A UNIFICATION HYPOTHESIS OF PIGMENT DISPERSION
BY Robert Ritch, MD
Purpose: To synthesize recent findings regarding pigment dispersion syn-
drome in order to arrive at a hypothesis concerning the nature of an
underlying genetic predisposition.
Methods: The literature on the subject was reviewed and analyzed.
Results: Eyes with pigment dispersion syndrome differ from normal in that
theyhave a larger iris, a midperipheral posterior iris concavity that increas-
es with accommodation, a more posterior iris insertion, increased irido-
lenticular contact that is reversed by inhibition of blinking, possibly an
inherent weakness of the iris pigment epithelium, and an increased inci-
dence of lattice degeneration of the retina.
Conclusion: A gene affecting some aspect ofthe development ofthe mid-
dle third ofthe eye early in the third trimester appears at the present time
to be the most likely cause.
Pigment dispersion syndrome (PDS) is a unique and fascinating entity. Far
more prevalent than previously suspected,' it is the first common disease
leading to glaucoma for which we are on the verge of a coherent overall
explanation ofpathogenesis and pathophysiology. This paper is an attempt
to tie together many interesting and sometimes disparate and/or appar-
ently anomalous findings in order to synthesize a coherent portrait of the
PDS and pigmentary glaucoma (PG) are characterized by disruption
of the iris pigment epithelium (IPE) and deposition of the dispersed pig-
ment granules throughout the anterior segment. The classic diagnostic
'From the Departments ofOphthalmology, NewYork Eye and Ear Infirmary, NewYork, and
New York Medical College, Valhalla. Supported by the New York Glaucoma Research
Institute and an unrestricted grant from Allergan.
TR. AM. OPHTH. SOC. VOL. XCIV, 1996
triad consists of corneal pigmentation (Krukenberg spindle); slitlike, radi-
al, midperipheral iris transillumination defects, and dense trabecular pig-
mentation. The iris insertion is typically posterior, and the peripheral iris
tends to have a concave configuration. The basic abnormality in this
hereditary disorder remains unknown.
In 1899, Krukenberg' described spindle-shaped pigment deposition on
the cornea. In 1901, von Hippel3 suggested that pigment obstructing the
aqueous outflow system could lead to elevated intraocular pressure (IOP).
Levinsohn4 first suggested that pigment in the anterior chamber angle of
patients with glaucoma originated from the IPE. A cause-and-effect rela-
tionship between pigment and glaucoma found both support56 and oppo-
In 1949, Sugar and Barbour"° described 2 young, myopic men with
Krukenberg spindles, trabecular hyperpigmentation, and open angles,
whose IOP increased with mydriasis and decreased with pilocarpine. The
investigators identified the disorder as a rare, distinct form of glaucoma,
which they termed pigmentary glaucoma. More patients were subse-
quently reported, and in 1966 Sugar" reviewed 147 cases in the world lit-
erature, mentioning several additional features, including bilaterality, fre-
quent association with myopia, greater incidence in men than in women,
and a relatively young age at onset. These features were confirmed by
Scheie and Cameron."
In the 1950s, the discovery of iris transillumination defects led to the
concept that the trabecular pigment originated from the IPE and perhaps
the ciliary body.'3"4 Congenital atrophy or degeneration of the IPE was
suggested as a cause of loss of iris pigment."5"6
In 1979, Campbell'7 proposed the pathogenesis to involve mechanical
damage to the IPE during rubbing ofthe posterior iris against the anteri-
or zonular bundles during physiologic pupillary movement. Subsequently,
the autosomal dominant inheritance, natural history, reversibility, and
more precise therapeutic approaches have become increasingly delineat-
ed. Ultrasound biomicroscopic studies are presently revealing new insights
into the pathophysiology ofPDS.
Loss of iris pigment appears clinically as a midperipheral, radial, slitlike
pattern oftransillumination defects seen most commonly inferonasally and
Pigment Dispersion Syndrome
more easily in blue eyes than in brown ones (Fig 1). Although the defects
can sometimes be seen by retroillumination, they are more easily detect-
ed by a dark-adapted examiner using a fiberoptic transilluminator in a
darkened room. Infrared videography provides the most sensitive method
of detection.'8 Pigment particles deposited on the iris surface tend to
aggregate in the furrows.""9 Rarely, this pigment can be dense enough to
darken the iris or to cause heterochromia when involvement is asymmet-
ric."'20 Iris vascular hypoperfusion on fluorescein angiography has been
reported,2' a finding which awaits verification.
