Corneal cross-linking (CXL) is a procedure developed to halt the progression of corneal ectatic diseases such as keratoconus, pellucid marginal degeneration, and corneal ectasia induced by refractive surgery.1 Clinical results have shown that CXL is generally effective2 and has resulted in the reduction of corneal transplantation over the long term.3 One impediment to the more widespread use of CXL is the duration of the procedure—the standard Dresden protocol takes more than 60 minutes.4 Higher ultraviolet (UV) irradiance has been used to decrease the exposure time. However, this led to decreased corneal stiffening.5 This is to be expected because oxygen diffusion limits the rate of the CXL reaction, which is oxygen dependent. The study by Richoz et al6 showed that in an anoxic environment, the UV-riboflavin reaction does not produce corneal stiffening. The photochemical reaction rapidly consumes oxygen in the corneal stroma. Kamaev et al7 showed that oxygen concentration at the 100-µm depth becomes effectively zero after the initial 10 to 15 seconds of UV exposure. For the reaction to continue, additional oxygen molecules must diffuse into the cornea from air. To increase oxygen diffusion depth, pulsed UV delivery has been used.8 However, this does not increase the average rate of oxygen diffusion into the corneal stroma, which limits the rate of corneal stiffening. To accelerate oxygen transport into the cornea, it is necessary to increase the atmospheric oxygen concentration at the corneal outer surface.
Two devices have been described for oxygen enrichment during CXL. Friedman and Adler9 invented a mask device worn over both eyes to deliver oxygen gas during clinical CXL. Another invention by Lopath10 described oxygen enrichment over the cornea through a perfluorocarbon fluid reservoir under a scleral contact lens. Both of these devices include disposable components in contact with the patient.
The current study describes a simpler apparatus to increase atmospheric oxygen concentration over the corneal surface. Our device hovers over the operative eye but does not come into contact with the patient. The level of oxygen enrichment achievable by our noncontact device is investigated in an experimental mock-up.
Materials and Methods
The oxygen delivery device comprises a clear poly-carbonate tube with various attachments (Figure 1A). The bottom of the tube opens toward the cornea. The length of the apparatus tube is 14 cm and the inner diameter is 1.58 cm. The top of the tube is air-sealed. Compressed oxygen gas (99% purity) is delivered to the tube through a side opening connected by flexible tubing to a flow regulator and oxygen tank. In an actual CXL set-up, the top of the tube would be connected to a UV delivery device. During the CXL procedure, both UV light and oxygen would pass through the tube lumen (Figure 1B). The oxygen gas would be humidified to prevent corneal drying and filtered to remove microbes.
(A) Photograph of the experimental apparatus to measure atmospheric oxygen concentration in a simulated corneal cross-linking (CXL) set-up. Oxygen gas is delivered through a transparent plastic tube to a scleral contact lens that simulates the cornea. The face is simulated by a mannequin. (B) Schematics of the proposed CXL device with coaxial ultraviolet light and oxygen delivery through the transparent tube.
In our experimental mock-up, a rigid mini scleral contact lens of 15 mm diameter and 5.4 mm base curve was used to simulate the cornea. The scleral contact lens was placed on a mannequin face to simulate human facial contours (Figure 1A). The oxygen concentration over the surface of the scleral contact lens was measured by a gas oximeter (Unitec RM 8000) with a suction rate set at 0.1 L/min. The oximeter was attached to a sampling tube of 1.5 mm inner diameter. The opening of the sampling tube was placed at 3 mm above the apex of the scleral lens. During each experimental run, the oximeter reading was recorded manually every 30 seconds for 4.5 minutes after the regulator was opened to the preset flow rate. Flow rates of 0.25, 0.50, and 1 L/min were tested. Another experimental parameter was the cornea-tube distance measured between the lower end of the polycarbonate tube and the top of the contact lens. The three cornea-tube distances tested were 8, 10, and 14 mm.
The measured baseline oxygen concentration was 20.9%, which was consistent with the known oxygen concentration in ambient air. The time course of oxygen concentration, averaged over three experimental trials conducted with the cornea-tube distances of 8, 10, and 14 mm, is shown in Figure 2. The oxygen concentration reached plateau levels after 2 to 3.5 minutes, depending on the flow rate. The time to plateau was shorter for higher oxygen flow rate. Therefore, in subsequent analysis the oxygen measurements were averaged over the plateau phase between 3.5 and 4.5 minutes (Figure 3).
