Multimodal imaging analysis and structure-function correlation in patients exposed to pentosan polysulfate sodium

Purpose

To study the anatomic and functional retinal changes in patients exposed to pentosan polysulfate (PPS) using multimodal imaging and mesopic microperimetry.

Design

A cross-sectional consecutive case series.

Methods

Patients exposed to PPS with and without maculopathy underwent color fundus photographs (CFP), optical coherence tomography (OCT), fundus autofluorescence (FAF; Spectralis OCT2, Heidelberg Engineering), macular swept-source OCT angiography (SS-OCTA; PLEX® Elite 9000, Carl Zeiss Meditec) using 3 × 3, 6 × 6, and 12 × 12-mm protocols, microperimetry (MAIA, CenterVue), and 10–2 Humphrey visual field. All patients were given a genitourinary pain index questionnaire. Vessel density (VD), vessels skeletonized density (VSD), in the whole retina (WR), and in the superficial (SCP) and deep (DCP) capillary plexus, FAZ metrics, choriocapillaris flow deficit percentage (CCFD%), and retinal sensitivity were measured.

Results

A total of 32 eyes of 16 patients (94 % female; age: 61.6 ± 13.0 years) were included, with 4 patients (25 %) having signs of maculopathy. The four patients with maculopathy were 10.6 years older than the non-maculopathy group (p = 0.21). Mean length of PPS exposure and mean cumulative PPS lifetime dose were significantly greater in patients with maculopathy than without (23.4 ± 2.4 years vs 9.2 ± 6.4 years, p < 0.005, and 2436.4 g ± 414.4 versus 812.1 g ± 677.7, p < 0.01, respectively). Mean VA was lower in the maculopathy group (69.8 ± 21.3 ETDRS letters, or 20/40 Snellen) compared to the non-maculopathy (82.5 ± 2.8 ETDRS letters, or 20/23 Snellen), p = 0.001. Ten eyes (31.25 %) showed abnormal CFP findings, including drusen (4 eyes, 40 %), all in patients without PPS-related maculopathy, and RPE irregularities (6 eyes, 60 %), in patients with and without PPS-related maculopathy. One eye (3.1 %) had a lamellar hole, and another (3.1 %) had a full-thickness macular hole, both in the maculopathy group. FAF showed hypoautofluorescent spots in 6 eyes (18.8 %) and hyperautofluorescent spots in 5 eyes (15.6 %). Structurally, the maculopathy group showed a significantly lower outer nuclear layer thickness, lower VD and VSD in the SCP, DCP, and WR (p < 0.001), a greater CCFD% in the 1-mm, 3 × 3-mm, and 6 × 6-mm circles, as well as in the inner and the outer rings (p < 0.05). Functionally, the maculopathy group showed a significantly decreased sensitivity on microperimetry (p = 0.012), lower mean deviation (p = 0.001) and higher pattern standard deviation (p = 0.027) on 10–2 Humphrey visual field. Length of exposure and total cumulative dose were negatively associated with VA and VD, and positively correlated with CCFD% in the 6 × 6 mm. Structure-function correlations were observed between VA, mean deviation, mesopic sensitivity with VD, VSD, ONL thickness, FAZ area, and CCFD%.

Conclusions

Patients with PPS-related maculopathy demonstrated a lower outer nuclear layer thickness, vessel density, VA, retinal sensitivity, visual field, as well as a greater choriocapillaris flow deficit compared to the non-maculopathy PPS exposed group. Multimodal imaging might provide biomarkers for early detection of macular toxicity in patients exposed to PPS and may suggest pathogenesis of PPS maculopathy is related to choriocapillaris injury. Most patients that discontinued PPS did not have worsening genitourinary symptoms suggesting a need to continually evaluate the clinical need.