November 1985
Vol. 26/11
Investigative Ophthalmology
b Visual Science
A Journal of Dosic and Clinical Research
Articles
Interaction of Ultraviolet Laser Light with the Cornea
Ronald W. Krueger, Stephen L. Trokel, and Hermann D. Schubert
The effect of pulsed ultraviolet (UV) laser light on the cornea depends on wavelength (photon energy),
irradiance (photon flux), and pulse firing rate. At the available excimer laser wavelengths of 193, 249,
308, and 351 nanometers, the authors have varied the irradiance per pulse (10 to 2000 mj/cm2) as well
as pulse frequency (1, 10, 25 Hz) and determined the thresholds for coagulation and ablation of the
corneal stroma. The latter ablative action creates a groove resembling an incision and was present at
all wavelengths studied. The threshold for ablation increased for longer wavelengths and lower pulse
frequencies, except for 193-nm exposure, which was characterized by a constant threshold independent
of laser pulse rate. The grooves at 193 nm were both biomicroscopically and histologically smooth and
no coagulation effects were noted. Some degree of coagulation of adjacent tissues was noted at 249, 308,
and 351 nm. Invest Ophthalmol Vis Sci 26:1455-1464, 1985
Shortly after the corneal epithelium was found1 to
have an unusual sensitivity to 193-nanometer (nm)
laser light, this laser frequency was shown2 to be capable
of precisely etching plastics. It seemed reasonable that
this etching effect would similarly occur with biological
tissue. This was shown3 to be true in a series of experiments in which the corneal stroma was ablated with
an accuracy comparable to plastic. In this study, we
report the results of our examination of the interaction
of ultraviolet laser emissions with the cornea over a
range of pulse frequency and irradiance.
The excimer laser is the most convenient source of
ultraviolet (UV) laser light and has 193-nm, 249-nm,
308-nm, and 351-nm light as major emission lines.
These wavelengths span the UV spectrum and provide
a suitable range with which to examine corneal tissue
interactions. Each wavelength is produced by filling
the laser cavity with a different combination of a noble
gas and a halogen gas (Table 1). The far ultraviolet line,
at 193 nm is produced by charging the laser with argon
and fluorine gases; 249 nm, with krypton and fluorine;
From the Department of Ophthalmology, College of Physicians
and Surgeons, Columbia University, New York, New York.
Submitted for publication: April 1, 1985.
Reprint requests: Stephen Trokel, MD, 635 West 165th Street,
New York, NY 10032.
1455
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
308 nm, with xenon and chlorine; and 351 nm, with
xenon and fluorine. The gases react together in a high
voltage field to produce an excited diatomic molecular
species (hence "excimer" for "excited dimer"), which
is the active laser medium.
These four wavelengths can be produced at sufficiently high irradiance to cause corneal tissue ablation.
The mechanism of this effect is presumed similar to
the ablation studied in synthetic plastics and other organic polymers. It has been hypothesized that at 193
nm, laser light interacts with the macromolecules of
organic polymers by breaking intramolecular bonds
and setting the resulting fragments into motion. When
the irradiance is above threshold, the molecules will
decompose before bond reformation can occur. The
phenomenon as described by Srinivasan4 is termed
"ablative photodecomposition" and is seen to produce
a localized removal of material exactly corresponding
to the laser irradiated area.
This sharply controlled cutting ability of the excimer
laser is in contrast to the longer wavelength infrared
lasers which are also absorbed by corneal tissue. The
CO 2 laser, emitting light at 10,600 nm, produces an
incision with irregular jagged edges, which chars the
collagenous material of the stroma.5 This effect is quite
unlike the uniform incision produced by the 193-nm
emission of the excimer laser. Consequently, the ex-
INVESTIGATIVE OPHTHALMOLOGY b VI5UAL SCIENCE / November 1985
1456
Table 1. Excimer laser output
Wavelengths (nm)
Gas mixture
193
249
308
351
ArF
KrF
XeCI
XeF
cimer laser shows potential for corneal surgery. However, before clinical applications can be explored, a
thorough investigation of the physical parameters of
corneal interaction is needed.
Materials and Methods
The excimer laser used in this study was the LambdaPhysik (Acton, MA 01720) model EMG 102E. The
output for the four major emission lines was rated at
100 to 250 mj per pulse with pulse widths of 10 to 16
nsec duration. The laser emits a rectangular beam of
light of gaussian distribution which can be focused and
controlled with a supersil lens and mask aperture. The
mode structure of the beam can be varied by using
either stable or unstable resonator optics in the cavity.
We used unstable resonator optics to decrease the beam
divergence and a cylindrical lens with a 14 mm X 1
mm rectangular mask to focus the laser light into a
narrow high energy slit.
