- Email: [email protected]
Visualization of femtosecond laser pulse–induced microincisions inside crystalline lens tissue
LABORATORY SCIENCE
Visualization of femtosecond laser pulse–induced microincisions inside crystalline lens tissue Oliver Stachs, PhD, Silvia Schumacher, PhD, Marine Hovakimyan, PhD, Michael Fromm, MSc, Alexander Heisterkamp, PhD, Holger Lubatschowski, PhD, Rudolf Guthoff, MD
PURPOSE: To evaluate a new method for visualizing femtosecond laser pulse–induced microincisions inside crystalline lens tissue. SETTING: Laser Zentrum Hannover e.V., Hannover, Germany. METHOD: Lenses removed from porcine eyes were modified ex vivo by femtosecond laser pulses (wavelength 1040 nm, pulse duration 306 femtoseconds, pulse energy 1.0 to 2.5 mJ, repetition rate 100 kHz) to create defined planes at which lens fibers separate. The femtosecond laser pulses were delivered by a 3-dimension (3-D) scanning unit and transmitted by focusing optics (numerical aperture 0.18) into the lens tissue. Lens fiber orientation and femtosecond laser–induced microincisions were examined using a confocal laser scanning microscope (CLSM) based on a Rostock Cornea Module attached to a Heidelberg Retina Tomograph II. Optical sections were analyzed in 3-D using Amira software (version 4.1.1). RESULTS: Normal lens fibers showed a parallel pattern with diameters between 3 mm and 9 mm, depending on scanning location. Microincision visualization showed different cutting effects depending on pulse energy of the femtosecond laser. The effects ranged from altered tissue-scattering properties with all fibers intact to definite fiber separation by a wide gap. Pulse energies that were too high or overlapped too tightly produced an incomplete cutting plane due to extensive microbubble generation. CONCLUSIONS: The 3-D CLSM method permitted visualization and analysis of femtosecond laser pulse–induced microincisions inside crystalline lens tissue. Thus, 3-D CLSM may help optimize femtosecond laser–based procedures in the treatment of presbyopia. J Cataract Refract Surg 2009; 35:1979–1983 Q 2009 ASCRS and ESCRS
The development of presbyopia and the restoration of accommodation have become a recent focus of research.1–3 The accepted main cause of presbyopia is progressive sclerosis of the crystalline lens tissue.4,5 One strategy for restoring accommodation and overcoming lens stiffening is to use femtosecond laser pulses to separate the collagen fibrils by microbubbles6,7 or to induce microincisions inside the lens that act as gliding planes to restore tissue flexibility.8 The microincisions inside the lens tissue are created with femtosecond laser pulses in a manner similar to flap generation in laser in situ keratomileusis.8 The first in vitro studies of porcine8 and human9 donor lenses showed increased lens deformability after treatment. At present, 2 approaches are used to visualize the microincisions created by the femtosecond laser. In the first approach, which uses light microscopy and optical coherence tomography (OCT), no physical manipulation is necessary to obtain images of the Q 2009 ASCRS and ESCRS Published by Elsevier Inc.
incisions. Both technologies show the incisions; however, because the contrast is weak, the internal structure of the lens cannot be resolved.10 The second approach entails difficult and time-consuming preparation of the tissue. Histologic sections and scanning electron micrograph (SEM) images resolve the internal structure of the lens, allowing imaging of single lens fibers and laser spots. These methods require physical and chemical manipulation of the specimens. Considerable information about the spatial arrangement of anatomical structures can be lost during specimen preparation, especially during mechanical cutting of the tissue. In addition, lens tissue begins to degenerate immediately after enucleation. Although these established methods provide information about microarchitecture, they reveal less detail about the spatial arrangement of lens fiber structure and femtosecond laser–induced microincisions. There is, therefore, a major incentive to develop a method that resolves the 0886-3350/09/$dsee front matter doi:10.1016/j.jcrs.2009.06.019
1979
1980
LABORATORY SCIENCE: FEMTOSECOND LASER PULSE–INDUCED MICROINCISIONS
internal structure of the lens and can be performed without tissue manipulation. The noninvasive alternative is cell imaging in vivo using reflected light from within the tissue. Evaluating changes in the refractive index or using fluorescence agents permits recognition of intercellular and intracellular details and has the potential to yield sufficient information to diagnose several types of tissue alterations. Confocal microscopy has been established as a valuable tool for obtaining high-resolution images and 3-dimension (3-D) reconstructions of biologic specimens.11 The principle behind the confocal microscope was proposed by M. Minsky (Available at: http://web.me dia.mit.edu/wminsky/papers/ConfocalMemoir.html. Accessed July 28, 2009). This eventually evolved into a new optical technique for the clinical examination of the human eye.12 The original concept was augmented by the development of the scanning laser ophthalmoscope by Webb et al.13,14 in which a coherent laser is used as a high-intensity light source and the laser beam is scanned by a set of galvanometer mirrors, providing fast scanning around the x–y plane. The reflected light refocused by the microscope objective is scanned by the galvanometer mirrors again and imaged on a pinhole aperture located in front of the photomultiplier. The aim of the present study was to evaluate the utility of confocal laser scanning microscopy (CLSM) for the nondestructive visualization of femtosecond laser pulse–induced microincisions inside crystalline lens tissue. MATERIALS AND METHODS In this study, CLSM was performed using a Retina Tomograph (Heidelberg Engineering GmbH), a well-established
Submitted: March 16, 2009. Final revision submitted: June 18, 2009. Accepted: June 30, 2009.
