Endocrinology in the critically ill
Critical illness is any condition requiring support of failing vital organ systems, without which survival would not be possible. It is characterized by striking alterations in the hypothalamic-anterior-pituitary axes that are known to contribute to the high risk of morbidity and mortality.
For a long time, these endocrine changes were considered to be part of a uniform stress response that is sustained throughout intensive care and that reflects a beneficial adaptation of the human body, contributing to survival. However, it has become clear that this is not correct. Research during the past years has elucidated a biphasic neuroendocrine response to critical illness (1, 2). During the acute phase of critical illness, an actively secreting pituitary, together with the development of target-organ resistance, results in low concentrations of peripheral effector hormones. These endocrine alterations may reduce energy and substrate expenditure, an effect likely to be beneficial for short-term survival.
About 30% of critically ill patients do not recover within a few days, and instead enter a chronic phase of critical illness, during which they remain dependent on vital-organ support and face a more than 20% risk of death. The high mortality observed during this prolonged phase is usually attributed to nonresolving failure of multiple organ systems and vulnerability to infectious complications, rather than to the type or severity of the disease for which patients were originally admitted to the intensive care unit. During the prolonged phase of illness, low serum levels of peripheral effector hormones are caused by uniform suppression of the neuroendocrine axes, primarily of hypothalamic origin. The prolonged phase of critical illness is further characterized by persistent hypercatabolism, despite feeding, which leads to a substantial loss of lean body mass in the presence of relative preservation of adipose tissue. This ‘wasting syndrome’ is likely to compromise vital functions and delay recovery, and as such to contribute to the increased morbidity and mortality.
The different patterns of the neuroendocrine responses in the acute and prolonged phase of critical illness underlie the pathophysiology of these neuroendocrine changes. Indeed, erroneous extrapolation of the changes observed in the acute-disease state to the prolonged phase of critical illness has misled investigators to apply certain endocrine treatments that unexpectedly increased rather than decreased mortality (3, 4). In addition, patients admitted to intensive care units may suffer from pre-existing central and/or peripheral endocrine diseases, making the puzzle even more complex and contributing to the major challenge of endocrine function testing in a critically ill patient. Furthermore, the inability to identify the neuroendocrine changes either as adaptation or as pathology makes the issue of treatment even more controversial. Therefore, knowledge of the underlying pathophysiology is of vital importance for the development of therapeutic interventions to correct these alterations, and to open perspectives to improve survival (Box 10.2.3.1).
In this chapter, an overview is given of the dynamic neuroendocrine alterations that occur during the course of critical illness. In addition, it highlights the complexity of the differential diagnosis with pre-existing endocrine diseases, and the available evidence of benefit and/or harm of some endocrine interventions.
Pathogenesis, clinical features, and treatment options
The Somatotropic Axis
Growth hormone, which is secreted by the somatotropes in the anterior pituitary, is essential for growth during childhood, and serves a number of other important, mainly anabolic functions throughout life. The pulsatile nature of growth hormone release, with peak serum levels alternating with virtually undetectable troughs, is important for its metabolic effects. The release of growth hormone is stimulated by the hypothalamic growth hormone-releasing hormone (GHRH), and is inhibited by somatostatin. Moreover, several synthetic growth hormone-releasing peptides (GHRPs) and nonpeptide analogues with potent growth hormone-releasing activity have been developed (5). A highly conserved endogenous ligand of the growth hormone secretagogue receptor is ghrelin, which originates both in peripheral tissues and in the hypothalamic arcuate nucleus. Ghrelin appears to be a third key factor in the complex physiological control of pulsatile growth hormone release (6).
Apart from its direct actions, growth hormone also exerts indirect effects that are mediated mainly through insulin-like growth factor 1 (IGF-1), the bioactivity of which is regulated by several IGF-binding proteins (IGFBPs).
The somatotropic axis in acute critical illness
The first hours to days after an acute insult, such as surgery, trauma or infection, are hallmarked by a dramatically changed growth hormone profile (Fig. 10.2.3.1) (1, 2, 7). The growth hormone pulse frequency is increased, and the peak levels and interpulse concentrations are high. Concomitantly, a state of peripheral growth hormone resistance develops, which is suggested to be triggered by cytokines such as tumour necrosis factor-alpha (TNF-α), interleukin-1 (IL-1), and interleukin-6 (IL-6). Serum concentrations of IGF-1, growth hormone-dependent IGFBP-3, and the acid-labile subunit (ALS) of the ternary complex are low during the acute phase of critical illness, in spite of the clearly enhanced growth hormone secretion. An enhanced clearance of IGF-1, in part related to elevated circulating levels of small IGFBPs such as IGFBP-1, IGFBP-2, and IGFBP-6, also contributes to its low serum levels. These events are preceded by a decrease in serum levels of GHBP, which is thought to reflect reduced growth hormone receptor expression in peripheral tissues.
It remains unclear as to which factor ultimately controls the stimulation of growth hormone release in response to acute stress. Nevertheless, it can be inferred that reduced negative feedback inhibition, caused by reduced expression of the growth hormone-receptor and subsequent low levels of circulating IGF-1, is the primary event inducing the abundant release of growth hormone in the acute phase of illness. The high growth hormone levels may then exert direct lipolytic, insulin-antagonizing, and immune-stimulating actions, resulting in increased fatty acid and glucose levels in the circulation, whereas the indirect, IGF-1-mediated effects of growth hormone are attenuated. This explanation is plausible in that such changes prioritize essential substrates such as glucose, free fatty acids, and amino acids (glutamine) toward survival rather than costly anabolism, which is mainly mediated by IGF-1 and considered less vital at this time. Therefore, from a teleological point of view, the response within the growth hormone axis to acute illness seems highly appropriate in the struggle for survival.
