Full text of DOI:10.1007/s00417-002-0505-8

Graefe’s Arch Clin Exp Ophthalmol
(2002) 240:533–538
C L I N I C A L I N V E S T I G AT I O N
DOI 10.1007/s00417-002-0505-8
D. Pauleikhoff
D. Löffert
Pigment epithelial detachment in the elderly
G. Spital
Clinical differentiation, natural course
and pathogenetic implications
M. Radermacher
J. Dohrmann
A. Lommatzsch
A. C. Bird
Abstract Background: A prospec-
tive analysis was performed to char-
acterize the angiographic appear-
in similar visual loss. Avascular
PEDs were smaller than vascular
PEDs, and the latter were smaller
Received: 28 June 2001
Revised: 16 April 2002
Accepted: 8 May 2002
Published online: 25 June 2002
© Springer-Verlag 2002
ance, natural course and prognosis of than PCV-PEDs. During follow-up
serous pigment epithelial detach-
ments (PEDs) in elderly patients.
The aim was to differentiate PEDs
these differences were always pres-
ent, but all PEDs enlarged initially
followed by regression. This course
according to their angiographic char- was associated in all PEDs with pro-
acteristics and to analyze the specific gressive visual loss, accompanied by
clinical, visual and morphologic
course of the different PEDs.
the development of RPE atrophy in
avascular PEDs or disciform scars or
RPE tears in the two other types.
Conclusion: Despite different associ-
ations, all PEDs have a similar clini-
cal course with respect to visual loss
and enlargement or regression. This
is compatible with the proposed
common pathogenetic background
with a hydrophobic barrier in
Methods: Fluorescein and indocya-
nine green angiography were per-
formed in 101 consecutive patients
(53–87 years; 63 female, 38 male)
with clinical signs of serous PED
and drusen. Results: Different types
of serous PED were identified:
polypoidal choroidal vasculopathy
D. Pauleikhoff ( ) · D. Löffert · G. Spital
✉
M. Radermacher · J. Dohrmann
A. Lommatzsch
Department of Ophthalmology,
St. Franziskus Hospital,
Hohenzollernring 74,
48145 Münster, Germany
e-mail: dapauleikhoff@muenster.de
Fax: +49-251-9330819
(PCV)-associated PED in 14 patients Bruch’s membrane causing fluid re-
(13.9%), vascular PED in 72
sulting from RPE pumping activity
to accumulate between the pigment
epithelium and Bruch’s membrane.
(71.2%), and avascular PED in 15
(14.9%). All PEDs resulted initially
A.C. Bird
Moorfields Eye Hospital, London, UK
ICG angiographic characteristics, and to correlate the
findings with the visual and morphologic course of the
disease and the proposed pathogenetic concepts of PED.
Introduction
Serous pigment epithelial detachment (PED) in the el-
derly is a common manifestation of age-related macular
degeneration (AMD) [5, 7, 13, 22, 30, 33]. Recently
PED has been described as part of the spectrum of the
clinical manifestation of polypoidal choroidal vasculopa-
thy (PCV) [1, 21, 23, 25, 34, 35]. Whether or not the le-
sions occurring in these two settings have a common
pathogenic background is in doubt.
Patients and methods
Serous PED were analyzed and prospectively observed during fol-
low-up in 101 consecutive patients with unilateral disease. Pa-
tients were included who were seen in our department because of
recent visual loss due to a newly developed serous PED with asso-
ciated drusen, with or without clinical evidence of associated
CNV. The duration of visual loss had to be less than 3 months, but
no limit as to the extent of visual loss was assigned. In all patients
the fellow eye demonstrated signs of AMD, either early (67 pa-
The purpose of the present prospective longitudinal
study was to subdivide serous PED in elderly patients
according to their morphology by their fluorescein and

534
tients) or late (34 patients). Patients were asked for written consent
and the protocol of the study was approved by the local ethics
committee.
The group consisted of 63 women (62.4%) and 38 men
(37.6%). Their age was between 53 and 87 years (mean 70), and no
other associated eye disease was present. The duration of follow-up
was between 3 and 67 months (mean 10.7 months). The number of
patients at each follow-up visit was comparable between the differ-
ent PED groups, with 101 patients (100%) reviewed for 3 months,
80 (79.2%) for 6 months, 65 (64.4%) for 9 months, 48 (47.5%) 12
months, 40 (39.6%) for 15 months, 32 (31.7%) for 18 months and
10 (9.9%) for 24 months. At the initial examination and follow-up
examinations every 3 months the following investigations were
performed: ETDRS visual acuity, ophthalmoscopic fundus exami-
nation and fluorescein and indocyanine green (ICG) angiography.
