Clinical comparison of manual and laser-cut corneal tunnel for intrastromal air injection in femtosecond laser-assisted deep anterior lamellar keratoplasty (DALK)

Purpose The most crucial step in deep anterior lamellar keratoplasty (DALK) is to achieve a bare Descemet's membrane. We aimed to assess a new femtosecond laser software that allows for a precise intrastromal tunnel creation for big bubble (BB) air injection using a real-time microscope-integrated optical coherence tomography. Materials and methods A retrospective review of 61 eyes of 61 patients with keratoconus. Before introducing the new software update, DALK was performed using a partial-assisted femtosecond laser (partial-thickness circular cut followed by a lamellar cut) with manual intrastromal tunnel creation (partial FS-DALK group). After the software update, the femtosecond laser created the intrastromal tunnel (full FS-DALK group). Results In the full FS-DALK group, the BB's formation was significantly higher (64.3% vs. 36.4%, p =0.04), and surgery time was shorter (21.8 +/- 5.1 vs. 25.6 +/- 6.8 min, p= 0.025) than in the partial FS-DALK. Penetrating keratoplasty conversion rate (7.1% vs. 15.1%, p= 0.432) was similar between the groups. Both groups showed statistically significant improvement in uncorrected and corrected distance visual acuity, central corneal thickness, surface asymmetry, and regularity indices. Endothelial cell density loss at 12 and 18 months was lower in the full compared with the partial FS-DALK group (12 months:10.0% vs. 16; 18 months: 10.7 vs. 16.5%, p < 0.001 for both comparisons). Conclusions Creating the intrastromal guiding tunnel using FS laser for air injection resulted in a higher rate of BB formation, reduced long-term endothelial cell loss, and operating room time.

Automated summary

Creating an intrastromal tunnel using a new femtosecond laser software in deep anterior lamellar keratoplasty (DALK) resulted in a higher rate of big bubble formation, reduced endothelial cell loss, and shorter surgery time.