- Email: [email protected]
HPA hyperactivity with increased plasma cortisol affects dexamethasone metabolism and DST outcome
Journal of Psychiatric Research 36 (2002) 417–421 www.elsevier.com/locate/jpsychires
HPA hyperactivity with increased plasma cortisol affects dexamethasone metabolism and DST outcome Peter E. Stokes*, Carolyn Sikes, Betty Lasley, Peter Stoll New York Presbyterian Hospital, Weill Medical College of Cornell University, 21 Bloomingdale Road, White Plains, NY 10605, USA Received 4 April 2002; received in revised form 19 August 2002; accepted 23 August 2002
Abstract Data suggests that dexamethasone bioavailability or pharmacokinetic factors contribute importantly to the outcome of the dexamethasone suppression test, and a relationship between plasma cortisol and plasma dexamethasone levels has been shown. To evaluate these data further, we studied plasma dexamethasone pharmacokinetics in 24 patients with major depression (15 suppressors and nine nonsuppressors) who received a 1 mg IV dexamethasone bolus at 09:00 h with blood samples collected at intervals over the next 14 h. We found that nonsuppressors had significantly shorter plasma dexamethasone half-life (P=0.003) as well as significantly lower dexamethasone levels 10 h (P=0.02) following IV dexamethasone administration. Moreover, upon clinical improvement of patients, the shortened dexamethasone half-life and lower dexamethasone levels disappeared in the five patients who switched from nonsuppression to suppression and were restudied by IV bolus. These 10-h post IV plasma dexamethasone level findings paralleled the results of the 1 mg overnight oral DST performed in these depressed patients (N=22) where we found significantly lower 10 h plasma dexamethasone levels in nonsuppressors on admission compared to suppressors (P=0.002) and again at discharge (P=0.007). Interestingly, in the few patients who switched from suppression to nonsuppression over the course of hospitalization, 10-h post dose plasma dexamethasone levels simultaneously dropped. No difference in dexamethasone half-life was observed in the patients studied by oral and IV dexamethasone administration. These findings support the concept that metabolism of dexamethasone is significantly related to the activity of the HPA axis (particularly by plasma cortisol levels), and that dexamethasone pharmacokinetics can be modified by state-dependent phenomena. # 2002 Elsevier Science Ltd. All rights reserved. Keywords: DST; Dexamethasone half-life; Hypercortisolemia; Cortisol; Depression
1. Introduction Bioavailability of dexamethasone has in recent years been of definite interest in psychiatry particularly as regards the design and interpretation of the dexamethasone suppression test (DST). A large number of dexamethasone studies have reported lower 08:00 and 16:00 h plasma dexamethasone levels in nonsuppressors compared to suppressors and have revealed an inverse relationship between post dexamethasone plasma cortisol levels and plasma dexamethasone levels during the DST (Arana et al., 1984; Berger et al., 1984; Johnson et al., 1984; Stokes et al., 1987; Wiedemann and Holsboer,
* Corresponding author. Tel.: +1-914-997-4392; fax: +1-914-9975958. E-mail address: [email protected] (P.E. Stokes).
1987; Cassidy et al., 2000). In a study of 31 depressed patients who converted from dexamethasone nonsuppression to suppression on re-test after clinical remission of depression, significantly lower plasma levels of dexamethasone were reported during the depression than after recovery and normalization of the DST (Holsboer et al., 1986). Also, nonsuppressors were found to have shorter dexamethasone half-lives than suppressors. Pre-dexamethasone plasma cortisol levels have been reported to be statistically significant predictors of the outcome of DST in depressed patients (Poland et al., 1987) and are probably a contributing factor in development of depression (Stokes, 1995) and associated cognitive impairment (Wolkowitz et al., 1990; Sikes et al., 1990). Animal studies reveal that cortisol levels before and during the DST are probably the causal agent controlling dexamethasone metabolism (Stokes et al., in press). These findings emphasize the influence of variable
0022-3956/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved. PII: S0022-3956(02)00059-6
418
P.E. Stokes et al. / Journal of Psychiatric Research 36 (2002) 417–421
dexamethasone bioavailability secondary to altered metabolism and pharmacokinetics in determining DST results. Variations in the absorption and distribution of dexamethasone (early kinetics) do not appear to be significant factors determining decreased dexamethasone bioavailability. No differences were found during the absorption and distribution phase in peak plasma levels, time to peak, or area under the curve after oral dexamethasone in suppressors vs. nonsuppressors (Wiedemann and Holsboer, 1987). Body weight, surface area, body mass index, age, and sex during DST have shown no correlation to plasma dexamethasone levels (Morris et al., 1986), or only a weak negative correlation to body weight (Stokes et al., 1984). Thus data reporting shorter plasma dexamethasone half-lives and associated lower plasma dexamethasone concentrations appeared to us to be primarily related to changes in the elimination phase of dexamethasone kinetics, i.e. faster dexamethasone metabolism or excretion. We have reported data from animal studies supporting this concept (Stokes et al., in press). This study’s principal aim was to explore this possibility further in humans.
