Objective To explore the value of a Nomogram model in predicting the death risk of patients with severe thrombocytopenia.

Methods The clinical materials of 340 patients with severe thrombocytopenia diagnosed in authors' hospital from May 2020 to May 2022 were retrospectively summarized, and they were randomly divided into model group (n=238) and validation group (n=102) according to the ratio of 7 to 3. The clinical materials of death and survival patients in the model group were analyzed by single factor and multiple factor Logistic regression to screen the main risk factors, and R software was used to construct the Nomogram model.

Results The mortality in the model group was 34.0% (81/238), which showed no significant difference when compared to 29.4% (30/102) in the validation group (P=0.405). Multivariate Logistic regression analysis showed that cerebrovascular diseases (OR=1.986, 95%CI, 1.524 to 2.659, P < 0.001), malignant tumors (OR=2.056, 95%CI, 1.744 to 2.789, P < 0.001), mechanical ventilation (OR=2.324, 95%CI, 1.856 to 3.121, P < 0.001), vasopressors (OR=2.759, 95%CI, 2.425 to 3.562, P < 0.001), continuous renal replacement therapy (OR=2.421, 95%CI, 2.012 to 3.123, P < 0.001) and prolonged coagulation time (OR=1.649, 95%CI, 1.232 to 2.011, P < 0.001) were the independent risk factors of death in patients with severe thrombocytopenia. The total score of the Nomogram was 240 points, and the C-index values of the model group and the validation group calculated by Bootstrap method were 0.912 and 0.879 respectively, indicating that the predictive efficiency of the model was good. The calibration curve showed that the predictive probabilities of the model group and the validation group were basically consistent with the measured values. The area under the curve (AUC) calculated by receiver operating characteristic (ROC) curve of the model group and the validation group were 0.889 and 0.856 respectively, indicating that the predictive accuracy was high. Compared with the traditional Sequential Organ Failure Assessment (SOFA) score and the Simplified Acute Physiology Score Ⅱ (SAPS Ⅱ) score, the AUC value of the Nomogram was significantly larger (P < 0.001). The decision curve showed that the clinical net benefit ratios of the model group and the validation group were better.

Conclusion Severe thrombocytopenia has a high risk of death. Application of the Nomogram model can better guide clinicians to identify people with high risk of death in the early stage and actively implement intervention for improvement of the prognosis.