Optical biometry is a test that uses optical interference to measure axial length, corneal curvature, anterior chamber depth, lens thickness, and other biometric data of the eye non-invasively.
Most devices are equipped with swept-source optical coherence tomography (SS-OCT), enabling highly accurate and reproducible measurements. Its main applications include not only IOL power calculation but also obtaining axial length correction values for glaucoma OCT analysis in highly myopic eyes, preoperative refractive surgery evaluation, and follow-up of myopia control treatment with low-concentration atropine.
When Harold Ridley performed the first IOL implantation in 1949, the patient had a refractive error of about 20 D. In the late 1960s, IOL power estimation using vergence formulas began, marking the starting point of modern calculation methods 1). In the 1970s, ultrasound A-mode biometry was established, and later the IOLMaster using partial coherence interferometry (PCI) appeared, standardizing optical measurements. In recent years, third-generation devices with SS-OCT have become widespread, achieving further improvements in accuracy.
It measures axial length (AL), corneal refractive power (K value), anterior chamber depth (ACD), lens thickness (LT), and corneal diameter (white-to-white: WTW). These parameters are used to predict the effective lens position (ELP) and calculate the required IOL power. Some devices can also measure central corneal thickness (CCT).
| Parameter | Abbreviation | Normal Range | Significance in IOL Calculation |
|---|---|---|---|
| Axial Length | AL | 22–25 mm (average emmetropic eye ~24 mm) | Most important. A 1 mm error affects IOL power by approximately 2.5–3 D. |
| Corneal curvature | K value | Anterior average 7.5 mm (approx. 44 D) | Second most important. A 1 D error is reflected almost 1:1. |
| Anterior chamber depth | ACD | Emmetropic 3–4 mm | Required for ELP prediction |
| Lens thickness | LT | Approx. 4–5 mm | Additional variable in new-generation formulas |
| Corneal diameter | WTW | Approx. 11–12 mm | Used for ELP prediction and IOL size selection |
| Central corneal thickness | CCT | Approximately 530–550 μm | Depends on the device (used for glaucoma evaluation, etc.) |
IOLMaster 700 (Carl Zeiss Meditec)
Measurement method: SS-OCT (wavelength 1,050 nm band)
Features: Integrates with the cataract surgery support system “CALLISTO eye.” Provides centering guidance for toric IOLs and multifocal IOLs.
Strengths: Adaptability to advanced cataracts and posterior uveitis cases. With swept-source technology, it can measure more cataract eyes than the previous generation PCI 3).
ARGOS (Alcon Japan)
Measurement method: Equipped with SS-OCT
Features: Integration with the astigmatic axis alignment system “VERION.” Implements segmented axial length measurement (applying individual refractive indices to each segment).
Strengths: Segmented correction in long and short eyes is expected to improve calculation accuracy. VERION integration enhances intraoperative astigmatic axis management.
Optical biometry has been shown to provide significantly higher accuracy and operator-independent results compared to A-mode ultrasound3).
When using IOLMaster, adopt measurements with signal-to-noise ratio (SNR) ≥ 5. When using optical biometers, use optical-specific IOL constants. The A-constants provided by IOL manufacturers are only recommended values; optimization based on surgeon’s experience or use of the ULIB database (User Group for Laser Interference Biometry) is beneficial3). Measuring and comparing axial lengths of both eyes helps in early detection of measurement errors.
In the following cases, sufficient signal cannot be obtained, making optical measurement difficult or impossible.
For these cases, A-mode ultrasound biometry is used. The ESCRS guidelines recommend using ultrasound biometry when optical methods are not applicable for mature or advanced cataracts1).
In cases of dense cataract or poor fixation, optical measurement may be difficult. The alternative is A-mode ultrasound biometry, with the immersion method recommended over the applanation method due to less compression error. Some reports indicate no statistically significant difference between immersion A-scan performed by an experienced operator and optical biometry1).
| Item | Optical (SS-OCT/PCI) | Ultrasound A-mode (Immersion) | Ultrasound A-mode (Applanation) |
|---|---|---|---|
| Principle | Optical interference (wavelength 1,050–1,310 nm) | Measurement of sound wave propagation time | Measurement of sound wave propagation time |
| Contact | Non-contact | Non-contact (immersion probe) | Contact (corneal compression) |
| AL error | 0.01–0.02 mm | Equivalent to optical (experienced user) | 0.1–0.2 mm (with compression error) |
| User dependency | Low | Moderate | High |
| Mature cataract compatibility | Difficult to impossible | Possible | Possible |
| Infection risk | None | Low (managed with disinfection) | Yes (contact) |
Strengths:
Limitations:
In ultrasound methods, the sound velocity in the medium directly affects measurement accuracy.