2. Patients and methods Twenty-four patients hospitalized for treatment of their major depression (DSM and meeting RDC criteria) were approached and included in the study protocol after obtaining their informed consent if HAM-D was greater than 21. All procedures were approved by the Institutional Review Board of the Weill Medical College of Cornell University. No patients were on any sedatives or other medications known to alter the pharmacokinetics of steroids. A few patients were on antidepressants (mainly tricyclics) on admission, and all were on antidepressants at discharge. All patients had completed physical examination, biochemical and hematological profiles and had no significant acute or chronic medical illness. Each patient underwent a standard overnight (23:00 h) oral 1 mg DST within 3 days of admission and was classified as nonsuppressor if post dexamethasone 09:00 h plasma cortisol was > 5.0 mcg/ dl. Three days after this oral DST and before instituting any change in pharmacotherapy, all patients (15 suppressors, nine nonsuppressors) underwent an IV 1 mg dexamethasone bolus at 09:00 h (injected over 3 min) with plasma dexamethasone sampling 1, 2, 3, 4, 7, 10 and 14 h later (i.e. at 23:00 h) via indwelling venous catheter. Plasma levels of cortisol were determined by radioimmunoassay (Clinical Assays, Baxter International, Cambridge MA, 02139). All cortisol levels from any single patient were assayed in duplicate in a single run.
Intra-assay coefficients of variation were 11.6 and 7.6% for plasma pools containing levels of cortisol at the low and high ends of the range of values observed. Dexamethasone levels were determined in extracted samples. The dexamethasone assay followed a double antibody method outlined by the IgG Corporation (Nashville, TN) as previously described (Stokes et al., in press). Dexamethasone half-lives were derived from dexamethasone levels determined in duplicate from samples assayed within the same run. Intra-assay coefficients of variation were 12.3 and 15.0% for plasma pools containing levels of dexamethasone at the low and medium points of the range of values observed. Dexamethasone levels were plotted against time, and half-lives were derived from the slopes of the lines best fit to the decay curves by the power curve method. Statistical analyses consisted of Student’s t-tests and Pearson correlation coefficients with two tailed tests of significance.
3. Results In the 1 mg IV dexamethasone bolus studies (09:00 h), the ten hour dexamethasone levels (23:00 h) were significantly lower (1.33 1.0 ng/ml) in nine nonsuppressors than in 15 suppressor (2.44 1.22 ng/ml) depressed patients (P=0.02), a reduction of about 45%. In these patients, dexamethasone half-life was significantly shortened from mean 4.75 h ( 1.4) in suppressors to 3.25 h ( 1.66) in nonsuppressors (P=0.003). A high correlation was observed between dexamethasone half-life and 10 hour post dexamethasone level (r=0.93, P=0.007, n=24). No difference in half-life was observed in oral (mean=3.63 h) vs. IV (mean=3.73 h) dexamethasone half-life in (three) patients studied by both oral and IV dexamethasone. After 1 mg oral overnight DST on admission, 09:00 plasma dexamethasone levels were significantly lower in nonsuppressors (1.24 0.4 ng/ml; n=8) compared to suppressors (3.42 2.08; n=14; P < 0.002). Five of the eight oral nonsuppressors on admission who had oral overnight DST repeated at discharge showed significant increase in plasma dexamethasone level when the DST was suppressing, and their depression had responded to treatment with HDRS decreased by 50% from admission to lower than 10 at discharge (admission nonsuppression=1.24 0.39 ng/ml; discharge suppression= 2.51 1.06 ng/ml). Eleven of the 14 oral suppressors on admission had a repeat oral DST at discharge and remained suppressor showing no significant change in plasma dexamethasone levels (2.91 1.44 ng/ml on admission; 3.37 1.84 ng/ml at discharge). These 11 patients admission plasma dexamethasone levels were not significantly different from that found in all 14 suppressors on admission (Table 1).
419
P.E. Stokes et al. / Journal of Psychiatric Research 36 (2002) 417–421 Table 1 Summary of major findings Admission IV dexamethasone studies # Pts
Dex test status
9
NS
15
Supp
X 10 h post dex Plasma dex level
X Dex half-Life
1.33 (P=0.02) 2.44
3.25 (P=0.003) 4.75
Oral dexamethasone studies On admission
On discharge
# Pts
Dex test Status
X 10 h post dex Plasma dex level
8
NS
1.24
14
Supp
3.42
! (P=0.002) !
