The Thyroid and its Diseases
	Chapter 5a-- Effects of the Environment, Chemicals and Drugs on Thyroid Function Revised by David Sarne, 1 Nov 2004 Chapter 5b-- The Non-Thyroidal Illness Syndrome Revised by Leslie J. De Groot, M.D., 1 Jan 2005 TO OBTAIN A DOWNLOAD OF THIS CHAPOTER IN PDF OR WORD, CLICK HERE

Preface- The following section was published in a similar form in the Journal of Clinical Endocrinology and Metabolism. The Editor of the Journal and the Endocrine Society have kindly given permission to use the basic text in THYROIDMANAGER.

The author of this article takes the position that patients with severe NTIS  most likely metabolically hypothyroid due top hypothalamic dysfunction and reduced peripheral generation of T3, and that they should receive replacement thyroid hormone therapy. It is appropriate to acknowledge that this disagrees with a common interpretation of the condition, and readers may wish to review other opinions, several of which are referenced in the bibliography. A recent version of this opinion is found in reference 1b.

Introduction-

Low T3 States

Non-thyroidal Illness Syndrome With Low Serum T4

As the severity of illness, and often associated starvation, progresses, there is the gradual development of a more complex syndrome associated with low T3 and low T4 levels. In this state serum free T4 levels are commonly below normal but may be normal, or rarely above normal as described below. Generally TSH levels are low or normal, despite the low serum hormone levels, and reverse T3 levels are normal or elevated. The depression of serum T3 alone represents the least marked abnormality in NTIS, but there is no clear separation of this response from the more severe syndrome. Rather, there seems to be a gradual progression from just a low T3, to the most advanced condition in serious illness, associated with extremely low T3 and T4 levels. Most patients with serious illness in the hospital have low serum T3. A large proportion of patients in an intensive care unit setting have various degrees of severity of NTIS with low T3 and T4. Girvent et al note that NTIS is highly prevalent in elderly patients with acute surgical problems, and is associated with poor nutrition, higher sympathetic response, and worse postoperative outcome (11.1).

The reason for interest in this syndrome is not simply to understand its physiology. A marked decrease in serum T4 is associated with a high probability of death. NTI was found in a group of 20 patients with severe trauma, among whom 5 died, and the drop in T3 correlated with the Apache II score(11.1).  NTI found in patients undergoing bone marrow transplantation associated with a high probability of fatal outcome (11.2)   NTIS was typical in elderly patients undergoing acute surgery and associated with a worse prognosis (11.3). When serum T4 levels drop below 4 g/dl, the probability of death is about 50%, and with serum T4 levels below 2 g/dl, the probability of death reaches 80% (12-15). Obviously this raises the question as to whether replacement of thyroid hormone would be beneficial in such patients, and could increase their chance of survival. The dogma in endocrinology, accepted and supported by many individuals in the field over the past three decades (15-17), has been that this is a beneficial physiologic response and that "it is difficult to advocate or even defend treatment of NTI patients" (18). However, as described below, there is no factual basis for this dogma.

Five conceptual explanations of NTIS can be followed through the literature:

1. The abnormalities represent test artifacts, and assays would indicate euthyroidism if a proper test were employed.

2. The serum thyroid hormone abnormalities are due to inhibitors of T4 binding to proteins, and tests do not appropriately reflect free hormone levels. Proponents of this concept may or may not take the position that a binding inhibitor is present throughout body tissues, rather than simply in serum, and that the binding inhibitor may also inhibit uptake of hormone by cells or prevent binding to nuclear T3 receptors, and thus inhibit action of hormone.

3. In NTIS, T3 levels in the pituitary are normal because of enhanced local deiodination. Thus the pituitary is actually euthyroid, while the rest of the body is hypothyroid. This presupposes enhanced intrapituitary T4 T3 deiodination as the cause.

4. Serum hormone levels are in fact low, and the patients are biochemically hypothyroid, but this is (teleologically) a beneficial physiologic response and should not be altered by treatment.

5. Lastly, the patient's serum and tissue hormone levels are truly low, tissue hypothyroidism is present, this is probably disadvantageous to the patient, and therapy should be initiated if serum thyroxine levels are depressed below the danger level of 4 g/dl.

Serum T3 and Free T3: With few exceptions, reports on NTIS indicate that serum T3 and Free T3 levels are low (19-24). Chopra and coworkers have recently reported that freeT3 levels were low (Fig.5b-1) (25), or in a second report, normal (26). However it is important to note that in the latter report the patients with "NTIS" actually had average serum T4 levels that were above the normal mean. While it is uncertain which study should be given precedence, it is clear that most of the subjects in the latter report did not have severe NTIS and can not be considered at risk.

Figure 5b-1.  Free T3 concentrations in different groups of patients, as reported by Chopra et al, reference 25. In this report, patients with NTIS have significantly lowered Free T3 levels than do normal subjects.

Serum rT3 is normal or elevated, and is not a reliable indicator of abnormal thyroid hormone supply (25a). While it may be expected that rT3 should always be elevated, this is not true, and more commonly it is within the normal range. The enzymatic machinery for deiodination of T4 to rT3 is reduced. It is usually assumed that further metabolism of rT3 via the 5'-deiodinase is inhibited by decreased function of the same enzyme that generates T3. However formation of rT3 is limited by the low level of substrate (T4) in serum and in tissues, and perhaps by inhibition of T4 entry into cells. Experience treating patients with NTIS (unpublished) shows that when T4 is given and repletes serum hormone levels, generation of rT3 rapidly increases, and levels become significantly elevated.

Serum T4: Serum T4 levels are reduced in NTIS in proportion to the severity and, probably length of the illness (17-21). In acute, short term, trauma such as cardiac bypass (27), or short term starvation (28), there is no drop in serum T4. However, with increasing severity of trauma, illness, or infection, there is a drop in T4 which may become extreme. As indicated, serum T4 levels below 4 ug/dl are associated with a marked increased risk of death (up to 50%), and once T4 is below 2, prognosis becomes extremely guarded. In neonates, low total T4 and TSH are associated with a greater risk of death and severe intraventricular hemorrhage.  It is suggested that thyroid hormone supplementation might be a potential benefit in infants with the lowest T4 values (21.1)  

 Total serum T4 is reduced in part because of a reduction in TBG. One reason for this reduction appears to be because of cleavage of TBG. Schussler’s group recognized a rapid drop TBG to 60% of baseline within 12 hours after bypass surgery, and their data suggest that this is due to cleavage of TBG by protease, which causes TBG to lose its T4 binding activity (28.1).  Further studies by this group demonstrated the presence of a cleaved form of TBG present in serum of patients with sepsis (28.2).  One hypothesis from this work is that, in sites of inflammation with proteases present, TBG could be cleaved, thyroxine liberated, deiodinated, and that the iodide would be involved in use by leukocyte myeloperoxidase  in a bacteriacidal peroxidase reaction. 

