Screening criteria for SMILE are similar to those used for femto-flap-based LASIK, but there are also some limitations to the procedure that will reduce the number of potential candidates.
|After the femtosecond pass, surgeons use an instrument to break the tissue connections above the lenticule.|
(All images: Jesper Hjortdal, MD, PhD, and Anders Ivarsen, PhD)
One of SMILE’s current limitations is that the most reliable nomograms for it are primarily for myopic correction, and it’s only approved in the United States for the correction of myopia up to -8 D with no more than 0.5 D of astigmatism. Also, there’s a hardware limitation in that, in order to perform it, you need to use the Carl Zeiss Meditec Visumax femtosecond laser.
In terms of safety screening, we won’t perform SMILE on someone with a cornea thinner than 480 µm, or whose refraction hasn’t been stable for at least a year. Also, since we mainly perform SMILE on high myopes at our practice, we’re reluctant to perform it on patients younger than 25, because of the potential risk of forme fruste keratoconus that might manifest in the postop years. When screening for potential ectasia risks, we look for oblique astigmatism and use the Pentacam to look for a difference in power between the inferior and superior corneas. Also, we inquire whether anyone in the patient’s family has a history of keratoconus.
We take these precautions during screening because, even though SMILE is less invasive than LASIK and should, theoretically, result in a more stable cornea biomechanically, it does loosen the fibers in the anterior part of the cornea. This means the post-SMILE cornea may still be weakened compared to non-dissected cornea. Also, even though we’ve never experienced it in our practice, there have been reports about ectasia occurring after SMILE.1 However, it’s worth noting that, when one reviews these reports, it appears that the patients in question had some signs of subclinical keratoconus before the SMILE procedure, such as corneal asymmetry.
There are certain measures you can take to help the procedure go more smoothly, and to avoid potential difficulties.
|Surgeons say it’s easier if you free the posterior side of the lenticule after the anterior side.|
First, I’ve found it helps to give the patient a detailed explanation of what’s going to occur at each phase of the procedure. We do this both in our office in the days before the procedure as well as during the actual SMILE surgery. During the procedure, I’m constantly telling the patient to keep calm, to look at the blinking green light and to not follow the light if it appears to move. I keep her looking straight during the docking step. After docking, when the patient is moved under the laser, I then explain what’s going to happen in the next few seconds. This relaxes the patient and helps her to keep her gaze straight.
• Docking and suction. When the time comes to perform the procedure, we instill two drops of anesthesia: one three minutes before the case and another one minute before we insert the speculum. During this period, it’s important to evaluate the situation in a way similar to a LASIK protocol, to make sure you have good access to the eye and that the suction device on the laser won’t collide with the speculum. Move the patient’s head a few degrees to the side opposite the eye to be treated, then adjust the head so it’s exactly coaxial with the laser beam.
We then perform a mild cleaning of the corneal surface to ensure there’s no debris, such as mucus, in the tears. We instill a little BSS to ensure a nice wet surface, and then move the patient’s laser bed from the surgical microscope to beneath the laser. When this occurs, you’ll notice that the screens on the laser have changed, and you’ll then have another view through the microscope that’s down the laser path. You then elevate the patient bed toward the suction device and make sure to center the suction head along the visual axis, which you can accomplish by asking the patient to look at the blinking green light. The suction device then makes contact with the cornea. Make sure you’ve centered the procedure over the pupil and then elevate the patient bed a little more. Then, just before you have a full, curved applanation, activate the suction. Have a look at the pupil to make sure that everything is centered.
• The femtosecond step. At this point, it’s important to tell the patient to relax. He’ll be able to see with the operative eye because the suction isn’t that high with SMILE. Again, tell him the procedure is about to begin, so he shouldn’t move the eye or try to follow the light if it appears to move. Let him know that when the femtosecond is operating, his vision may become blurred due to small bubbles in the cornea and that it’s perfectly normal. Keep him calm and apprised of how much time remains in the procedure, which usually takes 20 to 25 seconds, depending on the spot pattern. Then, when the lenticule cut is complete and the side incision has been made, the laser will stop and suction will automatically decrease. Reassure the patient that things went well and that the most difficult part is over. You then move the patient bed down beneath your surgical microscope.
