11.
12.
13.
14.
vascular disease in University and Community Hospitals. Stroke 3:
747-758, 1972
Truscott BL, Kretschmann CM, Toole JF, Pajak TF: Early reha­
bilitative care in community hospitals: Effect on quality of survi­
vorship following a stroke. Stroke 5: 623-629, 1974
Garraway WM, Wiebers DO: A critique of the design and conduct
of therapeutic trials in acute ischaemic cerebrovascular disease. In
preparation
Kennedy FB, Pozen TJ, Gabelman EH, Tuthill JE, Zaentz SD:
Stroke intensive care — An appraisal. Am Heart J 80: 188-196,
1970
Drake WE, Hamilton MJ, Carlsson M, Blumenkrantz J: Acute
stroke management and patient outcome: The value of neurovascu­
lar care units. Stroke 4: 933-945, 1973
15. Garraway WM, Akhtar AJ, Prescott RJ, Hockey L: Management
of acute stroke in the elderly: Preliminary results of a controlled
trial. Br Med J 280: 1040-1043, 1980
16. Garraway WM, Whisnant JP, Drury Ivo: The continuing decline in
the incidence of stroke. Mayo Clinic Proceedings, in press
17. Anderson GL, Whisnant JP: A comparison of trends in mortality
from stroke in the United States and Rochester, Minnesota. Stroke
13: 804-809, 1982
18. Drury Ivo, Whisnant JP, Garraway WM: Further observations on
intracranial hemorrhage in Rochester, Minnesota with CT scan
confirmation. In preparation
19. Sundt T, Whisnant JP: Subarachnoid hemorrhage from intracranial
aneurysms: Surgical management and natural history of disease. N
Engl J Med 299: 116-122, 1978
Downloaded from http://stroke.ahajournals.org/ by guest on June 14, 2017
Hemiplegics after a First Stroke: Late Survival and
Risk Factors
P A B L O SOLZI, M . D . , H A I M RING, M . D . , THEODORE NAJENSON, M . D . , AND Y A I R L U Z , I . E .
SUMMARY
Scanning 3000 cases admitted for rehabilitation after cerebrovascular accident over a 20
year period produced a sample of 1369 subjects, without age restrictions, admitted within six months of a
first stroke of thrombotic etiology.
In this sample, survival rates showed no significant difference between men and women. Age at onset,
however, clearly influenced survival chances; the expected mean survival was 6 years at 40 and 2 at age 80;
average loss of life was 14 years for the whole sample, meaning a vital prognosis two to three times worse
than that of the general population.
At least 86% of the sample presented one or more of five etiological antecedents to stroke: hypertensive
heart disease, peripheral vascular disease, diabetes mellitus, myocardial infarction and atrial fibrillation.
In 87% of those, HHD and/or PVD were present. Presence of hypertension significantly lowered life
expectancy and so did PVD; their influence is felt from the earliest stages. In contrast, diabetes mellitus, the
next most common factor, has a late influence, starting about the fifth year after stroke. MI and AF were
present in relatively fewer patients, but they contributed towards a considerable decrease in life expectancy,
evident from the first stages, the more drastic reduction being observed in the AF group.
Stroke Vol 14, No 5, 1983
STROKE is recognized as both a killer disease and a
major cause of disability. Consequently, in recent
years, more research has been directed towards the
study of its etiology, incidence, effect on mortality
rates and, finally, of means which might help to pre­
vent its occurrence. At the same time, great progress
has been recorded in the rehabilitation of stroke-dis­
abled patients and in the improvement of their chances
of both surviving and living a satisfactory life after­
wards.
This work studies in retrospect data on patients after
cerebrovascular accidents causing hemiplegia who
were hospitalized at the Loewenstein Rehabilitation
Center (Ra'anana, Israel) over a period of twenty
years. Its purpose is twofold: to assess the survival
From the Loewenstein Rehabilitation Center, Ra'anana, and TelAviv University Sackler School of Medicine, Ramat Aviv, Israel. (Dr.
Najenson, Professor of Rehabilitation, Dr. Solzi, Senior Physician and
Head, Dept. Hemiplegics, Dr. Ring), and the Occupational Health and
Rehabilitation Institute at the Loewenstein Hospital, Ra'anana, Israel
(Yair Luz, Senior Research Engineer and Statistician).
