Cerebrovascular accidents (CVA) have kept the first rank of the fatality rate in Japan. Although the population of cerebral thrombosis has gradually increased to date, cerebral hemorrhage is still taking important place in CVA. It has been well experienced that fevering, unconsciousness and pathological respiration were fatal in the patients with CVA.
In this study an attempt was made to determine the survival and mortality of cerebral hemorrhage by the method of linear discriminant function using clinical symptomes and laboratory findings on the onset. The patients of cerebral hemorrhage admitted to Osaka University Hospital were investigated. The clinical data were obtained from 63 patients (44 males, 19 females) including 19 death cases (13 males, 6 females). Age, sex, level of consciousness, vomiting, convulsion, incontinence of urine, abnormality of the pupils and light reflexes, pathological respiration, proteinuria, glucosuria, pulse rate, body temperature, systolic and diastolic blood pressure, erythrocyte sedimentation rate, white cell count and percentage of neutrophils were used as parameters of routine examination. Regarding quantification of each symptom, the method of“0 or 1” was adopted. For example, “0”or“1”represented that symptom was“absent”or“present”respectively. The correlation analysis and the Chi-square test between death or life and parameters mentioned above produced following results.
Disturbance of consciousness, incontinence of urine, abnormality of light reflex of pupils, pulse rate, body temperature, blood pressure and white cell count correlated significantly with the fatal prognosis of the life. Especially incontinence of urine, pulse rate or body temperature was significant at the 0.1 percent level. By using above eight parameters, linear discriminant function was led to decide the survival of patient during one month after the onset. The formula was as follow:
Score=-0.488×incont. unine
-2.06×abnorm. of light reflex
-1.33×level of consc.
-0.0964×pulse rate (/min.)
-1.59×body temp.(°C)
+0.00122×systol. b. p.(mmHg)
-0.0239×diastol. b. p.(mmHg)
-0.000135×white cell count
+75.6
“Score”was calculated in each patient according to this formula. If the score is above or equal zero, the patient was judged to be alive. If the score is below zero, the patient will die during one month. The probability of accurate diagnosis was calculated at 88.9% in the internal sample, white it was 83.9% in the external sample.
View full abstract