The somatotropic axis in prolonged critical illness
When recovery is not achieved within a few days and patients enter a prolonged phase of critical illness, different changes are observed within the somatotropic axis (1, 2). The nonpulsatile fraction remains somewhat elevated, and although the number of pulses is still high, the pulsatile release of growth hormone is strongly suppressed (Fig. 10.2.3.1). Indeed, the mean nocturnal growth hormone serum concentrations are scarcely elevated compared with the healthy, nonstressed condition, and are substantially lower than in the acute phase of stress. Furthermore, although the growth hormone resistance of acute illness may be partially reversed in the chronic phase, as indicated by increased serum levels of GHBP, the levels of IGF-1, IGFBP-3, and ALS are even lower in prolonged critical illness patients. There is a strong positive correlation between the pulsatile fraction of growth hormone secretion and circulating levels of IGF-1, IGFBP-3, and ALS in this phase, meaning that the smaller the growth hormone pulses, the lower the circulating levels of growth hormone-dependent IGF-1 and ternary complex binding proteins. This clearly no longer represents a pure state of growth hormone resistance, and suggests that loss of pulsatile growth hormone release in the prolonged phase of critical illness contributes to the low levels of IGF-1, IGFBP-3, and ALS. Since the robust release of growth hormone in response to growth hormone secretagogues (GHS) (Fig. 10.2.3.2) excludes a possible inability of the somatotropes to synthesize growth hormone, the origin of the relative hyposomatotropism is likely located within the hypothalamus. Furthermore, the release of growth hormone in response to GHRH injection appears to be less pronounced than that to GHRP-2 injection in prolonged critical illness, suggesting that a hypothalamic deficiency or inactivity of endogenous GHRP-like GHS is a more plausible cause of the hyposomatotropism than is GHRH deficiency.
Chronic relative growth hormone deficiency is believed to contribute to the pathogenesis of the ‘wasting syndrome’ that characterizes prolonged critical illness (1, 2). This is suggested by the observation that low serum levels of IGF-1 and ternary-complex-binding proteins (IGFBP-3, ALS, and IGFBP-5) are closely correlated to biochemical markers of impaired anabolism, such as low serum osteocalcin and leptin concentrations, during prolonged critical illness. Furthermore, although total growth hormone output is indistinguishable between male and female patients, there appears to be a gender dissociation; men show a greater loss of pulsatility and regularity within the growth hormone secretion pattern than women, and concomitantly have lower circulating IGF-1 and ALS levels (1, 2). However, it remains unknown whether there is a casual or a causal association between this paradoxical sexual dimorphism within the growth hormone/IGF-1 axis, and the fact that males seem to be at higher risk of an adverse outcome of prolonged critical illness.
Therapeutic interventions: treatment with growth hormone during critical illness
The assumption of sustained growth hormone resistance in the presence of normal or adaptively altered pituitary function during the catabolic condition of prolonged critical illness, was the main rationale for the administration of pharmacological doses of growth hormone, in an attempt to restore anabolism in intensive care patients. However, a large multicentre study that investigated the effects of this high-dose growth hormone treatment in prolonged critical illness patients found that instead of improving outcome, this intervention increased both morbidity and mortality (4). Since it is clear now that the growth hormone resistance of acute illness is at least partially resolved in the prolonged phase, it is likely that administration of these high doses of growth hormone evoked toxic side effects. Indeed, high doses of growth hormone administered in the prolonged phase of critical illness can induce supranormal IGF-1 levels, excessive fluid retention, hypercalcaemia, and pronounced insulin resistance with hyperglycaemia. As a consequence, the glucose counter-regulatory side effects may have exceeded any possible beneficial effects of this therapy. In addition, in view of the broad spectrum of growth hormone target tissues, and taking into account the pre-existing impairment of vital organ functions during critical illness, the excessive doses of growth hormone may have further deteriorated the function of multiple organs. Another treatment is the combined administration of growth hormone and IGF-1, which are additive in their anabolic actions and neutralize each other’s side effects (8). Furthermore, treatment with hypothalamic releasing factors to reactivate the pituitary rather than administration of pituitary or peripheral hormones may be more effective and safer (1, 2). Indeed, infusions of GHS not only restored pulsatile growth hormone secretion, but also increased IGF-1, IGFBP-3, and ALS, which is indicative of restored peripheral responsiveness.
The thyroid axis
Thyroid hormones play a key role in the regulation of energy and substrate metabolism, and are essential for the stimulation of normal growth and development (9). Thyrotropin-releasing hormone (TRH) is secreted by the hypothalamus and stimulates the pituitary thyrotropes to produce and secrete thyroid-stimulating hormone (TSH). TSH in turn drives the thyroid gland to synthesize and secrete thyroid hormones. Although these comprise mainly thyroxine (T4), the biological activity of thyroid hormones is largely exerted by triiodothyronine (T3). Different types of deiodinases (D1-D3) are responsible for the peripheral activation of T4 to either T3 or to the biologically inactive reverse T3 (rT3). TRH and TSH secretion are controlled by negative feedback from the thyroid hormones.
The thyroid axis in acute critical illness
Early after the onset of severe physical stress, the thyroid axis responds with a rapid decrease in serum levels of T3 and an increase of rT3 levels, predominantly because of altered peripheral conversion of T4 (1, 2, 10). TSH and T4 levels are elevated briefly and subsequently return to normal, though in the more severely ill patients T4 levels may also fall (Fig. 10.2.3.3). Although at this point mean serum levels of TSH are normal, the TSH profile is already affected as shown by the absence of the normal nocturnal TSH surge. The low T3 levels persist beyond TSH normalization, a configuration often referred to as ‘the low T3 syndrome’. The magnitude of the T3 decrease within 24 hours after the insult is related to the severity of illness, and appears to correlate with mortality.
