CASE REPORT
Corneal perforation by an astigmatic keratotomy performed with an optical coherence tomography–guided femtosecond laser Daniel G. Cherfan, MD, Samir A. Melki, MD, PhD
We present a case of corneal perforation secondary to an intrastromal astigmatic keratotomy performed with an optical coherence tomography–guided femtosecond laser. The keratotomy was concomitant with cataract surgery and resulted in a flat anterior chamber prior to the start of lens extraction. Interrupted nylon sutures were placed to seal the keratotomy prior to phacoemulsification. Escape of cavitation bubbles into the anterior chamber or the liquid interface can alert the surgeon to the possibility of unintended perforation of the endothelium or the epithelium, respectively. Financial Disclosure: Neither author has a financial or proprietary interest in any material or method mentioned. J Cataract Refract Surg 2014; 40:1224–1227 Q 2014 ASCRS and ESCRS Online Video
Corneal astigmatism can be surgically reduced or corrected through several methods, which can be incisional or ablative. Incisional techniques concurrent with cataract surgery include arcuate keratotomy,1 limbal relaxing incision,2 and opposite clear corneal incision.3 Arcuate keratotomies can be performed either manually using a keratome or a diamond knife or using a femtosecond laser.2,4,5 Astigmatic keratotomy (AK) with a femtosecond laser is becoming a popular choice in clinical practice due to its presumed advantages over manual methods.2 Ocular coherence tomography (OCT)– guided femtosecond lasers offer more detailed corneal imaging, which may improve accuracy and precision of the corneal incisions. We present a case of corneal perforation following creation of
Submitted: November 6, 2013. Final revision submitted: February 14, 2014. Accepted: February 16, 2014. From the Massachusetts Eye and Ear Infirmary and the Boston Eye Group (Melki), Boston, Massachusetts, USA; the Royal College of Surgeons (Cherfan), Dublin, Ireland. Corresponding author: Samir A. Melki, MD, PhD, 1101 Beacon Street 6W, Brookline, Massachusetts 02446, USA. E-mail: samir_ [email protected].
1224
Q 2014 ASCRS and ESCRS Published by Elsevier Inc.
an intrastromal AK incision using an OCT-guided femtosecond laser–assisted cataract extraction. CASE REPORT A 50-year-old woman with a nuclear sclerotic cataract in both eyes was scheduled for femtosecond laser–assisted cataract surgery in her left eye due to difficulty driving at night. The corrected distance visual acuity at the time of surgery was 20/40 in the left eye with a refraction of –1.75 1.25 62. The ocular examination was normal except for a lenticular cataract graded as 1C. Automated keratometry revealed 0.5 diopter (D) of astigmatism with a steep meridian of 44.5 @ 137. Scheimpflug pachymetry (Pentacam, Oculus Surgical, Inc.) showed 0.7 D of regular astigmatism with a steep meridian of 44.70 @ 142. Keratometry from partial coherence interferometry (IOLMaster, Carl Zeiss Meditec AG) revealed 0.59 D of astigmatism with a steep meridian of 44.88 @ 126. The pachymetry map on Scheimpflug pachymetry showed a corneal thickness of approximately 600 mm in the area of the planned keratotomy. A plan for cataract extraction with AK at the 159-degree meridian was made using the LRI calculator (Abbott Medical Optics, Inc.), taking into account 0.5 D of astigmatism induced by the cataract incision placed at the 110-degree meridian. An OCT-guided femtosecond laser (Catalys, Optimedica Corp.) was used to create the main corneal incision, 2 paracentesis incisions, as well as the AK. The Catalys laser uses a liquid interface that docks the patient to the system and is designed to provide a clear optical path for real-time video, OCT imaging, and laser treatment. This is performed by filling in corneal surface irregularities with liquid and creating a wide field of view, enabling all cuts to be 0886-3350/$ - see front matter http://dx.doi.org/10.1016/j.jcrs.2014.04.021
CASE REPORT: CORNEAL PERFORATION WITH OCT-GUIDED FEMTOSECOND LASER
