The femtosecond laser was introduced into ophthalmology a decade ago and has become a popular alternative to the mechanical microkeratome for flap creation in laser in situ keratomileusis (LASIK).1 Although it has been considered a safer and more precise method for flap creation, there are several possible complications uniquely specific to the femtosecond laser. These include mechanical, inflammatory, and optical complications such as flap tears, gas-bubble breakthrough, incomplete dissection and flap lifting, macular hemorrhage, transient light sensitivity syndrome, diffuse lamellar keratitis, interface haze and light scattering, rainbow glare, anterior chamber bubbles, and inability of pupil tracking in Asian eyes.2–9 Although these potential complications are often less severe than those seen with mechanical microkeratomes, they present a challenge to the surgeon in terms of their proper management. One of the most challenging is that of gas-bubble breakthrough because the bubbles can interfere with the flap creation.3,10 Although all instances of gas-bubble breakthrough do not preclude proper dissection and lifting of the flap, it is important to determine the appropriate management for each case so the visual outcome will not be compromised. We report a case of femtosecond laser LASIK in which bilateral gas-bubble breakthrough occurred during flap creation, requiring a different form of management in each eye.
A 50-year-old woman, left eye dominant, was referred for laser vision correction. Her ocular history was significant for dry eyes since the age of 16 and anterior basement membrane dystrophy. Evidence supporting the diagnosis of anterior basement membrane dystrophy was noted preoperatively and postoperatively by the presence of geographic and fingerprint lines within the superior paracentral area of both corneas. There was no history of recurrent erosions. The patient was asymptomatic at the time of presentation. The corrected distance visual acuity (CDVA) was 20/20 in both eyes with a manifest refraction of −7.00 +1.00 × 142 in the right eye and −7.00 +1.75 × 53 in the left eye. The manifest refraction was stable over a year.
Slitlamp examination showed mild inferior punctate epithelial erosions and mild superior anterior basement membrane dystrophy in both eyes. Preoperative corneal topography revealed regular astigmatism with mean keratometric readings of 45.57 diopters (D) and 46.43 D in the right eye and left eye, respectively. Preoperative ultrasound pachymetry was 568 μm and 568 μm, respectively. The patient was started on cyclosporine 0.05% twice a day and artificial tears 4 times a day.
On a subsequent preoperative visit after the patient had been using the artificial tears and topical cyclosporine, there were no further signs of punctate epithelial erosions, although a mild topographic irregularity due to anterior basement membrane dystrophy was still present. At no preoperative visit did the patient report symptoms of epithelial erosions.
The treatment options were discussed with the patient, and photorefractive keratectomy (PRK) with topical mitomycin-C (MMC) in both eyes was advised, targeting monovision in the right eye. The patient requested LASIK over PRK because of her desire for faster recovery time and on mutual agreement, an informed consent was obtained for femtosecond LASIK.
A flap was created with the Wavelight FS-200 laser (Alcon Laboratories, Inc.) with an intended flap thickness of 110 μm and a diameter of 8.7 mm in both eyes. During flap creation, a superior bubble breakthrough appeared beneath the applanation interface in both eyes, extending in front of the advancing laser edge in the right eye (Figure 1) and behind the advancing laser edge in the left eye (Figure 2). The decision was to abort the flap separation and excimer laser ablation portion of the procedure on that day. The patient was counseled concerning the reason for the delay and sent home with topical antibiotics and steroids.
Four days later, the patient returned with a CDVA of 20/20+ in each eye and a stable refraction. The eyes were not inflamed, and the patient was asymptomatic.
The following day, a PRK with MMC was performed in the right eye with the Allegretto Wave Eye-Q excimer laser system (Alcon Laboratories, Inc.). Following the laser ablation, a fine 4.0 mm circular demarcation line was noted centrally, outlining the residual unablated flap edge around the area of deep central ablation. In the left eye, the flap was easily lifted and the previously planned LASIK ablation was performed.
