If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Physical Medicine and Rehabilitation, Shuang Ho Hospital, Taipei Medical University, New Taipei City, TaiwanGraduate Institute of Injury Prevention and Control, College of Public Health and Nutrition, Taipei Medical University, Taipei, Taiwan
Department of Physical Medicine and Rehabilitation, Wan Fang Hospital, Taipei Medical University, Taipei, TaiwanDepartment of Physical Medicine and Rehabilitation, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan
Wu C-H, Liou T-H, Chen H-H, Sun T-Y, Chen K-H, Chang K-H. Stroke risk in poliomyelitis survivors: a nationwide population-based study.
Objectives
To assess the prevalence and risk of stroke among adults with polio and controls.
Design
A prospective, probability-sampling, 6-year population-based cohort study.
Setting
A National Health Insurance Research Database consisting of 316,355 randomly selected enrollees. The database is related to a National Health Insurance program with more than 22 million participants.
Participants
After excluding patients under 40 years of age, polio patients (N=212) (mean age ± SD, 54.0±10.2y; 57.1% men) were identified from the database from January 1, 2003 to December 31, 2008. For each polio patient, 2 age- and sex-matched patients were recruited as controls. Control patients did not have any neuromuscular diseases commonly found in childhood. The frequencies of patients with potential risk factors for stroke were assessed.
Intervention
None.
Main Outcome Measure
The prevalence and the adjusted odds ratio of ischemic stroke among polio patients and the controls were estimated.
Results
Polio patients had a higher prevalence of stroke (10.8% vs 2.4%, P<.001) than the controls. Polio patients with hypertension had a much higher prevalence of stroke (23.0%). The risk of stroke was higher for polio patients compared with the controls, yielding an adjusted odds ratio of 4.17 (95% confidence interval, 1.84–9.45, P<.001). Polio was a significant risk factor for stroke independent from hypertension, diabetes mellitus, hyperlipidemia, and cardiac diseases.
Conclusions
Adults with polio had a high prevalence of ischemic stroke. Polio was an additional risk factor for stroke. Polio patients with hypertension might potentiate the risk of stroke. Developing a health promotion program, suitable for polio patients, to increase participation in activities and exercises may be essential, especially for polio patients with hypertension.
Differing in other cardiovascular disease (CVD) victims, stroke patients are often left with some neurologic impairments, such as altered consciousness or cognition, motor weakness or incoordination, sensory impairment, and sphincter dysfunction. All or part of these distinctive features can make stroke patients dependent on activities of daily living. One third of the stroke patients or half of the stroke survivors thus have a major disability in their life after stroke.
Being restrained from participating in physical exercise, adults who are disabled since childhood are more likely to have premature CVDs. Adults with previous poliomyelitis (polio) tend to experience hypertension and coronary heart disease,
and the mean ages of polio patients studied in those published articles were younger than 65 years. To our knowledge, articles in the literature describing stroke in polio survivors were scarce. This issue might be important because the combined loss of function in stroke patients with preexisting polio might mutually potentiate rather than simply add up each individual functional loss and could further compromise their participation in activities of daily living.
Polio was a common cause of physical disability in children in Taiwan in the 1950s and 1960s. Most polio patients are now 50 years or older. The prevalence of polio was estimated to be 1.17 to 1.63 per 1000 patients in Taipei.
In Taiwan, polio patients often experienced the onset of polio before the age of 3 years. Most survivors had the spinal form of polio with muscular atrophy and weakness of the legs.
Varied clinical patterns, physical activities, muscle enzymes, electromyographic and histologic findings in patients with post-polio syndrome in Taiwan.
We hypothesized that adults with previous polio would be more likely to have stroke. To conduct a nationwide sampling study, the Taiwan National Health Insurance Research Database (NHIRD) was used to test our hypothesis. The aims of this study were to assess the prevalence and risk of stroke among adults with polio and age- and sex-matched controls.
Methods
Taiwan NHIRD
The National Health Insurance (NHI) program is a mandatory comprehensive medical care program with more than 22 million (99% of Taiwan's population) participants. One million enrollees in the NHI program in Taiwan in the year 2005 were randomly selected into the NHIRD (fig 1).
Every person in Taiwan has an equal chance to be sampled into the database. All of the health record data, including longitudinal medical claims for each selected enrollee, from January 1, 2003 to December 31, 2008, were compiled. The NHIRD thus consists of all the information on diagnosis and medical service, including ambulatory care and hospital inpatient care, of 1 million randomly selected enrollees for 6 years. For public purchase, data of those 1 million enrollees was equally divided into 25 data subsets with random numbers. Therefore, every data subset contains 40,000 enrollees' medical information from January 1, 2003 to December 31, 2008. With 8 data subsets of the Taiwan NHIRD, a nationwide population-based cohort study was conducted to evaluate the prevalence of ischemic stroke among adults with previous polio.
Fig 1Summary of the construction of the Taiwan NHIRD and the processes of recruitment of polio and control patients in the study.
Data of 316,355 patients of those 8 data subsets randomly selected from the Taiwan NHIRD from January 1, 2003 to December 31, 2008 were used in this study (see fig 1). Personal identification data of the NHIRD were delinked information for public purchase. Therefore, this study was exempt from full review by the institutional review board. The data of the patients with previous polio, including an International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) code of 138,
were extracted from those 8 data subsets of the NHIRD with an electronic search. For each patient with polio (polio group), 2 age- and sex-matched patients were recruited as controls (control group). The birthday of each control patient was no more than 7 days of age before or after that of the matched polio patient. Control patients did not have cerebral palsy, intellectual disability, and any other neuromuscular diseases commonly found in childhood. Starting from January 1, 2003 forward, a research assistant searched manually and selected the 2 matched patients first met in the database into the control group. These data, other than sex and birthday of each control patient, were not examined in the recruitment processes. For each polio patient, the searching processes stopped when the 2 matched control patients had been met. The date of each participant's first medical claim in the database was used to calculate his or her age using the formula: (the date of having first medical claim – birthday)/365. Excluded were those under 40 years of age.
Patients who had had an ischemic stroke (ICD-9-CM code of 433 or 434)
in both the polio and control groups were identified. For each patient, the end date of the study coincided with the recorded date of stroke in the database. For patients without a stroke during the study period, the end date of the study for both groups was December 31, 2008. The frequencies of patients with risk factors for stroke,
including dysphagia (787.2), speech disturbance (784.5), visual field defects (368.4), neurologic neglect syndrome (781.8), abnormality of gait (781.2), and hemiplegia (342.9).
Data Analyses
The differences in clinical characteristics between the polio and control groups were compared with a Pearson chi-square test or a Fisher exact test. Logistic regression analyses with enter method were used to assess potential risk factors of having stroke. Those variables found to have significant association with stroke in the univariate analyses were entered into the model with a stepwise method. The data were analyzed with SAS software version 9.1.a The differences between the 2 groups were considered significant if P values were smaller than .05.
Results
A random sample of 212 polio patients (mean age ± SD, 54.0±10.2y; 57.1% men) between January 1, 2003 and December 31, 2008 were identified from the 8 data subsets (including 316,355 randomly selected patients) of the Taiwan NHIRD. The control group (n=424) had the same distribution of age and sex as the polio group. Table 1 compares the clinical characteristics of patients with polio and the matched controls. Polio patients experienced more frequencies of stroke (10.8% vs 2.4%, χ2=20.71, P<.001) and hypertension (χ2=14.06, P<.001) than the controls.
Table 1Comparison of Clinical Characteristics Between Patients With Poliomyelitis and the Matched Controls
Hemiplegia (n=3), dysphagia (n=1), and speech disturbance (n=1) were found only in polio patients having stroke. Moreover, polio patients with hypertension had much higher prevalence of stroke (20 of 87) than the controls (6 of 112) with hypertension (23.0% vs 5.4%) (fig 2). The prevalence for those without hypertension in both the polio (3 of 125) and control (4 of 312) groups (2.4% vs 1.3%) were not significantly different (Fisher exact test, P=.414). The hypertensive to normotensive ratio of the prevalence of stroke was 9.58 for polio patients and 4.19 for controls (χ2=16.90, P<.001).
Fig 2Comparison of the prevalence of ischemic stroke among polio and control patients with and without hypertension. The prevalence of stroke for polio with hypertension was 4-fold that of controls with hypertension (χ2=13.40, P<.001), 10-fold that of polio without hypertension (χ2=22.48, P<.001), and 18-fold that of controls without hypertension. *P<.05 (Fisher exact test), ***P<.001 (Pearson χ2 test).
Using univariate analyses, the associations between each clinical characteristic and stroke were shown in table 2. Table 2 is constructed by the simple logistic regression for each individual variable. Table 3 is the results of the stepwise method for model building. We have poliomyelitis, hypertension, atrial fibrillation, and myocardial infarction in the final model. The risk of stroke was higher for polio patients compared with controls, yielding an adjusted odds ratio of 4.17 (95% confidence interval, 1.84–9.45, P<.001) (see table 3). The logistic regression model had a good fit (Hosmer-Lemeshow test, χ2=2.22, P=.33). The interaction of hypertension by polio was not significant (β=1.02, P=.263) after adding it into the logistic regression model. The interaction of atrial fibrillation by polio was also insignificant (β=41.05, P=.999).