Address correspondence to: Dr. P. Solzi, Loewenstein Hospital,
P.O. Box 3, Ra'anana 43100. Israel.
Received April 30, 1982; second revision accepted March 3, 1983.
rates of these patients after their admittance for reha­
bilitation, and to identify risk factors which, besides
provoking C.V.A. onset, might in the long run affect
these patients' mortality and survival rates.
Materials and Methods
About 3,000 records built up in the years 1958-1977
were screened according to the following criteria:
stroke was of a clearly vascular origin, the patient had
had a single stroke prior to admittance, and the patient
had been admitted for rehabilitation within six months
from onset of C.V.A.
This selection left 1700 well-documented cases an­
swering the criteria and constituting the sample for this
investigation. Each file was attentively studied. The
first item to be checked was the patient's status, using
the State of Israel population registry; death dates of
those deceased were noted down. The following item
was etiology. Cases whose etiology could not be incontestably established were left out and the remaining
cases were classified into five groups, in whom stroke
had been caused by:
1. Congenital malformation of cerebral arteries (24
cases).
TABLE 1
Break Down by Etiology
and Sex (Average
Age
J^i
patients
Cause of accident
Malformation of brain arteries
Valvular heart disease
Hemorrhage
Embolism
Men
N
%
ABS
Given)
A g e (years)
ABS
%
Mean
Women
A g e (years)
N
SD
ABS
%
Mean
SD
15
24
100
11
46
37
17
13
54
45
109
100
28
26
50
11
81
74
53
14
72
100
41
57
55
12
31
43
58
12
126
100
77
61
61
12
49
39
58
14
Thrombosis
1369
100
828
61
64
10
541
39
66
10
Total
1700
100
985
58
63
11
715
42
63
11
Downloaded from http://stroke.ahajournals.org/ by guest on June 14, 2017
2. Valvular heart disease (generally causing embo­
lism, though not altogether precluding thrombo­
sis) (109 cases).
3. Clear-cut cases of brain hemorrhage (72 cases).
4. Clear-cut cases of embolism not included in
Group 2 (126 cases).
5. Cerebral arterial thrombosis (1369 cases). After
the previous groups had been denned, the re­
maining patients may be assumed to constitute a
homogeneous group in the sense that the stroke
mechanism is uncontestably of the thrombotic
type. Diagnosis is limited, of course, to the
methods and techniques available during the 20year period of data-gathering, and cannot now be
refined in the light of modern advances.
Tables 1 and 2 present general characteristics of the
complete population of 1700 cases. In table 1, etiologi­
cal groups are broken down by sex, in table 2, by the
incidence of major pathological findings preceding the
first cerebrovascular accident. Note that for thrombo­
sis the prevalence of diabetes is double that for other
groups. Myocardial infarct is most frequent in embo­
lism-induced strokes and hypertensive heart disease is
the major finding prevalent in cases of hemorrhage.
Virtually all cases of arterial malformation were char­
acterized by retinopathy, also highly prevalent in all
other etiological types.
Statistical analyses of survival rates were performed
by means of programs available at the Computer Cen­
ter of Tel-Aviv University: BMDP1L and SPSS
'' SURVIVAL.'' The results of these methods are quite
similar; both programs are limited to one-way analy­
sis, i.e. can handle one factor at a time, for any number
of patient groups. In the course of preparing this paper
multivariate survival analysis (BMDP2L) became
TABLE 2
Prevalence
of Major Pathological
Findings
Total no
of patients
Cause of accident
Malformation of brain arteries
Valvular heart disease
Hemorrhage
Results
In figure 1 we present the life expectancy rates ob­
tained for the five etiological categories described
above, with a total of 1,700 patients. The difference in
survival between those groups is highly significant.
This, together with those shown in tables 1 and 2, adds
up to a considerable list of group-distinguishing char­
acteristics setting the thrombotic group apart from all
the other etiological categories: distribution of sex and
mean age, distribution of CVA-concomitant diseases,
Preceding
First
Accident
Myocardial
infarction
Diabetes
Arterial
hypertension
ABS
%
ABS
%
ABS
24
100
2
8
1
4
6
10
109
100
15
14
11
10
22
20
72
100
11
15
18
25
51
71
38
ABS
%
available and was used to complete the findings from
the one-way techniques.
