Nonmechanical trephination cut with energy of 100 mJ per pulse, 1.3-mm spot size, and 2-Hz repetition rate. The external ring represents the cut edge. Top, Recipient cornea with metal mask. Note the presence of "orientation teeth" at the cut edge (arrows). Bottom, Smoother cut edges with ceramic mask (arrows) (original magnification ×8).
Macroscopic appearance of recipient cornea. Note the relatively smooth cut edge (arrows) with a mild white zone of stromal thermal effect increasing with depth (arrowheads). Laser settings were 400 mJ per pulse, 3.2-mm spot size, and 2-Hz repetition rate. A metal mask was used (original magnification ×16).
Stromal thermal effect at the edge of the cut (arrows) in a donor cornea with the use of a ceramic mask and laser settings of 200 mJ per pulse, 2-Hz repetition rate, and 3.2-mm spot size (periodic acid–Schiff, original magnification ×400).
Transmission electron microscopy of the cut edge showing 2 distinct layers: an electron-dense area of collagen fiber disarrangement, with fuzzy contours in the area of the cut (between arrowheads), and a granular material layer at the laser-exposed area (between arrows) (100 mJ per pulse, 2-Hz repetition rate, and 1.3-mm spot size) (bar indicates 3 µm; original magnification ×4500).
Scanning electron microscopy of the cut edge. Left, Cut edge of a donor button, showing the regularity of the cut with ceramic mask and the focal and regular ablative process in all depths of the stroma (s; arrows) with the use of 200 mJ per pulse, 2-Hz repetition rate, and 3.2-mm spot size; e indicates epithelium (bar indicates 300 µm). Right, Recipient bed with ceramic mask, with a smooth cut surface (arrow) with the use of 200 mJ per pulse and the same repetition rate and spot size (bar indicates 100 µm).
Behrens A, Küchle M, Seitz B, Langenbucher A, Kus MM, Amann T, Schlötzer-Schrehardt U, Rummelt C, Naumann GOH. Stromal Thermal Effects Induced by Nonmechanical (2.94-µm) Erbium:YAG Laser Corneal Trephination. Arch Ophthalmol. 1998;116(10):1342-1348. doi:10.1001/archopht.116.10.1342
To determine stromal thermal changes after erbium (Er):YAG laser corneal trephination with the use of 2 open masks.
Corneal trephination was performed in 89 enucleated pig eyes with an Er:YAG laser (400-microsecond pulse duration), 4 open masks (2 metallic and 2 ceramic) for both donors and recipients, and an automated globe rotation device. Different combinations of laser settings were used: pulse energy, 100, 200, and 400 mJ; repetition rate, 2 and 5 Hz; and spot size, 1.3 and 3.2 mm. Thermal effects in corneal stroma and regularity of the cut edges were quantitatively assessed by light microscopy, transmission and scanning electron microscopy.
Best regularity and minimal thermal effects of the cut were observed with the use of ceramic masks at 200 mJ, 2 Hz, and 3.2-mm spot size, with middepth thermal changes of 18±2 µm. Effects increased with cut depth and were lower in donor corneas and with the use of ceramic masks (P<.001). Regularity of the cut was higher in the donors (P=.05) with lower repetition rates (P<.001).
Even with the "free-running" Er:YAG laser mode, features of the trephination cut resembling those created by the 193-nm excimer laser along metal mask were achieved. Ceramic masks may be more suitable than metal masks. The Er:YAG laser seems to have the potential to be a compact and low-cost alternative in nonmechanical trephination for penetrating keratoplasty.
Thermal effects after corneal trephination with the free-running Er:YAG laser (2.94 mm) are limited and predictable.
