(1)
Charlotte Eye Ear Nose & Throat Associates, Charlotte, NC, USA
Abstract
Lack of agreement in defining 4-aminoquinoline retinopathy has impeded progress in understanding and preventing the condition. There are stages of retinopathy, but unclear boundaries separate them. An international working group to standardize definitions would help lay the foundation for clearer communication and synthesis of knowledge.
Abbreviations
4AQ
4-Aminoquinoline (chloroquine or hydroxychloroquine)
4AQR
4-Aminoquinoline retinopathy
FA
Fluorescein angiography
HC
Hydroxychloroquine
mfERG
Multifocal electroretinogram
RPE
Retinal pigment epithelium
SAP
Standard automated perimetry
SD-OCT
Spectral domain optical coherence tomography
This chapter covers the definitions of retinopathy encountered in the literature on chloroquine and hydroxychloroquine (HC) retinopathy. The name 4-aminoquinoline (4AQ) will be used to refer to both chloroquine and hydroxychloroquine. The definitions of 4-aminoquinoline retinopathy (4AQR) apply to both drugs [1]. Commonly used abbreviations in this chapter are collected in “Abbreviations” for reference. Each term will be first used in its full form, along with its abbreviation.
Confusion arises from the number and variety of definitions for hydroxychloroquine and chloroquine retinopathy [2–5]. An example has to do with whether an abnormality on an ancillary test is the same as toxicity and whether toxicity is the same as retinopathy. It has been stated that an abnormality on multifocal electroretinography or other electrodiagnostic tests is different from toxicity [6, 7]. An abnormality, in this understanding, might represent a pharmacologic effect of the drug. However, because all toxic effects of drugs are pharmacologic effects (i.e., harmful ones), the distinction amounts to a tautology. The effects of hydroxychloroquine on the multifocal electroretinogram (mfERG) are not desirable. Therefore, they represent toxicity. The clinically important issue is not the semantic one, but rather reversibility [7]. Whether one calls an effect pharmacologic, toxic, or retinopathic, what matters is whether it goes away if the drug is stopped. Because we do not know at the time the mfERG is obtained whether it is reversible or not, it is difficult to see how such a distinction is useful.
Lack of agreement on the definition of various stages of chloroquine and hydroxychloroquine retinopathy makes it difficult to compare results across series [8]. For example, Henkind and colleagues could not compare their color vision results to those of Okun because of variability in case severity between the two series [8, 9]. Rates of retinopathy can vary more than tenfold depending on how inclusive or exclusive the definitions of retinopathy are (see Chap. 5) [10]. For example, using a definition of retinopathy based entirely on the mfERG, Lyons and Severn concluded that 37 of 131 eyes (28 %) of patients taking hydroxychloroquine had retinopathy [11]. Using a definition based on cessation of 4AQs after consideration of the totality of the evidence, including mfERG, Browning found that 2 of 183 patients (1.1 %) had retinopathy [12]. Even when clinicians agree on a definition, the variability in interpretation of findings upon which the definitions of retinopathy are based implies frequent disagreements over whether a particular case has retinopathy or not [5, 13, 14].
In general, there are three stages of 4AQR (see Chap. 6). The term premaculopathy implies that a functional change has occurred but no morphologic change (Fig. 4.1a, b). Reversibility is part of the definition. However, premaculopathy is a fuzzy concept and difficult to apply reproducibly. If one stipulates a morphologic criterion as part of the definition of retinopathy, then one has probably failed in the attempt to detect reversible effects. There is scant evidence that morphologic changes on spectral domain optical coherence tomography (SD-OCT), fundus photography, or fluorescein angiography (FA) resolve [15]. The advantage of morphologic criteria is that they are more reproducible than functional definitions. The price paid—irreversibility of the change in most cases—may be acceptable in setting the definition if progression can be averted when more advanced changes than purely functional ones have occurred.