in vivo confocal imaging system. The device has a 670 mm diode laser designed to acquire and evaluate topographic measurements of the optic nerve head to detect glaucomatous damage. By the addition of a detachable objective system, the tomograph device becomes a high-resolution confocal laser scanning microscope. The cornea module, an optical system developed in the Department of Ophthalmology at the University of Rostock,15 shifts the focal plane closer to the objective lens system and increases the magnification (Figure 1). In this study, the objective lens system comprised a water-immersion objective (Achroplan 63/ 0.95 W/AA 1.45 mm, Carl Zeiss Meditec) with a long focal length and large aperture. For in vivo imaging, the front lens was coupled to the tissue to be evaluated via a poly(methyl methacrylate) plate with interposition of transparent gel (Vidisic, Mann Pharma). For 3-D imaging, a scanning device moved the focal plane perpendicularly to the x–y plane identically to the optic disk topography of the original retina tomograph’s functionality. The scanned volumes had a maximum size of 400 mm 400 mm 100 mm and a voxel size of 0.78 mm 0.78 mm 0.95 mm. Optical sections were analyzed and digitally reconstructed in 3-D using Amira visualization software (version 4.1.1, Mercury Computer Systems, Inc.). In this ex vivo study, porcine eyes were obtained from a local abattoir. The crystalline lenses were extracted from the bulb of the eye, placed in a special lens holder, and slightly applanated during the laser process. A 3-D cutting pattern consisting of rings, cylinders, and square planes was applied to each lens. The rings (top and bottom planes of the structure) had an outer diameter of 2.0 mm and an inner diameter of 0.3 mm. Two cylinders were placed at the rims of the ring at 2.0 mm and 0.3 mm, respectively. In addition, 12 square planes perpendicular to the rings were added to the 3-D structure, which had a total depth of 240 mm. Ripken et al.8 gave a detailed description of the pattern. The structure was placed 100 mm below the lens capsule. The laser (mJewel D-400, IMRA America, Inc.) with a wavelength of 1040 nm, pulse duration of 306 femtoseconds, and a repetition rate of 100 KHz was coupled to a 3-D scanning device. The scanning device consisted of a galvanometer scanner (x–y plane) (Scanlab AG) and a linear translation stage (z plane) (Physik Instrumente GmbH & Co.). The beam was focused by an f-theta optic (numerical aperture 0.18) into the lens tissue with a minimum spot size of 4 mm. The pulse energy varied between 1.0 mJ and 2.5 mJ, with spot separation laterally ranging from 5 to 7 mm and axially from 30 to 50 mm.
From the Department of Ophthalmology (Stachs, Hovakimyan, Guthoff), University of Rostock, Rostock, and Laser Zentrum Hannover e.V. (Schumacher, Fromm, Heisterkamp, Lubatschowski), Hannover, Germany. No author has a financial or proprietary interest in any material or method mentioned. Supported by BMBF FKZ 13N8709 and 13N8712 and in part by the DFG (Transregio 37, Micro- und Nanosystems in Medicine–Reconstruction of Biological Functions). David Beattie provided editorial support. Corresponding author: Oliver Stachs, PhD, Department of Ophthalmology, University of Rostock, Doberaner Straße 140, D-18057 Rostock, Germany. E-mail: [email protected].
Figure 1. The CLSM optical setup used for crystalline lens imaging.
J CATARACT REFRACT SURG - VOL 35, NOVEMBER 2009
LABORATORY SCIENCE: FEMTOSECOND LASER PULSE–INDUCED MICROINCISIONS
1981
Figure 2. Representative confocal images of fresh porcine lenses. A: Lens capsule morphology. B: Crystalline lens fiber morphology shows a wellaligned pattern. C: Surface suture at periphery of the lens.