DESPITE MANY advances in corneal transplant surgery, postoperative visual outcome is still limited by corneal surface irregularity. Multiple confounding factors seem to be involved in the appearance of astigmatism. From our clinical studies on nonmechanical 193-nm excimer laser trephination for penetrating keratoplasty (PK), we have evidence that suturing technique is a most important factor for the immediate and midterm postoperative period. However, corneal wound discrepancy caused by mechanical trephination is one of the most important factors influencing the "suture-out" optical results after PK.1- 5 Different mechanical devices have been developed to make this procedure precise and standardized, but results are still controversial.6- 8 With the development of excimer lasers as surgical instruments because of their ablation properties,9,10 improved nonmechanical methods for corneal trephination are currently a promising alternative to conventional techniques since they reduce the disparity of the cut edge, subsequent corneal topographic distortions, and difficulties in the mechanical procedure itself.11- 17 In addition, nonmechanical trephination with open metal masks allows performance of different shapes of trephination and changes in the configuration of the cut, making elliptical trephination with variable cut angles a new possibility.18- 21 After the development of nonmechanical trephination for PK, more than 304 such procedures were successfully performed in our department.
Nonetheless, the use of excimer laser for corneal surgery has disadvantages. Presence of toxic gases in the operating room, high operative and maintenance costs, and the bulky size of the equipment are drawbacks inherent in this type of laser and have spurred researchers to investigate other delivery systems.
Mid-infrared laser radiation has been studied for the ablation process in corneal surgery.22- 25 Hence, the erbium (Er):YAG laser (2.94 µm) has recently received much attention because of its low penetration in corneal tissue and relatively low thermal damage.26,27 Its wavelength corresponds closely to the highest absorption coefficient of water, the main constituent of the cornea, which could ensure low optical penetration to other segments of the eye.28
Superficial keratectomy with this device has been performed in an attempt to reproduce the results obtained with excimer laser refractive surgery.29- 32 However, higher ablation rates are common undesired effects with Er:YAG laser for this type of intervention31 but could represent an additional advantage for corneal trephination.33
The purpose of this study was to perform nonmechanical Er:YAG laser corneal trephinations in corneas of enucleated pig eyes, similar to the technique with the 193-nm excimer laser for PK currently used for all nonvascularized corneal processes in our department.1,12,34,35 Thermal effects in the corneal stroma and regularity of the trephination edges were assessed with 2 mask materials and different combinations of laser settings.
A total of 89 pig eyes, enucleated immediately after death, were obtained from the local slaughterhouse and used for Er:YAG laser corneal trephination, following the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research. The intraocular pressure was increased to 18 to 25 mm Hg (Schiötz tonometer) by injecting saline solution through the optic nerve with a 26-gauge cannula. The globes were placed in a metallic cylindrical support, part of a motorized device rotating at a constant speed of 1 rotation per minute. After the globes were in place, 1 drop of saline solution was instilled on the corneal surface and a mask was positioned over the cornea and centered according to the pupillary aperture.
Two mask materials were used: stainless steel and ceramic (silicon carbide hardened with aluminum). From each material, masks for donors (covering the central cornea) and recipients (covering the peripheral cornea) were used. Metal masks had 8 peripheral orientation teeth of 0.30×0.15 mm in donor masks, with correspondent notches in the recipient masks.1,35 Ceramic masks lacked both orientation teeth and notches. The donor masks had a diameter of 8.1 mm, were provided with a central opening of 3 mm, and weighed 0.18 g (metal) and 0.06 g (ceramic). The recipient masks had a diameter of 13 mm, were provided with a central opening of 8.0 mm, and weighed 0.4 g (metal) and 0.13 g (ceramic). All masks were 0.5 mm thick.
The laser delivery system consisted of a 2.94-µm-wavelength Er:YAG device (NWL Laser Technologie GmbH, Eckental, Germany), with a 650-nm diode laser as aiming beam. This laser emits fluences of up to 800 mJ at a fixed pulse duration of 400 microseconds. Variations of 3 energy settings, 2 spot sizes, and 2 repetition rates were used (Table 1). The laser equipment with its specular transmission system was fixed by a metal stand in a 90° angle to the surface of the cornea. The globe rotation device was placed under the laser arm, and the guiding beam was directed to the rim of the mask, with approximately half of the spot on the mask and half on the corneal surface.
With the rotation device in action, the trephination process was performed under careful visual control. The procedure was stopped immediately as soon as aqueous was observed in the trough, as a sign of local corneal perforation.
