Are you measuring your OCT precisely in Heyex?

In my previous article, I discussed the complexities of converting pixel data to physical measurements (micrometers) in optical coherence tomography (OCT) imaging. Building on that technical foundation, this article focuses on the clinical-academic implications for ophthalmological measurements, particularly when using the Heidelberg Eye Explorer (Heyex) software.

Accurate measurements are crucial in ophthalmology, and it's essential to avoid common pitfalls. Let's consider an example of a patient with retinal vein occlusion. When measuring in the 1:1 pixel setting, even slight image tilt can lead to significantly different results due to scaling effects explained in the previous article.

Axial measurement

In this example we are measuring central retinal thickness in 1:1 pixel settings with caliper:

However when we switch from 1:1px to 1:1μm in Heyex2 even with slight tiltation we have very different results because of scaling that was explained in last article.

After correct measurment in this case, the difference is 65 micrometers. This might seems like a trivial difference, however image is not even much tilted and yet we have an 11% measurement error, despite minimal image tilt.

What I'm emphasizing is that measurements should never be taken in the 1:1 pixel setting, only in the 1:1 μm setting. Additionally, the measurement setting used should always be clearly documented in studies.

Transversal (horizontal) measurement

When performing transversal measurements in ophthalmology, it is imperative to adjust for individual eye characteristics that can affect image magnification. Various ocular components—including ocular refraction, corneal curvature, the refractive indices of ocular media, axial length, and anterior chamber depth—can influence the magnification of retinal images. To account for these factors, correction formulas like the adapted Littmann formula is commonly used.

The Littmann formula is expressed as:

The eye-specific magnification factor q can be estimated using Bennett’s formula:

q = 0.01306 x (axial length - 1.82)

In this equation, 1.82 mm is the assumed constant distance between the corneal apex and the eye's second principal point. The axial length can be either the individual patient's measurement or the standard axial length assumed by the OCT system.

Different OCT devices calibrate their systems based on varying standard axial lengths, leading to differences in the q factor. For instance:

• Cirrus HD-OCT assumes an axial length of 24.46 mm
• Spectralis OCT uses 24.385 mm
• Optovue OCT is based on 23.82 mm (Ctori et al., 2015; Odell et al., 2011; Sampson et al., 2017).

These variations mean that the magnification factors — and consequently the measurements— can differ between devices.

Practical Example - transversal measurement

Scenario: A patient has a measured axial length of 31.60?mm (longer than the standard axial length), and a macular hole measuring 400?μm on the OCT image. The OCT device is a Spectralis OCT, which assumes a standard axial length of 24.385?mm.        

1. Calculate q using the device's standard axial length:

• q(s) = 0.01306 x (axial length - 1.82)
• q(s) = 0.01306 x (24.385 - 1.82) = 0.2947

if axial length would be a device constant, we can think that t = s (what we measure is actually real size). In that case we could also say that p = 1/q(s).

in our case p = 1 / 0.2957 = 3.3818

2. Calculate q using the device's real patient axial length:

• q(r) = 0.01306 x (axial length - 1.82)
• q(r) = 0.01306 x (31.6 - 1.82) = 0.01306 x 25.18 = 0.3889

3. Determine true size t

• t = p*q(r)*s
• t = 3.3818 x 0.3889 x 400
• t = 526,10 μm

Result: The true size (t) of the macular hole is 526.1?μm, which is larger than the measured size (s) of 400?μm due to the longer axial length of the patient's eye.

Interpretation: The macular hole is actually larger than what is measured on the image because the eye's longer axial length reduces the magnification, making structures appear smaller on the OCT image.

While the Littmann formula is widely used, it has its limitations. It primarily incorporates axial length and assumes a fixed position for the principal point (1.82 mm from the corneal apex), neglecting other ocular factors such as refractive error, anterior chamber depth, and corneal curvature. These factors can significantly impact the magnification, especially in eyes that have undergone cataract or refractive surgery.

Therefore, when conducting transversal measurements, it is crucial to:

• Input individual ocular parameters into the OCT system when possible.
• Apply appropriate correction formulas to adjust for individual variations.
• Be aware of the device-specific assumptions and calibrations of the OCT system being used.

