity

6 192 Doc Ophthalmol (2017) 135: Table 1 mferg responses following a distribution in rings that were numbered 1 5 Rings Eyes with abnormal responses in SLE group (n = 3) Eyes with normal responses in SLE group (n = 67) Eyes with abnormal responses in RA group (n = 10) Eyes with normal responses in RA group (n = 70) Eyes in control group (n = 20) R1 (Mean ± SD) ± 20.98*.34 ± ± 25.57* ± ± R2 (Mean ± SD) ± 9.35* ± ± 10.13* ± ± R3 (Mean ± SD) ± 4.85* ± ± 6.99* ± ± 6.27 R4 (Mean ± SD) 9.95 ± 5.08* ± ± 5.33* ± ± 4.89 R5 (Mean ± SD) 6.78 ± 4.34* ± ± 3.85* ± ± 3.89 SD standard deviation, n number * Statistical difference with respect to eyes with normal responses and controls at p \ 0.05 Fig. 4 Receiver operating characteristic (ROC) curves for mferg responses in SLE patients. AUC Area under the ROC curve, R1 Ring 1, R2 Ring 2, R3 Ring 3, R4 Ring 4, R5 Ring 5 results suggest that genetic, environmental and many other factors that we do not know yet may be effective in the development of retinal toxicity. Therefore, we believe that further analysis and research are needed to explain why retinal toxicity is more prevalent in patients with RA. In conclusion, mferg and OCT are more sensitive tests for the early detection of HCQ-related retinal toxicity compared to other tests. Using either mferg or OCT, or combination of them may be recommended at least once a year in order to detect retinal changes earlier.

7 Doc Ophthalmol (2017) 135: Fig. 5 Receiver operating characteristic (ROC) curves for mferg responses in RA patients. AUC Area under the ROC curve, R1 Ring 1, R2 Ring 2, R3 Ring 3, R4 Ring 4, R5 Ring 5 Table 2 Optical coherence tomography parameters between groups Eyes with IS OS junction irregularities in SLE group (n = 4) Eyes without IS OS junction irregularities in SLE group (n = 31) Eyes with IS OS junction irregularities in RA group (n = 9) Eyes without IS OS junction irregularities in RA group (n = 31) Eyes in control group (n = 20) CFT (lm) mean ± SD ± 8.6** ± 16.8* ± 6.6** ± 16.3* ± 20.1 RNLF (lm) mean ± SD Superior ± ± ± ± ± 14.0 Nasal 69.8 ± ± ± ± ± 13.5 Inferior ± ± ± ± ± 14.9 Temporal 67.3 ± ± ± ± ± 14.2 SD standard deviation, CFT central foveal thickness, RNLF retinal nerve fiber layer, n number * Statistical difference with respect to controls at p \ 0.05 ** Statistical difference with respect to eyes without IS OS junction irregularities and controls at p \ Compliance with ethical standards Conflict of interest All authors certify that they have no affiliations with or involvement in any organization or entity with any financial interest (such as honoraria; educational grants; participation in speakers bureaus; membership, employment, consultancies, stock ownership or other equity interest; and expert testimony or patent-licensing arrangements) or nonfinancial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) in the subject matter or materials discussed in this manuscript.

