Journal of Refractive Surgery

The articles prior to January 2012 are part of the back file collection and are not available with a current paid subscription. To access the article, you may purchase it or purchase the complete back file collection here

Comparison of Yellow Dye, Continuous Wave Nd=YAG, and Argon Green Laser on Experimentally Induced Corneal Neovascularization

Neal M Krasnick, MD; Alan V Spigelman, MD

Abstract

ABSTRACT

BACKGROUND: Corneal neovascularization is generally undesirable because it can lead to corneal scarring, lipid deposits, and corneal graft failure. To eliminate these vessels, several techniques are available including laser photocoagulation.

METHODS: This prospective study was designed to compare the effectiveness of three laser wavelengths (continuous wave Nd:YAG, 1064 nm; argon green 514 nm; and yellow dye, 570 nm) to obliterate experimentally induced corneal neovascularization in the rabbit. Corneal vascularization was created in 12 rabbits by placing 7-0 silk sutures through two quadrants of the cornea. Once neovascularization was complete, the suture was removed and one of the three lasers was applied to occlude vessels at one of the neovascular sites. The other site was used as a control.

RESULTS: The yellow wavelength, when compared with the green required fewer exposures to occlude corneal vessels. At no time during observation was any laser more effective than the control. In the continuous wave Nd:YAG group, tissue necrosis was needed to achieve closure of vessels.

CONCLUSIONS: Yellow and green laser light are equally effective in eliminating the corneal vessels. Continuous wave Nd:"XAG, as used here, appears to be a poor choice. [J Refract Surg. 1995;11:45-49.]

Abstract

ABSTRACT

BACKGROUND: Corneal neovascularization is generally undesirable because it can lead to corneal scarring, lipid deposits, and corneal graft failure. To eliminate these vessels, several techniques are available including laser photocoagulation.

METHODS: This prospective study was designed to compare the effectiveness of three laser wavelengths (continuous wave Nd:YAG, 1064 nm; argon green 514 nm; and yellow dye, 570 nm) to obliterate experimentally induced corneal neovascularization in the rabbit. Corneal vascularization was created in 12 rabbits by placing 7-0 silk sutures through two quadrants of the cornea. Once neovascularization was complete, the suture was removed and one of the three lasers was applied to occlude vessels at one of the neovascular sites. The other site was used as a control.

RESULTS: The yellow wavelength, when compared with the green required fewer exposures to occlude corneal vessels. At no time during observation was any laser more effective than the control. In the continuous wave Nd:YAG group, tissue necrosis was needed to achieve closure of vessels.

CONCLUSIONS: Yellow and green laser light are equally effective in eliminating the corneal vessels. Continuous wave Nd:"XAG, as used here, appears to be a poor choice. [J Refract Surg. 1995;11:45-49.]

Corneal neovascularization is undesirable in many pathologic situations. It can lead to corneal transplant rejection, corneal melting, or corneal opacification. A convenient and safe method of eliminating corneal neovascularization would be helpful.

This study was designed to compare the ability of three different laser's wavelengths to eliminate corneal neovascularization. Previous investigations have demonstrated equivocal results using argon laser to ablate corneal blood vessels.1-5 Knowing yellow light, available with the dye laser, is absorbed maximally by oxyhemoglobin, we postulated this modality would be efficacious in occluding blood vessels, and in providing a better chance of permanent neovascular regression. Continuous wave Nd.YAG laser, which provides a thermal effect, can photocoagulate blood vessels and, therefore, was also used in comparison.6

MATERIALS AND METHODS

An animal model was created using 14 pigmented rabbits. Surgical anesthesia consisted of IM Rompum (10 mg/kg) and IM Ketamine (50 mg/kg). Two 7-0 sutures were placed through one cornea of each rabbit 3 mm from the limbus at the 10 and 2 o'clock positions. Gentamicin ophthalmic ointment was applied to the operated eye twice a day for 1 week. Neovascularization up to the suture was complete by 3 weeks. All sutures were then removed. Two rabbits in which only a minimal amount of vascularization developed were excluded. The remaining rabbits had a broad 2- to 3-millimeter band of vessels. At this time, laser was applied to one neovascular site on each cornea, the other being used as a control. Argon green (514 nm) was applied to 4 eyes, yellow dye (570 nm) to 4 eyes (both with Coherent dye laser, model 920, Palo Alto, Calif), and continuous wave NdrYAG (1064 nm) (CVI Laser Corp, Albuquerque, NM) through a fiber-optic probe to 4 eyes (Tables 1 and 2). Laser energy was adjusted to produce a cessation of blood flow through the vessels. A 1- to 2-millimeter wide segment of each vessel was treated as it emerged from the limbus. Whenever possible, large trunks were obliterated individually, but in most eyes where there were many small vessels close together, we placed contiguous shots across the vascularized cornea. Observations made and photographs taken before and after treatment were used to document the effects of treatment (Fig 1). The rabbits were anesthetized and corneas evaluated with an operating microscope at day 0 and at 1, 2, 4, and 6 weeks after laser surgery. To follow the changes in the corneal vessels, we adapted a grading system as used previously by Reed et al.2 This system graded the most severe and dense vascularization as 4 + and the absence of vessels as 0. Between these extremes were grades of 1 + , 2 + , and 3 + .

