|Year : 2013 | Volume
| Issue : 3 | Page : 216-221
A comparison of clinical diagnosis with Computed Tomography findings in stroke patients
Tomar Nyodu, Ksh. Birendra Singh, Jeetenkumar Singh, S Kenny, Ch. Dhanaraj Singh, M Kulachandra Singh
Department of Medicine, RIMS, Imphal, Manipur, India
|Date of Web Publication||19-Feb-2014|
Department of Medicine, RIMS, Imphal - 795 004, Manipur
Source of Support: None, Conflict of Interest: None
Objective: To compare clinical diagnosis with computed tomography findings in identifying the types of stroke. Materials and Methods: The study was carried out after obtaining approval from the Institutional Ethical Committee, in the Department of Medicine, RIMS, Imphal, Manipur, from October 2010 to September 2012. The study sample consisted of 150 patients presenting with first-ever acute stroke who were clinically evaluated. An interviewer-structured questionnaire was administered and Siriraj stroke score computed. Clinical diagnosis of stroke-type (hemorrhagic or ischemic) was made based on Siriraj stroke score. The stroke-type was classified and compared with CT scan diagnosis. Data were analyzed using SPSS 16 software. Results: There were 150 patients in the study sample, out of them, 71 were hemorrhagic stroke patients and remaining 79 were ischemic stroke patients diagnosed by CT scan. The correlation between stroke-type, headache, level of consciousness at presentation, vomiting, and diastolic blood pressure achieved statistical significance. Sensitivity, specificity, positive predictive value, and negative predictive values of Siriraj stroke score for cerebral hemorrhage were 78.87%, 91.13%, 88.88%, and 82.75%, respectively. Conclusion: Hemorrhagic and ischemic stroke cannot be distinguished clinically with a simple clinical evaluation, and it is virtually impossible for all stroke patients to have a computed tomography scan immediately after admission, especially in rural area. When computed tomography is not immediately available and the clinician wishes to start anti-thrombotic treatment, the clinical diagnosis based on the Siriraj stroke score can be useful to identify patients at low risk of intracerebral hemorrhage.
Keywords: Computed tomography, Siriraj stroke score, Stroke
|How to cite this article:|
Nyodu T, Singh K, Singh J, Kenny S, Singh C, Singh M K. A comparison of clinical diagnosis with Computed Tomography findings in stroke patients. J Med Soc 2013;27:216-21
|How to cite this URL:|
Nyodu T, Singh K, Singh J, Kenny S, Singh C, Singh M K. A comparison of clinical diagnosis with Computed Tomography findings in stroke patients. J Med Soc [serial online] 2013 [cited 2020 Oct 19];27:216-21. Available from: https://www.jmedsoc.org/text.asp?2013/27/3/216/127397
| Introduction|| |
WHO defined stroke as "rapidly developed clinical signs of focal (or global) disturbance of cerebral function; lasting more than 24 hours or leading to death, with no apparent cause other than vascular origin." 
After coronary heart disease and all cancers, stroke is the third most common cause of death in the world, causing about 4 millions deaths in 1990, three-quarters of them in developing countries. It is the most common life-threatening neurological disease, and is the most important single cause of severe disability.
Stroke is mainly classified as hemorrhagic stroke (Primary cerebral hemorrhage and subarachnoid hemorrhage) and ischemic stroke (Cerebral thrombosis or embolism). Therapeutic decisions regarding management of stroke require accurate diagnosis of stroke types (Hemorrhagic or ischemic stroke) and exclusion of mimics. This helps in opting for relevant therapeutic options like thrombolysis, anticoagulation, carotid endarterectomy etc. Appropriately-timed computed tomography is a safe, non-invasive gold standard, most accurate in distinguishing hemorrhagic stroke from ischemic stroke.
