Impact Of Descemet’S Membrane Perforation On Corneal Densitometry In Deep Anterior Lamellar Keratoplasty (Dalk): A Retrospective Study

Deep anterior lamellar keratoplasty (DALK) preserves corneal endothelium, reducing the risk of rejection. Descemet’s membrane perforation is a common intraoperative complication, with a reported incidence ranging from 4% to 40%, occurring as micro- or macroperforations, particularly during the learning curve. However, in many cases, conversion to penetrating keratoplasty is not required. Corneal densitometry serves as an objective method for assessing corneal transparency and the host-donor interface quality. This study aims to determine whether corneal densitometry differs between corneas  with and without micro- or macroperforations  during DALK

A retrospective study was conducted at a tertiary hospital in patients who underwent DALK over the past five years, with a minimum postoperative follow-up of one year. All surgeries were performed by the same surgeons using the big-bubble technique or manual dissection.

Intraoperative perforations were documented and classified as microperforations or macroperforations. Topographic and tomographic parameters, as well as corneal densitometry, were assessed using Pentacam HR (Oculus, Wetzlar, Germany), evaluating all layers (anterior, mid, posterior) and all regions (central, paracentral, peripheral). Best-corrected visual acuity (BCVA) and endothelial cell count (ECC) were measured using specular microscopy. Statistical analysis was performed using SPSS (IBM, version 19). Shapiro-Wilk was used for normality assessment, while Kruskal-Wallis and Mann-Whitney U tests were applied for group comparisons (p < 0.05).

A total of 18 patients were analyzed, with 4 patients from the non-perforation group excluded due to insufficient follow-up or retransplantation. The final cohort included 9 patients without perforation and 5 with perforation (3 microperforations, 2 macroperforations); 6 females and 8 males, mean age 59.71 years. No significant differences in corneal densitometry were observed between patients with and without intraoperative perforations (p > 0.05), nor among the three groups (no perforations, microperforations, macroperforations). No differences were found in any corneal layer or region. Median BCVA (logMAR) was 0,4 (non-perforation), 0,3 (microperforation), and 0,55 (macroperforation) (p = 1, two-group; p = 0.866, three-group).

Intraoperative perforations during DALK do not significantly affect corneal transparency measured by densitometry one year postoperatively, regardless of the corneal layer or region. Additionally, BCVA showed no significant differences between groups.