A novel clinical perspective on subarachnoid hemorrhage complicated by myocardial injury: independent predictors and short-term prognostic assessment

Figures

• 

  Figure 1. Flowchart of patient inclusion and exclusion. Patients with spontaneous subarachnoid hemorrhage screened between 2022 and 2024 were evaluated according to the predefined inclusion and exclusion criteria, including exclusion of pregnant patients if present. After applying the exclusion criteria, 208 patients who underwent endovascular aneurysm embolization within 24–48 h were included in the final analysis.

• 

  Figure 2. Exploratory nomogram for predicting short-term neurological prognosis at hospital discharge in patients with aSAH. The model incorporates age, hydrocephalus, GCS score, and hsCRP and requires external validation before routine clinical use.

• 

  Figure 3. Exploratory nomogram for predicting myocardial injury risk in patients with aSAH. The model incorporates NPE, PCT, and RONSD and is intended for preliminary risk stratification pending external validation.

• 

  Figure 4. Calibration curve for myocardial injury risk prediction nomogram. Calibration curve assesses agreement of predicted/observed probabilities of myocardial injury in SAH patients diagnosed with myocardial injury according to nomogram. The solid line represents bias-corrected estimates; dotted line represents apparent performance; dashed diagonal line represents perfect calibration. Close alignment of solid curve with dashed diagonal indicates good calibration/reliable model. Mean absolute error = 0.036 indicates there is very little deviation between predicted and observed outcomes.

• 

  Figure 5. It depicts the calibration curve for predicting short-term prognosis based on the nomogram in patients with SAH. According to this graph, there appears to be good agreement between the predicted risk and the actual risk (measured outcomes) for this population. The bias-corrected curve closely approximated the ideal reference line, indicating excellent calibration and validity of the model. The mean absolute error was found to be 0.016 based on the bootstrap sample of the data (n) of 208 resamples obtained from the sample of 1,000.

• 

  Figure 6. It depicts the decision curve analysis (DCA) performed on the myocardial injury risk prediction nomogram developed for patients with SAH. The DCA illustrates the standardized net benefit (SNB) of the myocardial injury risk prediction nomogram based on a variety of threshold probabilities (0–1) for use when predicting myocardial injury. The myocardial injury risk prediction nomogram demonstrated a net benefit greater than the treat-all and treat-none strategies at threshold probabilities approximately between 0.10 to 0.75. Therefore, these results suggest that the myocardial injury risk prediction nomogram has significant clinical utility in assisting healthcare providers in assessing risk and making decisions regarding patients with SAH.

• 

  Figure 7. The short-term prognosis nomogram was used to assess decision curves (DCA) of primary outcome data in subarachnoid hemorrhage patients to identify potential clinical utility for guided systematic risk assessments and/or treatment selections. The short-term prognosis nomogram provided greater than expected “net” benefit relative to both treat-all and treat-none based treatment selection strategies over a wide range of threshold probabilities (from as low as 0.10 to as high as 0.90).

• 

  Figure 8. The myocardial injury prediction multivariable logistic regression model was found to possess good discrimination ability (AUC = 0.823, 95% CI = 0.748–0.898) as shown by the area under the receiver operating characteristic (ROC) curve.

• 

  Figure 9. The poor prognosis prediction multivariable logistic regression model exhibited excellent discrimination ability (AUC = 0.940, 95% CI = 0.907–0.974) as presented by the area under the receiver operating characteristic (ROC) curve.