The area of PED was analyzed on the ICG angiogram, and the PED
volume was measured using the Heidelberg Retina Tomograph
(HRT Heidelberg, Germany) using a method resulting in reproduc-
ible volume measurements as demonstrated in earlier studies [27].
The total number of eyes with visual loss in each PED group was
recorded, and the cumulative risk was analyzed using Kaplan–
Meier statistics. In addition the distribution of other associated fac-
tors in the different groups was analyzed using chi-square statistics
[2, 3]. A P value <0.05 was regarded as statistically significant. Be-
cause of the rapid diffusion of the dye into the serous PED, with
fluorescein angiography an exact distinction of the PED and the
type of the associated neovascular complex might be difficult. Be-
cause ICG demonstrates only minor diffusion into a PED the ICG
angiograms were used to distinguish between vascular and avascu-
lar lesions as well as identifying the presence of PCV [30]. Subdi-
vision of the lesions was undertaken as follows:
Fig. 1 Incidence of visual loss during follow-up (Kaplan–Meier
analysis; P=0.9)
Table 1 Initial visual acuity in the different types of PED (P=0.6)
Visual acuity
Avascular PED Vascular PED PCV-PED
Type 1 PCV-associated PED (PCV-PED): PED associated with
polypoidal choroidal vasculopathy.
Type 2 vascular serous PED: PED with angiographically visible
choroidal neovascularisation (CNV).
<20/200
20/200–20/60 6 (40%)
>20/60 8 (53.3%)
1 (6.7%)
10 (13.9%)
30 (41.7%)
32 (44.4%)
0 (0%)
7 (50%)
7 (50%)
Type 3 avascular serous PED: PED without evidence on fluo-
rescein or ICG angiography of an associated neovascular
complex. Because of the indistinct transition between
larger confluent drusen and non-vascularized, drusenoid
PED, these PEDs were defined as being >500 µm in
diameter.
ent groups of visual acuity was comparable (P=0.6)
(Table 1). This was despite significant differences in the
size (measured in mm2) of the different types. The mean
size of the PED was 13.9± mm2 in PCV-PED,
9.8±5.4 mm2 in vascular serous PED and 6.0±4.5 mm2
in avascular serous PED (P<0.001).
The mean volume of the PED also differed among
the three types: PCV-PED 3.53±1.8 mm3, vascular
serous PED 2.10±2.1 mm3 and avascular serous PED
1.13±1.1 mm3; P=0.005).
The initial visual acuity and the morphologic characteristics of the
PED size and volume were recorded according to this classifica-
tion. The natural course of the disease with relationship to the vi-
sual acuity, size and volume and morphologic development were
documented.
Results
Initial characterization and differentiation of serous PED
Follow-up characterization of different PED
The PED in the group of 101 patients were classified as
follows:
Loss of visual acuity (<3 lines) was observed during re-
view in 55 patients (54.5%): 6 patients (40%) with avas-
cular PED, 38 patients (52.8%) with vascular PED and
11 patients (78.6%) withPCV-PED (P=0.1). Also com-
paring the incidence of visual loss in relation to the dura-
tion of follow-up, no significant difference between the
different types of PED could be seen (Fig. 1) (P=0.9).
The mean final visual acuity was 1/18 in avascular PED,
Type 1 PCV-associated PED (PCV-PED): 14 patients
(13.9%); mean follow-up 9.1 months
Type 2 Vascular serous PED: 72 patients (71.2%); mean
follow-up 10.8 months
Type 3 Avascular serous PED: 15 patients (14.9%);
mean follow-up 11.9 months
The period of follow-up in the different groups was 1/24 in vascular PED and 1/30 in PCV-PED (P=0.1).
similar (P=0.6) (Table 1) The mean initial visual acuity The distribution of visual acuity also did not differ
in all groups was 20/60, and the distribution in differ- among the three groups (Table 2) (P=0.3).

535
Fig. 2 a Incidence of enlargement in size during follow-up (Kaplan–Meier analysis; P=0.1). b Mean size initially and during follow-up
(P<0.01)
Fig. 3 a Incidence of enlargement in volume during follow
(Kaplan–Meier analysis; P=0.06). b Mean volume initially and
during follow-up (P<0.01)
All three types demonstrated enlargement of the PED
during the first months of follow-up, followed by a slight
decrease in size. Increase in diameter developed similarly in
all groups and was observed in 10 avascular PEDs (66.7%),
32 vascular PEDs (44.4%) and 4 PCV-PEDs (28.6%)
(P=0.1). The time course of the incidence of enlargement
also did not differ among the three groups (Fig. 2a) (P=0.1).
