| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Hanson Institute (J.T.H., H.A.-M., C.J.B., D.J.T.), The University of Adelaide (J.T.H., H.A.-M., C.J.B., D.J.T.), and the Endocrine and Metabolic (J.T.H., H.A.-M., T.Q., D.J.T.) and Intensive Care Unit (M.J.C., P.D.T.), Royal Adelaide Hospital, North Terrace, Adelaide, SA 5000, Australia; and Canterbury Health Laboratories (J.G.L.), Christchurch, New Zealand
Address all correspondence and requests for reprints to: Dr. Jui Ho, Endocrine and Metabolic Unit, Royal Adelaide Hospital, North Terrace, Adelaide 5000, South Australia, Australia. E-mail: Jui.Ho{at}imvs.sa.gov.au.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Hypothesis: Free plasma cortisol may correspond more closely to illness severity than total cortisol, comparing SS and sepsis (S).
Design: This was a prospective study.
Setting: This study took place in a tertiary teaching hospital.
Patients: Patients had SS (n = 45) or S (n = 19) or were healthy controls (HCs; n = 10).
Aim: The aim of the study was to compare total with free cortisol, measured directly and estimated by Coolens’ method, corticosteroid-binding globulin (CBG), and albumin in patients with SS (with and without RAI) and S during acute illness, recovery, and convalescence.
Results: Comparing SS, S, and HC subjects, free cortisol levels reflected illness severity more closely than total cortisol (basal free cortisol, SS, 186 vs. S, 29 vs. HC, 13 nmol/liter, P < 0.001 compared with basal total cortisol, SS, 880 vs. S, 417 vs. HC, 352 nmol/liter, P < 0.001). Stimulated free cortisol increments varied greatly with illness category (SS, 192 vs. S, 115 vs. HC, 59 nmol/liter, P = 0.004), whereas total cortisol increments did not (SS, 474 vs. S, 576 vs. HC, 524 nmol/liter, P = 0.013). The lack of increase in total cortisol with illness severity is due to lower CBG and albumin. One third of patients with SS (15 of 45) but no S patients met a recently described criterion for RAI (total cortisol increment after tetracosactrin 
248 nmol/liter). RAI patients had higher basal total cortisol (1157 vs. 756 nmol/liter; P = 0.028) and basal free cortisol (287 vs. 140 nmol/liter; P = 0.017) than non-RAI patients. Mean cortisol increments in RAI were lower (total, 99 vs. 648 nmol/liter, P < 0.001; free, 59 vs. 252 nmol/liter, P < 0.001). These differences were not due to altered CBG or albumin levels. Free cortisol levels normalized more promptly than total cortisol in convalescence. Calculated free cortisol by Coolens’ method compared closely with measured free cortisol.
Conclusions: Free cortisol is likely to be a better guide to cortisolemia in systemic infection because it corresponds more closely to illness severity. The attenuated cortisol increment after tetracosactrin in RAI is not due to low cortisol-binding proteins. Free cortisol levels can be determined reliably using total cortisol and CBG levels.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Relative adrenal insufficiency (RAI) is conceptually defined by some investigators as an inadequate cortisol response to severe illness associated with rapid clinical and hemodynamic improvement after stress-dose glucocorticoid therapy (6, 7, 8, 9, 10). High plasma cortisol levels and blunted cortisol responses to corticotropin stimulation are linked to higher mortality in SS (9, 10, 11, 12, 13). A multicenter, randomized, prospective study in SS patients showed that the use of stress-dose corticosteroids significantly improved 28-d survival and facilitated withdrawal of vasopressor support in patients with RAI (defined here by a cortisol increment of 248 nmol/liter or less, i.e. 
248) but not in patients without RAI (14).