8 194 Doc Ophthalmol (2017) 135: Statement of human rights All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Statement on the welfare of animals This article does not contain any studies with animals performed by any of the authors. Informed consent Informed consent was obtained from all individual participants included in the study. References 1. Payne JF, Hubbard GB, Aaberg TM, Yan J (2011) Clinical characteristics of hydroxychloroquine retinopathy. Br J Ophthalmol 95: Geamanu A, Popa-Cherecheanu A, Marinescu B, Geamanu C, Voinea L (2014) Retinal toxicity associated with chronic exposure to hydroxychloroquine and its ocular screening. Review. J Med Life 7: Grassmann F, Bergholz R, Mändl J, Jägle H, Ruether K, Weber BH (2015) Common synonymous variants in ABCA4 are protective for chloroquine induced maculopathy (toxic maculopathy). BMC Ophthalmol 15:18 4. Farrell DF (2012) Retinal toxicity to antimalarial drugs: chloroquine and hydroxychloroquine: a neurophysiologic study. Clin Ophthalmol 6: Schroeder RL, Gerber JP (2014) Chloroquine and hydroxychloroquine binding to melanin: some possible consequences for pathologies. Toxicol Rep 1: Lai TY, Chan W-M, Li H, Lai RY, Lam DS (2005) Multifocal electroretinographic changes in patients receiving hydroxychloroquine therapy. Am J Ophthalmol 140: Kellner S, Weinitz S, Kellner U (2009) Spectral domain optical coherence tomography detects early stages of chloroquine retinopathy similar to multifocal electroretinography, fundus autofluorescence and near-infrared autofluorescence. Br J Ophthalmol 93: Kellner U, Renner AB, Tillack H (2006) Fundus autofluorescence and mferg for early detection of retinal alterations in patients using chloroquine/hydroxychloroquine. Investig Ophthalmol Vis Sci 47: Pasadhika S, Fishman G (2010) Effects of chronic exposure to hydroxychloroquine or chloroquine on inner retinal structures. Eye 24: Rodriguez-Padilla JA, Hedges TR, Monson B et al (2007) High-speed ultra-high-resolution optical coherence tomography findings in hydroxychloroquine retinopathy. Arch Ophthalmol 125: Lai TY, Ngai JW, Chan WM, Lam DS (2006) Visual field and multifocal electroretinography and their correlations in patients on hydroxychloroquine therapy. Doc Ophthalmol 112: Maturi RK, Yu M, Weleber RG (2004) Multifocal electroretinographic evaluation of long-term hydroxychloroquine users. Arch Ophthalmol 122: Lyons JS, Severns ML (2007) Detection of early hydroxychloroquine retinal toxicity enhanced by ring ratio analysis of multifocal electroretinography. Am J Ophthalmol 143: Lyons JS, Severns ML (2009) Using multifocal ERG ring ratios to detect and follow Plaquenil retinal toxicity: a review. Doc Ophthalmol 118: Mavrikakis I, Sfikakis PP, Mavrikakis E et al (2003) The incidence of irreversible retinal toxicity in patients treated with hydroxychloroquine: a reappraisal. Ophthalmology 110: Wolke F, Marmor MF (2010) Rates and predictors of hydroxychloroquine retinal toxicity in patients with rheumatoid arthritis and systemic lupus erythematosus. Arthr Care Res 62: Chang WH, Katz BJ, Warner JE, Vitale AT, Creel D, Digre KB (2008) A novel method for screening the multifocal electroretonogram in patients using hydroxychloroquine. Retina 28: Nebbioso M, Livani ML, Steigerwalt RD, Panetta V, Rispoli E (2011) Retina in rheumatic diseases: standard full field and multifocal electroretinography in hydroxychloroquine retinal dysfunction. Clin Exp Optom 94: Hart WM Jr, Burde RM, Johnston GP, Drews RC (1984) Static perimetry in chloroquine retinopathy. Perifoveal patterns of visual field depression. Arch Ophthalmol 102: Marmor MF, Chien FY, Johnson MW (2013) Value of red targets and pattern deviation plots in visual field screening for hydroxychloroquine retinopathy. JAMA Ophthalmol 131: Marmor MF, Kellner U, Lai TY, Lyons JS, Mieler WF (2011) Revised recommendations on screening for chloroquine and hydroxychloroquine retinopathy. Ophthalmology 118: Stepien KE, Han DP, Schell J, Godara P, Rha J, Carroll J (2009) Spectral-domain optical coherence tomography and adaptive optics may detect hydroxychloroquine retinal toxicity before symptomatic vision loss. Trans Am Ophthalmol Soc 107: Chen E, Brown DM, Benz MS et al (2010) Spectral domain optical coherence tomography as an effective screening test for hydroxychloroquine retinopathy (the flying saucer sign). Clin Ophthalmol 4: So SC, Hedges TR, Schuman JS, Quireza ML (2003) Evaluation of hydroxychloroquine retinopathy with multifocal electroretinography. Ophthalmic Surg Lasers Imaging 34: Grassmann Felix, Bergholz Richard, Mändl Julia, Jägle Herbert, Ruether Klaus, Bernhard HF, Weber B (2015) Common synonymous variants in ABCA4 are protective for chloroquine induced maculopathy (toxic maculopathy). BMC Ophthalmol 15:1