Intraocular Lens Power Calculation Formulas—A Systematic Review

Author: Stopyra W, Langenbucher A, Grzybowski A

Geographical coverage: Not reported

Sector: Cataract surgery

Sub-sector: Intraocular lens

Equity focus: Not reported

Study population: Patients with cataracts

Review type: Effectiveness review

Quantitative synthesis method: Narrative synthesis

Qualitative synthesis method: Not applicable

Background:

Refractive accuracy is critical for successful cataract surgery, yet about 26% of cases miss the ±0.5 D target. These errors often arise from biometric inaccuracies, particularly in axial length, keratometry, and anterior chamber depth, as well as from incorrect estimation of effective lens position (ELP). Numerous IOL power calculation formulas have been developed, which are now often classified by their calculation logic rather than by generation. While most formulas perform well in eyes with average axial length, accuracy drops in short and long eyes. The Barrett Universal II and Kane formulas show strong results for myopic and hyperopic eyes, respectively. Despite advancements, no consensus exists on a universally best formula.

Objective:

To describe and compare the accuracy of most of the formulas on the market.

Main findings:

The review analysed 29 studies comparing the accuracy of various IOL power calculation formulas. Findings showed that in general patient populations, newer formulas such as Barrett Universal II, Kane, and PEARL-DGS demonstrated superior accuracy compared with traditional formulas (e.g., Hoffer Q, SRK/T, Haigis). Barrett Universal II consistently achieved the lowest mean absolute error (MAE), while PEARL-DGS showed the highest percentage of eyes within ±0.5 D, particularly in short eyes (AL < 22.0 mm). In long eyes (AL > 26.0 mm), the Kane formula performed best, closely followed by Barrett Universal II and EVO.

Subgroup analyses revealed that third- and fourth-generation formulas (e.g., Holladay 1, SRK/T) were gradually outperformed by AI-based and hybrid formulas (e.g., Hill-RBF, Ladas Super Formula). However, no single formula was universally optimal across all axial lengths. The study highlighted that accuracy varied significantly with biometric factors such as corneal curvature and the estimation of effective lens position (ELP). Overall, the findings underscored the importance of selecting formulas based on individual patient characteristics, particularly axial length.

Methodology:

Searches were conducted in PubMed, Web of Science, SciELO, Crossref, Google Scholar, and the Cochrane Library to identify studies on the accuracy of IOL power calculation formulas. Peer-reviewed articles published in English between 2015 and 2022 were included in the review. Articles identified through these searches were screened, and relevant data were extracted. The primary outcomes of interest were mean absolute error (MAE) and the percentage of eyes achieving refractive predictions within ±0.5 D and ±1.0 D of the target. Findings were synthesised narratively.

Applicability/external validity:

The review did not discuss the applicability or external validity of its findings.

Geographic focus:

The review did not apply any geographical limits; however, it did not report the geographical distribution of the included studies.

Summary of quality assessment:

Overall, confidence in this review’s conclusions is low. The search strategy was comprehensive, with clearly defined inclusion and exclusion criteria. Characteristics of the included studies were well presented. Findings were synthesised narratively, and heterogeneity was discussed. However, the review did not report whether a formal assessment of study quality or risk of bias was conducted. It was also unclear how many reviewers independently screened the studies or extracted data. While a list of included studies was provided, excluded studies were not listed, and there was no mention of contacting authors or screening reference lists. The search was limited to peer-reviewed articles published in English only.