| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
bner,
J. Britto,
M. Levin,
N. Klein and
A. Aynsley-Green
The London Centre for Paediatric Endocrinology and Metabolism and Great Ormond Street Hospital for Children NHS Trust (E.J.L.-K., A.A.-G.), London; Centre for Paediatric Epidemiology and Biostatistics (T.J.C.), Institute of Child Health, London WC1N 1EH; Department of Endocrinology (C.C.-H.), St. Bartholemews Hospital, London EC1A 7BE; Department of Paediatrics (J.B., M.L.), Imperial College School of Medicine, London W2 1PG; and Department of Infectious Diseases and Microbiology (N.K.), Institute of Child Health, London WC1N 1EH, United Kingdom
Address all correspondence and requests for reprints to: E. J. Lichtarowicz-Krynska, Department of Biochemistry, Metabolism, and Endocrinology, Institute of Child Health, 30 Guilford Street, London WC1N 1EH, United Kingdom. E-mail: e.lichtarowicz-krynska{at}ich.ucl.ac.uk.
| Abstract |
|---|
|
|
|---|
The aim of the study was to define adrenocortical hormone status at presentation. Sixty children admitted to the pediatric intensive care unit were studied. Children were divided into two groups: group A (n = 31), with meningococcal sepsis, mean age 4.4 yr (range 0.514.4), predicted risk of mortality mean 32.3% (range 0.599.3%); and group B (n = 29), with other diagnoses (post major surgery and with severe respiratory infections), mean age 4.1 yr (range 0.316.3), predicted risk of mortality mean 9.4% (range 0.283%). The groups were not significantly different for age. Plasma levels of aldosterone and cortisol were determined by RIA.
The mean plasma aldosterone concentration on admission in group A was 427.5 ± 88.1 pg/ml, with 96.7% of values within the normal range for age for healthy children and were significantly lower than group B mean, 1489.2 ± 244.2 pg/ml (P < 0.0001), with 59.3% of values above the normal range.
In group A there was no correlation with plasma concentrations of sodium, potassium, or volume of colloid infused in the previous 8 h. In group A mean serum cortisol mean values were 799.5 ± 75.9 nmol/liter and in group B cortisol levels were 703.4 ± 78.6 nmol/liter (P = n.s.). We conclude that children with meningococcal disease present with lower plasma aldosterone concentrations than other patients in the pediatric intensive care unit, for which there is no clear explanation. Further work is needed to elucidate the mechanisms underlying this finding and to examine its clinical implications.
| Introduction |
|---|
|
|
|---|
Patients with meningococcal sepsis have features that would be consistent with adrenal dysfunction. At presentation they frequently have hypotension, poor perfusion, oliguria due to loss of intravascular circulating volume, and decreased cardiac function. Postmortem adrenal gland histopathology reveals a range of abnormalities from mild hemorrhage or thrombi, through to degeneration of adrenocortical cells due to massive bilateral adrenal hemorrhages (2). Adrenal damage is thought to be diffuse and there are no reports of zonal damage in the literature. The frequency of histological abnormalities in survivors is unknown.
The problem of functional adrenal insufficiency in septic shock has recently received much attention. In several studies basal serum cortisol concentrations are substantially elevated. However, dynamic testing of the adrenocortical axis indicates a degree of functional adrenal insufficiency (3). The reported prevalence of this in septic shock of all causes varies depending on the test used (4), the criteria used to define adrenal insufficiency (3, 5, 6), and the age of the population studied. In meningococcal sepsis there is a consensus that response to synacthen stimulation is severely attenuated (7, 8, 9). Evidence for functional adrenal insufficiency in meningococcal infection dates back to 1978 when patients with petechiae showed a poor plasma cortisol rise to im ACTH (7). More recent publications have shown that 52% of children had inadequate plasma cortisol response to short synacthen provocation (8), whereas others report a lower incidence of 14% failing low-dose synacthen (9)
In at least one pediatric study, the degree of functional adrenal insufficiency, as defined by poor response to synacthen stimulation, was associated with increasing vasopressor requirement and duration of shock (8).
Defects in the mineralocorticoid axis have been described in a subset of critically ill adults, in whom hyperreninemic hypoaldosteronism has been associated with a high mortality rate (10). This subset showed little or no aldosterone rise to ACTH, a feature that was reversible in surviving patients on recovery.
