Transepithelial surface ablation (TESA) is a new method for performing transepithelial photorefractive keratectomy (t-PRK). It has recently gained impetus as a surface treatment for refractive errors in a single-step process without having to use any surgical instruments contacting the cornea.1-3 Not only do surface ablations have the advantages of maintaining corneal biomechanical strength and eliminating the risks associated with flap-related complications but also enable clinicians to optimise the way the epithelium is removed. Over the past decade, co-workers4-6 assessed the clinical outcomes following t-PRK largely on broad-beam based excimer laser systems. Lee et al4 evaluated that there were no differences in pain, corneal haze formation or uncorrected visual acuity and found t-PRK to yield over-corrections. On the Nidek excimer laser systems (Nidek, Tokyo, Japan) EC5000 and CXIII, Ghadhfan et al5 and Buzzonetti et al6 respectively showed better outcomes following t-PRK. This article discusses the principles involved and the recent experience gained treating myopic refractive errors on the Schwind Amaris excimer laser platform (Schwind eye-tech-solutions, Kleinostheim, Germany).
In essence, TESA achieves epithelial and stromal ablation in a single uninterrupted stage that consists of uniform precise epithelial removal followed by stromal ablation. This one-step no-touch technique goes some way to address the main aims of corneal refractive surgery: to reduce the risk of corneal dehydration, shorten the treatment duration and minimise mechanical manipulation of corneal tissue and hence deliver a precise refractive result.
How is epithelial ablation achieved?
The Amaris laser system applies an epithelial thickness ablation profile that mimics a hyperopic treatment of approximately +0.75D as a compensatory effect against the corneal epithelial thickness profile that is thicker in the periphery and thinner in the centre.7 High frequency digital ultrasonography has shown that in normal corneas the epithelial thickness averaged 53 microns centrally and 58 microns peripherally at 6mm.8 Therefore, assuming that there is a constant epithelial thickness profile, the algorithm of the t-PRK enables optimisation of the epithelial removal thereby avoiding myopic over-corrections to -0.75D. Within this context, it important to remind ourselves why does flat-depth phototherapeutic keratectomy (PTK) induce a hyperopic shift.7
Firstly, the central epithelial thickness is 10 microns less than the periphery and a flat-depth PTK will induce a -0.75D myopic correction and a corresponding hyperopic shift of +0.75D. Secondly, loss of laser energy occurs on the peripheral cornea due to the oblique incident angle and further distance for the laser to travel. This will increase the difference of ablation depth with respect to the centre and can induce +0.6D of hyperopic shift. Thirdly, central corneal tissue removal induces a shorter axial length and induces an insignificant hyperopia. The effect is minimal at <0.25D per 100 microns of tissue ablated.9
The resultant effect from the flat-depth PTK is a hyperopic shift in the region of +1.00D to +1.50D. In summary, the Schwind Amaris laser’s ORK-CAM software applies:
• An aspheric ablation profile
• A pre-defined epithelial thickness profile based on measurements from large population-based studies An epithelial profile that ablates 55µm at the centre and 65µm at the periphery
• Compensatory photoablation rates for the epithelium and stroma.
How is the epithelium prepared prior to ablation?
Transepithelial PRK/TESA is assumed to be a ‘no-touch’ technique and it follows that it is important to have the epithelium in an optimal condition immediately prior to the ablation. Fadlallah et al2 describe instilling one drop of ofloxacin and proparacaine three times five minutes apart. This is followed by scrubbing of the eyes, draping and keeping the lids apart with a speculum with suction.
Aslanides et al1 describe preparing the eyes with iodine and topical tetracaine 0.5 per cent instillation followed by draping and speculum insertion. A maximally wetted Merocel sponge applied ‘with three slow painting movements’ on the corneal surface to reduce the risk of uneven epithelial surface wetting. Similarly, epithelial preparation is described by Luger et al3 whereby a Merocel sponge dipped in balanced salt solution was maximally wetted and applied with ‘three slow painting-like movements’ on the corneal epithelium.
Our experience has been to warn patents not to rub the eyes when seen on the day of treatment and after prepping the eyes, instilling two drops of topical proxymetacaine 0.5 per cent and applying a speculum carefully not to touch the cornea while keeping the eyelids parted away from the cornea thereby avoiding irregular corneal wetting with the eyelid margins. The cornea is not wiped at any stage of the procedure.
It would appear that the choice of topical anaesthesia is also important although it is not clear from the literature as to the optimal method in preparing the corneal epithelium. Birchall et al10 showed there was less pain and reflex watering using topical proxymetacaine 0.5 per cent. Blojka et al11 showed on electron microscopy that topical amethocaine 0.5 per cent showed loss of epithelial microvilli and increased epithelial desquamation. It is possible that both of these factors can affect the smoothness and evenness of the corneal epithelium.
