Vitreous Research: The Oxygen Story

Know the role of the vitreous humor to better understand certain ocular pathologies.


Vitreous Research: The Oxygen Story

Know the role of the vitreous humor to better understand certain ocular pathologies.

Nancy M. Holekamp, MD

Nancy M. Holekamp, MD, is director of retina services at the Pepose Vision Institute and professor of clinical ophthalmology at Washington University School of Medicine in Saint Louis. She reports no financial interests in any products mentioned in this article. Dr. Holekamp can be reached via e-mail at

Many retina specialists regard the vitreous body as simply a sticky gel that needs to be removed to fix serious retinal pathology. In other words, the vitreous gel is dispensable. What if, perhaps, we were to ascribe to the vitreous gel a very important physiologic property that was essential to the health of the eye? Would our regard for the vitreous gel change?

This article will first present data showing that the vitreous gel does, indeed, perform an important physiological task that is essential to the eye: it consumes oxygen and protects the eye from oxidative damage. The article will then discuss the implications of the vitreous gel having this property only in the gel state and not when liquefied or removed surgically.

Specifically, age-related cataract and some cases of primary open-angle glaucoma may be attributed to oxidative damage caused by vitreous liquefaction or vitrectomy surgery. Finally, the potential benefits to the health of the eye of preventing vitreous liquefaction will be proposed.


Traditionally the functions of the postnatal vitreous gel have been thought to be biophysical in two ways: (1) to serve as a support structure for the retina, and (2) to serve as a medium for light transmission. Then, in the April 2009 issue of Archives of Ophthalmology, my colleagues at the Washington University School of Medicine and I reported a new and provocative finding: gel vitreous consumes oxygen.1

We performed a series of experiments in which the tip of an oxygen sensor (the size of a 30-gauge needle) was easily fit through trocars at the time of 23- or 25-gauge vitrectomy surgery. In more than 60 patients undergoing vitrectomy surgery for a variety of retinal conditions, we measured intraocular oxygen tension prior to turning on the infusion cannula in two locations: adjacent to the pars plana and in the mid-vitreous cavity.

The results were surprising. First, we noted that the inside of the eye exists under hypoxic conditions. The oxygen tension was less than 5% (room air is 21% oxygen). Second, we noted that there was an oxygen gradient in the vitreous cavity. The oxygen tension was higher near the pars plana because of the nearby vascularized tissues, and it was lower in the center of the vitreous cavity.2

If diffusion alone were the only principle at work, then the oxygen tension should have been the same throughout the vitreous cavity. Thus, we hypothesized that the vitreous must actively consume oxygen.

This hypothesis led to a series of experiments in which we put cadaver vitreous into an oxygen-impermeable glass tube and placed the oxygen sensor inside. The oxygen tension in the vitreous started out very high, at 120 mm Hg, and within two hours fell to zero. The oxygen tension in the control tube filled with saline remained high at 120 mm Hg.

If the cadaver vitreous was stirred overnight in a beaker exposed to room air and then tested the next day, no oxygen consumption occurred. If ascorbate was added to the stirred, air-exposed vitreous, the ability to consume oxygen was restored. If different concentrations of ascorbate were added, the vitreous consumed oxygen at different rates in a linear fashion (Figure 1). Boiling the vitreous did not alter its ability to consume oxygen. The addition of metal chelating agents did not alter the ability of the vitreous to consume oxygen.

Figure 1. Vitreous consumes oxygen in an ascorbate-dependent manner. The rate of oxygen consumption varies in a linear fashion with the concentration of ascorbate in the vitreous gel.

These findings led us to conclude that vitreous consumes oxygen in an acellular, ascorbate-dependent chemical reaction. The reaction we believe to be continuously taking place in the vitreous gel is seen in Figure 2. Note that both ascorbate and oxygen are consumed in the reaction and that the ultimate product is water.

Figure 2. This is the proposed biochemical reaction occurring in the vitreous gel by which oxygen and ascorbate are consumed, and water is produced.

Ocular physiologists have known for decades that the vitreous gel contains extremely high levels of ascorbate, or vitamin C. In fact, the vitreous-to-plasma ratio for ascorbate is 40:1 owing to active transport. The body is constantly pumping ascorbate into the vitreous gel. The function of such a high ascorbate level in the eye was previously speculative. Now, it is clear that vitreous ascorbate is the eye's antioxidant.

The early experiments were performed with cadaver vitreous. Would the findings hold true in vivo? This question prompted a series of experiments in which 3 mL of undiluted vitreous were removed from patients undergoing vitrectomy surgery for a variety of retinal conditions. While in the operating room, the undiluted vitreous was immediately placed in the same oxygen-impermeable glass tube with an oxygen sensor.

We made an interesting observation: Formed or “gel-like” vitreous consumed oxygen at a faster rate than liquefied vitreous fluid. The difference in the rate of oxygen consumption was even more striking when comparing vitreous gel to the fluid removed from a previously vitrectomized eye. This evoked the question: Was the “gel” nature of the vitreous important? Were we born with a vitreous “gel” for a reason?

The answer is yes. Subsequent assays of ascorbate concentrations in the vitreous gel or fluid indicated that a gel vitreous has a higher concentration of ascorbate than liquid vitreous, and the rate of oxygen consumption varies accordingly. A gel vitreous with a high concentration of ascorbate will consume oxygen more rapidly than a liquid vitreous with a low concentration of ascorbate (Figure 3).

