Surgical correction of high astigmatism remains challenging. There are different surgical approaches, including arcuate and limbal astigmatic incisions and laser in situ keratomileusis, or intraocular procedures such as toric phakic intraocular lens implantation or refractive lens exchange with toric intraocular lens implantation. Small incision lenticule extraction (SMILE) has been used for the management of myopia and myopic astigmatism for many years.1 More recently, the safety and efficacy of SMILE have been investigated in hyperopia.2,3 A team at the Aarhus University Hospital in Denmark recently introduced a new concept for high astigmatism correction based on SMILE lenticule rotation.4 The spherocylindrical lenticule, created using a femtosecond laser, is rotated by 90° within the corneal stroma. This results in tissue addition in the steep meridian and tissue subtraction in the flat meridian in the periphery of the optical zone. Because the lenticule is not extracted and therefore no tissue is removed, the central corneal thickness is not significantly reduced. The impact on corneal stability should thus be lower compared to conventional SMILE surgery with the lenticule extraction. The lenticule can theoretically be rotated more than once in the desired position, making this procedure open to adjustment or even reversal to the preoperative state.
A 31-year-old woman came to our clinic requesting refractive surgery. Her left, non-dominant, amblyopic eye had a refraction of +2.00 −5.00 × 10° (spherical equivalent −0.50 diopters [D]), uncorrected distance visual acuity (UDVA) of 0.8 logMAR, and corrected distance visual acuity (CDVA) of 0.7 logMAR. She had slight horizontal nystagmus. Her uncorrected near visual acuity (UNVA) was worse than Jaeger 13 (> 1.0 logMAR), correcting to Jaeger 8 (0.5 logMAR). The limited potential of refractive surgery to improve her vision due to amblyopia was explained to the patient. She agreed to have the SMILE procedure and signed the informed consent.
SMILE surgery was performed using a 500-kHz VisuMax femtosecond laser system (Carl Zeiss Meditec AG) on the above-mentioned left eye with high mixed astigmatism. The lenticule parameters were set to −3.00 −3.00 × 10°, with an optical zone of 6.5 mm, transition zone of 0.1 mm, minimum lenticule thickness of 30 µm, and cap thickness of 130 µm with a 7-mm diameter. Two incisions of 2.5 and 3 mm width were created by the femtosecond laser system at the 90° and 0° meridians, respectively. The second incision was created as a spare, to be used in case a surgeon needed extra access for easier manipulation, but was not used. The incision at 90° was opened by a spatula. The anterior and posterior surfaces of the lenticule were separated from the stromal bed using a broad, rounded tip spatula. Following this, the lenticule and the cap were marked at the 90° axis by two overlapping radial marks created by forceps stained from the inside with corneal marking ink. Then, the lenticule edges were separated from the corneal stroma all around the lenticule circumference and the lenticule was rotated by 90° so both radial marks were 90° apart (Figure 1). The Stodulka ReLEx Smile double spatula (Rumex; cat. n. 20-207) was used for lenticule separation and the Stodulka ReLEx Smile forceps (Rumex; cat.n. 4-2012S) was used for marking and lenticule rotation. Finally, the lenticule was fully unfolded as evenly as possible by the spatula and by gently touching the overlying cornea.
(A) An initial radial mark at both the lenticule and the cap. (B) The lenticule mark perpendicular to the initial cap mark after the 90° lenticule rotation.
After the surgery, combined antibiotics (tobramycin 0.3%) and steroid (dexamethasone 0.1%) topical eye drops (Tobradex; Novartis Pharmaceuticals UK Ltd) were given to the patient and used three times a day until used up. Furthermore, at 1 month the patient was administered corticosteroid (0.1 mg/mL fluorometholone) topical eye drops (Flarex; Novartis Pharmaceuticals UK Ltd) that were used once a day until used up.
One week after the surgery, the refractive astigmatism was 0.00 D cylinder. A significant myopic shift resulted in distance refraction of −4.00 D (spherical equivalent). Her UDVA stayed the same at 0.8 logMAR and CDVA improved to 0.5 logMAR. Keratometric astigmatism was 0.75 D cylinder (measured using Tonoref III; Nidek). The patient noted slight UDVA improvement but a significant uncorrected reading improvement that was better than preoperative values (UNVA > 1.0 to 0.4 logMAR).
One month after surgery, refractive astigmatism remained 0.00 D cylinder and distance spherical equivalent refraction was −3.75 D. Her UDVA and CDVA remained unchanged. Keratometric astigmatism was 0.25 D cylinder. Subjectively, the patient reported a slight UDVA improvement but a significant uncorrected reading improvement and UNVA at 1 month was 0.3 logMAR.