The cytokines TNF-α, IL-1, and IL-6 have been proposed to play a role in the pathogenesis of the low T3 syndrome. Although they are capable of mimicking the acute stress-induced response of the thyroid axis, cytokine antagonists failed to restore normal thyroid function after endotoxaemic challenge (12). Other factors that have been investigated as potential triggers for the low T3 syndrome at the tissue level include low concentrations of thyroid hormone binding proteins and inhibition of hormone binding, transport, and metabolism by elevated levels of free fatty acids and bilirubin.
The alterations observed in the thyroid axis during acute critical illness are similar to those seen during fasting. Indeed, during starvation, the immediate fall in circulating T3 has been regarded as an attempt by the body to reduce its energy expenditure and prevent protein wasting. Therefore, it could be interpreted as a beneficial and adaptive response that warrants no intervention. Although acute illness is also accompanied by temporary starvation, the validity of extrapolating this interpretation from simple starvation to acute critical illness remains a controversial issue. Indeed, although short-term intravenous T3 administration to patients during elective coronary bypass grafting improves postoperative cardiac function (13), the doses of T3 resulted in supranormal serum T3 levels, which supports, but does not prove, an adaptive nature of the ‘acute’ low T3 syndrome.
The thyroid axis in prolonged critical illness
Patients who remain in the intensive care unit and enter a prolonged phase of critical illness show a different set of changes within the thyroid axis (1, 2). In addition to the absent nocturnal TSH surge, the pulsatility of the TSH secretion pattern is dramatically reduced (Fig. 10.2.3.3), which is related to low serum levels of both T3 and T4. In particular, the decline in T3 correlates positively with the diminished pulsatile release of TSH. The prognostic value of the disturbed thyroid axis with regard to mortality is now illustrated by the reduced TSH, T4, and T3 levels, and the higher rT3 levels in patients who ultimately die as compared to those surviving prolonged critical illness. Possible explanations for these findings include an alteration in the set-point for feedback inhibition, an impaired capacity of the thyrotropes to synthesize TSH, inadequate TRH-induced release of TSH, or an elevated somatostatin tone. Fliers and colleagues (14) showed reduced expression of the TRH gene in hypothalamic paraventricular nuclei from prolonged critically ill patients, whereas this was not the case after death from acute insults. In addition, they observed that the TRH mRNA levels in the paraventricular nuclei correlated positively with blood levels of TSH and T3. Together, these findings indicate a predominantly central origin of the suppressed thyroid axis, which is similar to the alterations in the somatotropic axis. This concept is supported by the rise in TSH secretion and in peripheral thyroid hormone levels after TRH administration in prolonged critically ill patients (15) (Fig. 10.2.3.4). Furthermore, reduced GH secretagogue action may also be involved, as the pulsatility of the TSH secretion pattern is only improved when TRH is infused together with GHRP (1, 2). The exact mechanisms underlying the neuroendocrine pathogenesis of the low thyroid hormone levels in prolonged critical illness, however, are unknown. Circulating cytokine levels are usually much lower at this stage compared to the acute phase, indicating that other factors operating within the central nervous system are more likely to be involved. These factors may involve endogenous dopamine and prolonged hypercortisolism, as it is known that exogenous dopamine and glucocorticoids provoke or severely aggravate hypothyroidism in critical illness (1, 2).
In addition to the resetting of hypothalamic control, another factor contributing to the low T3 syndrome in the chronic phase of critical illness is a disturbed peripheral metabolism of thyroid hormone (Fig. 10.2.3.3) (16). This is indicated by a reduced activity of D1, responsible for peripheral conversion of T4 to T3, and an increase of D3 activity, which mediates conversion of T4 to inactive rT3, in prolonged critically ill patients. These alterations in enzyme activity result in a reduced ratio of active to inactive thyroid hormone (T3/rT3), indicating that changes in thyroid hormone metabolism are contributing to low T3 syndrome in the prolonged phase of critical illness. In addition, both the T3/rT3 ratio and serum levels of rT3 correlate with tissue deiodinase activity. D2 activity, on the other hand, is increased during prolonged critical illness and does not appear to play a role in the pathogenesis of the low T3 syndrome in this phase.
Interestingly, simultaneous infusion of TRH and GHRP-2 not only increased TSH, T4, and T3 levels, but also prevented the rise in rT3 seen with TRH alone (Fig. 10.2.3.4) (1, 2). These results suggest that deiodinase activity may be affected by GHRP-2, either directly or indirectly, through its effect on the somatotropic axis. In a rabbit model of prolonged critical illness, the down-regulation of D1 and up-regulation of D3 were reversed by the simultaneous administration of TRH and GHRP-2 (17). This indicates that D1 suppression in critical illness is related to alterations within the thyroid axis, whereas D3 is increased under joint control of the somatotropic and thyroid axes.
The regulation of thyroid hormone action at the level of the thyroid hormone receptor (TR) is also changed during critical illness. Alternative splicing gives rise to two TR isoforms, with TR-1 being a bona fide T3 receptor, and TR-2 acting as a dominant negative isoform. Therefore, the ratio of these splice variants may have a significant influence on T3-regulated gene expression. This is interesting in view of the changing thyroid hormone metabolism during critical illness. Recently, an inverse correlation was observed between the T3/rT3 ratio and the TR-1/TR-2 ratio in liver biopsies of prolonged critically ill patients (18). Furthermore, sicker and older patients presented with higher TR-1/TR-2 ratios compared to less sick and younger ones. These findings indicate that prolonged critically ill patients may adapt to the low T3 levels by increasing the expression of the active form of the TR gene, and in this way possibly increasing the cellular thyroid hormone sensitivity
Therapeutic interventions: treatment with thyroid hormone or releasing factors during prolonged critical illness
The acute changes within the thyroid axis, uniformly present in all types of acute illnesses, could be looked upon as a beneficial and adaptive response that does not warrant intervention. The prolonged phase of critical illness, however, is in a way an unnatural condition, brought by the development of intensive care medicine. The alterations observed during prolonged critical illness can therefore not be interpreted as merely selected by evolution, and such as it is unlikely that they represent an adaptive response. Indeed, the constellation of increased expression of the active form of the TR gene in association with the decreasing thyroid hormone levels in prolonged critically ill patients does not support an adaptive nature of the low T3 syndrome.
