Construction and validation of a prediction model for the risk of citrate accumulation in patients with hepatic insufficiency receiving continuous renal replacement therapy with citrate anticoagulation.

BACKGROUND: To construct and validate a prediction model of the risk of citrate accumulation in patients with hepatic dysfunction receiving continuous renal replacement therapy with regional citrate anticoagulation (RCA-CRRT), which reduces the risk of citrate accumulation.

METHODS: All patients who received RCA-CRRT from 2021 to 2022 and were hospitalized in the First Affiliated Hospital of Zhejiang University were considered for study participation. Logistic regression analysis was used to identify the risk factors for citrate accumulation, based on which a nomogram model was constructed and validated in the validation group.

RESULTS: Six factors were finally identified, from which a nomogram was created to predict the risk of citrate accumulation. The area under the curve of the prediction model was 0.814 in the training group and 0.819 in the validation group, and the model showed acceptable agreement between the actual and predicted probabilities. Decision curve analysis also demonstrated that the model was clinically useful.

CONCLUSIONS: The model constructed from six factors reliably predicted the risk of citrate accumulation in patients with hepatic insufficiency who received RCA-CRRT.