Electron Microscopy of Laser Capsulotomy Edge: Inter-Platform Comparison
Narrative Responses:
Purpose
To compare the edge of capsulotomies generated by femtosecond lasers with manually generated capsulorhexis using a scanning electron microscope (SEM).
Methods
54 capsules were collected using multiple laser platforms and software versions; 10 LenSx “pre-soft-fit” (Alcon, version 2.16), 10 LenSx “post-soft-fit” (version 2.20), 4 Victus I (Bausch & Lomb, version 2.5), 10 Victus II (version 2.7SP2) 10 Catalys (Optimedica, version 2.15.13) and 10 manual. Scanning electron microscopy images were analyzed by calculating, the coefficient of variation (CoV) of pixilation along the capsular edge and homogeneity using Grey Level Co-Occurrence Matrix Analysis. A qualitative assessment for anomalies such as tags was also made.
Results
Regression analysis demonstrated manual capsulotomy CoV (101.6±0.6%, mean ± stdev) to be smoother than LenSxpre-soft-fit (105.2±1.2%, p <0.001), LenSxpost-soft-fit (102.7±1.4%; p = 0.04), VictusI (104.9±0.9%; p<0.001), Catalys (104 .5±1.6%; p<0.001) and VictusII (104.7±1.6%; p<0.001). All lasers (0.19 to 0.23; p<0.001) except for the LenSxpost-soft-fit (0.25±0.03, p=0.23) generated less homogenous capsulotomies than the manual technique (0.27±0.04). Tags were seen with LenSxpre-soft-fit (5 of 10), LenSxpost-soft-fit (1 of 10), Victus I (1 of 4) and VictusII (3 of 10). Misdirected pulses were seen with the Catalys (4 of 10) and Victus II (3 of 10), which use fluid interfaces.
Conclusion
Laser capsulotomies are approaching the smoothness of manual capsulorhexis. LenSx “post-soft-fit” demonstrated the least anomalies and the smallest difference to the manual method for the metrics of CoV and Homogeneity.