46

Hemolytic or Ghost-Cell Glaucoma

David L. Epstein, MD, MMM

CLINICAL FINDINGS

The term hemolytic glaucoma was introduced by Fenton and Zimmerman¹ to describe the clinical and histologic findings in a patient who had spontaneous hemorrhage into the vitreous humor and several weeks later had high intraocular pressure (IOP) and pain, leading to enucleation. Hemorrhagic debris and pigment-laden macrophages were found in sections of the anterior chamber angle. Fenton and Zimmer man¹ postulated that the glaucoma was caused by a mechanical blockage of aqueous outflow by broken-down red blood cells and macrophages, analogous to phacolytic glaucoma, which was conceived of as being due to obstruction by lens material and macrophages. Clinical and experimental investigations by Campbell and colleagues^2–5 have led us to believe that ghost cells, resulting from de generation of red blood cells, are a particularly important factor in obstructing aqueous outflow and causing glaucoma as an occasional consequence of hemorrhage in the vitreous humor, and also in some cases of glaucoma associated with hemorrhage into the anterior chamber from blunt injury to the globe.

Retinal vascular disease or injury often produces hemorrhage into the vitreous humor, but the blood seldom passes the barrier of the anterior hyaloid membrane to reach the anterior chamber. However, a few such cases have been reported in phakic eyes.^6,7 Most commonly, the anterior hyaloid is not intact, often from accidental injury or surgery, particularly after cataract extraction or vitrectomy, and thereby the products of hemorrhage can enter the anterior chamber and produce severe open-angle glaucoma. Fresh hemorrhage into the vitreous humor appears red when viewed with the slit-lamp biomicroscope, but within 2 to 4 weeks, the color of the material in the vitreous changes to a light tan or khaki color due to loss of the red-colored hemoglobin and degeneration of the red cells to ghosts. The red blood cells change from biconcave discs to spherical forms with Heinz bodies. Ghost-cell glaucoma tends to develop within 3 to 4 weeks after the occurrence of a hemorrhage into the vitreous humor in an eye with a defective anterior hyaloid membrane. Myriad small cells appear in the aqueous humor, and the IOP rises into the range of 30 to 70 mm Hg, with the angle remaining open.

In many instances, the appearance of myriad cells in the aqueous humor associated with a severe open-angle glaucoma has been misinterpreted as an iritis or uveitis, in the belief that the cells were inflammatory cells. In such cases, topical or systemic treatment with corticosteroids has often been used, generally to no avail. In fact, corticosteroids may delay the reabsorption of the blood cells from the eye. (It is interesting to contemplate how corticosteroids might interfere with this blood reabsorption within the trabecular meshwork [TM] by interfering with nor mal trabecular cell actions. Similar cellular actions may be involved in steroid-induced open-angle glaucoma or even primary open-angle glaucoma.)

Ghost-cell glaucoma with such tan-colored cells in the anterior chamber may be misinterpreted as post surgical endophthalmitis. In these cases, there are no inflammatory keratic precipitates. If, with the slit-lamp biomicroscope the color of the light tan, old blood in the vitreous is compared with that of the fine cells circulating in the anterior chamber, they have the same hue, characteristic of ghost cells.

Aspiration of material from the anterior chamber in a series of cases of this sort and immediate examination of the fluid by phase-contrast microscopy without drying, staining, or filtering have shown great numbers of ghost cells with characteristic Heinz bodies. These cells have occasionally been called erythroclasts. Their identity has been confirmed by transmission and scanning electron microscopy both in the aqueous and vitreous humors. Variable amounts of amorphous debris accompany the ghost cells, but in at least 15 samples of aqueous humor from eyes with hemolytic glaucoma examined by Camp bell and colleagues,2 macrophages were remarkably scarce and there was no cellular evidence of inflammatory reaction.

By gonioscopy, the angle may appear normal, but if there are many ghost cells in the anterior chamber, they may form a khaki-colored layer on the filtration portion of the TM, occasionally resembling a layer of butter. In more extreme cases, the ghost cells may fill the dependent angle. The tan or yellowish color distinguishes this accumulation from an exudate or hypopyon of inflammatory cells.

The main causes of ghost-cell glaucoma in our experience have been the following:

• Cataract extraction complicated by hemorrhage into the anterior chamber or vitreous cavity
  
• Blunt or penetrating trauma with hemorrhage into the anterior chamber or vitreous cavity
  
• Closed vitrectomy for removal of vitreous hemorrhage for the purpose of improving vision

Ghost-cell glaucoma can also result from extensive hyphema formation alone, but it is most commonly associated with vitreous hemorrhage that acts as a reservoir for these cells to come into the anterior chamber after a subsequent interval of time.