Table

Table 1Comparison of Lasers Used in Treatment of Corneal NeovascularizationTable 2Number of Laser Exposures Required to Close All Vessels in Each of 12 Rabbit EyesFigure 1 : (A) Typical appearance of corneal vessels induced by silk suture in rabbit eye prior to treatment (arrows). (B) Reduced neovascularization following treatment with yellow dye laser (arrows).

Table 1

Comparison of Lasers Used in Treatment of Corneal Neovascularization

Table 2

Number of Laser Exposures Required to Close All Vessels in Each of 12 Rabbit Eyes

Figure 1 : (A) Typical appearance of corneal vessels induced by silk suture in rabbit eye prior to treatment (arrows). (B) Reduced neovascularization following treatment with yellow dye laser (arrows).

RESULTS

All treated rabbits survived and were included in the results. The results comparing the three laser modalities are displayed in Figure 2. The raw data are displayed in Table 3. A comparison of the green versus the yellow laser group and both versus the control, revealed no significant difference at any time during 6 weeks observation. All rabbits in these groups were graded between 0 and 1+ vascularization. The control site in the Nd:YAG group actually regressed further than the treated site. Continuous wave Nd:YAG was clearly disadvantageous when compared with the other two laser groups. Statistical comparison was made using Kruskal one-way analysis of variance.

Figure 2: Corneal neovascularization fn each group over time. Comparison of healing time between control in each group showed no significant difference (p-value = .31). Healing with yellow versus green wavelength revealed no significant difference (p-value = .60). Healing between the YAG and the other treatment groups revealed a significant difference (p-value = .0006) (p-value is obtained using ch/-square distribution with two degrees of freedom).Table 3Intensity of Corneal Neovascularization (0+ to 4 + ) for Each Rabbit Eye*

Figure 2: Corneal neovascularization fn each group over time. Comparison of healing time between control in each group showed no significant difference (p-value = .31). Healing with yellow versus green wavelength revealed no significant difference (p-value = .60). Healing between the YAG and the other treatment groups revealed a significant difference (p-value = .0006) (p-value is obtained using ch/-square distribution with two degrees of freedom).

Table 3

Intensity of Corneal Neovascularization (0+ to 4 + ) for Each Rabbit Eye*

We evaluated the ease and faculty of the treatment sessions and found the yellow wavelength was more efficient at producing closure of the corneal vessels. Multiple exposures of green laser were required to coagulate the vessel walls, whereas yellow required only one or two exposures to produce the same effect. An average of 132.5 exposures per eye were required to reach our endpoint with argon green, and 108 exposures per eye were required using yellow dye. In the continuous wave Nd:"YAG group, we were able to obliterate the corneal vessels but not without causing a fair amount of tissue necrosis and corneal shrinkage (Fig 3).

Figure 3: Corneal scarring and vascularization seen after exposure to continuous wave Nd:YAG (arrow).

Figure 3: Corneal scarring and vascularization seen after exposure to continuous wave Nd:YAG (arrow).

The only ocular side effect noted in the first two groups was an occasional peripheral iris burn noted directly under the treatment site. No corneal scars were noted and the eyes remained quiet. No thinning of the cornea was noted in the area of treatment. In contrast, two of the Nd:YAG treated eyes developed a significant inflammatory response with an increase in corneal vessels at the margins of treatment along with dense corneal scarring and some thinning.

DISCUSSION

Corneal neovascularization is seen in many pathologic conditions including inflammatory disorders, chemical injuries, infections, and keratitis sicca. This neovascularization is the cause of significant ocular morbidity, and is a problem which has not responded well to medical management. Laser therapy has been an option to be evaluated. Previous work has focused on the use of argon blue-green laser to photocoagulate corneal blood vessels.15 This proved to be tedious and in the long run often ineffective in inducing complete remission of vessels.

Photothrombosis of corneal vessels, employing intravenous rose bengal and argon irradiation, has been shown to be effective in animal models.7 8 This technique induces thrombosis formation via a photochemical interaction between a photosensitizing dye and laser light irradiation. This interaction leads to endothelial damage and secondary platelet aggregation. Although this can be accomplished with lower intensities of laser light, compared with those required for photocoagulation, it does require the use of an intravenous agent with potential systemic side effects.

