Gastrointestinal Hemorrhage is Associated with Mortality after Acute Ischemic Stroke.

OBJECTIVE

Gastrointestinal (GI) hemorrhage is serious during the acute phase and is reported to be related to an increased risk of death during the acute phase of acute ischemic stroke in particular. Our study was designed to investigate the relationship between GI hemorrhage and the mortality of acute ischemic stroke, assessing the influence of cerebrovascular risk factors, brain herniation and oral anticoagulation on the onset of GI hemorrhage. The identified risk factors for the occurrence of GI hemorrhage help to elucidate their respective roles in the mortality of acute ischemic stroke.

METHODS

A total of 15993 consecutive patients with acute ischemic stroke, including 216 cases and 15777 controls, were enrolled in the study from October 2010 to December 2018. Basic clinical and examination data were collected at the time of study enrollment. GI hemorrhage was diagnosed according to the presence of clinical features and endoscopy. Chi-square test and multiple logistic regressions were conducted to explore the associations between the GI hemorrhage occurrence and known risk factors. Kaplan-Meier was used to assess the influence of GI hemorrhage on the age of mortality of acute ischemic stroke.

RESULTS

GI hemorrhage cases among patients with acute ischemic stroke accounted for 1.35%. Male patients with ischemic stroke were more likely to have GI hemorrhage than their female counterparts (odds ratio (OR): 1.79; P = 0.000). Patients with atrial fibrillation (AF) had a higher incidence of GI hemorrhage than their counterparts without AF (3.03% vs. 1.20%; P < 0.05). Use of oral anticoagulants was related to increased risk for GI hemorrhage (OR: 1.96; P = 0.00). After adjusting for age and sex, both AF and oral anticoagulant use maintained associations with increased risk for GI hemorrhage (2.59-times and 2.02-times risk respectively; P = 0.00). Patients with hyperlipidemia had a lower incidence of GI hemorrhage than their counterparts without hyperlipidemia (0.62% vs. 1.60%; P < 0.05). Hyperlipidemia was associated with a reduced risk of GI hemorrhage (OR: 0.38, 95% confidence interval (CI): 0.25-0.58; P = 0.00), even after adjusting for age and sex (OR: 0.41; P = 0.00). Patients with brain herniation had a 6.54-times increased risk for GI hemorrhage (P = 0.00). GI hemorrhage was associated with 10.98-fold risk for mortality of acute ischemic stroke (P = 0.00). There was an interaction between GI hemorrhage and brain herniation and increased 26.91-fold risk for the mortality after acute ischemic stroke (P = 0.00).

CONCLUSION

AF, oral anticoagulant use, brain herniation and male sex increase GI hemorrhage risk, while hyperlipidemia reduces risk. GI hemorrhage itself increases the risk for mortality of acute ischemic stroke. The interaction between GI hemorrhage and brain herniation increased the risk for the mortality after acute ischemic stroke.