Take home message

• All measurements needs to be measured in 1:1 μm setting.
• Axial measurement is not depending on ocular magnification.
• Transversal measurement is depending on ocular magnification.
• The transversal measurement is imprecise in patients with very short or extremely long axial lengths.
• Volume / layer thickness (which is combination of both axial and transversal measurments) is depending on ocular magnification.
• Considering most of experimental errors in Bennett formula, the total uncertainty is around ±3.4%, which is acceptable for clinical and research purposes.

References:

Elksne, E., Stingl, J.V., Schuster, A.K. et al. Do biometric parameters improve the quality of optic nerve head measurements with spectral domain optical coherence tomography?.BMC Ophthalmol 22, 56 (2022). https://doi.org/10.1186/s12886-022-02281-6

Bennett AG, Rudnicka AR & Edgar DF (1994): Improvements on Littmann's method of determining the size of retinal features by fundus photography. Graefes Arch Clin Exp Ophthalmol 232: 361–367.

Knaapi, L., Aarnisalo, E., Vesti, E. and Leinonen, M.T. (2015), Clinical verification of the formula of Bennett et al. (1994) of determining the size of retinal features by fundus photography. Acta Ophthalmol, 93: 248-252. https://doi.org/10.1111/aos.12555

Littmann H. Zur Bestimmung der wahren Gr?sse eines Objektes auf dem Hintergrund des lebenden Auges [Determination of the real size of an object on the fundus of the living eye]. Klin Monbl Augenheilkd. 1982 Apr;180(4):286-9. German. doi: 10.1055/s-2008-1055068. PMID: 7087358.

Ctori I, Huntjens B. Repeatability of Foveal Measurements Using Spectralis Optical Coherence Tomography Segmentation Software. PLoS One. 2015 Jun 15;10(6):e0129005. doi: 10.1371/journal.pone.0129005. PMID: 26076457; PMCID: PMC4468112.

Odell D, Dubis AM, Lever JF, Stepien KE, Carroll J. Assessing Errors Inherent in OCT-Derived Macular Thickness Maps. J Ophthalmol. 2011;2011:692574. doi: 10.1155/2011/692574. Epub 2011 Aug 17. PMID: 21869920; PMCID: PMC3157761.

Rudnicka AR, Burk RO, Edgar DF, Fitzke FW. Magnification characteristics of fundus imaging systems. Ophthalmology. 1998 Dec;105(12):2186-92. doi: 10.1016/S0161-6420(98)91214-3. PMID: 9855145.

Sampson DM, Gong P, An D, Menghini M, Hansen A, Mackey DA, Sampson DD, Chen FK. Axial Length Variation Impacts on Superficial Retinal Vessel Density and Foveal Avascular Zone Area Measurements Using Optical Coherence Tomography Angiography. Invest Ophthalmol Vis Sci. 2017 Jun 1;58(7):3065-3072. doi: 10.1167/iovs.17-21551. PMID: 28622398.

Savini G, Barboni P, Parisi V, Carbonelli M. The influence of axial length on retinal nerve fibre layer thickness and optic-disc size measurements by spectral-domain OCT. Br J Ophthalmol. 2012 Jan;96(1):57-61. doi: 10.1136/bjo.2010.196782. Epub 2011 Feb 24. PMID: 21349942.

Salmon AE, Sajdak BS, Atry F, Carroll J. Axial Scaling Is Independent of Ocular Magnification in OCT Images. Invest Ophthalmol Vis Sci. 2018 Jun 1;59(7):3037-3040. doi: 10.1167/iovs.17-23549. PMID: 30025118; PMCID: PMC6005622.

Niyazmand H, Lingham G, Sanfilippo PG, Blaszkowska M, Franchina M, Yazar S, Alonso-Caneiro D, Mackey DA, Lee SS. The effect of transverse ocular magnification adjustment on macular thickness profile in different refractive errors in community-based adults. PLoS One. 2022 Apr 13;17(4):e0266909. doi: 10.1371/journal.pone.0266909. PMID: 35417477; PMCID: PMC9007368.

[1] Scoles D, Mahmoud TH. Inaccurate Measurements Confound the Study of Myopic Macular Hole. Ophthalmol Retina. 2022 Feb;6(2):95-96. doi: 10.1016/j.oret.2021.10.009. PMID: 35123728.