| PURPOSE:1.To investigate the biological characteristics of posterior corneal astigmatism(PA)in patients with 1.00 diopter(D)or more of corneal astigmatism preparing for cataract surgery and compare the differences between keratometric astigmatism(KA)and total corneal astigmatism(TA).2.To observe the effect of cataract incision on anterior cornea and posterior cornea,especially the change of posterior corneal surface and its influence on SIA,and to compare the difference between keratometric astigmatism SIA and total corneal astigmatism SIA,so as to provide a theoretical basis for accurately evaluating the total corneal astigmatism that needs to be corrected before Toric IOL implantation.3.To compare the residual astigmatism errors of three Toric calculatons with and without PA considered.Methods1.A rotating Scheimplfulg camera(Pentacam HR)was used to measure 1029 eyes of 1029 patients with with 1.00 diopter(D)or more of corneal astigmatism preparing for cataract surgery from December 2016 to December 2017.The distribution of PA,TA and KA,and analyse the correlation between them.The error produced by estimating TA using KA was calculated through vector and arithmetical analysis.According to the axial length(AL),they were divided into high myopia group(AL>26mm)and control group(20≤AL<26mm).94 eyes in the high myopia group and 121 eyes in the control group were observed for the differences of PA,total TA and KA between the high myopia group and the control group.2.52 patients(88 eyes)underwent phacoemulsification combined with astigmatism corrective IOL implantation From January 2017 to September 2018.Pentacam was used to measure PA,KA and TA before and 3 months after surgery.Vector analysis was used to calculate the SIAPA,SIAKA,SIATA.3.109 eyes of 78 patients who underwent phacoemulsification and Toric IOL implantation from January 2017 to September 2018 were collected.IOL-Master 500 was performed before and 3 months after surgery.The results of IOL-Master 500 were entered into Original Alcon Toric calculaton,New Alcon Toric calculaton,Barrett Predicted PA calculaton and Barrett Measured PA calculaton respectively.The residual astigmatism and axis corresponding to the implanted Toric model were recorded and compared with 3 months after surgery.The vector errors of four calculatons are calculated by vector analysis of postoperative manifest astigmatism.The KA was divided into two groups:the≥2.5D group and the<2.5D group.The accuracy of different calculatons was compared.To compare the accuracy of four different calculatons for patients with WTR PA and to propose a preliminary optimization method.RESULTS1.The mean magnitude of PA before surgery for cataract patients with corneal astigmatism(>1.0D)was(0.33±0.22D).The steepest meridian was vertically aligned in 79.11%of cases and the magnitude of 14.3%of cases exceeded 0.5D.The magnitude(0.40±0.29D@91°)in the with-the-rule(WTR)group was larger than that in the against-the-rule(ATR)group(0.08±0.24D@94).The vector difference between KA and TA was 0.50±0.55D@920,of which 33.8%of patients≥0.5D.The absolute error of KA and TA was the smallest(0.27±0.27D)in the WTR group,and the ATR group was the largest(0.30±0.26D).The KA in the WTR group overestimated the TA by 0.08±0.29D,the KA in the ATR group underestimated the TA by 0.11 ±0.44D,and the KA in the oblique astigmatism group underestimated the total corneal astigmatism by 0.04±0.48D.The mean magnitude of PA in high myopia group was-0.33±0.20D and that in normal axis group was-0.31 ±0.18D.There was no significant difference between the two groups(t=0.589,P=0.557).PA was negatively correlated with age(r=-0.039,P<0.001),and positively correlated with the difference between KA and TA(r=0.221,P<0.05),but not with axial length(r=0.067,P=0.160).The vector errors of KA and TA were positively correlated with age,Km(KA),the difference between corneal apex and the thinnest thickness(r=0.208,0.144,0.138,P<0.05),but not with axial and KA(r=-0.083,0.020,P>0.05).The mean magnitude of the PA of the WTR was 0.22±0.20D and 84.27%of the patients had anterior surface astigmatism of the ATR type.For the patients whose PA was in WTR group and anterior corneal surface was in ATR,there was a moderate positive correlation between PA and KA(r=0.438,P=0.033).2.The steep meridian of PA almost did not change before and after cataract surgery incision,and most of them were still irregular astigmatism.SIAPA was 0.09±0.06D(95%Cl:0.08-0.11D),which was positively correlated with preoperative posterior corneal astigmatism(r=0.228,P=0.032).There were significant differences in the vector errors of KA,TA and PA before and after surgery in the ATR group(P<0.05)The difference between SIAKA and SIATA was statistically significant in WTR group(P<0.05).SIAKA underestimated SIATA by 