Automated keratomileusis in situ: clinical study of 142 eyes.

Purpose: To compare a modified automated lamellar keratoplasty (ALK) technique that uses two blades with the original technique, which uses one blade.

Setting: Buzard Eye Institute, Las Vegas, Nevada.

Methods: This study comprised 142 eyes of 85 patients who had ALK: 107 using the original one-blade technique (Group 1) and 35 eyes using the modified two-blade technique (Group 2). Mean follow-up was 11 months in Group 1 and 3 months in Group 2.

Results: Mean preoperative spherical equivalent was -8.93 diopters (D) +/- 2.80 (SD) in Group 1 and -8.33 +/- 2.80 D in Group 2. No patient had worse than 20/80 best corrected visual acuity preoperatively. Mean spherical equivalent at 1 month was -2.06 +/- 2.00 D in Group 1 and -0.79 +/- 1.20 D in Group 2 (P < .05). Mean postoperative spherical equivalent at last follow-up was -0.43 +/- 0.90 and -0.65 +/- 1.10 D, respectively. At last follow-up, 104 eyes (97%) in Group 1 and 34 (97%) in Group 2 had a spherical equivalent between +1.00 and -3.00 D, and 90 eyes (84%) in Group 1 and 27 (77%) in Group 2 had 20/40 uncorrected visual acuity. Seven eyes (7%) in Group 1 and 1 (3%) in Group 2 lost two or more lines of best corrected visual acuity at last follow-up. Three months after ALK (for equal comparison), 25 eyes (23%) in Group 1 and 3 (8%) in Group 2 required radial keratotomy; 20 (19%) and 2 (6%), respectively, needed ALK revision; and 45 eyes (42%) and 17 (47%) needed astigmatic keratotomy.

Conclusions: The results suggest that ALK is capable of impressive myopic corrections across a broad range of refractive error. The procedure is not accurate enough with a single microkeratome pass and requires enhancement procedures including recuts of the bed with the microkeratome and astigmatic and radial keratotomies. Using two blades to achieve an even cut gives better spherical equivalent and best corrected visual acuity results and lower enhancement rates.