(1)
Charlotte Eye Ear Nose & Throat Associates, Charlotte, NC, USA
Abstract
Overdosing is the most important factor contributing to 4-aminoquinoline retinopathy (4AQR). Proper interpretation of ancillary test results requires estimation of the pretest probability of 4AQR based on height, weight, daily dosage, duration of therapy, and the presence of renal and liver disease. An abnormal trend of an ancillary test over time carries greater weight than a single abnormal test. Shortening follow-up intervals and adding complementary ancillary tests are proper responses to a mild suspicion of 4AQR after screening. Reducing dosage, shortening follow-up, and adding complementary tests are proper responses to moderate suspicion of 4AQR without certainty. Cessation of the 4AQ is the proper response if the estimated probability of 4AQR is high. Obtaining the correct test is useless if it is misinterpreted; therefore, refining skills in interpreting tests is worthwhile. A checklist of risk factors can help to avoid overlooking information affecting the probability of retinopathy. The normal fellow eye is scrutinized for evidence of retinopathy in screening patients with unilateral preexisting maculopathy. In cases with bilateral preexisting maculopathy, more weight is placed on optimizing modifiable risk factors (e.g., safer daily dosing based on the lesser of ideal and actual body weight). Although rare, cases of 4AQR do occur in properly dosed patients, especially if the cumulative dose is high. One should not screen for 4AQR with the 24-2 or 30-2 visual field; it is better to use the 10-2 VF. In patients with glaucoma who take 4AQs, it is better to follow the glaucoma and screen for 4AQR with separate visual field protocols dedicated to the particular purpose rather than to use the 24-2 or 30-2 VF to investigate both problems.
Abbreviations
4AQR
4-Aminoquinoline retinopathy
4AQs
4-Aminoquinolines (chloroquine and hydroxychloroquine)
AAO
American Academy of Ophthalmology
ABW
Actual body weight
AG
Amsler grid
C
Chloroquine
HC
Hydroxychloroquine
IBW
Ideal body weight
RA
Rheumatoid arthritis
RPE
Retinal pigment epithelium
SLE
Systemic lupus erythematosus
The first nine chapters of this book were thematic. This chapter takes a topical approach, presenting clinical examples that illustrate the principles developed in the earlier chapters. In many cases, the details of the clinical scenario modify a straightforward extrapolation of a theoretical assertion from an earlier chapter. By reviewing many cases, the clinician can learn to assess the risk of retinopathy and detect it early in patients who take 4-aminoquinolines (4AQs). The format in the chapter will be case reports with ancillary test images followed by an analysis of the case, and when available, a description of the outcome and follow-up.
The overriding message of the cases to be presented is how often patients are overdosed, and how the problems entailed are potentially avoidable if the principle of dosing based on the lesser of ideal body weight (IBW) and actual body weight (ABW) is followed (see Chap. 7) [1]. When both hydroxchloroquine and chloroquine are under discussion, they will be termed 4-aminoquinolines and their retinopathies will be termed 4-aminoquinoline retinopathy (4AQR). 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.
10.1 A Case of Prolonged Toxic Dosing with Evidence of Premaculopathy
A 48-year-old woman with systemic lupus erythematosus (SLE) had been taking hydroxychloroquine for 20 years at 400 mg/day. She was 5 ft 3 in. tall and weighed 167 lb. She had no renal or liver disease, nor any preexisting macular abnormalities on funduscopy. She had annual 10-2 visual fields (10-2 VFs), which were normal (Fig. 10.1). When the American Academy of Ophthalmology (AAO) guidelines were revised in 2011 to add multifocal electroretinography (mf ERG) and spectral domain optical coherence tomography (SD-OCT), where available, these were also obtained. Three consecutive 10-2 mf ERGs showed progressive decrease in N1P1 amplitude for rings R 1 to R 3 in both eyes and progressive increase in the R 1/R 2 ratio (Fig. 10.1). By the third study, these variables were all in the abnormal range in both eyes except for the R 1/R 2 ratio for the right eye (Fig. 10.1). Nevertheless, the 10-2 VFs were normal for both eyes. The SD-OCT showed paracentral thinning temporally in the right eye, but no abnormality of the inner segment/outer segment junction (IS/OS junction) (Fig. 10.1). How would you manage this patient?
