Stromal opacity after cross-linking.

To the Editor: The same stromal opacities found in the rabbit study by Salomão et al, 1 which appeared in the June 2011 issue of the Journal of Refractive Surgery, have been previously described as frequent but asymptom-atic complications after corneal cross-linking (CXL) in cases of steep corneas with keratometry values Ͼ54.00 diopters (D). 2-4 Raiskup et al 4 demonstrated that infusion with a hypo-osmolar ribofl avin solution preserved cor-neal tissue from stromal opacity development even in severe keratoconus cases with ultra-thin corneas, with no reduction of CXL effectiveness. The same phenomenon was observed by our group 5 in patients with postopera-tive LASIK ectasia. Corneas were plumped with hypo-tonic ribofl avin before CXL irradiation. 6 None of the patients developed any stromal opacity, despite their mean central corneal thickness (CCT) was less than the recognized safe cutoff depth of 400 μm. We found that the usual development of deep stromal opacities after CXL was related to a reduction of CCT at 1 month after CXL. Therefore, we decided to study CCT behavior during CXL in 45 corneas with keratoconus. Mean patient age was 27 years (range: 19 to 37 years). After 20 minutes of soaking with 0.1% ribofl avin solution (10 mg ribofl avin-5-phosphate in 20% dextran-T-500 10 mL), all corneas with CCT Ͻ350 μm were plumped to 400 μm with a hypo-osmolar solution (ribofl avin-5-phosphate) for a mean time of 8 minutes. Intraoperative CCT was measured with ultrasound pachymetry after epithelial removal (t1), after 20 minutes of ribofl avin infusion (t2), after corneal expansion (t3), and at the end of irradiation (t4) (Fig). Posterior stromal opacity formation was evaluated by anterior segment optical coherence tomography up to 12 months postoperatively. No signifi cant difference (P=.003) was found after epi-thelium removal. A mean CCT reduction of 102.11 μm was measured after 20 minutes of ribofl avin infusion (PϽ.00001). No signifi cant change in CCT was measured at the end of corneal expansion (P=.51) or irradiation (P=.016). Eyes were also divided into 3 groups of 15 eyes each according to preoperative CCT (group I Ͻ450 μm, group II between 450 and 500 μm, and group III Ͼ500 μm). The decrease in CCT after the fi rst 20 minutes of ribofl avin infusion was signifi cantly higher in group III, with a 28% CCT reduction. Groups I and II showed a 22% and 25% decrease in CCT, respectively. None of the patients …