In defining risk factors on the strength of the pa­
tients' medical histories up to the first stroke, we ap­
plied the following criteria:
1. Hypertensive heart disease (HHD) was diag­
nosed if a past history of elevated blood pressure
was recorded (at least three findings of diastolic
B.P. of over 100 mm Hg); or in the presence of
either ECG evidence of left ventricle hyper­
trophy or hypertensive retinopathy. The final
classification was based on the consensus of two
independent medical researchers.
2. Diabetes mellitus (DM), blood hyperglycemia
and glucosuria were diagnosed on the evidence
of at least three samples showing sugar level.
3. Peripheral vascular disease (PVD) was diag­
nosed on the strength of clinical data and, where
performed, of oscillometric findings.
4. Myocardial infarction (MI) was diagnosed only
where ECG showed characteristic changes.
5. Atrial fibrillation (AF) was also diagnosed on the
base of ECG changes; in most patients it ap­
peared in chronic form.
%
126
100
23
18
43
34
48
Thrombosis
1369
100
429
31
298
22
798
58
Total
1700
100
480
28
371
22
925
54
Embolism
S U R V I V A L - EFFECT
SUAVIVAL- EFFECT OF STROKE PATHOGENESIS
CHI-SOUARE = 9 J
OF MYPERT.
C H I - S Q U A R E • 9.7
HEART
OIS<HHD)
P-O.OQ2
P^OOS
" — . . . . N O HHD
HHD
(YEARS)
FIGURE 3 .
Survival
effect of hypertensive
heart
disease.
1- MAIN MTIRY MALFOMMTMM 2-HCMONmMCC 3-tMLVUUR HEMT M U M 4-EMMKISM S - T I M W D M
F I G U R E 1.
Survival
effect of stroke
pathogenesis.
Downloaded from http://stroke.ahajournals.org/ by guest on June 14, 2017
distribution of life expectancy rates, and proportion in
the total number of patients accepted for rehabilitation
(the thrombotics constitute 80% of the total).
We feel therefore justified in concentrating now on
this category, which is large and homogeneous in com­
position. Sixty-one percent are male and 39% female;
both constitute the oldest strata in all the etiological
types (about 65 ± 10 years of age). The incidence of
diabetes prior to the advent of stroke is by far the
highest (31%), and that of arterial hypertension, while
less than that found among the embolism patients, is
still very high (58%). Myocardial infarction occurred
in 22% of all thrombotics.
In both one-way and multivariate analysis, we found
that patient sex does not affect survival. On the other
hand, age at time of stroke is a significant factor, as
seen from figure 2. Its relationship with other factors
will be examined further on.
The risk factor most frequently encountered was
hypertensive heart disease (HHD). Figure 3 compares
the survival times of those who had and those who had
not HHD. The hypertensive patients' survival expec-
SURVIVAL EFFECT OF AGE AT TIME OF STROKE
(FROM
MULTIVARIATE
ANALYSIS: GLOBAL CHI SQUARE = 17.8)
P<0 01
tancy is significantly lower than that of the non-hyper­
tensives.
Peripheral vascular disease (PVD) also significantly
influences survival (fig. 4). Diabetes, the next most
common factor, is doubly interesting: it is very fre­
quent among thrombotic CVA patients, and its influ­
ence on survival is felt at a later stage, roughly after the
fifth year. It may be appreciated from figure 5 that
during the earlier stage there is very little difference
between diabetics and other "thrombotics"; at later
stages the differences becomes highly significant.
Both myocardial infarction and atrial fibrillation
(fig. 6 and 7 respectively) are represented by relatively
small groups; while this certainly points to their rela­
tively low frequency as factors preceding stroke, their
presence nevertheless considerably affects fatality
after stroke.
We can appreciate the influence of each risk factor
in table 3: patients with atrial fibrillation had the most
drastic reduction in life expectancy; those with diabe­
tes had the longest average survival period.
Using linear approximations, we derived regression
curves enabling us to compute mean survival chances
(fig. 8) and to predict the average loss of life expectan­
cy relative to the expectancy of the general population
(fig. 9) as functions of age at time of stroke. In figure 8
we see that the mean survival to be expected at age 40
is just under 6 years after stroke, the curve gradually
dropping to just 1 year after 90. The average loss in life
S U R V I V A L - EFFECT
OF P E R I P M . V A S C - D I S . I P V D)
C H I - S Q U A R E = 7.7
P=O005
UP TO 54
55-64
65-74
OVER 74
4
6
SURVIVAL
F I G U R E 1.