Nevertheless, it remains controversial whether correction of the low serum and tissue concentrations of T3 in critically ill patients by thyroid hormone administration is beneficial. Pioneering studies using T4 administration have so far failed to demonstrate clinical benefit within an intensive care setting (19). Administration of T3 substitution doses in dopamine-treated paediatric cardiac surgery patients revealed improvement in postoperative cardiac function (20). However, a benefit of T3 treatment in iatrogenic, dopamine-induced hypothyroidism still does not provide evidence for clinical benefit of treating the noniatrogenic low T3 levels that are characteristic of prolonged critical illness.
Rather than administration of thyroid hormones, a safer method for treatment of illness-associated hypothyroidism may be the infusion of hypothalamic-releasing factors, since this preserves the normal feedback systems. Indeed, by continuous infusion of TRH in combination with a GHS, not only were thyroid hormone levels restored to normal physiological levels, but markers of hypercatabolism were also reduced (1, 2). This suggests that low thyroid hormone levels contribute to rather than protect from the hypercatabolism of prolonged critical illness. In addition, the peripheral tissue responses to the normalization of serum levels of IGF-1 and its binding proteins (IGFBPs) via GHRP infusion seem to depend on the coinfusion of TRH and the simultaneous normalization of the thyroid axis (1, 2). Although infusion of GHRP-2 alone causes identical increases in growth hormone secretion and in serum concentrations of IGF-1, IGFBP-3, and ALS, none of the anabolic tissue responses, evoked by the combined infusion of GHRP and TRH, are present. Further studies will be needed to assess the clinical benefits on morbidity and mortality of TRH infusion alone or in combination with GHS in prolonged critical illness.
In view of the hypothalamic-pituitary suppression occurring during prolonged critical illness in patients with and without previous endocrine disease, it is virtually impossible to diagnose pre-existing central hypothyroidism during intensive care. Patients with pre-existing primary hypothyroidism, myxoedema coma being the extreme presentation, are expected to have low serum levels of thyroid hormones in combination with very high TSH concentrations. However, severe nonthyroidal critical illness may conceal this increase in TSH in patients suffering from primary hypothyroidism. Furthermore, serum T3 levels may be undetectable and T4 may be dramatically reduced in patients with prolonged non-thyroidal critical illness. In patients with myxoedema coma and severe comorbidity, serum T3 and T4 levels are also very low and could therefore be indistinguishable from those values observed in prolonged critical illness. Whereas serum TSH is significantly increased in uncomplicated primary hypothyroidism, it is paradoxically normal or even decreased in severely ill patients. In the severe hypothyroid condition of patients with myxoedema coma and concomitant illness, serum TSH may therefore be much lower than expected, indicating that normal or low TSH levels during intercurrent critical illness do not necessarily exclude primary hypothyroidism. Another problem that is faced when a (pre-existing) thyroid disease is suspected is the limited diagnostic accuracy of measured values of thyroid hormones or TSH. Therefore, in many patients, no definite laboratory diagnosis can be established, and further clues for the presence or absence of thyroid disease must be given by history, physical examination, and the possible presence of thyroid antibodies. To confirm the diagnosis, thyroid function tests must be repeated after recovery from nonthyroidal illness.
It still remains controversial when and how to treat primary thyroidal illness during the course of an intercurrent nonthyroidal critical illness, because controlled studies on the optimal treatment regimen are lacking. One exception is a presumed diagnosis of myxoedema coma, for which there is a general agreement that patients should be treated with a parenteral form of thyroid hormone. In any other case, the primary uncertainty relates to the type of thyroid hormone that should be given: T4, T3, or a combination of both. A second issue involves the optimal initial dose of any thyroid hormone replacement regimen. Many clinicians prefer a loading dose of 300 to 500 μg of intravenous T4 in order to quickly restore circulating levels of T4 (21), followed by 50 to 100 μg of intravenous T4 daily until oral medication can be given. Higher doses do not seem to be beneficial, but do not increase cardiovascular risk in severely ill hypothyroid patients (22). Some authors, however, suggest the use of T3 in addition to T4, because T3 does not require conversion to a biologically active form by 5′-deiodinase enzymes. Indeed, an animal experimental study showed that replacement therapy for hypothyroidism with T4 alone did not ensure euthyroidism in all tissues (23), whereas this was induced with a combined treatment with both T4 and T3. These findings may be explained by tissue-specific deiodinase activity, which acts as a local regulatory mechanism.
The lactotropic axis
Prolactin is a well-known stress hormone that is produced and secreted by the lactotropes in the pituitary in a pulsatile and diurnal pattern. The main function of prolactin is to stimulate lactation, but it is also presumed to have immune-enhancing properties. The immunosuppressive drug cyclosporine is known to compete with prolactin for a common binding site on T cells, which may explain part of its effects. Physiological regulation of prolactin secretion is largely under the control of dopamine, although it can be modulated by several other prolactin inhibiting and releasing factors (24).
The lactotropic axis in acute critical illness
Acute physical or psychological stress causes prolactin levels to rise, which may contribute to altered immune function during critical illness (1, 2). This increase is possibly mediated by vasoactive intestinal peptide, oxytocin, and dopaminergic pathways, but also by cytokines or as-yet uncharacterized factors. The rise in prolactin levels following acute stress is believed to contribute to the vital initial activation of the immune cascade early in the disease process, although this remains speculative.
The lactotropic axis in prolonged critical illness
In the prolonged phase of critical illness, the pulsatile fraction of prolactin release becomes suppressed, and serum prolactin levels are reduced compared to the acute phase (1, 2). It is unclear whether blunted prolactin secretion contributes to the immunosuppression or increased susceptibility to infections that is associated with prolonged critical illness. However, this remains a tempting speculation, since exogenous dopamine, frequently infused as an inotropic drug in intensive care-dependent patients, further suppresses prolactin secretion and concomitantly aggravates T-lymphocyte dysfunction and impaired neutrophil chemotaxis.
