Objective:To analyze the correlation between electromagnetic perturbation index and intracranial pressure (ICP) in patients with acute ischemic stroke (AIS) after mechanical thrombectomy, and to explore their values in early warning of decompressive craniectomy (DC).Methods:Forty-three patients with AIS after mechanical thrombectomy admitted to our hospital from January 1, 2018 to December 31, 2019 were enrolled in our prospective cohort study. Electromagnetic perturbation index and invasive ICP were continually monitored for 1-5 d in all patients 24 h after mechanical thrombectomy. According to the mean ICP on the first d, all patients were divided into normal ICP group (ICP<15 mmHg), mild increased ICP group (15≤ICP≤22 mmHg) and moderate to severe increased ICP group (ICP>22 mmHg). According to the implementation of DC, these patients were divided into decompressive craniectomy group and non-decompressive craniectomy group. The differences in clinical data of patients in different ICP groups were compared. The correlation between electromagnetic perturbation index and ICP was analyzed by Pearson correlation method. Receiver operating characteristic (ROC) curve was drawn to analyze the diagnostic value of electromagnetic perturbation index in increased ICP (ICP>22 mmHg). Multivariate Logistic regression analysis was used to determine the independent influencing factors for DC after mechanical thrombectomy in AIS patients. ROC curve was used to evaluate the diagnostic values of electromagnetic perturbation index and ICP in DC implementation after mechanical thrombectomy in AIS patients.Results:(1) In these 43 patients, 8 had normal ICP, 13 had mild increased ICP, and 22 had moderate to severe increased ICP. There were significant differences in baseline Glasgow Coma Scale (GCS) scores, baseline National Institutes of Health Stroke Scale (NIHSS) scores, baseline Alberta stroke program early CT scale (ASPECTS) scores, percentage of patients accepted DC, and electromagnetic perturbation index among the 3 groups ( P<0.05). Correlation analysis showed that electromagnetic perturbation index was negatively correlated with ICP ( r=-0.699, P=0.000). ROC curve showed that the area under curve (AUC) of electromagnetic perturbation index in diagnosing ICP>22 mmHg was 0.850 ( 95%CI: 0.690-1.000, P=0.000), enjoying the optimal cutoff value of 126. (2) Among the 43 patients, 27 were in the decompressive craniectomy group and 16 were in the non-decompressive craniectomy group. Multivariable Logistic regression analysis showed that baseline NIHSS scores, baseline ASPECTS scores, electromagnetic perturbation index, and ICP were independent risk factors for DC implementation after mechanical thrombectomy in AIS patients ( P<0.05). ROC curve showed that the AUC of ICP in predicting DC implementation after mechanical thrombectomy was 0.851 ( 95%CI: 0.728-0.973, P=0.000), enjoying the optimal cutoff value of 18.5 mmHg; the AUC of electromagnetic perturbation index in predicting DC implementation after mechanical thrombectomy was 0.764 ( 95%CI: 0.609-0.919, P=0.004), enjoying the optimal cutoff value of 137.5. Conclusion:There is a good correlation between electromagnetic perturbation index and ICP, which can be used as reference indexes for early warning of DC after mechanical thrombectomy in AIS patients.