Figure 3. In these graphs, the state of the vitreous gel is graded: <3 represents gel-like vitreous; >3 represents liquefied vitreous gel; and “2nd surgery” represents liquid vitreous after vitrectomy. The rate of oxygen consumption is slower as the vitreous gel liquefies and contains less ascorbate.

The take-home message is that vitreous liquefaction and vitrectomy surgery are associated with less intraocular oxygen consumption. In those two “states,” the eye is exposed to oxidative stress and subsequent damage.


Oxygen and Nuclear Sclerotic Cataract

The crystalline lens exists in a hypoxic state thought to be essential for lens clarity. It is well established that age-related or nuclear sclerotic cataract is caused by oxidative damage to the nuclear proteins in the lens (Figure 4). Although the final damage-causing step may involve free radical oxygen species, the source is molecular oxygen diffusing from the vascularized tissues of the eye.

Figure 4. It is well established that nuclear sclerotic cataract is due to oxidation of proteins in the nucleus of the crystalline lens.

For example, the retina and choroid are two of the most vascular and highly oxygenated tissues in the body. Approximately 20 mm away is the crystalline lens, which exists in less than 5% oxygen. Only the vitreous gel lies between these two vastly different oxygen microenvironments.

Evidence now suggests that the function of the vitreous gel is to consume the molecular oxygen diffusing from the retinal surface to protect the lens from the oxidative damage that leads to nuclear cataract. When the vitreous gel is surgically removed, the lens is newly exposed to very high levels of molecular oxygen and rapidly accelerated nuclear sclerosis ensues in anyone over the age of 50. Vitreoretinal surgeons know this to be true, as vitrectomy surgery in anyone older than 50 invariably leads to cataract extraction in more than 90% of patients within two years.

However, when the vitreous gel slowly liquefies with age, the lens is gradually exposed to slightly ever-increasing amounts of molecular oxygen, and slowly progressive nuclear sclerosis ensues. Thus, the underlying cause of nuclear sclerotic cataract is vitreous liquefaction.

Oxygen and Open Angle Glaucoma

Not only does the vitreous gel consume oxygen, but the crystalline lens does too. If an individual older than 50 has vitrectomy surgery in a phakic eye, that eye will likely undergo lens removal (ie, cataract surgery) within two years. Consequently, the eye will have lost the two most important structures for protecting itself from oxidative damage: the vitreous gel and the crystalline lens.

Stanley Chang, MD, in the 2006 AAO Jackson Memorial Lecture, presented evidence that roughly 15% of eyes undergoing both vitrectomy surgery and cataract surgery will develop a form of primary open-angle glaucoma if followed for a long enough period of time.4 My colleagues at Washington University and I explored the “Chang Hypothesis” with the oxygen sensor in patients undergoing cataract surgery, glaucoma surgery and vitrectomy surgery.

We measured oxygen tension in the anterior chamber angle and found it to be significantly higher in eyes that had undergone both vitrectomy surgery and cataract surgery, compared to eyes with no history of surgery, vitrectomy surgery alone, or cataract surgery alone.3 We hypothesized that the mechanism of glaucoma is outflow obstruction secondary to oxidative damage to the trabecular meshwork.

The Oxygen Hypothesis

Based on the research discussed above, the “oxygen hypothesis” has been formulated: Loss of the vitreous gel through age-related liquefaction or surgery allows increased levels of molecular oxygen from the retina and the ciliary body to reach the lens. The oxidative damage leads to nuclear sclerotic cataract. When the cataract is removed, the in creased levels of molecular oxygen reach the trabecular meshwork. The oxidative damage leads to open angle glaucoma in up to 15% of individuals.


If vitreous liquefaction represents an early step in the pathogenesis of nuclear cataract and some cases of open angle glaucoma, perhaps research efforts should focus on preventing vitreous liquefaction. Currently, there is far more interest in pharmacologic vitreolysis. Perhaps the field of retina, and all of ophthalmology, would benefit from doing exactly the opposite: preventing vitreous liquefaction and posterior vitreous detachment.

The possible benefits of being able to maintain a vitreous gel throughout life are many: no nuclear sclerotic cataracts, reduced cases of open angle glaucoma, and reduced cases of retinal tears, retinal detachments, macular holes, vitreomacular traction, macular pucker, bleeding from proliferative diabetic retinopathy, etc. The list goes on.

Unfortunately, there is currently very little understanding of the mechanism of vitreous liquefaction and very few research efforts in this area. The importance of a gel vitreous to the health of the eye has long been unexamined. Vitreous research is the unexplored frontier of the eye. I am sure the vitreous gel will continue to surprise us. RP


1. Shui YB, Holekamp NM, Kramer BC, et al. The gel state of the vitreous and ascorbate-dependent oxygen consumption: relationship to the etiology of nuclear cataract. Arch Ophthalmol. 2009; 127:475-482.
2. Holekamp NM, Shui YB, Beebe DC. Vitrectomy surgery increases oxygen exposure to the lens: a possible mechanism for nuclear cataract formation. Am J Ophthalmol. 2005; 139:302-310.
3. Siegfried CJ, Shui YB, Holekamp NM, Bai F, Beebe DC. Oxygen distribution in the human eye: relevance to the etiology of open-angle glaucoma after vitrectomy. Invest Ophthalmol Vis Sci. 2010; 51:5731-5738.
4. Chang S. LXII Edward Jackson lecture: open angle glaucoma after vitrectomy. Am J Ophthalmol. 2006; 141:1033-1043.