Serum Free Thyroxine: The major problem in understanding the NTIS is in analyzing data on the level of free T4. Free T4 is believed by most workers to represent hormone availability to tissues. The results of Free T4 assays in NTIS are definitely method dependent, and may be influenced by a variety of variables including (alleged) inhibitors present in serum, or the effect of agents such as drugs, metabolites, or free fatty acids in the serum or assay. Assays which employ a resin uptake method to estimate free hormone usually return low values for calculated free thyroxine in NTIS. Methods using T3 analogs in the assay also give levels that are depressed. The free T4 level determined by dialysis varies widely, as does T4 measured by ultrafiltration (19-23), but the majority of reports are of normal or low, and in some samples, even elevated values.

In theory, methods utilizing equilibrium dialysis may allow dilution of dialyzable inhibitors. Compounds such as 3-carboxy-4-methyl-5-propyl-2-furan-propanoic acid, indoxyl sulfate and hippuric acid, can accumulate in severe renal failure (29). However these compounds probably do not interfere with serum hormone assays. Free fatty acids, if elevated to 2 - 5 mmol/l, can displace T4 binding to TBG and elevate free T4. Free fatty acids almost never reach such levels in vivo (30, 31). However, even small quantities of heparin (0.08 U/kg given iv, or 5,000 U given sc) can lead to in vitro generation of free fatty acids during extended serum dialysis, and falsely augment apparent free hormone levels (32). Since heparin is so universally employed for prevention of thrombotic episodes in patients in intensive care units and in other settings during severe illness, this is probably a widespread and serious problem which may explain many instances of apparently elevated free T4 levels in patients with acute illness.

One of the most thorough comparative studies of serum T4 assays was reported in 1982 by Melmed et al (20). Free T4 was measured by six methods including dialysis, and was found to be uniformly reduced as measured by all methods in patients in the MICU, while results were more variable for patients with liver disease or chronic renal failure (see below). A problem to be noted in reviewing these reports has to do with the categorization of patients. Patients reported with NTIS who have normal serum T4 typically will not have reduced free T4 by most assay methods. However, when patients with low serum T4 are studied separately, the results become more uniform. In an extensive comparison of methods by Kaptein and associates (21), free T4, measured by five methods, was extremely low in patients with NTIS who had a serum T4 under 3 g/dl. However, free T4 was in the normal range in patients measured by two commercial methods and by equilibrium dialysis. Uchimura et al. (33) studied the effect of dilution of serum on free T4 and found that it caused up to a 30% reduction in apparent free T4. This reduction caused by dilution of course also applied to serum standards. Thus values obtained by study of undiluted serum, diluted serum, or using indirect methods for establishing free T4 concentration, all gave values which closely correlated. Nelson and Weiss (34) also studied the effect of serum dilution on free T4. They found that using a tracer dialysis method, there was progressive reduction in free T4 values with serum dilution. The change with dilution of free T4 in sera from a normal patient, and a patient with NTIS, varied in parallel. Thus, by this method, despite dilution, values for the NTIS patient appeared low. However, using a method which they believe is more appropriate, measuring T4 in the dialysate by direct RIA, serum from patients with low T3 syndrome frequently gave high values when undiluted, and normal or even low values when diluted. Nelson and Weiss are convinced that the direct RIA method is correct, and that the alterations reflect the presence of dialyzable inhibitors in the serum altering the measurement of free T4.
    Wang et al (34.1) recently investigated the reliability of several commercial methods for measuring free T4 in the presence of altered binding proteins, and found that results were often erroneous. Using a dialysis method which they believe is highly accurate, they found free T4 levels in NTI patients, with an average T4 of 4 ug/dl, to be in the normal range.

Results obtained using ultrafiltration also are variable. Wang et al. (35) found that, in patients with NTIS, free T4 measured by ultrafiltration was uniformly low (average of 11.7 ng/liter), but when measured by equilibrium dialysis, free T4 was near normal, at 18 ng/liter. By ultrafiltration, free T3 was also, not surprisingly, found to be low and similar to free T3 by radioimmune assay. The authors suggest that the observations with ultrafiltration are more apt to be erroneous due to the effect of inhibitors of binding, in contrast to the results of dialysis, which they assume are correct. Chopra et al. (25) recently reported free T3 measured by dialysis in patients with NTIS, and found free T3 to be markedly reduced whereas free T4 was within the normal range. However, it must be noted that, in this study, their patients had an average T4 in the normal range (6.9 g/dl), and these patients would not be expected to have low free T4 levels. The second study from this group recently published is noted above. Surks et al. (19) studied free T4 levels by equilibrium dialysis and by ultrafiltration of undiluted serum. Although the authors report that the results in patients with NTIS were "similar to or higher than those in 12 normal subjects", in fact 7 of 9 patients had levels below the normal mean, ± 2 SD, when measured by dialysis, 6 of 9 were low when measured by ultrafiltration, and 7 of 9 were low when measured by standard resin-uptake-corrected free T4. The means of the NTIS patients in this study were clearly below the mean of normal.

Thus it is still a question as to whether the free T4 in patients with NITS is actually low, or normal, and even sometimes elevated. It is of interest that this problem does not carry over to estimates of free T3, which are depressed in most studies. There might be two reasons for this difference. Firstly, the depression of total T3 is proportionately greater than of total T4. Secondly, factors which affect thyroid hormone binding are more apt to alter T4 assays than T3, since T4 is normally more tightly bound to TBG than is T3.

In patients with advanced renal disease who have not been recently dialyzed, there is possibly the accumulation of substances, as noted above, which can alter binding of T4 (29). These materials could be dialyzed out promptly during assays of free hormone, and therefore cause the assay to record an apparently low free T4. Evidence for dialyzable and non-dialyzable inhibitors of T4 binding has been presented by Chopra (36). The material in serum was thought possibly to be fatty acids. In contrast, Mendel and colleagues (37) found no evidence for an inhibitor of thyroxine binding to serum proteins in a study of a series of 111 patients from acute care wards. It should be noted that almost all subjects had T4 values within the normal range. Only 3 had values below 4 g/dl. Thus the patients may not have been optimal for studying evidence of a binding inhibitor. As reviewed by Mendel et al (37), one of the main concerns regarding an inhibitor of binding is the potential effect of elevated free fatty acid (FFA) levels in starving NTIS patients. Levels of FFA above 5 mmol/l, with a molar ratio of FFA to albumin of > 5, may produce this abnormality. In the patients studied by Liewendahl (30) and by Csako et al (38), and in the study of Mendel et al (37), the FFA levels were below this level. Thus the FFA levels in the serum sample taken from patients ordinarily are not high enough to cause a problem, although remarkably elevated FFA levels were found in the series of patients reported by Chopra et al (39). A more serious problem may occur if low doses of heparin have been given, as noted above. FFA can be generated during the incubation procedure, as reported by Jaume et al (32). In this situation, there may be progressive increase in FFA during prolonged dialysis, causing a spurious increase in free thyroxine fraction. Mendel et al (37) carefully review the studies that have claimed the presence of dialyzable inhibitors of binding and point out that many of these studies must be viewed with caution. Numerous artifacts are present in both dialysis assays and ultrafiltration assays. They also point out, that, while the low free T4 by resin uptake assays found in NTIS generally do not agree with the clinical status of the patient, it is equally true that clinical assessment generally does not fit with the high free T4 results found by some equilibrium dialysis assays in NTIS.
    Administration of salsalate to humans causes acute displacement of T4 and T3 from TBG, a drop in total T4 and T3, transient elevation of fT4, and transient suppression of TSH. This has been proposed as a model of NTI (37.1), and  does model the effect of a binding inhibitor. Whether this relates to any other aspect of chronic NTI is unknown.
    Acute reduction in TBG is seen in the post-operative state (40.1), and also after isolated limb perfusion using TNF for cancer therapy (40.2). In the later condition Il- 6 becomes elevated transiently, free T 4 is acutely elevated, TSH is transiently suppressed, and then TSH rebounds above normal suggesting prior induction of transient hypothyroidism. The sudden drop of TBG occuring over 10 min in this model, and lasting 1 day, appears to cause the elevation of free hormone.  A dilution effect would not by itself elevate free hormone, but it is possible that serum dilution introduced an artifact in measurement of free hormone, as noted above (34.1).