4. Discussion The principal findings in this study are the significantly shorter dexamethasone half-life and lower 10 h post dexamethasone levels found in nonsuppressor vs. suppressor depressed patients in both IV and oral DST studies. In addition, we found no difference in dexamethasone half-life between oral vs. IV dexamethasone studies done in the same three patients. Further, the change from nonsuppressor on admission to suppressor status at discharge was associated with an increase in 10 h plasma dexamethasone level while no changes in plasma dexamethasone level was found in suppressors on admission who remained suppressing at discharge. These findings extend and support previous data showing shorter dexamethasone half-life and lower plasma dexamethasone levels in nonsuppressors (Wiedemann and Holsboer, 1987; Holsboer et al., 1986; Poland et al., 1987; Stokes et al., 1987; Cassidy et al., 2000). Wiedemann and Holsboer (1987) reported no difference in dexamethasone kinetics during the absorption and distribution (early) phase after oral dexamethasone administration but found that the later elimination phase of dexamethasone from the circulation was significantly enhanced (i.e. shorter half-life) in six oral DST nonsuppressors resulting in an association of decreased plasma dexamethasone with elevated post dexamethasone cortisol levels. They also reported five IV nonsuppressors and eight IV suppressors whose plasma dexamethasone kinetics were indistinguishable and in the same order of magnitude as those of (14) nonsuppressors after oral DST. Examination of their IV dexamethasone data reveals a trend to lower plasma dexamethasone levels in the five nonsuppressors studied as compared to the eight IV nonsuppressors. We find significantly shorter half-life (i.e. increased dexamethasone metabolism) and lower plasma dexamethasone levels in
# Pts
Dex test Status
X 10 h post dex Plasma dex level
5
S
11
S
2.51 (P=NS) 3.37
nonsuppressors during the later elimination phase after IV dexamethasone administration, and paralleling this, lower dexamethasone levels in oral dexamethasone nonsuppressors consistent with shorter dexamethasone half-life and the report of Cassidy et al. (2000). There are several possible reasons for the reported differences in findings between the Wiedemann and Holsboer study and our current study. First, only five of their total number of IV dexamethasone studies (13) were nonsuppressors by their definition of nonsuppression. They defined nonsuppression as a post dexamethasone plasma cortisol > 4 ucg/dl at any one of the three post dexamethasone time points sampled (i.e. 08:00, 16:00, and 23:00 h). We have used the more demanding definition of a required 09:00 h plasma cortisol level greater than 5 ucg/dl to define nonsuppression as discussed previously (Stokes et al., 1984). One third (11/33) of their total patients were nonsuppressing by their criteria of nonsuppression. By our more rigorous definitions of nonsuppression, only three of their oral and only two of their IV dexamethasone study patients (5/33 or 15%) were nonsuppressing, and therefore comparable to our patients. Further, none of their IV dexamethasone and oral dexamethasone studies were compared in the same patients. Plasma cortisol levels during the day following nighttime administration of dexamethasone (circa 23:00 h) in Wiedemann and Holsboer’s study reveal marked fluctuations after the 08:00 h sample the next morning (see their Table 2). This variability was previously discussed (Stokes et al., 1984) where post dexamethasone plasma cortisol ‘‘patterns’’ were identified and found variable, especially in DST nonsuppressors. Thus, following evening administration of dexamethasone cortisol patterns from early morning to mid day to evening the next day may be described as showing various permutations of nonsuppression or suppression levels of cortisol at the three post dexamethasone sampling times. This prob-
420
P.E. Stokes et al. / Journal of Psychiatric Research 36 (2002) 417–421
ably reflects the initial and major early post administration suppressing affect of the dexamethasone followed by decreasing dexamethasone suppression as plasma and tissue levels fall. Superimposed on this is the recovering central ‘‘drive’’ of inherent brain pituitary activity overlain by transient sensory inputs (exciting or inhibiting) of the day (Stokes and Sikes, 1987) and by inadvertent sampling at various times during a secretory phase. The result is the fluctuation in post dexamethasone cortisol levels. A mechanism for shortened dexamethasone half-life in nonsuppressing patients has recently been proposed by our group (Stokes et al., in press). Many tissues metabolize glucocorticoids but experimental evidence supports the liver as the most likely site for the bulk of dexamethasone metabolism. Although cortisol itself has not been directly shown to induce hepatic microsomal enzymes, chronic administration of various synthetic glucocorticoids has been shown to be associated with increased metabolism of dexamethasone even in humans with renal insufficiency and no increase in renal steroid excretion (Araki et al., 1966). It is more than probable that the degradation of cortisol and dexamethasone share common pathways (Simmons et al., 1984). We postulated that the decreased dexamethasone half-life and lower plasma dexamethasone levels in psychiatrically ill patients who are DST nonsuppressors could be secondary to the frequently persistent moderate hypercortisolemia associated with but not specific to depression (Stokes et al., 1976; 1984) and the consequent potential probable induction of hepatic enzymes metabolizing dexamethasone. Our studies in rabbits (Stokes et al., in press), under pharmacologically induced hypercortisolemia, supported our hypothesis. In that study, dexamethasone plasma levels and half-life decreased (mean=41%) from control baseline values after induction of hypercortisolemia and returned to control values after cessation of the induced hypercortisolemia. Further, pre-dexamethasone cortisol values correlated negatively and highly with dexamethasone half-life. Interestingly, the percentage decrease in dexamethasone half-life in nonsuppressor patients compared to suppressors described by Holsboer et al., (1986; 43%) and in our current study (33%) is of similar magnitude (42%) we found in our rabbits during experimentally induced hypercortisolemia compared to their pre-hypercortisolemic state and consistent with that (41%) observed by Araki et al. (1966) in human subjects treated chronically with various glucocorticoids. The relatively small number of patients studied in our intravenous protocol and in other such studies (Wiedemann and Holsboer, 1987) suggests that confirmatory investigation be completed in larger numbers of clearly dexamethasone nonsuppressing versus suppressing patients. It will also be useful to confirm our findings of
no difference in dexamethasone half-life between IV and oral dexamethasone administration in the same patients. In addition, a study of dexamethasone half-life in patients before and during glucocorticoid treatment, preferably with hydrocortisone, would provide additional clarifying data. We wanted to assay urinary levels of the cortisol metabolite 6BOH cortisol in these suppressors and nonsuppressors as a further measure of altered hepatic enzyme activity (probably CYP3A4) since glucocorticoids have been shown to induce CYP3A, but were unable to accomplish this portion of the study.