Little information exists, however, on how the pediatric mineralocorticoid axis behaves in critical illness. The aim of this study was to define adrenocortical hormone status of children admitted to pediatric intensive care units (PICU), with emphasis on cases of meningococcal sepsis.
| Patients and Methods |
|---|
|
|
|---|
Sixty children (34 boys; 26 girls) admitted to the PICU of Great Ormond Street Hospital for Children, London, and St. Marys Hospital London were recruited to the study. Patients were classified into two groups. Group A consisted of 31 children (15 boys; 16 girls) with meningococcal disease, mean age 4.4 yr (range 0.514.4) with a predicted risk of mortality calculated from the PRISM score, mean value 32.3% (range 0.599.3%). Group B consisted of 29 critically ill children without sepsis and included children post surgery (n = 12) (mainly gastrointestinal and laryngological) and with respiratory (n = 13) and neurological (n = 4) diagnoses. Their mean age was 4.1 yr, (range 0.316.3 yr), with a predicted risk of mortality 9.4% (range 0.283%).
The volume of colloid received by each patient in the preceding 8 h and the level of inotropic support required to maintain normotension were recorded. Patients with meningococcal disease were subclassifed on the basis of inotrope requirements into four groups, as defined in Table 1
.
|
Children who had been ill for more than 24 h or who had preexisting renal, liver, metabolic, and or immunological disease were excluded from the study.
Management protocols
Ethical approval for the study was obtained from the local Research Ethics Committees of Great Ormond Street Hospital for Children NHS Trust and the Institute of Child Health and St. Marys Hospital NHS Trust. Informed written consent was obtained from the parents.
Methods
Blood samples were obtained from indwelling arterial lines on admission to the PICU and at eight hourly intervals thereafter. Plasma was stored at -70 C until the samples were analyzed.
Cortisol were measured by a solid-phase 125I RIA using the Coat-A-Count kit no. TKCO1 (Diagnostics Product Corp., Caernarfon, Wales, UK). The sensitivity of the assay was 13 nmol/liter. The within-assay coefficient of variation was 8.4, 5.3, and 11.1% at serum concentrations of 138, 529 and 910 nmol/liter. The interassay coefficient of variation (CV) was 9.1, 8.9, and 11.5% at serum concentrations of 132, 527, and 982nmol/liter.
Aldosterone was measured by a coated tube RIA kit (no. DSL-8600, Diagnostic Systems Laboratories, Webster, TX).
The sensitivity of the assay was 25 pg/ml. The within-assay CV was 7.0, 3.6, and 8.3% at concentrations of 50.4, 254.6, and 777.9 pg/ml. The interassay CV was 10.4, 7.3, and 10.0% at concentrations of 50.6, 257.6, and 850.9 pg/ml.
Plasma renin activity (PRA) was measured in a subset of 15 patients according to the method described by Menard and Catt (11).The sensitivity of the assay was 1 nmol/h/ml. The within-assay CV was 5.4% at 1.23 nmol/h/ml, 5.6% at 8.24 nmol/h/ml, and 7.5% at 13.8 nmol/h/ml. The interassay CV was 6.0% at 1.23 nmol/h/ml, 6.2% at 8.24 nmol/h/ml, and 8.0% at 13.8 nmol/h/ml.
Statistical analysis
Continuous variables were summarized as the mean ± SEM. Students t test and one-way ANOVA were used to compare group means. Pearson correlations were calculated. Statistical significance was defined as P < 0.05.
The program DATADESK 6.1.1 for Windows (Ithaca, NY) was used for analysis of the data.
| Results |
|---|
|
|
|---|
Figure 1
shows that in 96.7% of meningococcal patients, plasma aldosterone concentrations fell within or below the range for healthy recumbent children (12). In contrast, patients in group B had significantly higher concentrations, with nearly 60% of values above the normal range. In all meningococcal cases, the presenting aldosterone level proved to be the highest value within the time frame measured (Fig. 2A
). This was not always the case for patients in the nonmeningococcal group.
|
|
PRA was measured in 15 patients with meningococcal disease. Values were either normal or high at presentation and, in contrast to aldosterone, did not necessarily fall with time. As shown in Fig. 2B
, levels were highest in younger patients. At presentation, there was no correlation between plasma aldosterone concentration and plasma renin activity.
