Penetrating Keratoplasty Offers Better Results, Higher Rejection Risk Than DALK

Patients with advanced keratoconus can expect better visual and refractive outcomes after PK than DALK; however, risk of graft rejection is higher with PK.

Penetrating keratoplasty (PK) for the treatment of advanced keratoconus associated with superior visual acuity (VA) and refraction outcomes compared with deep anterior lamellar keratoplasty (DALK), but PK is more prone to graft rejection, according to a study published in the American Journal of Ophthalmology.

This retrospective case series reviewed records of patients (N=378; n=411 eyes) who underwent PK (n=218; mean age, 30.1±9.9; 121 men, 97 women) or DALK (n=193; 28.5±8.7 years; 119 men, 74 woman) for the treatment of advanced stage keratoconus at Shahid Beheshti University of Medical Sciences in Iran between 1994 and 2020. The outcomes of VA, refraction, keratometry, and complications were compared between surgical approaches.

The PK and DALK recipients had uncorrected distance VAs of 1.69±0.34 and 1.69±0.33 logMAR, corrected distance VA of 1.61±0.41 and 1.60±0.36 logMAR, and keratometry of 64.39±2.53 and 63.46±2.69 D, respectively.

The donor trephination size was similar for the PK (mean, 8.19 mm) and DALK (mean, 8.22 mm) groups, but the PK donor qualities were superior (excellent, 9.6%; very good, 79.8%) to the DALK donor qualities (excellent, 7.3%; very good, 55.4%; P <.001).

Both techniques of keratoplasty resulted in a significant improvement in visual and refractive parameters in patients with advanced keratoconus.

At follow-up, the PK group was associated with superior uncorrected distance VA (mean, 0.57 vs 0.75 logMAR; P =.003), corrected distance VA (mean, 0.18 vs 0.26 logMAR; P <.001), spherical equivalent refraction (mean, -2.89 vs -4.58 D; P <.001), and keratometry (mean, 45.32 vs 47.74 D; P <.001) compared with DALK, respectively.

Overall, 37.3% of DALK interventions were converted to PK due to extensive Descemet’s membrane tears.

Among the DALK recipients, 17.1% had visible folds in the donor-recipient interface but folds only persisted to the last follow-up among 4.7%, 3.6% had interface haze with clear overlaying graft, and 1.6% developed double anterior chambers.

Compared with PK, more DALK recipients had suture complications (72.0% vs 48.6%; P <.001) and graft overriding (5.7% vs 0.9%; P =.006), respectively.

Conversely, more PK eyes had graft rejection than DALK eyes (33.5% vs 19.7%; P =.002), rejection occurred more quickly (mean, 10.8 vs 12.8 months; P =.003), more PK recipients experienced multiple rejection events (15.6% vs 8.3%; P =.02), and rejection-free graft survival was lower at 180 months (61.1% vs 78.8%; P =.001), respectively. Stratified by type of rejection, only the PK group had endothelial graft rejection (P <.001) and subepithelial (P =.006) and stromal (P =.01) rejections were more frequent in the DALK group

Overall, graft failure occurred among 1.8% of the PK and 5.2% of the DALK cohorts (P =.06).

The limitations of this study included the retrospective design and lack of randomization.

“Both techniques of keratoplasty resulted in a significant improvement in visual and refractive parameters in patients with advanced keratoconus,” according to the study authors. “Our results showed that visual and refractive outcomes as well as the rate of suture-related complications were significantly better after PK than DALK; however, the safety profile including risk of graft rejection […] favored the DALK procedure.”


Feizi S, Javadi MA, Karimian F, Bayat K, Bineshfar N, Esfandiari H. Penetrating keratoplasty versus deep anterior lamellar keratoplasty for advanced stage of keratoconus. Am J Ophthalmol. 2022;S0002-9394(22)00447-0. doi:10.1016/j.ajo.2022.11.019