Acknowledgements From the Laboratory of Psychobiology—Endocrinology in the Department of Psychiatry at Weill Medical College of Cornell University. Supported by grants from the Citicorp-Wriston Endowment Scholarship Fund, George F. Baker Trust, and gifts from Robin Jaffe-Cohen. The authors wish to thank, Hermina Ombid, our laboratory assistant, and research assistant, Alexandra I. Barsdorf for statistical analyses, library research, and manuscript preparation.
References Araki Y, Yokota O, Kato T, Kashima M, Miyazaki T. Dynamics of synthetic corticosteroids in man. In: Pincus G, Nakao T, Tait JF, editors. Steroid dynamics. New York: Academic Press; 1966. p. 474–7. Arana GW, Workman RJ, Baldessarini RJ. Association between low plasma levels of dexamethasone and elevated levels of cortisol in psychiatric patients given dexamethasone. American Journal of Psychiatry 1984;141:1619–20. Berger M, Pirke KM, Doerr P, Krieg JC, von Zerssen D. The limited utility of the dexamethasone suppression test for the diagnostic process in psychiatry. British Journal of Psychiatry 1984;145:372– 82. Cassidy F, Ritchie JC, Verghese K, Carroll BJ. Dexamethasone metabolism in dexamethasone suppression test suppressors and nonsuppressors. Biological Psychiatry 2000;47:677–80. Holsboer F, Wiedemann K, Gerken A, Boll E. The plasma dexamethasone variable in depression: test–retest studies and early biophase kinetics. Psychiatry Research 1986;17(2):97–103. Johnson GF, Hunt G, Lerr K, Caterson I. Dexamethasone supression test (DST) and plasma dexamethasone levels in depressed patients. Psychiatry Research 1984;13:305–13. Morris H, Carr V, Gilliland J, Hooper M. Dexamethasone concentrations and the dexamethasone suppression test in psychiatric disorders. British Journal of Psychiatry 1986;148:66–9. Poland RE, Rubin RT, Lesser IM, Lane LA, Hart PJ. Neuroendocrine aspects of primary endogenous depression: II. Serum dexamethasone concentrations and hypothalamic-pituitary-adrenal cortical activity as determinants of the dexamethasone suppression test response. Archives of General Psychiatry 1987;44:790–5. Sikes CR, Stokes PE, Lasley B, Carson RC. Cognitive sequelae of hypothalamic-pituitary-adrenocortical dysregulation. The 143rd Annual Meeting of the American Psychiatric Association, New York, NY, 12–17 May, 1990.
P.E. Stokes et al. / Journal of Psychiatric Research 36 (2002) 417–421 Simmons PS, Miles JM, Gerich JE, Haymond MW. Increased proteolysis: an effect of increases in plasma cortisol within the physiologic range. Journal of Clinical Investigation 1984;73:412. Stokes PE. The potential role of excessive cortisol induced by HPA hyperfunction in the pathogenesis of depression. European Neuropsychopharmacology 1995;5(Suppl):77–82. Stokes PE, Sikes C. Hypothalamic-pituitary-adrenal axis in affective disorders. In: Meltzer HY, editor. Psychopharmacology: a third generation of progress. New York: Raven Press; 1987. p. 589– 607. Stokes PE, Sikes CR, Stoll PM, Lasley BJ. Dexamethasone half-life affects DST response. Presented at the 42nd annual meeting of the Society of Biological Psychiatry, Chicago, IL May, 1987. Stokes PE, Stoll PM, Koslow SH, Maas JW, Davis JM, Swann AC, Robins E. Pretreatment DST and hypothalamic-pituitary-adreno-
421
cortical function in depressed patients and comparison groups: a multi-center study. Archives of General Psychiatry 1984;41:257–67. Stokes PE, Stoll PM, Mattson MR, Sollod RN. Diagnosis and psychopathology in psychiatric patients resistant to dexamethasone. In: Sachar EJ, editor. Hormones, behavior, and psychopathology. New York: Raven Press; 1976. p. 225–9. Stokes PE, Stoll PM, Schluger JH, Lasley B. Hypercortisolemia decreases dexamethasone half-life in rabbit. Journal of Psychiatric Research 2002 (in press). Wiedemann K, Holsboer F. Plasma dexamethasone kinetics during the DST after oral and intravenous administration of the test drug. Biological Psychiatry 1987;22:1340–8. Wolkowitz OM, Reus VI, Weingartner H, Thompson K, Breier A, Doran A, Rubinow D, Pickar D. Cognitive effects of corticosteroids. American Journal of Psychiatry 1990;147(10):1297–303.
