Abstract
Purpose :
To evaluate accuracy and tendency of prediction by 4 intraocular lens (IOL) calculation formulas using a new optical biometer (IOL Master 700, Carl Zeiss Meditec) depends on 3 common IOLs, AMO Tecnis 1 ZCB00V (ZCB), STAAR KS-SP (KS), Alcon SN60WF (SN), and biometric factors.
Methods :
This retrospective study used 253 eyes (71 ZCB, 100 KS and 82 SN). Using IOL Master 700, 3 preoperative parameters, axial length (AL), anterior chamber depth (ACD) and keratometry (K), were measured. Preoperative IOL power calculations were done with 4 formulas, Haigis (HG), SRK/T (S/T), HofferQ (HQ) and Holladay2 (H2). Mean absolute estimation error (MAE) and mean postoperative refraction shift (MPRS) were compared, and correlation between postoperative refraction shift (PRS) and the parameters were analyzed among combinations of each formula and each IOL. Wilcoxon signed-rank test, Paired T test and Spearman’s rank correlation were used to analyze accordingly.
Results :
In ZCB, MAE was significantly lower in S/T and H2 compared to HQ (p<0.01). In KS, MAE was significantly higher in HG compared to other formulas (p<0.0001). In SN, no significance in MAE was shown among formulas. In ZCB, MPRS showed significant hyperopic shift in HG and HQ compared to H2 showing slight hyperopic shift (p<0.0001). In KS, MPRS showed significant myopic shift in S/T compared to other formulas showing hyperopic shift (p<0.0001). In SN, MPRS showed significant hyperopic shift in S/T compared to other formulas showing myopic shift (p<0.0001).In ZCB and SN, significantly positive correlation between PRS and AL was observed in HQ and H2 (r>0.3 p<0.01). In ZCB, significantly positive correlation between PRS and ACD was observed in S/T, HQ and H2 (r>0.3 p<0.01). In SN, the same was true in S/T and HQ. In each IOL, positive correlation between PRS and K in HG and HQ, and negative correlation between them in S/T and H2 was observed. Especially, the significance (p<0.01) was observed in HG (r>0.3) and S/T (r<-0.3) using each IOL, in HQ (r>0.3) using KS, and in H2 (r<-0.3) using SN.
Conclusions :
Accuracy and tendency of prediction by the formulas using IOL Master 700 varies depending on the IOLs. The parameters can also influence tendency of prediction in different ways. Conversance to these factors is vital to improve prediction of IOL calculation.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.