Experimental oxygen concentration time profile recorded using three oxygen flow rates at cornea-tube distances of (A) 8, (B) 10, and (C) 14 mm. Error bars represent the standard deviation of the three trials.
Steady state oxygen concentration as a function of oxygen flow rates and distances between cornea and tube. Error bars represent the standard deviation of the three trials.
After cutting off oxygen flow at 4 minutes and 30 seconds, it took 1 to 2 minutes for oxygen concentration to return to baseline. Oxygen enrichment was found to be strongly dependent on the oxygen flow rate up to 1 L/min, at which point 99% concentration could be achieved (Figure 3). At 1 L/min flow rate, 99% concentration was achieved at the cornea-tube distance of 8 to 10 mm, but decreased to 90% at 14 mm. The results indicate that approximately 1 L/min and 8 to 14 mm cornea-tube distance constitutes an acceptable operating range for our device.
Our results show that enrichment of atmospheric oxygen to greater than 90% concentration is feasible using our noncontact gas delivery device. Ninety percent oxygen is 4.3 times the ambient air level. This level of enrichment could potentially allow for greatly accelerated CXL procedures using high UV irradiance without depleting corneal stromal oxygen.
During CXL, oxygen is consumed in the corneal stroma by a photochemical reaction and must be replenished by transport from the air–corneal interface. The oxygen concentration at 100-µm depth level quickly drops to near zero after 10 to 15 seconds of UV exposure using the standard CXL regimen.7 Thus oxygen transport is a rate-limiting process. The rate of oxygen transport into the cornea is driven by the gradient between the atmospheric and stromal oxygen concentrations. Because the concentration is near zero in mid-stroma at a steady state, the rate of transport should be approximately proportional to the atmospheric concentration. Thus ambient oxygen enrichment should allow accelerated CXL without sacrificing CXL depth and strength.
Experimental results have demonstrated approximate proportionality between ambient air oxygen concentration and stromal oxygen concentration. In an ex vivo experiment in an atmosphere controlled chamber, Hill et al11 showed that enriching the environmental oxygen concentration from 20% to 90% increases the stromal oxygen concentration by a factor of approximately 5 both prior to and during accelerated CXL with UV irradiation at 30 mW/cm2. The oxygen enriched environment maintained the stromal oxygen concentration at 13% at a depth of 230 µm.
The approximate proportionality of the ambient connection between atmospheric and stromal oxygen is also expected on theoretical grounds. The diffusion-reaction equation for oxygen is given by Equation 1. The equations states that the steady-state oxygen input by diffusion (left side) is equal to the oxygen consumption (right side)12:
is the partial oxygen pressure within the cornea, D
is the diffusion coefficient of the cornea, α
is the oxygen solubility constant, and k P
is the oxygen consumption rate, assuming that oxygen consumption is proportional to oxygen concentration (first order kinetics). The solution to this diffusion-reaction equation (Equation 2) for first order kinetics and in one spatial dimension z
is known in the literature.12
The one-dimensional approximation is permissible because the treatment zone diameter (approximately 10 mm) is much larger than the oxygen diffusion depth (small fraction of a millimeter).
is depth below the anterior corneal surface and P0
is the partial oxygen pressure at the corneal surface. Equation 2 shows that as we increase the atmospheric oxygen at the corneal surface, the oxygen concentration at stromal depths also increases proportionally.
