Table 4.
Summary of ICL sizing and postoperative vault prediction using anterior segment imaging domains
| Study | Number of study participants | Country | Anterior segment imaging device | Algorithm | Variables for ICL sizing and vault prediction |
|---|---|---|---|---|---|
| Manufacturer (STAAR nomogram) | - | - | - | - | WTW, ACD |
| Dougherty, et al. (2007) [26] | 73 eyes of 48 patients | USA | UBM (VuMax-II) | Linear regression | Sulcus-to-sulcus, ICL power |
| Kojima, et al. (2012) [11] | 47 eyes of 25 patients | Japan | UBM (VuMax-II) | Linear regression | ACD, Sulcus-to-sulcus, Sulcus-to-sulcus lens rise |
| Lee, et al. (2012) [10] | 129 eyes of 75 patients | Korea | UBM (Carl Zeiss model 835) | Pearson’s correlation | Sulcus-to-sulcus |
|
Igarashi, et al. (2019) [30] (KS formula v1) |
44 eyes of 23 patients | Japan | AS-OCT (CASIA2) | Spearman rank correlation | ATA (= ACA distance) |
| Nakamura, et al. (2018) [29] (NK formula v1) | 46 eyes of 23 patients | Japan | AS-OCT (CASIA2) | Linear regression | ACW, CLR |
| Nakamura, et al. (2020) [15] (NK formula v2) | 81 eyes of 35 patients | Japan | AS-OCT (CASIA2) | Linear regression (stepwise variable selection) | ACW, CLR |
| Oleszko, et al. (2020) [20] | 81 eyes of 43 patients | Poland | Pentacam and AS-OCT (Visante OCT) | Partial least square regression | SE, ATA (= ACA distance), ACD, LE (Visante OCT), axial length, Keratometry (AS-OCT), Corneal radius (Pentacam), ACV |
|
Igarashi, et al. (2021) [31] (KS formula v2) |
121 eyes of 65 patients | Japan | AS-OCT (CASIA2) | Corrected KS formula for each ICL size (discrete function) |
ATA (= ACA distance), vault prediction by KS formula × 0.8 at a size of 12.1 mm and KS formula × 1.3 at sizes of 13.2 and 13.7 mm |
| Kamiya, et al. (2021) [18] | 1745 eyes of 1745 patients | Japan & South Korea | AS-OCT (CASIA2) | Random forest | Age, sex, refractive power (sphere and cylinder), SE, BCVA, toric lens, WTW, ACD, ATA (= ACA distance), CLR, ACW, LV, CCT, AOD500, TIA500 |
| Kang, et al. (2021) [5] | 3506 eyes of 1753 patients | South Korea | AS-OCT (CASIA2) | XGBoost + LightGBM | age, sex, SE, ACD, ACW, ATA, WTW, CLR, pupil size, CCT, toric lens, ICL power |
| Shen, et al. (2021) [32] | 6297 eyes of 3536 patients | China | Pentacam, IOL-Master and UBM (Quantel) | Random forest | ACA, pupil size, axial length, Keratometry, refractive power (sphere and cylinder), ACD, CCT, WTW, SE, toric lens, ICL power, time after surgery |
| Reinstein, et al. (2022) [16] | 147 eyes | United Kingdom | Ultrasound scanner (ArcScan) | Linear regression | ciliary body inner diameter, Sulcus-to-sulcus lens rise, pupil diameter |
| Rocamora, et al. (2022) [33] | 115 eyes of 59 patients | Belgium | AS-OCT (MS-39) | LASSO | Keratometry, Corneal volume, CLR, pupil size, iris diameter, ICL power |
| This study | 894 eyes of 471 patients | South Korea | AS-OCT (ANTERION) | LASSO | AQD, ACA distance, LT |
ACA anterior chamber depth, ACD anterior chamber angle, ACD anterior chamber depth, ACV anterior chamber volume, ACW anterior chamber width, AOD500 average nasal and temporal angle open distance at 500 μm, AQD aqueous depth, AS-OCT anterior segment optical coherence tomography, ATA angle-to-angle distance, CCT central corneal thickness, CLR crystalline lens rise, LASSO least absolute shrinkage and selection operator, LE lens elevation, LT lens thickness, LV lens vault, SE spherical equivalent, TIA500 average nasal and temporal trabecular iris angle at 500 μm, UBM ultrasound biomicroscopy, WTW white-to-white distance