In 249 eyes, Kane had the lowest absolute error and four modern formulas generally outperformed traditional methods; EVO and Barrett performed better in long eyes.
doi: 10.1155/joph/6375056In a multicenter database, modern formulas outperformed Haigis-L and Shammas-PL, while versions using posterior corneal curvature were more accurate than corresponding no-PCC methods.
doi: 10.1016/j.ajo.2026.07.059In an independent 275-eye validation set, selecting formulas by AL, K and ACD reduced mean absolute error to 0.24 D and placed 89.1% within ±0.50 D.
doi: 10.1097/j.jcrs.0000000000002055In 59 eyes after secondary Yamane fixation, EVO, Barrett, Hoffer Q and SRK/T did not differ in accuracy; every formula showed directional myopic prediction error.
doi: 10.1186/s12886-026-04934-2In 98 eyes with a capsular tension ring, Barrett Universal II and Holladay II had lower absolute error, but all four formulas showed a mild hyperopic shift.
doi: 10.4274/tjo.galenos.2026.33338In a 325-eye series, O Formula had lower error than four contemporary formulas, but did not differ significantly from SRK/T and was tested at one center.
doi: 10.1016/j.ajo.2025.08.049Across 271 short eyes, EVO and Holladay 1 had the lowest overall mean error, while Haigis, K6, Kane and PEARL-DGS led in particular axial-length ranges.
doi: 10.1007/s00417-026-07505-9In a 2,550-eye retrospective multicenter dataset, Moussa had the lowest mean absolute error, but its margin over the closest formulas was measured in hundredths of a diopter.
doi: 10.1097/j.jcrs.0000000000002079In Swedish compensated events, endophthalmitis, wrong IOL power and incomplete lens removal were most frequent; severe harm also arose outside the operating room.
doi: 10.1007/s00417-026-07493-wIn 164 eyes measuring 26.04–29.72 mm, Holladay 1 NLR and SRK/T WK2 achieved an RMSAE of 0.393 D; some adjusted older formulas underperformed newer methods.
doi: 10.1016/j.apjo.2026.100301