DIAGNOSIS

The diagnosis of hemolytic or ghost-cell glaucoma should be suspected whenever hemorrhage into the vitreous cavity has occurred and has been followed within 3 or 4 weeks by elevated IOP in the range of 30 to 70 mm Hg, associated myriad very small cells circulating in the anterior chamber, with no conglomerate keratic precipitates on the corneal endothelium. There may be a tan hyphema, sometimes mixed with red blood (Figure 46-1). When there has been trauma, gonioscopy may also show contusion-disruption of angle structures, but this should not distract attention from the importance of the fine cells in the aqueous humor. The diagnosis can be most securely established by aspirating the aqueous humor from the anterior chamber and examining it immediately by phase-contrast microscopy without centrifuging, filtering, drying, or staining. Characteristically, no inflammatory cells are seen, but a rare macrophage may be found with innumerable erythrocyte ghost cells.

MECHANISM

Experiments by Campbell and colleagues^2,3,5 have provided evidence that permits one to explain the development of high IOP in this condition. Normal red blood cells are so pliable that they pass readily through spaces smaller than the ordinary diameter of these cells (that are presumably present in the outflow pathway), but when they de generate to ghost cells, they lose this pliability and are unable to pass through the same small spaces. This has been well-demonstrated in vitro with standard microporous filters. Campbell et al have confirmed that fresh human red blood cells pass readily from the anterior chamber through the outflow system to the aqueous veins in enucleated normal human eyes, producing only a modest decrease in facility of aqueous outflow, and they have demonstrated that human erythrocyte ghosts differ from the fresh cells in being rigid and unable to pass from the anterior chamber through the aqueous outflow system.² Instead, they obstruct and cause a marked reduction in facility of aqueous outflow, leading to the elevation of IOP.

TREATMENT

Treatment may be medical or surgical, depending on the seriousness of the situation. If IOP is not high, it may respond to medical therapy (aqueous suppressant therapy). Although miotics have been used with occasional success in the past by presumably mechanically widening the outflow channels to allow egress, this also may be counterproductive as in any eye with inflammation. If the IOP can be kept down to the 30- to 40-mm Hg range with this treatment, the cornea is not edematous, and the patient is not in pain, the ghost cells may gradually clear in the course of some weeks (but remember the above potential contrary effect of corticosteroids) and the IOP may fall. However, if the amount of old hemorrhagic material in the vitreous is large, it may take too long to clear by slow diffusion forward into the anterior chamber and out the drainage system. When the IOP is in the 60- to 70-mm Hg range, it is rare for medical treatment to reduce the IOP sufficiently and surgical treatment is required. Systemic hyperosmotic agents usually help only temporarily. Surgical treatment is indicated also in those cases in which IOP in the 40- to 50-mm Hg range persists for weeks despite medical treatment. In eyes prone to retinal or optic nerve vascular occlusion, such as those with sickle cell anemia or trait, there is a much lower IOP threshold for surgical intervention (24 to 30 mm Hg).

This inference is drawn from a study8 in which patients who had old hemorrhage in the vitreous cavity, but no glaucoma, were subjected to closed vitrectomy to improve vision. In some of these cases, glaucoma developed within 2 to 10 days after the procedure. This was associated with a flooding of the anterior chamber with ghost cells, presumably entering through some breach in the anterior hyaloid membrane. This complication occurred less frequently when thorough irrigation was performed to remove as much old blood as possible from the vitreous cavity. In cases of ghost-cell glaucoma precipitated by closed vitrectomies with less thorough irrigation, the IOP characteristically was greater than 40 mm Hg, sometimes higher than 60 mm Hg. Some of these cases were controlled medically and others required repeated irrigation of the anterior chamber. When only a small amount of the old blood remained in the vitreous cavity, there was less tendency for recurrence of glaucoma than when a larger reservoir of ghost cells was left in the vitreous cavity.

After vitrectomy, an increase of IOP can be produced in other ways. One must also consider the possibilities of corticosteroid glaucoma, lens-induced glaucoma, neovascular glaucoma, and obstruction of the outflow system by sickle cells in a patient with this disease.⁹ However, persistent ghost-cell glaucoma has been the most commonly observed cause of an increase of IOP after vitrectomy.

If, despite these measures, there is reaccumulation of many cells in the anterior chamber and high IOP, other procedures must be considered. Cyclocryotherapy or newer ciliary destructive techniques have been effective in reducing the IOP in some cases, presumably by reducing the rate of aqueous formation.

The following case illustrates a typical course of ghost-cell glaucoma from contusion of the eye.