A strong argument against the importance of factors in serum inhibiting binding of thyroid hormone is provided in the clinical study of Brendt and Hershman (Fig.5b-2)(40). These researchers gave 1.5 g of T4 per kg body weight to 12 of 24 patients with severe NTIS and followed serum hormone levels over 14 days. T4 levels returned to the normal range within three days of therapy. Thus the thyroxine pool was easily replenished, and T4 levels reached normal values. Not surprisingly, because of reduced T4 T3 deiodination, T3 levels did not return to the normal range until the end of the study period in the few patients that survived. However, the ability of intravenous thyroxine to restore the plasma pool to normal clearly shows that an inhibitor of binding could not be the predominant cause of low serum T4 in this group of severely ill patients.

Figure 5b-2: Patients with severe NTIS were randomized and left untreated or given T4 iv over two weeks. Serum T3, T4, and TSH concentrations are shown for the survivors of the control ( , 1 – 3), and T4-treated ( o, 4 – 6), groups during the study period and at the time of follow-up. The shaded area designates the normal range. Note the prompt recovery of T4 values to the normal range immediately following i.v. treatment with T4. Also note the elevated TSH levels in some patients. T3 levels did not return to normal following T4 treatment for up to two weeks. (Reference 40).

Serum TSH in NITS is typically normal or reduced and may be markedly low, although usually not less than .05 U/ml (19, 20, 22, 25, reviewed in 17 and 41). However, to use usual endocrinological logic, these TSH levels are almost always inappropriately low for the observed serum T4 and T3. Third generation assays with sensitivity down to .001 U/ml may allow differentiation of patients with hyperthyroidism ( a rare problem in differential diagnosis) to be separated from those with NTIS, although there can be overlap in these very disparate conditions (42). There is a suggestion that serum TSH in patients with NTIS may have reduced biological activity, perhaps because of reduced TRH secretion and reduced glycosylation. Some patients are found with a TSH level above normal, and elevation of TSH above normal commonly occurs transiently if patients recover (Fig.5b-3)(17,23, 40). This elevation of TSH strongly suggests that the patients are recovering from a hypothyroid state, during which the ability of the pituitary to respond had been temporarily inhibited.

Figure 5b-3: T3 and TSH concentrations are shown in patients with nonthyroidal illness who were eventually discharged from hospital (left panels). The broken line indicates ± 2 SD of the mean value in the normal subjects. The right panel displays T3 and TSH concentrations in patients with NTIS who died. Subjects are indicated by numbers. Note the elevated TSH in some patients who recovered, and the generally dropping T3 and low TSH levels in patients who died. (Reference 23)

Responsiveness of the pituitary to TRH during NTIS is variable: many patients respond less than normal (43) and others respond normally (44). Normal responsiveness in the presence of low TSH may suggest that there is an hypothalamic abnormality as a cause of the low TSH and low T4. There is also a diminution, or loss, of the diurnal rhythm of TSH (45), and in some studies there is evidence for reduction of TSH glycosylation with lower TSH bioactivity (46). That TSH is not elevated in the presence of low T4 is taken to mean that the patients are not hypothyroid. An easy and perhaps more logical alternative explanation is that the low TSH is in fact the proximate cause of the low thyroid hormone levels. As will be shown later, there is reason to believe that hypothalamic function is impaired in patients with NITS, and that this may, because of low TRH, result in low TSH and thus low output of thyroid hormones by the thyroid.

There is other evidence of diminished hypothalamic function in patients with serious illness. Serum testosterone drops rapidly, as does FSH and LH (47, 48). Typically serum cortisol is elevated as part of a stress response, but this is not always the case. Some patients develop hypotension in association with apparent transient central hypo-adrenalism, and have low or normal serum ACTH, and cortisol levels under 20 ug/dl. The patients respond dramatically to cortisol replacement, and may manifest normal adrenal function at a later time if they recover.
    Centrally-mediated hyposomatotropism, hypothyroidism, and pronounced hypoandrogenism were observed in a study of patients in the catabolic state of critical illness.  In these patients, pulsatile LH secretion and mean LH secretions are very low, even in the presence of extremely low circulating total testosterone and low estradiol.  Pulsatile GH and TSH secretion are also, as is known, suppressed.  IL-1b levels are normal, whereas IL-6 and tumor necrosis factor-a are elevated.  Exogenous iv GnRH partially returned the serum testosterone levels toward normal, but did not completely overcome the hypoandrogenism, suggesting that combined efficiency of GH, GNRH, and TSH secretagogues may be important in this low androgen syndrome (48.1).

The daily turnover (tissue supply) of thyroid hormone can be estimated from the serum hormone concentration, and disappearance curve of injected isotopic labelled T4 or T3. Daily degradation of T4 and T3 has long been considered the most exact methods for analyzing the supply of thyroid hormone to the body tissues. In numerous studies, there is a marked correlation with clinical status in patients with normal function, hyper- or hypothyroidism. There are few studies of thyroxine and T3 metabolism in patients with NTIS. Among those available are the outstanding studies by Kaptein et al. (49, 50), who studied a group of patients who were critically ill, all of whom had total T4 below 4 /dl, low FT4 Index, free T4 by dialysis which was low normal, and TSH which was normal or slightly elevated. In these patients, the mean T4 by dialysis was significantly below the normal mean. There was on average a 35% decrease in thyroxine disposal per day. Although the authors state that the thyroxine production rate was "normal", the T4 production rate in NTIS was significantly below the mean of 17 normal subjects (p < .005)(Table 5b-1). The metabolic clearance rate of thyroxine from serum was more rapid in the critically ill patients, which may in part be related to reduced TBG levels. In a similar study of T3 kinetics (50), free T3 was found to be 50% of normal serum values. The production rate of T3 was reduced by 83%(Table 2). Metabolic clearance rate of T3 during the period after initial distribution was actually slower than in normal subjects, in contrast to the findings with T4. These two studies document a dramatic reduction in provision of T4 and T3 to peripheral tissues, which would logically indicate that the effects of hormone lack (hypothyroidism) should be present. However, the authors observe that "use of T4 therapy would not appear to be appropriate, since there is no proof of an overt deficiency of free T4", and the "low T3 levels may be of adaptive significance in reducing protein catabolism, potentially making T3 therapy detrimental" (50). The reasons to object to this teleological analysis have been given, and whether reduced protein catabolism could be beneficial or not, will be discussed below. One study reported normal thyroidal secretion of T3 in patients with NTIS due to uremia (Table 3) (51). However, this was a calculated, rather than directly measured value, was exceedingly variable, and does not negate the extreme reduction in T3 supply due to diminished T4 T3 conversion.