On the first postoperative day, the patient experienced mild light sensitivity and blurry vision in both eyes with an uncorrected distance visual acuity (UDVA) of 20/40+1 in the right eye (PRK) and 20/20−3 in the left eye (LASIK). At 1 week, the patient was very happy and had no complaints; the manifest refraction was plano with a UDVA of 20/20 in the right eye and 20/15−1 in the left eye. At 1 month, the patient was pleased with both distance and intermediate/near vision; the UDVA was 20/30−1 ([Jaeger] J1) in the right eye (monovision) and 20/20 (J3) in the left eye (distance). The manifest refraction was −0.75 D and +0.50 D, respectively.
Although gas-bubble breakthrough is not a newly reported complication of femtosecond laser LASIK, it is a controversial one with regard to its proper management when encountered. Most surgeons will acknowledge its presence, but the significance of its effect on the laser disruption process during flap creation is not well differentiated. Previous reports of gas-bubble breakthrough mostly report an attempt to lift the flap, which results in resistance due to incomplete flap creation.3,10 Our report outlines the diagnostic value of visualizing the extent of gas-bubble formation beneath the interface plate in light of the leading edge of the femtosecond laser pulsing. In the right eye, the gas bubble was noted to advance in front of the leading edge of the laser pulsing (Figure 1), whereas in the left eye it was noted to remain behind the leading edge of the laser (Figure 2). Although the surgeon may notice the bubble after the flap creation is complete, this alone gives no information about the completion of the laser disruption beneath the bubble. The essential information is the temporal dynamics of the bubble during the laser advancement, which requires astute observation throughout the flap creation process. Even careful inspection of the figures showing the completed flap does not conclusively show the leading edge relative to the bubble location except, perhaps, for the apparent demarcation line in Figure 2 (as demonstrated by the arrows).
Another important observation to consider in our case is the presence of anterior basement membrane dystrophy, which can lead to a thicker epithelium than is typically found in the average LASIK patient. Although we did not perform anterior segment optical coherence tomography (OCT) on this patient, the technology now exists to measure the profile of epithelial thickness,11 which would have been beneficial in planning for a thicker flap setting than customary. Usually, a surgeon would consider a thicker flap setting in the presence of an anterior stromal scar so the epithelial plug within this scar does not become the site of gas-bubble breakthrough. Although the scar may be a more obvious target to avoid, the presence of anterior basement membrane dystrophy as a risk for gas-bubble breakthrough has not been reported or considered.
In the past, anterior basement membrane dystrophy would be considered a relative contraindication for LASIK, as the mechanical microkeratome would often slough off the weakly adherent epithelium, thus complicating the healing process. With the advent of femtosecond lasers for flap creation, the patient with anterior basement membrane dystrophy and no history of recurrent erosion might be considered for femtosecond laser flap without risk for epithelial sloughing. In our patient, we initially offered a more conservative management of her refractive error by avoiding the creation of a flap with PRK. When she strongly requested LASIK to reduce the visual recovery time, it was not improper to grant her request, but compensation for the anterior basement membrane dystrophy should have been considered when setting the flap thickness. It is quite likely that a 20 or 30 μm deeper flap thickness setting might have prevented this complication.
Once the gas-bubble breakthrough was experienced and the temporal dynamics of the gas expansion noted in reference to the leading edge of the laser treatment, the next step was to consider what management would be best. Knowing the gas expansion in front of the leading edge of the laser would block the laser disruption in that area, any attempt to lift this flap would be unadvisable. Although we chose to abort the excimer laser ablation in both eyes, it is possible that we could have proceeded with LASIK in the left eye in which the bubble expansion remained behind the advancing laser pulses. Because a refractive correction in 1 eye and not the other would create a high degree of anisometropia, this option was not considered. Performing PRK in the right eye immediately after the femtosecond laser flap creation was also not selected as this could complicate the healing process and lead to anterior stromal scarring (personal experience). Although we elected to abort the procedure on that day, Chang and Lau12 report the use of a microkeratome immediately after femtosecond laser flap creation with a vertical gas breakthrough without adverse sequelae. Because the use of a microkeratome is much less common and no longer available in many modern-day LASIK practices such as ours, we did not consider this alternative.
