Accelerated corneal crosslinking causes pseudoprogression in keratoconus within the first 6 weeks without affecting posterior corneal curvature.

PURPOSE

To evaluate the effectiveness of epithelium-off (epi-off) accelerated corneal crosslinking (A-CXL, 9 mW/cm, 10 min) in adult keratoconus (KC) patients.

METHODS

The study included 151 KC corneas (124 patients) after A-CXL. The parameters best corrected visual acuity (BCVA) and the tomographic readings (Pentacam HR, Oculus, Germany) were analysed at 24, 12 and 6 months preoperatively, prior to surgery; and 6 weeks, 6 months, 1, 2 and >2 years postoperatively. The demarcation line was assessed by anterior segment optical coherence tomography (Tomey SS-1000, CASIA 2 (Tomey, Nagoya, Japan)).

RESULTS

Comparing pre- to postoperative findings 6 weeks after A-CXL with paired t-test, the anterior steep (46.8 ± 4.0|47.1 ± 4.1), flat (50.2 ± 4.3|50.6 ± 4.6) and maximal keratometry (57.6 ± 6.8|58.3 ± 6.8) increased (p < 0.05), while the thinnest pachymetry decreased significantly (459 ± 39|444 ± 42, p < 0.05). Lateron, however, there was a decreasing anterior flat (1, 2 and >2 years; p < 0.0001), mean (1 year; p = 0.01 and 2 years; p = 0.03) and maximal keratometry (1, 2 and >2 years; p < 0.0001). The posterior corneal keratometry readings did not change significantly until >2 years after A-CXL (MANOVA; steep, p = 0.008; flat, p = 0.027; mean, p = 0.007). The mean depth of the demarcation line was 242 ± 62 µm (53.6%). The preoperative logMAR BCVA (0.35 ± 0.02) decreased 6 weeks after A-CXL (0.39 ± 0.03) followed by a continuous improvement until the latest follow-up (0.18 ± 0.04).

CONCLUSION

A-CXL constitutes a successful method for KC stabilization. Signs of KC progression occur within the first 6 weeks postoperatively ("pseudoprogression"), but this is not indicative of the long-term effect.