From the Department of Medicine, Umeå University, Umeå,
Sweden.
Correspondence to Markku Peltonen, Department of Medicine, Umeå University, S-901 87 Umeå, Sweden. E-mail markku.peltonen{at}medicin.umu.se
Methods—Within the framework of the population-based WHO
Multinational Monitoring of Trends and Determinants in
Cardiovascular Disease (MONICA) Project,
all acute stroke events were recorded in the age group 25 to 74
years in northern Sweden during the period 1985 to 1994. All first-ever
stroke patients were followed for information on vital status (minimum
follow-up time was 1 year). Survival time was related to time period of
stroke onset, stroke diagnosis, and concomitant diseases.
Results—Survival times for a total of 6819 first-ever stroke
patients (4057 men and 2762 women) were analyzed. Age-adjusted
odds ratio for death within 1 year after stroke was 0.70 (95%
confidence interval [CI], 0.55 to 0.88) in the period 1993 to 1994 as
compared with the period 1985 to 1986 in men and 0.69 (95% CI, 0.53 to
0.90) in women. Corresponding odds ratios were 0.73 in men and 0.70 in
women among those who survived the first 28 days. Similar improvements
were seen for 3- and 5-year survival. Improvements in survival over
time were most marked among patients with ischemic stroke.
There was no improvement in survival over time among patients with the
most severe deficits at onset.
Conclusions—Gradually improved survival, both short and long
term, was observed during the 10-year study period. The improvements
are not explained by changes in known confounding prognostic factors.
In Sweden, stroke mortality declined in both men and women during the
period 1969 to 1993.14 Stroke incidence remained
relatively constant in the Northern Sweden MONICA study during 1985 to
1991, whereas short-term case fatality declined among patients with
first-ever stroke.15 In our study, we
analyzed time trends in long-term survival among stroke
patients between 1985 and 1994 in the population-based Northern Sweden
MONICA study. In particular, we wanted to study whether there were any
changes in long-term survival after stroke, independent of changes in
short-term survival (ie, 28-day case fatality) and after adjusting for
other prognostic factors.
Diagnostic Criteria
Each case was classified into 1 of the 3 categories: "definite
stroke," "unclassifiable," or "not stroke." The category
"unclassifiable" was mainly used in fatal cases in which acute
cerebrovascular disease was given as the cause of death on the death
certificate but the clinical information was too limited to classify
the event as a definite stroke. In the present paper, only cases
classified as definite stroke were included in the nonfatal events,
whereas fatal events also included cases coded as unclassifiable (fatal
cases were persons who died within 28 days of stroke onset). Only
first-ever strokes are considered in this study. Multiple strokes
occurring within 28 days of the onset of the first attack were
considered as 1 event.
The subtypes of acute stroke were based on the following examinations:
SAH, ICD code 430, bloodstained cerebrospinal fluid and an
aneurysm or an arteriovenous malformation found on angiography
or positive finding on CT scan or necropsy; ICH, ICD 431, positive
finding on CT scan or at autopsy; brain infarction, ICD 434, no signs
of hemorrhage on CT scan or at autopsy; unspecified stroke, ICD
436, not investigated by CT scan or autopsy. ICD-8 was used until the
end of 1986 and ICD-9 thereafter. In the analyses by subtype of
stroke, 3 categories were defined: SAH (ICD 430), ICH (ICD 431), and
possible IS (ICDs 434 and 436).
Information on the severity of the stroke and on other diseases such as
diabetes and history of atrial fibrillation was obtained from medical
records. The extent of motor deficits and presence of aphasia or
dysphasia were recorded by the attending physician when the patient
was first seen (in most of the events) or early during the
hospitalization.
