Quantitative analysis of the Effect of Ocular Rotation on Postoperative Residual Astigmatism in Small Incision Lenticule Extraction for Myopia Correction

Caiyu Liu¹Xiaoying Xu¹Hao Gu²Chi Wang¹Lu Lu¹Kaizhen Ye³Yan Zheng³Haiyan Wang¹Wei Chen¹Meiyan Li⁴Xiaolin Zhou¹Shangkun Ou^2*Fangwen Yang^1*

• ¹The Second Affiliated Hospital of Guizhou Medical University, Kaili, China
• ²The Affiliated Hospital of Guizhou Medical University, Guiyang, China
• ³Guizhou Medical University, Guiyang, China
• ⁴Fudan University Eye Ear Nose and Throat Hospital, Shanghai, China

Objective: To conduct a systematic investigation into the effects of ocular rotation on postoperative residual astigmatism in patients undergoing small incision lenticule extraction (SMILE). Methods: A prospective observational cohort study involved 79 patients (153 eyes) with myopia and astigmatism who underwent SMILE surgery. Ocular rotational magnitude was measured using manual corneal and scleral marking with slit-lamp microscopy assessment. Preoperative and postoperative (1-month and 3-month) assessments included uncorrected distance visual acuity (UDVA), best-corrected visual acuity (BCVA), refractive error, and other relevant ocular parameters. Results: Residual astigmatism showed significant correlations with ocular rotation magnitude (r = 0.429, P < 0.001), preoperative intraocular pressure (r=-0.178, P=0.032), and preoperative cylindrical lens power (r=0.175, P=0.035). A multiple linear regression analysis indicated that rotation magnitude significantly impacted postoperative residual astigmatism (P<0.001). However, preoperative intraocular pressure (P=0.349) and spherical equivalent (P=0.105) were not significantly related to residual astigmatism. Linear regression analysis further demonstrated significant positive correlations between rotation amplitude and various astigmatism parameters at both 1-and 3-months postoperative follow-ups (all P < 0.05). Specifically, the relationships were quantified as follows: Cylindrical Lens (CYL[D]) (1 month: y = 7.058x + 17.480, P<0.001; 3 months: y=7.464x+13.610, P<0.001), Target Induced Astigmatism (TIA[D]) (1 month: y=0.112x+1.275, P=0.012; 3 months: y=0.097x + 1.217, P = 0.026), Surgically Induced Astigmatism (SIA[D]) (1 month: y = 0.094x + 0.936, P < 0.001; 3 months: y = 0.059x + 0.911, P = 0.022), and Difference Vector (DV[D]) (1 month: y = 0.041x + 0.289, P = 0.005; 3 months: y = 0.037x + 0.866, P = 0.011). Notably, rotation amplitude exhibited the strongest association with postoperative CYL. ROC analysis determined the optimal thresholds for rotation magnitude in predicting residual astigmatism to be 1.5° at 1 month (AUC = 0.753; sensitivity 79.7%, specificity 58.2%) and 2.5° at 3 months (AUC = 0.929; sensitivity 92.9%, specificity 83.5%). Conclusion: The magnitude of rotation shows a notably positive correlation with residual astigmatism during both the 1-month and 3-month postoperative follow-ups. Thresholds of 1.5 ° (1 month) or 2.5 ° (3 months) prove predictive of residual astigmatism, with enhanced diagnostic precision at the later follow-up.