Iris transillumination defects. Typical defect is midperipheral, radial, and slit-like. Some
defects, especially inferiorly, have peripheral clublike endings, giving them the appearance of
an exclamation point. These peripheral transillumination dots might result from iridociliary
Anisocoria may occur with asymmetric involvement, the larger pupil
corresponding to the eye with greater pigment loss from the iriS.22-24
Alward and Haynes22 suggested the presence of an efferent defect in the
eye with the larger pupil. The pupil may be distorted in the direction of
maximal iris transillumination.'5 This would be consistent with the pres-
ence of hyperplasia of the iris dilator muscle (see below).26
Corneal endothelial pigment generally appears as a central, vertical,
brown band (Krukenberg spindle), the shape being attributed to aqueous
convection currents (Fig 2). The pigment is phagocytosed by endothelial
cells,27,28 but endothelial cell density and corneal thickness remain
unchanged compared with controls.'9 Coincident PDS and megalocornea
have been reported,12162930as have subluxated lenses.'2'3'
The anterior chamber is deeper both centrally and peripherally than
can be accounted for by sex, age, and refractive error. Davidson and asso-
ciates32 compared the central and peripheral anterior chamber depths of
patients with PDS to statistical controls. The anterior chamber was signif-
icantly deeper, and the anterior chamber volume was significantly greater
in the PDS group, the difference being greatest inferiorly.
The angle is characteristically widely open, with a homogeneous,
dense hyperpigmented band on the trabecular meshwork (Fig 3). Pigment
may also be deposited on Schwalbe's line. The iris insertion is posterior,
and the peripheral iris approach is often concave. The iris is most concave
in the midperiphery. In younger patients, the scleral spur may be poorly
Pigment Dispersion Syndrome
Pigment reversal sign in 48-year-old man. A. Inferior angle. B. Superior angle. Pigment is
denser in superior angle. Note that pigment band has sharp anterior and posterior margins
and appears smooth, indicating that pigment was deposited in past and is now localized to
region of filtering portion oftrabecular meshwork. Iris is inserted posteriorly.
IPE. These changes consisted of transformation of the normally partially
pigmented and partially smooth-muscularized anterior pigment epitheli-
um into a complete smooth muscle cell, often in a hyperplastic form, and
anteriorly migrated into the iris stroma. We felt that this type of dysplas-
tic change could not have occurred secondarily but had to be a primary
congenital abnormality ofthe iris. This finding of a congenital iris abnor-
mality certainly fits in with Dr Ritch's hypothesis. Furthermore, one could
propose that the anteriorly and centrally placed dilator muscle could
account, perhaps in part, for the posterior bowing ofthe iris that is seen in
PDS. A different degree of IPE congenital abnormality in each eye of a
patient could account for the asymmetry often seen in a patient's eyes.
Dr Ritch's point about the high incidence ofPDS, lattice degeneration
ofthe retina, and retinal detachment is indeed interesting and also may be
explained by Dr Ritch's overall genetic hypothesis. Greater numbers of
patients with this association are needed for further meaningful analysis.
Many other fascinating points are raised in this paper, too many to discuss
in this brief time. I would like to ask Dr Ritch the following questions:
In light of his hypothesis, how does a peripheral iridectomy work to
cause a flattening ofthe iris?
2.Does a laser peripheral iridectomy reduce the amount of pigment
being shed? If so, how?
3.Does the development of a normal iris stroma depend on the prior
development of a normal iris PE, or vice versa?
expressed in myopia?
Although intuitively I agree with his statement that the diagnosis of
PDS often is missed, does he have evidence to back this statement?
In closing, I congratulate the author on pulling together a wealth of
material into a working hypothesis.
Is the genetic defect linked to myopia or only more likely to be
Brini, A, Porte A, Roth A. Atrophie des couches epitheliales de l'iris. Jttude d'un cas de
glaucome pigmentaire au microscope optique et au microscope 6lectronique. Doc
Opthalmol 1969; 26:403-423.
Fine BS, Yanoff M, Scheie HG: Pigmentary "glaucoma": a histologic study. Trans Am
Acad Ophthalmol OtQlaryngol 1974; 110:314-325.
Kupfer C, Kuwabara T, Kaiser-Kupfer M. The histopathology ofpigmentary dispersion
syndrome with glaucoma. AmJ Ophthalmol 1975; 80:857-862.
Rodrigues MM. Spectrum of trabecular pigmentation in open-angle glaucoma: a clini-
copathologic study. 1976; 81:258-276.
