help button home button Endocrine Society JCEM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Journal of Clinical Endocrinology & Metabolism , doi:10.1210/jc.2005-0888
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by den Brinker, M.
Right arrow Articles by Hokken-Koelega, A. C. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by den Brinker, M.
Right arrow Articles by Hokken-Koelega, A. C. S.
Related Collections
Right arrow Thyroid
The Journal of Clinical Endocrinology & Metabolism Vol. 90, No. 10 5613-5620
Copyright © 2005 by The Endocrine Society

Euthyroid Sick Syndrome in Meningococcal Sepsis: The Impact of Peripheral Thyroid Hormone Metabolism and Binding Proteins

Marieke den Brinker, Koen F. M. Joosten, Theo J. Visser, Wim C. J. Hop, Yolanda B. de Rijke, Jan A. Hazelzet, Venje H. Boonstra and Anita C. S. Hokken-Koelega

Department of Pediatrics, Divisions of Endocrinology (M.d.B., V.H.B., A.C.S.H.-K.) and Pediatric Intensive Care (M.d.B., K.F.M.J., J.A.H.), Erasmus MC-Sophia Children’s Hospital, and Departments of Internal Medicine (T.J.V., Y.B.d.R.) and Epidemiology and Biostatistics (W.C.J.H.), Erasmus MC, 3000 CB Rotterdam, The Netherlands

Address all correspondence and requests for reprints to: Drs. Marieke den Brinker or Anita C. S. Hokken-Koelega, Erasmus MC-Sophia Children’s Hospital, Department of Pediatrics, Divisions of Endocrinology and Pediatric Intensive Care, Room SP-3435, P.O. Box 2060, 3000 CB Rotterdam, The Netherlands. E-mail: m.denbrinker{at}erasmusmc.nl or a.hokken{at}erasmusmc.nl.


   Abstract
Top
 Abstract
Introduction
 Patients and Methods
 Results
 Discussion
 References
 
Context and Objectives: The objective of this study was to elucidate the influence of disease severity, deiodination, sulfation, thyroid hormone binding, and dopamine use on thyroid function in euthyroid sick syndrome.

Setting: The study was performed at a university-affiliated pediatric intensive care unit (PICU).

Design: This was an observational cohort study.

Patients: Sixty-nine children with meningococcal sepsis were studied.

Main Outcome Measures: Differences in thyroid function among nonsurvivors, shock survivors, and sepsis survivors on PICU admission were the main outcome measures.

Results: The main study group consisted of 45 non-dopamine-treated children. All children had decreased total T3 (TT3)/rT3 ratios without elevated TSH. T4 sulfate levels were decreased in 88%. Nonsurvivors had paradoxically higher TT3/rT3 ratios than shock survivors (0.71 vs. 0.30); this ratio also correlated with shorter duration of disease (r = –0.43). TT4 and T4-binding globulin (TBG) levels declined with increasing disease severity. TBG levels correlated inversely with elastase levels (r = –0.46). Only TSH levels were significantly lower in 24 dopamine-treated children compared with non-dopamine-treated children (0.65 vs. 0.84), whereas other thyroid hormones did not significantly differ. Both higher TT3/rT3 ratios and lower TT4 levels were predictive for mortality, but this disappeared when IL-6 was entered into the regression model.

Conclusions: All children with meningococcal sepsis showed signs of euthyroid sick syndrome. Alterations in peripheral thyroid hormone metabolism related inversely to the duration of disease and seemed to be enacted by profound induction of type 3 deiodinase rather than by down-regulation of type 1. Lower TT4 levels were related to increased turnover of TBG by elastase. Dopamine was found to suppress only TSH secretion, not other thyroid hormone levels, on PICU admission. Both the TT3/rT3 ratio and TT4 levels were predictive for mortality, but were not superior to IL-6.


   Introduction
Top
 Abstract
 Introduction
Patients and Methods
 Results
 Discussion
 References
 
MENINGOCOCCAL DISEASE IS one of the most severe infectious diseases in children. During critical illness, many endocrine changes take place. The initial phase is characterized by peripheral down-regulation of most hypothalamic-pituitary hormones, except cortisol (1). The changes in thyroid hormones during critical illness are called the euthyroid sick syndrome or nonthyroidal illness. Initially, serum total T3 (TT3) values decline, and rT3 values increase (2). When the condition deteriorates, total T4 (TT4) values may decrease as well, producing a combined low TT3-low TT4 state that typically does not result in compensatory higher levels of TSH, which generally remain normal. Alterations in peripheral thyroid hormone metabolism, such as deiodination, play an eminent role in the development of the euthyroid sick syndrome (3, 4). Altered sulfation may also lead to changes in peripheral thyroid hormones (5). Data on thyroid hormone sulfation in critically ill adults are scarce (6), whereas data in critically ill children do not exist.

Furthermore, because thyroid hormones are mainly bound to carrier proteins, a decline in serum thyroid hormones may be the result of decreased concentrations of or decreased binding to serum carrier proteins due to serum inhibitors, such as nonesterified fatty acids (NEFA) (2). Additionally, elastase, a serine protease released by activated neutrophils, has been shown to cleave T4-binding globulin (TBG) during inflammation, resulting in decreased affinity of TBG for thyroid hormones as well as increased clearance of TBG (7, 8). Besides this, medication such as dopamine is known to influence serum thyroid hormones as well, via suppression of TSH (9, 10).