Survival
i
8
TIME
effect of age at time of
—
r
J R V I V A L
1
8 ^ " T ^ " ™ I T
T I M E
stroke.
FIGURE 4.
Survival
effect of peripheral
(YEARS)
vascular
disease.
SURVIVAL-EFFECT
1.0
OF
CHI-SQUARE « O.S
DIABETES
MELLITUS
S U R V I V A L - EFFECT
(DM)
OF A T R I A L
C H I - SQUARE = 7.2
F>)OOS
FIBRILLATION
P = 0.007
>
8
I'
z 7
o
!•*
o
J .5-
S I N U S RHYTHM
> .4
V
3 ->
10
11 12 13 14 15 16
T I M E
17 18 ]»
(YEARS)
{ V E A R I I
FIGURE 6 .
FIGURE 5 .
Survival
effect of diabetes
Downloaded from http://stroke.ahajournals.org/ by guest on June 14, 2017
MYOCARDIAL
C H I - S Q U A R E = 17.9
FIGURE 7 .
Survival
INFARCTION
Survival
effect of myocardial
infarction.
mellitus.
years as related to age at onset was similarly computed
and is presented in figure 9.
A general exposition of the relative and joint preva­
lence of the five risk factors discussed is in order. Out
of 1369 subjects, in 1179 cases (86% of the sample),
the presence of at least one factor was ascertained. For
the remainder, the absence of the five risk factors may
be safely assumed. Out of the 1 1 7 9 , 4 9 6 presented one
factor only, while 683 disclosed various combinations
of two to four factors. None had all the five, and in
only 8 cases (out of 59) did AF appear alone (table 4).
If we take the ratio of appearance of a factor alone
vs. all the cases where it was present, we find 35% for
HHD, round 19% for PVD and diabetes, and round 13
for MI and AF.
Further food for thought can be found in the compo­
sition of the multi-factor combinations (table 5):
1. Out of 26 cases with four-factor combinations,
atrial fibrillation was absent in 21 and PVD in
only 2. All contained HHD.
2. In 188 cases three-factor combinations were
present; only 20 did not include HHD and 33
lacked PVD.
3. Among the 469 patients showing two-factor
combinations, about a third had no HHD and
almost a half lacked PVD.
4. Out of 496 appearance of single factors, 56%
were of HHD and 20% of PVD.
We tested whether age differentials might not under­
lie the significance found for the various risk factors.
SURVIVAL- EFFECT OF
W / / / / / / / / / / / / / / / / / /
ATRIAL FIBRILL
(Mil
For each factor, the age distribution of subjects with
and without that factor was checked (table 6). At the
bottom of the table, we may see that the subjects dead
and censored also had similar distributions. Not sur­
prisingly, those free of the various diseases were, on
the average, slightly younger. Though a synergistic
interaction between age and the other risk factors (as
between those factors themselves) is not to be dis­
counted, it appears from the evidence that age differen­
tials cannot, by themselves, explain the significant
effects detected for the risk factors. Again, this is con­
firmed by the results of age stratification tests shown in
figure 2.
Multivariate analysis allows us to formulate a joint
model and to test the contribution of each factor or
combination of factors to the complete model. We find
that all the risk factors examined are of importance,
and their relative weight is assessed by the program, as
shown in table 7. The most important is age, with
HHD, PVD and diabetes occupying a slightly lower
position. MI and atrial fibrillation come last, because
of their lower frequency and therefore reduced overall
impact.
However, it is also important to keep in mind the
influence of each factor in those cases where it is
present: those factors that appear the less important for
the whole sample involve the strongest risk for the
limited sub-groups presenting them. This should not
be overlooked or overshaded by the results of multivar­
iate analysis, since it greatly affects the prognosis of
such patients.
Discussion
P = 0.001
effect of atrial
fibrillation.
