|Year : 2007 | Volume
| Issue : 3 | Page : 154-159
Surveillance of stroke: WHO STEP-wise approach: A Chennai stroke unit report
Deepak Arjundas1, Uma Pandiyan1, G Arjundas2, Baux Henry3
1 Department of Neurology, Vijaya Health Center, Chennai, India
2 Mercury Nursing Home, Chennai, India
3 Stroke Unit, Vijaya Health Center, Chennai, India
Mercury Nursing Home, 36, Pantheon Road Egmore, Chennai - 600 008
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Stroke is potentially the most devastating consequence of vascular disease, causing long-term disability and placing high medical, emotional, and financial burdens on the patient. WHO STEPS stroke surveillance study, provided the framework for our study of stroke patients in Chennai. In our center, we followed only the STEP 1 study and only hospital-based stroke data were collected. STEP 2 and 3 involved study of strokes in the community. We found that 78% of our patients were in the 6th to 8th decades of life; 72% had hypertension, either alone or in combination with other risk factors; 4 weeks after the stroke, 51% of our patients were independent with regard to mobility and the activities of daily living, while 14% had severe disability. We found a mortality rate of 14%. We concluded that the incidence of stroke in the community can be decreased further by effective risk factor modification. The admission delay needs to be shortened from >48 h to less than 6 h. Our data also gave us fresh information about the recovery of motor function in stroke survivors at 28 days after the stroke.
Keywords: Disability, modified Rankin scale, NIH scale, risk factor, stroke surveillance, TOAST
|How to cite this article:|
Arjundas D, Pandiyan U, Arjundas G, Henry B. Surveillance of stroke: WHO STEP-wise approach: A Chennai stroke unit report. Ann Indian Acad Neurol 2007;10:154-9
|How to cite this URL:|
Arjundas D, Pandiyan U, Arjundas G, Henry B. Surveillance of stroke: WHO STEP-wise approach: A Chennai stroke unit report. Ann Indian Acad Neurol [serial online] 2007 [cited 2019 Nov 18];10:154-9. Available from: http://www.annalsofian.org/text.asp?2007/10/3/154/34794
Stroke is potentially the most devastating consequence of vascular disease, causing serious long-term disability and placing extremely high medical, emotional, and financial burdens on the patient. It is the second leading cause of death and one of the commonest causes of disability in adults.  In 2005, stroke accounted for 5.7 million deaths worldwide, which is equivalent to 9.9% of all deaths. Two-thirds of those deaths occurred in people living in developing countries.  While India is still struggling with the problems of communicable diseases, noncommunicable diseases are on the rise. Stroke has a multifactorial etiology. Combinations of risk factors over time influence a person's likelihood of suffering a stroke. Hypertension is the most important risk factor for stroke. Control of hypertension can decrease the morbidity and mortality due to stroke. Age is an important and independent risk factor for stroke.  Among the hospital stroke population, 20-30% are below the age of 40 years. 
The objective of the study was to analyze the data profile of our stroke patients and to assess the outcome of our protocol for treating patients with stroke.
| Materials and Methods|| |
This WHO STEP-wise stroke surveillance/ICASS-II study is a sequel to the ICASS-I study conducted in 2002. ICASS is the acronym for Indian Cooperative Acute Stroke Study, which was done by the Indian Stroke Association. The WHO stroke surveillance part of the study focuses on hospitalized stroke patients only (STEP-1). We did not undertake the evaluation of strokes in the community (STEP-2 and STEP-3) due to various reasons. The STEP-I study has three modules. The first module contains information on age, gender, vital status (on the 10 th and 28 th days of illness), and whether it is a confirmed stroke. The second module gives information on treatment details before admission, during hospitalization, and after discharge. The severity of stroke and the functional ability after the stroke are also evaluated in the second module. The third module gives details about the type and size of stroke as assessed by neuroimaging. The ICASS-2 part of the study focused on admission delay and risk-factor assessment.
