The Accuracy of Serum Matrix Metalloproteinase-9 for Predicting Hemorrhagic Transformation After Acute Ischemic Stroke: A Systematic Review and Meta-Analysis

Background Hemorrhagic transformation is a serious complication of acute ischemic stroke, which may cause detrimental outcomes and the delayed use of anticoagulation therapy. Early predicting and identifying the patients at high risk of hemorrhagic transformation before clinical deterioration occurrence become a research priority. Objective To study the value of plasma matrix metalloproteinase-9 predicting hemorrhagic transformation after ischemic stroke. Methods We searched PubMed, Ovid, Cochrane Library, and other 2 Chinese databases to identify literatures published up to September 2017 and performed meta-analysis by STATA (version 12.0, StataCorp LP, College Station, TX). Results Twelve studies incorporating 1492 participants were included and 7 studies were included in the quantitative statistical analysis. The pooled sensitivity was 85% (95% confidence interval [CI]: 75%, 91%) and the pooled specificity was 79% (95% CI: 67%, 87%). The area under the receiver operating characteristic curve was .89 (95% CI .86, .91). Significant heterogeneity for all estimates value existed (all the P value < .05 and I2 > 50%). There is no threshold effect with P value greater than .05 of the correlation coefficient. Meta-regression and subgroup analysis showed cut-off value and hemorrhagic subtype contributed to heterogeneity. Deeks' funnel plot indicated no significant publication bias for 7 quantitative analysis studies. Conclusions Matrix metalloproteinase-9 has high predictive value for hemorrhagic transformation after acute ischemic stroke. It may be useful to test matrix metalloproteinase-9 to exclude patients at low risk of hemorrhage for precise treatment in the future clinical work. Hemorrhagic transformation is a serious complication of acute ischemic stroke, which may cause detrimental outcomes and the delayed use of anticoagulation therapy. Early predicting and identifying the patients at high risk of hemorrhagic transformation before clinical deterioration occurrence become a research priority. To study the value of plasma matrix metalloproteinase-9 predicting hemorrhagic transformation after ischemic stroke. We searched PubMed, Ovid, Cochrane Library, and other 2 Chinese databases to identify literatures published up to September 2017 and performed meta-analysis by STATA (version 12.0, StataCorp LP, College Station, TX). Twelve studies incorporating 1492 participants were included and 7 studies were included in the quantitative statistical analysis. The pooled sensitivity was 85% (95% confidence interval [CI]: 75%, 91%) and the pooled specificity was 79% (95% CI: 67%, 87%). The area under the receiver operating characteristic curve was .89 (95% CI .86, .91). Significant heterogeneity for all estimates value existed (all the P value < .05 and I2 > 50%). There is no threshold effect with P value greater than .05 of the correlation coefficient. Meta-regression and subgroup analysis showed cut-off value and hemorrhagic subtype contributed to heterogeneity. Deeks' funnel plot indicated no significant publication bias for 7 quantitative analysis studies. Matrix metalloproteinase-9 has high predictive value for hemorrhagic transformation after acute ischemic stroke. It may be useful to test matrix metalloproteinase-9 to exclude patients at low risk of hemorrhage for precise treatment in the future clinical work.