Concerning the relationship between thyroid hormone levels and mortality, a few studies in critically ill children showed an association between low levels of TT4 (11) and TT3 (12) and mortality, whereas this relation was not found in some other studies (13, 14). We previously reported higher TT3 and TT4 levels in a small group of nonsurvivors of meningococcal septic shock compared with survivors (15). To date, little is known about factors influencing changes in thyroid hormone levels in relation to outcome in critically ill children.

We therefore evaluated thyroid function in relation to disease severity and assessed the influence of deiodination, sulfation, and thyroid hormone binding on thyroid hormone levels at pediatric intensive care unit (PICU) admission in a large group of acutely ill children with sepsis or septic shock, not treated with dopamine. In addition, we investigated the influence of dopamine on thyroid hormone levels in a separate group who did receive dopamine. Finally, we assessed the predictive value of thyroid hormone levels on mortality.


   Patients and Methods
Top
 Abstract
 Introduction
 Patients and Methods
Results
 Discussion
 References
 
Patients

The group consisted of previously healthy children primary admitted or referred to the PICU of Erasmus Medical Center-Sophia Children’s Hospital between October 1997 and February 2000 and between October 2001 and January 2004, suffering from meningococcal sepsis, defined as sepsis with petechiae/purpura. Sepsis was defined as temperature of less than 36.0 C or more than 38.5 C with tachycardia and tachypneu. In addition, children were determined to have septic shock if they had persistent hypotension or evidence of poor end-organ perfusion, as described previously (15). Children were not eligible for the study if they had endocrine or chromosomal abnormalities or had received radiation or chemotherapy within the previous 6 months. The local medical ethics committee approved the study, and written informed consent was obtained from the parents or legal representatives.

Clinical parameters

To access disease severity, we determined the pediatric risk of mortality (PRISM II) score (15, 16) and the sepsis-related organ failure assessment (SOFA) score (17) and analyzed plasma IL-6, lactate, and C-reactive protein (CRP) levels. We recorded the interval between appearance of first petechia and PICU admission, respiratory support, and medication use.

Sample collection and laboratory assays

Arterial blood samples were collected on admission; after centrifugation, serum/plasma was stored at –80 C until determination of thyroid hormones, IL-6, and elastase-{alpha}-1-antitripsine complex (elastase). All other laboratory parameters were determined immediately in a certified laboratory of clinical chemistry (ISO 17025 and 9001).

Serum levels of TT4, T4 sulfate (T4S), T3, rT3, and TBG were determined by RIA, free T4 (FT4) by single-step analog binding assay and TSH, as described previously (18, 19). Thyroid hormone SD scores (z-scores) were calculated with data from a control group comprised of healthy age-matched children and considered normal between –2 and +2.

Serum CRP was determined by immunoturbidimetric assay, and serum NEFA by calorimetric assay (NEFA-C kit, WAKO Diagnostics, Richmond, VA), both on a Hitachi 912 analyzer (Roche, Indianapolis, IN). Arterial lactate and glucose were determined on blood gas analyzer (ABL 625, Radiometer, Copenhagen, Denmark). The reference values were less than 10 mg/liter for CRP, 35–50 g/liter for albumin, 0.2–1.2 mmol/liter for NEFA, less than 2.0 mmol/liter for lactate, and 2.6–11.0 mmol/liter for glucose. Plasma IL-6 levels were determined with ELISA (Sanquin, Amsterdam, The Netherlands), with a detection limit of 10 pg/ml. Plasma elastase levels were determined by RIAs (Sanquin, Amsterdam, The Netherlands; normal range, <100 ng/ml) in patients included between October 2001 and January 2004.

Caloric intake

On PICU admission, children received glucose iv (4–6 mg/kg·min), but no enteral or parenteral feeding until the second day.

Statistics

Results, analyzed with SPSS 11.5 (SPSS, Inc., Chicago, IL), are expressed as medians unless specified otherwise. We used Mann-Whitney U, {chi}2, or Fisher’s exact test, when necessary, and Spearman’s correlation coefficient (r). For backward multiple linear regression analysis and logistic regression analysis, data were log transformed when necessary. A value from two-tailed t test of P < 0.05 was considered statistical significant.


   Results
Top
 Abstract
 Introduction
 Patients and Methods
 Results
Discussion
 References
 
In the study period, 69 previously healthy children were admitted to the PICU with meningococcal sepsis. The total group was divided into the main study group (n = 45), consisting of non-dopamine-treated children, and a dopamine-treated group (n = 24). The latter group was separately analyzed and is presented at the end of Results.

Non-dopamine-treated group

Clinical parameters. The main study group (non-dopamine-treated) consisted of 30 boys and 15 girls. Children were divided according to the presence of shock and survival into the following disease severity groups: nonsurvivors (n = 8), shock survivors (n = 30), and sepsis survivors (n = 7). Concomitant therapy before study enrollment included antibiotics and administration of fluids in all children, inotropics, not dopamine, in 36 children with septic shock and in one child with sepsis, and glucocorticoids in seven children with septic shock. Twenty-three children with septic shock were mechanically ventilated and sedated with benzodiazepines.

