The relationship between NTIS, thyroid hormones and outcomes in critically ill patients
| Authors | Country | Subjects | Age | Main Findings |
|---|---|---|---|---|
| Wang et al.[10] | Shanghai, China | 480 ICU patients | 71.71 years (mean) | T3, T4, FT3, FT4, TSH levels were lower in non-survive compare to survive patients |
| FT3 (cut-off <3.33 pmol/L) has the largest AUC (0.762) to predict mortality (specificity = 0.78, sensitivity = 0.62) | ||||
| Praven et al.[11] | Hyderabad, India | 119 critically ill patients | 60.15 years (mean) | NTI prevalence was lower in survive (63%) compared to non-survive (93%) patients |
| T3 and TSH levels were lower in non-survive compare to survive patients | ||||
| T3 has the largest AUC (0.677) to predict mortality (specificity = 0.56, sensitivity = 0.83) | ||||
| Sayarifard et al.[12] | Tehran, Iran | 35 children with critical illness | 4 months to 15 years | T3 levels were lower among non-survive patients in the 1st day of hospitalization while T4 levels were lower in the 3rd day |
| Carreras et al.[13] | Spain | 103 PICU patients | median 8.51 years (higher mortality risk score group) and 5.78 years (lower mortality risk score group) | FT4 (cut-off <16.60 pmol/L) has the largest AUC (0.655) to predict higher mortality risk (specificity = 0.615, sensitivity = 0.76) |
| Subjects with NTIS had 6.04 times higher mortality risk while those with FT4<16.60 pmol/L had 4.92 times higher mortality risk | ||||
| Wang et al.[14] | Wuxi, China | 51 ICU patients | 15 to 88 years | APACHE II score has positive correlation with rT3 (r = 0.379) and negative correlation with TSH (r =−0.256), T3 (r =−0.370), T4 (r =−0.364) (all p≤0.05) |
| Krug et al.[15] | Leipzig, German | 1790 ICU patients | 44–77 years (interquartile range) | TSH, FT3, and FT4 levels were lower in NTIS compared to non-NTIS |
| NTIS condition was associated with longer ICU length of stay but not associated with higher ICU mortality rate | ||||
| Vidart et al.[16] | 6869 adult with critical illness (meta-analysis of 25 studies) | Subjects with NTIS had 2.21 times higher mortality risk compared to non-NTIS | ||
The association between NTIS, thyroid hormones and outcomes in patients with infection and sepsis
| Authors | Country | Subjects | Age | Main Findings |
|---|---|---|---|---|
| Yao et al.[17] | Shanghai, China | 1219 septic patients, 318 with DIC, 831 without DIC | 70 years (mean) | Sepsis patients with NTIS had 3.19 times higher risk to suffer DIC. The higher the DIC score the higher NTIS prevalence |
| Abdelgawad et al.[18] | Cairo, Egypt | 40 children with sepsis and septic shock | 6 to 120 months | NTI prevalence was lower in sepsis (30%) compared to septic shock (65.5%) patients. |
| FT3 levels were higher in 5th day, rT3 levels were lower in 1st and 5th day in survive compared to non-survive patients | ||||
| Sharma et al.[19] | New Delhi, India | 40 neonates patients with sepsis, 40 control subjects | neonates | FT4 and FT3 levels were lower in septic patients, and also lower in non-survive compared to survive patients |
| The higher the CRP levels the lower the FT3 levels in survive-septic shock and non-survive patients | ||||
| Lee et al.[20] | Seoul, South Korea | 213 ICU patients | mean 58.9 years (euthyroid group), 69.1 years (mild NTIS), 66.9 years (moderate NTIS), 64 years (severe NTIS) | Subjects with infection had higher prevalence of moderate (39.8% vs 30.7%) and severe NTIS (8% vs 0%) compared to non infection patients. Subjects with moderate to severe NTIS had 3.1 times higher mortality risk compared to normal-mild NTIS |
Changes in thyroid hormone patterns in NTIS
| Thyroid Hormones | Acute Phase of Critical Illness | Chronic Phase of Critical Illness | Recovery Phase of Critical Illness |
|---|---|---|---|
| TSH | N or slightly ↓ | N or slightly ↓ | N or slightly ↑ |
