Endocrine Abstracts

Background: Elevated reverse triiodothyronine (rT3) is a characteristic feature of euthyroid sick syndrome (ESS), which is associated with mortality in critical patients. However, the lack of reference values limits the use of rT3 for ESS diagnosis and the prognostic value of rT3 is unclear. This study aimed to determine the reference range of rT3 in the ESS identification and mortality prediction in critically ill patients.

Methods: This ongoing prospective cohort study was conducted from April 2022 at eight hospitals in China. Patients who had been hospitalized in intensive care unit for at least 24 hours without thyroid disorders were consecutively enrolled. Demographic characteristics, APACHE II, and thyroid function were evaluated on admission. Patients were followed up at the time of ICU discharge, hospital discharge, and 90 days after discharge. Study outcomes included ESS diagnosis (defined as decreased free triiodothyronine on admission without thyroid itself lesion), in-hospital death, and 90-day death. The 2.5th percentile (P_2.5) of rT3 in ESS patients was calculated as early warning. The optimal cut-off points of rT3 for the ESS diagnosis and mortality prediction were determined using receiver operating characteristic (ROC) curve and time-dependent ROC curve, respectively.

Results: This preliminary analysis presented the results of data collected up to January 2024. A total of 959 critical patients were screened, of which 849 (mean age 64.63 years, 3.25% male) were eligible and 34 were lost to follow-up after discharge. Of those, 394 (46.41%) had ESS on admission, 79 (9.31%) died during hospitalization, and 250 (30.67%) experienced 90-days deaths. The P_2.5 of rT3 levels in ESS patients was 0.57 ng/mL. The optimal cut-off point of rT3 for diagnosing ESS was 1.06 ng/mL, with an area under the curve of 0.745, sensitivity of 73.32%, and specificity of 67.13%. The optimal cut-off points for predicting in-hospital and 90-days mortality were 1.92 and 1.62 ng/mL, respectively. The log-rank test showed that patients with high rT3 level had higher in-hospital and 90-days mortality than those with low rT3 (P< 0.001). Sensitivity analysis (excluding patients discharged from ICU to home) did not significantly change the results.

Conclusion: In critically ill patients, rT3 had clinical value in identifying ESS and predicting mortality, which would contribute to risk stratification and care. The different levels of rT3 may have different clinical implications: 0.57 ng/mL for ESS early warning, 1.06 ng/mL as the optimal cut-off point for ESS diagnosis, and 1.62 ng/mL for mortality risk stratification.