KEY CONCEPTS
- •
Before performing any refractive intervention in keratoconus, it is important to recognize the progressive nature of the disease and to treat appropriately.
- •
The only surgical treatment that has been shown to halt the progression of keratoconus is corneal collagen cross-linking.
- •
Surface laser ablation in patients with keratoconus is safe in nonprogressive early stages.
- •
Intrastromal corneal ring segment implantation is an option in patients with best corrected visual acuity equal to or worse than 20/40, achieving a significant effect in visual acuity improvement, correction of high cylinder, and reduction of corneal aberrations.
- •
The use of phakic intraocular lenses may be a useful tool for the correction of high ametropia in patients with keratoconus and can be combined with corneal regularization techniques to improve the result.
- •
In advanced keratoconus and intolerance to scleral lenses, keratoplasty is the treatment option. Depending on the depth of the corneal scar, deep anterior lamellar keratoplasty or penetrating keratoplasty can be performed.
Background
Surgical procedures play a fundamental role in the treatment of keratoconus. The final result in visual rehabilitation depends on an appropriate indication as well as surgical performance. Eye rubbing has proven to be a trigger factor for the disease, and warning the patient to stop eye rubbing is a nonsurgical measure that should always be present as a mainstay of treatment. However, in cases of progression despite cessation of eye rubbing or in patients who are not adherent to changes in lifestyle, corneal cross-linking (CXL) has proven to be the only procedure that halts the progression of ectasia. ,
The patient with keratoconus is a daily challenge in clinical practice, because of high ametropia and anisometropia, which makes correction difficult with conventional optical aids. Despite progress in their fitting, the use of contact lenses requires a learning curve and daily insertion and removal routines that often lead the patient to request a “more definitive” therapeutic option. Surface ablation for higher order aberrations and refractive error, intracorneal ring segments (ICRSs), phakic intraocular lenses (pIOLs), and keratoplasty, in advanced stages, are the procedures that are currently employed. This chapter will deal with the planning and the decision-making necessary when considering surgical intervention in keratoconus. After presenting the surgical solutions that are currently available, we will offer a flowchart that can assist the reader in choosing the proper strategy based on the patient’s needs and condition.
ASSESSING SEVERITY AND PROGRESSION
The management of a keratoconus patient varies according to disease severity. Therefore the first step is to determine the disease stage. Although the Amsler-Krumeich classification system is widely accepted, it does not combine tomographic indicators or biomechanical parameters. Newer classification systems combine other parameters. The ABCD grading system incorporates anterior and posterior curvature, the thinnest pachymetric values, and distance visual acuity. It consists of five stages (0 to 4). In Table 22.1 both classification systems are shown.
| Stage | Amsler-Krumeich Classification System | ABCD Grading System |
|---|---|---|
| 0 | ARC >7.25 mm (<46.5 D) PRC >5.9 mm Thinnest pachymetry >490 µm BCDVA ≥20/20 Scarring − | |
| 1 | Eccentric steepening Myopia, induced astigmatism, or both <5.00 D Mean central k readings <48 D | ARC >7.05 mm (<48.0 D) PRC >5.7 mm Thinnest pachymetry >450 µm BCDVA <20/20 Scarring −, +, ++ |
| 2 | Myopia, induced astigmatism, or both from 5.00 to 8.00 D Mean central k readings <53 D Absence of scarring Corneal thickness >400 µm | ARC >6.35 mm (<53.0 D) PRC >5.15 mm Thinnest pachymetry >400 µm BCDVA ≥20/40 Scarring −, +, ++ |
| 3 | Myopia, induced astigmatism, or both from 8.00 to 10.00 D Mean central k readings >53 D Absence of scarring Corneal thickness 300–400 µm | ARC >6.15 mm (<55.0 D) PRC >4.95 mm Thinnest pachymetry >300 µm BCDVA ≥20/100 Scarring −, +,++ |
| 4 | Refraction not measurable Mean central k readings >55 D Central corneal scarring Corneal thickness <200 µm | ARC <6.15 mm (>55.0 D) PRC <4.95 mm Thinnest pachymetry 450 µm BCDVA ≤20/400 Scarring −, +, ++ |
Many studies show that a criterion for ectasia progression is a change of at least 1 diopter (D) in the maximum and/or steeper and/or mean keratometry or corneal apex power. Another sign of progression is a 0.5-D change in manifest spherical equivalent (SE) and thinning of more than 2% or 30 µm in central corneal thickness from the baseline, in a 1-year period. The Global Consensus on Keratoconus and Ectatic Diseases defines ectasia progression by a consistent change in at least two of the following parameters: (1) steepening of the anterior corneal surface, (2) steepening of the posterior corneal surface, and (3) thinning and/or an increase in the rate of corneal thickness change. In the majority of the criteria used, progression of keratoconus usually leads to a deterioration in best spectacle-corrected visual acuity (BSCVA).