The experiments were carried out by irradiating the
corneas of freshly enucleated calf eyes. The irradiance
Vol, 26
values were determined by recording the amount of
energy exposed to the cornea and the area of laser exposure as measured from photographic film. The average energy per pulse was measured with a GenTec
Joulemeter (Ste-foy; Quebec, Canada) both before and
after corneal irradiation. The energy density (irradiance) was varied by changing the focal distance and
consequent area of the slit at the irradiation site.
The effects of the laser energy on the cornea were
observed for the different wavelengths at several laser
pulse rates. Laser pulse rates of 1 Hz, 10 Hz and 25
Hz were tested to observe differences in corneal tissue
interaction. Threshold measurements were made for
the observed corneal effects of discoloration or whitening, coagulation, and ablation. Measurements were
made by varying the irradiance per pulse at each pulse
rate and wavelength considered. The threshold were
determined by observing the effects after 2000 to 3000
laser pulse exposures. The effects were examined by
gross observation and inspection with a slit lamp, classified, and graphically analyzed to determine the
threshold. Photographs were taken of representative
examples. After laser irradiation and observation, the
eyes were placed in formalin and prepared for histopathologic analysis.
Results
The corneal tissue response to U V laser light follows
a general pattern of tissue interaction. At low irradiance
Fig. 1. Linear cracks (arrows) extending laterally
from a 249-nm ablated incision. These appear shortly
above threshold and disappear at higher irradiances.
They have been unique to the
249-nm irradiated tissues.
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
1457
ULTRAVIOLET LASER EFFECT ON CORNEA / Krueger er ol.
No. 11
Table 2. Irradiance thresholds (mJ/cm 2 ) of UV laser light on calf corneas
Wavelength (nm)
Opacity
Coagulation
lHz
193
249
308
351
25 ± 5
<10
30 ± 15
360 ± 40
XXX
XXX
400 ± 50
780 ± 80
50 ±
185 ±
540 ±
1050 ±
13
20
80
100
lOHz
193
249
308
351
22 ± 6
<10
25 ± 15
150 ±30
XXX
200 to 400*
150 ±25
600 ± 50
50 ±
120 ±
500 ±
1000 ±
10
20
50
100
25Hz
193
249
308
351
20 ± 5
<10
25 ± 15
100 ± 30
XXX
150 to 300*
150 ±20
350 ± 50
55±
80 ±
420 ±
900 ±
10
25
40
100
Ablation
Corresponds to coagulation range rather than threshold value.
levels, the cornea remains grossly unaffected by the
laser energy. As irradiance of the laser exposure is progressively increased, we observe a faint clouding of the
exposed corneal tissues. This ranged from surface discoloration at 193 and 249 nm to a penetrating stromal
clouding at 308 and 351 nm. With further increased
irradiance, coagulation effects were seen at all wavelengths except 193 nm. These effects were manifest at
249 nm by the appearance of linear tracks of fine bubbles radiating away from the exposed tissues. At 308
and 351 nm, large superficial bubbles were seen with
brown discoloration or charring of the exposed surface.
The adjacent unexposed tissues showed distortion and
shrinkage.
This was distinguished from the results of yet higher
irradiances, when an ablation of corneal tissues was
found which grossly was restricted to the area of laser
exposure. The pattern has a unique deviation at 249
nm, in that the fine linear bubbles were not present at
irradiances below the ablation threshold. Rather, the
bubbling effects and tissue ablation occurred together
within a short range of irradiance above threshold. Fine
linear bubbles are seen radiating outward (Fig. 1) from
the ablated area within this range of irradiances. With
increased irradiance, this bubbling effect no longer occurs. The cutting thresholds at 249 nm vary widely
with pulse rate, having more than a twofold threshold
difference at 1 Hz (185 ± 20 mj/cm2) compared to 25
Hz (80 ± 25 mj/cm2). The irradiance thresholds at
which each of these effects are first recognized are listed
in Table 2. The threshold values decrease as the laser
pulse rate is increased, with the exception of tissue
ablation at 193 nm.
At 193 nm, coagulation phenomena were not observed under any condition of exposure. At low irradiance values, only a slight surface clouding of the cor-
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
neal epithelium was seen. As the irradiance was increased, tissue ablation occurred with no intervening
tissue coagulation. The ablation threshold remained
constant (50 ± 10 mJ/cm 2 ) as the laser pulse rate was
increased, in contrast to the decrease measured at longer wavelengths.
2,000 r
1,000
^500
!
5
IOO
50
160
200
240
280
320
360
X(nm)
Fig. 2. Irradiance in millijoules per square centimeter for ablative
action of the four major wavelengths of the excimer laser. Note the
linear semilogarithmic relationship at 1 Hz. The departure from the
logarithmic relationship with lowering of the threshold at 249 nm
implies a thermal component of the laser-tissue interaction.