performed in a single dock. The OCT generates an image that is automatically registered and analyzed using the Catalys Integral Guidance interface to create safety zones in both the cornea and the posterior aspect of the lens capsule. The safety zones are used to design the treatment by focusing the femtosecond laser based on the patient's ocular anatomy. The intraoperative axial and sagittal OCT images of the cornea are seen in Figure 1, A and B. The femtosecond laser was also used for the capsulotomy and the lens fragmentation steps. Once the safety zones were identified by OCT, the incisions were made in the following sequence: anterior capsulorhexis, lens fragmentation, AK, and penetrating incisions. Examination of the OCT imaging revealed a displacement of the lines identifying the epithelial and endothelial layers (Figure 1, C). This was presumably due to patient movement after the initial OCT scan and prior to initiation of the keratotomy. The AK incision was programmed to be intrastromal. Table 1 shows the laser settings used, indicating that the arcuate incision was planned to extend from 20% below
1225
the epithelium to 20% above the endothelium. Air bubbles were seen emanating from the anterior and posterior aspects of the arcuate incision (Figure 2, A). The procedure was not aborted as the bubbles were not noticed intraoperatively. The patient was then moved to the operating microscope, where the anterior chamber was noted to be flat. The cataract corneal incision was not patent. A keratome was used to enter the anterior chamber. The latter was reformed with an ophthalmic viscosurgical device. The AK was inspected with a cyclodialysis spatula, which revealed a fullthickness incision along the full length of the incision (Figure 3). Two interrupted 10-0 nylon sutures were placed to seal the wound (Video 1, available at: http://jcrsjournal. org). The cataract was extracted by phacoemulsification as planned with no further complications. Postoperatively, the sutures were removed at weeks 5 and 6. At the 3-month follow-up visit, the uncorrected visual acuity was 20/50. The keratometry was 44.25 in both meridians, and Scheimpflug pachymetry showed 0.4 D of astigmatism with a steep meridian of 44.4 D at the 184-degree meridian.
Figure 1. Intraoperative OCT images of the cornea. Axial (A) and sagittal (B) views indicating that the epithelial as well as the endothelial surfaces were initially adequately identified by the OCT. C: Superior displacement of the cornea after the initial OCT imaging and prior to the initiation of laser treatment.
J CATARACT REFRACT SURG - VOL 40, JULY 2014
1226
CASE REPORT: CORNEAL PERFORATION WITH OCT-GUIDED FEMTOSECOND LASER
Table 1. Femtosecond laser parameters used to perform the AK. One asymmetric intrastromal incision was created aiming for 60% intrastromal incision. Centration Meridian Length Optical zone Uncut anterior (%) Uncut posterior (%) Time
Limbus 159 degrees 45 degrees 9.0 mm 20 20 8.7 seconds
The manifest refraction was 0.75 D sphere, resulting in 20/ 20 visual acuity.
DISCUSSION Corneal perforation, a potential complication of incisional keratotomy, has been described during manual radial and arcuate keratotomy.6,7 Vaddavalli et al.8 reported a full-thickness incision during an AK performed using the femtosecond laser. To our knowledge, this is the first report of a corneal perforation during an intrastromal AK performed with an OCT-guided femtosecond laser. Incisional keratotomy requires accurate measurement of corneal thickness in the area of the incision to allow adequate blade or laser calibration, thereby avoiding perforation. This is usually performed preoperatively by ultrasound pachymetry or Scheimpflug imaging. These modalities may be inaccurate as they may not measure the exact location of the planned AK. An OCT-guided femtosecond laser should, in theory, be more precise as it is able to image and measure the area of the planned incision intraoperatively. In our case, a full-thickness incision was created despite calibration for an incision 20% deeper than the epithelium and 20% shallower than the endothelium. It is not clear why it resulted in a full-thickness incision. Possibilities include ocular or head movement
Figure 2. Air bubbles seen at the posterior aspect of the AK incision indicating posterior perforation.
