In the present study, out of the total sample, 28 patients (70%) had NTI patterns. The observed patterns included various combinations of low FT3, low FT4, and low TSH levels, with percentages ranging from 2.50% to 22.50%. Specifically, 7.50% of patients had low FT3, normal FT4, and normal TSH levels, while 22.50% had low FT3, low FT4, and normal TSH levels. In 5.00% of patients, all three levels (FT3, FT4, and TSH) were low. Other observed patterns included low FT3 with normal FT4 and low TSH (7.50%), normal FT3 with low FT4 and low TSH (2.50%), and normal FT3 with low FT4 and normal TSH (7.50%). Additionally, 15% of patients had normal FT3, normal FT4, and low TSH levels.

It is noteworthy that all participants in the study had increased levels of rT3, which indicates its potential as a sensitive and early indicator for detecting acute alterations in thyroid hormone metabolism. The short half-life of rT3, approximately 3 h, further enhances its effectiveness in promptly identifying these changes [28].

In the study by El-Ella et al. [12], NTI patterns were observed in 62.9% of critically ill children. These patterns included low FT3 only (26.9%); low FT3 and FT4 (15%); low FT3, FT4, and low TSH (1.9%); low FT3 and TSH (5.8%); low FT3 with high TSH (1.9%); low TSH only (1.9%); and low FT4 only (1.9%).

However, Den Brinker et al. [13] described NTI patterns in 69 children with meningococcal sepsis, including low serum T4 levels (80%), normal FT4 levels (89%), low T3 levels (100%), high rT3 levels (89%), and low T3/rT3 ratios (100%). None of the patients had elevated TSH levels. The divergence may be explained by differences in the characteristics of the study populations.

Consistent with our findings, Zargar et al. [29] analyzed circulating T3, T4, and TSH levels in 382 adult patients with NTI. Different patterns were observed at the onset of illness, as well as in the 3rd and 24th weeks. These patterns included low T3 (29.6%); low T3 and T4 (13.1%); low T3, T4, and TSH (2.6%); high T4 (7.3%); low TSH (6.8%); high TSH (6.5%); and low T4 (1%).

Overall, the comparison highlights the variations in NTI patterns across different patient populations and age groups. These differences may be attributed to the underlying conditions, severity of illness, and individual physiological responses. Therefore, it is important to consider these variations when interpreting and applying the findings to clinical practice.

These findings provide valuable insights into the prevalence and distribution of NTI patterns in both children and adults, highlighting the complexity of thyroid hormone alterations during non-thyroidal illness. This wide variability could be attributed to various factors. These factors include differences in the age of the patients being studied, the underlying critical illness, the size of the study sample, the technique used for measuring free T3 levels, as well as other factors such as ethnicity.

In the present study, an unexpected finding was the elevated FT4 levels despite low FT3 and low or normal TSH levels. This can be explained by the fact that even small amounts of heparin administration could lead to in vitro generation of free fatty acids during extended serum dialysis, resulting in falsely increased apparent free hormone levels [30]. This is a significant issue as heparin is commonly used for thrombosis prevention in intensive care units, suggesting that this problem may be widespread and severe. Another possible explanation is that the decrease in total T3 levels is more significant than that of total T4 levels. Additionally, factors that impact thyroid hormone binding are more likely to affect T4 assays, as T4 is tightly bound to thyroxine-binding globulin compared to T3 [31].

Zargar et al. [29] concluded that despite clinical improvement in the majority of patients with NTI patterns by the third week, there was a continued decline in T3 and T4 levels. It was only after six months that significant evidence of recovery could be observed. Even at this point, although mean T4 levels were comparable to control values, T3 levels remained lower compared to controls.

In the present study, the median length of stay (LOS) was found to be 10.5 days. After the reversal of shock, the follow-up period varied for each patient and was not a fixed point in time. A third sample was taken after five days. The authors did not anticipate a complete recovery of thyroid hormone levels, but rather looked for a reassuring trend indicating a pattern of recovery.

By comparing the mean of thyroid hormones between the three samples, A, B and C, we concluded that mean T4 levels in sample B were significantly higher than sample A, and that of sample C was significantly higher than sample A. Furthermore, mean FT3 and FT4 levels were significantly higher in sample C than in sample A. Also, the mean of TSH in sample C was significantly higher than in sample B and sample A.

In our study, we observed a dynamic pattern of FT3 levels during the course of shock reversal and recovery in pediatric patients. These findings shed light on the intricate relationship between thyroid function and critical illness and provide valuable insights for clinicians and researchers.

