Figure 3. Fluorescein angiography of the left eye at one minute and thirty seconds demonstrates leakage in the temporal macula as well as peripheral window defects.

Workup, Diagnosis and Treatment

Ancillary imaging, including fluorescein angiography (Figure 3) and OCT (Figure 4), was obtained. The former showed late leakage in the superior macula of the left eye. OCT demonstrated cystoid macular edema in both eyes. Goldmann visual fields showed mild, scalloped, concentric peripheral visual field loss, most prominent superiorly.

The work-up of this patient with a history of HIV, syphilis and sarcoidosis, presenting with bilateral insidious peripheral vision loss and nyctalopia due to wide-spread peripheral, pigmentary RPE atrophy, began by excluding infectious etiologies. Given the patient’s history of syphilis, an RPR and Venereal Disease Research Laboratory (VDRL) titer were drawn. (Fluorescent treponemal antibody-absorption [FTA-ABS] remains reactive from an initial infection, so it’s not useful in monitoring for recurrence.) The RPR titer was 1:4, decreased from 1:16 previously. A computed tomography scan of the chest showed hilar lymphadenopathy and interstitial lung disease consistent with known sarcoidosis.

The patient’s medication list was reviewed, and only ritonavir was associated with a retinopathy. Ritonavir toxicity is reported to have a characteristic pattern of perifoveal whitening and late bull’s-eye macular atrophy. No association with macular edema or peripheral atrophy has been made with this medication. Further review of his previous highly active antiretroviral therapy revealed that he had been treated with didanosine, the use of which has been associated with a gyrate-like peripheral retinopathy. Given the low suspicion for other diagnoses and the similarity of previous case reports of didanosine toxicity to this case, a presumed diagnosis of didanosine toxicity was made. However, didanosine has never been reported to cause macular edema, as was seen in this patient.

The patient was started on topical dorzolamide to treat the macular edema, to limit interactions with his other medications and to avoid steroids, given his history of CSCR. In addition to the dorzolamide for macular edema, he also received one week of oral ibuprofen that successfully treated his scleritis, which was the cause of his presentation to the Wills ER, but incidental to the retinal findings. 

He was followed closely for three years without any progression of the peripheral retinal findings. The macular edema fluctuated in severity, despite continuing topical
dorzolamide.

Figure 4. Optical coherence tomography of the left eye shows cystoid macular edema, as well as outer retinal atrophy temporally.

Discussion

Didanosine is a nucleoside reverse transcriptase inhibitor (NRTI) most well-known for its role as a component of HAART used in the management of HIV. It’s an adenosine analogue that terminates viral transcription when it’s incorporated into a growing nucleic acid chain. Introduced in 1991, the drug was the second HIV medication approved by the Food and Drug Administration, preceded only by zidovudine.¹ It became the first generic HIV medication in 2001. The most well-known adverse effect of didanosine is acute pancreatitis, which has resulted in a black-box warning. Other gastrointestinal symptoms can also occur, such as nausea, vomiting, diarrhea and abdominal pain. Peripheral neuropathy is reported in a quarter of patients, and case reports of optic neuritis have been described.¹

There are 22 case reports of didanosine-related ophthalmic toxicity, all of which include a peripheral retinal atrophy with pigmentary deposition.² The pigmentary changes often take on a gyrate-like pattern, as described in this case, and never involve the macula. For this reason, symptoms are limited to nyctalopia and photopsias, with central visual acuity remaining largely unaffected.² It’s believed that the process primarily affects rods, due to their high mitochondrial activity, as didanosine inhibits mitochondrial function in vitro.¹

Signs and symptoms in these patients are typically stable or slowly worsen. There are no cases where the retinal findings or symptoms resolved. The most important step in management is to diagnose didanosine as the offending medication and discontinue its use, although case reports have documented progression of retinal atrophy despite cessation of therapy.³

This raises the question of whether the retinopathy is due to didanosine alone or is associated with the entire class of NRTIs. There are no documented cases of this pattern of retinal toxicity in patients on HAART not taking didanosine, but it’s possible that the use of other medications that affect mitochondrial function may increase the risk of didanosine toxicity.³ In our patient, didanosine had already been discontinued at the time of diagnosis, but he didn’t know when it was stopped. Fortunately, the toxicity didn’t progress in our patient. 

The cystoid macular edema in this case remains something of a diagnostic mystery. The patient was taking ritonavir, which is known to cause a maculopathy. However, this has a characteristic pattern described as perifoveal whitening, telangiectasias, intraretinal crystals and RPE disruption with late retinal atrophy. Progression leads to a bull’s eye maculopathy due to retinal atrophy, but no reports document macular edema.⁴ Macular edema is not associated with any other HAART medications, including didanosine.⁵ As the patient had no other conditions associated with cystoid macular edema, it seems possible that this could be a novel presentation of didanosine toxicity.

This case illustrates the importance of a broad differential diagnosis in HIV-positive patients. The most important first step is to rule out an infectious process. Pathogens affecting the retina include syphilis, toxoplasmosis, cytomegalovirus, herpes simplex and the varicella-zoster virus. The Herpes viridae have characteristic presentations that aren’t consistent with the retinopathy or insidious time-course seen in this case. Toxoplasmosis can cause a bilateral necrotizing chorioretinitis in HIV patients, but the mid-peripheral atrophic appearance of the fundus in this patient didn’t appear compatible with toxoplasmosis scars. Syphilis was the most likely possible infectious process, especially given the patient’s history of syphilis treated multiple times with only intramuscular penicillin. Ocular manifestations of syphilis include anterior or intermediate uveitis, optic neuritis and posterior subretinal placoid lesions. Pertinent to this case, syphilis leading to a non-necrotizing bilateral process resulting in large areas of retinal atrophy followed by RPE hyperplasia has been described. Inflammatory findings are typically present, including retinal vasculitis or vitritis. Decreasing RPR titers, the gyrate-like appearance and lack of progression weren’t consistent with syphilis or other infectious processes.

In conclusion, this is a case of bilateral pigmentary retinopathy in an HIV-positive patient with a history of syphilis and sarcoidosis. After infectious, inflammatory and neoplastic processes were ruled out, a diagnosis of didanosine toxicity was presumed. An important point is that patients, especially those with HIV and autoimmune diseases, may have numerous simultaneous ocular processes. While Occam’s razor states that a single underlying diagnosis may explain numerous seemingly unrelated findings, this may not be the case in HIV patients. This case presented a management dilemma in balancing medication side effects with the necessity of steroid therapy and HAART. Co-management with infectious disease and other specialties is critical in proper management of these medically complex patients.  REVIEW

 

^{1. Kakuda TN. Pharmacology of nucleoside and nucleotide reverse transcriptase inhibitor-induced mitochondrial toxicity. Clin Ther 2000;22:685-708.}

^{2. Haug SJ, Wong RW, Day S, et al. Didanosine retinal toxicity. Retina 2016;36 Suppl 1:S159-S167.}

^{3. Gabrielian A, MacCumber MM, Kukuyev A, et al. Didanosine-associated retinal toxicity in adults infected with human immunodeficiency virus. JAMA Ophthalmol 2013;131:255-259.}

^{4. Pinto R, Vila-Franca M, Oliveira Afonso C, et al. Ritonavir and bull’s eye maculopathy: Case report. GMS Ophthalmol Cases 2013;3:Doc01.}

^{5. Whitcup SM, Butler KM, Caruso R, et al. Retinal toxicity in human immunodeficiency virus-infected children treated with 2’,3’-dideoxyinosine.  Am J Ophthalmol 1992;113:1-7.}