Nonsurvivors were significantly younger than survivors and had significantly shorter time from first petechia to admission than shock survivors (Table 1Go). Parameters of disease severity were significantly higher in nonsurvivors compared with shock survivors and sepsis survivors for PRISM and SOFA scores, plasma IL-6, and arterial lactate, whereas CRP was significantly lower in nonsurvivors compared with shock survivors. IL-6 levels correlated strongly with PRISM score (r = 0.80; P < 0.001), and CRP levels correlated strongly with time from first petechia to admission (r = 0.52; P < 0.001).


View this table:
[in this window]
[in a new window]
  		TABLE 1. Patients’ characteristics of the main non-dopamine-treated study group on admission, divided into nonsurvivors, shock survivors, and sepsis survivors

 
TT4, FT4, TT3, rT3, and TSH levels. Compared with age-matched reference values, the vast majority of the children in the main, non-dopamine-treated study group had low serum TT4 levels (80%), normal FT4 levels (89%), low TT3 levels (100%), high rT3 levels (89%), and low TT3/rT3 ratios (100%), whereas none had elevated TSH levels. TT4 levels were lower with increasing disease severity; nonsurvivors had significantly lower TT4 levels than survivors, and shock survivors had significantly lower TT4 levels than sepsis survivors (Table 2Go), whereas FT4, TT3 and TSH levels did not significantly differ among the three groups. Median rT3 levels were significantly lower in nonsurvivors than in shock survivors, whereas rT3 levels were significantly higher in shock survivors than in sepsis survivors. As a result, the TT3/rT3 ratios were significantly higher in nonsurvivors compared with shock survivors and tended to be lower in shock survivors than in sepsis survivors (P = 0.057).


View this table:
[in this window]
[in a new window]
  		TABLE 2. Thyroid hormone levels and SD scores of the main non-dopamine-treated study group on admission, divided into nonsurvivors, shock survivors, and sepsis survivors

 
Parameters of disease severity correlated negatively with TT4 levels, but not with FT4, TT3, rT3, or TSH levels or TT3/rT3 ratios (Table 3Go). Time from first petechia to admission correlated positively with rT3 levels and negatively with TT3/rT3 ratios. Serum cortisol levels correlated positively with TT4 levels, but not with FT4, TT3, rT3, or TSH levels or TT3/rT3 ratio, whereas age did not correlate with any thyroid parameter.


View this table:
[in this window]
[in a new window]
  		TABLE 3. Significant (P < 0.05) Spearman’s correlation coefficients of thyroid hormones with clinical and laboratory parameters of the main non-dopamine-treated study group on admission

 
T4S levels. On admission, serum T4S levels were measured in 34 non-dopamine-treated children (one nonsurvivor, 26 shock survivors, and seven sepsis survivors). T4S levels were decreased in the majority of the children (88%) compared with age-matched reference values and did not significantly differ between shock survivors and sepsis survivors (Table 2Go).

T4S levels correlated positively with time from first petechia to admission, but not with age or any parameter of disease severity (Table 3Go). Furthermore, T4S levels tended to correlate with rT3 levels (r = 0.32; P = 0.069), but not with other thyroid hormone values or ratios.

TBG and albumin. Serum TBG levels were decreased in 62% of the children, and serum albumin levels were decreased in 54% compared with age-matched reference values, whereas none of the children had elevated TBG or albumin levels. TBG levels were lower with increasing disease severity; nonsurvivors had significantly lower TBG levels than survivors, and shock survivors had significantly lower TBG levels than sepsis survivors (Fig. 1Go), whereas albumin levels did not significantly differ among the three groups (Table 1Go). TT4/TBG ratios tended to be lower in nonsurvivors than in shock survivors (P = 0.051), but did not differ from sepsis survivors (Fig. 1Go). TT3/TBG ratios were significantly higher in nonsurvivors compared with shock survivors and sepsis survivors, whereas rT3/TBG ratios were not significantly different in nonsurvivors compared with shock survivors or sepsis survivors.



View larger version (22K):
[in this window]
[in a new window]
  		FIG. 1. Serum TBG levels (A), TT4/TBG ratios (B), TT3/TBG ratios (C), and rT3/TBG ratios (D) in the main, non-dopamine-treated study group on admission. Nonsurvivors are depicted in the left box, shock survivors in the middle box, and sepsis survivors in the right box. In the box-whisker plots, the boxes indicate the 25–75th percentile with the median, and the attached whiskers indicate the complete range with exclusion of outliers ({circ}) and extremes (*). Significant differences between nonsurvivors and shock survivors ({dagger}), between shock survivors and sepsis survivors (#) and between nonsurvivors and sepsis survivors ({ddagger}) are indicated.

 
TBG levels correlated negatively with elastase levels and parameters of disease severity, but not with age or time from first petechia to admission (Table 3Go). TT4/TBG ratios correlated positively with FT4 levels (r = 0.46; P < 0.001).