1458
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / November 1985
Vol. 26
\
'
•
*
Fig. 3. Photomicrograph of a 193-nm
laser lesion. The lesion produced is similar at all supra-ablation irradiances and
laser firing rates. Note the smooth featureless wall with no distortion of the
adjacent corneal stroma (X24),
mi '
WT.it
Graphic analysis of the ablation threshold data reveals a logarithmic relationship with wavelength. In
Figure 2, we see a linear increase in the log of the
threshold at 1 Hz as the wavelength increases. However,
as the pulse rate increases, threshold values of ablation
tend to decrease with the exception of the 193 nm
threshold. This decrease is most prominent at the 249nm wavelength. At the 193-nm wavelength, the curves
converge to one point and there is no threshold variation with pulse rate.
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
The irradiated corneas at each of the observed wavelengths were examined histologically. At 193 nm, all
tissue sections studied showed smooth uniform walls
(Fig. 3), without regard to pulse rate or supra-ablation
irradiance level. These smooth even walls are in marked
contrast to the texture of the stroma seen at 249 nm.
In Figure 4A, the walls of the tissue ablated at 249 nm
are irregular and vacuolated, yet roughly parallel. As
the irradiance is increased, the incision bulges (Fig. 4B),
probably due to increased gas pressure preventing rapid
No. 11
ULTRAVIOLET LASER EFFECT ON CORNEA / Krueger er ol.
1459
Fig, 4, The stromal lesions produced with 249nm laser radiation. A, A linear incision with parallel
walls produced at supra-ablation irradiance. B, The
bulging of the walls as the irradiance is increased.
The bulging is presumably due to increased back
pressure of gases formed by the greater thermal interaction. C, The lesion produced when both irradiance and firing rate is increased. The walls and
floor of the lesion show increasing thermal effects
with actual melting and dissolution of the stroma
(X24).
clearing of ablated products from the incision depths.
In Figure AC, the irradiance and pulse rate were both
increased which caused a considerable thermal effect
with melting of the walls and floor of the ablated
stromal tissue.
Corneal coagulation is dramatically demonstrated
when a rapidly pulsed laser emitting 308 nm light of
low irradiance burns the cornea (Fig. 5A). An extensive
area of corneal coagulation surrounds a central charred
zone, and distortion of the cornea radiates from the
coagulated area. Histologically, this is seen in Figure
5B as extensive disorganization and destruction of the
stroma. When the irradiance is raised above ablation
threshold, linear incisions are formed which appear
grossly similar (Fig. 6A) to those incisions formed with
193 and 249 nm light. Histopathology (Fig. 6B) shows
the stromal disorganization due to thermal interaction
adjacent to the ablated area at this wavelength.
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
Low irradiance exposure at 351 nm causes through
and through whitening of the cornea along the laser
transmission path. As the radiance increases, corneal
burning occurs. Histopathology (Fig. 7A) of 351-nm
corneal laser burns shows extensive thermal destruction
of corneal tissue. At supra-ablation irradiances and low
firing rates, tissue is removed with a narrow zone of
thermal disorganization (Fig. 7B). When the laser firing
rate is increased to 25 Hz, the thermal interaction zone
surrounding the ablated area is considerably broader
(Fig. 7C).
Discussion
The mechanisms by which tissues are ablated when
exposed to high irradiances of ultraviolet light are not
clear and controversy surrounds various current models. Both thermal and photochemical interactions have
1460
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / November 1985
Vol. 26
Fig. 5. A. A corneal photograph of a 308-nm irradiated cornea below ablation threshold. The central exposure (arrow) at irradiance below
ablation threshold shows white coagulation surrounding a brown zone of char with distortion of surrounding cornea, B, A histologic section
which shows the broad zone of necrosis, lammellar disorganization, and swelling of the central lesion of Figure 5 A (X48).
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
ULTRAVIOLET LASER EFFECT ON CORNEA / Krueger er ol.
No. 11
14tM
B
Fig. 6. If the irradiance at 308 nm is increased, incisions are produced as in A. The narrower irradiated areas (arrows) have higher irradiances
and grossly resemble those formed at 249 and 193 nm. Distortion of the stroma in B (X24) provides evidence for heating of adjacent tissue in
contrast to the undisturbed adjacent stroma seen for 193 nm exposure (Fig. 3).
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
1462
INVESTIGATIVE OPHTHALMOLOGY G VISUAL SCIENCE / November 1985
Vol. 26
B
Fig. 7. A, Extensive disorganization of the stromal structure by 351-nm ultraviolet light below the ablation threshold. There is widespread heating and destruction of the corneal stroma. B, A low firing rate exposure of 351-nm light above the ablation threshold shows tissue
removal with local changes in the stroma. Gas bubbles are seen extending into the stroma, and an obvious thermal interaction zone can be
identified (X24).