T4 Entry Into Cells

Using deiodination of T4 as an index of cellular transport of T4 into rat hepatocytes, Lim et al. (52) and Vos et al. (53) found that serum

Sera from critically ill NTI patients caused reduced T4 uptake compared to control sera in one study, and the authors considered elevated NEFA and bilirubin, and reduced albumin, to play a role. Serum from patients with mild NTIS did not cause impaired deiodination of T4 and T3 (54). Inhibition of uptake of T4 into hepatocytes caused by sera of patients with NTIS also was observed by Sarne and Refetoff (55). Kinetic analysis in animals and man indicate  reduced uptake of T4 into tissues leading to reduced availability of T4 for deiodination to T3 .  Tissue uptake is an energy dependent process. There is evidence for a drop in tissue ATP levels, in both animals and man, during starvation.The diminution in tissue uptake is likely related to low ATP levels in tissues, including liver.  Generation of the bioactive form of thyroid hormone, T3, is greatly reduced in NTIS.  There is a diminution in the “reducing equivalents” available for the deiodination of T4 to T3 in liver, and presumably elsewhere, thus lowering the function of the Type I iodothyronine deiodinase (55.1).  In animals (and probably in man), there is also a drop in the level of Type I iodothyronine deiodinase enzyme  , apparently due to hypothyroidism, since it can be reversed by giving T3. Recently a study was performed on blood, liver, and skeletal muscle biopsies of patients immediately after death in intensive care unit settings.  Liver T4 Deiodinase 1 was found to be down-regulated, and T4 Deiodinase 3 was induced, especially in situations associated with poor tissue perfusion.  These changes presumably contribute to the low T3 syndrome of severe illness (55.2)
In theory reduced cellular uptake would cause tissue hypothyroidism, reduced T3 generation and serum T3 levels, and elevated serum T4. Except for the serum T4 levels, this hypothesis would explain many of the changes in hormone economy seen in NTIS, and would also suggest a need for replacement hormone therapy to provide adequate cellular hormone.
 It is likely that reduced hormone supply in NTIS is caused by multiple factors, and that reduced cell uptake is one of the factors. On the other hand, it is also clear that T4 must enter cells to some extent. The patients are not myxedematous. T4 is converted to T3, although inefficiently. In addition, T4 is rapidly converted to rT3, by an intracellular process, suggesting that entry into cells is not seriously impaired, but the pathways of intracellular deiodination are abnormal.

Thyroid Hormone In Tissues

Only one study has provided significant data on thyroid hormone in tissues of patients with NTIS (56). The general finding was of a dramatically reduced level of T3 in all tissues.(Table 4). While most samples had very low levels of T3 compared to normal tissues, some patients with NTIS showed sporadically and inexplicably high levels of T3 in certain tissues, especially skeletal muscle and heart. These levels exceeded a level that could be brought about by contamination with serum T3, and suggest, if the assays are correct, that there may have been, for some reason, a deposition of T3 in these tissues. This mysterious and important observation awaits clarification, but the main finding of this study is the generally low level of T3 in tissues.   
    Prolonged severe illness is associated with reduced tissue Diodinase 1 and increased Deiodinase 3 actvity, which contribute to lowering TS production and increasing rT3 production. Elevated expression of the D2 gene and D2 activity in skeletal muscle of prolonged, but not acute, critically ill patients was observed in the face of low circulating thyroid hormone levels, indicating that reduced D2 activity in muscle does not play a role in the pathogenesis of the low T(3) syndrome of critical illness. (Mebis L, Langouche L, Visser TJ, Van den Berghe G. J Clin Endocrinol Metab. 2007 Aug;92(8):3330-3. Epub 2007 May 15 The type II iodothyronine deiodinase is up-regulated in skeletal muscle during prolonged critical illness.)
    Information on expression of TRs in human tissues  during illness is limited. Increased expression of the mRNA for thyroid hormone receptor a1, a2, and b1 in cardiac tissue of patients with dilated cardiomyopathy has been reported.  a1 and a2 isoforms also had increased expression in ischemic heart disease.  The cause of this alteration is unknown.  Presumably it might be followed by increased protein expression, but this is not known (56.1). Thyroid hormone receptor levels in humans during NTIS are not known with certainty.  One study suggests that levels are normal.  In animals, starvation and illness are associated with a reduction in thyroid hormone receptor levels. In experimental studies in mice, LPS induces NTIS and this is associated with an early decrease in binding of the RXR/TR dimer to DNA due to limiting amounts of RXR, and later an up to 50% decrease in levels of RXR and TR protein (55.3-55.4)
    Thijssen-Timmer et al studied the expression of the isoforms TRbeta1, TRalpha1, and TRalpha2 and 5'-deiodinase in postmortem liver biopsies of 58 patients who were critically ill and died in the intensive care unit using real-time PCR.: The THRA gene encodes two isoforms of the thyroid hormone receptor, TRalpha1 and TRalpha2. The ratio of these splice variants could have a marked influence on T(3)-regulated gene expression.. All ratios of the biopsies were higher than those found in three normal liver biopsies due to an increased TRalpha1 level. The TRalpha1/TRalpha2 ratio increased with age and severity of illness Furthermore, no relation was seen between TR isoforms and the T(3). In critically ill patients the ratio of TRalpha1/TRalpha2 expression increased with age and severity of illness, possibly indicating a mechanism to enhance sensitivity to T(3) in the oldest and sickest patients(55.5).

Are Patients with NTIS Hypothyroid?

It is straightforward that the usual clinical parameters of hypothyroidism are absent in patients with NTIS. However, these patients usually present with an acute illness and are diagnostically challenging in view of their complicated state.. Many are febrile, have extensive edema, have sepsis or pneumonia, may have hypermetabolism associated with burns, have severe cardiac or pulmonary disease, and in general, have features that could easily mask evidence of hypothyroidism. Further, the common clinical picture of hypothyroidism does not develop within even 2 - 3 weeks, but rather requires a much longer period for expression (57).

General laboratory tests are also suspect. Thus starvation or disease-induced alterations in cholesterol, liver enzymes, TBG, CPK, and even BMR generally rule out the use of these associated markers for evidence of hypothyroidism. Angiotensin converting enzyme levels are low (58), as seen in hypothyroidism, while TEBG and osteocalcin levels are not altered (59).

It is probable that the cause of NTIS is multifactorial, and may differ in different groups of patients. Specifically, the changes in liver disease and renal disease are probably somewhat different from those occurring in other forms of illness (v.i.).