|With both sides free of tissue connections, remove the lenticule with forceps and make sure it’s complete, with no missing pieces. |
It’s worth noting that, in rare cases (0.7 percent in a study we performed),2 you can lose suction during the procedure. To try to avoid this, the most important thing is to have the correct amount of hydration on the eye—not too much or too little. Also, you can help prevent it by, again, talking to the patient and keeping him calm. If suction loss does occur, however, your response depends on how much tissue you’ve cut and where you are in the procedure, per Carl Zeiss:
Stage 1 (posterior lenticule cut is less than 10 percent complete): Restart the procedure;
Stage 2 (posterior lenticule cut is greater than 10 percent done): Switch to LASIK;
Stage 3 (lenticule side cut): Repeat the lenticule side cut, and decrease the lenticule diameter by 0.2 to 0.4 mm;
Stage 4 (anterior lenticule cut): Repeat the anterior lenticule cut; or
Stage 5 (anterior lenticule side cut): Repeat the anterior lenticule side cut and decrease the lenticule diameter by 0.2 to 0.4 mm.
• Completing the dissection. It’s now time to complete the dissection and remove the lenticule. Remind the patient to keep looking at the light, and tell him he may experience movement of the eye during the dissection.
First, use a Sinskey hook or other sharp instrument to identify the plane of the lenticule. Most surgeons do this by going to the left side of the incision and making sure that they have a plane above the lenticule before they start dissection. Then you take a dissector, either flat or curved—we prefer curved—and go to the extreme left to a place where you know you’re above the lenticule and dissect with smooth movements from left to right over the anterior surface of the lenticule toward the periphery of your cap cut. This will break all the tissue connections above the lenticule. Having done that, you go below the lenticule, entering through the right side of your incision, and do the same type of dissection from right to left toward the edge of the lenticule, breaking all the connections below. Make sure you’ve dissected the full planes above and below the lenticule. You then remove the lenticule with forceps and place it on top of the cornea to make sure that it’s complete.
• Two-incision technique. You can also choose to have two incisions: one at 12 o’clock and one at 6 o’clock. Then, if you dissect from the 12 o’clock position but realize you’ve started with the wrong plane before you completely dissect the other plane, you’ll have a second chance at success by using the incision down at 6 o’clock.
• Low vs. high myopia. The thicker the lenticule, the easier the dissection. So, in low myopia, you should typically add a little to the base thickness to make the lenticule easier to manipulate. In the typical SMILE case, there’s a 15-µm base thickness. So, in low myopes, you should increase that to 20 or 25 µm, giving more bulk to the lenticule. It’s like a plano ablation, so this extra tissue shouldn’t have any refractive effect.
• Notable differences from LASIK. The dissection in SMILE is more difficult than just lifting a LASIK flap. Also, it’s important to realize if you really can’t get under the lenticule properly, the most likely reason is that the lenticule is still attached to the cap. So, if you find yourself having to be very aggressive and digging down under the lenticule, your first thought should be that the lenticule may actually be stuck to the cap. Therefore, instead of just continuing a rough dissection where there’s no dissection plane, go back and see whether the lenticule is actually stuck to the cap, and try to see if you can get into the plane between the lenticule and the cap at the very edge.
If it’s really difficult or almost impossible to dissect, it’s better to acknowledge that the lenticule wasn’t properly cut and just stop—don’t make a mess of the middle of the cornea. It’s better to come back in a month or so when you’ve have time to see what happened, and consider doing another procedure such as PRK or LASIK.
• Closing the case. After you’ve removed the lenticule, flush the interface with BSS and, using either a sponge or the head used for docking, push on the cornea to squeeze out any remaining saline. Wait 30 seconds to a minute before removing the speculum to make sure that everything is in place. In some cases, you might notice that the operation caused a small epithelial defect or abrasion. If this is the case, place a bandage contact lens postop. You can typically remove the bandage lens on postop day one because the abrasion is usually healed.
We then instill Vigamox and a mild, pain-relieving NSAID drop. We prescribe a mixture of tobramycin and dexamethasone q.i.d. for a week, and then b.i.d. for another week.
Though most of our SMILE cases go smoothly, it helps to be vigilant for postop problems in order to nip them in the bud. Here are the things you should be looking for.
• Watch for inflammation. Postop, we’ll take a look to see if there is any debris or inflammation in the interface. Sometimes, you see small fibers like the kind seen after LASIK. If the debris is central, or you have a feeling that there’s some epithelium in the interface, then flush the interface. (This happens rarely.) Then look for tears around your incision, which typically only require observation.
• Visual results. Visual recovery with SMILE is a little slower than LASIK. If, for example, the average LASIK patient sees 20/20 the day after surgery, a typical SMILE patient may see 20/25 or a little less. However, there are also SMILE patients who have excellent visual acuity on the first postop day. Since it’s somewhat difficult to predict who will do well and who won’t, we tell patients to be prepared that their vision may not be perfectly clear on day one, but they’ll be fine after the first week.