Despite these data, there is currently no consensus on diagnostic criteria or indications for treatment for RAI. Furthermore, the total cortisol increment after ACTH may not be the best guide to RAI because of reduced corticosteroid-binding globulin (CBG) and albumin levels. CBG and albumin transport over 90% of the circulating cortisol under normal circadian conditions (15, 16). In critical illness, mean CBG and albumin levels fall by around 50% but with marked interindividual variation. Hence, total cortisol may not reflect the biologically active free (unbound) cortisol. Indeed, a recent study found that baseline and ACTH-stimulated cortisol levels in a heterogeneous group of critically ill patients (n = 66) were lower in the subgroup with hypoalbuminemia (albumin < 2.5 g/dl, n = 36). However, free cortisol levels were similar in the hypoalbuminemic and normoalbuminemic groups (17). Our study was directed toward comparing basal and stimulated free and total cortisol levels, specifically in S and SS.
Although hydrocortisone treatment appears beneficial in a study of severe pneumonia-induced S, no cortisol data were available (18). Hence, we examined the rates of RAI, by one empiric definition, in our subjects with S.
The Coolens’ method may be practically useful in critical illness because it estimates free cortisol levels from total cortisol and CBG levels (15). If reliable in SS, use of Coolens’ method would obviate the need for complex nonautomated methods of free cortisol measurement such as ultracentrifugation or equilibrium dialysis. However, the method has not been validated in SS, and such validation was one objective of this study.
Our principal hypothesis was that the free cortisol would be higher than total cortisol in SS than S. Hence, free cortisol may be a better measure of circulating glucocorticoid activity. Our aims were: 1) to compare total and free plasma cortisol levels under basal and tetracosactrin-stimulated conditions, in patients with SS (with and without currently defined RAI), S, and in healthy controls (HCs); 2) to determine the rates of recovery of RAI in survivors; and 3) to evaluate the use of the Coolens’ method, a calculated free cortisol based on total cortisol and CBG concentrations, in estimating free cortisol in SS.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Patients and volunteers
Definitions of SS and S followed the American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference Committee criteria (19). SS was defined as the presence of systemic inflammatory response syndrome, documented infection or positive blood culture and organ dysfunction, hypoperfusion abnormality or S-induced hypotension refractory to adequate fluid resuscitation, or the use of inotropic or vasopressor support. Systemic inflammatory response syndrome is manifested by two or more of the following: fever (temperature > 38 C) or hypothermia (temperature < 35.5 C), tachycardia (>90 beats/min), tachypnea (>20 breaths/min), leucocytosis or leucopenia (white blood cell count > 12,000 or <4000 cells/µl), or immature neutrophils (bands > 10%). The criteria for organ dysfunction or hypoperfusion abnormality included one of the following: altered mental status, hypoxemia (PaO2 < 75 mm Hg), oliguria (urine output < 30ml/h), or elevated plasma lactate level (>2 mmol/liter). The definition of hypotension was: systolic blood pressure < 90 mm Hg or a reduction in systolic blood pressure of 40 mm Hg or more from baseline.
SS patients who met the entry criteria were recruited from ICU. S patients were recruited from general wards after referral from their physician. Tetracosactrin tests were performed while patients still met the diagnostic criteria for S, within 24 h of diagnosis. HCs were recruited by advertisement for the tetracosactrin stimulation test.
Exclusion criteria included: age under 18 yr, pregnancy, irreversible underlying disease such as advanced malignancy or AIDS, or conditions known to disrupt the hypothalamic-pituitary-adrenal axis such as the use of glucocorticoids and preexisting hypocortisolism. Patients with cirrhosis, pulmonary embolism, and acute myocardial infarction were excluded.
Tetracosactrin (ACTH 1–24) stimulation tests
Blood samples were collected immediately before iv tetracosactrin (250 µg ACTH 1–24, Synacthen, Novartis, North Ryde, NSW, Australia) and at 30 and 60 min postinjection. Surviving patients were offered a second ACTH stimulation test performed within 48 h of hospital discharge and a third ACTH stimulation test 6–12 wk after the second test. Follow-up ACTH stimulation tests were not performed in the control subjects.
Clinical evaluation
Information relating to: 1) age and gender, 2) infection site(s) and organisms, 3) severity of illness at baseline using the Acute Physiology and Chronic Health Evaluation (APACHE) II scoring system, 4) disease outcome including mortality, 5) use of hydrocortisone, and 6) time to discontinuation of vasopressor support was collected.