It is difficult to be sure clinically about the type of stroke in majority of cases as there is no specific differentiating feature. To overcome these difficulties, several weighted clinical scoring systems, based on discriminant analysis techniques have been developed. In the absence of CT scan, as is prevalent in this part of the world, the weighted clinical scoring system such as Siriraj stroke score (SSS)  may be used for improved diagnostic gain. We carried out this study;
- To compare clinical diagnosis based on Siriraj stroke score with CT scan in identifying the type of stroke, and
- To study the different clinical variables and its correlation with the types of stroke.
| Materials and Methods|| |
The study was carried out after obtaining approval from the Institutional Ethical Committee, in the Department of Medicine, RIMS, Imphal, Manipur, from October 2010 to September 2012. This was a cross-sectional study, in which 150 consecutive stroke patients fitting in the inclusion criteria, admitted in the medicine wards during the time interval of the study, were taken up for the study.
- Those who fulfill the WHO criteria for stroke.
- Those for whom CT scan brain was done.
- Causes of focal neurological deficit other than stroke.
- Repeated or recurrent stroke.
| Methods|| |
Detailed history of clinical information was taken, and detailed examination was conducted. If patient was aphasic or unconscious, information from a spouse or relative was taken.
After detailed history, general physical and neurological examination, Siriraj stroke score (SSS) was calculated by obtaining details of each clinical variable used for the calculation in the pretested performa, designed for the purpose (Annexure A).
The SSS was calculated as (2.5 × level of consciousness) + (2 × vomiting) + (2 × headache) + (0.1 × diastolic blood pressure) - (3 × atheroma markers) - 12. This was computed for each patient, and clinical diagnosis of stroke type (Hemorrhagic or ischemic stroke) was made based on the Siriraj stroke score. Cut-off level for Siriraj stroke score validated by original study  was used in this study.
- Score greater than + 1 = Hemorrhage
- Score less than - 1 = Infarction
- Score between + 1 and -1 = Uncertain or ill-defined
Every patient was subjected to undergo non-contrast computed tomography (NCCT) scan of brain preferably within 24-72 hours after admission but not later than two weeks after the onset of the stroke by using Brilliance Philips 64 slice CT scan machine in the department of radio diagnosis, RIMS, Imphal. Every NCCT scan was reported by consultant radiologist who was blinded to clinical diagnosis of stroke subtypes based on Siriraj stroke score.
NCCT scan brain was used as the gold standard for the diagnosis of hemorrhagic and ischemic stroke.
Diagnosis of stroke type based on Siriraj stroke score was compared with the results of CT scan.
SPSS 16.0 was used for statistical analysis. Results are reported as Mean ± SD for quantitative or continues variables and number of cases along with percentages for the categorical or qualitative variables. Independent sample t-test for continuous variables and χ2 - test for categorical variables were used for test of significance. In order to measure the magnitude of agreement/similarity of clinical diagnosis findings with the computed tomography findings, Kappa statistic is calculated. Again, to test the validity of the clinical diagnosis based on the gold standard (computed tomography findings), sensitivity and specificity along with positive predictive value and negative predictive value are also deliberated. The P-values of <0.05 are used as the cut-off values for significance.
| Results and Observation|| |
There were 150 patients in the study sample, out of which 71 were hemorrhagic stroke patients and remaining 79 were ischemic stroke patients diagnosed by CT scan, a gold standard, with a ratio of 1 : 1.1. The age of the patients ranged from 22 to 90 years with a mean of 62.61 ± 14.158 years. Ischemic stroke patients were significantly older than that of hemorrhagic stroke patients (P = 0.023) with their respective mean ages 65.1 years and 59.85 years [Table 1].
|Table 1: Stroke-wise distribution of cases according to demographic factor|
Click here to view
The level of consciousness at presentation between hemorrhagic and ischemic strokes is quite different since in the former type, drowsiness/stupor occur high percentage (49.3%), which is followed by coma (29.6%) and alertness (21.1%) respectively whilst in latter type, the highest percentage (62.0%) pertains to alertness followed by drowsiness/stupor and very minimal i.e., only 2.5% recorded as coma. The dissimilarity of consciousness between the two types of strokes is quite significant as evident by P < 0.001.
