Service d'aide à la décision clinique
Advertising


Free Article !

Gastroentérologie Clinique et Biologique
Volume 33, n° 6-7
pages 584-588 (juin 2009)
Doi : 10.1016/j.gcb.2009.03.011
Foie et voies biliaires
Assessing adrenal function in cirrhotic patients: Is there a reliable test?
Évaluation de la dysfonction surrénale chez les patients cirrhotiques : a-t-on un test fiable ?
 

T. Thevenot a, 1, ⁎ , S. Borot b, 1, A. Remy-Martin c, 1, R. Sapin d, 1, A. Penfornis b, 1, V. Di Martino a, 1, E. Monnet a, 1
a Service d’hépatologie et de soins intensifs digestifs, hôpital Jean-Minjoz, boulevard Fleming, 25030 Besançon cedex, France 
b Service d’endocrinologie-métabolisme et diabétologie-nutrition, hôpital Jean-Minjoz, 25030 Besançon cedex, France 
c Service de biochimie hormonale, hôpital Jean-Minjoz, 25030 Besançon cedex, France 
d Service d’exploration fonctionnelle par les isotopes, 1, place de l’Hôpital, 67091 Strasbourg cedex, France 

Corresponding author.
OutlineMasquer le plan
Introduction
Is the assessment of adrenal function using serum total cortisol reliable?
Is the assessment of adrenal function using salivary cortisol promising?
Adrenal failure in cirrhosis: what is the evidence?
Conflict of interests
Introduction

Adrenal dysfunction is an uncommon clinical condition that can arise in critical conditions such as severe infections, trauma and after aggressive surgery. Cortisol is essential for regulating the impaired vasomotor tonus of blood vessels, influencing the distribution of bodily fluids and sensitizing catecholamine receptors. However, few studies have assessed adrenal function in the setting of liver disease [1, 2, 3, 4, 5, 6, 7, 8, 9]. Adrenal dysfunction has been reported in acute liver disease and in hemodynamically stable cirrhotic patients, and is found with an impressive frequency of 77% in cirrhotic patients with septic shock [6]. The mechanisms leading to adrenal dysfunction in cirrhosis are poorly understood, but could include structural damage to the adrenal gland due to hemorrhage or infarction, increased levels of proinflammatory cytokines and impaired synthesis of cholesterol, the preferred source of steroidogenic substrate in the adrenal gland. At rest and during stress, around 80% of circulating cortisol is derived from plasma cholesterol, with the remaining 20% synthesized in situ from acetate and other precursors [10]. Low apolipoprotein A-I/HDL levels in critically ill patients may be pathogenetically linked to the high incidence of adrenal dysfunction in such patients, as it has been demonstrated that low HDL-cholesterol levels are associated with an attenuated response to synacthen [11].

This putative new concept of hepatoadrenal syndrome is supported by the many similarities between sepsis and end-stage liver disease such as the presence of hyperdynamic circulatory failure, decreased systemic vascular resistance, increased cardiac output and overproduction of interleukin (IL)-6 and tumor necrosis factor (TNF)-⍺. However, no proof of this theory is currently available, and there are no randomized controlled trials showing beneficial effects of corticotherapy in this patient group. In addition, the management of adrenal dysfunction in sepsis or septic shock remains controversial: the CORTICUS study group found no improvement in mortality, and only a speedier reversal of shock in patients receiving hydrocortisone. Moreover, steroids caused more episodes of superinfection, including new sepsis and septic shock [12].

The short synacthen test safely examines adrenal function and, in non-stressed subjects, adrenal dysfunction is defined as a post-short synacthen test cortisol level of less than 500 to 550nmol/L [13]. Up to now, no guidelines have been forthcoming from the French Societies of Endocrinology and of Anesthesia and Intensive Care, and there are only a few, limited US recommendations for diagnostic criteria to define adrenal dysfunction in critical illness, including a delta cortisol (cortisol increment after a short 250μg synacthen test) of less than 250nmol/L and a random total cortisol of less than 276nmol/L [14]. Moreover, a low-dose short synacthen test (1μg) is thought to be more sensitive for diagnosing infraclinical adrenal dysfunction in non-stressed patients. However, given the limited data available for the use of a low-dose short synacthen test, there is currently no consensus as to the appropriate high-dose or low-dose short synacthen test applicable as criteria for normal adrenal function. At this time, several questions still need to be addressed.

Is the assessment of adrenal function using serum total cortisol reliable?

More than 90% of the circulating cortisol in serum is bound to proteins (transcortin and albumin). This means that alterations in these binding proteins are likely to affect measures of serum total cortisol levels and the interpretation of the results of tests assessing adrenal function. Standard assays of serum cortisol determine total (bound plus free fractions) hormone concentrations and are probably inaccurate for assessing adrenal failure in patients with liver insufficiency, as transcortin and albumin are both synthesized in the liver. The present authors believe that the reported prevalences of adrenal dysfunction in cirrhotic patients are misleading and overestimated (Table 1), and need to be corrected by free cortisol measurement. Indeed, it was recently observed that nearly 40% of critically ill patients with hypoproteinemia had subnormal serum total cortisol values despite normal adrenal function, as determined by appropriately raised baseline and post-short synacthen test serum free cortisol levels [15]. Free cortisol is currently considered a biologically active hormone at the cellular level [14], and its serum measurement appears to be the most appropriate approach to avoid any drawbacks in cirrhotic patients with or without sepsis. However, as the current assays for measuring serum free cortisol levels are costly and time-consuming, they are not routinely performed. Furthermore, the results of this specialized test are usually delayed and not suitable for guiding acute clinical decisions. For this reason, attempts have been made to obtain information on free cortisol indirectly, using indices calculated from total cortisol and cortisol-binding globulin (also named transcortin), or directly, by analyzing urine or saliva.

Is the assessment of adrenal function using salivary cortisol promising?

Determining salivary cortisol concentrations is an easier and more reliable alternative means of measuring serum free cortisol, as cortisol levels in saliva are in equilibrium and correlate well with free cortisol levels [16]. In addition, venipuncture is more stressful than taking saliva samples and so could lead to an artifactual increase in cortisol. Furthermore, salivary cortisol testing can be carried out in standard hospital laboratories, and the results are quickly available, with a turn-around time of less than 3hours. Several commercial devices are suitable for the collection of saliva samples and subsequent cortisol analysis. Salivary cortisol concentrations are determined by direct radioimmunoassay, now often replaced by non-radioactive Elisa methods or by innovative technology such as liquid chromatography coupled with mass spectrometry [17]. Nevertheless, results need to be interpreted with caution, as wide variations in normal values are observed between laboratories using different assays or even between laboratories using the same technique [18]. To date, as there is no currently available universal assay to measure salivary cortisol, but only ‘home-made’ assays, it is mandatory for each laboratory to establish its own reference value. It should also be borne in mind that most commercial laboratory reference ranges are developed from healthy volunteer samples and generally do not account for age, gender or other specific medical conditions such as cirrhosis.

Despite these shortcomings, however, this easy test has already proved useful for evaluating critically ill patients with chronic renal failure, HIV infection and adrenal insufficiency [19]. Indeed, Galbois et al. [9] assessed adrenal function in 73 hemodynamically stable cirrhotic patients (74% Child–Pugh C) using salivary and serum total cortisol determinations. These two methods led to discrepant results, with adrenal dysfunction observed in 6.8% by cortisol salivary assessment compared with 31.5% using cortisol serum determinations, thereby confirming that the latter can overestimate adrenal insufficiency in cirrhotic patients. Serum total cortisol correlated with the salivary method only when the albumin level was more than 25g/L. In this study, the cutoff level of post-synacthen salivary cortisol was 12.9ng/mL (35.6nmol/L), but no correlation was performed with the serum free cortisol levels.

Another recent study assessed salivary cortisol and serum free cortisol levels in 51 critically ill patients (excluding liver disease), divided into two groups based on their serum albumin levels (less or equal to 25g/L or greater than 25g/L). The authors found that the rise in salivary cortisol and serum free cortisol levels were well correlated regardless of the cortisol-bound proteins [20]. This relationship between salivary and serum free cortisol was similar within the range of high serum free cortisol levels observed during critical illness (55 to 276nmol/L), but the data are lacking for such a correlation in the lower ranges. In contrast, serum total cortisol levels in patients with low albumin levels (less than 25g/L) were lower than in those with similar illnesses, but with near-normal albumin levels (greater than 25g/L), even though both groups of patients had similar elevated serum free and salivary cortisol concentrations. This study confirmed that measuring only the serum total cortisol in such hypoproteinemic patients led to overestimated diagnoses of adrenal dysfunction [20].

Two different approaches were deemed successful for measuring serum free cortisol levels and taking into account serum transcortin levels. One approach, the free cortisol index—calculated as cortisol concentration divided by transcortin level—is a promising surrogate marker [21]. However, this index needs to be further investigated in cirrhotic patients as it does not provide direct measurements of free cortisol and fails to take into account the frequently low serum albumin levels in these patients. The index is currently used more often as a correction factor to better interpret total cortisol. The other approach is the formula introduced by Coolens et al. [22], which estimates free cortisol levels from total cortisol and cortisol-binding globulin levels, thereby obviating the need for complex methods of serum free cortisol measurement such as centrifugal ultrafiltration or equilibrium dialysis. A well-designed study aimed at comparing total and free cortisol under basal and post-synacthen conditions evaluated the use of this method in patients with septic shock or sepsis and in healthy controls [23]. In this study, the Coolens calculation proved to be a reliable predictor of measured free cortisol even though some patients were hypoalbuminemic. However, additional data are needed to validate these findings in cirrhosis before suggesting the routine use of this complex formula (Table 2).

Adrenal failure in cirrhosis: what is the evidence?

Despite the difficulties of diagnosing adrenal dysfunction in septic cirrhotic patients, prompt identification of impaired cortisol production is of major clinical importance because of the possible key role of a blunted corticoid response in the clinical deterioration of such patients. The serum or salivary free cortisol assay is a promising test, but is not commonly used at present except in clinical research, probably because it is not readily available and the normal range is as yet undetermined. Pending further investigations of free cortisol dosages, we recommend implementing the American College of Critical Care Medicine suggestions for adrenal dysfunction diagnosis in critically ill cirrhotic patients [14]. However, the main drawback is that most critically ill patients are classified as adrenal dysfunction patients. As a result, they may be receiving unnecessary treatment with glucocorticoids and being exposed to the potential adverse effects of steroids (hyperglycemia, myopathy, neurotoxicity, fluid retention and immunosuppression).

As for serum free cortisol, two issues have yet to be clarified. The first deals with the relevant free cortisol range required to produce an effect, a point that is likely to remain unresolved for some time as it is not possible to measure tissue glucocorticoid resistance. Nevertheless, it has been said that, in sepsis, there may be a reduced affinity of cellular receptors for glucocorticoids. The second issue is the reference range of serum free cortisol that is considered the gold-standard dose in healthy and septic cirrhosis patients. During critical illnesses, adrenal dysfunction may be defined by a serum free cortisol level of less than 50 to 55nmol/L at baseline or less than 86nmol/L after synacthen stimulation [15, 16]. The reference range for baseline serum free cortisol in healthy subjects also varies—from 8.3–25.9nmol/L, according to Arafah [16], to 12.4–70.1nmol/L, according to the Strasbourg laboratory [24]. In addition, growing experience with salivary cortisol testing shows that a significant percentage of cirrhotic or intubated critically ill patients have abnormal salivary flow and, as a result, produce an inadequate salivary sample. Furthermore, gingival or buccal hemorrhage (leading to serum contamination) and oral candidiasis (likely to lead to underestimation of cortisol concentration) place additional limits on salivary sampling. Prospective trials in such patients are needed to ascertain whether or not salivary cortisol is a good surrogate marker—defined as a convincing association with a definitive clinical outcome—and to determine the best cutoff point for differentiating patients with and without adrenal dysfunction.

