[Logistic regression analysis on risk factors of cerebral hemorrhage complicated with stress ulcer].

OBJECTIVE

To explore the related risk factors of cerebral hemorrhage complicated with stress ulcer (SU).

METHODS

The clinical data of 1 185 patients with cerebral hemorrhage admitted to Department of Emergency Medicine of Nanjing General Hospital from March 2006 to March 2014 were retrospectively analyzed. Patients were divided into two groups according to whether patients complicated with SU or not. Data was collected within 8 hours after admission in two groups including gender, age, amount of bleeding, the bleeding site (basal ganglia, thalamus, brainstem, brain lobe, ventricle, subarachnoid, and cerebellum), disturbance of consciousness, acute physiology and chronic health evaluation II (APACHEII) score, systolic blood pressure (SBP), history of hypertension, and history of cerebral hemorrhage. The statistically significant risk factors found using univariate analysis was selected and was analyzed to find independent risk factors with multivariate logistic regression analysis. The receiver operating characteristic curve (ROC curve) was plotted to analyze the independent risk factors and evaluate their power of test.

RESULTS

1 185 patients with cerebral hemorrhage were enrolled in the study, 293 cases occurred SU, accounting for 24.7%, and 892 cases without SU, which accounted for 75.3%. As shown by univariate analysis, risk factors for cerebral hemorrhage complicated with SU included age, amount of bleeding, the bleeding site, disturbance of consciousness, APACHEII score, SBP. As to the site of bleeding, brain, thalamus, brainstem hemorrhage complicated with SU were higher proportion, 45.3% (43/95), 39.1% (63/161), 36.9% (48/130), which were significantly higher than those of the lobes of the brain [26.2% (33/126)], cerebellum [18.8% (15/80)], basal ganglia [16.1% (78/485)], arachnoid the inferior vena cava [12.0% (13/108)]. Multivariate logistic regression analysis showed that amount of bleeding [odds ratio (OR)=3.305, P=0.001, 95% confidence interval (95%CI) 2.213-48.634], the bleeding site (OR=1.762, P=0.008, 95%CI 0.123-2.743), SBP (OR=1.223, P=0.034, 95%CI 0.245-2.812) were independent risk factors of cerebral hemorrhage complicated with SU. The area under the ROC curve (AUC) of amount of bleeding and SBP were 0.846 and 0.597, suggesting that amount of bleeding has moderate diagnostic value and SBP has low diagnostic value.

CONCLUSIONS

Cerebral hemorrhage patients with large amount of bleeding, the bleeding site in the ventricle, thalamus or brainstem, high SBP are of great risk. We should lower blood pressure and give preventive treatment for SU as soon as possible.