| FT4 | N or slightly ↑ | N or slightly ↑ | mostly N |
| FT3 | N or gradual ↓ | ↓↓ | gradual ↑ to baseline |
| T3 | N or gradual ↓ | ↓↓ | gradual ↑ to baseline |
| T4 | N or slightly ↓ | ↓↓ | gradual ↑ to baseline |
| rT3 | N to gradual ↑ | ↑↑ | gradual ↓ to baseline |
The relationship between NTI, thyroid hormones and outcomes in patients with non-infectious diseases
| Authors | Country | Subjects | Age | Main Findings |
|---|---|---|---|---|
| Cardiovasular Disease | ||||
| Adawiyah et al.[22] | Kuala Lumpur, Malaysia | 85 ACS patients, 6-month cohort study | 58.3 years (mean) | NTIS prevalence was 56% in ST elevation myocardial infarction, 54% in non-ST elevation myocardial infarction, and 48% in unstable angina |
| Peak Troponin T levels was associated with FT3 levels (r =−0.22, p = 0.049) | ||||
| Dal et al.[23] | Istanbul, Turkey | 70 acute myocardial infarction patients | 64.46 years (mean) | Non-survive acute myocardial infarction patients had lower FT3 levels compared to survive ones |
| Zhang et al.[24] | Shanghai, China | 501 acute myocardial infarction patients, cohort study (mean follow up 10±2 months) | 69 years (mean) | Mortality rate in low FT3 subjects were higher than normal FT3 (14 vs 2.7%). FT3 level was determinant factor of mortality and MACE (HR = 0.142 and 0.748, p<0.05 respectively) during follow up |
| Okayama et al.[25] | Tokyo, Japan | 270 ADHF patients | mean 65 years (normal FT3 group) and 72.3 years (low FT3 group) | FT3 (cut-off <2.05 pg/mL) has the largest AUC (0.791) to predict mortality risk (specificity = 0.72, sensitivity = 0.85) |
| Zhao et al.[26] | Gansu, China | 594 ADHF patients with euthyroid | 57 years (mean) | Low FT3 level was not associated with mortality during hospitalization but associated with 1 year mortality (HR = 1.85) |
| Hayashi et al.[27] | Japan | 274 ADHF patients | 70 years (mean) | Subclinical hypothyroidism (HR = 2.31) but not low FT3, associated with cardiovascular events |
| Kidney Disease | ||||
| Obasuyi et al.[28] | Benin, Nigeria | 184 CKD subjects, 80 controls | 5 to 90 years | 42.4% CKD patients had NTIS (54.5% of stage 3; 51.1% of stage 4; and 37.2% of stage 5 CKD patients) |
| Yuasa et al.[29] | Tokyo, Japan | 510 CKD subjects | 67 years (median) | 33.1% of CKD patients had NTIS. Urine protein, eGFR, and age were determeninat factor of NTIS |
| Li et al.[30] | Chengdu, China | 384 nephrotic syndrome patients | 41.3 years (mean, normal thyroid subjects); 40.65 years (mean, thyroid dysfunction subjects) | NTIS prevalence ranged from 31% to 66.66% among different nephrotic syndrome types. Blood creatinine, cholesterol, platelet, hemoglobin, albumine, and urine protein are determinant factors of thyroid dysfunction |
| Other Diseases | ||||
| Yasar et al.[31] | Turkiye | 125 COPD subjects | 65 years (mean) | NTIS prevalence in COPD patients was 53.8%. NTIS was linked to higher APACHE II scores (29 in NTIS vs 24 in non-NTIS) |
| Langer et al.[32] | Germany | 437 liver cirrhosis patients | 52.1 to 55.8 years (mean, among liver cirrhosis type) | NTIS prevalence was 72.1% in ACLF and 39.3% in AD patients. Subjects with low FT3 had higher 3-month mortality rate |
| Deng et al.[33] | Guangzhou, China | 396 diabetic ketoasidosis and ketosis patients | 57.5 years (mean) | NTIS prevalence was 57.8% among diabetic ketoasidosis and ketosis patients. FT3 had negative association with blood hsCRP, urine albumine levels, and leukocytes count |
| Zhang et al.[34] | Nanjing, China | 223 SLE patients | 36.7 years (mean, subjects with NTIS); 36.9 years (subjects without NTIS) | Prevalence NTIS in SLE patients was 58.74%. FT3 had negative association with SLE activity (r =−0.313), creatinine (r =−0.298), blood urea nitrogen (r =−0.325), CRP (r =−0.200) and urine protein (r =−0.301) among NTIS patients |