What is known about the visual outcomes following t-PRK/TESA?
Fadlallah et al2 found the visual outcomes comparable between t-PRK and the conventional alcohol-assisted PRK group. The postoperative mean sphere and mean astigmatism in the t-PRK group was -0.21 ± 0.61 and +0.43 ± 0.62 respectively. There was significantly less postoperative pain and rapid complete epithelial healing in the t-PRK group. Uncorrected distance visual acuity (UDVA) was not significantly different between the two groups at three months. Postoperative corneal haze can also occur after PRK and their study found at postoperative three months, 10 per cent of eyes in the t-PRK group had grade 1 haze compared to 26 per cent in the alcohol-assisted PRK group.
[CaptionComponent="783"]Aslanides et al1 found both t-PRK and the alcohol-assisted PRK groups to have safe outcomes. Their primary finding was that in the t-PRK group, patients had less early postoperative pain and photophobia on the third postoperative day with rapid epithelialisation. Patients in this group also had better vision by three Snellen lines on this day.
Corneal haze was significantly less at one, three and six months (0.2 versus 0.43) but by year one there was no haze present in both groups. At postoperative one-month, there was no significant difference in the unaided Snellen visual acuity (0.94 versus 0.97). Similarly, Fadlallah et al2 also found no significant difference in visual acuity between the two groups at the one-month and three-month postoperative periods.
Luger et al3 evaluated that between the t-PRK group and alcohol-assisted PRK, the postoperative mean spherical equivalent one year after surgery was +0.07D ± 0.23 and +0.01D ± 0.27 respectively and 97 per cent of eyes in both groups achieved an UDVA of 0.1 logMAR or better.
We recently presented our unpublished clinical results with our more refined version of t-PRK which we have now termed TESA at the 15th International Schwind Users Meeting 2014 in Vancouver, Canada12 and at the XXXII Congress of the ESCRS 2014 in London, UK.13 In our retrospective analysis of patients treated in our Optimax Laser Eye Clinics in UK by five surgeons, 399 eyes underwent single-step laser epithelial removal and stromal ablation using the transepithelial PRK nomogram of the Amaris laser’s ORK-CAM software (Schwind eye-tech-solutions, Kleinostheim, Germany). All eyes underwent ablation with an aberration-free algorithm with the Schwind Amaris at a repetition rate of 750Hz pulse with 1050Hz eye tracking. The laser ablation was centred on the pupillary axis. The intended refractive aim for all eyes was emmetropia and there were no re-treatments included. Adjunct mitomycin C was not used in any patient. The preoperative manifest spherical equivalent (SE) was -3.88 ± 1.47 dioptres (D) (range: -1.25 to -8.00D). At one month the postoperative manifest SE was reduced to -0.20 ± 0.53D (range: -4.88 to 1.88) and at three months it was -0.17 ± 0.18D (range: 0.88 to -1.25D). The manifest SE was within 0.50D and 1.00D of emmetropia in 89 per cent and 99 per cent of eyes (Figure 1), respectively. At three months, the preoperative manifest sphere was reduced from -3.58 ± 1.44D (range: -0.50 to -7.75D) to -0.05 ± 0.33D (range: +1.25 to -1.00D) and the preoperative manifest astigmatism was reduced from -0.60 ± 0.53 (range: 0 to -3.50D) to -0.25 ± 0.25D (range: 0 to -1.75D). UDVA of 20/25, 20/20 and 20/16 or better (Figure 2) was achieved in 20 per cent, 45 per cent and 24 per cent of 399 eyes, respectively. A gain of one or more lines was observed in 25 per cent of eyes (Figure 3). Postoperative corneal haze of =1.5 was observed in 2 per cent of eyes only. Figures 4 and 5 display the visual outcomes of attempted versus achieved spherical equivalent (SEQ) and astigmatism. Stability of SEQ is shown in Figure 6.
[CaptionComponent="784"][CaptionComponent="785"]
Potential limitations of TESA
The ORK-CAM software employs one standard epithelial ablation algorithm for all eyes and there are factors that can influence the achieved optic zone (OZ) to that intended.
If the actual corneal epithelial profile is thinner than the applied epithelial ablation profile, up to 14 microns of wasted tissue can be ablated but is unrelated to the refractive error.7 If the actual corneal epithelial profile is thicker than the epithelial ablation profile, this can result in a reduced OZ and subsequent under-correction for small refractive errors. If the corneal epithelial profile is steeper than that of the epithelial ablation profile, this can induce a hyperopic shift or a myopic shift if vice-versa.