Qualitatively, we found yellow laser light was more efficient than argon green at photocoagulating experimentally induced corneal vessels. This stands to reason, if one looks at the absorption spectrum for oxyhemoglobin (Fig 4).9 The absorption coefficient of yellow light is approximately half a log unit higher than green light. This was the only advantage over argon green, since the long-term results of both were not significantly different from control at any time during 6 weeks of observation.

Our inability to demonstrate a difference between treatment and control, unlike previous studies, may be related to the difficulty of creating an appropriate model. Rabbits are known to be excellent wound healers; therefore, in future investigations, one might consider leaving the neovascular stimulus in place during the treatment and observation. In addition, our system of grading was not sensitive enough to uncover small differences between the groups.

As in humans, we observed different patterns of vessel growth from the limbus. In the case of vessels coming off one or two larger trunks, the approach to obliteration should be different from the case when there are multiple vessels without a common trunk. In the latter situation, as we found here, it is easier to use the laser to paint a band across the corneal vessels rather than hit each individually. More extensive treatment along the entire length of larger feeder vessels may be more appropriate than stopping blood now in a small !-millimeter segment as we did here.

Figure 4: Absorption coefficient of yellow light for hemoglobin is higher than for other wavelengths.9

Figure 4: Absorption coefficient of yellow light for hemoglobin is higher than for other wavelengths.9

The continuous wave Nd:YAG's effectiveness was limited by the large spot size (600 µp?) created by the probe required to carry the laser light. This compares with a standard spot size of 50 µp? used for capsulotomies or iridectomies. There was no way of avoiding damage to intervening corneal stroma. Manipulation of the spot size to one that is closer to the caliber of the treated blood vessels, about 50 to 100 µ??, must be undertaken if this modality is to be considered further.

Baer and Foster presented encouraging results using 577-nanometer yellow light for corneal vessel photocoagulation in humans.10 They ablated corneal vessels in a small group of eyes with either graft rejection, high-risk vascularization before corneal transplantation, neovascularízation with lipid keratopathy, or extensive neovascularization in poorly seeing eyes. The results were encouraging, especially in the first two groups where resolution of graft rejection or successful keratoplasty was noted in all 10 eyes.

We think either yellow or green laser light can be an effective means of eliminating corneal neovascularization in humans, even though in our rabbit model neither was better than the control. Presumably the rabbit model is a poor one. The development of a more sensitive animal model that simulates human conditions is needed.

REFERENCES

1. Cherry PMH, Garner A. Corneal neovascularization treated with argon laser. Br J Ophthalmol. 1976;60:464-472.

2. Reed JW, Fremer C, Klintworth GK Induced corneal neovascularization remission with argon laser therapy. ArcA Ophthalmol. 1975;93:1017-1019.

3. Ey RC, Hughes WF, Bloome MA, Tallman CB. Prevention of corneal vascularization. Am J Ophthalmol. 1968 ;66: 1118-1131.

4. Cherry PMH, Faulkner JD, Shaver RP, Wise JB, Witter SL. Argon laser treatment of corneal neovascularization. Ann Ophthalmol. 1973;5:911-920.

5. Nirankari VS, Baer JC. Corneal argon laser photocoagulation for neovascularization in penetrating keratoplasty. Ophthalmology. 1986;93:1304-1309.

6. Fankhauser F, Van Der Zypen E, Kwasniewska S, Loertscher H. The effect of thermal mode Nd:Yag laser radiation On vessels and ocular tissues. Ophthalmology. 1985;92:416-419.

7. Huang AJW, Watson BD, Hernandez E, Tseng SCG. Ph0tothrombosis of corneal neovascularization by intravenous rose bengal and argon laser irradiation. Arch Ophthalmol. 1988;106:680-685.

8. Carrent G, Rassel TJ, Tseng SCG, Watson BD. Promotion of graft survival by photothrombotic occlusion of corneal neovascularization. Arch Ophthalmol. 1989;107:1501-1506.

9. L'Esperance FA, Jr. Ophthalmic Lasers, 3rd ed. St Louis, Mo: CVMosby Company; 1989;1:283.

10. Baer JC, Foster CS. Corneal laser photocoagulation for treatment of neovascularization. Efficacy of 577 nm yellow dye laser. Ophthalmology. 1992;99:173-179.

Table 1

Comparison of Lasers Used in Treatment of Corneal Neovascularization

Table 2

Number of Laser Exposures Required to Close All Vessels in Each of 12 Rabbit Eyes

Figure 1 : (A) Typical appearance of corneal vessels induced by silk suture in rabbit eye prior to treatment (arrows). (B) Reduced neovascularization following treatment with yellow dye laser (arrows).

10.3928/1081-597X-19950101-12

Sign up to receive

Journal E-contents