Mortality Follow-up
Statistical Analyses
The proportion of patients surviving 28 days and 1 year after stroke
grouped by sex, age, and stroke subtype is shown in Table 2
In the Figure
In Table 3
In order to study whether the improved 1-year case fatality in
first-ever stroke patients was attributed only to an improved early
case fatality, we calculated survival times separately for those who
survived the first 28 days. As shown in Table 4
Secular trends in basal clinical characteristics among patients with
first-ever stroke are shown in Table 5
In Table 6
It should be noted that this study covers only ages up to 75
years. A decreasing case fatality over time has also been observed in
the elderly.23 24 Case fatality among patients
with acute stroke increases with age25 so that in
most aging societies the majority of stroke deaths occurs after the age
of 75 years.14 26 In Sweden, the decline in
age-specific stroke mortality has affected all age groups, also those
above 75 years.14 Projections up to the year
2003 based on an age-period-cohort model have shown that, despite the
demographic development, there will be no increase in the total number
of stroke deaths in Sweden among both men and women aged 25 to 89
years.14 In Sweden, the aging of the population
is now leveling off,27 but the situation is
different in other countries where there will be a more dramatic
increase in the number of elderly people in the coming years. Thus,
projections performed in Italy have shown a major increase in the
total number of stroke deaths over the next
decades.26
The strength of the population-based stroke register in the WHO MONICA
study is that uniform registration and diagnostic criteria
have been applied strictly over the study period. Case ascertainment in
the Northern Sweden MONICA study has been validated twice, showing that
the data are of high quality.17 18 Approximately
3% of the nonfatal cases in our study have been treated
out-of-hospital.17 This low proportion of
out-of-hospital stroke cases is in agreement with other
population-based stroke registers in
Sweden.28 29 30 A more frequent use of CT scan over
time could result in the identification of more milder strokes.
However, cases identified only by CT scan, without focal neurological
deficits, were not included in our study. Therefore, it is not likely
that the observed improvement in survival after stroke is because of
better case-finding over time.
Changes in survival after stroke could be affected by changes in the
severity of the disease. This could result from changes in risk factor
levels in the population or from an increase in public awareness of
stroke. Trends in risk factor levels for cardiovascular
disease in northern Sweden have been in part favorable, with a decline
in serum cholesterol levels in both men and women and a
modest decline in blood pressure levels among
women.31 However, body mass index has tended to
increase, especially among younger age groups.31
Smoking has shown diverging trends, declining in men but not changing
in women.31
Studies that support the hypothesis that stroke is becoming a less
severe disease have been published.12 15 32 33
However, in our study the proportion of patients found dead or
unconscious tended to have increased over time. There were no apparent
changes in case fatality during the first few days after stroke onset,
which is in accordance with the increased proportion of patients found
dead or unconscious at onset. Thus, in stroke patients with the most
severe strokes (ie, unconscious patients), there was no improvement in
survival. Therefore, the improvement reported here becomes even more
pronounced for patients with less severe strokes. Improved case
fatality over time has been reported previously to have occurred in
patients with less severe deficits, whereas no improvement occurred
among those with severe deficits.22
It has been hypothesized that the increase in hypertension detection
and treatment has been a major reason for declining stroke
mortality,34 35 affecting also case fatality
indirectly. However, in-depth analyses have shown that only a
small part of the stroke decline can be attributed to improved
hypertension control in the general
population.3 36 The introduction of specialized
stroke units, providing more accurate management of patients, has
reduced acute stroke mortality substantially at 3-month follow-up
according to a meta-analysis.37 In
Sweden, specialized stroke units and stroke teams have been widely
established during the last decade. As there have not been any other
major advances in medical or surgical treatment of acute stroke during
the study period, it is possible that the increasing establishment of
stroke units partly explain the improvement in survival after stroke,
also on the long-term. The fact that improved survival did not occur
until after the first few days is in accordance with this
contention.
Of the stroke subtypes studied, patients with IS had the most
marked decline in risk of death. This group was also the only group in
which changes reached statistical significance. In the 2 other groups,
SAH and ICH, numbers of cases per time period and sex were small,
thereby the power of the tests was low. However, there was a tendency
toward lower risk over time also in these 2 groups, especially among
women. An exception was short-term survival among men with ICH, where
an increased mortality was observed. These 2 groups also had higher
28-day case fatality as compared with patients with IS.