Contact (applanation) methods compress the cornea, artificially shortening the axial length. Immersion methods avoid compression errors because the probe does not touch the cornea directly, but require careful alignment. Reports indicate that immersion methods performed by experienced operators show no statistically significant difference from optical methods 1).
IOL power calculation formulas have evolved over generations, and currently new-generation formulas such as Barrett Universal II, Kane, and Hill-RBF demonstrate high predictive accuracy. The main characteristics of each formula are as follows 3).
| Formula Category | Representative Formula | Additional Variables | Indications |
|---|---|---|---|
| Third Generation (Old Generation) | SRK/T, Holladay I, Hoffer Q | None/ACD | Normal eyes (new generation recommended currently) |
| Fourth generation | Barrett Universal II, Haigis | ACD, LT, WTW | Good across all axial lengths |
| AI/Regression hybrid | Kane, Hill-RBF, Pearl-DGS | ACD, LT, WTW | Improved accuracy especially in extreme axial lengths |
Older generation regression formulas (SRK-II, SRK, Binkhorst, etc.) are no longer recommended 5). Newer formulas (such as Barrett Universal II) have been reported to improve accuracy, especially in eyes with extreme axial lengths 4).
Post-refractive surgery eyes: Because the anterior-to-posterior corneal curvature ratio is altered, standard formulas produce systematic errors. Specialized methods such as the ASCRS online calculator, Barrett True-K formula, and Haigis-L formula are required 1).
Toric IOLs: Indications include corneal astigmatism of ≥2.0 D with-the-rule or ≥1.5 D against-the-rule. Use of the Haigis-T, Barrett Toric, or Kane Toric formula is recommended 3).
Silicone oil-filled eyes: Optical biometry is most accurate. Because silicone oil acts as a minus lens, the IOL power should be adjusted by 3–5 D 1).
Refractive surgery (LASIK, PRK, RK) alters the anterior-to-posterior corneal curvature ratio. Keratometers estimate the posterior curvature from the anterior curvature alone, so they overestimate corneal power in post-refractive eyes. Additionally, many IOL formulas predict ELP from axial length and corneal power, but this relationship is altered after refractive surgery, leading to errors. Use of specialized methods (e.g., ASCRS online calculator) is recommended 1).
SS-OCT (Swept-Source Optical Coherence Tomography) measures each ocular interface with high precision based on the following principle.
This is the first-generation optical measurement principle adopted by the IOLMaster 500. It measures axial length using the interference pattern of a double beam. Compared to SS-OCT, it has lower penetration and is more likely to be unable to measure in eyes with advanced cataracts.
Conventional optical methods apply a uniform refractive index to the entire eye, which tends to cause overestimation in highly myopic eyes (AL ≥ 25 mm). The “segmented axial length measurement” implemented in ARGOS applies individual refractive indices to each segment (aqueous humor, lens, vitreous body). In long eyes, it displays approximately 0.50 mm smaller than conventional methods, and many calculation formulas have reported improvements in MAE (mean absolute error). However, evidence for this method is still accumulating, and future clinical studies are awaited.
AI-based formulas such as the Hill-RBF method (artificial intelligence pattern recognition), Kane formula, and Pearl-DGS formula have shown improved accuracy 4). Suzuki et al. (2025) conducted a retrospective evaluation of 80 eyes with extreme axial myopia (axial length ≥30.0 mm) and reported that the Kane and Hill-RBF formulas had significantly lower mean absolute errors (MAE) compared to the conventional SRK/T formula 7).
The percentage of eyes within ±0.5 D was 26.3% for SRK/T, 45.0% for Barrett Universal II, 55.0% for Hill-RBF, and 65.0% for Kane. In the subgroup with axial length ≥32 mm, Hill-RBF MAE was 0.49 D and Kane MAE was 0.44 D, which were the best results 7).
Ray tracing based on anterior segment OCT data (Anterion-OKULIX) has been reported to have a significantly lower arithmetic prediction error (−0.13 D vs −0.32 D) compared to the Barrett True K no-history formula in eyes after myopic LVC 6). By directly using full corneal shape data, it is expected to have a theoretical advantage in application to eyes after refractive surgery.
Intraoperative wavefront aberrometry using devices such as the Optiwave Refractive Analyzer is attracting attention as a means to complement preoperative biometry. Reports indicate that in routine adult cataract surgery, it achieves postoperative outcomes comparable to conventional biometry and may enable intraoperative correction of refractive errors 2).
Regular axial length measurements using optical biometers are applied to evaluate the efficacy of myopia control treatments such as low-dose atropine eye drops and orthokeratology. Monitoring axial length every 6 months to 1 year allows objective assessment of treatment effects. Specific measurement intervals and thresholds await future guideline development.
