Background: Since there are many controversies about the incidence and prevalence of thyroid hormone abnormalities in critically-ill patients, and because of lack of such information in our patients, this study was performed to determine the frequency of these abnormalities in patients admitted to Intensive Care Unit of Kashan Medical University hospitals in the second half of 2003.

Materials and Methods: This descriptive study (with new data)was conducted on patients with severe illness admitted in Intensive Care Unit. The information on age, sex, primary diagnosis, vital signs was obtained. A sample of blood to assess thyroid hormone was taken too. Another sample to assess thyroid hormone was taken 8 weeks later.

Results: The population under study included 99 cases with mean age of 58.04 ± 21 years, 61.6% of which were males and 49.5% of patients died in the course of admission. The prevalence of reduction in T4-T3-TSH-T3Ru on the beginning of admission was 22.2%, 31.3%, 29.3% and 22.2% respectively. The most common abnormality was decreased in T3 level and 71% of patients with this reduction died. Decrease related to T4 was 81.8% and related to simultaneous decreased of both hormones was 91.7%. The mean level of T4 was 76.10 ± 37.60 (nmol/lit) in died patients and 100.48 ± 37.84 (nmol/lit) in recovered patients (p<0.002). The measurement about T3 in died and recovered patients was 0.96 ± 0.58 (ng/ml) and 1.33 ± 0.65 (ng/ml) respectively (p<0.004). The mean of hormonal level in 50 recovered patients was 0.96 ± 0.58 (ng/ml) and 1.33 ± 0.65 (ng/ml) respectively (p<0.004). The prevalence of NTI according to hormonal assessment 8 week after discharge-related to T3,T4,T₃RU and TSH was 88.9, 66.7, 100, 0 percents respectively.

Conclusion: Our research showed that hormonal change in Nonthyroid illness (NTI) in intensive care unit is common, and has prognostic value. Therefore, recognition of this abnormality in critically-ill patients is needed. Experimental clinical trials are recommended to choose the appropriate approach.