Improvements in long-term survival could be merely a reflection of the
improved short-term survival. However, when the analyses were
restricted to the group surviving the first 28 days, an increased
proportion of long-term survivors was also observed. Hence, the
changing survival after stroke is not restricted to the improvements in
acute phase, but it is likely that the improved secondary prevention
after stroke has contributed to better long-term survival. Improved
scientific documentation of the beneficial long-term effects of
antiplatelet agents in patients with IS38 and
of oral anticoagulants in those with atrial
fibrillation39 has had a major impact on clinical
practice. In Sweden today, nearly all surviving patients less than 75
years of age who have had an IS are discharged from the hospital with
either antiplatelet drugs (63%) or oral anticoagulants (31%)
(unpublished data from the Northern Sweden MONICA Center). If the
effects demonstrated in clinical trials are directly transferable to
routine health care, the widespread use of antithrombotic agents has
contributed substantially to the improved long-term survival. On the
other hand, less than 3% of all stroke patients are subjected to
carotid surgery (unpublished data from the Northern Sweden MONICA
Center), so this procedure contributes little to improved survival on
the population level.
There were no clear changes in the distribution of causes of death
among those who died within 1 year from the stroke onset. The
proportion of deaths attributed to heart infarction declined somewhat
among men. This could be seen as a marker for improved management of
other cardiovascular diseases in conjunction with
stroke. On the other hand, among women no such shift in cause of death
was seen.
Comparison of survival rates between different studies is complicated
by different registration procedures, differences in
diagnostic criteria, and differences in study design.
Compared with other populations in the WHO MONICA study, with uniform
case ascertainment and registration procedures, 28-day case fatality in
northern Sweden is among the lowest.40 Stroke
incidence in northern Sweden has been rather
stable,15 although a modest increase in numbers
was observed in this study. Similarly, in southern Sweden, incidence of
stroke did not vary much during 1986 to 1990,41
and an increasing incidence has been reported for
women.28 Therefore, the declining stroke
mortality observed in Sweden14 can largely be
attributed to the improved survival after stroke.
In summary, improved survival, both short- and long-term, was observed
in the Northern Sweden MONICA Study during the period 1985 to 1994.
These improvements cannot be explained by changes in known confounding
prognostic factors. The improved long-term survival was not just
because of improvements in short-term survival.
Received March 2, 1998;
revision received April 10, 1998;
accepted April 10, 1998.
© 1998 American Heart Association, Inc.
Original Contributions
Time Trends in Long-term Survival After Stroke
The Northern Sweden Multinational Monitoring of Trends and Determinants in Cardiovascular Disease (MONICA) Study, 1985–1994
Abstract
Top
Abstract
Introduction
Subjects and Methods
Results
Discussion
References
Background and Purpose—Stroke
mortality rates and case fatality of stroke have declined since the
beginning of the 1970s in Sweden, but the incidence of stroke has been
stable. The aim of this study was to analyze trends in
long-term survival after stroke.
Key Words: stroke outcome • survival • case fatality • time trend • Sweden
Introduction
Top
Abstract
Introduction
Subjects and Methods
Results
Discussion
References
Cerebrovascular
disease mortality rates have declined in the United States, western
Europe, Japan, and Australasia during recent
decades.1 2 3 4 5 6 7 Changes in mortality rates over time
could be attributed to changes in disease incidence or case fatality
rate. Changes in risk-factor levels in the population because of, for
example, improvements in hypertension detection and treatment, could
result in a decreased incidence and less severe strokes. In several
studies, changes in stroke incidence have been insufficient to explain
the declining stroke mortality.5 8 9 10 As a
corollary, case fatality in the acute phase of stroke has been reported
to have declined markedly.4 11 12 13 Improved
medical management, change in the severity of disease, or a better
case-ascertainment, resulting in more mild strokes being identified,
could explain improved case fatality. Long-term trends in late survival
after stroke have only occasionally been
reported.11 12 13
Subjects and Methods
Top
Abstract
Introduction
Subjects and Methods
Results
Discussion
References
A population-based stroke registry in the 2 northern-most
counties in Sweden was established in 1985 as a part of the WHO MONICA
Project.16 In the Northern Sweden MONICA
study, all acute cases of stroke in both men and women aged 25 to 74
years during 1985 to 1994 were registered according to the WHO MONICA
protocol.16 The study population was 310 348 in
1985 and 319 413 in 1994.
