Optical coherence tomography-based misdiagnosis and morphological distinction in pachychoroid neovasculopathy vs. polypoidal choroidal vasculopathy

Jakob Siedlecki, Julian Klaas, Leonie Keidel, Ben Asani, Johannes Schiefelbein, Dominik Knebel, Nikolaus Luft, Siegfried Priglinger, Benedikt SchwormEye 2023 · DOI

Abstract

A relevant percentage of eyes diagnosed with PNV might instead suffer from PAT1/PCV. The detection of a maximum PED height ("peaking PED") exceeding approximately 150 µm, SHRM, SRRLS, and sub-RPE fluid might greatly aid in the production of a more accurate diagnosis.

Introduction

The "pachychoroid spectrum" describes a group of eye diseases characterized by a pathologically thickened choroid. Within this spectrum, Pachychoroid Neovasculopathy (PNV) and Polypoidal Choroidal Vasculopathy (PCV, specifically the PAT1 subtype) often present with overlapping clinical features, making them difficult to distinguish using standard imaging. Accurate differentiation is critical because PCV is generally more aggressive, carrying a higher risk of sudden vision loss due to hemorrhage, and often requires different management strategies, such as photodynamic therapy (PDT) in addition to anti-VEGF injections.

While Indocyanine Green Angiography (ICGA) remains the diagnostic gold standard for identifying the aneurysmal lesions (polyps) characteristic of PCV, its invasive nature and limited availability necessitate the search for reliable, non-invasive biomarkers. This study investigates the use of Spectral-Domain Optical Coherence Tomography (SD-OCT) to identify morphological features that can accurately distinguish PCV from PNV.

Multimodal imaging comparison of PNV (left) and PAT1/PCV (right) across four patients, highlighting differences in PED morphology and imaging signatures.

Methods

Researchers at the Ludwig-Maximilians University Munich conducted a retrospective cohort study re-evaluating 49 eyes from 44 patients initially diagnosed with PNV between 2017 and 2021. The diagnostic re-evaluation utilized multimodal imaging, including SD-OCT, Fluorescein Angiography (FA), and ICGA.

Key morphological parameters were analyzed, specifically focusing on pigment epithelium detachment (PED) characteristics such as maximum height and horizontal diameter. The study also looked for qualitative biomarkers, including sub-retinal hyperreflective material (SHRM), sub-retinal ring-like structures (SRRLS), and sub-RPE fluid. Statistical modeling via Receiver Operating Characteristic (ROC) analysis was employed to determine optimal numerical cutoffs for diagnostic differentiation.

Results

Upon re-evaluation with multimodal imaging, 7 out of 49 eyes (14.3%) originally diagnosed with PNV were reclassified as PAT1/PCV. The study found that while subfoveal choroidal thickness (SFCT) and PED horizontal diameter were comparable between the two groups, PED height was a significant differentiator.

The PAT1/PCV group exhibited significantly higher PEDs (199 ± 31 µm) compared to the PNV group (82 ± 46 µm). ROC analysis identified a PED height of 158 µm as the optimal threshold for a "peaking PED." This biomarker achieved a sensitivity of 1.0 and a specificity of 0.95, significantly outperforming horizontal diameter measurements.

ROC analysis comparing the diagnostic accuracy of PED diameter versus PED peak height, demonstrating the superior performance of peak height as a biomarker.

Furthermore, qualitative features were highly indicative of PCV. Sub-retinal ring-like structures (SRRLS) were present in 71.4% of PCV cases but were entirely absent in the PNV group. SHRM and sub-RPE fluid were also significantly more frequent in the PCV cohort.

Discussion

The 14.3% misdiagnosis rate observed in this study underscores the clinical challenge of distinguishing PNV from PCV without invasive imaging. The identification of the "peaking PED" (height > 158 µm) provides clinicians with a highly accurate, non-invasive tool for screening.

The presence of sharp, high elevations in PCV contrasts with the flatter, irregular "double layer sign" typically seen in PNV. While this study is limited by its retrospective nature and small sample size—particularly within the PCV subgroup—it highlights the potential for OCT-based biomarkers to refine diagnostic accuracy. These findings are specific to the pachychoroid phenotype and suggest that early identification of PCV through OCT could lead to more tailored and effective treatment strategies, potentially improving long-term visual outcomes for patients.

Summary updated

Abstract

A relevant percentage of eyes diagnosed with PNV might instead suffer from PAT1/PCV. The detection of a maximum PED height ("peaking PED") exceeding approximately 150 µm, SHRM, SRRLS, and sub-RPE fluid might greatly aid in the production of a more accurate diagnosis.
BibTeX
@inproceedings{siedlecki2023optical,
  author = {Jakob Siedlecki and Julian Klaas and Leonie Keidel and Ben Asani and Johannes Schiefelbein and Dominik Knebel and Nikolaus Luft and Siegfried Priglinger and Benedikt Schworm},
  title = {{Optical coherence tomography-based misdiagnosis and morphological distinction in pachychoroid neovasculopathy vs. polypoidal choroidal vasculopathy}},
  booktitle = {Eye},
  year = {2023},
  doi = {10.1038/s41433-023-02529-5},
}
Figure