RESULTS Normal Lens Fiber Patterns Figure 2 shows confocal images of fresh porcine lenses. Lens fibers ran parallel to each other. Fiber diameters were between 3 mm and 9 mm depending on the scanning location and penetration depth. Maximum penetration depths were between 500 mm and 700 mm depending on the scanning location and tissue transparency. Femtosecond Laser Pulse–Induced Microincisions Microincision visualization was achieved to varying degrees depending on lens transparency; however, it was possible to obtain images to a depth of 500 mm inside the lens tissue. Single-section images through the microincisions inside the lens showed that the cutting effect varied depending on pulse energy (Figure 3), allowing identification of distinct patterns. In the first pattern, the scattering properties of the tissue were
altered at the point of the laser–tissue interaction but all fibers were intact (Figure 3, A). In the second, separation of fibers was clearly visible but some fibers were intact (Figure 3, B). In the third, all fibers were clearly separated by a wide gap (Figure 3, C). The 3-D reconstruction showed well-aligned scan geometry in all spatial directions (Figure 4). DISCUSSION One experimental approach to the correction of presbyopia involves lens softening using femtosecond laser pulses to create 3-D cutting structures inside the lens tissue. The first experimental results were promising.6–8 Studies show that the procedure can be applied successfully in porcine,8 human cadaver donor,9 and rabbit10 eyes. The laser microincisions inside the crystalline lens were detectable on OCT and Scheimpflug imaging, which emphasizes the integral role these technologies play in targeting and characterizing
Figure 3. Representative confocal images of fresh porcine lenses with femtosecond laser–induced microincisions. A: Tissue changes at the laser– tissue interaction; all fibers are intact (spot separation laterally 7 mm; pulse energy 1.3 mJ). B: Some fibers are intact (spot separation laterally 7 mm; pulse energy 1.4 mJ). C: A wide gap separates all fibers (spot separation laterally 7 mm; pulse energy 1.5 mJ). J CATARACT REFRACT SURG - VOL 35, NOVEMBER 2009
1982
LABORATORY SCIENCE: FEMTOSECOND LASER PULSE–INDUCED MICROINCISIONS
Figure 4. The 3-D volume (400400100 mm) and different cross-sections (A to D) at intervals of 20 mm showing 2 well-aligned cutting planes (pulse energy 1.5 mJ).
postoperative tissue effects. Intralenticular imaging showed progressive fading of the incisional opacities generated by the femtosecond laser after 14 days with no detectable cataract formation. The rabbit experiments showed that it is possible to create microincisions inside the crystalline lens within an acceptably short (!30 seconds) treatment time. Follow-up 14 days after intralenticular laser treatment did not show undesirable side effects, such as cataract formation. Visualization of femtosecond laser pulse patterns presents significant challenges. Histologic sections and SEM images resolve the internal structure of the lens, permitting single lens fibers to be imaged and laser spot information to be obtained. However, these investigations require physical manipulation of the specimens, and considerable information about the spatial arrangement of femtosecond laser–induced patterns is lost during specimen preparation. Confocal laser scanning microscopy at a wavelength of 670 nm can solve this problem. With this technique, the ultrastructure of the crystalline lens fibers can be visualized without staining or invasive preparation procedures. Our findings correspond well with stateof-the-art knowledge of lens fiber orientation16,17 and are not discussed in detail here. The maximum penetration depth possible and the visibility of lens fibers and femtosecond laser–induced microincisions varied, depending on the lens evaluated. Overall, it was possible to obtain images inside the lens tissue to a depth of 500 mm. Penetration depth and image quality are highly tissue dependent. Image quality is characterized not only by resolution but also
by contrast. Contrast is dependent on the illumination level, refractive index, and reflectivity of the tissue studied. Penetration depth is limited by 2 factors; that is, the signal-to-noise ratio of the scattered light and the background intensity. Femtosecond laser–induced cutting inside the lens tissue is based on the nonlinear laser–tissue interaction that occurs at high intensities above 109 W/cm2. These high intensities can occur only if ultrashort laser pulses are tightly focused. At a certain threshold, multiphoton absorption and cascade ionization occur, accompanied by the induction of rapid plasma generation. The expanding plasma creates a shockwave followed by a cavitation bubble, thus disrupting the tissue. The size of the cavitation bubble (and therefore precision) depends on pulse energy. Optimum precision is achieved immediately above the threshold for nonlinear absorption. Scanning of the focal point through the tissue can create every conceivable 3-D structure/outline. The CLSM images through the microincisions inside the lens showed that the cutting effect varied depending on pulse energy and pulse overlap. In principle, 3 femtosecond laser-induced patterns were identified. First, the scattering properties of the tissue were altered at points where the laser–tissue interaction occurred below the threshold for optical breakdown and the fibers were still intact. Nevertheless, the electron density in the focal region induced a chemical change in the fiber cells that caused the increased scattering. In the second pattern, fiber separation was clearly visible but some fibers were intact. The laser pulses induced optical
J CATARACT REFRACT SURG - VOL 35, NOVEMBER 2009
LABORATORY SCIENCE: FEMTOSECOND LASER PULSE–INDUCED MICROINCISIONS
1983
crystalline lenses. In addition, optical alignments must be developed to increase the working distance for in vivo imaging of lens-softening effects. Finally, it is possible that 3-D CLSM will provide fresh insight into the aging of the lens, cataract development in the crystalline lens, and development of secondary cataract after cataract surgery. REFERENCES
Figure 5. Bubble formation resulting from a high-pulse energy (spot separation laterally 6 mm, pulse energy 1.5 mJ) causes large cavitation bubbles, which leave large cavities of residual gas that interfere with subsequent laser pulses.