The methods of case-finding and data quality control are
described in detail elsewhere.15 17 18 In short,
medical records from all suspected acute stroke events and death
certificates were validated by the WHO
criteria,18 and cases found to be acute stroke
were included in the register. The WHO definition of a stroke excludes
transient ischemic attacks and stroke caused by a trauma or
malignancy.
All subjects were followed up for information on vital status in
1997 with the help of the Cause of Death Register in
Sweden,19 which at that time included all deaths
up to and including year 1995. Thus, every individual had a follow-up
time of at least 1 year; median follow-up time was 4.7 years.
Individuals not identified as "deaths" in the Cause of Death
Register were assumed to be alive at the end of year 1995, as the
register has a coverage over 99%.19 For
individuals identified as dead, information on cause of death was
extracted. For individuals with unknown stroke onset date, this was
assumed to be the 15th of the month of onset. As end point, death from
any cause was used.
Time periods of stroke onset were grouped in 2-year periods:
1985 to 1986, 1987 to 1988, 1989 to 1990, 1991 to 1992, and 1993 to
1994. For univariate analysis, survival times were
computed with Kaplan-Meier product-limit estimate, and hypothesis
test of no change in survival over time was made by log-rank
test.20 Multivariable analyses of the
proportion surviving for a given period after stroke onset were
performed by logistic regression. Analysis of long-term
survival among those who survived the first 28 days was also performed
with logistic regression for 1-, 3-, and 5-year survival. Age at onset,
time period, stroke diagnosis, and other diseases were considered as
predictor variables. Results are presented as OR with 95%
CI. Analyses were performed with the statistics package
Stata.21
Results
Top
Abstract
Introduction
Subjects and Methods
Results
Discussion
References
During the study period 1985 to 1994, a total of 9362 stroke
patients (5599 men and 3763 women) was registered in the Northern
Sweden MONICA study. Of these, 6819 patients (4057 men and 2762 women)
with personal identification numbers were identified as first-ever
strokes. In Table 1
, the numbers of
patients with first-ever strokes are shown by time period and stroke
subtype. A modest increase in number occurred in the latter part of the
study period, but distribution of stroke subtypes did not change over
time. Median age of the patients remained approximately constant over
the study period. Women had a somewhat higher median age at onset than
men (Table 1).
View this table:
[in a new window]
Table 1. Number of First-Ever Strokes With Known Survival
Time and Median Age at Onset by Subtype in the Northern Sweden MONICA
Study, 1985–1994; Men and Women, Ages 25 to 74 Years . Overall, 85% of the men and 82% of
the women survived the first 28 days. The proportion surviving 28 days
varied from 46% among 65- to 74-year-old men with SAH to 95% among
25- to 54-year-old men with IS. For IS, survival rates were similar for
men and women, whereas women with SAH and ICH had lower survival than
men.
View this table:
[in a new window]
Table 2. Proportion of Patients Surviving 28 Days and 1 Year
After First-Ever Stroke in the Northern Sweden MONICA Study,
1985–1994, Ages 25–74 Years; Life Table Estimates the crude survival over the study period
is shown. There was a statistically significant gradual improvement in
survival over time (log-rank test, P=0.0006). There were no
apparent secular trends in case fatality during the first few days
after stroke onset. Instead, the survival curves started to diverge
from 1 week after onset (see insert of Figure).