Laboratory measurements
Free cortisol was measured by an ultrafiltration/ligand binding method. Plasma (0.5 ml) was equilibrated with 3H cortisol (0.1 µCi, TRK 407, GE Healthcare, Little Chalfont, Buckinghamshire, UK) for 30 min at 37 C followed by centrifugation in a preconditioned ultrafiltration device (UFC3LGC00, Millipore Corp., Billerica, MA) at 10,000 x g for 10 min at 37 C. Comparison of the ultrafiltrate radioactive counts (50 µl) with the equilibrated noncentrifuged plasma radioactivity (50 µl) determined the percent free cortisol. Radioactivity was counted in 2 ml of xylene-based scintillant (PCS II, GE Healthcare). The recovery of radioactivity was 100%. There was no adsorption of 3H cortisol to either the filter or the polypropylene used in the filter device and carrier tube.
Validation of the ligand binding assay against equilibrium dialysis method was performed using samples from 19 SS patients and 10 samples from normal individuals. For each sample, 1 ml plasma was dialyzed against 250 µl phosphate buffer solution using an Eppendorf tube device with a presoaked dialysis membrane (6–8000 molecular weight cutoff) for 18 h at 37 C. The total and postdialysis samples were analyzed by plasma direct ELISA.
Calculated free cortisol was derived using the Coolens’ equation: U2 x K (1 + N) + U[1 + N + K(G – T)] – T = 0, where T is cortisol, G is CBG, U is unbound cortisol, and K is the affinity of CBG for cortisol at 37 C (15). N is the ratio of albumin bound to free cortisol, and 1.74 is the value conventionally used. The value of N would be expected to change with altered concentrations of plasma albumin, as observed in SS. We addressed this by investigating the distribution of cortisol (600 nmol/liter) in varying concentrations of purified human serum albumin solutions (Sigma A-9511, Sigma-Aldrich, St. Louis, MO) using equilibrium dialysis. We used these experimentally derived values of N to further calculate free cortisol, thus compensating for variations in plasma albumin.
Plasma total cortisol was measured by enzyme-linked fluorescent assay (AxSYM Cortisol assay kit, Abbott Laboratories, Abbott Park, IL). The intra- and interassay coefficients of variation were 5 and 6.5%, respectively. Plasma CBG was measured by a monoclonal ELISA as previously described (20). Within- and between-assay variations were 5–12%, and the assay has a lower limit of detection of 50 nmol/liter. Plasma albumin was measured using the dye binding method with Bromocresol Purple (Integrated Sciences, Willoughby, NSW, Australia) and was quantitatively determined on an Olympus AU5400 analyzer.
Statistical analysis
The plasma cortisol increment was defined as the difference between the basal cortisol and the highest concentration after tetracosactrin. Descriptive statistics were reported as mean ± SE. The data from the three groups (SS, S, and control) were analyzed by Kruskal-Wallis one-way ANOVA on ranks to investigate group effects, time effects, and group by time interaction. RAI was empirically defined using the currently most validated criterion, that of Annane et al. (2). The RAI ( 248) and non-RAI ( > 248) groups were compared using nonparametric (Mann-Whitney rank sum test) analysis. Statistical significance was set at P < 0.05. Statistical analyses were performed using the Statistica software package (Statsoft, Tulsa, OK).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
. Control subjects were younger than patients, and more males than females were tested in the S group. Follow-up tetracosactrin stimulation tests were performed in 20 SS and 14 S patients.