This investigation tested the presence of six main
risk factors at the onset of cerebro-vascular accidents
(CVA) and their influence on the survival of the popu­
lation of patients accepted for rehabilitation at the
Loewenstein Rehabilitation Center over a period of
twenty years. From over 3,000 admittances, 1,700
were selected for our database and carefully document­
ed. A further refinement was the elimination of cases
of non-thrombotic etiology, leaving 1,369 clinically
defined thrombotics. This sample size should ensure a
fair representation of the possible modalities of throm­
bosis; this is regarded as a clinically homogeneous
category in need of rehabilitation.
TABLE 3 50-percent
Survival Periods for Patients
Given Risk Factors (in Months from Stroke)
Presenting
AVERAGE LIFE EXPECTANCY LOSS ««. AGE AT TIME OF STROKE
30
Survival period
25
Risk
factor
present
Not
present
Difference
|
Myocardial infarction
51
72
21
t<
Atrial fibrillation
38
69
31
o
Peripheral vascular dis.
56
74
18
Hypertensive heart dis.
50
69
19
Diabetes mellitus
62
72
10
Risk factor
Downloaded from http://stroke.ahajournals.org/ by guest on June 14, 2017
The study's main significance lies in the provision
of guidelines — from both the medical and care provi­
sion standpoints — to the life expectancy of patients
admitted for rehabilitation. Evidently, it applies to
subjects who survived the acute phase, with impair­
ments that cannot be attended to in ambulatory clinics,
yet show a reasonable capability for benefitting from
the rehabilitation process. The concept of rehabilita­
tion through the efforts of a multidisciplinary team, as
practised at the Loewenstein Hospital, is becoming
universally accepted; the results obtained with such a
large sample should prove of interest to other rehabili­
tation hospitals or wards.
The overall population from which our patients were
drawn covers about 75% of the Israeli population and
is not limited to any particular age group, social stra­
tum, region, trade, ethnic group or hospital area of
service. We trust that this and the other characteristics
described will make it an important forerunner for fur­
ther investigations into both general and specific popu­
lations of rehabilitation patients.
A further benefit derived from the uniform system of
registration throughout the Histadrut Sick Fund with
which our Hospital is affiliated is the completeness of
data on previous heart conditions. This probably ex­
plains our finding that only 14% of the patients had no
previous history of cardiovascular risk factors, while
Aho found 26% among aged patients and 36% among
younger ones.
Of the various problems that could be profitably
investigated, we have chosen two aspects: co-morbid1
20-
15
10
5
AGE AT TIME
FIGURE 9.
Average
OF STROKE
life expectancy
loss vs. age time of
stroke.
ity (the presence of various risk factors) and survival
expectancy. We do compare our findings with some of
the results published, without forgetting that, given the
different characteristics of the populations investigat­
ed, any comparisons must be made with the utmost
circumspection. Among works published there were
community studies; there were hospital studies; agedistributions were not always comparable or even re­
ported; some samples did not include patients older
than 65 years, which naturally enhances the life expec­
tancy. None specifically considered the population
surviving the first month (with its high mortality) and
referred for rehabilitation.
The Presence of Different Risk Factors
Practically all the authors concur on the paramount
importance of hypertension and HHD as risk factors
for stroke: Whisnant, Matsumoto, Eisenberg, Kagan, Rabkin, Okada, Sahs and their co-workers.
Beevers found significant differences in stroke recur­
rence after anti-hypertensive therapy (from 55% to
16%). Kannell" in one of the Framingham cohort stud­
ies found the risk of stroke three-fold in patients with
congestive heart failure or ECG signs of LVH; and
sixfold in the presence of acute or chronic atrial fibril­
lation. Diabetes mellitus has also been mentioned as a
risk factor. Olivares placed it as the third in impor­
tance; the Framingham t e a m
found a high risk as­
sociated with even a mild impairment of glucose toler­
ance. In another of their studies, Wolff' points out the
importance of AF in stroke etiology; we found that its
presence dramatically reduces life expectancy.
In our sample HHD appeared as the most frequent
2
5
6,7
3
8
4
9
10
12
13,14
5
AVERAGE YEARS OF SURVIVAL va. AGE AT TIME OF STROKE
TABLE 4 Frequency
or Coupled
of Appearance
of Five Risk Factors,
Singly
Number of cases
HHD
Single factor
(% of all appearances)
AGE AT
FIGURE 8.
Average
TIME
OF
years of survival
vs. age at time of
stroke.