occurring after the OCT scanning and prior to the laser incision. Examination of the OCT scan shows that the anterior and posterior surfaces of the cornea were initially correctly identified by the laser (Figure 1, A and B). A later image prior to initiation of the treatment shows a superior displacement of the cornea presumably due to patient movement (Figure 1, C). Air bubbles in the anterior chamber have been described as a sign of a perforated AK performed by a femtosecond laser.8,9 In our case, air bubbles were seen emanating from the endothelial aspect of the cornea into the anterior chamber as well as from the epithelial side into the liquid interface. Such air bubbles should be an indicator to the operating surgeon that the integrity of the endothelium and/or epithelium has been invaded. An undetected perforation may lead to postoperative incision leakage and increase the risk for a corneal infection and/or endophthalmitis. Astigmatic keratotomy is associated with a risk for corneal perforation even if performed with a femtosecond laser. A risk/benefit analysis is worth consideration, especially for lower levels of astigmatism in which keratometry measurement methods often disagree. The OCT imaging is performed prior to the laser activation rather than in a dynamic fashion. This may lead to an inaccurate keratotomy, capsulotomy, and lens fragmentation if eye movement occurs after the initial OCT scan. Real-time OCT with intraoperative motion detection would prompt an immediate pause in the procedure to avoid the occurrence of potentially serious complications during femtosecond-laser cataract surgery. REFERENCES
Figure 3. Full-thickness AK incision inspected with a spatula.
1. Duffey RJ, Jain VN, Tchah H, Hofmann RF, Lindstrom RL. Paired arcuate keratotomy; a surgical approach to mixed and myopic astigmatism. Arch Ophthalmol 1988; 106:1130–1135 2. Abbey A, Ide T, Kymionis GD, Yoo SH. Femtosecond laser-assisted astigmatic keratotomy in naturally
J CATARACT REFRACT SURG - VOL 40, JULY 2014
CASE REPORT: CORNEAL PERFORATION WITH OCT-GUIDED FEMTOSECOND LASER
occurring high astigmatism. Br J Ophthalmol 2009; 93:1566–1569 3. Khokhar S, Lohiya P, Murugiesan V, Panda A. Corneal astigmatism correction with opposite clear corneal incisions or single clear corneal incision: comparative analysis. J Cataract Refract Surg 2006; 32:1432–1437 €ckl T, Dexl AK, Bachernegg A, Reischl V, Riha W, 4. Ru Ruckhofer J, Binder PS, Grabner G. Femtosecond laser–assisted intrastromal arcuate keratotomy to reduce corneal astigmatism. J Cataract Refract Surg 2013; 39:528–538 5. Hoffart L, Proust H, Matonti F, Conrath J, Ridings B. Correction of postkeratoplasty astigmatism by femtosecond laser compared with mechanized astigmatic keratotomy. Am J Ophthalmol 2009; 147:779–787
1227
6. Hardten DR, Lindstrom RL. Surgical correction of refractive errors after penetrating keratoplasty. Int Ophthalmol Clin 1997; 37(1):1–35 7. Bahar I, Levinger E, Kaiserman I, Sansanayudh W, Rootman DS. IntraLase-enabled astigmatic keratotomy for postkeratoplasty astigmatism. Am J Ophthalmol 2008; 146:897–904 8. Vaddavalli PK, Hurmeric V, Yoo SH. Air bubble in anterior chamber as indicator of full-thickness incisions in femtosecondassisted astigmatic keratotomy. J Cataract Refract Surg 2011; 37:1723–1725 9. Srinivasan S, Rootman DS. Anterior chamber gas bubble formation during femtosecond laser flap creation for LASIK. J Refract Surg 2007; 23:828–830
J CATARACT REFRACT SURG - VOL 40, JULY 2014
Corneal perforation by an astigmatic keratotomy performed with an optical coherence tomography–guided femtosecond laser Daniel G. Cherfan, MD, Samir A. Melki, MD, PhD