This is a prospective study conducted in the stroke units of two private hospitals in Chennai. A total of 402 patients were evaluated at Vijaya Health Center and Mercury Nursing Home in Chennai, India. All patients had a focal neurological impairment of sudden onset, which had lasted more than 24 h. The strokes were confirmed by CT or magnetic resonance imaging (MRI). T1W and T2W diffusion, flair sequences, and MR angiogram of the carotids and cerebral circulation was done. The duration of this study was from December 2003 to December 2004. A pilot study with 10 cases was evaluated before the actual study started in December 2003.
The data on age, sex, occupation, socioeconomic status, race, and religion were also obtained for the surveillance study. The type and subtype of each stroke was established and analyzed. The severity of the stroke was evaluated by the modified NIH scale and the outcome by the Modified Rankin Scale (MRS).  The NIH scale was used for evaluation of severity on the day of admission and on the third day after admission; the higher the NIH score the more severe the stroke. The subtype of ischemic stroke was classified using the Trial of ORG 10172 in Acute Stroke Treatment (TOAST) classification. The outcome was evaluated by MRI 4 weeks after the onset of stroke; assessment was done by physical evaluation of the patient or by telephonic interview of the patient's attendant. In this study, the 4 th week follow-up was possible only in 354 of the 402 patients. This was because some patients had come from remote rural areas where communication links were inadequate, some patients were illiterate, or the contact addresses they had given were incomplete. On the MRS, a score of < 3 indicates that the patient is able to walk independently and requires no assistance for his/her activities of daily living (ADL), a score of 3-4 indicates that assistance is required for moving around, a score of 5 indicates that the patient is totally dependent on another person for his ADL, and the score of 6 is used to document mortality.
The risk factors of age, hypertension, diabetes mellitus, ischemic heart disease (IHD), dyslipidemia, anemia, and nicotine intake were evaluated. A blood pressure of 140/90 mm Hg and above was taken to denote hypertension. Blood sugar (postprandial venous sample) >180 mg/dl was taken as indicating diabetes. IHD was diagnosed by electrocardiogram and echo findings, the diagnosis being confirmed by a cardiologist.
Multiple logistic regression (forward stepwise addition method) was used for statistical analysis. P -values and odds ratios were calculated to assess the statistical significance of the various factors. Pearson's correlation analysis was also done to look for linear association between single and combinations of risk factors and stroke.
| Results|| |
There were 313 (77.8%) patients who were between 50 and 79 years of age. The youngest patient was 17 years and the oldest was 92 years old. Severe disability and death is most common in the 6 th to 8 th decades of life, as revealed by MRS scores of 5 and 6; the P -value is 0.04 which is statistically significant [Table - 1]. The mean age of stroke is 61.7± 13.4 (SD).
There were 265 (65.9%) male patients and 137 (34.1%) female patients. We examined the effect of gender on stroke outcome in the 354 patients whom we could follow up. 119 men had MRS scores of <3, 32 had a score of 5, and 37 died. Among female patients, 60 had MRS scores of <3, 20 had a score of 5, and 15 patients died. The association between gender and stroke outcome was not statistically significant.
Type of stroke
A total of 306 (77%) patients had ischemic strokes and 69 (17%) had hemorrhagic strokes. Ten (3%) were due to subarachnoid hemorrhage. Of the infarctions, 195 (60%) were due to large artery atherosclerosis and 97 (30%) were due to small vessel occlusion. This was classified as per the TOAST classification, where the size of infarct was identified after MRI. Cardioembolic stroke was present in 13 patients (4%).
Treatment history prior to stroke
Fifty-one patients were known hypertensives on treatment before the onset of stroke and 28% were diabetics on treatment prior to the stroke; 21% of our patients were on antiplatelet medication, 4% were on cholesterol-lowering medication, and 3% were on anticoagulants.
Severity of stroke
The severity of stroke was assessed by the NIH stroke scale. It was assessed on admission and on the third and seventh day of illness. The higher the NIH score, the more the impairment due to stroke; the lower the NIH score, the milder the stroke. The grading was as follows: a score of 0-5 was taken as indicating mild stroke, 6-15 was taken as indicating moderate stroke, and 16-42 was taken as indicating a severe stroke. By these criteria, 56% had a mild stroke, 26% had a stroke of moderate intensity, and 18% had a severe stroke.