In conclusion, in spite of the shortcomings mentioned above, cortisol salivary measurement is a promising means of diagnosing suspected adrenal dysfunction in cirrhotic patients, and is of particular value in cases of altered serum protein levels. Salivary cortisol testing has the following advantages: it is less invasive than the standard synacthen test, administered intravenously, and allows multiple samples to be collected in outpatients; it reflects the biological active steroid; and it may be better at detecting critically ill patients with true adrenal dysfunction, regardless of serum protein levels, thus avoiding prolonged and potentially harmful high-dose glucocorticoid therapy.

However, the optimal method and cutoff values for diagnosing possible adrenal dysfunction remain a matter of debate. Another concern is the reproducibility of the synacthen test in septic shock patients, as changes in cortisol production can occur within very short periods of time, which means that a single assessment by synacthen test could fail to detect adrenal dysfunction. The time interval between a patient’s admission to hospital and the synacthen test may also greatly affect observed cortisol values. In addition, a blunted response may be more indicative of adrenal exhaustion than insufficiency.

Nevertheless, the CORTICIR study is ongoing and should increase our knowledge of the relationship between free cortisol measured in saliva and serum, and serum total cortisol, in both cirrhotic patients with various degrees of liver insufficiency and septic cirrhotic patients.

Conflict of interests

None.


Acknowledgements

We thank Professor Jean-Louis Schlienger for his critical reading of the manuscript, and the following members of the CORTICIR study group for their contributions to the ongoing study: Jean-Paul Cervoni, Denis Cléau, Vincent Di Martino, Serge Fratte, Patrick Hillon, Stéphanie Kury, Stéphanie Lemaire, Anne Minello, Emilie Muel, Carinne Richou, Franck Schillo, Pascal Thibault, Simona Tirziu and Claire Vanlemmens.