We present a case of corneal perforation secondary to an intrastromal astigmatic keratotomy performed with an optical coherence tomography–guided femtosecond laser. The keratotomy was concomitant with cataract surgery and resulted in a flat anterior chamber prior to the start of lens extraction. Interrupted nylon sutures were placed to seal the keratotomy prior to phacoemulsification. Escape of cavitation bubbles into the anterior chamber or the liquid interface can alert the surgeon to the possibility of unintended perforation of the endothelium or the epithelium, respectively. Financial Disclosure: Neither author has a financial or proprietary interest in any material or method mentioned. J Cataract Refract Surg 2014; 40:1224–1227 Q 2014 ASCRS and ESCRS Online Video
Corneal astigmatism can be surgically reduced or corrected through several methods, which can be incisional or ablative. Incisional techniques concurrent with cataract surgery include arcuate keratotomy,1 limbal relaxing incision,2 and opposite clear corneal incision.3 Arcuate keratotomies can be performed either manually using a keratome or a diamond knife or using a femtosecond laser.2,4,5 Astigmatic keratotomy (AK) with a femtosecond laser is becoming a popular choice in clinical practice due to its presumed advantages over manual methods.2 Ocular coherence tomography (OCT)– guided femtosecond lasers offer more detailed corneal imaging, which may improve accuracy and precision of the corneal incisions. We present a case of corneal perforation following creation of
Submitted: November 6, 2013. Final revision submitted: February 14, 2014. Accepted: February 16, 2014. From the Massachusetts Eye and Ear Infirmary and the Boston Eye Group (Melki), Boston, Massachusetts, USA; the Royal College of Surgeons (Cherfan), Dublin, Ireland. Corresponding author: Samir A. Melki, MD, PhD, 1101 Beacon Street 6W, Brookline, Massachusetts 02446, USA. E-mail: samir_ [email protected].
1224
Q 2014 ASCRS and ESCRS Published by Elsevier Inc.
an intrastromal AK incision using an OCT-guided femtosecond laser–assisted cataract extraction. CASE REPORT A 50-year-old woman with a nuclear sclerotic cataract in both eyes was scheduled for femtosecond laser–assisted cataract surgery in her left eye due to difficulty driving at night. The corrected distance visual acuity at the time of surgery was 20/40 in the left eye with a refraction of –1.75 1.25 62. The ocular examination was normal except for a lenticular cataract graded as 1C. Automated keratometry revealed 0.5 diopter (D) of astigmatism with a steep meridian of 44.5 @ 137. Scheimpflug pachymetry (Pentacam, Oculus Surgical, Inc.) showed 0.7 D of regular astigmatism with a steep meridian of 44.70 @ 142. Keratometry from partial coherence interferometry (IOLMaster, Carl Zeiss Meditec AG) revealed 0.59 D of astigmatism with a steep meridian of 44.88 @ 126. The pachymetry map on Scheimpflug pachymetry showed a corneal thickness of approximately 600 mm in the area of the planned keratotomy. A plan for cataract extraction with AK at the 159-degree meridian was made using the LRI calculator (Abbott Medical Optics, Inc.), taking into account 0.5 D of astigmatism induced by the cataract incision placed at the 110-degree meridian. An OCT-guided femtosecond laser (Catalys, Optimedica Corp.) was used to create the main corneal incision, 2 paracentesis incisions, as well as the AK. The Catalys laser uses a liquid interface that docks the patient to the system and is designed to provide a clear optical path for real-time video, OCT imaging, and laser treatment. This is performed by filling in corneal surface irregularities with liquid and creating a wide field of view, enabling all cuts to be 0886-3350/$ - see front matter http://dx.doi.org/10.1016/j.jcrs.2014.04.021
CASE REPORT: CORNEAL PERFORATION WITH OCT-GUIDED FEMTOSECOND LASER
