Objective: To investigate the differences and consistency in measuring corneal astigmatism between autorefractors and corneal topographers, as well as in assessing total ocular astigmatism among autorefractors, wavefront aberrometers, and subjective refraction devices, so as to provide a basis for clinical instrument selection for astigmatism measurements. Methods: A retrospective study was conducted. A total of 245 patients (245 right eyes) who underwent corneal refractive surgery and related examination at the Department of Ophthalmology, The Second Affiliated Hospital of Chongqing Medical University and Chongqing Mingda Eye Hospital from January 2015 to November 2023 were enrolled, including 85 males and 160 females, aged (32.17±10.16) years, with astigmatism degrees ranging from -4.50 to -0.25 D. The patients were divided into three age groups: 17-29 years (96 cases), 30-39 years (77 cases), and 40-50 years (72 cases). They were also classified based on the astigmatism degree:-0.25 to -0.75 D (131 eyes), -1.00 to -1.75 D (65 eyes), and ≤-2.00 D (49 eyes). Corneal astigmatism was measured using an autorefractor and a corneal topographer, while total ocular astigmatism was measured using an autorefractor, a wavefront aberrometer, and a subjective refraction device. The horizontal-vertical astigmatism vector (J0) and 45° oblique astigmatism vector (J45) of the cornea and the entire eye were calculated. Differences in corneal and total ocular astigmatism were compared. The Pearson correlation analysis of results from different instruments was performed, and the consistency was evaluated using the Bland-Altman analysis. Results: For corneal astigmatism, there were no statistically significant differences in J0 measurements in the 30-39 years group and J45 measurements in the 40-50 years group between the two instruments; however, statistically significant differences in J0 and J45 measurements were observed in other age groups and the overall population (all P<0.05). Moreover, no statistically significant differences were found in J0 and J45 measurements between the instruments in the ≤-2.00 D group, while statistically significant differences were present in other astigmatism groups and the overall population (all P<0.05). The results presented in a polar coordinate system showed that the measurement values of the two instruments were close in all groups and the overall population. The Pearson analysis indicated that J0 and J45 in all groups and the overall population were significantly correlated (all P<0.05). The consistency analysis revealed relatively good consistency in the 95% confidence interval (CI) of the 95% limits of agreement (LoA) for the overall results presented in a polar coordinate system. For total ocular astigmatism, statistically significant differences in J0 measurements among the three instruments were observed in all age groups and the overall population (all P<0.05). Regarding J45 measurements, only the 30-39 years group showed no significant difference among the autorefractor (P>0.05). In addition, there were statistically significant differences in J0 measurements among the three instruments in all astigmatism groups and the overall population (all P<0.05). Regarding J45 measurements, no significant difference was observed in the overall population and the -1.00 D to-1.75 D group (P>0.05). J0 and J45 in all groups and the overall population showed significant correlations between each pair of the three instruments (all P<0.05). The consistency analysis demonstrated that the upper limit of the 95%CI of 95% LoA for J0 in all groups and the overall population was <0.50 D, and most of the lower limits were <0.50 D; the 95%CI of 95% LoA for J45 in all measurements was within 0.50 D, indicating good consistency among the three instruments. Conclusions: In the undilated pupil state, the autorefractor and corneal topographer have a high correlation and relatively good consistency in measuring corneal astigmatism and can be mutually referential in clinical practice. The autorefractor, subjective refraction device, and wavefront aberrometer have a high correlation and good consistency in the measurements of total ocular astigmatism. Among them, the autorefractor and wavefront aberrometer (objective refraction) can provide references for the subjective refraction device (subjective refraction). Clinically, the selection can base on the applicable scope of the instruments and clinical requirements.
目的： 探讨电脑验光仪与角膜地形图仪测量角膜散光的差异性与一致性，以及电脑验光仪、波前像差仪、综合验光测量全眼散光的差异性与一致性，为临床散光测量的仪器选择提供依据。 方法： 回顾性研究。选取2015年1月至2023年11月在重庆医科大学附属第二医院眼科和重庆明达眼科医院行角膜屈光手术并进行相关检查的245例屈光不正患者（245只右眼），其中男性85例、女性160例，年龄（32.17±10.16）岁，散光度数-4.50~-0.25 D。按年龄分为17~29岁（96例）、30~39岁（77例）、40~50岁（72例）3组；按散光度数分为-0.25~-0.75 D（131只眼）、-1.00~-1.75 D（65只眼）、≤-2.00 D（49只眼）3组。在小瞳孔状态下，角膜散光采用电脑验光仪与角膜地形图仪测量，全眼散光采用电脑验光仪、波前像差仪和综合验光仪测量。计算角膜及全眼的水平-垂直方向散光矢量（J0）和45°方向斜散光矢量（J45），比较角膜和全眼散光的差异，并做不同仪器结果的Pearson相关性分析；一致性评价采用Bland-Altman分析。 结果： 角膜散光测量中，除30~39岁组J0、40~50岁组J45的两种仪器测量差异无统计学意义外，其余年龄层及总体J0和J45测量差异均有统计学意义（P<0.05）；散光度数分组中，除≤-2.00 D组J0和J45测量差异无统计学意义外，其余各组及总体J0和J45测量差异均有统计学意义（均P<0.05）。显示两种仪器各组及总体测量值接近，Pearson分析示各组及总体测量结果均呈显著相关（均P<0.05），一致性分析示总体测量结果95%一致性界限（LoA）的95%置信区间（CI）一致性相对较好。全眼散光测量时，各年龄层及总体J0的3种仪器测量差异均有统计学意义（均P<0.05），J45仅30~39岁组差异无统计学意义（P>0.05）；散光度数分组中，各散光层及总体J0测量差异均有统计学意义（均P<0.05），J45仅总体与-1.00~-1.75 D组差异无统计学意义（P>0.05）。3种仪器两两间各分组及总体J0和J45均呈显著相关（均P<0.05），一致性分析示各分组及总体J0的95%LoA 的95%CI上限均<0.50 D、下限大部分<0.50 D，J4595%LoA的95%CI均在0.50 D内，3种仪器一致性良好。 结论： 小瞳孔状态下，电脑验光仪与角膜地形图仪测量角膜散光的相关性高、一致性较好，临床可互相参考；电脑验光仪、综合验光、波前像差仪测量全眼散光的相关性高、一致性良好，其中电脑验光仪与波前像差仪（客观验光法）可为综合验光（主观验光法）提供参考，临床可根据仪器适用范围及临床要求选择使用。.