Surgical Procedures
To provide an evidence-based review of the different procedures that can be performed in keratoconus, the level of evidence of each of the currently used treatments was classified according to the Oxford Centre for Evidence-Based Medicine (OCEBM) levels of evidence, shown in Table 22.2 .
| Level | Description |
|---|---|
| 1a | SR (with homogeneity) of RCTs |
| 1b | Individual RCT (with narrow confidence interval) |
| 1c | All or none |
| 2a | SR (with homogeneity) of cohort studies |
| 2b | Individual cohort study (including low-quality RCT, e.g., <80% follow up) |
| 2c | “Outcomes” research; ecological studies |
| 3a | SR (with homogeneity) of case-control studies |
| 3b | Individual case-control study |
| 4 | Case series (and poor-quality cohort and case-control studies) |
| 5 | Expert opinion without explicit critical appraisal, or based on physiology, bench research or “first principles” |
CROSS-LINKING
Whereas nonsurgical refractive solutions for keratoconus do not affect the disease progression, CXL does appear to halt the ectatic process. CXL increases the stiffness and rigidity of the anterior cornea by creating photochemical cross-linking and covalent binding between collagen fibers. Although it is an effective treatment for halting the progression of keratoconus, the improvement in visual acuity after CXL is usually not enough to improve patient quality of life or dependency on contact lenses.
The evidence available to evaluate the effectiveness of CXL in halting the progression of keratoconus has been reviewed in meta-analyses (Level 1a). In a study published by Li et al. in 2015, the authors found a significant reduction in K max and K min in the analysis of subgroups with homogeneity. In 2017 Kobashi and Rong published another meta-analysis of randomized controlled trials (RCTs, Level 1a). They found a statistically significant difference in the improvement of BSCVA in the CXL group and no difference in the minimum pachymetry and cylindrical refraction between the control group and the CXL group. The evidence for recommending CXL in patients with progressive keratoconus (Level 1a) is demonstrated by the decrease in K min and K max , without evidence of a significant change in thinnest pachymetry and refractive cylinder, as well as evidence of improvement in BSCVA when compared with the control group.
In 2017 McAnena et al. published a meta-analysis of CXL in children under 18 years of age in which they included case-control studies (Level 3a). They found improvement in uncorrected distance visual acuity (UCDVA) in the standard CXL group at 6 months and 1 year but with no difference at 2 years. K max did not show differences at 6 months and 1 year but did show a statistically significant reduction at 2 years in the standard CXL group. Henriquez et al. conducted a prospective cohort study (Level 2b) of CXL epi-off versus epi-on in children under 18 years of age. Their recently published results at 5 years found a significant flattening in the K mean in the epi-off group and a progression rate of 9.37% in the epi-on group compared with 0% in the epi-off group.
A randomized controlled trial (Level 1b) was published by Eissa and Yassin in which they compared accelerated CXL and conventional CXL in patients with keratoconus aged 9 to 16 years. They found an improvement in both groups in the UCDVA and the BSCVA, as well as flattening of the K max , with no cases of progression in either of the groups at 36 months of follow up.
Our group performs CXL in all patients with documented progression and those with a high risk of progression, defined as suspicious keratoconus by tomography and young age. We can do this alone or in combination with the fitting of a contact lens or a surgical procedure (like wavefront or corneal-guided photorefractive keratectomy [PRK], ICRS, pIOL) according to our goal.