Forty-nine percent of the patients reached our stroke units after a 48 h delay. The patients were referred by other physicians/ hospitals and some were from the neighboring states. Fourteen percent of the cases reached our hospitals within 24-48 h of onset of symptoms, 8% within 3-6 h, and 9% within the first 3 h of onset of stroke.
The stroke outcome was assessed by enquiring about the vital status on the 10 th and 28 th day after onset. On the 10 th day of illness we found a mortality rate of 7%. By the 28 th day, our stroke mortality rate had gone up to 14%. The MRS was used to assess the outcomes: 51% had scores of 0-2 (mild disability), 20% had scores of 3-4 (moderate disability), and 15% had an MRS score of 5 (severe disability). The MRS was assessed by physical examination by the physician or the physiotherapist or by enquiry over the telephone.
Modifiable risk factors
Hypertension was found to be a major risk factor of stroke in 72% of the patients. It was found to be associated with stroke both alone and in combination with diabetes mellitus and IHD. Diabetes mellitus was present in 49.8% and IHD in 33.8% of the stroke patients. Around 16% of the patients in this study had no significant risk factors that we could detect with the available facilities [Table - 2].
Of the 402 patients, 289 (72%) had hypertension. The 4 th week follow-up was possible in 354 patients. Out of this, 260 were diagnosed as stroke with hypertension as a risk factor (i.e., hypertension alone or in combination with other risk factors). MRS scores were as follows: 129 patients had scores of 0-2 (mild disability), 57 had a score of 3-4 (moderate disability), and 41 had a score of 5 (severe disability); 33 patients died. The Chi-square value was 5.31 (with 3 degrees of freedom) and the P -value was 0.15, which is not statistically significant.
Diabetes was present in 200 patients (49.8%). Among these, the 28 th day follow-up was possible in 173 patients: 76 patients had MRS scores of 0-2 (mildly disabled), 37 patients had scores of 3-4 (moderate disability), and 27 had score of 5 (severe disability); 33 patients died. The chi-square value was 7.87 (with three degrees of freedom) and this was seen to be statistically significant ( P = 0.049).
IHD was present in 136 patients (33.8%). These cases were diagnosed by electrocardiogram and echo findings and the diagnoses were confirmed by a cardiologist. Twenty-eighth day follow-up was possible in 126 patients. Fifty-three patients had MRS scores of 0-2 (mild disability), 27 had scores of 3-4 (moderate disability), and 23 patients had a score of 5 (severe morbidity); 23 patients died. The chi-square value of 6.37 was not significant ( P = 0.1) .
The individual components of the lipid profile (such as total cholesterol, HDL, LDL, VLDL, and TGL) were evaluated. Total cholesterol was evaluated in 271 patients, of whom 176 had hypercholesterolemia and 95 had normal cholesterol values. Ninety-six patients had MRS scores of <3, MRS of 3-4 , and 24 had a score of 5; 12 had a score of 6. These findings, when compared to that in patients who had normal cholesterol levels, were not statistically significant ( P = 0.21).
The relationship between LDL levels and stroke outcome was evaluated in 242 patients. MRS scores were as follows: 137 had scores of <3, 49 had scores of 3-4, 27 had a score of 5, and 29 had a score of 6. The P -value is 0.43, which is not significant.
The effect of triglyceride level on stroke outcome was analyzed in 163 patients. Eighty-seven had MRS scores of <3, 33 had scores of 3-4, and 21 had a score of 5; 22 had a score of 6. The P -value is 0.26, which is not significant.
The effect of VLDL level on stroke outcome was analyzed in 127 patients. Seventy-one had MRS scores of <3, 23 had scores of 3-4, 19 had scores of 5, and 14 had a score of 6. The P -value is 0.42, which is not statistically significant.
Around 100 patients were omitted from lipid profile evaluation either due to financial constraints or because of the severity of the stroke.
Tobacco use was prevalent in 23.6% of the patients; all three modes were seen, i.e., smoking cigarettes or beedis, chewing tobacco with betel leaves, or using snuff. Anemia was present in 10% of the patients. Forty-eight patients had a MRS score of <3, 16 had a score of 3-4, and 8 had a score of 5; 8 had a score of 6. The P -value is 0.42, which is statistically not significant.
On evaluating the effect of a combination of risk factors, i.e., diabetes and hypertension, the number of patients with 5 and 6 on the MRS showed severe morbidity and mortality than those with minimal disability <3.
There were only 8 stroke patients with IHD alone as the risk factor and 15 patients with diabetes plus IHD (out of 354) as the risk factors. This figure is too small to be statistically significant. However, though there is an association of diabetes and IHD in the multifactorial etiology of stroke, these risk factors, independently, do not add any additional risk to the stroke outcome.
When three or more risk factors are present, the mortality (MRS = 6) rises to one-fifth [Table - 3].
No known risk factor
There were no identifiable risk factors in 55 of our patients. Of these patients, 32 had a favorable outcome and 8 patients died.
The result of the multiple logistic regression analysis shows an increase in morbidity and mortality, with the MRS scores more than 5, in the presence of risk factors such as age , diabetes mellitus, and IHD. The nonsignificant variables are sex, hypertension, and elevated LDL (dyslipidemia). Hypertension is the most significant risk factor for the occurrence of stroke, but the stroke outcome was not significantly affected by the presence of hypertension, as the number of hypertensives with favourable outcomes and those with increased morbidity and mortality were about the same [Table - 4].
| Discussion|| |
Over the years, in our stroke unit, we have studied the common reasons for high mortality and morbidity and instituted changes in treatment and monitored therapy. The effects of these changes are reflected in the improvement in our mortality and morbidity rates over the last 20 years. Our mortality rate at the 10 th day of illness was 7% but rose to 14% by the 28 th day. This was probably due to the severity of the stroke and the site of the lesion, as well as the development of respiratory complications.
Fifty-one percent of our stroke patients had a score of 0-2 on the MRS. This means that these patients were able to carry out their day-to-day activities without assistance. Twenty percent of the patients had a score of 3-4 on the MRS, which signifies moderate disability, requiring assistance for performance of their bodily needs. Fifteen percent had a score of 5 on the MRS, indicating severe disability; these patients were bedridden, incontinent, and required constant nursing care and attention. Patients with severe impairments placed enormous socioeconomic stress on their families.
In our study, the maximum stroke incidence was in the age-group of 50-79 years, with a mean age at incidence of 61.7±13.4 (SD). The mean age of our stroke patients was 10 years lower than that reported from studies in high-income countries.  Only population-based incidence studies would be able to clarify whether our results reflect the actual situation in the source population or if it is due to the fact that elderly stroke patients are less likely to be admitted to health facilities.
In a study done by Srivastava et al . at AIIMS, New Delhi, in 2001, the average admission delay was 7.6 h, but 25% reached the hospital within 3 h; this was in contrast to our study, where only 8% came within 3 h.  Distance from hospital, contact with a local doctor, and low or no threat perception were independent factors associated with delay in arrival.
Gender differences are present in the incidence of stroke in our study: 65.9% were men and 34.1% were women. However this could reflect the gender bias existing in Indian society. There is no statistically significant difference in the stroke outcome due to gender.
The proportion of hemorrhagic strokes is lower than that reported from other hospital-based studies in low- and middle-income countries,  where nearly half the patients were diagnosed with hemorrhagic strokes.
A Malaysian study by Ong et al . that was reported in 2002, concluded that hypertension and diabetes are the most common risk factors of strokes admitted to a tertiary hospital.  The prevalence of hypertension in their stroke patients was 71.5% and that of diabetes was 40.2%, which was similar to our findings. They reported a mortality of 20.3% at one month, while we had a mortality of 14%.