|
Free cortisol levels corresponded more closely to illness severity than total cortisol (Fig. 1). Basal free cortisol levels were: SS, 186 ± 24 vs. S, 29 ± 5 vs. HC, 13 ± 2 nmol/liter; i.e. a 6.4-fold difference. SS vs. S (P < 0.001) and basal total cortisol levels were: SS, 880 ± 79 vs. S, 417 ± 45 vs. HC, 352 ± 34 nmol/liter, a 2.1-fold difference, SS vs. S (P < 0.001). A similar pattern was evident with cortisol increments after tetracosactrin; free cortisol levels were: SS, 192 ± 36 vs. S, 115 ± 12 vs. HC, 59 ± 7 nmol/liter (P = 0.004); and total cortisol increments were: SS, 474 ± 79 vs. S, 576 ± 36 vs. HC, 524 ± 54 nmol/liter (P = 0.013). Higher free cortisol with illness severity may be due to lower levels of cortisol binding proteins (Fig. 1). S patients had 35% lower CBG and 29% lower albumin levels than controls, and SS was associated with a further 19% fall in CBG and a 29% fall in albumin levels compared with the S group (Fig. 1). CBG levels were: SS, 335 ± 17 vs. S, 412 ± 40 vs. HC, 633 ± 43 nmol/liter (P < 0.001); and albumin levels were: SS, 19 ± 1; S, 27 ± 1; and HC, 38 ± 1 g/liter (P < 0.001).
|
One third (15 of 45, 33%) of SS patients met the current empiric definition of RAI ( 248). None of the S patients had RAI. Nine of the 15 patients with less than or equal to 248 (60%) and 50% (15 of 30) of patients with non-RAI ( > 248) received stress-dose hydrocortisone therapy. There was no difference in the duration of vasopressor support between the two groups (RAI, 8.3 ± 2.4 vs. non-RAI, 4.4 ± 1.0 d, P > 0.05). There was also no difference in the duration of hydrocortisone therapy between the two groups (RAI, 7.0 ± 2.2 vs. non-RAI, 3.0 ± 1.0 d, P > 0.05). Among patients with less than or equal to 248, 33% (three of nine) of hydrocortisone-treated compared with 50% (three of six) of nonhydrocortisone-treated patients died by d 28. In contrast, 50% (15 of 30) of patients with greater than 248 who received hydrocortisone and 20% (three of 15) of nonhydrocortisone-treated patients died by d 28. Overall, there was no significant difference in the 28-d mortality of SS patients with and without RAI (40 vs. 35%, respectively; P > 0.05).
Total and free cortisol, CBG, and albumin levels in SS patients are shown in Fig. 2. Patients with less than or equal to 248 had higher basal total cortisol (1154 ± 145 vs. 756 ± 87 nmol/liter, P = 0.028) and basal free cortisol levels (287 ± 51 vs. 140 ± 22 nmol/liter, P = 0.017). As expected by definition, total cortisol increments were less in the RAI group than the non-RAI group (99 ± 20 vs. 648 ± 101, P < 0.001). Free cortisol increments were also significantly less (59 ± 11 vs. 252 ± 48, P < 0.001). Lower cortisol increments in RAI were not related to CBG or albumin concentrations because these were similar: CBG, RAI, 350 ± 38 vs. non-RAI, 328 ± 18, P > 0.05; albumin, RAI, 19 ± 2 vs. non-RAI, 19 ± 1, P > 0.05.
|
less than or equal to 248 had a free cortisol increment after tetracosactrin of less than 110 nmol/liter, and 27 of 31 (87%) of patients with greater than 248 exceeded this threshold. Hence, a free cortisol increment of 110 nmol/liter categorizes RAI and non-RAI patients similarly to a total cortisol response of 248 nmol/liter. Results of repeated tetracosactrin tests after recovery from SS and S
By the time of hospital discharge, follow-up tests showed free cortisol levels had fallen to baseline, but total cortisol remained elevated (Fig. 3). Among survivors, mean basal total cortisol levels remained high at the second test (first test, SS, 880 ± 79 vs. second test, 464 ± 45 nmol/liter, P < 0.05), but mean basal free cortisol had significantly declined (first test, 186 ± 24 vs. second test, 38 ± 8 nmol/liter, P < 0.001). Basal total cortisol did, however, fall by the third tetracosactrin stimulation test. In SS patients, total cortisol increments after tetracosactrin did not change significantly on retesting (first test, 477 ± 79 vs. second test, 441 ± 36 nmol/liter, P = not significant), but there was a fall in free cortisol by the second test (first test, 192 ± 36 vs. second test, 84 ± 14 nmol/liter, P = 0.009). Persistently elevated total cortisol in the SS group, despite a fall in free cortisol, reflects the effect of rising CBG (first test, 335 ± 17 vs. second test, 502 ± 37, P < 0.001; second test vs. third test, 569 ± 24 nmol/liter, P > 0.05) and rising albumin (first test, 19 ± 1 vs. second test, 24 ± 2, P = 0.004; second vs. third test, 35 ± 2 nmol/liter, P < 0.001). Of the six patient (with 248) retested with tetracosactrin, all had cortisol greater than 248 nmol/liter at the time of the second test. S patients exhibited no significant change in basal or stimulated total and free cortisol on retesting. Among S patients, mean CBG levels rose slightly but nonsignificantly on retesting (first test, 412 ± 40 vs. second test, 526 ± 45 vs. third test, 576 ± 71 nmol/liter, P > 0.05). Albumin levels rose significantly (first test, 19 ± 1 vs. second test, 30 ± 2 vs. third test, 36 ± 0.4 nmol/liter, P = 0.05).