HPA hyperactivity with increased plasma cortisol affects dexamethasone metabolism and DST outcome Peter E. Stokes*, Carolyn Sikes, Betty Lasley, Peter Stoll New York Presbyterian Hospital, Weill Medical College of Cornell University, 21 Bloomingdale Road, White Plains, NY 10605, USA Received 4 April 2002; received in revised form 19 August 2002; accepted 23 August 2002
Abstract Data suggests that dexamethasone bioavailability or pharmacokinetic factors contribute importantly to the outcome of the dexamethasone suppression test, and a relationship between plasma cortisol and plasma dexamethasone levels has been shown. To evaluate these data further, we studied plasma dexamethasone pharmacokinetics in 24 patients with major depression (15 suppressors and nine nonsuppressors) who received a 1 mg IV dexamethasone bolus at 09:00 h with blood samples collected at intervals over the next 14 h. We found that nonsuppressors had significantly shorter plasma dexamethasone half-life (P=0.003) as well as significantly lower dexamethasone levels 10 h (P=0.02) following IV dexamethasone administration. Moreover, upon clinical improvement of patients, the shortened dexamethasone half-life and lower dexamethasone levels disappeared in the five patients who switched from nonsuppression to suppression and were restudied by IV bolus. These 10-h post IV plasma dexamethasone level findings paralleled the results of the 1 mg overnight oral DST performed in these depressed patients (N=22) where we found significantly lower 10 h plasma dexamethasone levels in nonsuppressors on admission compared to suppressors (P=0.002) and again at discharge (P=0.007). Interestingly, in the few patients who switched from suppression to nonsuppression over the course of hospitalization, 10-h post dose plasma dexamethasone levels simultaneously dropped. No difference in dexamethasone half-life was observed in the patients studied by oral and IV dexamethasone administration. These findings support the concept that metabolism of dexamethasone is significantly related to the activity of the HPA axis (particularly by plasma cortisol levels), and that dexamethasone pharmacokinetics can be modified by state-dependent phenomena. # 2002 Elsevier Science Ltd. All rights reserved. Keywords: DST; Dexamethasone half-life; Hypercortisolemia; Cortisol; Depression
1. Introduction Bioavailability of dexamethasone has in recent years been of definite interest in psychiatry particularly as regards the design and interpretation of the dexamethasone suppression test (DST). A large number of dexamethasone studies have reported lower 08:00 and 16:00 h plasma dexamethasone levels in nonsuppressors compared to suppressors and have revealed an inverse relationship between post dexamethasone plasma cortisol levels and plasma dexamethasone levels during the DST (Arana et al., 1984; Berger et al., 1984; Johnson et al., 1984; Stokes et al., 1987; Wiedemann and Holsboer,
* Corresponding author. Tel.: +1-914-997-4392; fax: +1-914-9975958. E-mail address: [email protected] (P.E. Stokes).
1987; Cassidy et al., 2000). In a study of 31 depressed patients who converted from dexamethasone nonsuppression to suppression on re-test after clinical remission of depression, significantly lower plasma levels of dexamethasone were reported during the depression than after recovery and normalization of the DST (Holsboer et al., 1986). Also, nonsuppressors were found to have shorter dexamethasone half-lives than suppressors. Pre-dexamethasone plasma cortisol levels have been reported to be statistically significant predictors of the outcome of DST in depressed patients (Poland et al., 1987) and are probably a contributing factor in development of depression (Stokes, 1995) and associated cognitive impairment (Wolkowitz et al., 1990; Sikes et al., 1990). Animal studies reveal that cortisol levels before and during the DST are probably the causal agent controlling dexamethasone metabolism (Stokes et al., in press). These findings emphasize the influence of variable
0022-3956/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved. PII: S0022-3956(02)00059-6
418
P.E. Stokes et al. / Journal of Psychiatric Research 36 (2002) 417–421
dexamethasone bioavailability secondary to altered metabolism and pharmacokinetics in determining DST results. Variations in the absorption and distribution of dexamethasone (early kinetics) do not appear to be significant factors determining decreased dexamethasone bioavailability. No differences were found during the absorption and distribution phase in peak plasma levels, time to peak, or area under the curve after oral dexamethasone in suppressors vs. nonsuppressors (Wiedemann and Holsboer, 1987). Body weight, surface area, body mass index, age, and sex during DST have shown no correlation to plasma dexamethasone levels (Morris et al., 1986), or only a weak negative correlation to body weight (Stokes et al., 1984). Thus data reporting shorter plasma dexamethasone half-lives and associated lower plasma dexamethasone concentrations appeared to us to be primarily related to changes in the elimination phase of dexamethasone kinetics, i.e. faster dexamethasone metabolism or excretion. We have reported data from animal studies supporting this concept (Stokes et al., in press). This study’s principal aim was to explore this possibility further in humans.