breakdown in some cases, and the tissue was disrupted in the focal region. In the third scenario, all fibers were clearly separated by a wide gap. Three-dimensional reconstruction showed well-aligned scan geometries in all spatial directions with a scanning device consisting of a galvanometer scanner (x–y plane) and a linear translation stage (z plane). The precision of the cut depends on the selected pulse energy and spot separation. For a precise cut, the combination of pulse energy and spot separation must be well adjusted. If the pulse energy is too high for a given spot separation or the spot separation is too low for a given pulse energy, large cavitation bubbles will be produced that will leave large cavities of residual gas. This effect causes interference with the beam path of subsequent laser pulses, resulting in an inhomogeneous cutting performance (Figure 5). No precise cut is achieved; rather, there is only an area of larger gas bubbles with intact fibers in between. Furthermore, the optical quality of the lens tissue also influences the threshold of the optical breakdown and therefore the cutting performance at a given combination of laser pulse energy and spot separation. In conclusion, 3-D CLSM at 670 nm permitted visualization and analysis of femtosecond laser pulse– induced microincisions inside crystalline lens tissue. Further developments are necessary to increase the penetration depth for imaging larger volume
1. Glasser A. Restoration of accommodation: surgical options for correction of presbyopia. Clin Exp Optom 2008; 91:279–295 2. Charman WN. The eye in focus: accommodation and presbyopia. Clin Exp Optom 2008; 91:207–225 3. Glasser A, Campbell MCW. Biometric, optical and physical changes in the isolated human crystalline lens with age in relation to presbyopia. Vision Res 1999; 39:1991–2015 4. Heys KR, Cram SL, Truscott RJW. Massive increase in the stiffness of the human lens nucleus with age: the basis for presbyopia? Mol Vis 2004; 10:956–963. Available at: http://www. molvis.org/molvis/v10/a114. Accessed July 27, 2009 5. Weeber HA, Eckert G, Pechold W, van der Heijde RGL. Stiffness gradient in the crystalline lens. Graefes Arch Clin Exp Ophthalmol 2007; 245:1357–1366 6. Myers RI, Krueger RR. Novel approaches to correction of presbyopia with laser modification of the crystalline lens. J Refract Surg 1998; 14:136–139 7. Krueger RR, Sun XK, Stroh J, Myers R. Experimental increase in accommodative potential after neodymium:yttrium-aluminumgarnet laser photodisruption of paired cadaver lenses. Ophthalmology 2001; 108:2122–2129 8. Ripken T, Oberheide U, Fromm M, Schumacher S, Gerten G, Lubatschowski H. fs-Laser induced elasticity changes to improve presbyopic lens accommodation. Graefes Arch Clin Exp Ophthalmol 2008; 246:897–906 9. Schumacher S, Oberheide U, Fromm M, Ripken T, Ertmer W, Gerten G, Wegener A, Lubatschowski H. Femtosecond-laser induced flexibility change of human donor lenses. Vision Res 2009; 49:1853–1859 10. Schumacher S, Fromm M, Oberheide U, Gerten G, Wegener A, Lubatschowski H. In vivo application and imaging of intralenticular femtosecond laser pulses for the restoration of accommodation. J Refract Surg 2008; 24:991–995 11. Dailey ME, Manders E, Soll DR, Terasaki M. In: Pawley JB, ed, Handbook of Biological Confocal Microscopy 2nd ed. New York, NY, Plenum Press, 1995; 381–398 12. Young JZ, Roberts F. A flying-spot microscope [letter]. Nature 1951; 167:231 13. Webb RH, Hughes GW, Delori FC. Confocal scanning laser ophthalmoscope. Applied Optics 1987; 26:1492–1499 14. Webb RH. Scanning laser ophthalmoscope. In: Masters BR, ed, Noninvasive Diagnostic Techniques in Ophthalmology. New York, NY, Springer Verlag, 1990; 438–450 15. Stave J, Zinser G, Gru¨mmer G, Guthoff R. Der modifizierte Heidelberg-Retina-Tomograph HRT. Erste Ergebnisse einer in-vivo-Darstellung von kornealen Strukturen. [Modified Heidelberg Retina Tomograph HRT. Initial results of in vivo presentation of corneal structures]. Ophthalmologe 2002; 99:276–280 16. Forrester JV, Dick AD, McMenamin PG, Lee WR. The Eye: Basic Sciences in Practice. London, UK, W.B. Saunders, 1996; 31–35 17. Kuszak JR. The development of lens sutures. Prog Retin Eye Res 1995; 14:567–591
J CATARACT REFRACT SURG - VOL 35, NOVEMBER 2009
Visualization of femtosecond laser pulse–induced microincisions inside crystalline lens tissue Oliver Stachs, PhD, Silvia Schumacher, PhD, Marine Hovakimyan, PhD, Michael Fromm, MSc, Alexander Heisterkamp, PhD, Holger Lubatschowski, PhD, Rudolf Guthoff, MD
PURPOSE: To evaluate a new method for visualizing femtosecond laser pulse–induced microincisions inside crystalline lens tissue. SETTING: Laser Zentrum Hannover e.V., Hannover, Germany. METHOD: Lenses removed from porcine eyes were modified ex vivo by femtosecond laser pulses (wavelength 1040 nm, pulse duration 306 femtoseconds, pulse energy 1.0 to 2.5 mJ, repetition rate 100 kHz) to create defined planes at which lens fibers separate. The femtosecond laser pulses were delivered by a 3-dimension (3-D) scanning unit and transmitted by focusing optics (numerical aperture 0.18) into the lens tissue. Lens fiber orientation and femtosecond laser–induced microincisions were examined using a confocal laser scanning microscope (CLSM) based on a Rostock Cornea Module attached to a Heidelberg Retina Tomograph II. Optical sections were analyzed in 3-D using Amira software (version 4.1.1). RESULTS: Normal lens fibers showed a parallel pattern with diameters between 3 mm and 9 mm, depending on scanning location. Microincision visualization showed different cutting effects depending on pulse energy of the femtosecond laser. The effects ranged from altered tissue-scattering properties with all fibers intact to definite fiber separation by a wide gap. Pulse energies that were too high or overlapped too tightly produced an incomplete cutting plane due to extensive microbubble generation. CONCLUSIONS: The 3-D CLSM method permitted visualization and analysis of femtosecond laser pulse–induced microincisions inside crystalline lens tissue. Thus, 3-D CLSM may help optimize femtosecond laser–based procedures in the treatment of presbyopia. J Cataract Refract Surg 2009; 35:1979–1983 Q 2009 ASCRS and ESCRS
The development of presbyopia and the restoration of accommodation have become a recent focus of research.1–3 The accepted main cause of presbyopia is progressive sclerosis of the crystalline lens tissue.4,5 One strategy for restoring accommodation and overcoming lens stiffening is to use femtosecond laser pulses to separate the collagen fibrils by microbubbles6,7 or to induce microincisions inside the lens that act as gliding planes to restore tissue flexibility.8 The microincisions inside the lens tissue are created with femtosecond laser pulses in a manner similar to flap generation in laser in situ keratomileusis.8 The first in vitro studies of porcine8 and human9 donor lenses showed increased lens deformability after treatment. At present, 2 approaches are used to visualize the microincisions created by the femtosecond laser. In the first approach, which uses light microscopy and optical coherence tomography (OCT), no physical manipulation is necessary to obtain images of the Q 2009 ASCRS and ESCRS Published by Elsevier Inc.
incisions. Both technologies show the incisions; however, because the contrast is weak, the internal structure of the lens cannot be resolved.10 The second approach entails difficult and time-consuming preparation of the tissue. Histologic sections and scanning electron micrograph (SEM) images resolve the internal structure of the lens, allowing imaging of single lens fibers and laser spots. These methods require physical and chemical manipulation of the specimens. Considerable information about the spatial arrangement of anatomical structures can be lost during specimen preparation, especially during mechanical cutting of the tissue. In addition, lens tissue begins to degenerate immediately after enucleation. Although these established methods provide information about microarchitecture, they reveal less detail about the spatial arrangement of lens fiber structure and femtosecond laser–induced microincisions. There is, therefore, a major incentive to develop a method that resolves the 0886-3350/09/$dsee front matter doi:10.1016/j.jcrs.2009.06.019
1979
1980
LABORATORY SCIENCE: FEMTOSECOND LASER PULSE–INDUCED MICROINCISIONS
internal structure of the lens and can be performed without tissue manipulation. The noninvasive alternative is cell imaging in vivo using reflected light from within the tissue. Evaluating changes in the refractive index or using fluorescence agents permits recognition of intercellular and intracellular details and has the potential to yield sufficient information to diagnose several types of tissue alterations. Confocal microscopy has been established as a valuable tool for obtaining high-resolution images and 3-dimension (3-D) reconstructions of biologic specimens.11 The principle behind the confocal microscope was proposed by M. Minsky (Available at: http://web.me dia.mit.edu/wminsky/papers/ConfocalMemoir.html. Accessed July 28, 2009). This eventually evolved into a new optical technique for the clinical examination of the human eye.12 The original concept was augmented by the development of the scanning laser ophthalmoscope by Webb et al.13,14 in which a coherent laser is used as a high-intensity light source and the laser beam is scanned by a set of galvanometer mirrors, providing fast scanning around the x–y plane. The reflected light refocused by the microscope objective is scanned by the galvanometer mirrors again and imaged on a pinhole aperture located in front of the photomultiplier. The aim of the present study was to evaluate the utility of confocal laser scanning microscopy (CLSM) for the nondestructive visualization of femtosecond laser pulse–induced microincisions inside crystalline lens tissue. MATERIALS AND METHODS In this study, CLSM was performed using a Retina Tomograph (Heidelberg Engineering GmbH), a well-established
Submitted: March 16, 2009. Final revision submitted: June 18, 2009. Accepted: June 30, 2009.
in vivo confocal imaging system. The device has a 670 mm diode laser designed to acquire and evaluate topographic measurements of the optic nerve head to detect glaucomatous damage. By the addition of a detachable objective system, the tomograph device becomes a high-resolution confocal laser scanning microscope. The cornea module, an optical system developed in the Department of Ophthalmology at the University of Rostock,15 shifts the focal plane closer to the objective lens system and increases the magnification (Figure 1). In this study, the objective lens system comprised a water-immersion objective (Achroplan 63/ 0.95 W/AA 1.45 mm, Carl Zeiss Meditec) with a long focal length and large aperture. For in vivo imaging, the front lens was coupled to the tissue to be evaluated via a poly(methyl methacrylate) plate with interposition of transparent gel (Vidisic, Mann Pharma). For 3-D imaging, a scanning device moved the focal plane perpendicularly to the x–y plane identically to the optic disk topography of the original retina tomograph’s functionality. The scanned volumes had a maximum size of 400 mm 400 mm 100 mm and a voxel size of 0.78 mm 0.78 mm 0.95 mm. Optical sections were analyzed and digitally reconstructed in 3-D using Amira visualization software (version 4.1.1, Mercury Computer Systems, Inc.). In this ex vivo study, porcine eyes were obtained from a local abattoir. The crystalline lenses were extracted from the bulb of the eye, placed in a special lens holder, and slightly applanated during the laser process. A 3-D cutting pattern consisting of rings, cylinders, and square planes was applied to each lens. The rings (top and bottom planes of the structure) had an outer diameter of 2.0 mm and an inner diameter of 0.3 mm. Two cylinders were placed at the rims of the ring at 2.0 mm and 0.3 mm, respectively. In addition, 12 square planes perpendicular to the rings were added to the 3-D structure, which had a total depth of 240 mm. Ripken et al.8 gave a detailed description of the pattern. The structure was placed 100 mm below the lens capsule. The laser (mJewel D-400, IMRA America, Inc.) with a wavelength of 1040 nm, pulse duration of 306 femtoseconds, and a repetition rate of 100 KHz was coupled to a 3-D scanning device. The scanning device consisted of a galvanometer scanner (x–y plane) (Scanlab AG) and a linear translation stage (z plane) (Physik Instrumente GmbH & Co.). The beam was focused by an f-theta optic (numerical aperture 0.18) into the lens tissue with a minimum spot size of 4 mm. The pulse energy varied between 1.0 mJ and 2.5 mJ, with spot separation laterally ranging from 5 to 7 mm and axially from 30 to 50 mm.
From the Department of Ophthalmology (Stachs, Hovakimyan, Guthoff), University of Rostock, Rostock, and Laser Zentrum Hannover e.V. (Schumacher, Fromm, Heisterkamp, Lubatschowski), Hannover, Germany. No author has a financial or proprietary interest in any material or method mentioned. Supported by BMBF FKZ 13N8709 and 13N8712 and in part by the DFG (Transregio 37, Micro- und Nanosystems in Medicine–Reconstruction of Biological Functions). David Beattie provided editorial support. Corresponding author: Oliver Stachs, PhD, Department of Ophthalmology, University of Rostock, Doberaner Straße 140, D-18057 Rostock, Germany. E-mail: [email protected].
Figure 1. The CLSM optical setup used for crystalline lens imaging.
J CATARACT REFRACT SURG - VOL 35, NOVEMBER 2009
LABORATORY SCIENCE: FEMTOSECOND LASER PULSE–INDUCED MICROINCISIONS
1981
Figure 2. Representative confocal images of fresh porcine lenses. A: Lens capsule morphology. B: Crystalline lens fiber morphology shows a wellaligned pattern. C: Surface suture at periphery of the lens.
RESULTS Normal Lens Fiber Patterns Figure 2 shows confocal images of fresh porcine lenses. Lens fibers ran parallel to each other. Fiber diameters were between 3 mm and 9 mm depending on the scanning location and penetration depth. Maximum penetration depths were between 500 mm and 700 mm depending on the scanning location and tissue transparency. Femtosecond Laser Pulse–Induced Microincisions Microincision visualization was achieved to varying degrees depending on lens transparency; however, it was possible to obtain images to a depth of 500 mm inside the lens tissue. Single-section images through the microincisions inside the lens showed that the cutting effect varied depending on pulse energy (Figure 3), allowing identification of distinct patterns. In the first pattern, the scattering properties of the tissue were
altered at the point of the laser–tissue interaction but all fibers were intact (Figure 3, A). In the second, separation of fibers was clearly visible but some fibers were intact (Figure 3, B). In the third, all fibers were clearly separated by a wide gap (Figure 3, C). The 3-D reconstruction showed well-aligned scan geometry in all spatial directions (Figure 4). DISCUSSION One experimental approach to the correction of presbyopia involves lens softening using femtosecond laser pulses to create 3-D cutting structures inside the lens tissue. The first experimental results were promising.6–8 Studies show that the procedure can be applied successfully in porcine,8 human cadaver donor,9 and rabbit10 eyes. The laser microincisions inside the crystalline lens were detectable on OCT and Scheimpflug imaging, which emphasizes the integral role these technologies play in targeting and characterizing
Figure 3. Representative confocal images of fresh porcine lenses with femtosecond laser–induced microincisions. A: Tissue changes at the laser– tissue interaction; all fibers are intact (spot separation laterally 7 mm; pulse energy 1.3 mJ). B: Some fibers are intact (spot separation laterally 7 mm; pulse energy 1.4 mJ). C: A wide gap separates all fibers (spot separation laterally 7 mm; pulse energy 1.5 mJ). J CATARACT REFRACT SURG - VOL 35, NOVEMBER 2009
1982
LABORATORY SCIENCE: FEMTOSECOND LASER PULSE–INDUCED MICROINCISIONS
Figure 4. The 3-D volume (400400100 mm) and different cross-sections (A to D) at intervals of 20 mm showing 2 well-aligned cutting planes (pulse energy 1.5 mJ).
postoperative tissue effects. Intralenticular imaging showed progressive fading of the incisional opacities generated by the femtosecond laser after 14 days with no detectable cataract formation. The rabbit experiments showed that it is possible to create microincisions inside the crystalline lens within an acceptably short (!30 seconds) treatment time. Follow-up 14 days after intralenticular laser treatment did not show undesirable side effects, such as cataract formation. Visualization of femtosecond laser pulse patterns presents significant challenges. Histologic sections and SEM images resolve the internal structure of the lens, permitting single lens fibers to be imaged and laser spot information to be obtained. However, these investigations require physical manipulation of the specimens, and considerable information about the spatial arrangement of femtosecond laser–induced patterns is lost during specimen preparation. Confocal laser scanning microscopy at a wavelength of 670 nm can solve this problem. With this technique, the ultrastructure of the crystalline lens fibers can be visualized without staining or invasive preparation procedures. Our findings correspond well with stateof-the-art knowledge of lens fiber orientation16,17 and are not discussed in detail here. The maximum penetration depth possible and the visibility of lens fibers and femtosecond laser–induced microincisions varied, depending on the lens evaluated. Overall, it was possible to obtain images inside the lens tissue to a depth of 500 mm. Penetration depth and image quality are highly tissue dependent. Image quality is characterized not only by resolution but also
by contrast. Contrast is dependent on the illumination level, refractive index, and reflectivity of the tissue studied. Penetration depth is limited by 2 factors; that is, the signal-to-noise ratio of the scattered light and the background intensity. Femtosecond laser–induced cutting inside the lens tissue is based on the nonlinear laser–tissue interaction that occurs at high intensities above 109 W/cm2. These high intensities can occur only if ultrashort laser pulses are tightly focused. At a certain threshold, multiphoton absorption and cascade ionization occur, accompanied by the induction of rapid plasma generation. The expanding plasma creates a shockwave followed by a cavitation bubble, thus disrupting the tissue. The size of the cavitation bubble (and therefore precision) depends on pulse energy. Optimum precision is achieved immediately above the threshold for nonlinear absorption. Scanning of the focal point through the tissue can create every conceivable 3-D structure/outline. The CLSM images through the microincisions inside the lens showed that the cutting effect varied depending on pulse energy and pulse overlap. In principle, 3 femtosecond laser-induced patterns were identified. First, the scattering properties of the tissue were altered at points where the laser–tissue interaction occurred below the threshold for optical breakdown and the fibers were still intact. Nevertheless, the electron density in the focal region induced a chemical change in the fiber cells that caused the increased scattering. In the second pattern, fiber separation was clearly visible but some fibers were intact. The laser pulses induced optical
J CATARACT REFRACT SURG - VOL 35, NOVEMBER 2009
LABORATORY SCIENCE: FEMTOSECOND LASER PULSE–INDUCED MICROINCISIONS
1983
crystalline lenses. In addition, optical alignments must be developed to increase the working distance for in vivo imaging of lens-softening effects. Finally, it is possible that 3-D CLSM will provide fresh insight into the aging of the lens, cataract development in the crystalline lens, and development of secondary cataract after cataract surgery. REFERENCES
Figure 5. Bubble formation resulting from a high-pulse energy (spot separation laterally 6 mm, pulse energy 1.5 mJ) causes large cavitation bubbles, which leave large cavities of residual gas that interfere with subsequent laser pulses.
breakdown in some cases, and the tissue was disrupted in the focal region. In the third scenario, all fibers were clearly separated by a wide gap. Three-dimensional reconstruction showed well-aligned scan geometries in all spatial directions with a scanning device consisting of a galvanometer scanner (x–y plane) and a linear translation stage (z plane). The precision of the cut depends on the selected pulse energy and spot separation. For a precise cut, the combination of pulse energy and spot separation must be well adjusted. If the pulse energy is too high for a given spot separation or the spot separation is too low for a given pulse energy, large cavitation bubbles will be produced that will leave large cavities of residual gas. This effect causes interference with the beam path of subsequent laser pulses, resulting in an inhomogeneous cutting performance (Figure 5). No precise cut is achieved; rather, there is only an area of larger gas bubbles with intact fibers in between. Furthermore, the optical quality of the lens tissue also influences the threshold of the optical breakdown and therefore the cutting performance at a given combination of laser pulse energy and spot separation. In conclusion, 3-D CLSM at 670 nm permitted visualization and analysis of femtosecond laser pulse– induced microincisions inside crystalline lens tissue. Further developments are necessary to increase the penetration depth for imaging larger volume
1. Glasser A. Restoration of accommodation: surgical options for correction of presbyopia. Clin Exp Optom 2008; 91:279–295 2. Charman WN. The eye in focus: accommodation and presbyopia. Clin Exp Optom 2008; 91:207–225 3. Glasser A, Campbell MCW. Biometric, optical and physical changes in the isolated human crystalline lens with age in relation to presbyopia. Vision Res 1999; 39:1991–2015 4. Heys KR, Cram SL, Truscott RJW. Massive increase in the stiffness of the human lens nucleus with age: the basis for presbyopia? Mol Vis 2004; 10:956–963. Available at: http://www. molvis.org/molvis/v10/a114. Accessed July 27, 2009 5. Weeber HA, Eckert G, Pechold W, van der Heijde RGL. Stiffness gradient in the crystalline lens. Graefes Arch Clin Exp Ophthalmol 2007; 245:1357–1366 6. Myers RI, Krueger RR. Novel approaches to correction of presbyopia with laser modification of the crystalline lens. J Refract Surg 1998; 14:136–139 7. Krueger RR, Sun XK, Stroh J, Myers R. Experimental increase in accommodative potential after neodymium:yttrium-aluminumgarnet laser photodisruption of paired cadaver lenses. Ophthalmology 2001; 108:2122–2129 8. Ripken T, Oberheide U, Fromm M, Schumacher S, Gerten G, Lubatschowski H. fs-Laser induced elasticity changes to improve presbyopic lens accommodation. Graefes Arch Clin Exp Ophthalmol 2008; 246:897–906 9. Schumacher S, Oberheide U, Fromm M, Ripken T, Ertmer W, Gerten G, Wegener A, Lubatschowski H. Femtosecond-laser induced flexibility change of human donor lenses. Vision Res 2009; 49:1853–1859 10. Schumacher S, Fromm M, Oberheide U, Gerten G, Wegener A, Lubatschowski H. In vivo application and imaging of intralenticular femtosecond laser pulses for the restoration of accommodation. J Refract Surg 2008; 24:991–995 11. Dailey ME, Manders E, Soll DR, Terasaki M. In: Pawley JB, ed, Handbook of Biological Confocal Microscopy 2nd ed. New York, NY, Plenum Press, 1995; 381–398 12. Young JZ, Roberts F. A flying-spot microscope [letter]. Nature 1951; 167:231 13. Webb RH, Hughes GW, Delori FC. Confocal scanning laser ophthalmoscope. Applied Optics 1987; 26:1492–1499 14. Webb RH. Scanning laser ophthalmoscope. In: Masters BR, ed, Noninvasive Diagnostic Techniques in Ophthalmology. New York, NY, Springer Verlag, 1990; 438–450 15. Stave J, Zinser G, Gru¨mmer G, Guthoff R. Der modifizierte Heidelberg-Retina-Tomograph HRT. Erste Ergebnisse einer in-vivo-Darstellung von kornealen Strukturen. [Modified Heidelberg Retina Tomograph HRT. Initial results of in vivo presentation of corneal structures]. Ophthalmologe 2002; 99:276–280 16. Forrester JV, Dick AD, McMenamin PG, Lee WR. The Eye: Basic Sciences in Practice. London, UK, W.B. Saunders, 1996; 31–35 17. Kuszak JR. The development of lens sutures. Prog Retin Eye Res 1995; 14:567–591
J CATARACT REFRACT SURG - VOL 35, NOVEMBER 2009