Elastase. Plasma elastase levels were measured in shock survivors and sepsis survivors and were elevated in all (Table 1Go). Median elastase levels did not significantly differ between shock survivors and sepsis survivors, but elastase levels correlated positively with parameters of disease severity, such as IL-6 levels (r = 0.66; P < 0.001) and SOFA score (r = 0.44; P = 0.021). Furthermore, elastase levels correlated negatively with TBG levels (Table 3Go).

NEFA levels and NEFA/albumin molar ratios. Serum NEFA levels were decreased in 5%, normal in 83%, and elevated in 13% of the children. NEFA levels were significantly lower in nonsurvivors than in survivors (Table 1Go). NEFA/albumin molar ratios ranged from 0.4–3.9 (median, 1.4) and were significantly lower in nonsurvivors compared with survivors. The NEFA/albumin molar ratios correlated positively with FT4 (r = 0.50) and TT4/TBG (r = 0.37; P = 0.022) and negatively with TT3/rT3 ratios (r = –0.55; Table 3Go).

Dopamine-treated group

Next to the main study group, in which none of the children received dopamine, 24 children with meningococcal septic shock received dopamine median, 3.7 h; (range, 0.25–8.7 h) before sampling. The dopamine-treated group consisted of two nonsurvivors and 22 shock survivors, but no sepsis survivors. Because all children of the dopamine-treated group experienced septic shock, we analyzed both the nonsurvivors and shock survivors together, comparing the dopamine-treated group with the non-dopamine-treated group. The mortality rate did not significantly differ between the dopamine-treated and the non-dopamine-treated group (by Fisher’s exact test, P = 0.29). The children with septic shock of the dopamine-treated group did not significantly differ in age (P = 0.55), time from first petechia to admission (P = 0.30), PRISM score (P = 0.97), SOFA score (P = 0.99), or lactate (P = 0.32), IL-6 (P = 0.56), or CRP levels (P = 0.77) compared with children with septic shock of the non-dopamine-treated group. Median TSH levels were, however, significantly lower in the dopamine-treated children with septic shock than in the non-dopamine-treated group (0.65 vs. 0.84 mU/liter; P = 0.025), and this applied also to the TSH/FT4 ratios (0.028 vs. 0.056; P = 0.005). However, median TT4 (56 vs. 52 nmol/liter; P = 0.86), FT4 (16.9 vs. 17.6 nmol/liter; P = 0.87), TT3 (0.34 vs. 0.42 nmol/liter; P = 0.10), rT3 (1.31 vs. 1.22 nmol/liter; P = 0.41), and TBG (11 vs. 10 mg/liter; P = 0.58) levels did not significantly differ between the dopamine-treated children with septic shock and the non-dopamine-treated group. Furthermore, the duration of dopamine use before admission did not correlate with TSH levels (P = 0.24), TSH/FT4 ratios (P = 0.35), or any other thyroid parameter on admission.

Total study group

Multivariate analysis. We investigated the contribution of thyroid hormone values on mortality with logistic regression analysis on the total group of 69 children. The odds for mortality against survival increased by a factor 3.7 for every doubling of the TT3/rT3 ratio and by 1.4 for every 10 nmol/liter lower TT4 level on admission. However, when admission values of IL-6 were added to the model, none of the thyroid hormone levels remained significantly related to mortality, and the odds for mortality against survival increased by a factor of 6 for every doubling of IL-6 levels.

We investigated the simultaneous contribution of pathophysiologically important factors to thyroid hormone levels on admission with multiple regression analysis on the total group of 69 children. Using multiple regression analysis, we found TT4 levels to decrease by 33% for every 10 mg/liter lower TBG level and by 4% for every halving of TSH levels. These two variables together explained 62% of the variation in TT4 levels, whereas TBG levels alone explained 58% of the TT4 variation. None of the additional variables investigated in this analysis (age, gender, IL-6 and cortisol levels, and NEFA/albumin molar ratios) were significantly related to TT4 levels. The relation between TT4 levels with TBG and TSH levels was not affected by dopamine administration before admission (P = 0.63).

We found TT3/rT3 ratios to decrease by 5% for every additional hour between first petechia to admission and by 29% for every point increase in NEFA/albumin molar ratios (R2 = 0.26). None of the additional variables investigated in the multiple regression analysis (age, gender, and IL-6, cortisol, and TT4 levels) were significantly related to TT3/rT3 ratios. The relation between TT3/rT3 ratios with time from first petechia to admission and NEFA/albumin molar ratios was not significantly affected by dopamine administration before admission (P = 0.19).

We found T4S levels to increase by 38% for every 10 g/liter lower albumin level (R2 = 0.17). None of the additional variables investigated in the multiple regression analysis (age, gender, cortisol and FT4 levels, and time between first petechia and admission) were significantly related to T4S levels. The relation between T4S levels and albumin levels was not affected by dopamine administration before admission (P = 0.31).

We found TBG levels to decrease by 2.3 mg/liter for every doubling in elastase levels (R2 = 0.15). None of the additional variables investigated (age, gender, and time from first petechia to admission) was significantly related to TBG levels. The relation between TBG levels and elastase levels was not affected by dopamine use on admission (P = 0.27).

Finally, we found only dopamine use to be significantly related to TSH levels, explaining 10% of the variation in TSH levels. None of the additional variables investigated in the multiple regression analysis (age, gender, time from first petechia to admission, or IL-6, cortisol, and FT4 levels) was significantly related to TSH levels.


   Discussion
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Discussion
References
 
In this study all critically ill children with sepsis or septic shock had signs of euthyroid sick syndrome on PICU admission, depicted by low TT3 and TT4 levels and high rT3 levels without compensatory elevated TSH levels. In the main, non-dopamine-treated study group, nonsurvivors appeared to lack time to develop full-blown euthyroid sick syndrome, resulting in paradoxically higher TT3/rT3 ratios in nonsurvivors than in shock survivors. In survivors, both T4S levels and TT3/rT3 ratios were decreased, which suggests alterations in peripheral thyroid hormone metabolism with a profound induction of type 3 deiodinase (D3), rather than down-regulation of type 1 deiodinase (D1) in the initial phase of meningococcal sepsis (see below). Furthermore, lower TT4 levels were related to lower TBG levels, which may be due to increased degradation of TBG by elastase. In the dopamine-treated group, dopamine was found to only have a suppressive effect on pituitary TSH secretion, although apparently not long enough to suppress thyroid hormone levels at the time of PICU admission. In the total group, both higher TT3/rT3 ratios and lower TT4 levels were predictive for mortality, but their predictive value for mortality disappeared when IL-6 levels were entered in the model.

The peripheral inactivation of circulating thyroid hormones plays an important role in the initial phase of critical illness (1, 3, 5). Increased levels of biologically inactive rT3 at the expense of decreased TT3 levels are a hallmark of the euthyroid sick syndrome, which are enacted by altered deiodination, the main pathway of peripheral thyroid hormone metabolism (4) (Fig. 2Go). We found decreased TT3/rT3 ratios in all children with meningococcal sepsis, suggesting altered deiodination. This is in agreement with studies in critically ill children and adults (11, 12, 20, 21). In contrast, however, we found significantly higher TT3/rT3 ratios in nonsurvivors than in shock survivors. A possible explanation for this might be the more fulminant course of disease in nonsurvivors, represented by a shorter time between the appearance of first petechia and admission in those children. The appearance of petechiae is considered to be the first clearly detectable clinical sign of meningococcal sepsis. Because the TT3/rT3 ratios were significantly inversely related to time from appearance of first petechia to admission, this might suggest that the peripheral thyroid hormone metabolism in nonsurvivors lacked the time to develop the full-blown euthyroid sick syndrome before PICU admission. Similarly, the acute phase protein CRP, which is known to be induced within 6–8 h, correlated inversely with time from first petechia to admission as well, indicating that nonsurvivors had less time to produce CRP before admission than survivors. We cannot, however, rule out the possibility that nonsurvivors could not sufficiently adapt to conditions of acute critical illness.



View larger version (20K):
[in this window]
[in a new window]
  		FIG. 2. Peripheral thyroid hormone metabolism during normal homeostasis (A) and during critical illness (B). During critical illness, deiodination is altered, with reduced D1 activity (thin arrow) and increased D3 activity (thick arrow), which both result in increased rT3 levels at the expense of TT3 levels (left side). Sulfated thyroid hormones, however, are exclusively deiodinated by D1 (right side), and reduced D1 levels will result in elevated levels of all sulfated thyroid hormones. Sulfation of thyroid hormones is indicated by dotted arrows.

 
Increased levels of cytokines (22, 23, 24) and glucocorticoids (3, 25), from either endogenous or exogenous sources, are both indicated as potential mediators in the alterations of thyroid function that occur in critical illness. In our study, major differences were seen in IL-6 levels, one of the key cytokines in the inflammatory response, between the disease severity groups. IL-6 levels were extremely high in nonsurvivors and much higher than in survivors, which is in agreement with studies in children (26) and adults (27) with life-threatening infections. Our study, however, does not support the postulation that cytokines, such as IL-6, are the potential mediators for altered peripheral deiodination during acute critical illness, because no significant relations were found between IL-6 and TT3 and rT3 levels or TT3/rT3 ratios. An explanation for this can be found in the different cortisol response in children during septic shock vs. adults. Although cortisol levels in critically ill adults are positively correlated with disease severity and outcome (20), in children with meningococcal sepsis, the opposite is true (15, 26, 28).

Next to deiodination, sulfation is another pathway of peripheral thyroid hormone metabolism, leading to the metabolites T4S, T3S, and rT3S (29). Data on thyroid hormone sulfation during critical illness are very scarce (6, 30). To our knowledge, we are the first to report T4S levels in the euthyroid sick syndrome. Contrary to our expectations, T4S levels were decreased in the vast majority of children with meningococcal sepsis compared with healthy age-matched children. Unpublished observations by H. van Toor and T. J. Visser revealed increased T4S levels in adults with the euthyroid sick syndrome. Recently, Peeters et al. (4) showed that the increased rT3 levels at the expense of TT3 levels in critical illness are not only due to down-regulation of D1, but also to induction of D3 (Fig. 2Go). Because iodothyronine sulfates are exclusively and rapidly deiodinated by D1 (31), the combination of elevated rT3 levels with low T4S levels might be more suggestive of a profound induction of D3 rather than down-regulation of D1 in the initial phase of meningococcal sepsis. Decreased sulfation of TT4 or increased T4S uptake in tissues, however, cannot be excluded, although the latter is less likely, because tissue uptake of thyroid hormones in critically illness is generally decreased (2).

