Development and validation of a survival prediction model for patients received mechanical ventilation in the intensive care unit: a large sample size cohort from the MIMIC database

Abstract

Background: Mechanical ventilation remains one of the primary management measures for critically ill patients in intensive care units (ICUs). However, previous studies on the prognosis prediction of ICU patients received mechanical ventilation were limited. This study was to develop and validate a nomogram for predicting short-and long-term survival among patients who received mechanical ventilation in the ICU. Methods: This was a retrospective cohort study with a 3-year follow-up. Demographic, laboratory, clinical data of 16,775 participants aged ≥18 years who received mechanical ventilation in the ICU were extracted from the Medical Information Mart for Intensive Care III (MIMIC-III) database. The outcomes of this study were 1-month, 3-month, 1-year, and 3-year survival. All eligible patients were randomly classified into the training and testing groups with a ratio of 7:3. A multivariate Cox regression in the training group was used to explore the predictors and develop the predictive nomogram. Internal and subgroup validations were performed, and the C-index was calculated to estimate the predictive performance of the nomogram. The time-dependent receiver operating characteristic curves were drawn, and corresponding areas under the curve (AUC) were calculated. Results: Totally 6,291 patients died during the follow-up duration. Age, gender, ethnicity, ICU type, comorbidity, days of mechanical ventilation, white blood cell count, blood urea nitrogen, the fraction of inspiration O2, Sequential Organ Failure Assessment scores, and the Glasgow coma score were predictors of the survival of ICU patients who received mechanical ventilation (P<0.05). The C-index of the nomogram was 0.819 and was validated in the testing group at 0.816. The AUCs for the prognostic nomogram for 1-month, 3-month, 1-year, and 3-year survival were 0.889, 0.892, 0.882, and 0.866, respectively. Conclusions: This nomogram showed good predictive performance for short-and long-term survival in ICU patients treated with mechanical ventilation, which may be a useful tool for clinicians to assess the prognosis of patients and to adjust treatment strategies in time.