Avoiding Intracorneal Ring Segment Complications

Complications with intracorneal ring segments (ICRS) are generally avoidable with correct keratoconus classification and proper surgical planning and technique, said Tiago Monteiro MD during an ESCRS eConnect Webinar.

Made of PMMA, ICRS are used to treat keratoconus by flattening the cornea and reducing higher-order aberrations such as coma and irregular astigmatism. Usually implanted at a stromal depth of 70–80% along the flattest topographic axis, the process now uses a femtosecond laser to create the tunnel, replacing earlier manual techniques.

Complications with ICRS can be functional or anatomical. Functional complications encompass loss of visual acuity or incorrect correction, inappropriate implant selection for the cornea type, incorrect procedure indication based on disease stage, and insufficient depth, centration, or symmetry of the intrastromal tunnel. Anatomical complications include improper implantation, extrusion, infection, inflammation, and neovascularisation.

Over the past 15 years, Professor José Alonso, Dr Monteiro, and their team have established a morphological classification system for ectasia phenotypes when identifying suitable ICRS specifications in keratoconus cases. This system uses corneal asphericity direction and assesses the relationship between topographic astigmatism and coma higher-order aberrations.¹

A 2021 study found refractive failure caused nearly 60% of ICRS complications. Dr Monteiro noted that for keratoconus stage 2 or less, issues often relate to inadequate optical zones, implant arc length, or thickness. For stages 3 or 4, deep anterior lamellar keratoplasty (DALK) is generally a better option.²

A 2018 study of 26 patients with unsuccessful ICRS outcomes found that exchanging ICRS for ones adjusted in thickness or arc length led to significant improvements in corrected distance visual acuity, refractive and topographic cylinder, and coma. He noted that in cases where the original topography is available and the tunnel was created correctly, the ICRS exchange can be performed in a single procedure. However, if the topography is not available and/or the tunnel was poorly centred or at incorrect depth, the explantation and implantation should be performed six months apart.³

Dr Monteiro explained how femtosecond lasers have greatly increased the precision of intrastromal tunnel creation for ICRS implantation. In a study he conducted with colleagues comparing manual and femtosecond laser techniques, only 15.40% of eyes in the manual tunnel group achieved tunnel depths within 10 microns of the intended measurement, whereas 67.92% of eyes in the femtosecond laser group met this criterion.⁴

“The implant depth is very important because if the implant is very shallow or very deep, we will have topographic overcorrection and higher postoperative astigmatism,” Dr Monteiro said. “And we are not going to achieve the refractive outcome we wanted.”

The precision of femtosecond lasers has also reduced the incidence of anatomic complications. A study comparing manual and femtosecond laser techniques found the manual approach delivered a 1.5% corneal perforation rate and a 5.7% ICRS extrusion rate, compared to the respective 0.9% and 0.0% rates with the femtosecond laser technique. The learning curve was a more important factor with manual technique as the complication rate was significantly lower during the first three years of a surgeon’s experience.⁵

The seminar is available at here.

Tiago Monteiro MD is based at Hospital de Braga, Braga, Portugal; Escola de Medicina Universidade do Minho, Braga, Portugal; and Hospital CUF Porto, Porto, Portugal. monteiro.tiago.pt@gmail.com

1. Alfonso JF, et al. Int J of Keratoconus Ectatic Corneal Dis, 2025; 11(1): 1–8. doi:https://doi.org/10.5005/jp-journals-10025-1207.

2. D’Oria F, et al. Am J Ophthalmol, 2021 Feb; 222: 351–358.

3. Monteiro T, et al. Cornea, 2018 Feb; 37(2): 182–188. doi:https://doi.org/10.1097/ICO.0000000000001449.

4. Monteiro T, et al. J Refract Surg, 2018 Mar 1; 34(3): 188–194. doi:https://doi.org/10.3928/1081597X-20180108-01.

5. Monteiro T, et al. Curr Eye Res, 2019 Dec; 44(12): 1291–1298.