|
|
Calculated free cortisol (Coolens’ method) and measured free cortisol (ultracentrifugation) were in close agreement (r = 0.9, P < 0.001, Fig. 4A). We also compared free cortisol by ultrafiltration with free cortisol calculated to take account of albumin concentration (Fig. 4B) using the experimentally derived values of N (Table 2). Furthermore, we compared the correlation between calculated and measured free cortisol in patients with hypoalbuminemia (serum albumin 25 g/liter) with those with normal albumin concentrations. As shown in Fig. 4C, the correlation coefficient (r) of 0.911, P < 0.001, is obtained for calculated vs. measured free cortisol in patients with hypoalbuminemia compared with r = 0.936 for the normoproteinemic group. Overall, adjustment of the Coolens’ equation constant N to take account of varying albumin concentrations had little effect on estimated free cortisol.
|
|
The ultrafiltration method of measuring free cortisol was compared with the gold standard of equilibrium dialysis using 19 samples from SS patients and 10 samples from normal individuals. The methods yield near identical cortisol values, and the correlation coefficient is r = 0.99 (Fig. 5).
|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
248) and non-RAI ( > 248) patients when basal free cortisol levels were used rather than basal total cortisol levels. In addition, we found that Coolens’ method predicts free cortisol, as measured by ultracentrifugation in a method validated by equilibrium dialysis, in SS. RAI is defined by some investigators based on mortality and treatment responses to hydrocortisone. Our study cannot support or refute the concept of RAI but suggests that free cortisol may be a better guide to cortisolemia.
Cortisol circulates in plasma in three fractions; 80–90% is bound to the 52- to 56-kDa glycoprotein, CBG, 5–10% is loosely bound to albumin, and up to 5% circulates as free or unbound cortisol, under unstressed conditions (21). High-affinity cortisol-CBG binding [association constant (Ka), 3 x 107 liter/mol] is saturated at 500–600 nmol/liter of cortisol because each molecule of CBG binds only a single molecule of cortisol (22, 23). In our group of critically ill patients, mean CBG levels were reduced by 48% in SS and by 35% in S. Correspondingly, the basal free cortisol fraction in plasma was 23% in SS, 7% in S, and 3.6% in controls. Despite increased free cortisol in SS, total cortisol levels may remain stable due to reductions in bound cortisol; our data show this because there was little difference in basal and stimulated total cortisol in S and SS but marked differences in free cortisol levels.