AF
Total
37
8
(13) (13)
496
Coupled factors
522
420
352
261
51
683
Total appearances
798
518
429
298
59
1179
All thrombotics
77
(19)
MI
98
(19)
No risk factor
STROKE
P V D Diab.
276
(35)
190
1369
TABLE 5 Frequency of Single and Multi-factor
Five Risk Factors, Showing Dominant
Factors
H H D present
Factor
combinations
Cases
N
Combinations
P V D present
%
N
%
Five-factor
0
—
—
—
—
Four-factor
26
26
100
24
92
Three-factor
188
168
89
155
82
Two-factor
469
328
70
241
51
Single-factor
496
276
56
98
20
No-factor
190
—
—
—
1369
798
—
—
Total
of
518
TABLE 7 Relative Weight of Risk Factors (Evaluated
iate Survival
Analysis)
Multivar­
Factor
Age
HHD
PVD
Diab
MI
AF
Coefficient
1.04
.945
.937
.890
.754
.759
Among others, we should mention the findings of
Sahs, Rabkin and Kannel" confirming the influence
of blood pressure, systolic blood pressure and ECG
changes on survival chances respectively. Regarding
survival times, findings published range very widely:
this underscores again the importance of a precise sam­
ple definition, as we attempted to achieve. Populations
studied by Marquardsen, Eisenberg and Yokoyama
survived 3 - 4 years (median). Abu-Zeid noted in
Manitoba that 50% of the men and 57% of the women
died between 3 and 4 years after cerebral infarction.
Robinson found a median survival of 4.1 years. Find­
ings higher than our own were contributed by Adams
and Merrett from Belfast (6.2 for men, 6.6 for wom­
en). Fugel-Meyer from Frederiksborg (Denmark)
puts median survival at over 6 years for the overall
cohort; if we disregard early-mortality cases, median
survival is even higher.
9
7
16
Downloaded from http://stroke.ahajournals.org/ by guest on June 14, 2017
factor, both in isolation and coupled with others; in
56% of all single-factor cases; in 70% of all two-factor
Combinations; in 89% of three-factor cases and in all
four-factor combinations. HHD appeared alone in onethird of all cases where it was present; PVD and diabe­
tes in 20%; myocardial infarction and atrial fibrillation
in only 13% of all instances. This might suggest a
complex mechanism whereby the most virulent factors
require, as a rule, the concurrence of other impair­
ments to cause stroke and to shorten survival chances.
by
4
17
18
19
20
21
Survival Expectancy
Evidently, the presence of various risk factors be­
fore stroke and their influence on ultimate survival are
different problems. The latter aspect has been less
highlighted by researchers.
In our sample of thrombotics, the median survival
time was 5.2 years for men and 4.9 years for women.
This represents an average loss of 14 years relative to
the life expectancy of general populations (ranging
from a loss of 5 years for those aged 80 to a loss of 30
years for those of age 40). Among risk factors, the
highest curtailment of life is due to the presence of
atrial fibrillation (31 months compared to other pa­
tients); myocardial infarction, hypertensive heart dis­
ease and peripheral vascular disease cause a median
loss ranging from 21 down to 18 months. The impact
of diabetes mellitus is felt mostly after the fifth year
from onset; its presence reduces life expectancy by 10
months.
TABLE 6 Age Distributions
of Subjects With or Without
Risk Factors, Dead and Censored (in percents)
Age group
Up to 5 4 5 5 - 6 4
Hypertensive
heart disease
+
Peripheral
vascular dis.
+
Diabetes
mellitus
+
Myocardial
infarction
+
Various
65-74
75 +
Mean
age
13.3
15.8
34.5
29.8
38.2
34.4
14.0
20.1
64.8
64.9
11.4
16.1
28.8
34.8
40.7
34.1
19.1
15.0
66.0
64.1
10.7
16.0
32.4
32.6
41.0
34.6
15.9
16.9
65.7
64.5
10.1
15.5
31.2
32.9
42.3
35.0
16.4
16.6
66.0
64.5
6.8
14.7
23.7
32.9
42.4
36.3
27.1
16.1
68.8
64.7
Dead
10.4
31.9
37.4
20.3
—
Censored
18.4
33.2
35.7
12.7
—
Atrial
fibrillation
+
-
Conclusion
Out of the sample of 1369 patients referred for reha­
bilitation within six months of a first stroke of throm­
botic etiology, at least 86% presented one or more of
the following antecedents: hypertensive heart disease,
peripheral vascular disease, diabetes mellitus, myocar­
dial infarction and atrial fibrillation (the first two were
present in 87% of these cases). The average loss of life
expectancy postulated on the predicted mean survival
time as function of age at time of stroke was 14 years.