Dissociation of aldosterone and renin has been previously quantified as an inappropriately low aldosterone/PRA ratio (12). In this study, it appears that 80% (12 of 15) of the aldosterone/plasma renin activity ratios were low (<2) on admission and continued to fall thereafter (after conversion of the units for both parameters to allow comparison with the previous study).
Serum cortisol levels at presentation were elevated in both groups, with the meningococcal group A having higher levels than the nonmeningococcal patients (mean 800.8 ± 75.7 mmol/liter vs. 703.4 ± 78.6 mmol/liter in group B, P < 0.05). High vasopressor requirement (to maintain normotension) on admission to PICU correlated with lower serum cortisol concentrations P < 0.05 (Fig. 3A
). Patients with the highest plasma aldosterone concentrations on admission also had the lowest serum cortisol concentrations (Fig 3B
). This was not related to the age distribution of the patients.
|
All patients recruited to the study survived.
| Discussion |
|---|
|
|
|---|
Patients with meningococcal sepsis are known to have poor perfusion, hypotension, oliguria due to loss of intravascular circulating volume, and depressed cardiac function. The expected response to the above is stimulation of the mineralocorticoid axis and consequently elevation in the plasma aldosterone level. Paradoxically, we found that in 96.7% of meningococcal patients, aldosterone values fell within or below the range for healthy recumbent children (13). In contrast, nonmeningococcal patients had significantly higher levels, with nearly 60% of values above the normal range. This would be consistent with previous observations in which plasma aldosterone levels are known to rise in response to stress (14).
In an attempt to understand this observation, we examined a number of parameters known to influence aldosterone production and release. These include the extent of intravascular depletion or hypoperfusion, hyperkalemia (15), and hyponatremia (16) and starvation (17). None of these correlated with aldosterone levels on admission and therefore did not appear to be responsible for the lower levels of this mineralocorticoid in the meningococcal patients. Whereas altered metabolic clearance rates could also influence aldosterone levels, data are lacking in critical illness or sepsis.
Aldosterone secretion and synthesis is primarily controlled by the renin-angiotensin system (18, 19). In health, both renin activity and plasma aldosterone levels correlate (20). In this study, however, children with meningococcal disease presented with a marked discordance between aldosterone levels, which were low, and elevated renin activity. An inappropriately low plasma aldosterone/PRA ratio, as used by Findling et al. (12)1987 to quantify this phenomenon, was detected in 80% of meningococcal patients on admission (although direct comparison was not possible with the previous study due to differing methodologies).
Conditions in which low levels of aldosterone exist in the context of elevated renin activity have been described. Congenital causes include a proportion of patients with congenital adrenal hyperplasia and those with aldosterone synthase defects (21). Future work may establish whether the activity of aldosterone synthase is decreased in cases of meningococcal sepsis by determination of the 18-corticosterone/aldosterone and corticosterone/aldosterone ratios (22). Acquired conditions include malnutrition (17), although the children in this study were not previously malnourished.
To explain the low plasma aldosterone finding, several potential underlying mechanisms are considered. Children with meningococcal sepsis invariably have generalized endothelial damage (23). Angiotensin-converting enzyme (ACE) is located on the surface of vascular endothelial cells. Destruction of endothelial bound ACE could influence aldosterone synthesis and release. However, in adults ACE levels were found to be similar in both hyperreninemic hypoaldosteronism (HH) patients and non-HH critically ill individuals (10), making this explanation less likely.
Cortisol and aldosterone have high affinities for the mineralocorticoid receptor. A high concentration of cortisol has the potential to saturate the enzyme 11-B dehydrogenase type II (steroid shuttle) at the mineralocorticoid receptor. This obviates the need for aldosterone (24), which is therefore suppressed. The clinical state of patients in whom the steroid shuttle becomes saturated is one of hypertension and hypokalemic alkalosis (25). This was not the clinical picture seen in these patients.
The most likely explanation for the renin-aldosterone dissociation is through disruption of aldosterone synthesis. This may be caused by selective destruction of the zona glomerulosa in the adrenal gland. This could be particularly pertinent to meningococcal sepsis in which damage to adrenal glands is known to occur (2), although this damage is thought to be diffuse and not zone specific. Another potential mechanism of aldosterone synthetic disruption is through cytokines. Patients with meningococcal sepsis have some of the highest proinflammatory cytokine levels recorded in humans. It is known that the immune system is known to exert a degree of paracrine modulation on the adrenal cortex (26), Although there is to date no evidence that aldosterone release in man is controlled by cytokines, work on rat adrenal cells in vitro showed that TNF was capable of blocking aldosterone release (27). This is an area for future study.