In our study, elevated total cholesterol, LDL, VLDL, TGL, and low HDL showed no relation to the incidence of ischemic or hemorrhagic stroke. Similar findings have been observed by Ross et al .  He found that elevated serum cholesterol was not linked to an increased stroke incidence. However hypercholesterolemia is related to the incidence of IHD, which is a risk factor for stroke. His study was a large, prospective observational study of middle-aged men and he found no relationship between plasma total cholesterol concentration and incidence of fatal or nonfatal stroke.
Two large meta-analyses , aggregated from very large cohorts failed to find a relationship between cholesterol and stroke. However a study  in 350,977 men aged 35-57 years found that a clear relationship emerged when stroke was categorized into ischemic and hemorrhagic types, the risk of ischemic stroke rising with increased cholesterol and that of hemorrhagic stroke rising as cholesterol values decreased.
In the WHO 2004 study by Lopez et al .,  they found a definite global and regional burden due to raised total cholesterol levels, leading to 4.4 million deaths annually.
Mcfarlane et al .  have concluded that the lowering of blood pressure to <130/80 mm of Hg is strongly recommended for the primary and secondary prevention of stroke. Lowering LDL to <100 mg/dl in diabetics without prior stroke and to <70 mg/dl in diabetics with stroke is currently recommended.
Tobacco, in various forms, was used by 23.6% of our patients. Our figures show increased mortality and morbidity in tobacco consumers compared to those who do not use tobacco in any form. The difference was statistically significant ( P = 0.009). Smokers, both current and past, have a higher risk of developing stroke. 
Lefkovits,  of Royal Melbourne Hospital Australia, found that age, cardiac disease, and diabetes independently worsen acute stroke outcome, which is consistent with our findings.
A Polish study by Szczepanska-Szerej et al .  in 2003 showed that diabetes has no effect on the course and outcome of ischemic stroke. In contrast, we found that in our population diabetes is an independent risk actor in increasing the morbidity. A Spanish study by Arboix et al .  on 393 diabetics with stroke found a different clinical pattern of ischemic stroke, with atherothrombotic stroke and lacunar infarcts being more frequent. This also affected the in-hospital mortality.
In our study, 16% of the patients had no known risk factors (cryptogenic stroke). Catalina and Adnan Qureshi et al .  have found that up to 30% of strokes remain cryptogenic after all the identifiable risk factors, including septal defects, have been ruled out. Unidentified environmental and genetic factors may play a role.
On evaluating the data profile and the predictors of stroke outcome in our stroke patients, the necessity for primary prevention of stroke amongst the patients' children is emphatically clear. The patient and his family must be educated regarding the measures for prevention of recurrent stroke.
| Conclusion|| |
The outcome of stroke essentially depends on the type of stroke, the severity, and the subtype. However the morbidity and mortality of stroke, as determined by the outcome at 4 weeks by the MRS, is significantly higher in elderly patients with diabetes and IHD and in smokers. The gender of the individual did not significantly alter the outcome. Hypertension definitely increased the incidence of stroke. Heightened public awareness of the risk factors and their management not only aids primary prevention of stroke, it also leads to early treatment and better stroke outcomes. Achieving these goals will require effective implementation of preventive interventions such as the global stroke initiative and programs similar to that recommended by the Center for Disease Control (CDC) in USA and the WHO. ,,