However, these very similar developments occurred on
three different levels, with avascular PEDs being the small-
est, vascular PEDs intermediate in size and PCV-PEDs the
Table 2 Final visual acuity in different types of PED (P=0.3)
Visual acuity Avascular PEDVascular PED PCV-PED
<20/200
20/200–20/60 5 (33.3%)
>20/60 6 (40%)
4 (26.7%)
28 (38.9%)
29 (40.3%)
15 (20.8%)
6 (42.9%)
7 (50%)
1 (7.1%)

536
avascular PEDs had the smallest diameter and volume, vas-
cular PEDs were intermediate and PCV-PEDs were the
largest. The differences among the three groups were al-
ways significant (Figs. 2b, 3b) (P<0.01).
The morphologic changes associated with the change in
size and volume, and the cause of the progressive visual
loss were predominantly the development of geographic at-
rophy in avascular PED and of RPE tears or disciform scars
in vascular PED and PCV-PED (Table 3). RPE tears only
developed in PEDs with large size and volume (Fig. 4).
Discussion
In agreement with other studies [1, 4, 7, 8, 9, 12, 13, 15,
20, 22, 23, 25, 30, 32, 33, 34, 35] we have found that
PEDs can occur under different circumstances. However,
all three types of PED were qualitatively similar in their
behavior in that the initial visual acuity and visual deteri-
oration during review were comparable. In addition all
three types followed were alike in the course of the dis-
ease, with enlargement in size and volume over several
months and later regression associated with either RPE
atrophy, disciform scarring or RPE tears (Fig. 5).
Fig. 4 Maximal size and volume initially of PED which devel-
oped a RPE tear (×) or not (✉) during follow-up
Table 3 Morphologic development of serous PED during follow-
up (P=0.3)
However, some quantitative differences were appar-
ent. PCV-PEDs were initially larger with less tendency
to further enlargement in area or volume [1, 23, 31, 34,
35]. A majority developed hypertrophic subretinal scar-
ring. This is in contrast to reports of spontaneous resolu-
tion in this type of PED [1, 33, 35], but these studies in-
cluded PCV-PED in younger patients as well as individ-
uals of non-Caucasian origin.
Findings
Avascular Vascular
PED PED
PCV-PED all PED
Persistent PED 8 (53.3%) 38 (52.8%) 5 (35.7%) 54 (53.4%)
RPE atrophy 5 (33.3%) 8 (11.1%) 1 (7.1%) 11 (10.9%)
Disciform scar 1 (6.7%) 17 (23.6%) 6 (42.9%) 24 (23.8%)
RPE tear 1 (6.7%) 9 (12.5%) 2 (14.3%) 12 (11.9%)
Vascular PEDs were the most common category in the
largest; the difference between the three groups ws always present study. They were initially medium sized, but area
significant (Fig. 2b) (P<0.01). A similar course was seen and volume of the PED enlarged with eventual regres-
concerning the volume of the PED. An enlargement in vol- sion. This resulted most often in a disciform scar, but as
ume was observed in 10 avascular PEDs (66.7%), 32 vas- found in other studies [27] RPE tears sometimes devel-
cular PEDs (44.4%) and 5 PCV-PEDs (35.7%) (P=0.1). oped. In general RPE tears most often developed in initial
The time course of enlargement did not differ among the large PED and increased further in volume, resulting in
three groups (Fig. 3a) (P=0.1). Throughout the evolution sufficient tangential stress to rupture the RPE [8, 9, 24].
Fig. 5 Schematic summary of
the development and course of
different serous PED entities

537
Avascular PEDs as demonstrated by ICG and fluores- ment of ions and material by the RPE. In vascular lesions
cein angiography were initially smaller. Despite their the new vessel complex would contribute to the accumula-
small size their effect on visual acuity was comparable to tion (Fig. 5). Therefore, it was hypothesized that the hydro-
that of the other two types of PED. They also enlarged phobic barrier in Bruch’s membrane is responsible for the
during follow-up but remained relatively small and finally fluid accumulation under the RPE, but another factor in
regressed. After several months of follow-up there was Bruch’s membrane, the RPE or the retina is responsible for
progressive visual loss, most commonly associated with development of choroidal neovascularization under the
RPE atrophy, although the final vision was somewhat bet- RPE (“dual pathogenetic pathway”). This may result in spe-
ter than in the other two types of PED. The small number cific therapeutic approaches in the different types of PED.
of avascular PEDs that grew to the size of the other forms
all developed evidence of new vessels during the study.