Other aldosterone regulatory pathways exist and may also be disrupted in meningococcal disease. In health, ACTH is known to stimulate not only the glucocorticoid but also the mineralocorticoid axis. In a recent study, 51% of critically ill children with meningococcal disease showed a suboptimal aldosterone response (<40% increment) to ACTH stimulation (28), suggesting that at least part of the explanation for low aldosterone levels is inadequate adrenal response to ACTH. Critically ill adult patients with low aldosterone levels also have impaired mineralocorticoid responses to ACTH stimulation, implying a defect or block in the pathway (10, 25). In adults dopamine receptor blockade has been shown to restore ACTH responsiveness (29, 30). Whereas a relationship between inotropic support and aldosterone was observed in our study (Fig. 3B
), it was patients with the highest requirement for inotropes, including dopamine, who had the highest aldosterone levels. This must, however, be interpreted with caution because responses to exogenously administered and endogenously released dopamine may differ.
The concept of HH was introduced by Zipser et al. (10) in 1981. They reported that critically ill adults who exhibited this phenomenon had a particularly poor prognosis, with a 30% increase in mortality in HH patients. Two decades later, therapeutic manipulation of adrenal steroids with stress doses of both hydrocortisone and fludrocortisone resulted in a 30% reduction in mortality (31). Future studies are required to further explore the causes and consequences of low circulating aldosterone levels in critically ill children.
| Conclusions |
|---|
|
|
|---|
Renin levels were high or normal, being appropriate for the clinical scenario of the patients but showing little relation to the low aldosterone levels.
The dissociation between plasma renin and plasma aldosterone has been previously described in adults although the mechanism underlying this finding is not known.
It remains to be determined whether low plasma aldosterone may result from pathway disturbances solely in meningococcal sepsis or is the end product of wider pathophysiological process.
Because HH in critically ill adults is associated with a high mortality rate (10), it is necessary in the future to establish the prevalence of low plasma aldosterone in critically ill septic children and determine its pathophysiological importance.
The clinical and therapeutic implications of these findings still need to be addressed.
Referencing of preliminary abstracts and posters at scientific meetings
Lichtarowicz-Krynska EJ, Bodamer O, Feillet F, Camacho-Hubner C, Britto J, Aynsley-Green A, 1999. Unexpectedly low aldosterone levels in children with meningococcal disease on admission to PICU. In: Care of the critically ill. Vol 15.4, p 145
Lichtarowicz-Krynska EJ, Bodamer O, Feillet F, Camacho-Hubner C, Britto J, Aynsley-Green A, 1999. Circulating aldosterone levels are unexpectedly low in children with meningococcal disease on admission to PICU. In: Hormone research. Vol 51, S2, p 83
Lichtarowicz-Krynska EJ, Bodamer O, Camacho-Hubner C, Britto J, Levin M, Aynsley-Green A, 2000 Circulating aldosterone (Ald) levels are surprisingly low in children with meningococcal disease (MD)the first 48 h in PICU. In: Pediatric research (Suppl). Vol 47 (59A), p 344
| Footnotes |
|---|
Abbreviations: ACE, Angiotensin-converting enzyme; CV, coefficient of variation; HH, hyperreninemic hypoaldosteronism; PICU, pediatric intensive care unit; PRA, plasma renin activity.
Received March 24, 2003.
Accepted December 23, 2003.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
E. A. Oral and M. Pietropaolo Metabolism Lessons for Survival: When Adults and Children Are Not Alike. J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 3763 - 3765. [Full Text] [PDF] |
||||
![]() |
M. den Brinker, K. F. M. Joosten, O. Liem, F. H. de Jong, W. C. J. Hop, J. A. Hazelzet, M. van Dijk, and A. C. S. Hokken-Koelega Adrenal Insufficiency in Meningococcal Sepsis: Bioavailable Cortisol Levels and Impact of Interleukin-6 Levels and Intubation with Etomidate on Adrenal Function and Mortality J. Clin. Endocrinol. Metab., September 1, 2005; 90(9): 5110 - 5117. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |