Background: Mechanical thrombectomy (MT) is an effective treatment for large-vessel occlusion in acute ischemic stroke, however, only some revascularized patients have a good prognosis. For stroke patients undergoing MT, predicting the risk of unfavorable outcomes and adjusting the treatment strategies accordingly can greatly improve prognosis. Therefore, we aimed to develop and validate a nomogram that can predict 3-month unfavorable outcomes for individual stroke patient treated with MT.
Methods: We analyzed 258 patients with acute ischemic stroke who underwent MT from January 2018 to February 2021. The primary outcome was a 3-month unfavorable outcome, assessed using the modified Rankin Scale (mRS), 3-6. A nomogram was generated based on a multivariable logistic model. We used the area under the receiver-operating characteristic curve to evaluate the discriminative performance and used the calibration curve and Spiegelhalter's Z-test to assess the calibration performance of the risk prediction model.
Results: In our visual nomogram, gender (odds ratio [OR], 3.40; 95%CI, 1.54-7.54), collateral circulation (OR, 0.46; 95%CI, 0.28-0.76), postoperative mTICI (OR, 0.06; 95%CI, 0.01-0.50), stroke-associated pneumonia (OR, 5.76; 95%CI, 2.79-11.87), preoperative Na (OR, 0.82; 95%CI, 0.72-0.92) and creatinine (OR, 1.02; 95%CI, 1.01-1.03) remained independent predictors of 3-month unfavorable outcomes in stroke patients treated with MT. The area under the nomogram curve was 0.8791 with good calibration performance (P = 0.873 for the Spiegelhalter's Z-test).
Conclusions: A novel nomogram consisting of gender, collateral circulation, postoperative mTICI, stroke-associated pneumonia, preoperative Na and creatinine can predict the 3-month unfavorable outcomes in stroke patients treated with MT.
Keywords: Mechanical Thrombectomy; Nomogram; Prediction; Stroke; Unfavorable outcome.
© 2022. The Author(s).