PHOTOREFRACTIVE KERATECTOMY
In PRK, the excimer laser is used to ablate tissue and change the profile of the central anterior part of the cornea. Although several studies have demonstrated that PRK in keratoconus patients may be considered a risk factor for progression of ectasia, , other studies have shown improvement in visual acuity, slowing of disease progression, and a decrease in high order aberrations.
Evidence for performing PRK in patients with suspected or diagnosed keratoconus as the only treatment is scarce; A study published by Guedj et al. retrospectively described a series of cases (Level 4) of patients who underwent PRK between 2004 and 2006, demonstrating stability and no progression of ectasia in the 62 eyes studied. The average age was 34.6 ± 15.1 years, mean SE 3.96 ± 3.05 D, and average thinnest point 522.14 ± 34.65 μm. Chelala et al. published a prospective cohort with a 5-year follow up of 78% (Level 4) in which they performed PRK in patients with grades 1 and 2 keratoconus according to the Amsler-Krumeich classification. The inclusion criteria were stable keratoconus, maximum ablation of 50 μm with a minimum residual stromal bed of 450 μm, and BCDVA better than or equal to 20/30. In this study, they reported that 2 eyes out of 119 treated (1.7%) presented with disease progression at 5 years of follow up.
This chapter’s authors recommend only performing PRK without CXL in patients with no associated risks for progression, older than 35 years, with mild keratoconus, without evidence of progression over at least 2 years of follow-up assessed with corneal topography and tomography, and with BSCVA equal or better than 20/30.
PRK AND CXL
The adjunctive use of CXL has been proposed to perform surface ablation safely on ectatic corneas while improving corneal stiffness and halting keratoconus progression. The objective of surface ablation in patients with keratoconus is to reduce corneal irregularity and/or, partially, refractive error with a maximum ablation of 50 μm in the center and minimal residual stromal bed between 350 and 400 μm. Between 2007 and 2012, different authors published case reports and case series of topography-guided surface ablation and CXL in patients with keratoconus both sequentially and simultaneously, demonstrating stability and refractive improvement up to 36 months of follow up. Non–topography-guided PRK has also been performed with success.
Kanellopoulos compared the effect of performing CXL and PRK in the same procedure versus performing PRK 6 months after CXL. He found that simultaneous treatment allows a single procedure, improving the patient’s visual rehabilitation time.
Gore et al. published a prospective case series (Level 4). They simultaneously performed CXL plus transepithelial PRK wavefront-guided only high order aberrations with average ablation at the apex of the cone of 35 ± 15 μm; This group was compared with a historical cohort of patients who underwent CXL alone. Both groups had 2 years of follow up, with three cases in each group that presented one keratometry parameter indicative of progression. Only one case in the CXL group presented two keratometry parameters indicative of progression. They found that the mean gain in BSCVA was just over one line, from 0.28 ± 0.21 preoperatively to 0.15 ± 0.14 ( P = 0.01) at 2 years. Clinically significant visual gains (≥2 lines of BSCVA) were more common after transepithelial PRK-CXL (30%) than after CXL alone (6%).
Besides the increased risk for ectasia mentioned earlier, when treating a patient with CXL-PRK combination, we also expose them to possible (but rare) CXL complications: postoperative infection/ulcer, corneal haze, endothelial damage, peripheral sterile infiltrates, herpes reactivation, and treatment failure. We prefer to perform corneal wavefront or ocular wavefront transepithelial PRK (after at least 6 months of CXL) or simultaneous with CXL focusing on correcting higher order aberration. We target improvement in BSCVA, so later we can correct the ametropia with spectacles or a pIOL. Fig. 22.1 shows a case of trans-PRK wavefront-guided treatment, 6 months after CXL.
Intracorneal Ring Segments (ICRS)
ICRS are miniature polymethylmethacrylate (PMMA) ring segments used to make the cornea’s surface more regular and make a correction in the patient refractive error. Because of the flattening effect of ICRS, this technique was first developed for myopia but is now a widely used treatment for keratoconus.
Corneal clarity and a minimal pachymetry of 400 µm in the insertion site are mandatory for ICRS and, therefore, it is commonly used in mild to moderate keratoconus. One or two ring segments are implanted into the stroma in a safe and reversible process. This procedure may improve best and uncorrected visual acuity (UCVA), contact lens tolerance, and high order corneal aberrations, and it can also delay the need for keratoplasty. There are two optional techniques for stromal tunnel creation: manual and femtosecond laser (FSL)–assisted. Using mechanical dissection is associated with higher rate of complications such as epithelial defects, depth asymmetry, and corneal perforation.