2. Patients and methods Twenty-four patients hospitalized for treatment of their major depression (DSM and meeting RDC criteria) were approached and included in the study protocol after obtaining their informed consent if HAM-D was greater than 21. All procedures were approved by the Institutional Review Board of the Weill Medical College of Cornell University. No patients were on any sedatives or other medications known to alter the pharmacokinetics of steroids. A few patients were on antidepressants (mainly tricyclics) on admission, and all were on antidepressants at discharge. All patients had completed physical examination, biochemical and hematological profiles and had no significant acute or chronic medical illness. Each patient underwent a standard overnight (23:00 h) oral 1 mg DST within 3 days of admission and was classified as nonsuppressor if post dexamethasone 09:00 h plasma cortisol was > 5.0 mcg/ dl. Three days after this oral DST and before instituting any change in pharmacotherapy, all patients (15 suppressors, nine nonsuppressors) underwent an IV 1 mg dexamethasone bolus at 09:00 h (injected over 3 min) with plasma dexamethasone sampling 1, 2, 3, 4, 7, 10 and 14 h later (i.e. at 23:00 h) via indwelling venous catheter. Plasma levels of cortisol were determined by radioimmunoassay (Clinical Assays, Baxter International, Cambridge MA, 02139). All cortisol levels from any single patient were assayed in duplicate in a single run.
Intra-assay coefficients of variation were 11.6 and 7.6% for plasma pools containing levels of cortisol at the low and high ends of the range of values observed. Dexamethasone levels were determined in extracted samples. The dexamethasone assay followed a double antibody method outlined by the IgG Corporation (Nashville, TN) as previously described (Stokes et al., in press). Dexamethasone half-lives were derived from dexamethasone levels determined in duplicate from samples assayed within the same run. Intra-assay coefficients of variation were 12.3 and 15.0% for plasma pools containing levels of dexamethasone at the low and medium points of the range of values observed. Dexamethasone levels were plotted against time, and half-lives were derived from the slopes of the lines best fit to the decay curves by the power curve method. Statistical analyses consisted of Student’s t-tests and Pearson correlation coefficients with two tailed tests of significance.
3. Results In the 1 mg IV dexamethasone bolus studies (09:00 h), the ten hour dexamethasone levels (23:00 h) were significantly lower (1.33 1.0 ng/ml) in nine nonsuppressors than in 15 suppressor (2.44 1.22 ng/ml) depressed patients (P=0.02), a reduction of about 45%. In these patients, dexamethasone half-life was significantly shortened from mean 4.75 h ( 1.4) in suppressors to 3.25 h ( 1.66) in nonsuppressors (P=0.003). A high correlation was observed between dexamethasone half-life and 10 hour post dexamethasone level (r=0.93, P=0.007, n=24). No difference in half-life was observed in oral (mean=3.63 h) vs. IV (mean=3.73 h) dexamethasone half-life in (three) patients studied by both oral and IV dexamethasone. After 1 mg oral overnight DST on admission, 09:00 plasma dexamethasone levels were significantly lower in nonsuppressors (1.24 0.4 ng/ml; n=8) compared to suppressors (3.42 2.08; n=14; P < 0.002). Five of the eight oral nonsuppressors on admission who had oral overnight DST repeated at discharge showed significant increase in plasma dexamethasone level when the DST was suppressing, and their depression had responded to treatment with HDRS decreased by 50% from admission to lower than 10 at discharge (admission nonsuppression=1.24 0.39 ng/ml; discharge suppression= 2.51 1.06 ng/ml). Eleven of the 14 oral suppressors on admission had a repeat oral DST at discharge and remained suppressor showing no significant change in plasma dexamethasone levels (2.91 1.44 ng/ml on admission; 3.37 1.84 ng/ml at discharge). These 11 patients admission plasma dexamethasone levels were not significantly different from that found in all 14 suppressors on admission (Table 1).
419
P.E. Stokes et al. / Journal of Psychiatric Research 36 (2002) 417–421 Table 1 Summary of major findings Admission IV dexamethasone studies # Pts
Dex test status
9
NS
15
Supp
X 10 h post dex Plasma dex level
X Dex half-Life
1.33 (P=0.02) 2.44
3.25 (P=0.003) 4.75
Oral dexamethasone studies On admission
On discharge
# Pts
Dex test Status
X 10 h post dex Plasma dex level
8
NS
1.24
14
Supp
3.42
! (P=0.002) !