Therapeutic interventions: prolactin as a therapeutic target?
Despite its immune-enhancing properties, prolactin is currently not available for therapy. Further studies will be needed to evaluate the therapeutic potential of TRH-induced prolactin release for optimizing immune function during prolonged critical illness. It also remains unclear whether patients on treatment for prolactinoma should interrupt or continue this treatment during an intercurrent critical illness.
The gonadal axis
Gonadotropin-releasing hormone (GnRH) is secreted in a pulsatory pattern by the hypothalamus, and stimulates the release of luteinizing hormone and follicle-stimulating hormone (FSH) from the gonadotropes in the pituitary. Again, the pulsatility in the secretion pattern of luteinizing hormone is important for its bioactivity. In women, luteinizing hormone mediates ovarian androgen production, whereas FSH drives the aromatization of androgens to oestrogens in the ovary. In men, luteinizing hormone stimulates the production of androgens (testosterone and androstenedione) by the Leydig cells in the testes, whereas the combined action of FSH and testosterone on Sertoli cells supports spermatogenesis. In turn, sex steroids exert negative feedback control on GnRH and gonadotropin secretion. Several other hormones and cytokines are involved in the complex regulation of the gonadal axis.
As most female patients in the critical care medicine unit are of high age and thus in the menopausal stage, clinical data on the changes within the gonadal axis are scarce in critically ill women. Therefore, we will focus on the changes that are documented in critically ill men.
The gonadal axis in acute critical illness
Acute physical stress in men causes an immediate fall in the serum levels of testosterone, even though luteinizing hormone levels are elevated (1, 2, 25). This observation suggests an immediate suppression of anabolic androgen production in Leydig cells, which may be interpreted as an attempt to reduce energy consumption and conserve substrates for more vital functions. The exact cause remains unclear, but again, inflammatory cytokines (IL-1 and IL-2) may be involved.
The gonadal axis in prolonged critical illness
When critical illness is prolonged, more dramatic changes develop within the male gonadal axis. Circulating levels of testosterone become extremely low, while mean luteinizing hormone concentrations and pulsatile release are suppressed. Estimated free oestradiol concentrations were shown to remain normal in one investigation, whereas other studies observed a remarkable rise in oestrogen levels (25). Since exogenous GnRH is only partially and transiently effective in correcting these abnormalities, they must result from combined central and peripheral defects within the male gonadal axis. There appears to be an increased aromatization of adrenal androgens to oestrogens in critically ill patients (26). As testosterone is the most important endogenous anabolic steroid, changes within the luteinizing hormone-testosterone axis in males may be relevant for the catabolic state of critical illness. Indeed, prolonged critical illness, and a variety of other catabolic states, is accompanied by low serum testosterone levels in men. Also, the high luteinizing hormone pulse frequency, with abnormally pulse amplitude, in prolonged critically ill men was interpreted as impaired luteinizing hormone hypersecretion in response to the very low serum testosterone levels. Again, it seems to be mainly impairment of the pulsatile component of luteinizing hormone secretion that occurs in response to the sustained stress of prolonged critical illness. The profound hypogonadotropism may be explained by multiple mechanisms. Endogenous dopamine, opiates, and preserved levels of circulating bioactive oestradiol may be involved, because exogenous dopamine, opioids, and oestrogens further decrease blunted luteinizing hormone secretion. Furthermore, animal data suggest that prolonged exposure of the brain to increased levels of cytokines, such as IL-1, may play a role in the suppression of GnRH synthesis (27).
Therapeutic interventions: sex steroid substitution therapy during critical illness?
It remains unknown whether the profound hypoandrogenism seen in male critically ill patients reflects adaptation or pathology. Therefore, it is not clear whether androgen substitution therapy for treatment of pre-existing hypogonadism should be interrupted or continued during the course of an intercurrent critical illness.
Pioneering studies evaluating the use of androgens in prolonged critical illness failed to demonstrate any conclusive clinical benefit (28, 29). It is, however, shown that exogenous pulsatile GnRH administration in prolonged critically ill men partially overcomes the hypogonadotropic hypogonadism. Moreover, when GnRH pulses were given together with GHRP2 and TRH infusion, target organ responses and anabolic effects followed (1, 2). These data again underline the importance of correcting all of the hypothalamic/pituitary defects rather than applying a single hormone treatment.
The adrenal axis
Under normal conditions, cortisol is secreted in the adrenal cortex according to a diurnal pattern. Cortisol release is induced by adrenocorticotropic hormone (ACTH or corticotropin), which is produced by the corticotropes in the pituitary under the control of the hypothalamic corticotropin releasing hormone (CRH). In turn, cortisol exerts negative-feedback control on both CRH and ACTH. Although only free cortisol is biologically active, more than 90% of circulating cortisol is bound to binding proteins such as corticosteroid-binding globulin (CBG) and, to a lesser extent, albumin.
The adrenal axis in acute critical illness
In the early phase of critical illness, the diurnal variation in cortisol secretion is lost (1, 2). Cortisol levels usually rise in response to an increased release of CRH and ACTH, either directly or via resistance to/inhibition of the negative-feedback mechanism exerted by cortisol. In addition, CBG levels fall substantially, resulting in proportionally much higher increases in the free hormone. The changes observed in the adrenal axis may be provoked by cytokines, since they are known to modulate cortisol production as well as glucocorticoid receptor number or affinity in acute illness (30, 31).
The stress-induced hypercortisolism acutely shifts carbohydrate, fat, and protein metabolism resulting in a delay of anabolism and the acute provision of energy to vital organs such as the brain. In addition, it offers haemodynamic advantages in the fight-and-flight reflex by induction of fluid retention and sensitization of the vasopressor response to catecholamines, and it protects against excessive inflammation by suppression of the inflammatory response (1, 2).