Topographical variations in the 3D structure of the epithelium can also affect corneal curvature. Corneal epithelium is flatter nasally8 and may exhibit toricity along flat and steep meridians. This can lead to variations in the achieved OZ if there are significant differences between the epithelial ablation profile and underlying epithelial toricity. Decentred ablations causing astigmatism and higher order aberrations such as coma are potentially inducible due to geometric disparities between the thinnest corneal epithelial point and the ablation centred on the entrance pupil.
[CaptionComponent="786"]Future of TESA
TESA for mild to moderate simple myopia or compound myopic astigmatism yields very safe, predictable and satisfactory visual outcomes. Short-term follow-up also indicates refractive stability and the development of clinically significant corneal haze does not appear to be common, but longer-term surveillance may be required. However, current studies express limitations as these are based on small population sizes, have limited follow-up time period and the need for fellow eyes as controls. The corneal epithelium provides an optimal smoothing agent in relation to any underlying stromal topographical irregularity and the aspheric ablation profiles transmit this smoothness to the underlying stroma because customised epithelial removal would only de-mask stromal irregularity and affect the accuracy of the refractive component of the TESA ablation profile.
There may also be cytological benefits in not using ethanol in that lower levels of keratocyte cell apoptosis have been observed in rabbit models.14 Ethanol has been shown to increase apoptosis and generate proinflammatory cytokines that can predispose to chronic ocular surface disease such as dry eye and predispose to recurrent erosions.15 A notable finding by Oh et al15 was that ethanol decreased the number of cells expressing stem cell markers in corneal epithelium and reduced the number of viable cells.
The future looks promising for a treatment that uses the epithelium as the smoothing agent, shortens the treatment time and reduces the risk of corneal dehydration. When we consider that the visual outcomes, safety and efficacy are comparable to alcohol-assisted PRK, TESA is just as an effective choice for a surface treatment.
[CaptionComponent="787"]References
1 Aslanides IM, Padroni S, et al. Comparison of single-step reverse transepithelial all-surface laser ablation to alcohol assisted photorefractive keratectomy. Clinical Ophthalmology, 2012; 67:973-980.
2 Fadlallah A, Fahed D, et al. Transepithelial photorefractive keratectomy: Clinical results. J Cataract Refract Surg, 2011; 37:1852-1857.
3 Luger MHA, Ewering T, Mosquera SA. Consecutive myopia correction with transepithelial versus alcohol-asssited photorefractive keratectomy in contralateral eyes: One year results. J Cataract Refract Surg, 2012; 38:1414-1423.
4 Lee HK, Lee KS, Kim JK, et al. Epithelial healing and clinical outcomes in excimer laser photorefractive surgery following three epithelial removal techniques: mechanical, alcohol, and excimer laser. Am J Ophthalmol, 2005; 139:56-63.
5 Ghadhfan F, Al-Rajhi A, Wagoner MD. Laser in situ keratomileusis versus surface ablation: visual outcomes and complications. J Cataract Refract Surg, 2007; 33:2041-2048.
6 Buzzonetti L, Petrocelli G, Laborante et al. A new transepithelial photorefractive keratectomy mode using the NIDEK CXIII excimer laser. J Refract Surg, 2009; 25:S122-S124.
7 Mosquera SA, Awaad ST. Theoretical analyses of the refractive implications of transepithelial PRK ablations. Br J Ophthalmol, 2013; 0:1-7.
8 Reinstein DZ, Archer TJ, et al. Epithelial thickness in the normal cornea: three-dimensional display with Artemis very high frequency digital ultrasound. J Refract Surg, 2008; 24:571-581.
9 Mosquera SA, de Ortueta D. Analysis of optimised profiles for ‘aberration-free’ refractive surgery. Ophthalmic Physiol Opt, 2009; 29:535-548.
10 Birchall W, Kumar V. A comparative study of proxymetacaine-fluorescein and lignocaine-fluorescein use during applanation tonometry. Br J Ophthalmol, 2001; 85:477-479.
11 Blojka M, Kolar G, Vidensek J. Toxic side effects of local anaesthetics on the human cornea. Br J Ophthalmol, 1994; 78:386-389.
12 www.eye-tech-solutions.com/en/home/information-centre/events/
13 www.escrs.org/london2014/programme/free-papers-details.asp?id=21925
14 Kim W-J, Shah S, Wilson SE. Differences in keratocyte apoptosis following transepithelial and laser-scrape photorefractive keratectomy in rabbits. J Refract Surg, 1998; 14:526-533.
15 Oh JY1, Yu JM, Ko JH. Analysis of ethanol effects on corneal epithelium. Invest Ophthalmol Vis Sci, 2013; 54(6):3852-3856.
Amir Hamid is consultant ophthalmic surgeon at North East London Treatment Centre, King George Hospital in Ilford, and medical director at Optimax and Ultralase, UK, Arif Sokwala is head optometrist, Vaishali Patel is IOL coordinator and Sajjad Mughal is a refractive surgeon, all three for Optimax