Free cortisol measurement, using either ultracentrifugation or equilibrium dialysis to separate free from bound cortisol, is not offered in routine clinical laboratories. Free cortisol measurement is time consuming and nonautomated. There are few clinical studies comparing free cortisol with total cortisol. However, we found that the Coolens’ calculation method, which relies on a number of assumptions, along with routine immunoassay of both total cortisol and CBG, applies in SS. The validity of this equation has been questioned particularly in critical illness where marked reduction of cortisol-binding proteins is observed (24). However, our data reveal that adjustment of the Coolens’ equation constant N to take account of varying albumin concentrations had little effect on estimated free cortisol (Fig. 4
). These data accord with in vitro data using albumin solutions and 3H cortisol distribution analyzed by ultrafiltration (25). Hence, the Coolens’ method, without adjustment of the N value, could be used in SS and may be more broadly applicable if free cortisol was shown to be superior to total cortisol in assisting the decision to use hydrocortisone therapy. We recognize that a constant value (K) used for the affinity of CBG for cortisol could constitute a limitation if CBG binding affinity changes in SS or in hereditary CBG variants with reduced affinity. The calculated free cortisol would be markedly underestimated if these abnormal forms of CBG can be detected immunologically. Because the frequency of these inherited CBG abnormalities is very low and the Coolens’ equation was reliable, these factors do not affect the clinical usefulness of the proposed method.
A recent report involving 66 critically ill patients, of whom 18 had S but none were reported to have SS, suggested that low cortisol binding proteins, rather than low free cortisol levels, may underlie reports of reduced total cortisol in critical illness (17). Our study supports this contention and suggests that free cortisol should be evaluated in treatment trials in SS. We also examined the possibility that diminished cortisol increments after ACTH may relate to low cortisol binding proteins. However, total cortisol increments less than or equal to 248 nmol/liter were not associated with lower cortisol-binding proteins in our study.
It has been proposed that low cortisol levels in critical illness may be due to adrenal hemorrhage or necrosis, pituitary ACTH deficits, and/or genetic polymorphisms that limit maximal adrenocortical hormone secretion (6). Our follow-up data reveal prompt recovery from RAI in SS survivors, although this observation involved only six patients. Moreover, follow-up testing of SS patients showed an overall return of free cortisol levels to normal values postillness, which is in keeping with the concept of a lack of functional adrenal reserve rather than adrenal damage during critical illness.
A design limitation of our study is that we cannot analyze free or total cortisol levels with respect to patient outcomes because therapy was not randomized, and our numbers are not sufficient to determine treatment effects. For example, we did not show a significant difference in mortality between RAI and non-RAI patients in SS. Although we found that a free cortisol increment of 110 nmol/liter categorized patients similarly to a total cortisol increment after tetracosactrin of 248 nmol/liter, a different free cortisol cutoff may correspond to the effects of hydrocortisone treatment, but to demonstrate such an effect would require a therapeutic trial. There were more males than females in the S group, and the controls were younger. However, due to the relatively minor effects of age and gender on the HPA axis function (26, 27, 28), we do not expect that this would have significantly altered the results.
The concept of RAI is controversial because it is not immediately apparent why reduced adrenal reserve, as evidenced by reduced cortisol response to ACTH despite high basal total cortisol, is predictive of responsiveness to parenteral hydrocortisone in SS. Traditionally, adrenal function would be considered normal with ACTH stimulation testing if basal cortisol levels are elevated irrespective of the magnitude of the increment in response to ACTH. On the other hand, the identification of patients who cannot increase cortisol secretion after exogenous ACTH may be revealing a subgroup that requires more cortisol despite high basal levels, perhaps due to acquired tissue glucocorticoid resistance. It is relevant here that one stress-dose hydrocortisone regimen (10 mg/h) produces total cortisol levels of approximately 3100 nmol/liter, well above the levels seen in our SS patients (29). Nevertheless, the concept has empirical support in prognostic (9, 10) and hydrocortisone (14) therapy studies. Although our data do not support or refute the concept of RAI, it appears relevant that free cortisol increments after tetracosactrin varied markedly with S severity, whereas total cortisol increments, on which the most validated empiric definitions of RAI rely, did not vary with S severity. Hence, we think that free cortisol measures may contribute usefully to studies that attempt to use plasma cortisol to predict responses to hydrocortisone.