Fig. 10.1
Ancillary testing in a patient on toxic dosing of hydroxychloroquine for 20 years. (a) The initial multifocal electroretinogram (mf ERG) from 10/14/2011. There is some 60 cycle noise evident in some of the hexagonal recordings (red arrows). The amplitudes and R 1/R 2 ratio for both eyes are normal. (b) The second mf ERG from 10-19-2012 is technically better than that of (a). The N1P1 amplitudes are decreased in comparison to those in (a). (c) The third mf ERG from 10-18-2013 is of good quality and shows a further trend in decreasing N1P1 amplitudes compared to those of (a, b). (d) Plots of the N1P1 amplitudes and the R 1/R 2 ratios for the three mf ERGs shown in (a–c). The green horizontal bars identify the lower limits of normal for the ring amplitudes and the upper limit of normal for the R 1/R 2 ratio. There is a three-study trend toward increasing abnormality consistent with hydroxychloroquine toxicity. (e) The false color maps of macular thickness from spectral domain optical coherence tomography (SD-OCT) show parafoveal thinning bilaterally (red arrows), which is most severe temporally in the right eye. (f) SD-OCT line scans showing normal inner segment/outer segment junction morphology (yellow arrows) at a time when parafoveal thinning was present. (g) 10-2 visual fields from 2005 through 2013 are normal
In this case, the trend in mfERG indices carried more weight than any single abnormal study. The SD-OCT documentation of parafoveal thinning despite normal outer retinal morphology was also compelling for toxicity. The 10-2 VFs in this case were technically excellent and were less sensitive than the other tests. The patient was taking a toxic dose of hydroxychloroquine. The lesser of IBW and ABW in this case was 135 lb using the National Heart Lung and Blood Institute table [2], implying that she had been on 6.52 mg/kg/day for 20 years, receiving a cumulative dose of 2,920 g. Reasonable options in this case would include reduction of dosage or cessation of drug. After discussion with the patient and the rheumatologist, it was decided to reduce dosing to 300 mg/day (4.89 mg/kg/day) and to recheck her ancillary tests in 3 months. The suspicion of 4AQR was high with a pretest probability of retinopathy estimated to be 60 %. At the 3 month follow-up, any further evidence of retinopathy (e.g., further parafoveal thinning on SD-OCT or a new paracentral scotoma on 10-2 VF) would raise the posttest probability of retinopathy into a range (e.g., greater than 85 %) compatible with cessation of drug. This follow-up had not occurred by the time of publication of this book.
10.2 The Importance of the Clinical Estimation of Pretest Probability of 4-Aminoquinoline Retinopathy for Proper Interpretation of Ancillary Tests
A 50-year-old man underwent a baseline screening for hydroxychloroquine retinopathy 8 months after beginning the drug at 400 mg/day for rheumatoid arthritis. He was 6 ft tall and weighed 205 lb. He had normal renal and liver function. There was no preexisting maculopathy. Visual acuity was 20/20 in each eye. The ancillary tests chosen were the mfERG and the 10-2 VF. There were abnormalities of each test. In the case of the 10-2 VF, there were paracentral scotomas bilaterally (Fig. 10.2). The mfERG had low N1P1 amplitudes for ring R 1 in the right eye and rings R 1 and R 2 in the left eye, but the R 1/R 2 ratios were normal in both eyes. How would you manage this case?
Fig. 10.2
Ancillary test images obtained at baseline in a 50-year-old man taking hydroxychloroquine for rheumatoid arthritis. (a) The 10-2 visual fields obtained with a III, red target have paracentral scotomas (circled in red) but the reliability indices are not excellent with 14 % and 17 % fixation losses in the right and left eyes, respectively, and one suspects a learning curve effect. (b) Based on N1P1 amplitude criteria, these mf ERGs are abnormal, but are not abnormal based on R 1/R 2 ratio criteria
The patient is properly dosed based on IBW at 5.05 mg/kg/day (see Chap. 7), and has only been on a 4AQR for 8 months. The pretest probability of 4AQR is on the order of 0.01 % or less (see Chap. 8). With the performance characteristics of 10-2 VF and mfERG, the posttest probability of 4AQR is no greater than 1 %. The clinician properly would not consider stopping the hydroxychloroquine or reducing the dose. Further testing with SD-OCT would be worthwhile. A normal SD-OCT would be associated with a negative predictive value (NPV) of close to 100 %, and reassure the clinician that follow-up in 1 year would be reasonable.
This case makes the point that the clinical situation is crucial in the correct interpretation of ancillary testing. Ancillary tests do not diagnose 4AQ. Clinicians considering a collection of evidence do. All ancillary tests have false positives, including 10-2 VF and SD-OCT. Based on the data reported in Chap. 8, the clinician should expect that 7.5 % of 10-2 VFs and 13.1 % of mf ERGs will have false positive results.