0.09±0.22D in patients with WTR group,and SIAKA overestimated SIATA by 0.03±0.18D in patients with ATR group.When PA is less than 0.4D,SIAPA can be neglected.When PA is greater than 0.4D,ignoring SIAPA can cause underestimation of SIAKA.When KA is less than 2.0D,there is no significant difference between SIATA and SIAKA,SIAPA can be neglected;when KA is greater than 2.0D,SIAPA will be underestimated.3.Among the four calculatons,the error of the original Alcon calculaton was the largest 0.21±0.55D@173°,0.26±0.51D@177°,0.11±0.59D@149°,and the error drifted to ATR.The difference between the original Alcon calculaton and the other three calculators was statistically significant(P<0.05),and the other three calculatons are all WTR error.For the WTR group,the centroid error of the Barrett Predicted PA calculaton is the smallest(0.24±0.52@94°),for the ATR group,the error of the Barrett Measured PA calculator is the smallest(0.22 ±0.26D@109°).For the Original Alcon calculaton,the WTR centroid error(0.26±0.51D@177°)is greater than that of the ATR group(0.11 ±0.59D@149°).For New Alcon calculaton and Barrett Predicted PA calculaton,the centroid errors in ATR group(0.30±0.61D@103°,0.34±0.63D@102°)were greater than those in WTR group(0.25±0.54D@93°,0.24±0.52D@94°).For Barrett Measured PA calculaton,the centroid error of ATR group(0.22±0.62D@109°)is less than that of WTR group(0.30±0.57D@92°).In whole sample and ATR group,the centroid errors of the four calculations in>2.5D group(Total:0.26 ± 0.68D@161°,0.30 ± 0.64D@102°,0.27 ± 0.67D@106°,0.28 ± 0.67D@102°.ATR group:0.2 ± 0.64D@121°,0.45±0.61D@102°,0.46±0.73D@107°,0.27±0.68D@117°)were greater than those in<2.5D group(Total:0.20±0.46D@1°,0.23±0.54D@93°,0.26±0.50D@92°,0.2 ± 0.55D@92°.ATR group:0.15 ± 0.54D@172°,0.22 ± 0.60D@104°,0.29±0.57D@97°,0.20±0.58D@102°).The difference of Original Alcon calculation between the two groups was significant(P<0.05).In WTR group,except the Original Alcon calculaton,the results of the other three calculatons are close.The Y-axis error of Barrett Predicted PA calculaton was positively correlated with the KA(r=0.197,0.377,P<0.04,0.03)in the whole sample and against-the-rule group,but not in WTR group(r=0.085,P=0.470),while the Y-axis error of Barrett Measured PA calculaton was not correlated with the KA in any case(P>0.05).The percentage of patients with WTR astigmatism was 9.17%.With the increase of astigmatism,the error values of the three new calculators also increase,and there is a strong correlation between them(r=0.788,0.788,0.658,P=0.007,0.007,0.039).Vector errors using New Alcon calculaton and Barrett Predicted PA calculaton are larger than those of Original Alcon calculaton(0.33±0.67D@130°)and Barrett Measured PA calculaton(0.33±0.70D@125°),which are contrary to the conventional results and far larger than the average value.After model adjustment,the vector errors are reduced(0.17±0.49D@171°,0.12±0.49D@147°).CONCLUSION1.Ignorance of PA in cataract patients with corneal astigmatism(>1.0D)leads to an overestimation of TA in about 70%of patients with WTR group,and an underestimation of TA in patients with ATR group,whereas about 30%of patients have the opposite.Individualized PA should be observed before implantation of Toric intraocular lenses to avoid ignoring PA and increasing postoperative refractive error.2.Ignoring the SIAPA caused by cataract incision can lead to SIAKA underestimating SIATA in WTR patients,while SIAKA overestimating SIATA in ATR patients.However,when PA is less than 0.4D or KA is less than 2.0D,the SIAPA can be neglected.3.Even though the new calculaton consider the influence factors and adjusts them,compared with the original Alcon calculaton,the over-correction of the WTR group and the under-correction of the ATR group,although all the new calculatons reduce the vector error,they show the under-correction of the WTR group and the over-correction of the ATR group.In most cases,the error values of New Alcon calculaton and Barrett Predicted PA calculaton are close,but the vector error values of the ATR group are larger than that of the WTR group.In WTR group,the Barrett Predicted PA calculator performed slightly better,while in ATR group,the vector error of Barrett Measured PA calculator was the smallest.When KA is greater than 2.5D,the vector errors of the four calculatons in the whole sample and ATR group are greater than those in the<2.5D group.In this case,the application of Barrett Measured PA calculaton will reduce the errors.For a few patients whose PA is WTR astigmatism,the errors of the three new calculatons increase with the increase of the PA.When implanting Toric IOL,the difference of TA between those and most patients should be taken into account,and the models should be adjusted to reduce the errors after surgery. |
PDF Full Text Request