Table 2Univariate Analyses of Risk Factors for Stroke
With this probability sampling database, we found that polio patients had a higher prevalence and an excess risk of experiencing stroke than the age- and sex-matched controls. Polio was a significant risk factor for stroke, independent of hypertension, diabetes mellitus, hyperlipidemia, and cardiac diseases. In addition, polio patients with hypertension might potentiate the risk of stroke.
Compared with the prevalence range of stroke of 1.3% to 2.6% among American Asians from 2003 to 2008,
we found that our control patients had a similar prevalence (see table 1). All these figures were lower than the prevalence of stroke of 10.8% among polio patients in this study. The occurrence of stroke in polio survivors may result in an additional loss of arm function, which is often normal in polio survivors in Taiwan,
Varied clinical patterns, physical activities, muscle enzymes, electromyographic and histologic findings in patients with post-polio syndrome in Taiwan.
Varied clinical patterns, physical activities, muscle enzymes, electromyographic and histologic findings in patients with post-polio syndrome in Taiwan.
Thus, polio survivors with a stroke might have an unfavorable walking outcome.
Being that the mean age of our polio participants was 54.0 years, the finding that 10.8% of these younger polio patients had a stroke may suggest that stroke developed early among polio patients. In addition, polio patients had an excess risk for stroke independent from hypertension (see table 3). Polio patients with hypertension had a much higher prevalence of stroke (see fig 2). The combined effect of preexisting polio and having hypertension on the risk of stroke thus might potentiate rather than add up each individual risk factor for stroke.
Obesity is common among persons with physical disability,
obesity may also be an important risk factor for stroke among polio patients. In contrast, the frequency of CVDs as a factor does not increase among adults with cerebral palsy.
reported that less than 5% of adults with cerebral palsy are obese and more than 15% of them are underweight. This anthropometric feature may be one of the reasons why adults with cerebral palsy do not have an increased prevalence of CVDs,
Therefore, an active program, suitable for polio patients, in order to increase participation in physical exercise and to reduce body weight, might be clinically important in preventing stroke among polio patients.
Study Limitations
This study has 3 limitations. First, the ICD-9-CM coding may be inaccurate when using the existing data from a large database such as NHIRD. But the Taiwan NHIRD is constructed of medical claims for payment in the NHI program. The authority of the NHI program audits those medical claims regularly. Therefore, the data accuracy of the Taiwan NHIRD might be acceptable. However, the coding not related to medical claims, such as the coding for disease severity, might be lacking under the fee-for-service payment mode in the NHI program. Second, probably because the medical service for obesity is not fully covered by the NHI program, obesity might not be well recognized in this study database. Those patients with lifestyle-related risk factors for stroke, such as smoking, physical inactivity, and an unhealthy diet, could not be recognized either. But the frequency of hypertension, diabetes mellitus, and hyperlipidemia is often influenced by those lifestyle-related factors. The associations between stroke and each clinical characteristic of hypertension, diabetes mellitus, and hyperlipidemia were analyzed in this study. Further studies are needed to evaluate the contribution of obesity and lifestyle-related factors to stroke among polio patients. And third, the genetic factor may account for up to 40% of the risk for stroke,
But to use the findings of this study globally, the influence of racial and socioeconomic differences on the occurrence of stroke needs to be evaluated.
Conclusions
Adults with polio had high prevalence of ischemic stroke. Polio was an additional risk factor for stroke, independent of hypertension. Therefore, developing a health promotion program, suitable for polio patients, in order to increase participation in activities and exercises may be essential, especially for polio patients with hypertension.
Varied clinical patterns, physical activities, muscle enzymes, electromyographic and histologic findings in patients with post-polio syndrome in Taiwan.
Supported by Wan Fang Hospital, Taipei Medical University (grant no. 101-wf-eva-12 ) and the Center of Excellence for Clinical Trials and Research in Neuroscience, Department of Health, Executive Yuan , Taiwan (grant no. DOH101-TD-B-111-003 ).
This study is based in part on data from the National Health Insurance Research Database provided by the Bureau of National Health Insurance, Department of Health, Taiwan, and managed by the National Health Research Institutes, Taiwan. The interpretation and conclusions contained herein do not represent those of Bureau of National Health Insurance, Department of Health, or National Health Research Institutes.
No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated.
In-press corrected proof published online on Aug 24, 2012, at www.archives-pmr.org.
P<.001 (vs controls; Pearson chi-square test).
00501-1&id=gr1.jpg){kind=link}
00501-1&id=gr2.jpg){kind=link}