Fig. 4.1
Fundus photographs and diagrams of retinal sensitivity to standardized automated perimetry corresponding to common definitions for the three stages of 4-aminoquinoline retinopathy. (a) Depiction of the normal retina. The macular appearance is normal and the hill of vision has the normal configuration with highest sensitivity at the fovea and gradually decreasing sensitivity with increasing eccentricity from the fovea. (b) Depiction of premaculopathy. The macular appearance is normal in this case, but it can also show mild mottling. There is a functional abnormality, with decreased retinal sensitivity corresponding to perifoveal locations (relative paracentral scotoma). The relative paracentral scotoma is reversible in some but not all cases. (c) Depiction of early maculopathy. The macular appearance is abnormal in this case with mottling of the retinal pigment epithelium, but it can also be normal. There is a more pronounced functional abnormality with a more pronounced decrease in perifoveal retinal sensitivity (deeper paracentral scotoma). The fundus pigmentary change and paracentral scotoma are irreversible, but have a low probability of progressing if drug is stopped at this stage. (d) Depiction of advanced retinopathy. The macula has a bull’s-eye abnormality that is irreversible. The foveal sensitivity is decreased (relative central scotoma) and the perifoveal retinal sensitivity is even less than in early retinopathy (deeper paracentral scotoma). Even with cessation of drug, at this stage there is a clinically important but poorly defined probability that progressive funduscopic and perimetric worsening will occur over time
Early retinopathy is more advanced than premaculopathy, but means different things to different investigators [16, 17]. Fundus changes can be present, but no bull’s-eye lesion (Fig. 4.1c) [3, 9, 18]. Many patients with definite early toxicity have normal fundi [19–22]. Other patients have no visual field abnormalities but have macular pigment stippling [23]. Some authors require that there be visual field scotomata, but do not allow fundus abnormalities [22]. Some authors describe patients in this group as having color vision problems but no visual field defects [17]. Patients often do not have symptoms at this stage, but many do and some authors require symptoms as part of their definition of early retinopathy [18, 24, 25]. Although some have said that early fundus changes are reversible in up to 50 % of such cases [26], skepticism is appropriate as there has not been photographic documentation to analyze [27]. Patients with early retinopathy may have mfERG abnormalities with or without visual field abnormalities [17].
The classically recognized bull’s-eye lesion of 4AQR signifies advanced retinopathy that never reverts to normal (Fig. 4.1d) [11, 28–32]. The presence of a bull’s-eye lesion is accompanied by an annular scotoma and usually by symptoms. Patients at this stage will have more severe mfERG abnormalities [17]. Although a bull’s-eye maculopathy is the hallmark of advanced 4AQR, it is not specific [33].
Table 4.1 compares definitions of chloroquine and hydroxychloroquine retinopathy across published studies. It is apparent that these papers are not addressing retinopathy at the same stage, and thus can only be compared with reservations. Some only consider cases with advanced retinopathy [15, 34]. Others are addressing patients with early retinopathy [35]. Some consider retinopathy to include all stages of toxicity, from functional, and reversible stages, to full-fledged, funduscopic changes with no hope of visual function recovery.
Table 4.1
Definitions of 4-aminoquinoline retinopathy
Year | Study/stage of retinopathy | Perimetric characteristic | mfERG | SD-OCT | Funduscopy | FA | Other criteria | Number of elements |
|---|---|---|---|---|---|---|---|---|
1968 | Carr [36]/NS | None | None | None | Pigmentary abnormality | None | None | 1 |
1971 | Marks [23]/NS | None | None | None | Pigmentary abnormality | None | None | 1 |
1985 | Bilateral reproducible visual field defects by two different techniques (e.g., Amsler grid and 10-2 VF with white II targets) | None | None | None | None | None | 1 | |
1985 | Finbloom [39]/NS | Elevated cone threshold with Goldman-Weeks adaptometer and a <1° red light | None | None | Pigmentary abnormality | None | Color vision abnormality, total dose >100 g | 3 |
1987 | Johnson [40]/NS | On 10-2 VF testing with red test object, two or more adjacent points of 5 dB loss or a single point of 10 dB loss | None | None | None | None | None | 1 |
1992 | Bernstein [41]/NS | Persistent central or pericentral scotoma to suprathreshold white stimuli | None | None | Bull’s-eye lesion qualifies even if no VF done | None | >9 months if daily dose >400 mg/d | 3 |
1993 | Easterbrook [2]/NS | Bilateral Amsler grid defects confirmed on perimetry | None | None | None | None | None | 1 |
2000 | Neubauer [16]/mild | No reproducible defects on Amsler grid or 10-2 visual field testing | None | None | Pigmentary abnormality | None | None | 1 |
2000 | Neubauer [16]/advanced | Reproducible, bilateral visual field defects | None | None | Typical bull’s-eye macular change and | None | None | 2 |
2003 | Mavrikakis [42]/NS | Two or more adjacent defects of 0.8–1.2 log units or one defect of 1.4–1.8 log units in an area of a previous scotoma | None | None | None | None | None | 1 |
2004 | Maturi [35]/NS | None | Any abnormality of mfERG—by hexagon or by ring, either amplitude or implicit time | None | None | None | None | 1 |
2004 | Araiza-Casillas [34]/NS | Scotoma at 10° | None | None | Pigmentary abnormality | Window defects | None | 3 |
2006 | Elder [43]/NS | Paracentral VF defects on 2 10-2 VFs separated by 2 months | None | None | None | None | None | 1 |
2007 | Shinjo [25]/NS | None | None | None | Pigmentary abnormality | None | Ocular symptoms | 2 |
2007 | Lyons [44]/NS | None | R1/R2 > 2.6 | None | None | None
 Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
Get Clinical Tree app for offline access
|