[Clinical study of suction loss in small incision lenticule extraction]

Objective: To investigate the classification, possible risk factors, managements and clinical outcomes of suction loss in small incision lenticule extraction (SMILE). Methods: In this nested case control study, subjects undergoing SMILE surgery between September 2013 and September 2017 were enrolled in the study. Eyes suffered from suction loss were included in the suction loss group, and eyes without intraoperative complications and operated at the same date were included in the control group. The refraction and visual outcomes were evaluated at preoperative and postoperative 1 day, 1 week, 1, 3 and 6 months. Independent t test and Pearson relation analysis were used for statistical analysis. Results: The study included 206 cases (344 eyes), of which 25 cases (27 eyes) were in suction loss group and 181 cases (317 eyes) were in control group. Suction loss occurred at: (1) scanning posterior surface of lenticule (5, 18.5%); (2) scanning lenticule side cut (1, 3.7%); (3) scanning anterior surface of lenticule (16, 59.3%);(4) scanning incision (5, 18.5%). The main reasons included: (1) abnormal eye movement (14, 51.9%); (2) extra fluid on corneal surface (13, 48.1%). There was significant difference in cap diameter between two groups (t=2.341, P<0.05). At postoperative 3 months, 3.7% (1 eyes) of eyes in cases lost one line; 2.6% (6 eyes) of eyes in controls lost one line. 92.6% (25 eyes) and 91.3% (209 eyes) of eyes had a UCVA of 0.10 LogMAR or better in cases and controls, respectively; 85.2% (31 eyes) and 93.4% (214 eyes) of eyes had a residual SE between±0.50D in cases and controls, respectively. Conclusions: Suction lost during anterior surface of cornea was most common in SMILE surgery. Extra fluid in cornea surface, sudden eye movement of patient and larger corneal cap may result in suction loss. And it may slow the visual recovery, but appropriate and effective managements will be benefited to the final visual outcomes. (Chin J Ophthalmol, 2018, 54: 890-896).