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
No. 11
ULTRAVIOLET LASER, EFFECT ON CORNEA / Krueger er al.
1463
Fig. 7C. The firing rate and irradiance are higher and a broader thermal zone is identified (X24).
been postulated to occur in varying combinations.
Certain of our findings support these mechanisms both
in the appearance of the irradiated cornea and the
threshold data.
At 193 nm, no tissue changes suggesting thermal
interaction are observed. Stromal integrity is intact adjacent to laser ablated sites. Low irradiance exposures
produce slight epithelial clouding with no discernible
change at light microscopic resolution, but irradiances
greater than 50 mJ/cm 2 cause tissue ablation. The 193nm ablation threshold is independent of the laser firing
rate, and suggests that heating of the irradiated tissues
is not an important factor. The low ablation threshold,
absence of a low irradiance coagulation effect, and the
independence of threshold with laser firing rate supports the photochemical theory of 193 nm laser tissue
interaction proposed earlier. The tissue loss therefore
follows the exact path of laser light exposure.
UV light of 249 nm can ablate the cornea as high
irradiances, but unwanted remote effects occur. Linear
tracts of bubbles extend from the edges of the irradiated
tissues at 249 nm and implies an explosive thermal
interaction. Histologic evidence supports this effect
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
with the appearance of incision bulging at higher irradiances (Fig. 4B) and stromal dissolution of adjacent
tissues (Fig. 4C). This is in marked contrast to 193 nm
where no thermal or distant effects could be demonstrated. The decrease in ablation threshold with increasing laser pulse rate (Fig. 1) is further evidenced
which suggests that ablation with 249 nm light has a
thermal component.
Laser light of 308 and 351 nm is largely transmitted
by the cornea. As irradiance levels are raised, corneal
absorption is sufficiently great to burn the cornea with
visible bubbling, burning, and charring. At higher irradiances, there is actual removal of corneal tissue
which grossly resembles the ablated lesions seen at 193
and 249 nm. The threshold decrease with increasing
laser pulse rate also suggests a thermal component to
ablation at higher laser pulse rates, the rapid light delivery allows less time for energy dissipation between
pulses, facilitating a localized temperature increase.
Histologic analysis shows considerable stromal dissolution with distortion of the collagen matrix. It is clear
that a thermal mechanism is involved in tissue removal
at these longer wavelengths, but it is not known whether
1464
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / November 1985
a concomitant photochemical effect is also present. The
thermal effects of longer wavelengths make precise
corneal cutting difficult to achieve. However, these
longer wavelengths may be desirable for other clinical
situations where tissue removal with heating would be
useful.
The relationship of decreasing ablation threshold
with decreasing wavelength of UV light (Fig. 2), is due
to the increased energy of the far UV photons. At low
laser pulse rates, ablation threshold logarithmically declines with decreasing wavelength. The relationship is
expressed as:
•Og It(abl) = « X X uv
where It(at>i) is the irradiance threshold for ablation, a
is the slope of the line, and Xuv is the irradiating wavelength of UV light.
The linear relationship alters as the pulse rate increases with maximal change at 249 nm. At this wavelength, nonphotochemical absorption produces heating
which explains the lowered ablation threshold as the
Downloaded From: http://iovs.arvojournals.org/ on 07/31/2017
Vol. 26
pulse rate increases. At 193 nm wavelength, the ablation threshold is independent of energy delivery rate
suggesting that heat buildup is not an interaction factor.
Key words: excimer laser, cornea, thresholds, histology, photoablation
References
1. Taboada J and Archibald CJ: An extreme sensitivity in the corneal
epithelium to far UV ArF excimer laser pulses. Proceedings of
the scientific program, Aerospace Medical Association, 1981,
San Antonio, Texas.
2. Srinivasan R and Mayne-Banton V: Self-developing photoetching
of Poly (ethylene terephthalate) films by far UV excimer laser
radiation. Appl Phys Lett 41:576, 1982.
3. Trokel SL, Srinivasan R, and Braren B: Excimer laser surgery
of the cornea. Am J Ophthalmol 96:710, 1983.
4. Srinivasan R and Leigh WJ: Ablative photodecomposition: action
of far-ultraviolet (193 nm) laser radiation on poly (ethylene terephthalate) films. J Am Chem Soc 104:6784, 1982.
5. Keates RH, Pedrotti L, Weichel H, and Possel W: Carbon dioxide
laser beam control for corneal surgery. Ophthalmic Surg 12:117,
1981.