After we discussed the safety concerns with our patient, she willingly agreed to defer the treatment for 5 days and have PRK with MMC in the right eye and completion of the LASIK procedure in the left eye. The decision to postpone the completion of the surgery by 5 days was to allow sufficient time for the immediate inflammation of the laser disruption to subside according to a schedule convenient for both patient and surgeon. A longer or shorter period could have been safely considered with sufficient postoperative treatment by topical steroids. Although we proceeded with the second eye after noting the gas-bubble breakthrough in the first, it would be advisable to stop after the first eye to avoid the possibility of a bilateral occurrence of this complication. Unfortunately, this is what happened, but this unique scenario gave us the opportunity to demonstrate the proper management specific to each eye.
In summary, the key point of this article is diagnosis of the blocked laser disruption from the gas-bubble breakthrough by observing its expansion in reference to the leading edge of the laser pulsing. Because gas-breakthrough is a rare complication, most surgeons may not notice the dynamic relationship of the bubble expansion and laser pulsing and hence will miss this important diagnostic information. Simply lifting the flap and proceeding with LASIK is preferable, but if the laser disruption is blocked over a significant area, it will result in an undesirable complication that would be best avoided.
The other important information in this article is gas-bubble breakthrough in the presence of anterior basement membrane dystrophy. Because the thickness of the epithelium can vary with anterior basement membrane dystrophy, corneal OCT should be considered to assess the thickness profile and, if necessary, to use a deeper than typical flap thickness setting. With proper intraoperative and preoperative diagnosis, the outcome of gas-bubble breakthrough can be managed without visual compromise.
1. Nordan LT, Slade SG, Baker RN, Suarez C, Juhasz T, Kurtz R. Femtosecond laser flap creation for laser in situ keratomileusis: six-month follow-up of initial U.S. clinical series. J Refract Surg
2. Chang JSM. Complications of sub-Bowman’s keratomileusis with a femtosecond laser in 3009 eyes. J Refract Surgery
3. Seider MI, Ide T, Kymionis GD, Culbertson WW, O’Brien TP, Yoo SH. Epithelial breakthrough during IntraLase flap creation for laser in situ keratomileusis. J Cataract Refract Surg
4. Stonecipher KG, Dishler JG, Ignacio TS, Binder PS. Transient light sensitivity after femtosecond laser flap creation: clinical findings and management. J Refract Surg
5. de Medeiros FW, Kaur H, Agrawal V, Chaurasia SS, Hammel J, Dupps WJ Jr, Wilson SE. Effect of femtosecond laser energy level on corneal stromal cell death and inflammation. J Refract Surg
6. Rocha KM, Kagan R, Smith SD, Krueger RR. Thresholds for interface haze formation after thin-flap femtosecond laser in situ keratomileusis for myopia. Am J Ophthalmol
7. Krueger RR, Thornton IL, Xu M, Bor Z, van den Berg TJTP. Rainbow glare as an optical side effect of IntraLASIK. Ophthalmology
8. Lifshitz T, Levy J, Klemperer I, Levinger S. Anterior chamber gas bubbles after corneal flap creation with a femtosecond laser. J Refract Surg
9. Ramos-Esteban JC, Bamba S, Krueger RR. Tracking difficulties after femtosecond laser flap creation with the LADARVision excimer laser system. J Refract Surg
10. Shah SA, Stark WJ. Mechanical penetration of a femtosecond laser-created laser-assisted in situ keratomileusis flap. Cornea
11. Rocha KM, Perez-Straziota E, Stulting RD, Randleman JB. SD-OCT analysis of regional epithelial thickness profiles in keratoconus, postoperative corneal ectasia, and normal eyes. J Refract Surg
12. Chang JSM, Lau S. Intraoperative flap re-cut after vertical gas breakthrough during femtosecond laser keratectomy. J Cataract Refract Surg