Because TT4, TT3, and rT3 are mainly bound to carrier proteins, differences in these hormone levels might also be based on decreased concentrations of or less binding to serum carrier proteins, such as TBG and albumin. Changes in serum levels of TBG, the most important carrier protein (32), explained 58% of the variation in TT4 levels, whereas increasing disease severity was correlated with both lower TBG and TT4 levels. Elastase levels were elevated and positively related positively to disease severity and inversely to serum TBG levels. This suggests that with increasing disease severity, TBG was increasingly cleaved by elastase, a serine protease that is released by activated neutrophils (33), resulting in lower TBG levels. Dilution and capillary leakage might be alternative processes resulting in lower TBG levels. Another, but less important, carrier protein for TT4, TT3, and rT3 is albumin. Although albumin levels were decreased in half the children, it is unlikely that albumin levels contributed to the changes in TT4, TT3, and rT3, because albumin levels did not significantly differ between the groups, whereas TT4, TT3, and rT3 levels did.

Besides decreased levels of binding proteins, less binding of thyroid hormones to their binding proteins by circulating inhibitors, such as NEFA, has been reported in the euthyroid sick syndrome (34). It has been reported that increased NEFA levels compete directly with TT4 for binding to TBG, especially when NEFA/albumin molar ratios exceed 5 (2, 35). Although the NEFA/albumin molar ratios in all of our patients were less than 5, they ranged widely from 0.4–3.9. Because the NEFA/albumin molar ratios were more strongly positively related to FT4 than to TT4/TBG levels, it cannot be ruled out that NEFA, even at relatively low NEFA/albumin molar ratios, might also be involved in the displacement of thyroid hormones from TBG. Furthermore, we found a significant inverse relation between NEFA/albumin molar ratios and TT3/rT3 ratios, suggesting that NEFA-induced displacement of thyroid hormones from their plasma binding proteins in addition to declined TBG levels might accelerate the inactivation of thyroid hormones in the euthyroid sick syndrome. Because nonsurvivors had significantly lower NEFA levels compared with survivors, this might also have contributed to the less decreased TT3/rT3 ratios in nonsurvivors.

Finally, in an additional group of children with meningococcal septic shock who received dopamine, we found significantly lower TSH levels and TSH/FT4 ratios than in those who did not receive dopamine, whereas other thyroid hormone levels did not significantly differ. The suppressive effect of dopamine infusion on the pituitary gland has been previously described in critically ill children and adults (9, 10). Dopamine directly inhibits anterior pituitary function through inhibitory dopamine receptors, resulting in diminished TSH release (36, 37). In children recovering from cardiac surgery, cessation of dopamine therapy resulted in an abrupt rise in TSH levels, followed by increases in TT4 and TT3 levels and in the TT3/rT3 ratio 24 h thereafter (9). Although our study was not designed to investigate the effects of dopamine, we found evidence for a suppressive effect on TSH levels, but not on TT4 or TT3 levels in critically ill children on PICU admission. An explanation for this might be the relatively short duration of dopamine administration at the time of sampling (median, 3.7 h) compared with the previously published data on dopamine administration of at least 21 h (9, 10).

In conclusion, our study shows that all critically ill children with sepsis or septic shock have signs of euthyroid sick syndrome on PICU admission. Alterations in peripheral deiodination were related to duration of illness and seemed to be enacted by a profound induction of D3 rather than down-regulation of D1. Low TT4 levels were related to increased cleavage of TBG by elastase. Dopamine-treated children showed only a reduction of TSH levels and no difference in other thyroid hormone levels compared with the non-dopamine-treated children at the time of PICU admission. Values of TT3/rT3 and TT4 were predictive for outcome, but this disappeared when IL-6 was entered in the model.


   Acknowledgments
 
We thank M. Maliepaard (research nurse) for her technical assistance; C. L. Vermont (M.D., Ph.D.), E. D. de Kleijn (M.D., Ph.D.), and their colleagues as well as E. M. N. Bannink, R. D. van Beek, M. van Dijk, D. A. M. Festen, D. C. van der Kaay, and A. S. Slingerland (all M.D.s) for helping to collect samples on weekends; and the nursing staff of the PICU for their support. We also thank H. van Toor and Y. B. de Rijke (Ph.D.) for providing age-matched reference values of thyroid parameters.


   Footnotes
 
This was an investigator-initiated project funded by an unconditional grant from Pharmacia, now Pfizer.