In comparison with a large therapeutic trial (n = 299) in SS, the prevalence of RAI (33 vs. 77%) and mortality rate (37 vs. 55%) were lower, despite comparable patient selection criteria (14). Furthermore, in our patients with RAI, mean basal total cortisol levels were higher in RAI than in non-RAI, whereas in the study of Annane et al. (14), basal total cortisol was lower in RAI. The illness severity of the patients in the two studies is difficult to compare as different severity indices were used, although our mean APACHE II scores were similar to other studies of SS (8). It should be recognized that etomidate, an inhibitor of 11-ß hydroxylase, was administered to some patients in the study by Annane et al. (14), and this could have contributed to the observed higher frequency of RAI. In the current study, no patient received etomidate.
In summary, free cortisol more closely reflects sepsis severity and falls to normal more quickly after recovery than total cortisol. Lower total cortisol increments in RAI, using one current definition, were due to reduced free cortisol increments rather than lower cortisol binding proteins. Our data indicate that the Coolens’ calculation is a reliable predictor of measured free cortisol and would be more practical than direct measurement. In the development of the notion that cortisol increments after tetracosactrin and basal cortisol measures may predict responsiveness to hydrocortisone treatment, our data suggest that free cortisol may be a better index of cortisolemia than the total cortisol measures hitherto advocated.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online November 1, 2005
Abbreviations: APACHE, Acute Physiology and Chronic Health Evaluation; CBG, corticosteroid-binding globulin; HC, healthy control; RAI, relative adrenal insufficiency; S, sepsis; SS, septic shock.
Received February 7, 2005.
Accepted October 20, 2005.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  I. Wallace, S. Cunningham, and J. Lindsay The diagnosis and investigation of adrenal insufficiency in adults Ann Clin Biochem, September 1, 2009; 46(5): 351 - 367. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Keenan, F. Roelfsema, B. J. Carroll, A. Iranmanesh, and J. D. Veldhuis Sex defines the age dependence of endogenous ACTH-cortisol dose responsiveness Am J Physiol Regulatory Integrative Comp Physiol, August 1, 2009; 297(2): R515 - R523. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. E. Marik Critical Illness-Related Corticosteroid Insufficiency Chest, January 1, 2009; 135(1): 181 - 193. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Bendel, S. Karlsson, V. Pettila, P. Loisa, M. Varpula, E. Ruokonen, and For the Finnsepsis Study Group Free Cortisol in Sepsis and Septic Shock Anesth. Analg., June 1, 2008; 106(6): 1813 - 1819. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Cooper and P. M. Stewart Adrenal Insufficiency in Critical Illness J Intensive Care Med, November 1, 2007; 22(6): 348 - 362. [Abstract] [PDF]
|
|
|
|
 |
|
 K. Klouche, E. F. Da Mota, R. Durant, L. Amigues, P. Corne, O. Jonquet, and J. J. Beraud Hypothalamic pituitary adrenal axis reactivity and dehydroepiandrosterone sulfate plasma concentrations in the critically ill elderly Age Ageing, November 1, 2007; 36(6): 686 - 689. [Full Text] [PDF]
|
|
|
|
 |
|
 B. M. Arafah, F. J. Nishiyama, H. Tlaygeh, and R. Hejal Measurement of Salivary Cortisol Concentration in the Assessment of Adrenal Function in Critically Ill Subjects: A Surrogate Marker of the Circulating Free Cortisol J. Clin. Endocrinol. Metab., August 1, 2007; 92(8): 2965 - 2971. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Vanhorebeek, R. P. Peeters, S. Vander Perre, I. Jans, P. J. Wouters, K. Skogstrand, T. K. Hansen, R. Bouillon, and G. Van den Berghe Cortisol Response to Critical Illness: Effect of Intensive Insulin Therapy J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 3803 - 3813. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. M. Arafah Hypothalamic Pituitary Adrenal Function during Critical Illness: Limitations of Current Assessment Methods J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 3725 - 3745. [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 |
denote P < 0.05 for SS vs. S and SS vs. control, respectively. The basal total (A) and free cortisol (B) concentrations were significantly higher in the SS group than in the S and control groups. Free cortisol increments (D) corresponded to S and its severity, whereas total cortisol increments did not (C). CBG and plasma albumin levels (E and F) fell significantly with illness severity.