Certainly one important cause of the drop in serum T3 (clearly proven in animals, and probable but not proven in man) is a decreased generation of T3 by Type 1 iodothyronine deiodinase in liver and reduced degradation of reverse T3. The net result is a drop in serum T3, and if substrate thyroxine is present in sufficient amount, an increment in serum levels of reverse T3. This drop in T3 is induced by starvation, especially by carbohydrate starvation, and is possibly related to the reduction in "reducing equivalents" needed in the liver in the enzymatic process for thyroxine deiodination to T3 (60). Possibly, as described above, entry of thyroid hormone into cells is abnormal, so that T4 substrate is not adequately provided to the intracellular enzymes. However, it is logical to assume that, if reduced entry into cells was a primary event and the major problem, then serum T4 levels would become elevated rather than suppressed. Some studies have suggested that individuals with NTIS may have selenium deficiency and that this may contribute to a malfunction of the selenium- dependent iodothyronine deiodinase (61). However supplements of 500 mg of selenium, given to patients in a surgical ICU during the first five days after serious injury, caused only modest changes in thyroid hormones, but an earlier normalization of T4 and reverse T3 levels.  The data did not suggest a major role for selenium deficiency in this condition (61a).

As described above, another major hypothesis is that part of the change in serum hormone levels is due to the presence of inhibitors of binding of T4, and perhaps T3, to serum proteins. This evidence has been discussed above, and need not be reviewed again here. The most compelling evidence against this concept as a major problems in humans is the observations by Brent and Hershman (40). Repletion of thyroxine intravenously served to elevate T4 levels to normal in patients with NTIS. Seemingly this rules out a binding inhibition as a major factor in the depression of hormone levels.

An alteration in binding of hormones to serum might logically affect fractional turnover rate of T4 and T3, but not the micrograms of hormone being degraded each day. Such an effect is seen in patients who have low TBG levels, and low serum hormone levels, but who are metabolically normal. In fact, as described above, metabolic clearance rate for thyroxine, the liters of serum cleared of thyroid hormone per day, is augmented in patients with NTIS, and is normal for T3. The changes recognized in the study by Kaptein et al (49, 50) are modest, and may reflect only an alteration in serum binding protein levels, rather than another effect. However, it is the total micrograms of T3 and T4 produced each day, rather than the kinetics, that correlate with metabolic effect.
    Schussler and co-workers have observed a sharp drop in TBG levels during cardiac bypass surgery, which their studies indicate is due to some selective consumption of TBG.  It is possible that this occurs because of activation of SERPIN proteases at sites of inflammation which cleave the TBG into an inactive form (50.1).

The overall degradation of thyroid hormone, both thyroxine and T3, is radically diminished in the NTIS syndrome in the presence of low hormone serum levels. The reduced degradation cannot produce the lowering of serum hormone levels; a primary reduction in degradation would increase serum hormone. The change in degradation must be secondary to the low hormone supply. The findings from turnover studies, in terms of ug of hormone used by the body each day, are similar to observations inpatients with severe hypothyroidism.

Considerable evidence suggests that an alteration in hypothalamic and pituitary function causes the low production of T4,which in turn causes the low production of T3. In rats, starvation reduces hypothalamic mRNA for TRH, reduces portal serum TRH, and lowers pituitary TSH content (62). A recent study documents low TRH mRNA in hypothalamic paraventricular nuclei (63) in NTI patients(Fig 5b-4). Responses to administered TRH vary in different reports, being suppressed or even augmented (43, 44). Administration of TRH has been suggested as an effective means of restoring serum hormone levels to normal in individuals with NTIS. A recent report of great significance by Van den Berghe and co-workers proves that administration of TRH to patients with severe NTIS leads directly to increased TSH levels, increased T4 levels, and increased T3 levels(Fig.5b-5, see below) (64). This data is strong support (albeit not proof) for the role of diminished hypothalamic function as an important factor causing NTIS.

Figure 5b-4: In situ hybridization study demonstrating mRNA for TRH in the periventricular nuclei of a subject who died with NTIS in Panel A, and a subject who died accidentally in Panel B. mRNA for TRH is significantly reduced in patients with NTIS. (Reference 63)

Figure 5b-5: The study demonstrates the effect of 1 µg/kg/hr infusion of TRH compared with placebo, or TRH plus GHRP-2 (1 µg/kg/hr), or the combined treatment. Values for mean serum TSH, basal and pulsatile TSH secretion, are shown in the upper panel, and 24 hour changes in peripheral thyroid hormone levels in the three study groups are shown in the lower panel. TRH infusion increased TSH secretion, TSH concentration, T4, T3, and rT3. (Reference 64)

Quite possibly the production of TRH, and responses to TRH, are reduced by cytokines, to be discussed below, or by glucocorticoids (65). The diurnal variation in glucocorticoid levels at least in part controls the normal diurnal variation in TSH levels, perhaps by affecting pituitary responsiveness to TRH (66). High levels of glucocorticoids in Cushing's disease suppress TSH and cause a modest reduction in serum hormone levels (67). High levels of glucocorticoids are known to suppress pituitary response to TRH in man (65). Stress induced elevation of glucocorticoids in animals causes suppression of TSH and serum T4 and T3 hormone levels (68). Thus, possibly, stress induced glucocorticoid elevation may be one factor affecting TRH and TSH production.

Pituitary production of TSH is probably radically suppressed in most patients with the euthyroid sick syndrome, who have low levels of TSH in the presence of reduced levels of serum T3 and T4. At a minimum, pituitary responsivity must be abnormal, considering that TSH is normal or suppressed when it should be elevated, in the presence of low serum hormone levels. So far as we have been able to ascertain, no studies on the effect of administered human TSH have been reported. (NTIS may constitute yet another use of rhTSH.)

Why should pituitary production of TSH be diminished in the presence of low serum thyroid hormone levels? One idea, without proof, is that it represents a response to hyperthyroidism, which has not been documented. Another possibility is that there is augmented intrapituitary conversion of T4 to T3, thus allowing the pituitary to remain "euthyroid", while the rest of the body is actually hypothyroid. There is experimental support for this idea in a uremic rat model of NTIS (69). Another suggestion is that some other metabolite of thyroxine may be involved in control of pituitary responsiveness. For example, possibly triac or tetrac generated by metabolism of thyroxine could control pituitary responsiveness (70,70a), but there is no experimental proof of this idea, and even if true, would mean that the pituitary was normal but the rest of the body hypothyroid. As suggested above, elevated serum cortisol levels could play a role. The most obvious possibility is that low TSH stems from diminished TRH production, as described above. It must also be remembered that the defect in pituitary function is not restricted to TSH, but that LH, and FSH, are also suppressed in seriously ill patients, and testosterone is reduced, in contrast to the generally augmented glucocorticoid response. Quite possibly these changes are the effect on the hypothalamus of neural integration of multiple factors including stress, starvation, glucocorticoids, and cytokines.
Vandenberg  has proposed that the changes in endocrine function seen during severe illness have a biphasic course.   Quite possibly the initial suppression of T3 levels   represents a genetically engineered adaptive response of the organism, allowing reduced metabolic rate, and conservation of energy and protein stores for a longer period of time, while the animal or man goes through a period of starvation.  As noted, this is not known to be associated with increased mortality.  However, the circumstances surrounding  severe illness, and the resuscitative efforts applied in an intensive care unit over one or more weeks,  presumably have not resulted in some genetically induced metabolic response, since survival under such extreme organ failure is a very recent phenomenon.  This second phase of the syndrome, with associated suppression of thyroid hormone and other pituitary hormones, and a variety of other changes, represents in this construction a maladaptive response.  Patients in this situation tend to have elevated insulin levels, nitrogen wasting, retention of fats if calories are made available, and a variety of other metabolic abnormalities including neuropathy and cardiomyopathy.  These authors consider that provision of multiple hormonal support, including thyroid hormone, growth hormone, and androgens, may be beneficial(70.1,72.2)