In a study of 1,800 SMILE eyes at our practice, the average preoperative spherical equivalent refraction was -7.25 ±1.84 D. Postop, the average was -0.28 ±0.52 D, with a mean error of treatment of -0.15 ±0.5 D. At the three month follow-up visit, 86 percent of the cases had the same or improved best-corrected distance vision.2
In another study of SMILE that included 722 cases with an average preop error of -6.8 D, at three months 88 percent of eyes were within 0.5 D of the target refraction. Sixty-three percent saw 20/25 or better at day one, which improved to 83 percent at three months postop. Though the average gain in best-corrected vision between preop and three months postop was 0.07 (logMAR), 12 eyes (1.6 percent) lost two or more lines of best-corrected vision during that period.3 In a smaller study of 45 eyes of 35 patients with an average preop error of -7 D, 86 percent of the cases saw 20/20 or better. Although 32 percent of patients gained a line of vision, 2 percent lost a line.4
In our study, 1.5 percent (24 eyes) lost two or more lines of BCVA. We noted, though, that of these, significantly more were treated with laser setting 1 (lower spot energy and closer spacing) than setting 2 (higher spot energy and wider spacing). Also, eight of these eyes were within our first 100 eyes treated, so there may have been a learning curve at work. It’s important to note, however, that by an average of 18 months postop, BCVA was within a line of preop levels in all eyes.2
The most common complications in our study were epithelial abrasions at the incision site in 6 percent of cases (n: 114), difficulty removing the lenticule in 1.8 percent (n: 34), small tears at the incision site in 1.8 percent and, in some cases, a combination of those complications. The corneal cap was perforated during surgery in four eyes and in one eye a major tear from the edge of the incision nearly divided the cap in two. However, neither of these issues resulted in late visual symptoms or a loss of BCVA. Postoperatively, the main complications in our series were corneal haze (grade 0.5 to 1) in 7 percent of cases (n: 127) and a dry ocular surface in 4 percent (n: 75).2
Overall, there were perioperative complications in 10.8 percent of the SMILE surgeries, which might appear to be relatively high when compared to LASIK. However, 59 percent of them were minor abrasions and 17 percent were small tears at the incision and weren’t associated with any sequelae.
• Enhancements. Fortunately, enhancements are rare in our SMILE patients, because they’re usually high myopes who are happy if their result is even -0.75 D. That said, SMILE enhancements can be a challenge due to the nature of the procedure and its use of a very fine lenticule to cause refractive change. Usually, the amount of postop error is very small, so the lenticule you’d need to remove for the enhancement would be very thin and challenging to work with. Because of this, most surgeons will use an excimer procedure for an enhancement.
In our practice, if we have to enhance a SMILE, we perform transepithelial PRK with mitomycin-C. Alternately, you could change the SMILE cap into a flap and perform a LASIK enhancement, though this would depend on how thick you made the initial SMILE cap. The Visumax laser comes with a circle option that allows you to use it to cut down into the SMILE interface and create a flap. Then, you can use an excimer to do a LASIK. Though surface ablation enhancement works for us, the best procedure for enhancing a previous SMILE is still unclear.
With thousands of SMILE procedures under our belts now, we’ve learned a lot about the procedure and its idiosyncrasies. Though it does put some demands on the surgeon in terms of new instrumentation and the need to learn new techniques, we’ve found it an effective treatment for the myopes in our practice. REVIEW
Dr. Hjortdal is a clinical professor at Aarhus University, and is medical director of the Danish Cornea Bank. Aarhus University Hospital has specified research agreements with Carl Zeiss Meditec.
1. Mattila JS, Holopainen JM. Bilateral ectasia after femtosecond laser-assisted small incision lenticule extraction (SMILE). J Refract Surg 2016;1:32:7:497-500.
2. Ivarsen A, Asp S, Hjortdal J. Safety and complications of more than 1500 SMILE procedures. Ophthalmology 2014;121:4:822-8.
3. Hansen RS, Lyhne N, Grauslund J, Vestergaard AH. SMILE: Outcomes of 722 eyes treated for myopia and myopic astigmatism. Graefes Arch Clin Exp Ophthalmol 2016;254:2:399-405.
4. Yıldırım Y, Alagöz C, Demir A, et al. Long-term results of SMILE in High Myopia. Turk J Ophthalmol 2016;46:5:200-204.