The globes were fixed in a buffered 10% paraformaldehyde solution, and macroscopic analysis was performed 24 hours after fixation. Globes were bisected, embedded in paraffin, cut in 8-µm sections, and stained with periodic acid–Schiff. All slides were analyzed by light microscopy, and regularity of the cut edges and the extension of thermal damage were recorded in a masked fashion. Regularity was semiquantitatively assessed in 4 levels, in ascending order of the degree of irregularity: 1 (regular cut edge), 2 (mild irregularity), 3 (moderate irregularity), and 4 (highly irregular cut edge). Thermal damage was defined as the thickness of the dark staining zone present in contiguity with the cut edge. Measurements were taken in 3 stromal depth levels: upper (150 µm), intermediate (400 µm), and lower (700 µm).
Selected cases were submitted for transmission and scanning electron microscopy. For transmission electron microscopy, specimens were postfixed in 2% buffered osmium tetroxide, dehydrated in graded alcohols, and embedded in epoxy resin (Epon 812, FLUKA, Buchs, Germany). Semithin sections were stained with toluidine blue; ultrathin sections were stained with uranyl acetate–lead citrate and were examined with a transmission electron microscope (EM9 A, Carl Zeiss, Oberkochen, Germany). A quantitative analysis of the thermal effect at the cut edge as well as cellular changes of keratocytes in the vicinity of the cut was performed.
For scanning electron microscopy, specimens were dehydrated in graded acetone 50% to 100%, dried to critical point, sputtered with gold, and examined with a scanning electron microscope (CAM SCAN, Dortmund, Germany). Comparisons were performed between different mask materials.
All data were entered in a common relational database system (Microsoft Access 2.0, Unterschleissheim, Germany). For statistical analysis, SPSS/PC 6.1.3 (Windows) was used. Measurements of variables in different groups of laser settings were described with mean, SD, and minimum and maximum values. Comparisons between groups or variables were performed with nonparametric tests (Mann-Whitney U test for unpaired samples, Wilcoxon test for paired samples). For bivariate correlation analysis, Spearman rank correlation coefficient r was used. For formulation of regression equations, linear regression coefficients were calculated. A P value of .05 or less was considered statistically significant.
Trephination time until focal perforation was achieved varied from 61±5 seconds in the 5-Hz repetition rate, 15-J/cm2 group, to 7 minutes 15±8 seconds in the 2-Hz repetition rate, 2.5-J/cm2 group. The total time of the trephination procedure until corneal perforation correlated with the energy density used (P<.001). The total energy used until perforation occurred varied from 49±1 J to 179±16 J. With the laser settings at 100 mJ per pulse and 3.2-mm spot size, no ablation could be obtained since it was below the threshold, whereas the groups with 400 mJ per pulse and 1.3-mm spot size resulted in spontaneous focal perforation before the laser completed the first circumference. These last 2 series were excluded from statistical analysis.
A qualitative relation of the macroscopic thermal effect and regularity of the cut with the mask material was observed. Regularity and smoothness at the cut surface were higher with the ceramic mask, in cases of low pulse energy and low repetition rate settings (Figure 1). A higher repetition rate of 5 Hz resulted in a more irregular undulating appearance of the cut edge with both mask materials. A stromal white zone of thermal effect was observed in bisected globes, increasing from superficial level to deeper stromal lamellae, and reaching up to 0.8 mm thick from the cut edge. This zone was more pronounced in cases with high pulse energy (400 mJ), metal masks, and recipient corneas (Figure 2). In addition, stromal carbonization detected by the presence of brown remnants of irregularly distributed burned tissue on the cut surface was observed when high pulse energy was applied.
Thermal changes were represented by a dark band of violet staining in contiguity with the cut edge. This band appeared to increase with the depth of ablation, and a highly significant difference in thickness (P<.001) was noted between each of the 3 depth levels of the cut. A statistically significant correlation between energy density and maximum thermal effect was detected (P=.04). These effects were more pronounced in recipient corneas (P<.001).
In donor series (Figure 3), thermal effects of 13±2 µm superficially, 18±2 µm in the intermediate depth, and 22±5 µm at the bottom of the cut edge were observed with ceramic mask (200 mJ per pulse, 2-Hz repetition rate, and 3.2-mm spot size), representing the lowest values detected in donors. A significant correlation between the extension of thermal effect and pulse energy was found with both types of donor mask materials (P<.001).