Much current attention is centered on the role of cytokines in developing the euthyroid sick syndrome, through an effect on the hypothalamus, the pituitary, or possibly elsewhere. In a series of septic patients studied shortly after admission to an ICU, total T4, free T4, total T3, and TSH were depressed, and IL-1b, sIL-2R, IL-6, and TNFa were elevated(70b).  The data suggest central suppression of TSH as the cause of the problem, but the relation to cytokines is unclear, as seen in  the following reports.  The hypothalamo-pituitary-adrenal axis was activated as expected (13). Hermus et al. (71) showed that continuous infusion of IL-1 in rats cause suppression of TSH, T3, and free T4. Higher doses of IL-1 were accompanied by a febrile reaction and suppression of food intake, which presumably played some role in the altered thyroid hormone economy. IL-1 did not reproduce the diminution in hepatic 5'-deiodinase activity believed to be so characteristic of NTIS. IL-1 is also known to impair thyroid hormone synthesis by human thyrocytes, and is enhanced in many diseases associated with NTIS (73). Van der Poll et al. (74) studied the effect of IL-1 receptor blockade in human volunteers, to determine if it could alter the NTIS induced by endotoxin. Blockade of IL-1 activity was achieved by infusing recombinant human IL-1 receptor antagonist, but this did not prevent the drop in T4, free T4, T3, and TSH, or rise in reverse T3 caused by endotoxin. This is evidence against an important role for IL-1.

Interleukin-2  is produced during inflamatory responses. In a study in patients with HIV infection, administration of  (9,000,000 IU per day) caused an increase in TSH, and FT4 for four days, with normalization after two weeks.   This study was done in asymptomatic HIV infected patients, and its relevance to the euthyroid sick syndrome is uncertain, since it seems to establish a stimulatory effect on thyroid function (73.1). 

TNF is another pro-inflammatory cytokine that is thought to be involved in many of the illnesses associated with NTIS. Infusion of recombinant TNF in man, by Vanderpool et al., produced a decrease in serum T3 and TSH, and increase in rT3. Free T4 was transiently elevated in association with a significant rise in FFA levels. These studies suggest that TNF could be involved when recombinant IL-6, given to humans, activates the hypothalamic pituitary axis and, as noted above, this could secondarily suppress TSH production. However, Chopra et al. (76) did not find TNF to be closely correlated with hormone changes in NTIS.  Van der Poll et al (73.1) gave human subjects endotoxin, which caused lowering of T4, free T4, T3 and TSH. TNF blockade by a recombinant TNF receptor-IgG fusion protein did not alter the response, indicating that TNF did not cause the changes in hormone econony induced by administration of endotoxin.

Serum IL-6 is often elevated in NTIS (77), and its level is inversely related to T3 levels (78). Stouthard et al. (79) gave recombinant human IL-6 chronically to human volunteers. Short term infusion of IL-6 caused a suppression of TSH, but daily injections over 42 days cause only a modest decrease in T3, and a transient increase in reverse T3, and in free T4 concentrations(Fig.5b-6). IL-6 could be involved in the NTIS syndrome, although the mechanism was not defined. In an animal model of NTIS studied by Wiersinga and collaborators (80), antibody blockade of IL-6 failed to prevent the induced changes in thyroid hormone economy typical of NTIS. Boelen et al. studied the levels of IFN , IL-8, and IL-10 in patients with NTIS and found no evidence that they had a pathogenic role (81). Short term administration of recombinant IFN-gamma to normal subjects caused a minimal elevation of Il-6, no alteration in TNF, and did not significantly alter thyroid hormone levels(81.1). Michalaki et al observed that serum T3 drops early after abdominal surgery as an early manifestation of the NTIS syndrome, prior to an increase in serum IL-6 or TNFa, suggesting that these changes in cytokines do not induce the drop in T3 (81.2).

Figure 5b-6: IL-6 was administered over six weeks, and changes in thyroid hormone levels and TSH were recorded. Except for a transient elevation in rT3 and minimal suppression of T3, no significant alteration in hormone levels was produced.

The potential interaction between cytokines and the hypothalamic pituitary thyroid axis is certainly complicated, and cytokines themselves operate in a network. For example, IL-1 and TNF can stimulate secretion of IL-6. Activation of TNF and IL-1 production is associated with the occurrence of cytokine inhibitors in serum, which are actually fragments of the cytokine receptor, or actual receptor antagonists. "Soluble TNF receptor" and "IL-1 RA" are receptor antagonists, which can inhibit the function of the free cytokines. These molecules are increased in many infectious, inflammatory, and neoplastic conditions. Boelen et al. (82) found evidence that the NTIS is "an acute phase response" generated by activation of a cytokine network. Soluble TNF , soluble TNF receptor, soluble IL-2 receptor antagonist, and IL-6 all inversely correlated with serum T3 levels. The authors concluded that the elevation of soluble TNF receptor and IL-6 were independent "determinants" of serum T3 and accounted for "35%" and "14%", respectively, of the change in T3. At least we can be convinced that these cytokine changes co-occur with changes in T3 and may play a pathogenic role by mechanisms yet unknown.

Altered CNS metabolism-In healthy men going through two 4.5 hour long sessions of induced hypoglycemia, TSH, fT3 and fT4 are significantly reduced  . Perinatal asphyxia, recognized by low Apgar scores, is associated with  a depression of TSH, T4 and T3, and the reductions are greatest in infants with hypoxic/ischemic encephalopaty. In this study 6 of 11 infants with FT4 < 2ng/dl died  . These data suggest that reduced substrate or O2 supply to the CNS could induce hypothalamic/pituitary dysfunction.(82.1,82.2)

Administration of glucagon to dogs caused a significant fall in serum T3, suggesting that the stress-induced hyperglucagonemia may__e a contributor to the NTIS syndrome by altering intracellular metabolism of T4 (83).

Dopamine given in support of renal function and cardiac function must play a role in many patients who develop low hormone levels while in an intensive care unit setting. Dopamine inhibits TSH secretion directly, depresses further the already abnormal thyroid hormone production, and induces significant worsening of the low hormone levels. Withdrawal of dopamine infusion is followed by a prompt dramatic elevation of TSH, a rise in T4 and T3, and an increase of the T3/rT3 ratio (78). All of these changes suggested to Van den Berghe et al. (84) that dopamine makes some patients with NTIS hypothyroid, inducing a condition of iatrogenic hypothyroidism, and that treatment (presumably by administering thyroid hormone), "should be evaluated".