PAUL LICHTER, M.D. The curiosity about pigmentary glaucoma has been
that way for years and years. It is one ofthe more interesting entities that
Pigment Dispersion Syndrome
we see in ophthalmology and it does give us curiosity as to why this pig-
ment is there and what it does. But pigment dispersion syndrome, which
Dr. Ritch indicated had an incidence of 2.54% in his screening ofwhites,
is a fairly common finding. And so is glaucoma. The gene frequency for
glaucoma is quite high. It is a widespread gene. One of the parts of the
pigment dispersion syndrome that most interests us is that it can, if asso-
ciated with glaucoma, lead to blindness. I want to ask Dr. Ritch why can't
there simply be two genes, because pigment dispersion is by far, much
more frequent than pigmentary glaucoma. So why can't individuals with
pigment dispersion simply have the same incidence of glaucoma as the
ROBERT RITCH, M.D. Dr. Yanoffasked four questions. First, how does an
iridotomy work in pigment dispersion syndrome? Remember that pupil-
lary block and reverse pupillary block are two sides ofthe same coin. Just
as iridotomy eliminates relative pupillary block and allows aqueous to flow
from the posterior to the anterior chamber in angle-closure glaucoma, it
relieves the reverse pupillary block in pigment dispersion syndrome and
allows aqueous to flow from the anterior to the posterior chamber. When
the iris is penetrated, one can actually see the pigmentflowing from the
anterior into the posterior chamber, the same phenomenon but opposite
direction of the "mushroom cloud" of aqueous and pigment which flow
into the anterior chamber when the iris is penetrated in pupillary block.
In addition to allowing equilibration of aqueous humor between the two
chambers, the accentuation of the iris concavity with accommodation is
significantly reduced if not totally eliminated.
Second, does laser iridotomy completely eliminate pigment disper-
sion? No, it does not. Drs. William Haynes and Lee Alward together with
our group reported such a patient recently in Ophthalmic Surgery and
Lasers. This patient had massive pigment liberation while playing basket-
ball and this was completely eliminated by low-dose Pilocarpine pretreat-
ment. Nevertheless, I have performed only about 20 iridotomies. My pre-
ferred treatment is pilocarpine in those patients who do not have periph-
eral retinal abnormalities, but ofcourse the younger patients cannot toler-
ate miotic drops because they cause accommodative spasm and induced
myopia. However, they do very well with Ocuserts. We have started hun-
dreds ofpatients on these.
Third, does the development of the normal iris stroma depend on the
normal development ofthe iris pigment epithelium? I just cannot answer
that question. I am not a developmental biologist and I don't know if any
one has ever looked at this. It is possible that normal development ofone
is dependent on the other, or it is possible that both of them respond to
the same transmitter during differentiation and that their differentiation
occurs together as a parallel process.
Fourth, is the genetic defect linked to myopia or is myopia more like-
ly to be expressed in glaucoma? It is possible that the two are linked, since
it is very rare to see hyperopes with pigment dispersion.
toward the hypothesis that hyperopes can be carriers, but that myopia pre-
disposes to the phenotypic expression. Accommodation may play a much
greater role than we have previously thought and may account for why
myopes manifest the phenotypic expression.
between myopia and intelligence; pigment dispersion patients tend to be
highly intelligent, and prolonged accommodation leads to myopia. There
also seems to be a psychological profile which many patients with pigment
dispersion fit. They are often highly goal oriented, more tense than aver-
age, and somewhat hypomanic. Surgeons and lawyers are more likely to
have pigment dispersion than psychiatrists and accountants. There seems
to be a generalized adrenergic hypersensitivity. Future studies may clari-
Dr. Lichter asked the "how-many-hits-does-it-take-to-make-glaucoma"
question. I have always thought in terms ofthe three hit theory. You need
something to affect the trabecular meshwork, such as pigment in pigmen-
tary glaucoma or a gene causing decreased trabecular function in juvenile
"primary" open-angle glaucoma (JPOAG).
doubt prompted by the fact that not everyone with pigment dispersion,
even when the pigment on the meshwork appears massive, develops ele-
vated IOP. So the second hit is one which makes the meshwork suscepti-
ble to developing elevated IOP. I might add at this point, parenthetically,
that people who have pigment dispersion who do not develop elevated
IOP nevertheless may go on later to develop glaucoma at an earlier age
than they would have, had they not had some trabecular damage by pig-
ment earlier in life. We are presently looking at whether patients with pig-
ment dispersion go on to have a higher incidence of normal-tension glau-
coma later in life. The third hit would be susceptibility ofthe disc to dam-
age, since not all patients with elevated pressure develop disc damage.
However, I think that we are dealing with two different primary hits,
and it's not necessary to have two genes, at least not the genes we are
thinking of in this discussion. Pigment dispersion is very common in the
population but most of them remain undiagnosed. Most affected people
go through life never knowing they have it; in fact, most people have never
even heard of it. The juvenile glaucoma gene is common also, but proba-
I have leaned
There is a correlation
Dr. Lichter's question is no
Pigment Dispersion Syndrome
bly less common than the one for pigment dispersion. Nevertheless, the
only hard evidence we have are two studies presented at ARVO in 1995 by
Drs. Zeev Stegman and Joseph Sokol, in which we looked at trabecular
meshwork heights. Patients with pigmentary glaucoma had normal mesh-
work heights, while those with JPOAG had meshworks which were signif-
icantly smaller than normal.
I would like to take this opportunity to thank the Society for inviting
me to speak today and to thank Dr. Yanoff for his excellent discussion.