Case 46-1

A middle-aged man was struck in one eye by a block of wood, causing a small conjunctival laceration, moderate hyphema, and IOP below normal. Surgical exploration of the outer surface of the globe showed no rupture. In a few days, the hyphema spontaneously cleared from the anterior chamber. Gonioscopy showed contusion-disruption of angle structures superiorly, and the anterior chamber was abnormally deep superiorly, with a slight posterior subluxation of the upper portion of the lens. The pupil was not appreciably distorted and the lens was clear, but there was no fundus reflex due to extensive hemorrhage into the vitreous cavity. In about 2 weeks, the IOP became normal. In 3 weeks, the IOP was in the 60s with diffuse corneal edema, pain, and myriad fine cells in the anterior chamber, but no keratic precipitates. Although the eye did not appear inflamed, the cells were misinterpreted as a sign of “iritis,” and anti-inflammatory treatment was started with atropine and frequent application of corticosteroid. Oral glycerin produced only transient reduction of IOP. After 2 or 3 days with fundamentally no change, the patient was seen by one of us. A presumptive diagnosis of ghost-cell glaucoma was made, and diagnostic aspiration of aqueous humor and therapeutic washing out of the anterior chamber were done. Phase-contrast microscopy of the aqueous humor immediately after removal from the anterior chamber showed innumerable ghost cells but no inflammatory cells or macrophages. For a day or two, IOP was normal, but soon more ghost cells leaked from the reservoir in the vitreous cavity to the anterior chamber, presumably through a defect in the anterior hyaloid membrane produced when the lens was subluxated, and the IOP increased again.

In contemplating the pathogenesis of various types of glaucoma and the mechanism by which they respond to various treatments and in thinking about how to evolve better treatments, it is interesting to consider that, both in ghost-cell glaucoma and in lens-induced glaucoma, there is a good clinical response to washing out the anterior chamber, provided that there is no reservoir of ghost cells or lens material to replenish what was washed out.10 Usually in the following days, the IOP is normal, suggesting that the aqueous outflow system has become cleared of the obstructing material. Yet, when the conditions are reproduced experimentally in an excised human eye by introducing ghost cells or lens proteins¹¹ into the anterior chamber, clearly obstructing outflow of aqueous fluid from the anterior chamber, we find no immediate relief of the obstruction when we wash out the anterior chamber. Therefore, in vivo, there must be some vital aid to clearing the outflow channels after the bulk of the material has been removed from the anterior chamber. Corticosteroids may interfere with these trabecular cellular processes. We wonder if macrophages are helping to clear the outflow system and whether there could be some advantage in stimulating their activity. Other therapeutic approaches might conceivably include use of enzymes to break down ghost cells or lens proteins to fragments that might more easily pass through the aqueous outflow system and obstruct it less.

HEMOSIDEROTIC GLAUCOMA

The discussion in this chapter has been limited to the acute problems posed by erythrocyte ghost cells; we have not considered the late consequence of leaving blood degenerating in the eye for a long time that may lead to a different condition called hemosiderotic glaucoma, which is usually diagnosed histologically after enucleation of a blind eye. That condition is characterized by degeneration of the TM with positive staining for iron.^12,13

REFERENCES

1.      Fenton RH, Zimmerman LE. Hemolytic glaucoma. Arch Ophthalmol. 1963;70:236-239.

2.      Campbell DG, Simmons RJ, Grant WM. Ghost cells as a cause of glaucoma. Am J Ophthalmol. 1976;81:441-450.

3.      Campbell DG, Essigmann EM. Hemolytic ghost cell glaucoma. Further studies. Arch Ophthalmol. 1979;97:2141-2146.

4.      Campbell DG. Ghost cell glaucoma following trauma. Ophthalmology. 1981;88:1151-1158.

5.      Lambrou FH Jr, Aiken DG, Woods WD, Campbell DG. The production and mechanism of ghost cell glaucoma in the cat and primate. Invest Ophthalmol Vis Sci. 1985;26:893-897.

6.      Mansour AM, Chess J, Starita R. Nontraumatic ghost cell glaucoma—a case report. Ophthalmic Surg. 1986;17:34-36.

7.      Frazer DG, Kidd MN, Johnston PB. Ghost cell glaucoma in phakic eyes. Int Ophthalmol. 1987;11:51-54.

8.      Campbell DG, Simmons RJ, Tolentino FI, et al. Glaucoma occurring after closed vitrectomy. Am J Ophthalmol. 1977;83:63-69.

9.      Wilensky JT, Goldberg MF, Alward P. Glaucoma after pars plana vitrectomy. Trans Am Acad Ophthalmol Otolaryngol. 1977;83:114-121.

10.    Phelps CD, Watzke RC. Hemolytic glaucoma. Am J Ophthalmol. 1975;80:690-695.

11.    Epstein DL, Jedziniak JA, Grant WM. Obstruction of aqueous outflow by lens particles and by heavy molecular weight soluble lens protein. Invest Ophthalmol Vis Sci. 1978;17:272-277.

12.    Benson WE, Spaiter HF. Vitreous hemorrhage. Surv Ophthalmol. 1975;15:297-311.

13.    Wollensak J. Phakolytisches und Hamolytisches Glaukom. Klin Monatsbl Augenheilkd. 1976;168:447-452.