View larger version (21K):
[in a new window]
Figure 1. Secular trends in survival after first-ever stroke during
1985 to 1994 in the Northern Sweden MONICA study, ages 25 to 74
years. , age-adjusted ORs for death
for different time periods compared with the period 1985 to 1986 are
shown. There were no clear changes in 2-day and 7-day case fatalities
over time among patients with SAH and ICH, except a statistically
significant increase among men with ICH. Among patients with IS, 7-day
case fatality decreased over time, and the decrease was statistically
significant among men. For 28-day and 1-year case fatalities, ORs
decreased gradually over time for IS in both men and women. In all
strokes, the OR for 1-year case fatality for the period 1993 to 1994 as
compared with 1985 to 1986 was 0.70 (95% CI, 0.55 to 0.88) in men and
0.69 (95% CI, 0.53 to 0.90) in women.
View this table:
[in a new window]
Table 3. Trends in Case-Fatality During the First Year After
First-Ever Stroke in the Northern Sweden MONICA Study, 1985–1994, Ages
25–74 Years; Age-Adjusted ORs With Period 1985–1986 as Reference , there was a similar gradual
improvement in long-term survival for both men and women over time. In
all strokes, the OR for 1-year case fatality was 0.73 (95% CI, 0.51 to
1.04) in men and 0.70 (95% CI, 0.44 to 1.09) in women for 1993 to 1994
as compared with 1985 to 1986.
View this table:
[in a new window]
Table 4. Trends in Case-Fatality After First-Ever Stroke
Among Subjects Surviving 28 Days in the Northern Sweden MONICA Study,
Ages 25–74 Years; Age-Adjusted ORs With Period 1985–1986 as Reference .
In both women and men, there was a statistically significant increase
over time in the proportion with a history of atrial fibrillation. The
proportion of patients found dead or unconscious increased somewhat
over time. When adjusting for these factors, the increase in survival
seen in Tables 3
and 4 became more marked, although the estimates of
relative risk were of the same magnitude. When analyzed
separately, there was no improvement in survival among patients with
the most severe strokes (OR for 28-day case fatality period 1985 to
1986 as compared with 1993 to 1994, 0.90; 95% CI, 0.44 to 1.87 for men
and women together).
View this table:
[in a new window]
Table 5. Clinical Characteristics of First-Ever Stroke
Patients by Time Period in the Northern Sweden MONICA Study, Ages
25–74 Years , trends in the underlying
cause of death (grouped into 3 groups: cerebrovascular disease ICDs 430
to 438, ischemic heart disease ICDs 410 to 414, and other
causes) among those who died within 1 year from the onset are shown.
There were no clear changes in the distribution of causes of death,
although the proportion of deaths from ischemic heart disease
tended to decrease in men, from 20% in 1985 to 1986 to 13% in 1993 to
1994 (not statistically significant).
View this table:
[in a new window]
Table 6. Time Trends in Underlying Cause of Death Among Those
Who Died Within 1 Year From Onset Northern Sweden MONICA Study, Ages
25–74 Years
Discussion
Top
Abstract
Introduction
Subjects and Methods
Results
Discussion
References
In this population-based study of first-ever stroke, both short-
and long-term survival were found to have improved over the period 1985
to 1994 among patients aged 25 to 74 years. The observed improvement in
survival could not be explained by changes in age distribution, by
stroke subtype, or by changes in other important prognostic factors. An
improving survival of patients with stroke has been observed in studies
worldwide, of which several were
hospital-based,4 11 12 and a few were
population-based.13 22 In Sweden, a decreasing
case fatality was observed during the 1970s and
1980s9 13 and between 1985 and
1993.22
Selected Abbreviations and Acronyms
CI
=
confidence intervals
ICD
=
International Classification of Diseases
ICH
=
intracerebral hemorrhage
IS
=
ischemic stroke
MONICA
=
Multinational Monitoring of Trends and Determinants in
Cardiovascular Disease
OR
=
odds ratios
SAH
=
subarachnoid hemorrhage
WHO
=
World Health Organization
Acknowledgments
This study was supported by grants from the Swedish National
Public Health Institute, the Swedish Medical Research Council
(27X-07192, 27P-12314, 27PE-12281), the Heart and Chest Fund, the
Swedish Stroke Patients' Association, the Joint Committee of the
Northern Sweden Health Care Region, and the Norrbotten and
Västerbotten County Councils. We are indebted to Prof Måns
Rosén at the Epidemiological Center of the Swedish Board of
Health and Welfare for his contributions to the design and conduct of
this study.
