Вы находитесь на странице: 1из 3

BRIEF REPORTS

BRIEF
AmREPORTS
J Psychiatry 154:6, June 1997

Meta-Analysis of Postmortem Studies of Alzheimers


Disease-Like Neuropathology in Schizophrenia

Ross J. Baldessarini, M.D., James D. Hegarty, M.D., M.P.H.,


Edward D. Bird, M.D., and Francine M. Benes, M.D., Ph.D.

Objective: To evaluate the hypothesis that patients with schizophrenia who have been treated
with neuroleptics have a high rate of Alzheimers disease-like neuropathology. Method: Neuro-
pathological studies indicating the presence or absence of Alzheimers disease-like neuropathol-
ogy in the postmortem brains of patients with schizophrenia, normal comparison subjects, and
comparison subjects who had affective disorder were evaluated with Mantel-Haenszel chi-square
and odds ratio analyses. Results: Ten studies with relevant data were reviewed; none of eight
with comparisons indicated that Alzheimers disease-like neuropathology was more likely to be
found in the brains of patients with schizophrenia than in the brains of comparison subjects.
Conclusions: Suggestions that cerebral plaques and neurofibrillary tangles are more common in
schizophrenia in association with neuroleptic treatment were not supported.
(Am J Psychiatry 1997; 154:861863)

P ositive long-term outcomes in studies of patients


with schizophrenia may have declined since the
1970s. Several reasons have been suggested for the de-
ropathology in schizophrenia, paying particular atten-
tion to evidence of secular trends that might be consistent
with a rising incidence of such neuropathology in the cur-
cline, including the possibility that it is related to narrow- rent era of long-term antipsychotic therapy.
ing of diagnostic criteria and that more severely or
chronically ill subjects are now available for study (1, 2).
Another suggestion is that there has been an increase in METHOD
the rate of dementia associated with neuropathological
changes in schizophrenia. These include an excess of the A computer-assisted literature search identified studies of postmor-
type of senile plaques and neurofibrillary tangles seen in tem neuropathology in schizophrenia. We evaluated studies of the
presence or absence of Alzheimers disease-like neuropathology in the
Alzheimers diseaseparticularly since the 1950s, when postmortem brains of patients with schizophrenia, normal comparison
neuroleptic treatment was introduced (24). In view of subjects, and comparison subjects with affective disorders. Mantel-
the fact that a century of research seeking a specific Haenszel chi-square analysis was applied for risk differences in the in-
neurohistopathology in schizophrenia has yielded largely dividual studies, for an overall effect of schizophrenia, and for homoge-
negative or inconclusive findings (5), and because of the neity of rates across studies. Odds ratios and their 95% confidence
intervals also were computed. Correlations between the incidence of
potential public health significance of the possibility that Alzheimers disease-like changes and the year of the report and the sub-
there has been an increase in neuropathological changes jects mean age were tested with Spearman rank correlation analysis; t
in schizophrenia related to long-term neuroleptic treat- tests were used compare means and standard deviations.
ment, this possibility requires closer consideration. Ac-
cordingly, we conducted a meta-analysis of available
data concerning the risk of Alzheimers disease-like neu- RESULTS

Nine reports provided pertinent data (4, 613); we


Received May 20, 1996; revision received Sept. 24, 1996; accepted also used new, unpublished data (E.D. Bird and F.M.
Oct. 1, 1996. From the Departments of Psychiatry and Neurology and Benes, 1996), based on diagnoses and tissue acquisition
the Neuroscience Program, Harvard Medical School, Boston, and the
Mailman Research Center, McLean Division of Massachusetts Gen- detailed elsewhere (14) (tables 1 and 2). Two of the
eral Hospital, Belmont, Mass. Address correspondence to Dr. Baldes- published reports lacked comparison groups (4, 10)
sarini, Mailman Research Center, McLean Hospital, 115 Mill St., Bel- and were excluded from chi-square analysis. Overall,
mont, MA 02178; rjb@mrc309.mclean.org (e-mail).
Supported in part by NIMH grants MH-31154 and MH-47370, the
the eight remaining reports found similar rates of Alz-
Bruce J. Anderson Foundation, and the McLean Hospital Private Do- heimers disease-like changes in the brains of patients
nors Neuropharmacology Research Fund. with schizophrenia (14.5%, range=0%34%) (table 2)
The authors thank Dr. Patrick McGeer, who provided prepublica- and comparison subjects, including those with affective
tion information on his reviews of this topic, Dr. Martin Teicher, who
provided a microcomputer program for the Mantel-Haenszel analy- disorder (16.5%, range=0%32%) or excluding these
sis, and Dr. Deborah Yurgelun-Todd, who provided helpful comments. subjects (12.5%, range=0%19.9%) (table 1). None of

Am J Psychiatry 154:6, June 1997 861


BRIEF REPORTS

TABLE 1. Description of Studies of Alzheimer-Like Neuropathology in Postmortem Brains of Patients With Schizophrenia and Comparison
Subjects
Diagnostic Criteria Age of Subjects (years) % Men
For Alzheimers
Comparison For Disease-Like Patients With Comparison Patients With Comparison
Study Year Group Schizophrenia Neuropathology Schizophrenia Subjects Schizophrenia Subjects
Corsellis (6) 1962 Patients with Unspecified Unspecified 65 53 50 52
affective clinical criteria clinical criteria
disorder
Jellinger (7) 1985 Normal St. Louis Unspecified 64
subjects criteria (14) clinical criteria
Buhl and Bojsen- 1988 Patients with ICD-8 Unspecified 80 70 39 43
Mller (8) affective clinical criteria
disorder
Soustek (9) 1989 None Unspecified Unspecified 58 58
clinical criteria clinical criteria
Bruton et al. (10) 1990 Normal St. Louis Unspecified 72 71 59 57
subjects criteria (14) clinical criteria
Prohovnik et al. (11) 1993 Patients with Unspecified National Insti- 78 75 46 30
affective clinical criteria tute on Aging
disorder criteria (14)
Purohit et al. (12) 1993 Normal DSM-III-R National Insti- 77 75 54
subjects tute on Aging
criteria (14)
Arnold et al. (13) 1994 Normal DSM-III-R National Insti- 77 71 60 40
subjects tute on Aging
criteria (14)
Wisniewski et al. (4) 1994 None Unspecified Unspecified 79
clinical criteria clinical criteria
Bird and Benes (un- 1996 Normal St. Louis National Insti- 50 66 78 54
published)a subjects criteria (14) tute on Aging
criteria (14)
b
Mean 70 69 54 50
aNew data (diagnostic methods reported in reference 14).
bMean ages of the schizophrenic and comparison groups were not significantly different (t=0.28, df=15, p=0.78).

TABLE 2. Findings of Alzheimer-Like Neuropathology in Studies of Postmortem Brains of Patients With Schizophrenia and Comparison
Subjects
Subjects With Alzheimers
Disease-Like Neuropathology Analysis
Number of Subjects Schizophrenia Comparison 95%
Odds Confidence
Study Schizophrenia Comparison N % N % Ratio Interval
Corsellis (6) 16 27 4 25.0 8 29.6 0.79 0.183.12
Jellinger (7) 172 117 16 9.3 8 6.8 1.40 0.593.55
Buhl and Bojsen-Mller (8) 23 14 8 34.8 4 28.6 1.33 0.326.12
Soustek (9) 225 44 19.6
Bruton et al. (10) 48 56 16 33.3 18 32.1 1.06 0.462.40
Prohovnik et al. (11)a 544 47 52 9.6 7 14.9 0.60 0.271.53
Purohit et al. (12) 13 12 0 0.0 0 0.0
Arnold et al. (13) 15 5 1 6.7 0 0.0
Wisniewski et al. (4) 102 35 34.3
Bird and Benes (unpublished)b 135 201 12 8.9 34 16.9 0.48 0.230.94
Total 1,293 479 188 14.5 79 16.5 0.86 0.651.15
aCorrected for 33% verification of Alzheimers disease findings in subsample.
bNew data (diagnostic methods in reference 14).