Finally, at the 3- and 6-month assessments, refractive astigmatism remained stable at 0.00 D cylinder and distance spherical equivalent refraction was −3.25 D. Keratometric astigmatism was 0.50 and 0.75 D cylinder, respectively. Subjectively, the patient felt a further improvement at 3 months and subsequently at 6 months, when her UDVA was 0.6 logMAR and UNVA was 0.1 logMAR. Objective refraction outcomes are presented in Table 1.
Objective Refraction and Anterior and Posterior Corneal Curvature
The difference in total corneal power map between preoperatively and 3 and 6 months postoperatively showed a decrease of refractive power at the periphery (Figure 2). Pentacam (Oculus Optikgeräte GmbH) keratometry values are summarized in Table 1. Central corneal thickness changes were clinically insignificant. Central corneal thickness was 520 µm preoperatively and 532, 535, 527, and 510 µm at 1 week and 1, 3, and 6 months postoperatively, respectively. The cornea remained clear throughout the follow-up period. A corneal slit-lamp photograph shows the edges of the corneal pocket and lenticule 1 month after surgery (Figure 3). Standard anterior segment optical coherence tomography (Casia; Tomey) showed no abnormalities in cross-section of the corneal interface and lenticule at any of the assessment intervals.
Total corneal power map preoperatively and 3 and 6 months postoperatively.
Slit-lamp photograph shows the edges of the corneal pocket and lenticule at 1 month after surgery.
To our knowledge, this is the first clinical case of astigmatism correction using SMILE lenticule rotation. This case of an amblyopic eye demonstrates that the procedure is capable of correcting both corneal and refractive astigmatism, and the short-term result seems stable.
The lenticule was not taken out of the pocket, although it is possible to remove the lenticule and reinsert it in the desired 90° position, as originally described by Damgaard et al.4 It was thought that the surgeon should be able to assess the completeness of the lenticule and thus decrease the manipulation time and risk of rotating the lenticule the other way around or upside down when removed from the pocket.
A significant myopic shift was noted, which will need further evaluation and explanation in the future. This was also noticed in the feasibility study of ex vivo human corneas. It was speculated that the myopic shift is caused by the presence of Bowman's membrane and the anterior stroma that resist the front curvature alteration when stromal tissue is added to the periphery.4
The lenticule cylindrical power was set to 3.00 D cylinder, slightly more than half of the refractive astigmatism value. We also set a higher myopic component (−3.00 D sphere of the lenticule) to increase its thickness for easier manipulation during lenticule rotation. It was the surgeon's impression that an even thicker lenticule would be easier to manipulate and rotate.
It might be time to evaluate the safety and efficacy of this novel technique in a prospective study with a longer follow-up and evaluate whether the technique could be considered in general practice.
- Reinstein DZ, Archer TJ, Gobbe M. Small incision lenticule extraction (SMILE) history, fundamentals of a new refractive surgery technique and clinical outcomes. Eye Vis (Lond). 2014;1(1):3. doi:10.1186/s40662-014-0003-1 [CrossRef]
- Reinstein DZ, Pradhan KR, Carp GI, et al. Small incision lenticule extraction for hyperopia: 3-month refractive and visual outcomes. J Refract Surg. 2019;35(1):24–30. doi:10.3928/10815 97X-20181025-01 [CrossRef]
- Pradhan KR, Reinstein DZ, Carp GI, Archer TJ, Dhungana P. Small incision lenticule extraction (SMILE) for hyperopia: 12-month refractive and visual outcomes. J Refract Surg. 2019;35(7):442–450. doi:10.3928/1081597X-20190529-01 [CrossRef]
- Damgaard IB, Ivarsen A, Hjortdal J. Intrastromal lenticule rotation for treatment of astigmatism up to 10.00 diopters ex vivo in human corneas. J Refract Surg. 2019;35(7):451–458. doi:10.3 928/1081597X-20190618-02 [CrossRef]
Objective Refraction and Anterior and Posterior Corneal Curvature
|Parameter||Preoperative||1 Month Postoperative||3 Months Postoperative||6 Months Postoperative|
|Objective refractiona||+1.25 −6.00 × 9°||−3.75 −0.50 × 26°||−2.75 − 1.00 x 12°||−3.00 −0.75 × 14°|
|Anterior cornea curvatureb|
| K1||43.40 D @ 10°||47.30 D @ 32°||47.60 D @ 28°||46.90 D @ 14°|
| K2||48.40 D @ 100°||48.30 D @ 122°||48.50 D @ 118°||48.40 D @ 104°|
| Astigmatism||5.10 D||1.00 D||0.90 D||1.50 D|
|Posterior corneal curvatureb|
| K1||−6.00 D @ 11°||−6.10 D @ 12°||−6.00 D @ 11°||−6.00 D @ 15°|
| K2||−7.10 D @ 101°||−7.10 D @ 102°||−7.10 D @ 101°||−6.90 D @ 105°|
| Astigmatism||1.00 D||1.00 D||1.10 D||0.90 D|