An appropriate activation of the hypothalamic-pituitary-adrenal axis and cortisol response to critical illness appears to be essential for survival, since both very high and very low cortisol levels have been associated with increased mortality (1, 2). High levels indicate more severe stress, whereas low levels reflect an inability to sufficiently respond to stress, also labelled ‘relative adrenal insufficiency’.
The adrenal axis in prolonged critical illness
In the prolonged phase of critical illness, hypercortisolism is usually sustained, but serum ACTH levels decrease, indicating that cortisol release and/or production may in this phase be driven by non-ACTH-mediated pathways (30, 31). Cortisol levels decrease slowly during chronic illness, but only reach normal levels during the recovery phase. CBG levels already recover during the chronic phase of critical illness.
The origin of the dissociation between ACTH and cortisol levels during prolonged critical illness is unclear, but a role for atrial natriuretic peptide or substance P has been suggested (32). Other possibilities that could explain the persistent hypercortisolism include a reduced cortisol clearance or, alternatively, an up-regulation of the peripheral cortisol regeneration by 11β-hydroxysteroid dehydrogenase (11β-HSD1). Indeed, 11β-HSD1 is hormonally regulated: insulin, growth hormone, and T3, all of which are decreased during prolonged critical illness, exert a suppressive effect on the activity of this enzyme (33). In addition, strict blood glucose control with intensive insulin therapy was shown to lower circulating cortisol levels in prolonged critically ill patients (34), again adding to the possibility of an important role for 11β-HSD1 in cortisol production during this phase of critical illness.
In contrast to the increased serum cortisol levels, circulating levels of adrenal androgens, such as dehydroepiandrosterone (which has immune-stimulatory properties on Th1-helper cells), are low during prolonged critical illness (1, 2). Furthermore, despite increased plasma renin activity, decreased concentrations of aldosterone are seen in protracted critical illness. This constellation suggests a shift of pregnenolone metabolism away from the mineralocorticoid and adrenal androgen pathway and towards the glucocorticoid pathway, orchestrated by an unknown peripheral drive. The fact that this type of relative adrenal insufficiency coincides with adverse outcome suggests that high levels of glucocorticoids remain essential for haemodynamic stability.
Whether the persisting elevation in cortisol is beneficial, remains uncertain. In theory, it could be involved in the increased susceptibility to infectious complications that are associated with prolonged critical illness. Furthermore, other possible disadvantages of prolonged hypercortisolism include impaired wound healing and myopathy, complications that are frequently observed during prolonged critical illness, although this remains to be proven.
Therapeutic interventions: treatment of adrenal failure during critical illness
In some specific cases, glucocorticoid treatment should be started or continued during critical illness, e.g. in patients with previously diagnosed primary or central adrenal insufficiency, or in patients who were previously treated with systemic glucocorticoids. Furthermore, it is obvious that patients suffering a true addisonian crisis need hydrocortisone treatment in severe stress conditions. Conversely, glucocorticoid therapy may aggravate the condition of patients with concomitant diabetes insipidus, since the lack of cortisol in these patients prevents polyuria. Another condition requiring special attention is the post-hypophysectomy phase for Cushing’s syndrome, characterized by a high vulnerability to an addisonian-like crisis. In these patients, drugs such as phenytoin, barbiturates, rifampicin, and thyroid hormone can increase the glucocorticoid replacement dose requirements due to an acceleration of the glucocorticoid metabolism. If this increased requirement is not met, an adrenal crisis may occur.
Initial trials using high doses of glucocorticoids in critically ill patients have shown that this strategy is ineffective and perhaps even harmful (3, 35). In contrast, the concept of relative hypothalamic-pituitary-adrenal insufficiency in patients with sepsis or septic shock, advocates short-term ‘low dose’ glucocorticoid replacement therapy as beneficial in patients with sepsis without a full blown adrenal failure. A recent randomized controlled trial on hydrocortisone therapy in patients with septic shock, however, could not confirm this benefit (36). It was shown recently that glucocorticoid treatment may down-regulate expression of the glucocorticoid receptor, and thus reduce glucocorticoid sensitivity, in the liver but not in muscle (37). Therefore, steroid-induced side effects, such as insulin resistance and catabolism, may be induced in muscle by pharmacologically high levels of cortisol, whereas in liver the reduced glucocorticoid sensitivity may protect against such side effects.
A problematic methodology for diagnosis of relative adrenal failure in acute stress conditions in part explains the controversy on this concept (38). Indeed, accurate ‘normal’ baseline cortisol levels in this type of stress, as well as normal reference values for cortisol responses to a classic ACTH-stimulation test, remain unavailable. An interesting alternative approach recommends a three-level test. The first level comprises the clinical suspicion of adrenal insufficiency; secondly, basal cortisol testing is performed; and finally, if the presence of corticosteroid insufficiency is doubtful from the measured cortisol levels, ACTH testing is performed, with the endpoint being peak cortisol responses (Fig. 10.2.3.5). In view of the ACTH-stimulation test, the dilemma of using a low (1 μg) or high (250 μg) dose ACTH test was recently reviewed by Steward and colleagues, who concluded in favour of the high-dose test (39). This conclusion was based on current evidence showing the inherent difficulty of reproducibility and the additional costs involved with the low dose test. Also, far more follow-up data are available for patients with ‘borderline’ cortisol values obtained by the high dose than by the low dose test.
Another controversial issue regarding the concept of relative adrenal failure in acute sepsis is the dose and duration of treatment once it has been initiated. Indeed, high dose glucocorticoid administration for a long period of time in patients with sepsis will conceivably worsen the loss of lean tissue, increase the risk of polyneuropathy and myopathy, extend intensive care unit dependence, and increase susceptibility to potentially lethal complications.