| References|| |
|1.||Lopez AD, Mathers CD, Ezzati M, Jamison DT, Murray CJ. Global and regional burden of disease and risk factors, 2001: Systematic analysis of population health data. Lancet 2006;367:1747-57. [PUBMED] [FULLTEXT]|
|2.||World health Organization. The World Health Report; 2002 Reducing risks: Promoting healthy life. World Health Organization: 2002. |
|3.||Wolf PA, Cobb JL, Agostino RB. Epidmiology of stroke in stroke. Pathophysiology, diagnosis and management. New York; 1992. p. 3-27. |
|4.||Nagaraja D, Taly AB. Stroke in the young. In: Sinha KK, editor. Progr Clin Neurosci 1988. p. 129-45. |
|5.||D'Ólaberriague L, Litvan I, Mitsias P, Mansbach HH. A reappraisal of reliability and validity studies in stroke. Stroke 1996;27:2331-6. |
|6.||Feigin VL, Lawes CM, Bennett DA, Anderson CS. Stroke epidemiology: A review of population-based studies of incidence, prevalence and case-fatality in the late 20th century. Lancet Neurol 2003;2:43-53. [PUBMED] [FULLTEXT]|
|7.||Srivastava AK, Prasad K. A study of factors delaying hospital arrival of patients with acute stroke. Neurol India 2001;49:272-6. [PUBMED] [FULLTEXT]|
|8.||Saposnik G, Del Brutto OH; Iberoamerican Society of Cerebrovascular Diseases. Stroke in South America: A systematic review of incidence, prevalence and stroke subtypes. Stroke 2003;34:2103-7. [PUBMED] [FULLTEXT]|
|9.||Ong TZ, Raymond AA. Risk factors for stroke and predictors of one-month mortality. Singapore Med J 2002;43:517-21. [PUBMED] |
|10.||Ross R. The pathogenesis of atherosclerosis: An update. N Engl J Med 1988;314:488-500. |
|11.||Cholesterol, diastolic blood pressure, and stroke: 13,000 strokes in 450,000 people in 45 prospective cohorts. Prospective studies collaboration. Lancet 1995;346:1647-53. [PUBMED] |
|12.||Atkins D, Psaty BM, Koepsell TD, Longstreth WT Jr, Larson EB. Cholesterol reduction and the risk for stroke in men. A meta-analysis of randomized controlled trials. Ann Intern Med 1993;119:136-45 |
|13.||Iso H, Jacob DR Jr, Wentworth D, Neaton JD, Cohen JD. Serum cholesterol levels and six-year mortality from stroke in 350,977 men screened for the multiple risk factor intervention trial. N Engl J Med 1989;320:904-10. |
|14.||McFarlane SI, Sica DA, Sowers JR. Stroke in patients with diabetes and hypertension. J Clin Hypertens 2005;7:286-92. |
|15.||Lefkovits J, Davis SM, Rossiter SC, Kilpatrick CJ, Hopper JL, Green R, et al. Acute stroke outcome: Effects of stroke type and risk factors. Aust N Z J Med 1992;22:30-5. [PUBMED] |
|16.||Szczepanska-Szerej A, Wojczal J, Belniak E, Krasinska-Czerlunczakiewicz H, Stelmasiak Z. Does diabetes mellitus affect the course and prognosis of ischemic stroke? Neurol Neurochir Pol 2003;37:327-37. |
|17.||Arboix A, Rivas A, Garcνa-Eroles L, de Marcos L, Massons J, Oliveres M. Cerebral infarction in diabetes: Clinical pattern, stroke subtypes, and predictors of in-hospital mortality. BMC Neurol 2005;5:9. |
|18.||Ionita CC, Xavier AR, Kirmani JF, Dash S, Divani AA, Qureshi AI. What proportion of stroke is not explained by classic risk factors? Prev Cardiol 2005;8:41-6. [PUBMED] [FULLTEXT]|
|19.||Bonita R, Mendis S, Truelsen T, Bogoulovsky J, Toole J, Yatsu F. The global stroke initiative. Lancet 2004;3:391-3. |
|20.||Center for disease control and prevention. CDC. US - Dept of health and human service Available from: http://www.cdc.gov/coh/action plan /index.htm. |
|21.||WHO - Report 2003, shaping the future. Geneva. |
[Table - 1], [Table - 2], [Table - 3], [Table - 4]
|This article has been cited by|
||Feasibility study of stroke surveillance: Data from Bangalore, India
| ||Nagaraja, D., Gururaj, G., Girish, N., Panda, S., Roy, A.K., Sarma, G.R.K., Srinivasa, R. |
| ||Indian Journal of Medical Research. 2009; 130(4): 396-403 |