References
Top
Abstract
Introduction
Subjects and Methods
Results
Discussion
References
This article has been cited by other articles:
 |
|
 |
|
  B. Hallstrom, A.-C. Jonsson, C. Nerbrand, B. Norrving, and A. Lindgren Stroke Incidence and Survival in the Beginning of the 21st Century in Southern Sweden: Comparisons With the Late 20th Century and Projections Into the Future Stroke, January 1, 2008; 39(1): 10 - 15. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. T. S. Kumar, A. D. Pandyan, and A. K. Sharma Biomechanical measurement of post-stroke spasticity Age Ageing, July 1, 2006; 35(4): 371 - 375. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Dhamoon, R. R. Sciacca, T. Rundek, R. L. Sacco, and M.S.V. Elkind Recurrent stroke and cardiac risks after first ischemic stroke: The Northern Manhattan Study Neurology, March 14, 2006; 66(5): 641 - 646. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. N. Struijs, M. L.L. van Genugten, S. M.A.A. Evers, A. J.H.A. Ament, C. A. Baan, and G. A.M. van den Bos Modeling the Future Burden of Stroke in the Netherlands: Impact of Aging, Smoking, and Hypertension Stroke, August 1, 2005; 36(8): 1648 - 1655. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Medin, A. Nordlund, and K. Ekberg Increasing Stroke Incidence in Sweden Between 1989 and 2000 Among Persons Aged 30 to 65 Years: Evidence From the Swedish Hospital Discharge Register Stroke, May 1, 2004; 35(5): 1047 - 1051. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Engstad, M. Viitanen, and E. Arnesen Predictors of Death Among Long-Term Stroke Survivors Stroke, December 1, 2003; 34(12): 2876 - 2880. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Stegmayr and K. Asplund Stroke in Northern Sweden Scand J Public Health, October 1, 2003; 31(5): 60 - 69. [Abstract] [PDF]
|
|
|
|
|
|
Bibliography Scand J Public Health, October 1, 2003; 31(5): 85 - 91. [PDF]
|
|
|
|
|
|
A. Terent Trends in Stroke Incidence and 10-Year Survival in Soderhamn, Sweden, 1975-2001 Stroke, June 1, 2003; 34(6): 1353 - 1358. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Hallstrom, B. Norrving, and A. Lindgren Stroke in Lund-Orup, Sweden: Improved Long-Term Survival Among Elderly Stroke Patients Stroke, June 1, 2002; 33(6): 1624 - 1629. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Bronnum-Hansen, M. Davidsen, and P. Thorvaldsen Long-Term Survival and Causes of Death After Stroke Stroke, September 1, 2001; 32(9): 2131 - 2136. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Hannerz and M. L. Nielsen Life Expectancies Among Survivors of Acute Cerebrovascular Disease Stroke, August 1, 2001; 32(8): 1739 - 1744. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Stegmayr, T. Vinogradova, S. Malyutina, M. Peltonen, Y. Nikitin, and K. Asplund Widening Gap of Stroke Between East and West : Eight-Year Trends in Occurrence and Risk Factors in Russia and Sweden Stroke, January 1, 2000; 31(1): 2 - 8. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||