# Intraocular Lens Power Calculation and Modern Formulas

## Biometry: Measuring the Eye

### Axial Length Measurement

Axial length is the most critical variable in IOL power calculation, as a 1 mm measurement error translates to approximately 2.5 D of refractive error. Optical biometry using instruments such as the IOLMaster and Lenstar is the gold standard, employing partial coherence interferometry (PCI) or optical low-coherence reflectometry (OLCR). These devices cannot measure through dense media such as mature cataracts or vitreous hemorrhage, in which case ultrasound A-scan biometry is needed. Applanation A-scan, where the probe contacts the cornea, may compress it and shorten the measurement by 0.1 to 0.3 mm, whereas immersion A-scan avoids this compression artifact and is more accurate. Normal axial length is approximately 23.5 mm, with short eyes less than 22 mm and long eyes greater than 25 mm.

### Keratometry

Keratometry measures corneal curvature to estimate refractive power, using a standard refractive index of 1.3375. It measures only the anterior corneal surface and assumes a fixed anterior-to-posterior curvature ratio, an assumption that breaks down after corneal refractive surgery.

### Anterior Chamber Depth (ACD)

ACD is measured from the corneal epithelium to the anterior lens surface and is used by newer formulas to predict the effective lens position. Lens thickness and white-to-white corneal diameter provide additional refinement, particularly in short and long eyes.

## IOL Power Formulas

The SRK/T formula combines regression and theoretical elements and performs well for average-length eyes (22 to 25 mm). The Holladay 1 uses axial length and K to predict a surgeon factor, while the Holladay 2 uses seven parameters for improved ELP prediction in short eyes and outliers. The Haigis formula uses three optimizable constants (a0, a1, a2) and is particularly useful for eyes with unusual ACD.

The Barrett Universal II is widely considered among the most accurate across all axial lengths, employing a thick-lens model that accounts for the IOL's principal planes. The Hill-RBF formula uses artificial intelligence and provides a confidence boundary indicating prediction reliability. The Kane formula combines theoretical optics with AI refinement and consistently ranks among the top performers. The EVO formula is a cloud-based AI-enhanced option with strong performance across multiple comparative studies.

## IOL Formula Comparison

| Formula | Generation | Key Variables | Best Use | Notes |
|---------|-----------|---------------|----------|-------|
| SRK/T | 3rd | AL, K | Average eyes (22-25 mm) | Regression + theoretical hybrid |
| Hoffer Q | 3rd | AL, K | Short eyes (<22 mm) | Personalizable ACD |
| Holladay 1 | 3rd | AL, K | Average eyes | Surgeon factor concept |
| Holladay 2 | 4th | 7 parameters | Short eyes, outliers | Improved ELP prediction |
| Haigis | 4th | AL, ACD, K | Unusual ACD | 3 optimizable constants (a0, a1, a2) |
| Barrett Universal II | Latest | Thick-lens model, multiple variables | All axial lengths | Consistently top performer |
| Kane | Latest | AI-enhanced theoretical | All axial lengths | Excellent across validation studies |
| Hill-RBF | Latest | AI (neural network) | All axial lengths | Provides confidence boundary |

## Effective Lens Position (ELP)

The ELP is the predicted postoperative IOL position and represents the most significant source of systematic error. It is estimated from preoperative data, and personalized A-constants derived from surgical outcomes refine its prediction for each surgeon.

## Special Situations

### Post-Refractive Surgery Eyes

Three error sources complicate IOL calculation after LASIK/PRK: incorrect keratometry (overestimates corneal power), invalid keratometric index (altered anterior/posterior ratio), and incorrect ELP prediction (flat K values cause underestimation). The ASCRS Post-Refractive IOL Calculator averages multiple methods and is invaluable. When pre-LASIK data are available, the clinical history method is most reliable. Surgeons should aim for slight myopia rather than hyperopia in these eyes.

### Short and Long Eyes

Short eyes carry higher risk of hyperopic surprise; Barrett Universal II, Hoffer Q, Holladay 2, and Kane perform best. Long eyes risk myopic surprise with traditional formulas; Barrett Universal II, Holladay 1 with the Wang-Koch adjustment, Kane, and Hill-RBF are preferred. Formula choice becomes increasingly important beyond 26 mm.

### Toric IOL Calculations

Accurate measurement of corneal astigmatism, including posterior corneal astigmatism (~0.3 D against-the-rule on average), is essential. The Barrett Toric Calculator accounts for posterior corneal astigmatism. Intraoperative aberrometry and preoperative axis marking are critical for accurate alignment.

<image>Schematic diagram of optical biometry showing the principle of partial coherence interferometry. A low-coherence light source sends a beam into the eye, reflecting off the cornea and retina. An interferometer detects the time delay between reflections to calculate axial length. Label the light source, beam splitter, reference mirror, corneal reflection, retinal reflection, and the detector. Show the interference signal graph with peaks corresponding to corneal and retinal surfaces.</image>

<image>Comparison diagram showing IOL power calculation in a normal eye versus a post-LASIK eye. Normal eye: standard keratometry accurately measures corneal power, and the formula correctly predicts ELP. Post-LASIK eye: three error sources illustrated: (1) keratometry overestimates true corneal power (measured K vs. true K), (2) altered anterior/posterior corneal ratio invalidates the standard keratometric index, (3) flat K values cause the formula to predict a more anterior ELP than actual. Arrows and labels show each error source contributing to a hyperopic surprise.</image>

<image>Bar chart comparing the mean absolute error (MAE) of major IOL formulas (SRK/T, Hoffer Q, Holladay 1, Holladay 2, Haigis, Barrett Universal II, Kane, Hill-RBF) across three axial length categories: short (<22 mm), medium (22-26 mm), and long (>26 mm). Barrett Universal II and Kane consistently show among the lowest MAE across all groups. Use color coding for each formula and error bars.</image>

## Clinical Pearls

Axial length is the most important measurement, and it should always be verified with repeat measurements while comparing eyes for symmetry. The Barrett Universal II or Kane formulas provide the best overall accuracy for most surgeons. In post-refractive surgery eyes, the ASCRS online calculator that averages multiple methods is invaluable. IOL constants should be optimized using the surgeon's own outcomes data. Targeting mild myopia (-0.25 to -0.50 D) is safer than emmetropia in post-refractive surgery eyes. Dense cataracts that prevent optical biometry require immersion (not applanation) A-scan. Always check for posterior staphyloma in highly myopic eyes, as it can cause A-scan errors.

## References
- Melles RB, et al. Accuracy of intraocular lens calculation formulas. Ophthalmology. 2018;125(2):169-178.
- Kane JX, et al. Intraocular lens power formula accuracy: comparison of 7 formulas. J Cataract Refract Surg. 2017;43(11):1490-1500.
- Wang L, Koch DD. Modified axial length adjustment for the Holladay 1 formula in long eyes. J Cataract Refract Surg. 2018.
- Shammas HJ, et al. Correcting the corneal power measurements for intraocular lens power calculations after myopic laser in situ keratomileusis. Am J Ophthalmol. 2003.
- AAO BCSC Section 3: Clinical Optics — Intraocular Lenses. 2023-2024.