Implications for clinical practice
The anterior pituitary responds biphasically to the severe stress of illness and trauma (Fig. 10.2.3.6). In the acute phase it is actively secreting, but target organs become resistant and concentrations of most peripheral effector hormones are low. In contrast, prolonged critical illness is characterized by a uniform suppression, predominantly of hypothalamic origin, of the neuroendocrine axes. These alterations contribute to low serum levels of the respective target organ hormones.
Although the differentiation between beneficial and harmful neuroendocrine responses to critical illness is difficult, it is important before considering any therapeutic intervention. The hypercatabolic reaction during acute critical illness is probably beneficial and, as such, provides no evidence that supports intervention. In prolonged critical illness, however, sustained hypercatabolism may compromise vital functions, cause weakness, and delay or hamper recovery. Theoretically, during this phase, a strategy of therapeutic intervention to correct these abnormalities could improve survival. Although it has been shown that coinfusion of GHRP2, TRH, and GnRH at least partially restores the three pituitary axes and reinitiates anabolism (1), the effect on survival remains unknown. Hence, because of the lack of appropriately designed and powered clinical trials, these and other interventions in the critically ill should at this time still be considered experimental. It underlines, however, the interaction that exists among the different endocrine axes and the importance of jointly correcting all hypothalamic-pituitary defects rather than applying a single hormone treatment.
In view of the adverse outcome of single-hormone treatment strategies, high doses of either growth hormone or glucocorticoids appeared to aggravate insulin resistance and hyperglycaemia that usually develop during critical illness (4, 35). However, the toxic side effects of glucose counter-regulation might have surpassed any possible benefits of these therapies. Although it had long been widely accepted that stress-induced hyperglycaemia is beneficial to organs that largely rely on glucose for energy supply but do not require insulin for glucose uptake, strict blood glucose control with intensive insulin therapy has shown to be beneficial, when maintained for at least a few days and avoiding excess hypoglycaemia (40–42). The optimal target level for blood glucose in the critically ill, however, remains a debated topic.
1. Langouche L, Van den Berghe G. The dynamic neuroendocrine response to critical illness. Endocrinol Metab Clin North Am, 2006; 35: 777–91.Find this resource:
2. Van den Berghe G, de Zegher F, Bouillon R. Clinical review 95: Acute and prolonged critical illness as different neuroendocrine paradigms. J Clin Endocrinol Metab, 1998; 83: 1827–34.Find this resource:
3. Roberts I, Yates D, Sandercock P, Farrell B, Wasserberg J, Lomas G, et al. Effect of intravenous corticosteroids on death within 14 days in 10008 adults with clinically significant head injury (MRC CRASH trial): randomised placebo–controlled trial. Lancet, 2004; 364: 1321–8.Find this resource:
4. Takala J, Ruokonen E, Webster NR, Nielsen MS, Zandstra DF, Vundelinckx G, et al. Increased mortality associated with growth hormone treatment in critically ill adults. N Engl J Med, 1999; 341: 785–92.Find this resource:
5. Bowers CY, Momany FA, Reynolds GA, Hong A. On the in vitro and in vivo activity of a new synthetic hexapeptide that acts on the pituitary to specifically release growth hormone. Endocrinology, 1984; 114: 1537–45.Find this resource:
6. Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, Kangawa K. Ghrelin is a growth–hormone-releasing acylated peptide from stomach. Nature, 1999; 402: 656–60.Find this resource:
7. Ross R, Miell J, Freeman E, Jones J, Matthews D, Preece M, et al. Critically ill patients have high basal growth hormone levels with attenuated oscillatory activity associated with low levels of insulin-like growth factor-I. Clin Endocrinol (Oxf), 1991; 35: 47–54.Find this resource:
8. Kupfer SR, Underwood LE, Baxter RC, Clemmons DR. Enhancement of the anabolic effects of growth hormone and insulin-like growth factor I by use of both agents simultaneously. J Clin Invest, 1993; 91: 391–6.Find this resource:
9. Yen PM. Physiological and molecular basis of thyroid hormone action. Physiol Rev, 2001; 81: 1097–142.Find this resource:
10. Rothwell PM, Lawler PG. Prediction of outcome in intensive care patients using endocrine parameters. Crit Care Med, 1995; 23: 78–83.Find this resource:
11. Van den Berghe G. Novel insights into the neuroendocrinology of critical illness. Eur J Endocrinol, 2000; 143: 1–13.Find this resource:
12. Van der Poll T, Endert E, Coyle SM, Agosti JM, Lowry SF. Neutralization of TNF does not influence endotoxininduced changes in thyroid hormone metabolism in humans. Am J Physiol, 1999; 276: R357–62.Find this resource:
13. Klemperer JD, Klein I, Gomez M, Helm RE, Ojamaa K, Thomas SJ, et al. Thyroid hormone treatment after coronary-artery bypass surgery. N Engl J Med, 1995; 333: 1522–7.Find this resource:
14. Fliers E, Guldenaar SE, Wiersinga WM, Swaab DF. Decreased hypothalamic thyrotropin-releasing hormone gene expression in patients with nonthyroidal illness. J Clin Endocrinol Metab, 1997; 82: 4032–6.Find this resource:
