1.

Introduction

Glaucoma is the second leading cause of blindness worldwide. It is a common ophthalmic disease, often associated with elevated intraocular pressure (IOP). More than two million people in the United States are affected by glaucoma.¹ Current medical and surgical therapies are directed at reducing IOP.² Drug treatments have several associated side effects.³ Traditional and laser surgical interventions also have unsatisfying disadvantages such as perioperative complications and limited longevity of achieved IOP reductions. In addition, surgical treatments may be associated with increased risk of intra-ocular infection.³

Over the last decade, the field of femtosecond eye surgery has expanded rapidly, supporting the advantage of combined high ablation precision and minimized collateral tissue effects.^4–9 The tissue can be removed cleanly and without thermal or mechanical damage by this effect. It was investigated and discussed by Teng et al., Stern et al., and Niemz et al.^10–14 By employing femtosecond laser with appropriate wavelength, photodisruption can be used to create subsurface incisions in the translucent sclera without damaging the overlying tissues. Femtosecond laser with a wavelength of 1700 nm has the ability to create partial-thickness intrascleral channels.^15,16 Femtosecond laser creates partial thickness fistulas that increased the rate of aqueous humor (AH) outflow in the cadaver eye and is mentioned in this report as AH drainage channel in the sclera.¹⁷

Perfusion experiment on the human cadaver eye was performed to explore the effect of femtosecond laser ablation drainage channels in the sclera on the outflow rate of aqueous humor (AH). Although perfusion on cadaver eye will increase the IOP, measurement results showed the decrease of IOP after the partial thickness intrascleral channels were created. The positive result indicates that IOP can be reduced by increasing the outflow facility with femtosecond laser created channels and that minimally invasive femtosecond laser treatments may have a future for surgical treatments of glaucoma.¹⁸

Dimensions and location of femtosecond-laser-induced partial-thickness intrascleral channels will affect the outflow rate resulting in IOP reduction. Analysis of the dimensions and location of such channels is important in understanding their effects. Observation of such channels will be important for optimization of laser treatment parameters and helpful for the future of femtosecond laser surgical treatments of glaucoma.

The two photon and confocal microscope integrates two functions: confocal microscopy and two-photon microscopy. A confocal microscope has the ability to optically section through a relatively thick, light-scattering tissue specimen to provide a three-dimensional (3-D) image, thereby eliminating the need for processing and sectioning procedures.^19,20 Two-photon microscopy forms an image from the nonlinear optical process induced by ultrahigh photon flux/multiphoton absorption. The second harmonic generation (SHG) signals from the sample are collected to form a microscopic image. This technology has a series of advantages, such as diffraction-limited resolution and deep depth discrimination; reduced photo-damage or bleaching effect outside the laser focus and without the need to fix, process, section, and stain the sample.^7,8,21,22 SHG signals can be obtained from sclera and used to study collagen organization.^23,24 In this report we use the confocal two-photon microscopy to investigate femtosecond-laser-created partial-thickness drainage channels in the sclera of human cadaver eye.

2.

Material and Methods

A femtosecond laser of wavelength 1700 nm was employed to create partial-thickness drainage channels in the sclera of a human cadaver eye. A confocal and two photon microscope was used to investigate the dimensions and location of these channels.

3.

Results

During laser irradiation, the sclera is ablated at the laser focus. This is because of the formation of cavitation bubbles, where the sclera tissue is transformed into gas and vapors. The gas or vapor either diffuses out of the tissue or is resorbed within hours. After femtosecond laser treatment, the eyeball was kept moist with 4% phosphate-buffered saline (PBS, pH 7.4) for about two hours to wait for the ablation plume to diffuse. Then the sclera was excised and fixed with freshly prepared paraformaldehyde (4% in PBS, pH 7.4) for future imaging process.

3.1.

Confocal Imaging

A rectangle intrascleral channel was created by a femtosecond laser 100 micron under the surface at a depth of 200 micron. Figure 2 shows a clear view of this subsurface channel at 110-micron depth count from the surface by confocal imaging. The image was obtained by illumination laser beam at wavelength 543 nm with $20\times (\mathrm{NA}=0.75)$ oil-immersion objective. The resolutions at both axes are 0.9 micron. The stack size has a width of 2303 micron and a length of 2764 micron. Most laser ablation surfaces are smooth and are not coagulated or damaged by thermal effect. There is still a little tissue remaining inside the channel, which illustrates the difficulty of subsurface laser ablation, and more research should be done focusing on the ablation laser energy and ablation scanning pattern.

Fig. 1

Schematic drawing of the two-photon confocal microscope experiment setup.

Fig. 2

Confocal imaging of subsurface channel at 110-micron depth.

Figure 3 shows the same intrascleral channel at 205-micron depth. All parameters are the same as in Fig. 2. Subsurface incision was cut without damaging the region adjacent to treated volume.

Fig. 3

Confocal imaging of subsurface channel at 205-micron depth.

A video of the whole channel was shown as Video 1 to show the total dimensions of AH drainage channel in the sclera.

3.2.

SHG Imaging

As a comparison with Fig. 3, a SHG imaging was observed at the same plane. All other parameters are the same except the wavelength of illumination laser beam is 800 nm. Figure 4 shows the SHG imaging.

Fig. 4

SHG imaging of subsurface channel at 205-micron depth.

A $40\times (\mathrm{NA}=0.75)$ oil-immersion objective was employed to investigate the edge of the intrascleral channel. The resolutions at both axes are 0.45 micron since a high-NA objective was involved. A series of SHG imaging was obtained to reconstruct the 3-D data sets. As shown in Fig. 5, SHG imaging is excellent in revealing intrascleral channel edge. The border region can be clearly recognized. The collagens adjacent to treated volume were untouched, and there is no indication of thermal damage.

Fig. 5

SHG imaging of subsurface channel edge.

Video 1

A video of the whole channel. (QuickTime, 186 KB). [ http://dx.doi.org/10.1117/1.JBO.17.8.081411.1]

4.

Discussion

Two-photon and confocal microscopy excel at high-resolution imaging hundreds of microns deep in the sclera of the human cadaver eye. It can generate confocal imaging or SHG imaging without processing and sectioning procedures. Because it avoids the shrinkage and distortion associated with conventional processing and sectioning procedures, it provides an important approach to evaluate dimensions and location of femtosecond-laser-induced subsurface channel in the sclera of human cadaver eye under the condition close to its physiological state. Such information will be important in understanding the femtosecond ablation effect in sclera on IOP reduction and helpful for the future of femtosecond laser surgical treatments of glaucoma.

As shown in Figs. 2 and 3, confocal microscopy provides a clear view. They demonstrated an excellent vertical ablation quality. The smooth ablation interface and high ablation precision confirmed the deterministic result from femtosecond laser ablation.

A lot of work has already been published concerning SHG imaging of the cornea.^{7,8,22,24,27,28} It is well known that the forward-scattered SHG is much more intensive than the backward-scattered SHG from the cornea. However, previous studies show that the intensity of the backward-scattered SHG from sclera is comparable to the forward-scattered SHG.²⁴ Our results also show that backward-scattered SHG imaging is possible to investigate the femtosecond laser intrascleral channel. Figure 4 shows that after femtosecond laser ablation, most collagen fibers adjacent to the intrascleral channel have remained nearly intact, and a minimal invasive surgery process was obtained. In a clinical situation, forward-scattered SHG imaging seems to be unrealistic for diagnostic purposes; backward-scattered SHG imaging may be a possible method to explore the femtosecond laser intrascleral channel.

Figure 4 shows SHG imaging from the same position as Fig. 3. The imaging is blurred in Fig. 4 compared with Fig. 3. It is difficult to acquire distinct imaging at the depth of the sclera. Two-photon microscopy penetrated less deeply into the sclera than the cornea, which is consistent with previous studies.²³ In contrast with corneal tissue, the sclera scatters visible light because Bragg scattering.²⁹ By proper preparation with dehydrating and a suitable wavelength chosen, the intrascleral channel can be introduced.^15,16,30,31 It is reasonable to hypothesize that we can use the same wavelength of 1700 nm as the femtosecond intrascleral channel created to generate SHG imaging of wavelength 850 nm in the sclera. Both wavelengths are more transparent for the sclera after proper preparation, and a distinct imaging will be obtained in deep depth of the sclera. The oscillator output at 800 nm is suitable to generate SHG imaging as commercial two-photon microscopy. Minimal invasive surgery and noninvasive diagnostics can be integrated in one femtosecond laser system for potential glaucoma femtosecond laser surgery. On the one hand, an amplified laser beam can produce an intrascleral channel to increase the outflow rate resulting in reduced IOP; one the other hand, an attenuated laser beam can introduce backward SHG imaging of such intrascleral channel. Therefore, therapeutic and diagnostic applications can be combined into one femtosecond laser and two-photon confocal microscope systems.