Leptin plays a key role in control of thyroid hormone levels during starvation in animals.  During starvation, leptin levels drop.  With this there is diminished stimulation of TRH, thus diminished secretion of TSH, and lowered thyroid hormone levels.  Administration of leptin appears to work via the arcuate nucleus to increase POMC expression, and works through the arcuate nucleus of the hypothalamus to induce production of POMC and thus aMSH, and reduce AgRP.  aMSH normally stimulates the melanocortin 4 receptor (MC4R), whereas AgRP suppresses it.  Presumably through these actions, a lack of leptin during starvation leads to diminished stimulation of the MC4R receptor on the TRH neurons in the pair of ventricular nuclear center, and thus diminished TRH secretion.  Administration of leptin partially reverses this sequence.   These actions appear to be part of an energy conserving scheme related to thyroid changes during starvation and are associated with leptin-induced increase in appetite, decreased energy expenditure, and modified neuroendocrine function.  Naturally the relevance of this to human physiology is as yet unclear, but the data strongly suggests that leptin is involved in the down-regulation of thyroid function during acute starvation.(84a,b,C) In clinical trials, stimulation of growth hormone secretion, by GH secretogogues lead to increased insulin and leptin levels in severely ill ICU patients  . Studies of leptin levels in patients with NTIS have to date indicated they are normal or elevated, not low. (85.1,85.2)
    Atrial natriuretic peptides, including amino acids 1 – 30, amino acids 31 – 67, known as vessel dilator, and 79 – 98 (kaliuretic hormone), and 99 – 126 (atrial natriuretic hormone), derived from the ANH prohormne, decreased significantly circulating concentrations of total T4, free T4, and free T3, when given to healthy humans for 60 minutes.  A reciprocal increase in TSH lasted for two or three hours after cessation of the administration of  these hormones, suggesting that the effect was a direct inhibition of thyroid hormone release from the thyroid gland, rather than an action of the hormones upon the hypothalamus or pituitary. No data is available on these factors in NTIS (84d).

Patients with alcoholic liver disease, as reported by Walfish et al. (85), tend to have low serum T3 levels, slightly reduced T4, and elevated FT4 Index because of low binding proteins. These changes were associated with increased mortality. In chronic biliary cirrhosis and chronic active hepatitis, as studied by Liewendahl (86), elevated TBG may be found associated with normal free T3 and free T4 levels. Chopra et al. (87) studied patients with hepatic cirrhosis and found free T4 to be significantly elevated, T3 to be markedly reduced, free T3 to be low, and TSH tended to be slightly above normal. Assessment of a variety of clinical parameters suggested that the patients were euthyroid. The authors conclude that, in this instance, euthyroidism is maintained by the high normal or slightly elevated serum free T4 levels. It should be noted that the mean free T4 levels in the patients studied by Chopra was 3.9 ng/100 ml, which falls well within the range of normal given by the authors of 1.8 - 4.2 ng/dl, and is not characteristic of NTIS. It is probable that some of the distinctive effects of liver disease on thyroid hormone economy are due to changes in synthesis of TBG, possibly the effect of hyper-estrogenism, and probably reduced deiodination of T4 to T3 in the liver.

Kaptein et al. (88) studied patients with acute renal failure (ARF) and found decreased serum T4 and T3, and normal or elevated levels of free T4 and TSH in patients with ARF but not critical illness. In this group of patients, reverse T3 levels tend to be normal. Ramirez et al. (89) studied patients on chronic hemodialysis and found a striking prevalence of goiter (58%), low serum thyroxine and T3, and TSH. TRH caused an increase of serum TSH and T3 levels, suggesting a suppression of pituitary function in these patients. Lim and co-workers (90) studied the thyroid hormone supply in a uremic rat model and found changes similar to those seen in uremic man, including a low serum T3, low serum T4, low serum TSH, and low liver T3 content. Triiodothyronine treatment of the animals increased low liver enzyme activity, and the authors conclude that the reduction in liver T3 content in the uremic rat, and the low enzyme activity, indicates hypothyroidism. T3 nuclear receptor binding capacity was also reduced in the uremic rat livers. Further studies found that pituitary T3 content was normal. Thus they hypothesize that pituitary type 2 deiodinase maintains an adequate level of T3 so that the pituitary is "euthyroid", while the rest of the body is "hypothyroid". In further studies, they presented data that intrapituitary T4 T3 deiodination is selectively increased in these animals (69). Not surprisingly, administration of 0.8 g T3/kg daily to uremic men increased nitrogen excretion, from increased protein catabolism (91). Presumably this is evidence for repair of hypothyroidism, and, if it represents a significant problem, could be covered by increased protein intake.

Clearly the question of exact free T4 levels in patients with NTIS remains uncertain, and most likely will be shown to be variable. In many patients all tests indicate that the hormone levels are low. Considering the range of assays applied, and their different response to inhibitors, it seems unlikely that inhibitors of T4 and T3 binding to serum proteins are universally important, causing a test artifact. There is no clear cut evidence for the role of any specific inhibitor, except possibly in uremic patients, or in patients previously treated with heparin (whose sera develop elevated FFA levels during in vitro dialysis). In point of fact, if the concept of heparin- induced FFA generation during dialysis proceedures is valid, it would produce an artifact contrary to that commonly offered to explain serum hormone discrepancies. In this case the usual T4 and FTI measurments would be reliable, but the determination of free T4 would be falsely elevated. Further, the test artifact hypothesis cannot explain the low T3, the suppressed TSH, or the low production of T4 and T3 in patients with NTIS.

The arguments against the binding inhibitor playing an important role have been spelled out above and in previous sections of this review. The salient points are that a binding inhibitor could not explain more than a fragment of the observed abnormalities, since it does not explain the reduced generation of T3, the low T3 levels, the low TSH levels, or the low production of T4 and T3. Most importantly, it is contradicted by the direct observation that replacement of T4 in patients with NTIS causes a return of serum T4 levels to normal in the patients reported by Brent and Hershman (40).

There is suggestive evidence that tissue hypothyroidism occurs because of low supplies of serum T4 and T3, low production levels of T4 and T3, and low tissue levels of T4 and T3. Much of the current research involving cytokines suggests the ability of these agents to induce a condition which is associated with low hormone supply in tissues. Nevertheless, absolute proof that tissues are chemically hypothyroid in humans with NTIS is clearly lacking as of this moment, primarily because such "tissue markers" are not available.

Assuming for the sake of argument that tissue hypothyroidism is present, can we assume that this is physiologically beneficial? We cannot take it for granted that metabolic changes occurring during illness are beneficial. Thus hyponatremia, hypoventilation, fever, hyper-metabolism of burn injury, and an endless array of other effects of illness are physiologically maladaptive. There are only two possible ways that we can know that the changes in NTIS are beneficial. The first is "revelation", and implies that we are given information, from a source that designed the system, that it is a beneficial response. This is not readily available! The second approach would be by obtaining convincing experimental evidence that the changes in thyroid economy lead to better physiologic performance. In contrast, the changes in thyroid hormone levels in NTIS, when they are extreme, are clearly associated with a marked increase in morbidity. If anything, the changes are associated with maladaptation (decreased survival), rather than beneficial adaptation. Of course correlation does not prove causation.

