Intra-corneal ring segments
Intra-corneal ring segments are crescent-shaped half rings made of polymethyl methacrylate (PMMA) that are inserted into channels deep in the corneal stroma (Figure 1). They were initially developed and used to correct myopia, as an alternative to excimer laser refractive procedures but they did not reach the levels of accuracy and predictability of excimer laser treatments. They now play a significant role in the management of keratoconus when patients become contact lens intolerant and when the central cornea is still clear. The goal of insertion of intra-corneal ring segments is not to eliminate corneal disease but to delay or eliminate the need for a corneal transplant by decreasing corneal abnormality and restoring contact lens tolerance. They may also have a use for correction of low myopia in forme fruste keratoconus.
The three most commonly used types are Intacs, Kerarings and Ferrara rings. Intacs (Addition Technology Inc, Chicago, IL, USA) were approved for use in low myopia by the Federal Drug Administration in 1999 and for keratoconus in 2004. They are hexagonal in cross section and have an arc length of 150 degrees with an inner diameter of 6.8 mm giving a 6.8 mm optical zone. The thickness ranges from 0.25 to 0.45 mm in 0.05 mm increments. Intacs SK have been developed more recently to work more effectively in cases of moderate to severe keratoconus. They have an elliptical design, give an optical zone of 6.0 mm and are available in two thicknesses (0.40 mm and 0.45 mm). By virtue of their position closer to the visual axis and corneal centre, Intacs SK segments exert a greater flattening effect on the central cornea. The elliptical design was incorporated to lessen halos and glare and enhance central corneal flattening. If the spherical equivalent refraction is -6.00 D or greater or the mean simulated K value is 55.00 D or more, then Intacs SK is the recommended choice. Kerarings (Mediphacos Ltda, Belo Horizonte, Brazil) and Ferrara rings (Ferrara Ophthalmics, Madrid, Spain) are available in variable arc lengths from 90 to 210 degrees, variable thicknesses from 0.15 to 0.35 mm and diameters to give an optical zone of either 5.0 mm or 6.0 mm. The smaller optical zone of these rings means they have a greater effect on the central cornea but may cause problems with glare and halos in eyes with larger pupils. They have a triangular cross section that theoretically induces a prismatic effect to attempt to reduce these photic phenomena.
It was generally thought that all these rings act as spacing elements to shorten the arc length of the anterior corneal surface, thus flattening and stretching the corneal apex, although the exact method of action is now debated. If this assumption is correct, inserting the intra-corneal ring segments into smaller and tighter corneal channels should enhance their effect; however, this has not been found to be the case, with channel size shown to have no effect on refractive outcome. Furthermore, if the apex flattening assumption is accepted, then it is likely that inserting two large intra-corneal ring segments would exert the largest flattening effect, but again this has not been found to be the case. In fact, significantly more improvements in uncorrected vision, visual acuity, K values and cylinder have been found with the use of a single inferior segment than double segments.
Successful implantation of intra-corneal ring segments depends on several factors, including correct placement and accurate depth of implantation. The intrastromal tunnels can be created manually with a spiral mechanical channel dissector or with a femtosecond laser. Complications with mechanical dissectors include epithelial defects, anterior and posterior perforation, asymmetric segment placement and extension of the incision toward the central visual axis or the limbus.[34, 35] The use of a femtosecond laser, such as the Intralase (Abbott Medical Optics, Inc, Santa Ana, CA, USA), allows the creation of tunnels of varying widths and more accurate channel depth throughout the length of the channel compared to manual dissection, with fewer surgical complications.[34, 36, 37] The femtosecond method may also be faster, easier and more comfortable for the patient. The complications related to femtosecond laser-assisted tunnel creation include incomplete channel formation, system malfunction, endothelial perforation and segment migration. When comparing the results of mechanical versus Intralase tunnels, there are no differences in visual or refractive results but there is a higher level of intra-operative complications with the mechanical method and possibly better results for primary spherical aberration, coma and other higher-order aberrations with the femtosecond laser method.[34, 39, 40]
The results of Intacs insertion are summarised in Tables 1 and 2. Most patients experience improvement in spherical and cylindrical refractive error, uncorrected vision, visual acuity and keratometric measurements. An important point to note is that significant refractive error may remain, and in fact insertion is not expected to correct more than 4.00 D of myopia in the best of circumstances. Results can be unpredictable and Intacs are unlikely to produce significant benefits when the maximum K values are over 55 D and with higher pre-operative sphere and cylinder. Loss of two or more lines of visual acuity occurs in 3.5 to 6.1 per cent of eyes.[37, 42, 43] The complete effect may take more than six months[43, 44] and there is no difference in results for different age groups.
Table 1. Summary of studies of visual outcome after insertion of Intacs
|Author||Year||Ring type||Eyes||Degree of keratoconus|| || || || |
|Zare and colleagues||2007||Intacs||30|| ||0.25||0.13||0.60||0.29|
|Ertan and Kamburoglu||2008||Intacs||306|| ||0.48||0.28||1.10||0.64|
|Sansanayudh and colleagues||2010||Intacs SK||10|| ||0.51||0.25||1.19||0.66|
|Khan and colleagues||2012||Intacs SK||31|| ||0.44||0.29||1.40||0.88|
Table 2. Summary of studies of refractive and keratometric outcomes after insertion of Intacs
|Author||Year||Ring type||Eyes||Degree of keratoconus|| ||Mean post-op SE (D)||Mean pre-op cyl (D)||Mean post-op cyl (D)|| || |
|Zare and colleagues||2007||Intacs||30|| ||-6.93||-3.23||-4.65||-3.90||49.84||47.90|
|Colin and Malet||2007||Intacs||100||Mild to severe||-6.93||-3.80||-4.62||-3.31||50.1||46.8|
|Ertan and Kamburoglu||2008||Intacs||306|| ||-6.04||-3.09||-4.11||-3.82||50.7||47.9|
|Sansanayudh and colleagues||2010||Intacs SK||10|| ||-8.08||-5.03||-5.05||-3.90||57.94||50.07|
|Khan and colleagues||2012||Intacs SK||31|| ||-6.57||-2.84|| || ||52.07||46.15|
Visual and refractive results are better when the inferior segment is thicker than the superior segment[49, 50] and under-correction can often be improved by exchanging the intra-corneal ring segments for thicker ones. Furthermore, there is now evidence that a single inferior ring has a better effect than double rings in paracentral and peripheral cones, whereas symmetrical double segments work best for central cones.[33, 50] The theory is that in paracentral and peripheral cones, placement of a segment superiorly in the flat topographic area exacerbates the topographic power asymmetry by causing further flattening superiorly.
