This retrospective availability-based series compared participant-level macular OCTA metrics between individuals with documented lifetime lens dislocation (LD+) and those without (LD−). The study analyzed five predefined OCTA endpoints generated by the device’s AngioAnalytics pipeline: whole-image, foveal, parafoveal, perifoveal superficial capillary plexus (SCP) vessel density, plus right-eye foveal avascular zone (FAZ) area. Device-reported medians were transcribed directly from the clinical export.
Investigators reported lower median SCP vessel-density values in LD+ participants across all four regional metrics. The foveal SCP vessel density showed the largest unadjusted between-group difference: median 19.38% (IQR, 18.68–20.04) in LD+ versus 30.20% (IQR, 29.20–31.85) in LD− (p = 0.036). That comparison met the conventional unadjusted threshold but was not robust to multiplicity sensitivity analysis (Benjamini–Hochberg q = 0.180). Whole-image, parafoveal, and perifoveal vessel-density comparisons did not achieve nominal significance (p = 0.328, 0.272, and 0.238, respectively; q = 0.410 each). Right-eye FAZ area, restricted to available OD exports, also did not differ significantly (LD+ median 0.148 mm2 vs LD− median 0.272 mm2; p = 0.606; q = 0.606).
The authors emphasized that the unadjusted foveal signal should be interpreted cautiously. Small sample size, multiplicity adjustment, and potential confounding by ocular biometry limited the confidence with which the apparent foveal reduction could be attributed to lens-dislocation status alone.
Archived routine-care genetic reports were available for seven LD+ participants and were reviewed descriptively without reclassification. Six of the seven reports documented a reported FBN1 variant; one report was assay-scope negative. The study team transcribed reported variant nomenclature and laboratory interpretations as issued by the clinical laboratory. No raw sequence reanalysis or application of updated ACMG/AMP or ClinGen FBN1-specific criteria was performed. Consequently, variant-level statements in the paper reflect the original clinical-report content rather than investigator-driven reinterpretation.
Because archived genetic reports were available only for LD+ participants, the authors did not perform between-group genetic comparisons. The report series included family-targeted and broader next-generation sequencing strategies; reported coverage and assay limitations were retained in the descriptive summary. The investigators explicitly declined to test individual-level genotype–OCTA associations because of the small, related, and heterogeneously tested sample.
All macular scans were obtained on an Optovue RTVue XR Avanti with a 6 × 6-mm acquisition. The single masked retina specialist reviewer applied prespecified quality criteria: signal strength index (SSI) ≥50, acceptable automated segmentation, preserved vascular continuity, adequate foveal centration, and absence of major motion artifacts. No manual segmentation correction was performed. Under these rules, 29 eyes from 15 participants contributed quantitative vessel-density measurements; one participant had no eye meeting the SSI threshold and therefore contributed no quantitative OCTA data.
Spherical equivalent and axial length differed markedly between groups. Median spherical equivalent was −7.75 D (IQR, −8.25 to −6.75) in LD+ versus −1.50 D (IQR, −1.75 to −1.38) in LD− (p = 0.001). Median axial length was 25.25 mm (IQR, 24.80–25.60) in LD+ and 23.65 mm (IQR, 23.55–23.65) in LD− (p < 0.001). These biometric divergences are relevant because refractive status and axial length influence OCTA signal and automated segmentation, and therefore they constitute plausible confounders for vessel-density comparisons.
The study used participant-level summaries derived by averaging valid bilateral measurements or using a single valid eye when only one eye passed quality control. Continuous outcomes were summarized as median and interquartile range and compared with two-sided Mann–Whitney U tests. Effect sizes were expressed as rank-biserial correlations with percentile-bootstrap 95% confidence intervals. Categorical variables used Fisher’s exact tests where applicable.
Investigators treated the five OCTA endpoints as exploratory and reported unadjusted p values as primary results. As a multiplicity sensitivity analysis, Benjamini–Hochberg q values were calculated across the five endpoints. The nominal two-sided p < 0.05 threshold was applied for unadjusted exploratory comparisons; the authors reported that the foveal SCP signal did not persist after the q-value adjustment. No multivariable models were fitted because the sample size was judged insufficient for stable adjusted estimation.
The authors listed multiple limitations that constrain interpretation. Sampling was availability-based rather than consecutive; the cohort was small (n = 16) and included related participants. Archived genetic reports were heterogeneous in assay scope and were not subjected to reclassification. Biometric and refractive differences between groups were large and likely to affect OCTA outputs. The FAZ analysis was limited to right-eye exports because left-eye FAZ values were not present in the retained analytical export.
The report did not establish OCTA as a diagnostic or prognostic test for fibrillinopathy, nor did it demonstrate an individual-level genotype–OCTA association. The study did not perform standardized prospective genetic testing, did not re-evaluate reported variants under contemporary criteria, and did not adjust vessel-density comparisons for axial length or refractive error. These specific unresolved questions remain: would a larger, prospectively ascertained cohort with uniform genetic testing and biometric adjustment reproduce the foveal signal? How much of any observed vessel-density difference is driven by axial length and segmentation artifacts rather than underlying microvascular change? The authors positioned their observations as hypothesis-generating and urged that such questions require targeted prospective study rather than inference from this small retrospective series.