Almost half (47.9%) of the hemorrhagic stroke patients have headache within two hours of onset of the symptom as against only below 20% ischemic stroke patient feels the same ailment. Similar in the case of vomiting too, 36.6% of the hemorrhagic patients have vomiting after onset while only 7.6% of the ischemic patients have the same ailment [Table 2].
|Table 2: Stroke-wise distribution of cases according to clinical presentation|
Click here to view
In case of blood pressure, diastolic level significantly differs from one type of stroke than the other type (P < 0.001) while systolic level does not (P = 0.054), which is just below significant level i.e., <0.05 [Table 2]. An important indication is that around 55% stroke cases associate with the history of hypertension.
Comparison of diagnostic findings between through CT scan and clinically is made and results are set forth in [Table 3]. Clinically, 10% of strokes are unable to detect definite types of stroke. The percentage fluctuation witnessed in the table is again tested and happens to be highly noteworthy (P < 0.001).
|Table 3: Diagnostic comparison of stroke by CT scan and clinical diagnosis|
Click here to view
For analysis purpose, ischemic stroke diagnosed by CT scan is termed as non-hemorrhagic stroke while ischemic stroke and undetermined by clinical diagnosis is pooled as non-hemorrhagic stroke [Table 4]. Here, Kappa = 0.704 indicates that there is a good agreement (beyond chance) of clinical diagnosis with the CT scan diagnosis to identify hemorrhagic stroke among the stroke patients. This statement is based on the thumb rule of ideal agreement of Kappa viz., if Kappa > 0.75, the agreement is excellence, if it is between 0.75 to 0.40 reveals intermediate to good agreement while < 0.40 implies poor.
|Table 4: Diagnostic agreement of hemorrhagic stroke by CT scan and clinical diagnosis|
Click here to view
Further validity of the clinical diagnosis test is measured by sensitivity, specificity, positive predictive value, and negative predictive value and their corresponding values are 78.87%, 91.13%, 88.88%, and 82.75%, respectively. Sensitivity = 78.87% implies that the test can detects 78.87 hemorrhagic strokes out of the 100 already known hemorrhagic stroke patients. Similarly, specificity (91.13%) infers that the test can identify 91.13 non-hemorrhagic strokes out of the 100 already known non-hemorrhagic stroke patients. The sensitivity and specificity are found to be excellence as both of them have more than 80%. The positive predictive value is found to be 88.88%, and negative predictive value is recorded as 82.75%.
When classification is made after elimination of 15 undetermined stroke cases by the clinical diagnosis, Kappa is noticed as 0.777 showing there is a good similarity of the test with the gold standard. At the same times, the sensitivity, specificity, positive predictive value, and negative predictive value are 87.50%, 90.14%, 88.88%, and 88.88%, respectively [Table 5].
|Table 5: Diagnostic agreement of stroke by CT scan and clinical diagnosis after elimination of undetermined stroke cases|
Click here to view
| Discussion|| |
Differentiation of cerebral infarction and cerebral hemorrhage is the most important first step in the management of acute stroke as clinical management of the two disorders differs substantially. The National Institute of Neurological Disorders and Stroke (NINDS) recombinant tPA (rtPA) Stroke Study showed a clear benefit for IV rtPA in selected patients with acute ischemic stroke. Based on the different study data, rtPA is being reviewed for approval in the 3-4.5-hour window in Europe, but is only approved for 0-3 hours in the United States and Canada. Use of IV tPA is now considered a central component of primary stroke centers as the first treatment proven to improve clinical outcomes in ischemic stroke and is cost-effective and cost-saving. 
In most developed countries, diagnosis is easily obtained by CT scanning, which allows the accurate distinction of hemorrhagic and ischemic types. However, quick access to CT scanning is not available in every country and hospital.
Since the clinical distinction between hemorrhagic and ischemic stroke cannot be achieved with a simple clinical evaluation, and it is virtually impossible to submit all stroke patients to CT, a weighted clinical score may offer some advantages to physicians who are involved in stroke management.
With this background, this study "A comparison of clinical diagnosis with computed tomography findings in stroke patients" was carried out in the Department of Medicine, RIMS, Imphal, Manipur during the period of October 2010 to September 2012.
