Development and validation of a prognostic nomogram for predicting poor outcomes following intravenous rt-PA in patients with acute ischemic stroke

TL;DR

This study developed a tool to predict which stroke patients will have poor outcomes after rt-PA treatment, using factors like blood clotting levels and medical history.

Contribution

A new nomogram was developed and validated to predict poor outcomes in rt-PA-treated stroke patients using baseline NIHSS, fibrinogen, and hypertension history.

Findings

The nomogram achieved high accuracy with AUCs of 0.948 in training and 0.959 in validation datasets.

Fibrinogen, baseline NIHSS, and hypertension history were identified as independent predictors of poor outcomes.

Calibration and decision curve analyses confirmed the model's reliability and clinical utility.

Abstract

Intravenous administration of recombinant tissue plasminogen activator (rt-PA) within 4.5 h of symptom onset is a standard treatment for acute ischemic stroke (AIS). However, certain patients continue to develop unfavorable outcomes despite timely rt-PA therapy. Identifying those at high risk is essential for developing individualized care plans and establishing appropriate follow-up.

This retrospective study included AIS patients treated with intravenous rt-PA at 0.9 mg/kg at our center. Outcomes at three months were evaluated using the modified Rankin Scale (mRS). Patients with mRS scores ≤2 were considered to have favorable outcomes, and those with scores >2 were considered to have poor outcomes. Univariable analysis and stepwise logistic regression were used to identify independent predictors of poor prognosis, and a nomogram was subsequently developed. The model’s discriminative power was assessed with area under the receiver operating characteristic curves (AUC-ROC), and its calibration was examined using calibration plots. Decision curves and clinical impact curves were applied to determine clinical utility.

Among 392 enrolled patients, 77 had poor outcomes three months after rt-PA therapy. Fibrinogen (Fg), baseline NIHSS, and a history of hypertension emerged as independent predictors of poor prognosis. The nomogram achieved an AUC of 0.948 (95% CI [0.910–0.985]), with sensitivity of 0.900 and specificity of 0.916 in the training dataset, and an AUC of 0.959 (95% CI [0.907–1.000]), with sensitivity of 0.943 and specificity of 0.947 in the validation dataset. Calibration plots demonstrated close agreement between predicted and observed probabilities, and decision curves indicated a wide range of net benefit threshold probabilities.

This nomogram, incorporating baseline NIHSS, Fg, and a history of hypertension, accurately predicts poor three-month outcomes in AIS patients treated with intravenous rt-PA. Its ease of use may facilitate early risk stratification and assist clinicians in formulating more targeted management strategies and follow-up protocols for patients likely to experience unfavorable outcomes.

Figures