In recipient series, thermal effects of 32±10 µm superficially, 45±8 µm intermediate, and 57±20 µm at the bottom depth of the cut were observed with the ceramic mask (200 mJ per pulse, 2-Hz repetition rate, and 1.3-mm spot size), and were the lowest values detected for recipients. Despite the existence of lower values of thermal effects in the intermediate level using recipient mask, this group showed the lowest dispersion of thermal change values comparing the 3 cut levels. A correlation with the extension of thermal changes and pulse energy could not be confirmed statistically in recipients (P=.31). However, a stromal zone of thermal alteration was observed below the contact area of the recipient metal mask to the cornea, at the rim of the cut, in a depth from 50 to 150 µm and increasing with the pulse energy, when repetition rate was 5 Hz. These changes were absent at the stromal level when the ceramic mask was used. Similar changes were observed with donor masks, at 2 points close to the center of the corneal buttons, where the central hole of the mask contacts the corneal surface. They reached up to 80 µm in depth and were absent in donor buttons trephined with ceramic mask. Results of thermal change at intermediate level of the cut are summarized in Table 2.
Vacuoles were observed between collagen lamellae and inside keratocyte cytoplasm in the cut edges, and were located more frequently in deeper levels of ablation. They were more pronounced with high pulse energies. Keratocytes in the zone of thermal effect showed a vacuolated and bulked cytoplasm, with the nuclei displaced to the periphery.
In terms of the regularity of the cut, in none of the cases was a regular pattern of a score of 1 achieved. The more regular cut surface observed macroscopically with the ceramic mask was confirmed histologically (P=.008). Regularity was higher with donor trephination (P<.001) and was found to be inversely correlated with the repetition rate (P<.05). The lowest values of irregularity (2.3±0.5) were found in donors and ceramic mask (200 mJ per pulse, 2-Hz repetition rate, and 3.2-mm spot size), which showed the least thermal effects (Table 3).
Interesting findings were spike-shaped dark-staining areas, identified by light microscopy, which represented an expansion of the thermal effect perpendicular to the cut and parallel to the corneal lamellae, extending up to 86±26 µm, especially with recipient trephination. The size of these spikes was correlated with the pulse energy (P=.02) and inversely with the spot size used (P=.04).
Transmission electron microscopy showed 2 distinct zones of thermal effect. One was an external coagulation zone (3-6 µm thick), closely in contact with the laser radiation, consisting of a coagulated material of granular, flocculent to amorphous appearance distributed along the stromal cut edge. Within this zone, virtually no keratocytes could be observed. The second zone showed a greater variability in thickness (9-15 µm) and consisted of an electron-dense zone with less sharply defined "fuzzy"collagen fiber contours, embedded in an amorphous, dense matrix, as well as disarrangement and separation of the lamellae or fiber bundles (Figure 4). Keratocytes in this zone showed changes in intracellular organelles and vacuolization of cytoplasm. These intracellular alterations seemed to be confined to the electron-dense zone previously described.
With scanning electron microscopy, the samples showed the focal nature of the ablative process in all depths of the cornea. The interlamellar spaces appeared swollen, and the presence of a coagulated material was seen at the cut edge, most likely corresponding to the granular material detected by transmission electron microscopy. The regularity of the ablation was greater in the ceramic mask specimens, and grooves like those frequently observed with metal mask on the surface of the cut edges were absent in the ceramic mask specimens (Figure 5).
Corneal trephination with 193-nm excimer laser as a noncontact trephine system for PK has shown potential benefits over the mechanical procedure.34- 36 Less persisting astigmatism after suture removal, more regular corneal topography, and higher best-corrected visual acuity have been achieved1,12; therefore, this method of trephination is now routinely applied for nonvascularized corneal processes. In an attempt to attain similar results with the advantages of a more user-friendly solid-state laser, we studied the potential of an Er:YAG laser for nonmechanical trephination for PK. Results obtained with this in vitro series showed that thermal effects induced by the trephination procedure in porcine corneal stroma were limited and predictable, dose-dependent by the laser energy-density level. On light microscopy, thermal changes were confined to a small zone of darker staining adjacent to the cut edge, corresponding to electron-dense irregular areas of collagen fiber disarrangement detected by transmission electron microscopy. Therefore, we used these areas on light microscopy to quantify the extent of the zones of thermal effect. The observed changes increased with the depth of ablation, probably because of cumulative laser energy exposure and induced dehydration in deeper areas of the cornea.