4. Discussion The principal findings in this study are the significantly shorter dexamethasone half-life and lower 10 h post dexamethasone levels found in nonsuppressor vs. suppressor depressed patients in both IV and oral DST studies. In addition, we found no difference in dexamethasone half-life between oral vs. IV dexamethasone studies done in the same three patients. Further, the change from nonsuppressor on admission to suppressor status at discharge was associated with an increase in 10 h plasma dexamethasone level while no changes in plasma dexamethasone level was found in suppressors on admission who remained suppressing at discharge. These findings extend and support previous data showing shorter dexamethasone half-life and lower plasma dexamethasone levels in nonsuppressors (Wiedemann and Holsboer, 1987; Holsboer et al., 1986; Poland et al., 1987; Stokes et al., 1987; Cassidy et al., 2000). Wiedemann and Holsboer (1987) reported no difference in dexamethasone kinetics during the absorption and distribution (early) phase after oral dexamethasone administration but found that the later elimination phase of dexamethasone from the circulation was significantly enhanced (i.e. shorter half-life) in six oral DST nonsuppressors resulting in an association of decreased plasma dexamethasone with elevated post dexamethasone cortisol levels. They also reported five IV nonsuppressors and eight IV suppressors whose plasma dexamethasone kinetics were indistinguishable and in the same order of magnitude as those of (14) nonsuppressors after oral DST. Examination of their IV dexamethasone data reveals a trend to lower plasma dexamethasone levels in the five nonsuppressors studied as compared to the eight IV nonsuppressors. We find significantly shorter half-life (i.e. increased dexamethasone metabolism) and lower plasma dexamethasone levels in
# Pts
Dex test Status
X 10 h post dex Plasma dex level
5
S
11
S
2.51 (P=NS) 3.37
nonsuppressors during the later elimination phase after IV dexamethasone administration, and paralleling this, lower dexamethasone levels in oral dexamethasone nonsuppressors consistent with shorter dexamethasone half-life and the report of Cassidy et al. (2000). There are several possible reasons for the reported differences in findings between the Wiedemann and Holsboer study and our current study. First, only five of their total number of IV dexamethasone studies (13) were nonsuppressors by their definition of nonsuppression. They defined nonsuppression as a post dexamethasone plasma cortisol > 4 ucg/dl at any one of the three post dexamethasone time points sampled (i.e. 08:00, 16:00, and 23:00 h). We have used the more demanding definition of a required 09:00 h plasma cortisol level greater than 5 ucg/dl to define nonsuppression as discussed previously (Stokes et al., 1984). One third (11/33) of their total patients were nonsuppressing by their criteria of nonsuppression. By our more rigorous definitions of nonsuppression, only three of their oral and only two of their IV dexamethasone study patients (5/33 or 15%) were nonsuppressing, and therefore comparable to our patients. Further, none of their IV dexamethasone and oral dexamethasone studies were compared in the same patients. Plasma cortisol levels during the day following nighttime administration of dexamethasone (circa 23:00 h) in Wiedemann and Holsboer’s study reveal marked fluctuations after the 08:00 h sample the next morning (see their Table 2). This variability was previously discussed (Stokes et al., 1984) where post dexamethasone plasma cortisol ‘‘patterns’’ were identified and found variable, especially in DST nonsuppressors. Thus, following evening administration of dexamethasone cortisol patterns from early morning to mid day to evening the next day may be described as showing various permutations of nonsuppression or suppression levels of cortisol at the three post dexamethasone sampling times. This prob-
420
P.E. Stokes et al. / Journal of Psychiatric Research 36 (2002) 417–421
ably reflects the initial and major early post administration suppressing affect of the dexamethasone followed by decreasing dexamethasone suppression as plasma and tissue levels fall. Superimposed on this is the recovering central ‘‘drive’’ of inherent brain pituitary activity overlain by transient sensory inputs (exciting or inhibiting) of the day (Stokes and Sikes, 1987) and by inadvertent sampling at various times during a secretory phase. The result is the fluctuation in post dexamethasone cortisol levels. A mechanism for shortened dexamethasone half-life in nonsuppressing patients has recently been proposed by our group (Stokes et al., in press). Many tissues metabolize glucocorticoids but experimental evidence supports the liver as the most likely site for the bulk of dexamethasone metabolism. Although cortisol itself has not been directly shown to induce hepatic microsomal enzymes, chronic administration of various synthetic glucocorticoids has been shown to be associated with increased metabolism of dexamethasone even in humans with renal insufficiency and no increase in renal steroid excretion (Araki et al., 1966). It is more than probable that the degradation of cortisol and dexamethasone share common pathways (Simmons et al., 1984). We postulated that the decreased dexamethasone half-life and lower plasma dexamethasone levels in psychiatrically ill patients who are DST nonsuppressors could be secondary to the frequently persistent moderate hypercortisolemia associated with but not specific to depression (Stokes et al., 1976; 1984) and the consequent potential probable induction of hepatic enzymes metabolizing dexamethasone. Our studies in rabbits (Stokes et al., in press), under pharmacologically induced hypercortisolemia, supported our hypothesis. In that study, dexamethasone plasma levels and half-life decreased (mean=41%) from control baseline values after induction of hypercortisolemia and returned to control values after cessation of the induced hypercortisolemia. Further, pre-dexamethasone cortisol values correlated negatively and highly with dexamethasone half-life. Interestingly, the percentage decrease in dexamethasone half-life in nonsuppressor patients compared to suppressors described by Holsboer et al., (1986; 43%) and in our current study (33%) is of similar magnitude (42%) we found in our rabbits during experimentally induced hypercortisolemia compared to their pre-hypercortisolemic state and consistent with that (41%) observed by Araki et al. (1966) in human subjects treated chronically with various glucocorticoids. The relatively small number of patients studied in our intravenous protocol and in other such studies (Wiedemann and Holsboer, 1987) suggests that confirmatory investigation be completed in larger numbers of clearly dexamethasone nonsuppressing versus suppressing patients. It will also be useful to confirm our findings of
no difference in dexamethasone half-life between IV and oral dexamethasone administration in the same patients. In addition, a study of dexamethasone half-life in patients before and during glucocorticoid treatment, preferably with hydrocortisone, would provide additional clarifying data. We wanted to assay urinary levels of the cortisol metabolite 6BOH cortisol in these suppressors and nonsuppressors as a further measure of altered hepatic enzyme activity (probably CYP3A4) since glucocorticoids have been shown to induce CYP3A, but were unable to accomplish this portion of the study.