The situation in PCV is less clear. In one histologic
study it has been reported that the vascular complex is inter-
It has been hypothesized that detachment of the RPE is nal to Bruch’s membrane [16]. If this is the case the venous
consequent upon the presence of a hydrophobic barrier to complex would contribute to the fluid. There is a clinical
the outward passage of fluid at the level of Bruch’s mem- impression that in some cases the complex is within the
brane, due to accumulation of neutral lipids. There is strong choroid, and both situations may exist. In the latter case, se-
circumstantial evidence to support this hypothesis [6, 10, rous detachment of the retinal pigment epithelium may not
11, 14, 17, 18, 19, 26, 28, 29], but this pathogenetic concept occur, but the lesion would present with hemorrhage or reti-
has never been shown to correlate with the natural course of nal edema. This difference in anatomic site may account for
different clinical PED entities. The similar clinical behavior the spontaneous resolution of some lesions.
of the different lesions during follow-up implies that this
In conclusion, the natural course of the different clini-
mechanism may be common to all three. The source of flu- cal entities of PED supports the concept that a specific
id may represent the only difference among them, and may significant hydrophobic barrier in Bruch’s membrane is
explain the quantitative differences. In the avascular lesions central to the pathogenesis of the lesion whether or not
the fluid is believed to accumulate from the outward move- there are new vessels of PCV.
References
1. Ahuja RM, Stanga PE, Vingerling JR,
Reck AC, Bird AC (2000) Polypoidal
choroidal vasculopathy in exudative
and haemorrhagic pigment epithelial
detachments. Br J Ophthalmol 84:479–
484
2. Altman DG (1999) Practical statistics
for medical research. Chapman and
Hall, London, p 245
8. Chuang EL, Bird AC (1988) Bilaterali- 16. Lafaut BA, Aisenbrey S, Vanden
ty of tears of the retinal pigment epi-
thelium. Br J Ophthalmol 72:918–920
9. Chuang EL, Bird AC (1988) The
Pathogenesis of tears of the retinal pig-
ment epithelium. Am J Ophthalmol
105:285–290
Broecke C, Krott R, Jonescu-Cuypers
CP, Reynders S, Bartz-Schmidt KU
(2001) Clinicopathological correlation
of retinal pigment epithelial tears in ex-
udative age related macular degenera-
tion: pretear, tear, and scarred tear.
Br J Ophthalmol 85:454–60
10. Curcio CA, Millican CL, Bailey T,
Kruth HS (2001) Accumulation of cho- 17. Moore DJ, Hussain AA, Marshall J
3. Altman DG (1999) Practical statistics
for medical research. Chapman and
Hall, London, p 365
4. Bandello F, Incorvaia C, Parmeggiani
F, Sebastiani A (2000) Idiopathic mul-
tiple serous detachments of the retinal
pigment epithelium followed by bilat-
eral central serous chorioretinopathy:
a case report. Ophthalmologica 214:362–
367
5. Bird AC (1991) Doyne Lecture: Patho-
genesis of retinal pigment epithelial
detachment in the elderly: the rele-
vance of Bruch’s membrane change.
Eye 5:1–12
6. Bird AC (1992) Bruch’s membrane
change with age. Br J Ophthalmol
76:166–168
lesterol with age in human Bruch’s
membrane. Invest Ophthalmol Vis Sci
42:265–274
(1995) Age-related variation in the hy-
draulic conductivity of Bruch’s mem-
brane. Invest Ophthalmol Vis Sci
36:1290–1297
11. Fisher RF (1987) The influence of age
on some ocular basement membranes.
Eye 1:184–189
12. Frederick ARJ, Morley MG, Topping
TM, Peterson TJ, Wilson DJ (1993)
The appearance of stippled retinal pig-
ment epithelial detachments. A sign of
occult choroidal neovascularization in
age-related macular degeneration.
Retina 13:3–7
13. Gass JDM (1997) Stereoscopic atlas of
macular diseases. Mosby, St. Louis
14. Holz FG, Sheraidah G, Pauleikhoff D,
Bird AC (1994) Analysis of lipid de-
posits extracted from human macular
and peripheral Bruch’s membrane.
