Abstract
BACKGROUND: Chronic critical illness (CCI) can manifest as dysfunction of thyroid hormones. This study aimed to investigate the predictive value of the nonthyroidal illness syndrome (NTIS) for the prognosis of CCI patients, establish a predictive model for the prognosis of CCI patients, and evaluate the efficacy of the model to provide a theoretical basis for clinical intervention. METHODS: This was a prospective observational study in which patients ≥18 years old who met the CCI criteria were enrolled. The primary outcome of the study was 90-day mortality after intensive care unit (ICU) admission. A nomogram was constructed to predict the prognosis of CCI patients, and the model was evaluated via the concordance index, calibration curve and decision curve analysis. RESULTS: A total of 545 patients were included, and NTIS patients accounted for 65.3% of the patients. CCI patients with NTIS had more ventilator days and higher 90-day mortality. Lower free triiodothyronine (FT3) levels (<1.19 pmol/L) or reduced free thyroxine (FT(4)) levels (<9.655 pmol/L) were significantly associated with reduced survival in CCI patients with NTIS. Older age, a higher Sequential Organ Failure Assessment (SOFA) score, an emergency other than a traumatic operation, and a lower FT4 and thyroid-stimulating hormone level were found to be independent prognostic factors for a fatal outcome in CCI patients. The C-index for the prediction nomogram was 0.734, and the bias-corrected C-index was 0.727. The area under the receiver operating characteristic curve of our prediction model was superior to that of the SOFA and Acute Physiology and Chronic Health Evaluation II scores. CONCLUSIONS: Decreased serum FT3 and FT4 concentrations in patients with CCI at admission to the ICU on day 10 are associated with 90-day mortality. Early detection of serum FT3 and FT4 levels could help clinicians target CCI patients at high risk of clinical deterioration.