Creating the tunnel with an FSL is also possible, predetermining the segment depth and orientation with high precision. This technique is considered by many to be more precise and more predictable than the mechanical technique.
PATIENT SELECTION
Choosing the right patient for ICRS implantation is important. Criteria are not fixed, but most authors suggest the following guidelines ( Table 22.3 ): corneal thickness at the thinnest point greater than 400 µm, K max lower than 60 D, refractive error in SE lower than 6 D, BSCVA equal or worse than 20/40, and absence of cornea scarring. In cases of higher keratometry values, thin cornea, and corneal scarring, ICRS is contraindicated. Other contraindications include pregnancy, uncontrolled autoimmune disease, or excessive eye rubbing.
| ICRS | Recommended | Relative Contraindication | Contraindicated |
|---|---|---|---|
| Corneal thickness | >400 µm | 350–400 µm | <350 µm |
| K max | <60 D | 60–65 D | >65 D |
| Refractive error (SE) | <6 D | >6 D | |
| Corneal transparency | Clear | Paracentral scars | Central scars or opacities |
INTRACORNEAL RING CALCULATION
Several companies supply ICRS, and each offers a nomogram that determines the recommended ring segment or segments according to the patient’s refraction, pachymetry, and topography. The surgeon can then customize the nomogram based on their surgical experience. A few parameters influence the ICRS mechanism of action that we need to understand to choose the right segment or segments for a specific patient. The Barraquer thickness law states that a flattening effect is achieved when adding material to the periphery of the cornea or subtracting material from the center. Therefore ICRSs have a flattening effect that is directly proportional to the thickness of the segment and inversely proportional to the corneal diameter where it is implanted.
The ring has a flattening effect at its parallel axis and between ring segments ( Fig. 22.2 ) and a steepening effect at the perpendicular axis—the shorter the arc, the lower the flattening effect. Another important factor is the distance from the center or the diameter, a smaller diameter will produce a higher flatening effect.
In some cases, implanting one segment is enough, and in others, two may be needed. Alió et al. demonstrated that the topographic pattern could help in making that decision. For example, in symmetrical cones, implantation of two segments offers better results, whereas in inferior steepening, one implant may be enough. Recently Izquierdo et al. showed a great flattening effect (average of 6 D), with the use of long arc rings with no differences when comparing two different techniques (tunnel vs. pocket) for two different rings (340 degrees and complete circular of 360 degrees) with an average of one to two lines of BSCVA improvement.
Izquierdo et al. published a systematic review of 18 trials, including clinical trials, cohort studies, and case-control studies (Level 2b) with high heterogeneity. UCDVA improved by 0.23 ± 0.28 logMAR, and corrected distance visual acuity improved by 0.06 ± 0.21 logMAR. Sphere improved by 2.81 ± 1.54 D, cylinder improved by 1.49 ± 0.83 D, and mean keratometry improved by 3.41 ± 2.13 D within 12 months of follow up. ICRS implantation combined with CXL improved UCVA, refraction, and keratometry to a greater degree than ICRS implantation alone.
Many studies have demonstrated the benefit of combining CXL with ICRS. Hashemi et al. published a systematic review and meta-analysis of clinical studies (Level 2a) where they evaluated the appropriate sequence of combining ICRS with CXL. In this study, they compared three groups: CXL first, ICRS first, and simultaneous CXL + ICRS, and found that at 12 months of follow up there were no statistically significant differences in UCVA, BSCVA, and refractive cylinder. However, when comparing spherical refractive error, flat keratometry, and steep keratometry, the group with simultaneous CXL + ICRS was superior to the others.
We implant ICRS for patients with more advanced disease than those for which we use PRK. On average, we expect two to three lines of BSCVA improvement using ICRS. Commonly, for asymmetric cones, we use a single ICRS (arc of 160 degrees or less). We use the longer arc (240 degrees or more) for central cones with higher spherical components with significant spherical component correction. Fig. 22.3 shows a case of keratoconus at diagnosis and 1 month after ICRS implantation.