Much of the basis for the argument that the changes are an adaptive mechanism have to do with the modest changes in thyroid hormone levels occurring in starvation. Even here the evidence is at best cloudy. With caloric restriction and weight loss, there is a modest drop in resting metabolic rate of about 10%, while serum T3 levels drop nearly 50% (92, 93). In animals, starvation induces a reduction in T3 binding capacity of the T3 nuclear receptors in liver due to a reduction in the quantity of nuclear receptor protein present (94). In rats the adaptation to starvation includes a decrease in TRH levels in hypothalamic portal blood, and thus decreased hypothalamic TRH synthesis and release, leading to decreased TSH production (62). Sanchez found that, in the brain, starvation did not alter content or binding capacity for T3, but illness (diabetes) did cause a decrease in TR content and T3 binding capacity of glial cell nuclei (95). This suggests that a decline in serum T3 during hypocaloric feeding is like hypothyroidism, and obviously this could be "adaptive". The fall in serum T3 during hypocaloric feeding in humans was shown by Osburne et al. (96) to cause apparent hypothyroidism as determined by "timing of the arterial sounds", and a decrease in pulse rate. Replacement doses of T3 (30 g/day), or T4 (100 g/day), promptly reversed these abnormalities. Gardener et al. found that fasting in normal males decreased serum T3 (97). Administration of 5 g T3 every 3 hours (40 g/day) brought T3 back to slightly higher than normal prefasting levels, and urea excretion was augmented. They suggested that the fasting induced reduction in T3 spared nitrogen. Burman et al. (98) conducted similar studies and showed decreased muscle catabolism during fasting, which was reversed by feeding doses of T3 which induced mild hyperthyroidism (60 - 100 g/day). Byerley and Heber (99) has presented rather contrasting data. During starvation in normal subjects, metabolic rate and CO₂ production decreased, but did not increase after T3 supplementation. Urinary nitrogen excretion decreased during fasting and did not increase with T3 supplementation (30 g of T3 qd). Their data suggests that the drop in T3 does not mediate the protein sparing found in fasting.

Thus it is clear that the fasting induces a drop in BMR, reduces nitrogen loss, and tends to decrease T3 levels, but replacement of T3 does not return the BMR to normal or necessarily alter protein metabolism. From these studies it cannot be proven that a drop in T3 exerts a specific "adaptive", physiological, protein sparing effect during fasting, although this remains a reasonable possibility. Even granted that this is true, any relationship of this to NTIS is extremely problematical. The changes in thyroid hormone supply induced by short term fasting in man are very modest and not comparable to the severe drop in hormone supply found in severely ill patients with T4 under 4 g/dl. Nor is there any evidence that such al decrease in T3 (in the absence of a drop in T4) increases the probability of death, as occurs in severe NTIS. Aside from the uncertainty about the relationship of T3 to protein sparing, and the lack of comparability to severe NTIS, a third more important point argues against the relevancy of this information in considering therapy for NTIS. While short term starvation is allowed in patients undergoing mild surgical intervention, or who present to the hospital with acute illness, starvation is not allowed to continue during illness. Patients are promptly supplemented with glucose, vitamins, lipids, amino acids, and every factor needed by every route possible, in order to maintain appropriate nutrition. Thus, while starvation may occur, it is not an accepted part of medical management of patients with NTIS, and in general NTIS patients are not, or at least should not be, starving.

Two valuable studies are available on replacement therapy using thyroid hormone in patients with NTIS.  In the study by Brent and Hershman (40), replacement with 1.5 g T4 i.v. per kilogram body weight, in 12 patients, promptly returned serum T4 levels to normal (thereby proving that a binding defect was not the cause of the low T4) , but did not normalize T3 levels over a period of 2 - 3 weeks. However, in both the treated and control group, mortality was 80% (40). Clearly, in this excellent small study, which used for primary therapy what would now be considered the wrong hormone, failed to show either an advantageous, or disadvantageous, effect. One can argue that the failure to show a positive effect was due to the failure of T3 levels to be restored to normal. In a study of severely burned patients given 200 ug T3 daily, again there was no evidence of a beneficial or disadvantageous effect (100). Mortality was not so great, as in the Brent and Hershman study, but it is entirely possible that the high levels of T3 provided worsened the hypermetabolism known to be present in burn patients, and could have, at these levels, been disadvantageous.

 An important study by Acker et al certainly advises caution regarding T4 therapy in patients with acute renal failure. Numerous studies in animals had previously documented a beneficial effect of T4 therapy  in experimental acute renal failure(99.1). In a randomized controlled prospective study of patients with acute renal failure, they treated patients received 150 mg of thyroxine four times intravenously over two days.  The single difference recognized in the subsequent laboratory data  was a suppression of TSH. T4 treatment had no effect on any measure of ARF severity. Among other questions, it is not clear that serum T3 levels were ever altered.   However, mortality was higher in the thyroxine group (43 vs. 13%) than in the control group.  It is of interest that, as the authors state, “the observed mortality in the controls in this study was less than that typically seen in our institution in ARF and ICU patients, whereas the 43% mortality noted in the thyroid group better approximates both our experience and that reported in the literature for ICU patients.”   It will be difficult to replicate this study, but it is uncertain whether the small dose of thyroxine administered over two days actually is related to the mortality, considering that the mortality in the treated group was that usually observed, whereas the control happened to have a much lower mortality (100.1).  The same group has also studied the effect of thyroid hormone treatment on post-transplant acute tubular necrosis.  T3 treatment during the post-transplant period did not alter outcome in a beneficial or derogatory manner (100.2).

  Studies from animals are often quoted in the literature as an argument against treatment of NTIS, or for the therapy. A study of sepsis induced in animals showed no difference in mortality, but some animals treated with thyroid hormone died earlier than did those that were untreated (101). Chopra et al. induced an NTIS in rats by injection of turpentine oil. The reduction in T4, T3, Free T4 Index, and TSH were associated with no clear evidence of tissue hypothyroidism, and urinary nitrogen excretion was normal. Thyroid hormone replacement with T4 or T3 did not significantly alter enzyme activities or urinary nitrogen excretion (102). Healthy pigs were subjected to 20 minutes of regional myocardial ischemia by Hsu and collaborators (103), and this was associated with a drop in T3, free T3, and elevated rT3. Some animals were treated with 0.2 g T3 per kilogram for five doses over two hours. While myocardial infarction size was not altered, the pigs treated with T3 showed a more rapid improvement in cardiac index (103). Oxygen consumption did not alter. It should be noted that the T3 levels fell back to normal levels within four hours of the last T3 dose, suggesting that more prolonged therapy might have been beneficial. Katzeff  et al (103.1) studied a model of NTI induced by caloric restriction in young rats. In these animals T3 was reduced, and there was a decrease in LV relaxation time, SERCA2 mRNA, and alpha-MHC mRNA. All changes were were reversed to normal values by supplementation with T3, suggesting that the low-T3 syndrome was related to the pathological cardiac c