performed in a single dock. The OCT generates an image that is automatically registered and analyzed using the Catalys Integral Guidance interface to create safety zones in both the cornea and the posterior aspect of the lens capsule. The safety zones are used to design the treatment by focusing the femtosecond laser based on the patient's ocular anatomy. The intraoperative axial and sagittal OCT images of the cornea are seen in Figure 1, A and B. The femtosecond laser was also used for the capsulotomy and the lens fragmentation steps. Once the safety zones were identified by OCT, the incisions were made in the following sequence: anterior capsulorhexis, lens fragmentation, AK, and penetrating incisions. Examination of the OCT imaging revealed a displacement of the lines identifying the epithelial and endothelial layers (Figure 1, C). This was presumably due to patient movement after the initial OCT scan and prior to initiation of the keratotomy. The AK incision was programmed to be intrastromal. Table 1 shows the laser settings used, indicating that the arcuate incision was planned to extend from 20% below
1225
the epithelium to 20% above the endothelium. Air bubbles were seen emanating from the anterior and posterior aspects of the arcuate incision (Figure 2, A). The procedure was not aborted as the bubbles were not noticed intraoperatively. The patient was then moved to the operating microscope, where the anterior chamber was noted to be flat. The cataract corneal incision was not patent. A keratome was used to enter the anterior chamber. The latter was reformed with an ophthalmic viscosurgical device. The AK was inspected with a cyclodialysis spatula, which revealed a fullthickness incision along the full length of the incision (Figure 3). Two interrupted 10-0 nylon sutures were placed to seal the wound (Video 1, available at: http://jcrsjournal. org). The cataract was extracted by phacoemulsification as planned with no further complications. Postoperatively, the sutures were removed at weeks 5 and 6. At the 3-month follow-up visit, the uncorrected visual acuity was 20/50. The keratometry was 44.25 in both meridians, and Scheimpflug pachymetry showed 0.4 D of astigmatism with a steep meridian of 44.4 D at the 184-degree meridian.
Figure 1. Intraoperative OCT images of the cornea. Axial (A) and sagittal (B) views indicating that the epithelial as well as the endothelial surfaces were initially adequately identified by the OCT. C: Superior displacement of the cornea after the initial OCT imaging and prior to the initiation of laser treatment.
J CATARACT REFRACT SURG - VOL 40, JULY 2014
1226
CASE REPORT: CORNEAL PERFORATION WITH OCT-GUIDED FEMTOSECOND LASER
Table 1. Femtosecond laser parameters used to perform the AK. One asymmetric intrastromal incision was created aiming for 60% intrastromal incision. Centration Meridian Length Optical zone Uncut anterior (%) Uncut posterior (%) Time
Limbus 159 degrees 45 degrees 9.0 mm 20 20 8.7 seconds
The manifest refraction was 0.75 D sphere, resulting in 20/ 20 visual acuity.
DISCUSSION Corneal perforation, a potential complication of incisional keratotomy, has been described during manual radial and arcuate keratotomy.6,7 Vaddavalli et al.8 reported a full-thickness incision during an AK performed using the femtosecond laser. To our knowledge, this is the first report of a corneal perforation during an intrastromal AK performed with an OCT-guided femtosecond laser. Incisional keratotomy requires accurate measurement of corneal thickness in the area of the incision to allow adequate blade or laser calibration, thereby avoiding perforation. This is usually performed preoperatively by ultrasound pachymetry or Scheimpflug imaging. These modalities may be inaccurate as they may not measure the exact location of the planned AK. An OCT-guided femtosecond laser should, in theory, be more precise as it is able to image and measure the area of the planned incision intraoperatively. In our case, a full-thickness incision was created despite calibration for an incision 20% deeper than the epithelium and 20% shallower than the endothelium. It is not clear why it resulted in a full-thickness incision. Possibilities include ocular or head movement
Figure 2. Air bubbles seen at the posterior aspect of the AK incision indicating posterior perforation.