References

[1]
Harry R., Auzinger G., Wendon J. The clinical importance of adrenal insufficiency in acute hepatic dysfunction Hepatology 2002 ;  36 : 395-402 [cross-ref]
[2]
Harry R., Auzinger G., Wendon J. The effects of supraphysiological doses of corticosteroids in hypotensive liver failure Liver Int 2003 ;  23 : 71-77 [cross-ref]
[3]
Marik P.E., Gayowski T., Starzl T.E. The hepatoadrenal syndrome: a common yet unrecognized clinical condition Crit Care Med 2005 ;  33 : 1254-1259 [cross-ref]
[4]
Fernandez J., and al. Adrenal insufficiency in patients with cirrhosis and septic shock: effect of treatment with hydrocortisone on survival Hepatology 2006 ;  44 : 1288-1295 [cross-ref]
[5]
Tsai M.H., and al. Adrenal insufficiency in patients with cirrhosis, severe sepsis and septic shock Hepatology 2006 ;  43 : 673-681 [cross-ref]
[6]
Thierry S., Giroux Leprieur E., Lecuyer L., Brocas E., Van de Louw A. Echocardiographic features, mortality, and adrenal function in patients with cirrhosis and septic shock Acta Anaesthesiol Scand 2008 ;  52 : 45-51
[7]
Sigalas A., Xirouchakis E., Manousou P., Corbani A., Calvaruso V., Patch D., and al. Adrenal impairment is frequent finding in stable cirrhosis and is related to disease severity Hepatology 2007 ;  46 : 573A
[8]
Jang J.Y., Cho W.Y., Jeong S.W., Kim S.G., Cheon Y.K., Kim Y.S., and al. Relative adrenal insufficiency in patients with chronic liver disease Hepatology 2008 ;  48 : 1088A
[9]
Galbois A., Massard J., Reignier S., Fulla Y., Bennani A., Bonnefont-Rousselo D., and al. Hepato-adrenal syndrome: myth or reality? Usefulness of salivary cortisol assay Hepatology 2008 ;  48 : 1054A
[10]
Borkowski A.J., Levin S., Delcroix C., Mahler A., Verhas V. Blood cholesterol and hydrocortisone production in man: quantitative aspects of the utilization of circulating cholesterol by the adrenals at rest and under adrenocorticotropin stimulation J Clin Invest 1967 ;  46 : 797-811
[11]
Van der Voort P.H., Gerritsen R.T., Bakker A.J., Boerma E.C., Kuiper M.A., de Heide L. HDL-cholesterol level and cortisol response to synacthen in critically ill patients Intensive Care Med 2003 ;  29 : 2199-2203 [cross-ref]
[12]
Sprung C.L., Annane D., Keh D., Moreno R., Singer M., Freivogel K., and al. Hydrocortisone therapy for patients with septic shock N Engl J Med 2008 ;  358 : 111-124 [cross-ref]
[13]
Annane D., Maxime V., Ibrahim F., Alvarez J.C., Abe E., Boudou P. Diagnosis of adrenal insufficiency in severe sepsis and septic shock Am J Respir Crit Care Med 2006 ;  174 : 1319-1326 [cross-ref]
[14]
Marik P.E., Pastores S.M., Annane D., Meduri G.U., Sprung C.L., Arlt W., and al. Recommendations for the diagnosis and management of corticosteroid insufficiency in critically ill adult patients: consensus statements from an international task force by the American College of Critical Care Medicine Crit Care Med 2008 ;  36 : 1937-1949 [cross-ref]
[15]
Hamrahian A.H., Oseni T.S., Arafah B.M. Measurements of serum free cortisol in critically ill patients N Engl J Med 2004 ;  350 : 1629-1638 [cross-ref]
[16]
Arafah B.M. Hypothalamic pituitary adrenal function during critical illness: limitations of current assessment methods J Clin Endocrinol Metab 2006 ;  91 : 3725-3745 [cross-ref]
[17]
Gröschl M. Current status of salivary hormone analysis Clin Chem 2008 ;  54 : 1759-1769
[18]
Nunes M.L., Vattaut S., Corcuff J.B., Rault A., Loiseau H., Gatta B., and al. Late-night salivary cortisol for diagnosis of overt and subclinical Cushing’s syndrome in hospitalized and ambulatory patients J Clin Endocrinol Metab 2009 ;  94 : 456-462
[19]
Raff H., Brock S., Findling J.W. Cosyntropin-stimulated salivary cortisol in hospitalized patients with hypoproteinemia Endocrine 2008 ;  34 : 68-74 [cross-ref]
[20]
Arafah B.M., Nishiyama F.J., Tlaygeh H., Hejal R. 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 2007 ;  92 : 2965-2971 [cross-ref]
[21]
Le Roux C.W., Chapman G.A., Kong W.M., Dhillo W.S., Jones J., Alaghband-Zadeh J. Free cortisol index is better than serum total cortisol in determining hypothalamic-pituitary-adrenal status in patients undergoing surgery J Clin Endocrinol Metab 2003 ;  88 : 2045-2048 [cross-ref]
[22]
Coolens J., Baelen H.V., Heyns W. Clinical use of unbound plasma cortisol as calculated from total cortisol and corticosteroid-binding globulin J Steroid Biochem 1987 ;  26 : 197-202 [cross-ref]
[23]
Ho J.T., Al-Musalhi H., Chapman M.J., Quach T., Thomas P.D., Bagley C.J., and al. Septic shock and sepsis: a comparison of total and free plasma cortisol levels J Clin Endocrinol Metab 2006 ;  91 : 105-114 [cross-ref]
[24]
Schlienger J.L., Sapin R., Gasser F., Briche-Prouveur S., Dreyfuss M. Intérêt du cortisol libre plasmatique en pratique clinique Sem Hop Paris 1989 ;  65 : 2067-2070

1  Members of the Corticir study group.


© 2009  Elsevier Masson SAS. All Rights Reserved.
EM-CONSULTE.COM is registrered at the CNIL, déclaration n° 1286925.
As per the Law relating to information storage and personal integrity, you have the right to oppose (art 26 of that law), access (art 34 of that law) and rectify (art 36 of that law) your personal data. You may thus request that your data, should it be inaccurate, incomplete, unclear, outdated, not be used or stored, be corrected, clarified, updated or deleted.
Personal information regarding our website's visitors, including their identity, is confidential.
The owners of this website hereby guarantee to respect the legal confidentiality conditions, applicable in France, and not to disclose this data to third parties.
Close
Article Outline