OCT and automated perimetry

2. Spectral-Domain Optical Coherence Tomography (SD-OCT)

Optical Coherence Tomography (SD-OCT) utilizes low-coherence near-infrared interferometry (wavelength $\lambda \approx 840\text{ nm}$) to acquire cross-sectional architectural tomograms of retinal microstructures with axial resolutions of $3\text{ to }5\,\mu\text{m}$.

1. Peripapillary Retinal Nerve Fibre Layer (pRNFL) Profiling

• Acquisition: Peripapillary scan diameter and segmentation depend on the device. Check centration, signal, artefacts and the same-device longitudinal comparison.
• The TSNIT Curve: In healthy eyes, pRNFL thickness displays a classic bimodal "double-hump" distribution along the Temporal $\to$ Superior $\to$ Nasal $\to$ Inferior $\to$ Temporal (TSNIT) trajectory:
  • The superior and inferior humps (normal thickness $110–135\,\mu\text{m}$) correspond to dense convergent bundles of arcuate axons entering the vertical poles.
  • Glaucomatous damage presents as flattening, notching, or complete collapse of the superior and/or inferior peaks.
• Normative Database Color-Coding:
  • Green ($p = 5\%–95\%$): Within normal limits.
  • Yellow ($p < 5\%$): Borderline thinning.
  • Red ($p < 1\%$): Outside normal limits (statistically significant structural loss).

2. Macular Ganglion Cell Analysis (GCIPL / GCC)

Over $50\%$ of all retinal ganglion cell bodies are concentrated within the central $4.5\text{ mm}$ of the macula, where the RGC layer forms a dense band up to $6$ cells thick:

• Parameters Measured: Depending on the OCT platform:
  • GCIPL (Ganglion Cell-Inner Plexiform Layer): Segments the RGC cell bodies (GCL) and their dendrites (IPL), excluding the variable nerve fiber layer.
  • GCC (Ganglion Cell Complex): Segments three layers: $\text{RNFL} + \text{GCL} + \text{IPL}$.
• Complementary macular analysis: Ganglion-cell measurements can help identify paracentral damage alongside peripapillary RNFL and fields. High myopia, staphyloma, macular disease and segmentation errors can affect both methods. Inspect the actual scans and longitudinal change rather than assuming macular analysis is uniformly more reproducible in every high myope.

3. The Structural Floor Effect

In advanced glaucoma, global RNFL measurements may approach a device-dependent measurement floor because residual non-neural tissue and surviving axons contribute to thickness. A stable global number does not prove stability or that all axons have died. Inspect the remaining measurable sectors, macular ganglion-cell scans and segmentation quality, together with fields and disc examination. A 10-2 field can sample central function densely; choose grids by the defect and avoid switching the whole monitoring strategy solely because of one RNFL value.

3. Standard Automated Perimetry (SAP): Humphrey Field Analyzer (HFA)

Standard Automated Perimetry (SAP) is the clinical gold standard for quantifying visual function in glaucoma. It measures differential light sensitivity across the central and peripheral visual field using static white-on-white automated testing.

Physical & Optical Principles of the HFA

• Stimulus Parameters: A common SAP stimulus is Goldmann size III, subtending about $0.43^\circ$, with a brief presentation of about $200\text{ ms}$. A white stimulus is presented on a standard bowl background. Brief presentation reduces opportunities for refixation, but saccadic latency varies; fixation monitoring and artefact review remain necessary.
• Background Luminance: Maintained at $31.5\text{ apostilbs (asb)}$ (equivalent to $10\text{ cd/m}^2$). This places retinal adaptation in the photopic/upper mesopic range where Weber's law of contrast sensitivity operates: $\frac{\Delta I}{I} = \text{Constant}$
• The Decibel Scale ($\text{dB}$): Retinal sensitivity is quantified on a logarithmic decibel scale: $\text{Sensitivity (dB)} = 10 \times \log_{10} \left( \frac{I_{\text{max}}}{\Delta I} \right)$ where $I_{\text{max}} = 10,000\text{ asb}$ (the maximum light intensity delivered by the perimeter projection system):
  • $0\text{ dB}$ represents maximum brightness ($10,000\text{ asb}$, meaning the patient can only perceive an intensely blinding light flash).
  • $30\text{ to }35\text{ dB}$ represents a normal central retinal threshold (the patient can detect an extremely dim stimulus of $\approx 10\text{ asb}$).
  • $50\text{ dB}$ represents the theoretical limit of maximum sensitivity ($0.1\text{ asb}$). A change of $10\text{ dB}$ represents a $10$-fold (one log unit) shift in retinal light sensitivity.

Perimetric Testing Protocols

• SITA-Standard (Swedish Interactive Thresholding Algorithm): Uses sophisticated Bayesian probability models based on extensive normative datasets. By updating prior probabilities based on responses to neighbouring test points, it cuts testing time to $4\text{ to }6\text{ minutes}$ per eye while preserving maximal test-retest reproducibility.
• SITA-Fast: Employs wider step sizes and abbreviated Bayesian algorithms, shortening test duration to $2\text{ to }3\text{ minutes}$ per eye. Useful in tired, elderly, or easily fatigued patients, though with slightly higher measurement variability.
• SITA-Faster: Eliminates blind-spot catch trials and false-negative trials entirely (monitoring fixation via automated corneal gaze tracking), reducing test time by an additional $30\%$.

Standard Testing Grids: 24-2, 30-2, and 10-2