There were 150 patients in the study sample, out of them, 71 were hemorrhagic stroke patients and remaining 79 were ischemic stroke patients diagnosed by CT scan, a gold standard, with a ratio of 1 : 1.1. O'Donnell et al. in a study of 3000 hospitalized first-in-a-lifetime stroke patients across 22 countries found that overall 22% of strokes were due to cerebral hemorrhage and 78% due to ischemia. The percentage due to cerebral hemorrhage was lowest in high-income countries, where 9% were due to cerebral hemorrhage and 91% due to ischemia. Sites in Africa had the highest percentage of strokes due to cerebral hemorrhage, at 34%. In our study, as compared to other studies,  the proportion of hemorrhagic stroke subtypes was more. This variability in the stroke subtypes may partly be due to the less sample size and in limited resources area; often only patients with more severe symptoms are admitted to hospital. Such patients are likely to have had a stroke due to cerebral hemorrhage rather than an ischemic stroke.
Age is probably the risk factor best correlated with stroke. The Framingham study showed that as a person ages, his or her risk of stroke increases, with incidence rates per 10,000 increasing from 22% to 32% to 83% in the age groups 45 to 55, 55 to 64, and 65 to 74 years, respectively. 
In our study, the age of the patients ranged from 22 to 90 years with a mean of 62 ± 14 years, and result of the present study is also similar with the findings of Emmanuel et al. in which the mean age of stroke was 64.2 years.
Men have a greater frequency of stroke than women, but because life expectancy is higher in women, women often outnumber men in many stroke studies.  During the pre-menopausal years, women have fewer strokes than men, but incidence levels off after age 60 years. Yukihiro et al. in their hospital-based study of the Care and Cost of Acute Ischemic Stroke in Japan found 69% male-to-female predominance. In our study also, male has more percentage (60%) of stroke than that of his counterpart female (40%); this is true in both types of strokes, but gender does not have any significant role to distinguish the types of strokes (P = 0.423).
Both ischemic and hemorrhagic strokes sometimes increase intracranial pressure indirectly as a result of brain edema. Hemorrhagic strokes can also increase intracranial pressure directly by adding extra vascular blood to the restricted intracranial space. Clinically, elevated ICP presents as headache, vomiting, and a decreased level of consciousness. In our study, the level of consciousness at presentation between hemorrhagic and ischemic strokes is quite different since in the former type, drowsy/stupor notices highest percentage (49.3), which is followed by coma (29.6%) and alert (21.1%) respectively whilst in latter type, highest percentage (62.0) pertains to alert followed by drowsy/stupor and very minimal i.e., only 2.5% recorded as coma. The dissimilarity of consciousness between the two types of strokes is quite significant as evident by P < 0.001.
In a prospective study of 240 consecutive patients with acute stroke, Arboix et al. found that headache occurred in 32% of 195 patients with ischemic stroke and 64.5% of 45 patients with hemorrhagic stroke (P < 0.0001). In our study, headache occurred in 16.5% of 79 patients with ischemic stroke and 47.9% of 71 patients with hemorrhagic stroke (P < 0.001) and vomiting occurred in 36.6% of hemorrhagic stroke and 7.6% of ischemic stroke cases (P =< 0.001).
Sallam et al. in their retrospective, hospital-based study conducted in Kuwait teaching hospital in Sana'a, over a four-year period (1999-2003), found that the major associated medical conditions in stroke patients were hypertension (HTN) in 68.3%, cardiac diseases that account for 42.4%, and diabetes mellitus (DM) in 24.4%. Hypertension was more common in the hemorrhagic stroke presents; 82.6%, versus 61.6% in the ischemic group. Smoking history was positive in 42%. In our study, history of hypertension was positive in 56% of stroke patients, diabetes in 16.7%, and smoking in 43.3%. The patients with the history of diabetes are more prone to have ischemic stroke than hemorrhagic stroke as the former attributes 27.8% history of diabetes while the latter associates only 4.2% history of the diabetes.