occurring after the OCT scanning and prior to the laser incision. Examination of the OCT scan shows that the anterior and posterior surfaces of the cornea were initially correctly identified by the laser (Figure 1, A and B). A later image prior to initiation of the treatment shows a superior displacement of the cornea presumably due to patient movement (Figure 1, C). Air bubbles in the anterior chamber have been described as a sign of a perforated AK performed by a femtosecond laser.8,9 In our case, air bubbles were seen emanating from the endothelial aspect of the cornea into the anterior chamber as well as from the epithelial side into the liquid interface. Such air bubbles should be an indicator to the operating surgeon that the integrity of the endothelium and/or epithelium has been invaded. An undetected perforation may lead to postoperative incision leakage and increase the risk for a corneal infection and/or endophthalmitis. Astigmatic keratotomy is associated with a risk for corneal perforation even if performed with a femtosecond laser. A risk/benefit analysis is worth consideration, especially for lower levels of astigmatism in which keratometry measurement methods often disagree. The OCT imaging is performed prior to the laser activation rather than in a dynamic fashion. This may lead to an inaccurate keratotomy, capsulotomy, and lens fragmentation if eye movement occurs after the initial OCT scan. Real-time OCT with intraoperative motion detection would prompt an immediate pause in the procedure to avoid the occurrence of potentially serious complications during femtosecond-laser cataract surgery. REFERENCES
Figure 3. Full-thickness AK incision inspected with a spatula.
1. Duffey RJ, Jain VN, Tchah H, Hofmann RF, Lindstrom RL. Paired arcuate keratotomy; a surgical approach to mixed and myopic astigmatism. Arch Ophthalmol 1988; 106:1130–1135 2. Abbey A, Ide T, Kymionis GD, Yoo SH. Femtosecond laser-assisted astigmatic keratotomy in naturally
J CATARACT REFRACT SURG - VOL 40, JULY 2014
CASE REPORT: CORNEAL PERFORATION WITH OCT-GUIDED FEMTOSECOND LASER
occurring high astigmatism. Br J Ophthalmol 2009; 93:1566–1569 3. Khokhar S, Lohiya P, Murugiesan V, Panda A. Corneal astigmatism correction with opposite clear corneal incisions or single clear corneal incision: comparative analysis. J Cataract Refract Surg 2006; 32:1432–1437 €ckl T, Dexl AK, Bachernegg A, Reischl V, Riha W, 4. Ru Ruckhofer J, Binder PS, Grabner G. Femtosecond laser–assisted intrastromal arcuate keratotomy to reduce corneal astigmatism. J Cataract Refract Surg 2013; 39:528–538 5. Hoffart L, Proust H, Matonti F, Conrath J, Ridings B. Correction of postkeratoplasty astigmatism by femtosecond laser compared with mechanized astigmatic keratotomy. Am J Ophthalmol 2009; 147:779–787
1227
6. Hardten DR, Lindstrom RL. Surgical correction of refractive errors after penetrating keratoplasty. Int Ophthalmol Clin 1997; 37(1):1–35 7. Bahar I, Levinger E, Kaiserman I, Sansanayudh W, Rootman DS. IntraLase-enabled astigmatic keratotomy for postkeratoplasty astigmatism. Am J Ophthalmol 2008; 146:897–904 8. Vaddavalli PK, Hurmeric V, Yoo SH. Air bubble in anterior chamber as indicator of full-thickness incisions in femtosecondassisted astigmatic keratotomy. J Cataract Refract Surg 2011; 37:1723–1725 9. Srinivasan S, Rootman DS. Anterior chamber gas bubble formation during femtosecond laser flap creation for LASIK. J Refract Surg 2007; 23:828–830
J CATARACT REFRACT SURG - VOL 40, JULY 2014