Energy density variables were positively correlated with the ablation rate, since the time to reach the perforation varied according to the fluences. In theory, it would be ideal to use high-energy densities to shorten the trephination time, but this increased the maximum thermal effect.
In our series, both thermal effect and irregularity of the cut edges of trephination were more pronounced with higher repetition rate and energy density in recipient corneas. Changes related to the heating of the mask were more pronounced with metal masks. These effects were reduced with the use of ceramic masks in both recipients and donors, where energy absorption was considerably lower (Table 4).
Temperature increase has been studied in corneal trephination with argon-fluoride excimer laser. Results indicated a greater increase when donor trephination was performed, and an inverse correlation between the temperature increase and the speed of the rotation.37 In contrast, thermal effects with Er:YAG laser appear to be lower in donor than recipient corneas at a constant speed of 1 rotation per minute.
Changes induced by Er:YAG laser exposure in corneal epithelial and endothelial layers with this technique are beyond the scope of our present study. Nevertheless, Peyman et al23 reported no changes in endothelial cells of rabbit corneas with ablation depths up to 300 µm for photorefractive procedures. The incisions performed with nonmechanical trephination are certainly deeper, and effects of radiation are expected to be likewise higher on the endothelial layer, as detected by Koch et al38 with perforating incisions with the use of the 193-nm excimer laser. Currently, studies to determine cellular alterations in this layer are being performed.
At first glance, our results seem to be inferior to the precision and low thermal effect possible with the excimer laser. Although there are some differences between corneal trephination for PK and laser refractive procedures, where the optical performance with minimal damage to the central cornea is crucial, in PK the cut is made far away from the visual axis and, thus, a certain amount of thermal effect causing opacity may not represent a serious drawback. However, possible effects in the wound healing process have yet to be analyzed, since thermal changes could stimulate greater degrees of postoperative inflammation resulting in wider scars that could induce irregular astigmatism. Previous experiments have shown that minor degrees of thermal effect in corneal stroma neither altered the outcome of corneal ablation nor adversely influenced wound healing in monkeys subjected to lamellar refractive procedures.29,39 Nevertheless, we find it necessary before attempting any human clinical trial to perform in vivo animal studies after nonmechanical trephination with Er:YAG laser to determine the exact role of these thermal changes in the corneal reparative process. To our knowledge, the wound healing response has not been previously assessed in animal models with this technique.
Alterations in cellular components under direct exposure to the laser emission were seen in our study, but these changes appeared to be confined within the limits of thermal change band.
The use of lasers with shorter pulse duration may be a possible factor in limitation of temperature effects on corneal tissue in the future. Some studies have reported no changes with pulse variation on the order of microseconds, but others indicated wide differences when the variation reaches the order of nanoseconds (Q-switched mode).26,40 This could be a promising alternative for reducing the thermal effects of the Er:YAG laser in the "free-running" mode.
The preliminary results obtained with our first in vitro series with the Er:YAG laser in nonmechanical trephination for PK show that, although still far from ideal, this procedure represents an encouraging step forward in the search for a nonmechanical, standardized technique with a compact machine, lower costs, and easier handling.
Accepted for publication April 16, 1998.
This study was supported by grant 229/97 from the Bayerische Forschungsstiftung, Munich, Germany, and grant 331 4 04 001 from the German Academic Exchange Service, Bonn, Germany (Dr Behrens).
We thank Erich Weimel from NWL Laser Technologie GmbH, Eckental, Germany, for providing the laser system and giving technical advice.
Reprints: Berthold Seitz, MD, University Eye Hospital, Schwabachanlage 6, D-91054 Erlangen, Germany (e-mail: email@example.com).