Acknowledgements From the Laboratory of Psychobiology—Endocrinology in the Department of Psychiatry at Weill Medical College of Cornell University. Supported by grants from the Citicorp-Wriston Endowment Scholarship Fund, George F. Baker Trust, and gifts from Robin Jaffe-Cohen. The authors wish to thank, Hermina Ombid, our laboratory assistant, and research assistant, Alexandra I. Barsdorf for statistical analyses, library research, and manuscript preparation.
References Araki Y, Yokota O, Kato T, Kashima M, Miyazaki T. Dynamics of synthetic corticosteroids in man. In: Pincus G, Nakao T, Tait JF, editors. Steroid dynamics. New York: Academic Press; 1966. p. 474–7. Arana GW, Workman RJ, Baldessarini RJ. Association between low plasma levels of dexamethasone and elevated levels of cortisol in psychiatric patients given dexamethasone. American Journal of Psychiatry 1984;141:1619–20. Berger M, Pirke KM, Doerr P, Krieg JC, von Zerssen D. The limited utility of the dexamethasone suppression test for the diagnostic process in psychiatry. British Journal of Psychiatry 1984;145:372– 82. Cassidy F, Ritchie JC, Verghese K, Carroll BJ. Dexamethasone metabolism in dexamethasone suppression test suppressors and nonsuppressors. Biological Psychiatry 2000;47:677–80. Holsboer F, Wiedemann K, Gerken A, Boll E. The plasma dexamethasone variable in depression: test–retest studies and early biophase kinetics. Psychiatry Research 1986;17(2):97–103. Johnson GF, Hunt G, Lerr K, Caterson I. Dexamethasone supression test (DST) and plasma dexamethasone levels in depressed patients. Psychiatry Research 1984;13:305–13. Morris H, Carr V, Gilliland J, Hooper M. Dexamethasone concentrations and the dexamethasone suppression test in psychiatric disorders. British Journal of Psychiatry 1986;148:66–9. Poland RE, Rubin RT, Lesser IM, Lane LA, Hart PJ. Neuroendocrine aspects of primary endogenous depression: II. Serum dexamethasone concentrations and hypothalamic-pituitary-adrenal cortical activity as determinants of the dexamethasone suppression test response. Archives of General Psychiatry 1987;44:790–5. Sikes CR, Stokes PE, Lasley B, Carson RC. Cognitive sequelae of hypothalamic-pituitary-adrenocortical dysregulation. The 143rd Annual Meeting of the American Psychiatric Association, New York, NY, 12–17 May, 1990.
P.E. Stokes et al. / Journal of Psychiatric Research 36 (2002) 417–421 Simmons PS, Miles JM, Gerich JE, Haymond MW. Increased proteolysis: an effect of increases in plasma cortisol within the physiologic range. Journal of Clinical Investigation 1984;73:412. Stokes PE. The potential role of excessive cortisol induced by HPA hyperfunction in the pathogenesis of depression. European Neuropsychopharmacology 1995;5(Suppl):77–82. Stokes PE, Sikes C. Hypothalamic-pituitary-adrenal axis in affective disorders. In: Meltzer HY, editor. Psychopharmacology: a third generation of progress. New York: Raven Press; 1987. p. 589– 607. Stokes PE, Sikes CR, Stoll PM, Lasley BJ. Dexamethasone half-life affects DST response. Presented at the 42nd annual meeting of the Society of Biological Psychiatry, Chicago, IL May, 1987. Stokes PE, Stoll PM, Koslow SH, Maas JW, Davis JM, Swann AC, Robins E. Pretreatment DST and hypothalamic-pituitary-adreno-
421
cortical function in depressed patients and comparison groups: a multi-center study. Archives of General Psychiatry 1984;41:257–67. Stokes PE, Stoll PM, Mattson MR, Sollod RN. Diagnosis and psychopathology in psychiatric patients resistant to dexamethasone. In: Sachar EJ, editor. Hormones, behavior, and psychopathology. New York: Raven Press; 1976. p. 225–9. Stokes PE, Stoll PM, Schluger JH, Lasley B. Hypercortisolemia decreases dexamethasone half-life in rabbit. Journal of Psychiatric Research 2002 (in press). Wiedemann K, Holsboer F. Plasma dexamethasone kinetics during the DST after oral and intravenous administration of the test drug. Biological Psychiatry 1987;22:1340–8. Wolkowitz OM, Reus VI, Weingartner H, Thompson K, Breier A, Doran A, Rubinow D, Pickar D. Cognitive effects of corticosteroids. American Journal of Psychiatry 1990;147(10):1297–303.