Arch Ophthalmol 112:402–406
15. Kunze C, Elsner AE, Beausencourt E,
Moraes L, Hartnett ME, Trempe CL
(1999) Spatial extent of pigment epi-
thelial detachments in age-related mac-
ular degeneration. Ophthalmology
106:1830–1840
18. Okubo A, Rosa RH, Bunce CV,
Alexander RA, Fan JT, Bird AC,
Luthert PJ (1999) The relationships of
age changes in retinal pigment epitheli-
um and Bruch’s membrane. Invest
Ophthalmol Vis Sci 40:443–449
19. Pauleikhoff D, Harper CA, Marshall J,
Bird AC (1990) Aging changes in
Bruch’s membrane. A histochemical
and morphologic study. Ophthalmology
97:171–178
20. Pauleikhoff D, Knebel C, Peuser M,
Schrenk M, Wessing A (1996) Fluores-
zenzangiographie der AMD. Studie zur
Häufigkeit koagulativ behandelbarer
Läsionen. Klin Monatsbl Augenheilkd
209:309–314
7. Bird AC, Marshall J (1986) Retinal
pigment epithelial detachments in the
elderly. Trans Ophthalmol Soc UK
105:674–682

538
21. Phillips WB, Regillo CD, Maguire JI
(1986) Indocyanine green angiography
of idiopathic polypoidal choroidal vas-
culopathy. Ophthalmic Surg Lasers
27::467–470
22. Poliner LS, Olk RJ, Burgess D, Gordon
ME (1986) Natural history of retinal
pigment epithelial detachments in age-
related macular degeneration. Ophthal-
mology 93:543–551
23. Schneider U, Gelisken F, Kreissig I
(1998) Indocyanine green angiography
and idiopathic polypoidal choroidal
vsculopathy. Br J Ophthalmol 82:98–
99
24. Schoeppner G, Chuang EL, Bird AC
(1989) Retinal pigment epithelial tears.
Risk to the second eye. Am J Ophthalmol
108:683–685
26. Spaide RF, Ho-Spaide WC, Browne
RW, Armstrong D (1999) Characteriza-
tion of peroxidized lipids in Bruch’s
membrane. Retina 19:141–147
27. Spital G, Brumm G, Radermacher M,
Mueller C, Lommatzsch A, Pauleikhoff
D (2000) Volumenbestimmung von
Pigmentepithelabhebungen bei AMD
mittels Laser Scanning Tomographie.
Ophthalmologe 97:173–180
28. Starita C, Hussain AA, Pagliarini S,
Marshall J (1996) Hydrodynamics of
ageing Bruch’s membrane: implica-
tions for macular disease. Exp Eye Res
62:565–572
29. Starita C, Hussain AA, Patmore A,
Marshall J (1997) Localization of the
site of major resistance to fluid trans-
port in Bruch’s membrane. Invest
Ophthalmol Vis Sci 38:762–767
30. Wolf S, Remky A, Elsner AE,
Arend O, Reim M (1994) Indocyanine
green video angiography in patients
with age-related maculopathy-related
retinal pigment epithelial detachments.
Ger J Ophthalmol 3:224–227
32. Yannuzzi LA, Slakter JS, Sorenson JA,
Guyer DR, Orlock DA (1992) Digital
indocyanine green videoangiography
and choroidal neovascularization.
Retina 12:191–223
33. Yannuzzi LA, Hope-Ross M, Slakter
JS, Guyer DR, Sorenson JA, Ho AC,
Sperber DE, Freund KB, Orlock DA
(1994) Analysis of vascularized pig-
ment epithelial detachments using in-
docyanine green videoangiography.
Retina 14:99–113
34. Yannuzzi LA, Ciardella A, Spaide RF
(1997) The expanding clinical spec-
trum of idiopathic polypoidal choroidal
vasculopathy. Arch Ophthalmol 10:18–
26
35. Yannuzzi LA, Wong DW,
Sforzolini BS, Goldbaum M, Tang KC,
Spaide RF (1999) Polypoidal choroidal
vasculopathy and neovascularized age-
related macular degeneration. Arch
Ophthalmol 117:1503–1510
25. Spaide RF, Yannuzzi LA, Slakter JS,
Sorenson J, Orlach DA (1995) In-
docyanine green videoangiography of
idiopathic polypoidal choroidal vascu-
lopathy. Retina 15:100–110
31. Yannuzzi LA, Sorenson J, Spaide RF,
Lipson B (1990) Idiopathic polypoidal
choroidal vasculopathy (IPCV). Retina
10:1–8