Ocular outcome and frequency of neurological manifestations in patients with acute posterior multifocal placoid pigment epitheliopathy (APMPPE)

Eighteen patients with APMPPE were identified in our database and were included in the study. Thirteen patients (72 %) participated in a follow-up exam; five did not agree to come or could not be contacted. In all patients, there was no doubt that the correct diagnosis was APMPPE. All 18 patients had bilateral involvement. Nine patients were female and nine were male. The mean age at time of diagnosis was 28.61 years (17–58 years). The mean follow-up time was 17.12 months (0–61.43 months).

Visual acuity in the worse eye on the date of presentation ranged between 1.54 and 0 log minimum angle of resolution (MAR) (mean, 0.34 logMAR, Table 1). On last examination, visual acuity was between 0.4 and 0 logMAR (mean, 0.09 logMAR). Taken together, visual acuity improved in 9 of 18 patients from a mean of 0.64 logMAR to a mean of 0.17 logMAR in the worse eye. In nine patients, it remained stable (mean, 0.03 logMAR). Near visual acuity was taken in nine patients (50 %) at follow-up exam and ranged from 0.52 logMAR to 0 logMAR in the worse eye (mean, 0.11 logMAR). All 9 patients (50 %) who were assessed by the Amsler test complained about metamorphopsias in the grid. Of the 13 patients who underwent a follow-up examination, all showed funduscopically small pigmented scars with no activity. The scars could be well visualized using FAF. The results of the fluorescein angiogram (17 patients; one patient refused the procedure) were all in agreement with the diagnosis of APMPPE. The angiograms that were taken in the acute phase all showed the typical staining pattern of early hypofluorescence and late hyperfluorescence. Angiograms that were taken later in the course of the disease showed inactive staining of scars. All 11 multifocal ERGs that were performed (ten at follow-up and one at a relatively acute phase) were pathological: mainly reduced sum responses and or reduced responses of the foveal and parafoveal regions. Fifteen out of 18 patients received a 30° static perimetry. These patients could be divided into three groups: group 1 (three patients) who only had perimetry at one of their first visits, group 2 (six patients) who had perimetry only at follow-up and group 3 (six patients) who had perimetry at one of their first visits as well as at follow-up. To be able to objectively compare the results of the perimetry, the mean defect (MD) values in decibels were analysed. In group 1, the mean MD was 1.1 dB (range, −0.6 to 4.1 dB). In group 2, the mean MD was 0.03 dB (range, −1.6 to 1.3 dB). The exact numbers are displayed in Table 1. In group 3, the mean MD was 4.14 dB (range, −1.2 to 10.8 dB) initially and improved to a mean of 1.77 dB (range, −1.8 to 7.6 dB), indicating that the visual fields improve after the acute phase of the disease. As can be seen in Table 1, all patients but one (patient 8) had an improvement of the visual fields at follow-up compared to the initial exam. One explanation why patient 8 did not improve is that he was our oldest patient and that his general health status at the day of the follow-up exam was poor. Especially in group 3, the MD shows a very broad range confirming that the disease can have varying extents. When looking at the visual fields of all patients that had a follow-up perimetry (i.e. groups 2 and 3), we felt that 6 out of 12 patients had remaining mostly small visual field defects. All but one patient had no more relapses of ocular disease at the end of the first treatment period. The one patient who relapsed had increasing visual field defects that improved after increasing the corticosteroid dosage.

Fifteen patients (83 %) were treated with systemic corticosteroids. The initial dosage ranged from 50 to 250 mg (mean, 106 mg) and was determined by the physician on duty who had the first contact with the patient. As far as it can be said retrospectively, the initial dosage was chosen independently of the presence of neurological symptoms. The treatment period ranged between 3 weeks and several months, whereas 47 % of the treated patients received treatment for approximately 1 month. The speed and amount of steroid taper were adjusted to the patient's individual visual and physical recovery (e.g. headaches resolved). The one relapsing patient received additional immunosuppressive medication for 38 weeks which then could be stopped without further relapse. The stroke patient received additional cyclophosphamide for CNS vasculitis.

Eleven patients (61 %) had neurologic symptoms of varying severity (Table 2). In most patients, the neurological symptoms started at the same time or close to the beginning of the ocular symptoms. However, the most severe neurological complication we witnessed—cerebral stroke with hemiparesis and aphasia—took place 2 months after the first symptoms (headaches and visual complaints). This 58-year-old patient (Table 2, patient 8) received 200 mg prednisone at presentation which was slowly tapered. He was on 80 mg prednisone per day when the stroke took place. Six days earlier, he had still complained about headaches. The MRI showed multiple areas of infarction consistent with vasculitis. Embolising cardiac disease, stenosis of cerebral vessels and paraneoplastic disease were ruled out as underlying causes of the stroke. A biopsy of the dura mater revealed lymphocytic infiltrates as a morphological sign of vasculitis.

The most common neurological symptom was headache occurring in nine (50 %) patients (Table 2). Three patients had sensory symptoms like hyp- or paraesthesias, one patient experienced psychosis and one patient complained about vertigo. Overall, 12 out of the 14 patients that had a brain MRI did not show signs of vasculitis. Four patients showed abnormalities in the MRI (Table 2, patients 8, 10, 13 and 16). The findings in patients 10 (gliotic changes) and 16 (pineal cyst) were considered insignificant. In patient 16, a CSF examination was performed due to abnormal TCD results, but CSF results were normal. Patient 8, the stroke patient, presented with bilateral contrast medium enhancement of optic nerve sheaths and nodular changes in the orbits, possibly granulomatous before he developed the stroke. CSF examination was normal at this point in time. For patient 13, who had unspecific bilateral signal intensities in the subcortical frontal area, possibly gliosis, a CSF examination was recommended but was declined by the patient.

In total, CSF examination was performed in six patients. Four had normal results. Two patients with headaches and normal MRI showed pleocytosis and/or oligoclonal bands. TCD sonography showed abnormalities in 6 of 14 tested patients. Interestingly, the MRI results, TCD and the CSF analysis did not necessarily correlate with each other. Electrophysiologic exams were not helpful in indicating CNS affection. More detailed results of neurological manifestations and technical abnormalities will be published separately.