The present study reported an increase in the percentage of patients maintaining normal FT3 levels after five days, highlighting the potential for recovery of thyroid function in the post-shock phase. Additionally, we noted a subset of patients transitioning from normal to low FT3 levels on shock reversal. While this transition was relatively small, it merits attention, as it could be indicative of thyroid dysfunction associated with the critical state. Further investigation is needed to determine the clinical implications of this transition.

Furthermore, among patients initially presenting with low FT3 levels, we observed variations in FT3 levels on shock reversal and five days afterward. Notably, a proportion of these patients experienced an increase in FT3 levels to the normal range five days after shock reversal. This intriguing finding suggests that thyroid function recovery is possible in some pediatric patients following the resolution of shock.

The PIM2 score showed significant negative correlations with FT3 and FT3on admission, as well as with T4, FT3, and FT4 on shock reversal. BUN levels were significantly negatively correlated with both T3 and FT3 on shock reversal, as well as with T3, T4, and FT3 five days after shock reversal. These findings suggest that higher PIM2 scores, indicating greater severity of illness, were associated with lower thyroid hormone levels, emphasizing the potential impact of critical illness on thyroid function.

It is worth mentioning that this study did not include patients with overt renal impairment. Still, since most of our study population suffered from septic shock or sepsis, their renal functions inevitably deteriorated. Sepsis is the most common precipitant for acute kidney injury in adults and children, and the progression of kidney disease in sepsis is a poor prognostic sign [32].

All patients in the study required vasoactive medication to maintain hemodynamic stability. Dopamine was not administered during their stay in the PICU. The inotropes used were dobutamine, noradrenaline, adrenaline, and milrinone. Among the patients, 22 (55%) received a single inotrope, 11 (27.5%) received a combination of two inotropes, and seven (17.5%) received a combination of three inotropes. Additionally, 22 patients received noradrenaline alone or in combination with another vasoactive medication.

The effects of these medications on thyroid hormones were evaluated in samples A, B, and C. It was found that dopamine and glucocorticoids suppress TSH secretion, while phenytoin and phenobarbital increase the hepatic metabolism of thyroid hormones, leading to decreased serum levels. Amiodarone and beta-adrenergic blocking agents inhibit deiodinase activity, thereby reducing T3 production. On the other hand, furosemide, heparin, and non-steroidal anti-inflammatory drugs transiently increase free thyroid hormone levels by inhibiting their binding to plasma transport proteins [33].

No significant correlations were observed between the thyroid hormones and the duration of medications administered to the study population during their PICU stay, CRP levels, WBC count, LOS, or the time until shock reversal.

However, El-Ella et al. [12] found a significant negative correlation between TSH and the LOS (r =  − 0.35; p = 0.011); they also found significant negative correlation between WBCs with both FT3 (r =  − 0.36; p = 0.002) and FT4 (r =  − 0.34; p = 0.005). However, no significant correlations were observed between thyroid hormones and CRP levels.

There was no significant correlation between the patients’ NTI patterns and either the LOS or the duration of shock reversal. On the other hand, den Brinker et al. [14] reported that T4 levels were an independent predictor of an unfavorable outcome, specifically a prolonged ICU stay.

A systematic review conducted by Angelousi et al. [11] reported that higher TSH levels, lower rT3 levels, or lower T3 and T4 levels were associated with an unfavorable outcome in patients with sepsis or septic shock. This suggests that, in some patients, thyroid hypofunction during sepsis or septic shock may independently influence the outcome of these conditions. Further studies are needed to evaluate this hypothesis in appropriately designed studies.

It is important to note that thyroid hormone assessments, like many other laboratory tests, may have a turnaround time that extends beyond the acute management phase. While our study focused on the potential clinical relevance of thyroid hormone assessments in pediatric shock patients, we acknowledge that these tests may not be immediately actionable during the acute phase of management. Therefore, their utility as real-time guidance for acute interventions may be limited.

The current study has several limitations, including being a single-center study with a small sample size. Conducting a larger multi-center study would provide stronger and more generalizable results. The study was also limited by the duration of patients; stay in the PICU, and a longer follow-up period would have provided more meaningful insights. Additionally, the exclusion of deceased subjects who did not survive to the second or third sample collection limits the understanding of mortality outcomes in relation to NTI. Furthermore, the study did not measure cortisol and cytokine values, which are important factors in understanding the underlying mechanisms of NTI. The lack of antibody testing in the assessment of thyroid function further limits the interpretation of these findings, especially in distinguishing between non-thyroidal illness and underlying thyroid dysfunction. These limitations highlight the need for further research to establish evidence-based guidelines for managing pediatric NTI in the PICU and to improve our understanding and management of thyroid dysfunction in critically ill children.