Post-operatively white/yellow channel deposits are often seen near the implants. These deposits primarily consist of intrastromal lipid accumulations and keratocytes and are thought to arise in response to corneal injury. They are non-progressive and do not require intra-corneal ring segments explantation.[37, 42, 43, 51, 52] Haze may also be seen in the area of the incision. More serious complications include migration toward the wound, incision melt, dislocation into the anterior chamber, stromal thinning, epithelial breakdown, extrusion of the segments, vascularisation and infection, although these complications do not occur frequently.[36, 37, 46, 48, 53-58] Adverse post-operative visual symptoms may include fluctuating vision, glare, diplopia and halos. These symptoms improve over time and in a study of patient post-operative satisfaction, 76 per cent of patients reported an improvement in their quality of vision after the Intacs procedure. If complications or patient dissatisfaction require removal of the Intacs, it is a relatively straightforward procedure and the cornea returns to its original shape. Rates of removal vary from one to 19 per cent and histological changes appear to be entirely reversible.[37, 42, 43, 48, 59]
While the aim of implantation of intra-corneal ring segments is to defer corneal transplantation and improve both spectacle-corrected visual acuity as well as contact lens tolerance, the presence of the segments can sometimes complicate the contact lens fitting procedure by altering the natural shape of the cornea. Most authors report improved contact lens fit and tolerance post-operatively in most patients but not in all patients.[42, 48] Intra-corneal ring segments create an abnormal corneal profile with the flattest area in the mid-periphery and abnormal relative steepening of the peripheral cornea. As a result, rigid contact lenses may tend to centre over the intra-corneal segments and steepened periphery rather than the corneal apex. Success has been reported in post-operative lens fitting and approaches to managing this have included increasing the lens diameter to obtain adequate corneal coverage and improve centration, mini-scleral design contact lenses, which minimise poor lens centration by vaulting over the entire irregular corneal surface and piggyback lenses.
When comparing Intacs, Kerarings and Ferrara rings, no one type is overall significantly better than any other. A study comparing Intacs SK to Kerarings found no statistical difference between the two in any post-operative outcome, including uncorrected vision, visual acuity, refractive results and keratometry. In contrast, another study found that patients implanted with Kerarings had greater improvement in visual acuity and greater decrease in maximum keratometric values compared with Intacs at one year post-operatively. Kaya and colleagues compared the results of Intacs and Ferrara rings and found no significant difference between the two groups with regard to uncorrected vision, visual acuity and keratometry; however, mean higher-order aberrations decreased in the Intacs group and increased in the Ferrara rings group. In both groups, post-operative scotopic contrast sensitivity decreased significantly when glare was introduced but this decrease was greater with Ferrara rings than Intacs. The decrease was also worse with larger pupils suggesting that in eyes with larger pupils, intra-corneal ring segments with a larger inner diameter are preferable. Conversely, Intacs have a more limited effect in correcting astigmatism than Ferrara rings.
There is debate about whether Intacs halt the progression of keratoconus. Bedi and colleagues reported on 105 eyes with Intacs over five years and found that 93 per cent of eyes with pre-operative progressive keratoconus showed no post-operative progression. Potential explanations may lie in keratocyte activation and new collagen formation, which are known to manifest after Intacs implantation or alternatively, at least in some of the patients, natural cross-linking with increasing age. On the other hand, Colin in a study of 100 eyes found that around 20 per cent of eyes had a reduction in uncorrected vision and visual acuity between one year and two years after insertion of Intacs, as well as a reduction in corneal thickness, all thought to be due to progression of the keratoconus and changes in irregular astigmatism.
The answer to post-operative stability may lie in corneal collagen cross-linking, which may be done simultaneously with insertion of ring segments or at a later date and success has been reported with both approaches. It is not yet clear which approach is better. A study comparing the results in eyes receiving an inferior Intacs segment alone, compared to eyes that received an inferior Intacs segment with simultaneous corneal collagen cross-linking showed that the eyes with Intacs and collagen cross-linking had a significantly greater reduction in cylinder and keratometric values than the Intacs only group. This may be explained by a simple additive effect of the procedures or alternatively, the collagen cross-linking induced increase in biomechanical rigidity in the cornea may lead to increased rigidity locally or across the Intacs segment, producing further flattening. Ertan, Karacal and Kamburoglu confirmed that corneal collagen cross-linking may have an additive effect when performed subsequent to the insertion of ring segments. Kilic, Kamburoglu and Akinci evaluated simultaneous collagen cross-linking and intra-corneal ring segment implantation in 131 eyes. After creation of the Intacs channels with the Intralase, riboflavin solution was injected into the channels prior to insertion of the Intacs, as well as standard topical application of riboflavin prior to exposure to UVA light. The spherical and cylindrical refraction improved, there was significant improvement in uncorrected vision and visual acuity and there were no complications.