First Published Online August 2, 2005

Abbreviations: CRP, C-Reactive protein; D1, type 1 deiodinase; D3, type 3 deiodinase; FT4, free T4; NEFA, nonesterified fatty acid; PICU, pediatric intensive care unit; PRISM, pediatric risk of mortality; SOFA, sepsis-related organ failure assessment; TBG, T4-binding globulin; T4S, T4 sulfate; TT3, total T3.

Received April 22, 2005.

Accepted July 22, 2005.


   References
Top
 Abstract
 Introduction
 Patients and Methods
 Results
 Discussion
 References
 

  1. Van den Berghe G, de Zegher F, Bouillon R 1998 Clinical review 95: acute and prolonged critical illness as different neuroendocrine paradigms. J Clin Endocrinol Metab 83:1827–1834[Free Full Text]
  2. Docter R, Krenning EP, de Jong M, Hennemann G 1993 The sick euthyroid syndrome: changes in thyroid hormone serum parameters and hormone metabolism. Clin Endocrinol (Oxf) 39:499–518[Medline]
  3. Bianco AC, Salvatore D, Gereben B, Berry MJ, Larsen PR 2002 Biochemistry, cellular and molecular biology, and physiological roles of the iodothyronine selenodeiodinases. Endocr Rev 23:38–89[Abstract/Free Full Text]
  4. Peeters RP, Wouters PJ, Kaptein E, van Toor H, Visser TJ, Van den Berghe G 2003 Reduced activation and increased inactivation of thyroid hormone in tissues of critically ill patients. J Clin Endocrinol Metab 88:3202–3211[Abstract/Free Full Text]
  5. Visser TJ 1996 Pathways of thyroid hormone metabolism. Acta Med Austr 23:10–16
  6. Santini F, Chiovato L, Bartalena L, Lapi P, Palla R, Panichi V, Velluzzi F, Grasso L, Chopra IJ, Martino E, Pinchera A 1996 Study of serum 3,5,3'-triiodothyronine sulfate concentration in patients with systemic non-thyroidal illness. Eur J Endocrinol 134:45–49[Abstract/Free Full Text]
  7. Afandi B, Schussler GC, Arafeh AH, Boutros A, Yap MG, Finkelstein A 2000 Selective consumption of thyroxine-binding globulin during cardiac bypass surgery. Metabolism 49:270–274[CrossRef][Medline]
  8. Janssen OE, Golcher HM, Grasberger H, Saller B, Mann K, Refetoff S 2002 Characterization of T4-binding globulin cleaved by human leukocyte elastase. J Clin Endocrinol Metab 87:1217–1222[Abstract/Free Full Text]
  9. Van den Berghe G, de Zegher F, Lauwers P 1994 Dopamine suppresses pituitary function in infants and children. Crit Care Med 22:1747–1753[Medline]
  10. Van den Berghe G, de Zegher F, Lauwers P 1994 Dopamine and the sick euthyroid syndrome in critical illness. Clin Endocrinol (Oxf) 41:731–737[Medline]
  11. Uzel N, Neyzi O 1986 Thyroid function in critically ill infants with infections. Pediatr Infect Dis 5:516–519[Medline]
  12. Yildizdas D, Onenli-Mungan N, Yapicioglu H, Topaloglu AK, Sertdemir Y, Yuksel B 2004 Thyroid hormone levels and their relationship to survival in children with bacterial sepsis and septic shock. J Pediatr Endocrinol Metab 17:1435–1442[Medline]
  13. Anand NK, Chandra V, Sinha RS, Chellani H 1994 Evaluation of thyroid functions in critically ill infants. Indian Pediatr 31:1233–1237[Medline]
  14. Zucker AR, Chernow B, Fields AI, Hung W, Burman KD 1985 Thyroid function in critically ill children. J Pediatr 107:552–554[CrossRef][Medline]
  15. Joosten KF, de Kleijn ED, Westerterp M, de Hoog M, Eijck FC, Hop WCJ, Voort EV, Hazelzet JA, Hokken-Koelega AC 2000 Endocrine and metabolic responses in children with meningoccocal sepsis: striking differences between survivors and nonsurvivors. J Clin Endocrinol Metab 85:3746–3753[Abstract/Free Full Text]
  16. Pollack MM, Ruttimann UE, Getson PR 1988 Pediatric risk of mortality (PRISM) score. Crit Care Med 16:1110–1116[Medline]
  17. Vincent JL, Moreno R, Takala J, Willatts S, De Mendonca A, Bruining H, Reinhart CK, Suter PM, Thijs LG 1996 The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med 22:707–710[Medline]
  18. den Brinker M, Dumas B, Visser TJ, Hop WCJ, Hazelzet JA, Festen DAM, Hokken-Koelega AC, Joosten KFM 2005 Thyroid function and outcome in children who survived meningococcal septic shock. Intensive Care Med 31:970–976[CrossRef][Medline]
  19. Williams FL, Simpson J, Delahunty C, Ogston SA, Bongers-Schokking JJ, Murphy N, van Toor H, Wu SY, Visser TJ, Hume R 2004 Developmental trends in cord and postpartum serum thyroid hormones in preterm infants. J Clin Endocrinol Metab 89:5314–5320[Abstract/Free Full Text]