10.3 Hydroxychloroquine Retinopathy Due to Long-Term Overdosing, Misinterpretation of 10-2 Visual Fields, and an Internist’s Unresponsiveness to a Recommendation to Stop Medication
A 79-year-old woman had been treated for rheumatoid arthritis for 15 years with hydroxychloroquine at 400 mg/day. She was 5 ft 5 in. tall and weighed 100 lb. She had no renal or liver disease, nor any preexisting maculopathy. She complained of blurred vision in 2012 and had best corrected visual acuity of 20/20 right eye, 20/25 left eye. She was screened yearly with 10-2 VF testing using the III, red target and was declared free of retinopathy by the screening ophthalmologist each year. The author, in reviewing the charts of patients taking hydroxychloroquine in the practice during 2012, noted the typical presence of annular scotomata in each eye (Fig. 10.3) and suggested to the screening ophthalmologist that discontinuation of the hydroxychloroquine be recommended, which was done. The prescribing internist reduced the dosing to 200 mg/day, but did not discontinue the medication. In 2013, the annular scotoma was worse and the visual acuity in the left eye had decreased to 20/30. How would you manage this case?
Fig. 10.3
10-2 visual fields (10-2 VFs) from a chronically overdosed patient on hydroxychloroquine. (a, b) Graytone displays of the left and right eyes, respectively. A suspicious annular scotoma is present on the first 10-2 VF from 2008 (red arrows), which progressively worsens for the next four fields. All of these 10-2 VFs were misinterpreted as normal. Red arrows show a shallow relative scotoma. The blue arrow shows a deeper relative scotoma. The orange arrow shows a broadening of a shallow relative scotoma. The green arrow shows an absolute scotoma. At 4-24-2012 the dosing was reduced to 200 mg/day but the annular scotoma worsened anyway over the next year. (c) Defect display of the left eye. This display is less sensitive than the graytone display. In 3-21-2008 the red circled points are within normal sensitivity ranges. By 9-26-2008 two of the same points have developed elevated thresholds (blue-circled area). By 3-30-2010 three of the four points have elevated thresholds and the two points already abnormal on 9-26-2008 study have even higher thresholds than on that date (orange-circled area). (d) Defect display of the right eye. This display is less sensitive than the graytone display. In 3-21-2008 the red-circled points are within normal sensitivity ranges. By 4-4-2011 one of the same points has developed an elevated threshold (blue-circled area). By 4-25-2013 three of the four points have elevated thresholds and the point already abnormal on the 4-4-2011 study has an even higher threshold than on that date (orange-circled area)
This case illustrates another overdosed patient. The weight to use in calculating adjusted daily dosing in this patient’s case is the ABW, not the IBW, because her ABW was less than her IBW. One should use the lesser of ABW and IBW (see Chap. 7) [1]. Using ABW, her daily dosing was 8.8 mg/kg/day. Her cumulative dosage was 2,190 g. The case also shows that obtaining a10-2 VF on a regular basis does not ensure detection of 4AQR if the result is misinterpreted. The diagnosis was delayed at least 3 years because of visual field misinterpretation. Note that the graytone display was more useful than the defect depth display for showing the scotoma when the 10-2 VF with the III, red target and FASTPAC display is used. Finally, the proper response was cessation of drug, not reduction of dosage. Reduction of dosage is a proper response when there is doubt about the diagnosis. In this case, a convincing series of progressively abnormal 10-2 VFs makes the diagnosis incontrovertible and the drug needed to be stopped. Ancillary testing with SD-OCT would have been useful to confirm the findings indicated by 10-2 VFs. If drug cessation had been stopped at 4-25-2013, it is probable that continued field loss and acuity loss would have occurred anyway, as the best corrected visual acuity had already dropped to 20/30 (see Chap. 6).
10.4 A Case of Hydroxychloroquine Retinopathy in an Overdosed, Regularly Monitored Patient in Whom the Fundus Changes Were Atypically Mild and in Which Progression of Damage Occurred Despite Cessation of Drug
A 68-year-old woman with SLE had been on hydroxychloroquine at a dosage of 400 mg/day for 8 years. Her height was 5 ft 3 in. and her weight was 135 lb. She had no renal or liver disease, nor any preexisting maculopathy. She was screened yearly by her optometrist, but the fields done from year to year were inconsistent with a mixture of 10-2 and 24-2 VFs. At the eighth year visual field scotomata were suspected but her fundi were considered to be normal. She was referred to an ophthalmologist who diagnosed retinopathy and recommended cessation of hydroxychloroquine. She was followed for an additional 18 months and despite cessation of drug had progression of retinopathy (Fig. 10.4). How would you analyze this case?