the eight studies with comparisons indicated a signifi- which the comparison patients with affective disorder
cantly higher rate of Alzheimers disease-like neuropa- were significantly older than the patients with schizo-
thology in the brains of patients with schizophrenia phrenia (mean=66.0 years, SD=16.0, versus mean=49.5
(overall Mantel-Haenszel 2=1.44, df=1, p=0.23), but years, SD=20.0) (t=8.4, df=334, p<0.001), we found a
there was some lack of homogeneity of rates across weakly lower rate of Alzheimers disease-like neuropa-
studies (Mantel-Haenszel 2=5.47, df=1, p=0.02). thology in the brains of patients with schizophrenia
In the new data (E.D. Bird and F.M. Benes, 1996), in (2=4.40, df=1, p=0.04) (table 2). However, among the

862 Am J Psychiatry 154:6, June 1997


BRIEF REPORTS

subjects with Alzheimers-disease neuropathology, the impairments (15). Cross-sectional assessments have not
mean ages of the patients with schizophrenia (76.5, indicated declining cognition in schizophrenia: in 10
SD=8.70) and the patients with affective disorder (77.3, studies published in 19641980 (15), IQ and year of
SD=9.4) were similar (t=0.03, df=44, p>0.90). study were uncorrelated (r=0.09, df=8, p>0.1). Instead,
Across studies, the risk of Alzheimers disease-like many psychotic patients show improvements in cogni-
changes did not increase significantly with age in tion during antipsychotic treatment, generally with oth-
schizophrenia (rs=0.40, df=8, p=0.23). The mean rates erwise improved clinical status (1618).
of Alzheimers disease-like neuropathology were simi- The information reviewed here provides little support
lar in the brains of patients with schizophrenia and for the proposal that Alzheimers disease-like neurode-
those of normal comparison subjects and patients with generative changes are more common in schizophrenia
affective disorder, and the overall odds ratio (0.86) was than in other idiopathic psychiatric disorders or in nor-
not elevated above 1.0 (table 2). mal subjects. Moreover, the unsettling proposal that
Reanalysis of the data of Wisniewski et al. (4) yielded such neuropathological changes may have increased
only weak support for the hypothesis that postmortem since the introduction of long-term maintenance treat-
incidence of Alzheimers disease-like changes has in- ment with neuroleptic agents remains unproved.
creased in association with antipsychotic drug treat-
ment in schizophrenia. The rates of plaques with or
REFERENCES
without neurofibrillary tangles in their cases autopsied
before 1952 (N=35 [57.4%] of 61) compared with 1. Hegarty JD, Baldessarini RJ, Tohen M, Waternaux CM, Oepen
those autopsied after 1954 (N=25 [61.0%] of 41) did G: One hundred years of schizophrenia: a meta-analysis of the
not differ (2=0.13, df=1, p=0.88). Incidence of plaques outcome literature. Am J Psychiatry 1994; 151:14091416
2. Oken RJ; Warner R; Papezova H; Dean CE: Schizophrenia: 100-
plus tangles tended to be greater in the more recent year retrospective; reply of JD Hegarty (letters). Am J Psychiatry
autopsies (N=26 [42.6%] of 61 versus N=9 [28.1%] of 1995; 152:16921695
32) (2=3.78, df=1, p=0.05). However, the force of this 3. Oken RJ, McGeer PL: Schizophrenia, Alzheimers disease, and
finding is limited by lack of comparisons, by a rather anti-inflammatory agents. Schizophr Bull 1996; 22:14
4. Wisniewski HM, Constantinidis J, Wegiel J, Bobinski M, Tar-
high incidence of Alzheimers disease-like changes nawski M: Neurofibrillary pathology in brains of elderly schizo-
(34%) in the mainly elderly patients with schizophre- phrenics treated with neuroleptics. Alzheimer Dis Assoc Disord
nia, and by diagnostic uncertainties in early samples. 1994; 8:211227
5. Shapiro RM: Regional neuropathology in schizophrenia: where
Omitting the report of Wisniewski et al. (4) from analy- are we? where are we going? Schizophr Bull 1993; 10:187239
sis because it included subjects autopsied in the pre- 6. Corsellis JAN: Mental Illness and the Aging Brain. London, Ox-
neuroleptic era suggested some lowering of the risk of ford University Press, 1962, pp 6972
Alzheimers disease-like changes over the 34 years sam- 7. Jellinger K: Neuromorphological background of pathochemical
studies in major psychoses, in Pathochemical Markers in Major
pled (19621996) (rs=0.62, df=7, p=0.08). Psychoses. Edited by Beckmann H, Riederer P. Heidelberg,
Springer, 1985, pp 123
8. Buhl L, Bojsen-Mller M: Frequency of postmortem Alzheimers
DISCUSSION disease in psychiatric patients. Dan Med Bull 1988; 35:288290
9. Soustek Z: Ultrastructure of cortical synapses in schizophrenics
brains. Zentralbl Allg Pathol 1989; 135:2532
We analyzed studies of postmortem brain tissue that 10. Bruton SJ, Crow TJ, Frith CD, Johnstone EC, Owens DG,
applied similar intrastudy methods to the comparison Roberts GW: Schizophrenia and the brain: prospective clini-
of subjects with schizophrenia, who presumably were coneuropathological study. Psychol Med 1990; 20:285304
11. Prohovnik I, Swork AJ, Kaufman MA, Willson N: Alzheimer-
routinely exposed to long-term neuroleptic treatment, type neuropathology in elderly schizophrenic patients. Schizophr
normal comparison subjects, and patients with psychi- Bull 1993; 19:805816
atric conditions unlikely to involve long-term neurolep- 12. Purohit DP, Davidson M, Perl DP, Powchik P, Haroutunian VH,
Bierer LM, McCrystal J, Losonczy M, Davis KL: Severe cognitive
tic exposure. The hypotheses 1) that Alzheimers disease- impairment in elderly schizophrenic patients: clinicopathological
like changes are now more prevalent in schizophrenia study. Biol Psychiatry 1993; 33:255260
than in other idiopathic psychiatric disorders or in nor- 13. Arnold SE, Franz BR, Trojanowski JQ: Elderly schizophrenia pa-
mal comparison subjects and 2) that Alzheimers dis- tients have infrequent neurodegenerative lesions. Neurobiol Ag-
ing 1994; 15:299303
ease-like neuropathological changes have increased 14. Benes FM, Bird ED: Analysis of the arrangement of neurons in
since the 1950s were not supported. It would be of in- the cingulate cortex of schizophrenic patients. Arch Gen Psychia-
terest to test the contrary hypothesis that coarse brain try 1987; 44:608616
15. Heaton RK, Drexler M: Clinical neuropsychological findings in
disease in individuals diagnosed with schizophrenia schizophrenia and aging, in Schizophrenia and Aging: Schizo-
may be declining with improving clinical and neuro- phrenia, Paranoia, and Schizophreniform Disorders in Later
pathological diagnostic methods. Life. Edited by Miller NE, Cohen GD. New York, Guilford
Neuropsychological data, although limited and in- Press, 1987, pp 145161
16. Kirkpatrick B, Golden RN, Fletcher RH: Is there a dementia of
consistent, also do not support increasing dementia-like schizophrenia? Psychiatr Med 1986; 4:253263
dysfunction in patients with schizophrenia during the 17. Cassens G, Inglis AK, Appelbaum PS, Gutheil TG: Neuroleptics:
neuroleptic era. Longitudinal IQ assessments indicate effects on neuropsychological function in chronic schizophrenic
improvements and decrements in individual patients patients. Schizophr Bull 1990; 16:477499
18. Buchanan RW, Holstein C, Breier A: The comparative efficacy
over time, largely in association with corresponding and long-term effect of clozapine treatment on neuropsychologi-
levels of psychotic disorganization and other functional cal test performance. Biol Psychiatry 1994; 36:717725

Am J Psychiatry 154:6, June 1997 863

Вам также может понравиться