  20. Rothwell PM, Udwadia ZF, Lawler PG 1993 Thyrotropin concentration predicts outcome in critical illness. Anaesthesia 48:373–376[CrossRef][Medline]
  21. Ray DC, Macduff A, Drummond GB, Wilkinson E, Adams B, Beckett GJ 2002 Endocrine measurements in survivors and non-survivors from critical illness. Intensive Care Med 28:1301–1308[CrossRef][Medline]
  22. Boelen A, Kwakkel J, Thijssen-Timmer DC, Alkemade A, Fliers E, Wiersinga WM 2004 Simultaneous changes in central and peripheral components of the hypothalamus-pituitary-thyroid axis in lipopolysaccharide-induced acute illness in mice. J Endocrinol 182:315–323[Abstract]
  23. Chopra IJ, Sakane S, Teco GN 1991 A study of the serum concentration of tumor necrosis factor-{alpha} in thyroidal and nonthyroidal illnesses. J Clin Endocrinol Metab 72:1113–1116[Abstract/Free Full Text]
  24. van der Poll T, Romijn JA, Wiersinga WM, Sauerwein HP 1990 Tumor necrosis factor: a putative mediator of the sick euthyroid syndrome in man. J Clin Endocrinol Metab 71:1567–1572[Abstract/Free Full Text]
  25. LoPresti JS, Eigen A, Kaptein E, Anderson KP, Spencer CA, Nicoloff JT 1989 Alterations in 3,3'5'-triiodothyronine metabolism in response to propylthiouracil, dexamethasone, and thyroxine administration in man. J Clin Invest 84:1650–1656
  26. van Woensel JB, Biezeveld MH, Biesterbos Alders AM, Eerenberg AJ, Endert E, Hack EC, von Rosenstiel IA, Kuijpers TW 2001 Adrenocorticotropic hormone and cortisol levels in relation to inflammatory response and disease severity in children with meningococcal disease. J Infect Dis 184:1532–1537[CrossRef][Medline]
  27. Gardlund B, Sjolin J, Nilsson A, Roll M, Wickerts CJ, Wretlind B 1995 Plasma levels of cytokines in primary septic shock in humans: correlation with disease severity. J Infect Dis 172:296–301[Medline]
  28. De Kleijn ED, Joosten KF, Van Rijn B, Westerterp M, De Groot R, Hokken-Koelega AC, Hazelzet JA 2002 Low serum cortisol in combination with high adrenocorticotrophic hormone concentrations are associated with poor outcome in children with severe meningococcal disease. Pediatr Infect Dis J 21:330–336[CrossRef][Medline]
  29. Visser TJ 1994 Role of sulfation in thyroid hormone metabolism. Chem Biol Interact 92:293–303[CrossRef][Medline]
  30. Chopra IJ, Wu SY, Teco GN, Santini F 1992 A radioimmunoassay for measurement of 3,5,3'-triiodothyronine sulfate: studies in thyroidal and nonthyroidal diseases, pregnancy, and neonatal life. J Clin Endocrinol Metab 75:189–194[Abstract]
  31. Visser TJ 1996 Role of sulfate in thyroid hormone sulfation. Eur J Endocrinol 134:12–14[Abstract/Free Full Text]
  32. Schussler GC 2000 The thyroxine-binding proteins. Thyroid 10:141–149[Medline]
  33. Jirasakuldech B, Schussler GC, Yap MG, Drew H, Josephson A, Michl J 2000 A characteristic serpin cleavage product of thyroxine-binding globulin appears in sepsis sera. J Clin Endocrinol Metab 85:3996–3999[Abstract/Free Full Text]
  34. Chopra IJ, Teco GN, Nguyen AH, Solomon DH 1979 In search of an inhibitor of thyroid hormone binding to serum proteins in nonthyroid illnesses. J Clin Endocrinol Metab 49:63–69[Abstract/Free Full Text]
  35. Lim CF, Stockigt JR, Curtis AJ, Wynne KN, Barlow JW, Topliss DJ 1993 A naturally occurring furan fatty acid enhances drug inhibition of thyroxine binding in serum. Metabolism 42:1468–1474[CrossRef][Medline]
  36. Wood DF, Johnston JM, Johnston DG 1991 Dopamine, the dopamine D2 receptor and pituitary tumours. Clin Endocrinol (Oxf) 35:455–466[Medline]
  37. Goldsmith PC, Cronin MJ, Weiner RI 1979 Dopamine receptor sites in the anterior pituitary. J Histochem Cytochem 27:1205–1207[Abstract]



This article has been cited by other articles:


Home page
 J Med MicrobiolHome page
A. L. de Souza and A. C. Seguro
Two centuries of meningococcal infection: from Vieusseux to the cellular and molecular basis of disease
J. Med. Microbiol., November 1, 2008; 57(11): 1313 - 1321.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by den Brinker, M.
Right arrow Articles by Hokken-Koelega, A. C. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by den Brinker, M.
Right arrow Articles by Hokken-Koelega, A. C. S.
Related Collections
Right arrow Thyroid

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology Endocrine Reviews J. Clin. End. & Metab.
Molecular Endocrinology Recent Prog. Horm. Res. All Endocrine Journals