In this study, in comparing the effectiveness of the clinical diagnosis by using Siriraj stroke score with the computed tomography as a gold standard to differentiate stroke subtype, the Siriraj stroke score had a sensitivity, specificity, positive predictive value, and negative predictive value of 78.87%, 91.13%, 88.88%, and 82.75% for hemorrhagic strokes and for the ischemic strokes, they were 81.08%, 88.73%, 88.88%, and 80.76%, respectively. Many previous studies reported varied results. Our finding is consistent with the finding of Raghuram et al. in which they reported sensitivity and specificity of 87.93% and 77.27% for ischemic strokes and 77.27% and 87.93% for the hemorrhagic strokes. Our finding is contradictory with the finding of Ogun et al. in which Siriraj stroke score had a sensitivity of 50% for cerebral hemorrhage and 58% for cerebral infarction with an accuracy of 54.2%. But their sample size was only 96 cases.
The diagnostic sensitivities of the Siriraj stroke score for intracranial hemorrhage and infarction were 85% and 90%, respectively, with an overall predictive accuracy of 88.5% in the study conducted by Hung et al. and this result is also similar with finding of the present study.
| Conclusion|| |
Hemorrhagic and ischemic stroke cannot be distinguished clinically with a simple clinical evaluation, and it is virtually impossible for all stroke patients to have a computed tomography scan immediately after admission, especially in rural area. When computed tomography is not immediately available and the clinician wishes to start anti-thrombotic treatment, the clinical diagnosis based on the Siriraj stroke score can be useful to identify patients at low risk of intracerebral hemorrhage. A prospective study with a larger sample size and modification of the discriminant variables is suggested for definitive conclusion.
| References|| |
|1.||Park K. Park's Textbook of Preventive and Social Medicine. 19 th ed. Jabalpur, India: Banarsidas Bhanot; 2007. |
|2.||Poungvarin N, Viriyavejakul A, Komontri C. Siriraj stroke score and validation study to distinguishing supratentorial intracerebral hemorrhage from infarction. BMJ 1991;302:1565-7. |
|3.||Smith WS, English JD, Johnston SC. Cerebrovascular disease. In: Longo DL, Fauci AS, Kasper DL, Hauser SL, Jameson JL, Loscalzo J, editors. Harrison's Principles of Internal Medicine. 18 th ed. New York: McGraw Hill; 2012. p. 3270-99. |
|4.||O'Donnel MJ, Xavier D, Liu L, Zhang H, Chin SL, Rao-Melacini P, et al. Risk factors for ischemic and intracerebral hemorrhagic stroke in 22 countries (the INTERSTROKE study): A case-control study. Lancet 2010;376:112-23. |
|5.||Kannel WB. Current status of the epidemiology of brain infarction associated with occlusive vascular disease. Stroke 1971;2:295-318. |
|6.||Emmanuel S, Papa SM, Christophe D, Khadi BF, Abdourahmane N, Sarah G, et al. Ischemic and Hemorrhagic strokes in Dakar, Senegal. A Hospital-Based study. Stroke 2005;36:1844-7. |
|7.||Sacco RL. Risk factors, outcomes, and stroke subtypes for ischemic stroke. Neurology 1997;49:39-44. |
|8.||Yukihiro Y, Toshiyuki U, Hiroshi Y, Yasushi K, Masayasu T, Etsuro M. Hospital-based study of the care and cost of acute Ischemic stroke in Japan. Stroke 2003;34:718-24. |
|9.||Arboix A, Massons J, Oliveres M, Arribas MP, Titus A. Headache in acute cerebrovascular disease: A prospective clinical study in 240 patients. Cephalalgia 1994;14:37-40. |
|10.||Sallam AR, Khalid AA, Awn H. The clinical profile of stroke: A Yemeni experiences. J Med J 2009;43:115-21. |
|11.||Raghuram PM, Biradar MS, Jeganathan J. Comparison of the Siriraj stroke score and the Guy's Hospital score in south India. J ClinDiagn Res 2012;6:851-4. |
|12.||Ogun S, Oluwole S, Aogunseyinde O, Fatade A, Ojini F, Odusote A. Accuracy of the Siriraj stroke in differentiating cerebral hemorrhage and infarction in African Nigerians. Asian J Neurosurg 2001;20:21-6. |
|13.||Hung LY, Wang PY, Wang Y, Chia LG. Clinical distinction between acute hemorrhagic and acute ischemic stroke by Siriraj stroke score. Zhonghua Yi Xue Za Zhi1995;55:248-52. |
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]