16. Mebis L, Langouche L, Visser TJ, Van den Berghe G. The Type II Iodothyronine Deiodinase Is Up-Regulated in Skeletal Muscle during Prolonged Critical Illness. J Clin Endocrinol Metab, 2007; 92: 3330–3.Find this resource:
17. Debaveye Y, Ellger B, Mebis L, Darras VM, Van den Berghe G. Regulation of tissue iodothyronine deiodinase activity in a model of prolonged critical illness. Thyroid, 2008; 18: 551–60.Find this resource:
18. Thijssen-Timmer DC, Peeters RP, Wouters P, Weekers F, Visser TJ, Fliers E, et al. Thyroid hormone receptor isoform expression in livers of critically ill patients. Thyroid, 2007; 17: 105–12.Find this resource:
19. Brent GA, Hershman JM. Thyroxine therapy in patients with severe nonthyroidal illnesses and low serum thyroxine concentration. J Clin Endocrinol Metab, 1986; 63: 1–8.Find this resource:
20. Bettendorf M, Schmidt KG, Grulich-Henn J, Ulmer HE, Heinrich UE. Tri-iodothyronine treatment in children after cardiac surgery: a double-blind, randomised, placebo-controlled study. Lancet, 2000; 356: 529–34.Find this resource:
21. Ringel MD. Management of hypothyroidism and hyperthyroidism in the intensive care unit. Crit Care Clin, 2001; 17: 59–74.Find this resource:
22. Kaptein EM, Quion-Verde H, Swinney RS, Egodage PM, Massry SG. Acute hemodynamic effects of levothyroxine loading in critically ill hypothyroid patients. Arch Intern Med, 1986; 146: 662–6.Find this resource:
23. Escobar-Morreale HF, Obregon MJ, Escobar del RF, Morreale de EG. Replacement therapy for hypothyroidism with thyroxine alone does not ensure euthyroidism in all tissues, as studied in thyroidectomized rats. J Clin Invest, 1995; 96: 2828–38.Find this resource:
24. Samson WK, Taylor MM, Baker JR. Prolactin-releasing peptides. Regul Pept, 2003; 114: 1–5.Find this resource:
25. Spratt DI. Altered gonadal steroidogenesis in critical illness: is treatment with anabolic steroids indicated. Best Pract Res Clin Endocrinol Metab, 2001; 15: 479–94.Find this resource:
26. Spratt DI, Morton JR, Kramer RS, Mayo SW, Longcope C, Vary CP. Increases in serum estrogen levels during major illness are caused by increased peripheral aromatization. Am J Physiol Endocrinol Metab, 2006; 291: E631–38.Find this resource:
27. Van den Berghe G, de Zegher F, Lauwers P, Veldhuis JD. Luteinizing hormone secretion and hypoandrogenaemia in critically ill men: effect of dopamine. Clin Endocrinol (Oxf), 1994; 41: 563–9.Find this resource:
28. Angele MK, Ayala A, Cioffi WG, Bland KI, Chaudry IH. Testosterone: the culprit for producing splenocyte immune depression after trauma hemorrhage. Am J Physiol, 1998; 274: C1530–6.Find this resource:
29. Ferrando AA, Sheffield-Moore M, Wolf SE, Herndon DN, Wolfe RR. Testosterone administration in severe burns ameliorates muscle catabolism. Crit Care Med, 2001; 29: 1936–42.Find this resource:
30. Vermes I, Beishuizen A. The hypothalamic-pituitary-adrenal response to critical illness. Best Pract Res Clin Endocrinol Metab, 2001; 15: 495–511.Find this resource:
31. Cooper MS, Stewart PM. Corticosteroid insufficiency in acutely ill patients. N Engl J Med, 2003; 348: 727–34.Find this resource:
32. Vermes I, Beishuizen A, Hampsink RM, Haanen C. Dissociation of plasma adrenocorticotropin and cortisol levels in critically ill patients: possible role of endothelin and atrial natriuretic hormone. J Clin Endocrinol Metab, 1995; 80: 1238–42.Find this resource:
33. Tomlinson JW, Walker EA, Bujalska IJ, Draper N, Lavery GG, Cooper MS, et al. 11beta-hydroxysteroid dehydrogenase type 1: a tissue-specific regulator of glucocorticoid response. Endocr Rev, 2004; 25: 831–66.Find this resource:
34. Vanhorebeek I, Peeters RP, Vander Perre S, Jans I, Wouters PJ, Skogstrand K, et al. Cortisol response to critical illness: effect of intensive insulin therapy. J Clin Endocrinol Metab, 2006; 91: 3803–13.Find this resource:
35. Minneci PC, Deans KJ, Banks SM, Eichacker PQ, Natanson C. Meta-analysis: the effect of steroids on survival and shock during sepsis depends on the dose. Ann Intern Med, 2004; 141: 47–56.Find this resource:
36. Sprung CL, Annane D, Keh D, Moreno R, Singer M, Freivogel K, et al. Hydrocortisone therapy for patients with septic shock. N Engl J Med, 2008; 358: 111–24.Find this resource:
37. Peeters RP, Hagendorf A, Vanhorebeek I, Visser TJ, Klootwijk W, Mesotten D, et al Tissue mRNA expression of the glucocorticoid receptor and its splice variants in fatal critical illness. Clin Endocrinol (Oxf), 2009; 71: 145–53Find this resource:
38. Mesotten D, Vanhorebeek I, Van den Berghe G. The altered adrenal axis and treatment with glucocorticoids during critical illness. Nat Clin Pract Endocrinol Metab, 2008; 4: 496–505.Find this resource:
39. Stewart PM, Clark PM. The low-dose corticotropin-stimulation test revisited: the less, the better. Nat Clin Pract Endocrinol Metab, 2009; 5: 68–9.Find this resource:
40. Van den Berghe G, Wouters P, Weekers F, Verwaest C, Bruyninckx F, Schetz M, et al. Intensive insulin therapy in critically ill patients. N Engl J Med, 2001; 345: 1359–67.Find this resource:
41. Van den Berghe G, Wilmer A, Hermans G, Meersseman W, Wouters PJ, Milants I, et al. Intensive insulin therapy in the medical ICU. N Engl J Med, 2006; 354: 449–61.Find this resource:
42. Vlasselaers D, Milants I, Desmet L, Wouters PJ, Vanhorebeek I, van den Heuvel I, et al. Intensive insulin therapy for patients in paediatric intensive care: a prospective, randomised controlled study. Lancet, 2009; 373: 547–56.Find this resource: