Nomogram Prediction of Prognosis After Surgical Operation for Cerebral Hemorrhage

Objective

This study aimed to investigate the risk factors for intensive care unit (ICU) mortality in patients with intracerebral hemorrhage after surgery and to construct a clinical nomogram.

Methods

The data in this retrospetive analysis were acquired from the Medical Information Mart for Intensive Care IV database, and the study controls were randomly divided into training and validation subsets in a ratio of 7:3. The primary clinical endpoint was all-cause ICU mortality. The prediction model was developed and a nomogram was generated based on findings of the logistic regression and least absolute shrinkage and selection operator regression analyses. Receiver operating characteristic curve was employed to assess model performance, and decision curve analysis was used to assess the clinical utility of the nomogram.

Results

This retrospective study comprised 859 participants, of whom 757 were survivors and 102 were nonsurvivors. The results showed that red cell distribution width (P = 0.014), Glucose (P < 0.001), mechanical ventilation ≥48 hours (P < 0.001), acute respiratory failure (P = 0.019), and Sequential Organ Failure Assessment (P = 0.017) were independent risk factors for death after intracerebral hemorrhage surgery. The results of the nomogram showed that blood glucose and red cell distribution width had the greatest impact on prognosis. The nomogram demonstrated strong discriminating for all-cause mortality in the ICU and showed a positive net benefit across a broad spectrum of threshold probabilities.

Conclusions

For patients with severe cerebral hemorrhage after craniotomy, we developed a distinctive nomogram model to forecast all-cause mortality in the critical care unit. It can simply and intuitively display the risk of poor prognosis for patients, providing clinicians with an important treatment tool for individualized treatment and outcome forecasting.