Journal of the International Neuropsychological Society (1996), 2, 565-570.

Copyright © 1996 INS. Published by Cambridge University Press. Printed in the USA.

CASE STUDY

Recovery of memory and executive function following

anterior communicating artery aneurysm rupture

MARK D'ESPOSITO, 1 MICHAEL P. ALEXANDER,2 RICHARD FISCHER,3

REGINA McGLINCHEY-BERROTH,4'5 AND MARGARET O'CONNOR 5
'Department of Neurology, University of Pennsylvania Medical Center, Philadelphia, PA 19104
department of Neurology, Boston University School of Medicine and Stroke Rehabilitation Program,
Braintree Hospital, Braintree, MA 02185
3
Department of Neuropsychology, New England Sinai Hospital, Stoughton, MA and Department of Neurology,
Tufts University School of Medicine, Boston, MA 02702
4
GRECC, West Roxbury VA Medical Center, West Roxbury, MA and Department of Psychiatry,
Harvard Medical School, Boston, MA 02132
5
Memory Disorders Research Center, Boston University School of Medicine, Boston, MA 02215
(RECEIVED July 5, 1995; REVISED December 19, 1995; ACCEPTED January 4, 1996)

Abstract
We studied the recovery of memory and executive function in 10 patients following anterior communicating artery
aneurysm (ACoA) rupture and repair. Patients were tested at 2 consecutive points in time following surgery
(approximately at 2 and 3 months). At the first testing, the patients divided into 2 groups based on the severity of
impairment on executive measures. Both groups had severe anterograde amnesia, but only patients with severe
executive impairments had retrograde amnesia with a temporal gradient. At second testing, both groups had
persistent severe anterograde amnesia. The dysexecutive group showed significant improvement in executive
deficits and in retrograde amnesia, with attenuation of the temporal gradient. Patients with more severe executive
impairments had more extensive bilateral frontal lesions than other patients. These results suggest that the cognitive
profile following ACoA rupture changes with time. Time postonset following aneurysm rupture and lesion site are
both critical for defining the neuropsychological profile, and determining the underlying cognitive mechanisms in
this neurological disorder. (JINS, 1996, 2, 565-570.)
Keywords: Anterior communicating artery aneurysm, Amnesia, Executive function, Recovery

