Objective: To compare the surgical induced astigmatism(SIA) of the total, anterior and posterior cornea between eyes with 2.2-mm and 3.0-mm clear corneal incisions in phacoemulsification. Methods: A prospective double-blind randomized controlled study was conducted on 131 patients (131 eyes) with age-related cataract from October 2017 to June 2018. They were randomized according to the random number table. Sixty-nine patients received phacoemulsification through a 2.2-mm clear corneal incision at 140 degrees; 62 patients were recruited in the control group with a 3.0-mm clear corneal incision at the same position. Lens density was graded and compared between the groups preoperatively. Total corneal refractive power (TCRP), simulated keratometry (SimK) and posterior corneal astigmatism (PCA) (Apex, 4-mm zone) were recorded through the Pentacam preoperatively and at 1 week, 1 month and 3 months postoperatively. Corneal astigmatism was decomposed to vertical/horizontal (X) and oblique changes (Y) using the power vector analysis. Independent sample t test and matched t test were used for statistical analysis. Results: A total of 123 patients (123 eyes) completed 3 months of follow-up including 65 in the 2.2-mm group and 58 in the control group. The age of the two groups was (69±9) and (71±10) years old, and the lens density was (2.08±0.47) and (2.12±0.46) grades, respectively. Both the age and the lens density of the two groups were not statistically significant (both P>0.05). Preoperative TCRP, SimK and PCA were not statistically significant (all P>0.05). SIA was calculated using the vector analysis. Centroid SIA conducted from TCRP, SimK and PCA in the 2.2-mm group was -0.11 D@146°, -0.11 D@151° and -0.03 D@67°, respectively, at 3 months postoperatively. In the 3.0-mm group, it was -0.25 D@158°, -0.24 D@147° and -0.04 D@47°, respectively. Statistical significance was found between the two groups in the Y polar value of SIA of the anterior surface [(-0.10±0.30) vs.(-0.22±0.37) D, t=-2.133, P=0.035] and the SIA of the posterior surface [(0.24±0.16) vs. (0.19±0.12) D, t=2.009, P=0.047] at 3 months postoperatively. In the comparison of the SIA of TCRP and anterior surface, statistical significance was found in the absolute value of SIA at all the intervals in the 2.2-mm group [(0.87±0.80) vs. (0.58±0.48) D, (0.58±0.48) vs. (0.50±0.28)D, (0.57±0.37) vs. (0.47±0.28) D, t=5.102, 4.155, 3.877, all P<0.01] and at 1 week and 1 month in the 3.0-mm group [(0.82±0.57) vs. (0.58±0.41) D, (0.59±0.36) vs. (0.50±0.28) D, t=5.034, 3.919, both P<0.01]. X components of SIA(TCRP) and SIA(simk) were significantly different at 3 month postoperatively in 3.0-mm group (P<0.05). Y components of SIA(TCRP) and SIA(SimK) were significantly different at 1 week postoperatively in the 2.2-mm group[(-0.48±0.85) vs.(-0.24±0.42) D, P<0.01] and 3.0-mm group [(-0.58±0.66) vs. (-0.37±0.42) D, P<0.01]. Conclusions: Compared with the 3.0-mm incision, the 2.2-mm coaxial micro-incision cataract surgery has a lower and more stable SIA. The effect of a clear corneal incision in cataract surgery on the total cornea is more obvious than that on the anterior surface of the cornea. The difference is significant in the early stage and gradually decrease with time. (Chin J Ophthalmol, 2019, 55: 495-501).