Regarding the effect of different traits on survival,
four main points emerge: (a) there is no clearcut differ­
ence between sexes; (b) all five risk factors significant­
ly influence survival; (c) AF appears with the least
frequency (usually with other factors) and when it
does, markedly curtails life expectancy; (d) the strong­
est single influence is that of age, but it was established
that the other five factors maintain the same relative
effect over different age strata. We should like to see
comparable results from other sources.
Acknowledgments
The authors gratefully acknowledge the contributions of Dr. Stella
(Dolly) Appelman-Yoffe in data collection, screening and coding, and
of Jedida Brachfeld, B.A., who prepared and edited the manuscript.
Useful contributions to the statistical side of this study were made by the
late Professor M.B. Tartakovsky.
References
1. Aho K, Harmsen P, Hatano S et al: Cerebrovascular disease in the
community: results of a WHO collaborative study. Bull World
Health Organiz 58: 113-130, 1980
2. Whisnant JP, Fitzgibbons JP, Kurland LT et al: Natural history of
stroke in Rochester, MN, 1945 through 1954. Stroke 2: 11-22,
1971
3. Matsumoto N , Whisnant JP, Kurland LT et al: Natural history of
stroke in Rochester, MN, 1955 through 1969: an extension of a
previous study, 1945 through 1954. Stroke 4: 2 0 - 2 9 , 1973
Downloaded from http://stroke.ahajournals.org/ by guest on June 14, 2017
4. Eisenberg H, Morrison JT, Sullivan P et al: Cerebrovascular acci­
dents: Incidence and survival rates in a defined population, Middle­
sex County, Conn. J Am Med Assoc 189: 883-889, 1964
5. Kagan A , Popper JS, Rhoads GG: Factors related to stroke inci­
dence in Hawaii Japanese men. The Honolulu heart study. Stroke
11: 14-21, 1980
6. Rabkin SW, Mathewson FAL, Tate RB: Long term changes in
blood pressure and risk of cerebrovascular disease. Stroke 9: 319—
327, 1978
7. Rabkin SW, Mathewson FAL, Tate RB: The relation of blood
pressure to stroke prognosis. Ann Internal Med 89 (1): 15-20,
1978
8. Okada H, Horibe H, Ohno Y et al: A prospective study of cerebro­
vascular disease in Japanese rural communities, Akabane and
Asahi. Part I: Evaluation of risk factors in occurrence of cerebral
hemorrhage and thrombosis. Stroke 7: 599-607, 1976
9. Sahs AL, Hartman EC, Aronson SM (Editors): Guidelines for
Stroke Care. Washington DC, US Dept Health, Education & Wel­
fare, 1976
10. Beevers DG, Fairman MJ, Hamilton M et al: Antihypertensive
treatment and the course of established cerebral vascular disease.
Lancet 1: 1407, 1973
11. Kannel W B , Wolf P, Dawber TR: Hypertension and cardiac im­
pairments increase stroke risk. Geriatrics 33: 7 1 - 8 3 , 1978
12. Olivares L, Castaneda E, Grife A, Alter M: Risk factors in stroke:
A clinical study in Mexican patients. Stroke 4: 773-781, 1973
13. Kannel WB, Dawber TR, Cohen ME et al: Vascular disease of the
brain — epidemiologic aspects: The Framingham study. Am J
Public Health 55: 1355-1365, 1965
14. Shurtleff D: Some characteristics related to the incidence of cere­
brovascular disease and death: Framingham study 18-year followup. An epidemiological investigation of cardiovascular disease.
Section 30, 1974. US Government Printing Office, Washington
DC
15. Wolf PA, Dawber TR, Emerson T, Kannel WB: Epidemiologic
assessment of chronic atrial fibrillation and risk of stroke: The
Framingham Study. Neurology 28: 973-977, 1978
16. Marquardsen J: The natural history of acute cerebrovascular dis­
ease: A retrospective study of 769 patients. Acta Neurol Scand 45
(suppl 38), 1969
17. Yokoyama I, Matsumata Y, Matsushima H: Follow-up study of
post-stroke patients after medical rehabilitation. Jap Med J Phys
Thermal Investig 38 (3^t): 4 6 - 5 2 , 1975
18. Abu-Zeid HAH, Choi NW, Hsu PH et al: Prognostic factors in the
survival of 1484 stroke cases observed for 30 to 48 months. I.
Diagnostic types and descriptive variables. Arch Neurol 35: 1 2 1 125, 1978
19. Robinson RW, Cohen WD et al: Life-table analysis of survival
after cerebral thrombosis — 10-year experience. JAMA 169:
1149-1152, 1959
20. Adams GF, Merrett JD: Prognosis and survival in the aftermath of
hemiplegia. Brit Med J 1 (5222); 309-314, 1961
21. Fugl-Meyer AR, Jaasko L, Norlin V: The post-stroke hemiplegic
patient. II. Incidence, mortality and vocational return in Gdteberg,
Sweden, with review of the literature. Scand J Rehab Med 7: 7 3 83, 1975
Factors Contributing to Stroke in Patients with
Atherosclerotic Disease of the Great Vessels:
The Role of Diabetes
JESSE WEINBERGER, M . D . , VICTORIA BISCARRA, R . N . , MICHAEL K . WEISBERG, B . S . ,
AND JULIUS H. JACOBSON, M . D .
SUMMARY
The incidence of carotid artery disease and cerebrovascular symptoms were determined in
102 consecutive patients with peripheral arterial disease. Symptoms were correlated with risk factors of
age, hypertension, smoking and diabetes and with the extent of disease at the carotid bifurcation. The
incidence of stroke with permanent neurological deficit was twice as high in diabetics as in non-diabetics
with equivalent atherosclerotic vascular disease (p < .05). In women, the incidence of stroke was three times
higher in diabetics (p < .02). The number of transient ischemic attacks was significantly higher in nondiabetics (p < .02). The total number of ischemic episodes in diabetics and non-diabetics was equivalent.
This indicates that diabetics are more prone to irreversible destruction of ischemic brain tissue regardless of
the nature of the circulatory disturbance.
Stroke, Vol 14, N o 5, 1983
THE ASSOCIATION OF ATHEROSCLEROTIC
DISEASE at the carotid artery bifurcation with both
transient ischemic attack and stroke is firmly estab­
1-3
l i s h e d . However, a large proportion of patients with
carotid disease remain asymptomatic. In addition, the
majority of patients with transient ischemic attacks do
4
not develop irreversible neurologic deficits.
5
fore, it is imperative to determine what factors are
responsible for permanent damage to neurons when the
brain is subjected to disruption of circulatory perfu­
sion.
In order to delineate the role of disease of the great
vessels in stroke, w e examined 102 consecutive pa­
There-
tients with atherosclerotic peripheral arterial disease
(PAD) who were referred to the vascular laboratory for
From the Departments of Neurology and Vascular Surgery, The
Mount Sinai School of Medicine of the City University of New York.
Address correspondence to Jesse Weinberger, M.D., Department of
Neurology, The Mount Sinai School of Medicine, 1 Gustave Levy
Place, New York, New York 10029.
Received August 13, 1982; revision accepted April 5, 1983.
non-invasive carotid artery testing. While this group of
patients is not randomly selected, they all had a similar
propensity to atherosclerotic arterial disease of the
great vessels so that valid comparisons of other factors
can be made.
Hemiplegics after a first stroke: late survival and risk factors.
P Solzi, H Ring, T Najenson and Y Luz
Downloaded from http://stroke.ahajournals.org/ by guest on June 14, 2017
Stroke. 1983;14:703-709
doi: 10.1161/01.STR.14.5.703
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1983 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located
on the World Wide Web at:
http://stroke.ahajournals.org/content/14/5/703
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not
the Editorial Office. Once the online version of the published article for which permission is being
requested is located, click Request Permissions in the middle column of the Web page under Services.
Further information about this process is available in the Permissions and Rights Question and
Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Stroke is online at:
http://stroke.ahajournals.org//subscriptions/