INTRODUCTION retrograde amnesia in ACoA patients as a group that was

The profile of cognitive impairment following ACoA rup-
ture and repair consists of a variable degree of amnesia, ex-
ecutive function impairment, and confabulation (Alexander
& Freedman, 1984). Several investigators have explored spe-
cific aspects of this cognitive impairment in detail in small
groups of patients. For example, Irle et al. (1992) described
a spectrum of anterograde amnesia in ACoA patients rang-
ing from a severe deficit comparable to that seen in global
amnesics (i.e., Korsakoff's syndrome) to no memory deficit
at all. Likewise, Gade et al. (1990) described a pattern of
Reprint requests to: Mark D'Esposito, Cognitive Neurology Section,
Department of Neurology, Hospital of the University of Pennsylvania, 3400
Spruce Street, Philadelphia, PA 19104-4283.
565
similar, to Korsakoff's syndrome. They observed a temporal
gradient of remote recall with information from earlier de-
cades being retrieved better than information from more re-
cent decades. The executive function impairments commonly
present following ACoA rupture have only been character-
ized in a few patients (Shoqeirat et alM 1990; Parkin et al.,
1994), and a clear understanding has not emerged. Many,
but not all, patients with ACoA rupture confabulate, but in
those who do, confabulation ranges from spontaneous and
fantastic to confabulations that are only obtained when pro-
voked (Fischer et al., 1995).
Most investigations of the "ACoA syndrome" have stud-
ied patients who were selected because they exhibited cog-
nitive deficits. Studies of unselected ACoA rupture patients
have, however, found persistent cognitive deficits present
566
in a large percentage of patients (Laiaconaet al., 1989; Sten-
houseetal., 1991). All studies have found a variable neuro-
psychological profile without a clear relationship between
the cognitive profile and the site of brain damage. Thus, the
effects of vasospasm or a "toxic" etiology have been put
forward as explanations for the cognitive impairments, but
these potential etiologies are unlikely. For example, 3 of 6
patients in one study with patients who exhibited global cog-
nitive impairments had no or mild vasospasm (Stenhouse
et al., 1991). Moreover, the exact nature of a possible "toxin"
has not been elucidated (Laiacona et al., 1989).
Several factors contribute to the difficulty in uncovering
clinicoanatomic relations in ACoA patients. First, most stud-
ies have not grouped patients according to lesion site, in-
cluding patients with a mixture of lesions, or even no lesions
at all (Stenhouse et al., 1991). Second, time postonset of
aneurysm rupture is not controlled; patients have been tested
at widely varying points during their recovery (Laiacona
et al., 1989; Stenhouse et al., 1991; Irle et al., 1992). Fi-
nally, lesion site may not have an unvarying relationship
with deficit. Different lesion sites may produce different def-
icits that also have differing potentials for recovery. We have
demonstrated such an interaction for lesion site, confabula-
tion, and time postonset in ACoA patient (Fischer et al.,
1995). In this study, it was our hypothesis that there are sim-
ilar interactions for executive function impairment, lesion
site and time postonset; and for retrograde amnesia, execu-
tive function impairments and time postonset in these pa-
tients. Only longitudinal assessment of ACoA patients will
characterize the cognitive profile of this neurological dis-
order and the relationship of cognitive deficits to lesion sites.
To our knowledge, there have been no reports that evaluate
the effect of recovery of specific cognitive functions in this
patient group.
METHODS
Research Participants
Ten consecutive patients admitted to Braintree Rehabilita-
tion Hospital who had had rupture and repair of an anterior
communicating aneurysm were studied. There were 4 men
and 6 women with a mean age of 47 years (range 27-66)
and 14 years of education (range 9-20). All patients were
right-handed.
Evaluation
Each patient underwent a comprehensive neuropsycholog-
ical evaluation:
1. Genera! intellectual functioning: Wechsler Adult Intel-
ligence Scale-Revised (Wcchsler, 1981)
2. Executive functions: backward digit span of the Wech-
sler Memory Scale-Revised (Wcchsler, 1987), FAS test
(Bcnton, 1968), Wisconsin Card Sorting Test (Heaton,
1981), and the Trail Making Test part B (Rcitan, 1958)
M. D'Esposito et al.
3. Anterograde amnesia: supraspan list learning test [first
testing, California Verbal Learning Test (Delis et al.,
1987); second testing, Rcy Auditory Verbal Learning Test
(Lczak, 1983)]
4. Retrograde amnesia: Famous Faces Test (Albert ct al.,
1979).
All scores obtained in patients were compared to pub-
lished normative values (Sprecn & Strauss, 1991) except
for the Famous Faces Test in which normative values were
obtained from 10 age- (t = .28, df = 18) and education-
matched (/ = .21, df = 18) control subjects. Each patient
was tested at two consecutive points in time after onset. Ini-
tial testing was performed at an average of 59 days (range
32-85) postonset and later testing was performed at an av-
erage of 89 days (range 60-110), resulting in a mean of 30
days between early and later testing.
All patients had a postoperative CT brain scan, on aver-
age 52 days (range 19-76 days) after surgery. Analysis was
done by two raters (MD and MPA), who were blind to each
other's rating and to the neuropsychological results. Le-
sions were scored along three dimensions: (1) latcralky; (2)
regional involvement of basal forebrain, striatal, medial fron-
tal, orbitofrontal or dorsolateral frontal structures by the cri-
teria of Irle et al. (1992); and (3) relative extent of frontal
lesions (0%, <25%, 25-75%, or >75%). There was 98%
agreement between raters, and when there was not com-
plete agreement, a consensus between raters was obtained,
RESULTS
Analysis of patients1 performance on tasks of executive func-
tion during early testing revealed two distinct groups. Pa-
tients in Group 1 (n ~ 5) were severely impaired ( > 2 SDs
from published normative means) on at least three of four
executive measures (dysexecutive group). In contrast, pa-
tients in Group 2 (n = 5), performed within normal limits
on at least three of four measures (see Table 1). There were
no significant differences on the remaining measures be-
tween these patient groups in age (t = 1.49, df = 8), edu-
cation (/ = .88, df = 8), general intellectual functioning
[VIQ: (f = .66); PIQ: (t = 1.93, df = 8)], or the number of
days between initial and later testing (/ = .77, df = 8).
First Evaluation (Table 1)
Both groups demonstrated severe deficits in new learning
as evidenced by their supraspan list learning. There was no
difference between groups in the percentage of items re-
called during Trial 5 (t = .83, df = 8), during delayed free
recall after distraction (/ = 1.73, df — 8), or the percentage
lost from Trial 5 to the delay (t = .23, df = 8). Only Group
1 showed evidence of retrograde amnesia during recall on
the Famous Faces Test as compared to controls (r = 2.11,
df - \3,p< .002). Regression analyses of the percentage
of faces correctly recalled across decades from the 1950s
through the 1980s was conducted. A significant negative
ACoA syndrome 567