Fick's first law gives the oxygen influx (Equation 3) at the corneal surface (z = 0) by differentiation of Equation 2. This shows that oxygen consumption, a proxy for the rate of CXL reaction, is also proportional to the oxygen concentration P0 at the surface:
We acknowledge that the above theoretical model is approximate and relies on some simplifying assumptions. The most crucial assumption is that of first-order kinetics, which occurs under the condition of relatively high oxygen consumption and low oxygen availability. At the other limit of low oxygen consumption and high oxygen availability, zero-order kinetics would prevail. But with zero-order kinetics, deep stromal oxygen concentration could be even more sensitive to ambient oxygen concentration.12
Thus, both experimental data and theoretical considerations suggest that the key to prevent stromal oxygen depletion and allow effective accelerated CXL is ambient oxygen enrichment. Our device appears to be a relatively simple and clinically feasible approach to achieve this. An oxygen flow rate of 1 L/min is commonly used for inhaled oxygen supplementation and easily achievable. The cornea-tube distances of 8 to 14 mm allows sufficient space for cannula delivery of riboflavin solution to the corneal surface and could be maintained by manual control. Further experimental work would be needed with direct measurement to confirm that stromal oxygen concentration would increase in proportion to atmospheric oxygen enrichment. Such experiments would need to be done using an oxygen-sensing microelectrode inserted at various stromal depths under various CXL conditions: riboflavin concentration, UV irradiance, and ambient oxygen concentration. Clinical trials are needed to verify that ambient oxygen enrichment could improve the efficacy of CXL, especially under epithelium-on and/or accelerated protocols. We hope our description of this set-up and our primary results will help other investigators implement oxygen-enriched experiments and clinical trials.
- Hersh PS, Greenstein SA, Fry KL. Corneal collagen cross-linking for keratoconus and corneal ectasia: one-year results. J Cataract Refract Surg. 2011;37(1):149–160. doi:10.1016/j.jcrs.2010.07.030 [CrossRef]
- Caporossi A, Mazzotta C, Baiocchi S, Caporossi T. Long-term results of riboflavin ultraviolet a corneal collagen cross-linking for keratoconus in Italy: the Siena Eye Cross Study. Am J Ophthalmol. 2010;149(4):585–593. doi:10.1016/j.ajo.2009.10.021 [CrossRef]
- Raiskup F, Theuring A, Pillunat LE, Spoerl E. Corneal collagen crosslinking with riboflavin and ultraviolet-A light in progressive keratoconus: ten-year results. J Cataract Refract Surg. 2015;41(1):41–46. doi:10.1016/j.jcrs.2014.09.033 [CrossRef]
- Wollensak G, Spoerl E, Seiler T. Riboflavin/ultraviolet-a-induced collagen crosslinking for the treatment of keratoconus. Am J Ophthalmol. 2003;135(5):620–627. doi:10.1016/S0002-9394(02)02220-1 [CrossRef]
- Kling S, Hafezi F. An algorithm to predict the biomechanical stiffening effect in corneal cross-linking. J Refract Surg. 2017;33(2):128–136. doi:10.3928/1081597X-20161206-01 [CrossRef]
- Richoz O, Hammer A, Tabibian D, Gatzioufas Z, Hafezi F. The biomechanical effect of corneal collagen cross-linking (CXL) with riboflavin and UV-A is oxygen dependent. Transl Vis Sci Technol. 2013;2(7):6–6. doi:10.1167/tvst.2.7.6 [CrossRef]
- Kamaev P, Friedman MD, Sherr E, Muller D. Photochemical kinetics of corneal cross-linking with riboflavin. Invest Ophthalmol Vis Sci. 2012;53(4):2360–2367. doi:10.1167/iovs.11-9385 [CrossRef]
- Zhu Y, Reinach PS, Zhu H, et al. Continuous-light versus pulsed-light accelerated corneal crosslinking with ultraviolet-A and riboflavin. J Cataract Refract Surg. 2018;44(3):382–389. doi:10.1016/j.jcrs.2017.12.028 [CrossRef]
- Friedman MD, Adler D, inventors;Avedro Inc, assignee. Systems and methods for treating an eye with a mask device. US patent US20170156926A1. 2017.
- Lopath PD, inventor;TECLens, LLC, assignee. Corneal cross-linking with oxygenation. US patent US20160175147A1. 2016.
- Hill J, Liu C, Deardroff P, et al. Optimization of oxygen dynamics, UV-A delivery, and drug formulation for accelerated epi-on corneal crosslinking. Curr Eye Res. 2020;45:450–458. doi:10.1080/02713683.2019.1669663. [CrossRef]
- Popel AS. Theory of oxygen transport to tissue. Crit Rev Biomed Eng. 1989;17(3):257–321.