Table 1. Demographic characteristics and ncuropsychological performance between two groups of ACoA patients

Patient group
Group 1 (dysexecutive) (3roup 2 (normal)
Time 1 Time 2 Time 1 Time 2
Variable M SD M SD M SD M SD

Age 52.4 7.7 — 42.4 12.9 -

Education 12.6 2.0 - 14.4 4.2 -
WAIS-R VIQ 96.6 10.1 - 102.8 17.8 -
WAIS-R PIQ 89.6 5.9 - 110.5 23.7 -
Digits backward 2.6 1.7 4.0 1.6 5.0 0.8 5.5 0.6
WCST (categories) 1.2 2.2 3.5 2.8 4.8 1.5 5.3 1.5
Verbal fluency (words) 13.8 3.3 26.2 10.8 28 10.2 33.8 12.8
Trails B (s) 324.8 105.7 153.6 55.1 83.4 11.4 83.3 37.0
Supraspan Trial 5 44.0% 12.1 51.1% 30.0 45.3% 17.3 64.9% 21.5
Supraspan delayed recall 0.0% 0.0 10.77o 12.1 17.8% 23.0 20.0% 15.6
Supraspan Trial 5—delay 44.0% 12.1 40.4% 6.7 27.5% 11.9 44.9% 24.6
Famous faces total recall 44.8% 21.0 61.4% 21.5 74.1% 29.9 82.4% 13.2

linear relationship was revealed for Group 1 (r = - . 7 1 , erage of 84.1%, whereas the improvement in Group 2 was
p < .01) but not for Group 2. The negative correlation in only 10.2%. Performance in Group 1 remained impaired on
Group 1 is consistent with the existence of a temporal gra- executive measures compared to Group 2, but this reached
dient (Figure 1). significance for only the Trail Making test (/ = 2.37, df -
8, p < .05). The degree of anterograde amnesia remained
severe in both groups. There was no significant improve-
Second Evaluation (Table 1) ment noted between testing during Trial 5 learning (Group
1: / = .31, df = 4; Group 2: / = 1.90, df = 4); delayed free
In Group 1 (dysexecutive group), performance improved on
recall (Group I: / = 2.38, df = 4; Group 2: / = .58, df= 4);
all executive measures from the first to the second evalua-
or the number of items lost from Trial 5 to the delay (Group
tion. On two of these measures, the extent of improvement
I: t = 1.33, df = 4); Group 2: / = 1.37, df = 4). In Group 1,
was statistically significant (FAS: / = 2.68, df = 4,/? < .05;
improvement was noted in the severity of retrograde amne-
Trails B: / = 2.86, df= A,p < .05). No significant change
sia (t = 3.21, df = 4, p = .06) but performance was still
was seen on these measures in Group 2. The amount of im-
impaired compared to controls (/ = 2.11, df = 13, p < .05).
provement across executive measures between the first and
Regression analyses revealed that the pattern of retrograde
second evaluation was calculated. Group I improved an av-
loss in this group no longer showed a significant linear re-
lationship (r = .31). This finding is consistent with an at-
tenuation of the temporal gradient in the dysexecutive group
100-, (Figure 1).
Lesion analysis revealed that Group I patients had bilat-
eral medial frontal lesions including the anterior cingulate
** so
o and corpus callosum, as well as striatal (caudate) and basal
o forebrain damage (Figure 2). Patients in Group 2 had le-
o 60- sions limited to basal forebrain (Figure 3) except for two
01
O> patients who also had small orbital frontal lesions. The le-
re sion profile for each patient is presented in Table 2.
c 40-
0)
u
l_
0)
Q. 20-
DISCUSSION
Following rupture and repair of anterior communicating an-
1950 1960 1970 1980 eurysms, our patients showed varying degrees of cognitive
Decade dysfunction in the domains of memory and executive func-
Fig. 1. Pattern of recovery of retrograde amnesia in Group (dys- tions. All patients exhibited significant anterograde amne-
executive group) on the Famous Faces Test. sia that persisted over time. In the early stage following
 568 M. D'Esposito et til.

Fig. 2. MRI scan of a patient in Group I demonstrating lesions in the Fig. 3. MRI scan of a patient in Group 2 demonstrating a lesion in
medial frontal lobes, basal forebrain (' septum) and striatum {caudate). the basal forebrain (septum).

surgery (1-2 months), two distinct groups could be identi-
fied, based on the magnitude of executive function im-
pairment. Further analysis indicated that this executive
impairment correlated with extent of frontal lobe damage.
Dyscxecutivc patients had extensive bilateral frontal le-
Table 2. Lesion sites for all patients
Frontal
Group Patient (L/R)
1 (dysexecutive) 1 f + f / 1
-) t f
3 + t / +
4 • / t
5 0
2 (normal) 6 • / •
7 t
8
0 «)
sions; patients who performed relatively normally on exec-
utive measures did not have significant frontal lesions. Only
patients with executive function impairments had signifi-
cant difficulty recalling remote information, and a signifi-
cant temporal gradient was observed in their pattern of recall
(poorer recall in more recent decades). Dysexecutive pa-
tients (Group 1) showed significant improvement in exec-
utive function within the next month. Some improvement
might have been due to practice effect (two testings, 1 month
apart). Group 2, however, had only modest improvement,
suggesting practice effects are small. Furthermore, the dys-
Stnatal Basal
executive patients were and remained densely amnesic, so
(L/R) forebrain
thai no explicit benefit of practice could be exploited. These
+ t,. + /() patients also showed improvement in recalling information
+ / f ()/ + from more recent decades, thereby flattening the slope of
* 0/ • their retrograde deficit. This change could not be due to
+•/() practice.
WO
Studying the recovery of cognitive function after an acute
0 s
brain insult provides a unique opportunity to gam insight
/() 0
regarding interactions among different cognitive processes,
i

0 0
F'or example, the temporally graded retrograde amnesia that
10 0 0 0
was observed only in the dysexecutive group supports the
hypothesis that this pattern of retrograde amnesia may he
related to the combination of executive and memory im-
l"i»i" t r o n t a l l e s i o n s . 0 no lesion. • • 25*4, • • IS IS'",
• * <• • 7V£ . F;or striatal and basal I'orebrain lesions, 0 absent. * pairments. 'This finding is consistent with the common
present observation of a temporally graded retrograde amnesia in
ACoA syndrome
Korsakoff's syndrome (Albert et al., 1979), another disor-
der in which patients have both executive and memory im-
pairments (Butters & Stuss, 1989). Improvement in our
patients' retrograde amnesia was concurrent with improve-
ment in executive function, despite a persistent anterogradc
memory deficit. Impairment in strategies for retrieving re-
mote information may contribute to their impaired perfor-
mance. Gade (Gade & Mortensen, 1990) also found a
temporal gradient pattern in remote recall in a large group
of ACoA patients. Their patients had significant antero-
grade memory impairments, but they were not tested on ex-
ecutive measures assessing frontal lobe function. Since lesion
site was not reported either, it also cannot be determined
whether their group of patients showed a similar relation-
ship between retrograde amnesia, executive function im-
pairment, and lesion site. Most of their patients were tested
approximately 2 years after injury, suggesting that retro-
grade amnesia can persist. The mechanisms underlying per-
sistent retrograde amnesia in ACoA patients remain unclear,
although persistent impairment in executive function would
seem the likely cause.
All of our patients had severe anterograde amnesia that
persisted up to 4 months following aneurysm rupture. In a
study by Irle (1992), anterogradc amnesia was investigated
in a large group of ACoA patients who were selected be-
cause of lesions identified in the ventromedial frontal lobe
or striatum. They found a clear relationship between lesion
site and severity of amnesia: Patients with combined le-
sions in the basal forebrain and striatum, or basal forebrain,
striatum and frontal lobes, had a severe memory deficit,
whereas patients with lesions in the basal forebrain or stri-
atum alone showed essentially no deficit. Our findings are
not consistent with these data since several of our patients
had lesions restricted to the basal forebrain, yet had severe
anterograde amnesia. Since the patients in Irle et al.'s study
were evaluated at times much later postonset (1-5 years af-
ter surgery) than our patients, it seems possible that antero-
grade amnesia continues to recover, especially in patients
with less extensive lesions.
Executive function has been investigated in detail in only
a few studies of ACoA patients. When assessed, patients have
shown impairments on tests thought to be sensitive to fron-
tal lobe function. In one study, a single patient was studied
3 years after aneurysm rupture (Parkin et al., 1994) and in
another, the time of testing was not reported (Shoqeirat
ct al., 1990). Neither study reported details about the lesion
site in their patients. Similar to the studies of memory func-
tioning in ACoA patients, it is apparent that impairments in
executive function can also persist. The factors that lead to
poor recovery of executive function in ACoA patients have
not been elucidated, but large frontal lesions are the likely
cause in most patients.
Our study demonstrates that the cognitive profile follow-
ing ACoA rupture changes with time. During the first 3
months following aneurysm rupture, there can be signifi-
cant recovery in specific cognitive functions such as exec-
utive function, whereas others, such as anterogradc memory,
569
can remain quite impaired. Although the number of patients
in our study is small, there appears to be a relationship be-
tween lesion site, time of testing after injury, and the pattern
of cognitive deficits. It will be crucial for future neuropsy-
chological studies of ACoA patients to account for both re-
covery and lesion site when investigating the cognitive and
behavioral profiles of the ACoA syndrome.
ACKNOWLEDGMENTS
Supported in part by NIH grants NS 26985 (Memory Disorders
Research Center, Boston University School of Medicine) and NS
01762 (M.D.). This study was presented in part at the Inter-
national Neuropsychological Society meeting in Cincinnati, OH,
February, 1994. The authors wish to thank Drs. Marie Walbridge
and Randall Otto for performing the neuropsychological testing
on many of these patients.
REFERENCES
Albert, M., Butters, N., & Levin, J. (1979). Temporal gradients in
the retrograde amnesia of patients with alcoholic Korsakoff's
disease. Archives of Neurology, 36, 211-216.
Alexander, M. & Freedman, M. (1984). Amnesia after anterior com-
municating artery aneurysm. Neurology, 34, 752-757.
Benton, A. (1968). Differential behavioral effects in frontal lobe
disease. Neuropsychologia, 28, 171-179.
Butters, N. & Stuss, D. (1989). Dienccphalic amnesia. In F. Boiler
& J. Grafman (Eds.), Handbook of neuropsychology (Vol. 3,
pp. 107-148). New York: Elsevier Science Publishers.
Delis, D., Kramer, J., Kaplan, E.. & Ober, B. (1987). California
Verbal Learning Test manual. San Antonio, TX: Psychological
Corporation.
Fischer, R., Alexander, M., D'Esposito, M., & Otto, R. (1995).
Neuropsychological and neuroanatomical correlates of confab-
ulation. Journal of Clinical and Experimental Neuropsychol-
ogy, 77,20-28.
Gade, A. & Mortensen, E. {1990). Temporal gradient in the re-
mote memory impairment of amnesic patients with lesions in
the basal forebrain. Neuropsychologia, 28, 985-1001.
Heaton, R. (1981). Manual for the Wisconsin Card Sorting Test.
Odessa. FL: Psychological Assessment Resources, Inc.
Irle, E., Wowra. B., Kuncrt. H., Hampl. J., & Kunze, S. (1992).
Memory disturbance following anterior communicating artery
rupture. Annals of Neurology, 31, 473-480.
Laiacona, M., Dc Santis, A., Barbarotto, R., Basso, A., Spagnoli,
D., & Capitani, E. (1989). Ncuropsychological follow-up of
patients operated for aneurysms of anterior communicating ar-
tery. Cortex, 25, 261-273.
Lczak, M. (1983). Neuropsychological assessment (2nd ed.). New
York: Oxford University Press.
Parkin. A., Ycomans. J., & Bindschaedler. C. (1994). Further char-
acterization of the executive memory impairment following fron-
tal lobe lesions. Brain and Cognition, 26, 23-42.
Rcitan, R. (1958). Validity of the Trail Making Test as an indica-
tion of organic brain damage. Perceptual and Motor Skills, S,
271-276.
570
Shoqeirat, M.f Mayes, A., MacDonald, C , Meudell, P., & Picker-
ing, A. (1990). Performance on tests sensitive to frontal lobe
lesions by patients with organic amnesia: Leng & Parkin revis-
ited. British Journal of Clinical Psychology, 29, 401-408.
Spreen, O. & Strauss, E. (1991). A compendium of neuropsycho-
logical tests: Administration, norms, and commentary. New
York: Oxford University Press.
M. D'Esposito et ah
Stenhouse, L., Knight, R., Longmore, B., & Bishara, S. (1991).
Long-term cognitive deficits in patients after surgery on aneu-
rysms of the anterior communicating artery. Journal of Neu-
rology, Neurosurgery, and Psychiatry, 54, 909-914.
Wechsler, D. (1981). Wechsler Adult Intelligence Scale-Revised
manual. New York: Psychological Corporation.
Wechsler, D. (1987). Wechsler Memory Scale-Revised manual. New
York: Psychological Corporation.